PS590AB

treor_

fawcett.pdf

Home >Psychology homework help >PS590AB

Mental Health Treatment for Children and Adolescents:

Cost Effectiveness, Dropout, and Recidivism by

Presenting Diagnosis and Therapy Modality

David Fawcett

A dissertation submitted to the faculty of Brigham Young University

in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

D. Russell Crane, Chair Roy A. Bean

Jeffry H. Larson Richard B. Miller James M. Harper

School of Family Life

Brigham Young University

December 2012

INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.

In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed,

a note will indicate the deletion.

unauthorized copying under Title 17, United States Code

ProQuest LLC. 789 East Eisenhower Parkway

P.O. Box 1346 Ann Arbor, MI 48106 - 1346

UMI 3553419

Published by ProQuest LLC (2013). Copyright in the Dissertation held by the Author.

UMI Number: 3553419

ABSTRACT

Mental Health Treatment for Children and Adolescents: Cost Effectiveness, Dropout, and Recidivism by

Presenting Diagnosis and Therapy Modality

David Fawcett School of Family Life, BYU

Doctor of Philosophy

As many as one in five children and adolescents may suffer from a mental health disorder, yet there are barriers that often prevent children from receiving optimal treatment. The current study explores the influence of practitioner license type, therapy modality, diagnosis, age, and gender on mental health therapy for children and adolescents. Data was provided by Cigna, a leading health care insurance provider in the United States. Participants include 106,374 boys (53.2%) and 93,753 girls (46.8%) ages 3 to 18 (M = 12.1, SD = 3.9) who were treated in outpatient facilities throughout the United States of America. Results indicate that there are differences in dropout, recidivism, cost, and treatment length by provider license, therapy modality, diagnosis, age, and gender. Specifically, results suggest that marriage and family therapists have the lowest percent recidivism and are among the lowest in terms of dropout and cost effectiveness. The results also suggest that family therapy is more cost effective than individual or mixed therapy and that mixed therapy has a much lower percent dropout than individual or family therapy. Analysis by diagnosis suggests a potential severity scale based on dropout, recidivism, and number of sessions. There are also significant differences in dropout and recidivism by age suggesting that younger children are more likely to dropout of treatment. These results provide valuable information about mental health treatment of children and adolescents. Specifically, utilizing a family based approach may help reduce the total length of treatment while utilizing a mixed mode approach to therapy may help reduce the risk of dropout from treatment. Also, some diagnoses appear to be more difficult to treat, with higher percentages of dropout and requiring more time and money for successful treatment. Limitations and future directions are discussed.

Key words: child adolescent therapy, mental health license type, therapy modality, family therapy, dropout, diagnosis, recidivism, mixed therapy, retrospective analysis, Cigna, cost, cost effectiveness, number of sessions, treatment length.

ACKNOWLEDGMENTS

I would like to express my profound appreciation to my advisor and mentor, Dr. Russell

Crane for his example and his attention to providing me with a valuable education. Through his

confidence and trust I have grown as a researcher and as an individual. I am grateful for the

many opportunities that he has helped provide to me. I have learned a great deal under his

excellent mentorship and look forward to continuing to learn from him. I am grateful to the

members of my committee, who continue to provide supportive and constructive feedback on my

clinical and scholarly work. I am grateful for the guidance and support from each of the faculty

members in the marriage and family therapy program. They have taught me much and for that I

am thankful.

I am especially grateful to my wife Elizabeth. There can be no question that without her

love and support I would not have made it to this point. She has been, and continues to be, an

example of excellence. Her patience and faith toward me are unparalleled. She is my best friend

and biggest support and I love her completely. I thank my three sons for their encouragement

and patience. I am grateful for my amazing parents and my siblings and for all of their examples

of hard work and for their continued support. Finally, I am grateful to my God for the many gifts

and subtle guidance in my life.

Table of Contents

Introduction………………………………………………………………………………………. 1

Literature Review........................................................................................................................ 3

Differences between children and adults in mental health care. ............................................. 3

Individual based treatment ...................................................................................................... 4

Family based treatment ........................................................................................................... 5

Effectiveness of family/systemic based treatments. ............................................................... 9

Combination of individual and family therapy. .................................................................... 12

Differences by license type. .................................................................................................. 13

Drop out and recidivism ........................................................................................................ 14

Cost effectiveness research with children and adolescents. .................................................. 20

The current project. ............................................................................................................... 23

Research Questions ................................................................................................................... 24

Method .......................................................................................................................................... 25

Design ....................................................................................................................................... 25

Sample....................................................................................................................................... 26

Participants. ........................................................................................................................... 26

Procedure .................................................................................................................................. 27

Data cleaning ........................................................................................................................ 27

Definitions................................................................................................................................. 27

Episode of Care (EoC). ......................................................................................................... 27

Therapy modality. ................................................................................................................. 27

Diagnoses. ............................................................................................................................. 28

Cost. ...................................................................................................................................... 28

Cost effectiveness. ................................................................................................................ 28

Dropout. ................................................................................................................................ 29

Types of treatment services. ................................................................................................. 29

Recidivism. ........................................................................................................................... 29

Number of sessions. .............................................................................................................. 30

Analysis................................................................................................................................. 30

Results ........................................................................................................................................... 30

Discussion ..................................................................................................................................... 37

Clinical Implications ................................................................................................................. 45

Limitations and Future Directions ............................................................................................ 47

References ..................................................................................................................................... 49

List of Tables

Table 1: Results for dropout and recidivism by license type and therapy modality.………..… 68

Table 2: Results of statistical analyses for outcomes by license type and therapy modality..… 69

Table 3: Tukey post-hoc analyses for comparisons in Table 2……………………………....... 70

Table 4: Results of statistical analyses for outcomes by diagnosis….………………………... 78

Table 5: Tukey post hoc analyses for comparisons by diagnosis………………………..……. 79

Table 6: A summary of the percentage of cases treated by each license type……………..… 102

Table 7: Analysis results by diagnosis and license type……………………………………... 103

Table 8: Results of statistical analyses for outcomes of family therapy by license type…..… 107

Table 9: Treatment outcomes by age……………………………………………….………... 108

Mental Health Treatment for Children and Adolescents: Cost Effectiveness, Dropout, and

Recidivism by Presenting Diagnosis and Therapy Modality

It has been estimated that up to 20% of children and adolescents suffer from a serious

mental health disorder (Belfer, 2008). However, most of these children do not receive treatment

from a mental health provider (Ani & Garralda, 2005; Kataoka, Zhang & Wells, 2002). Of

children who would benefit from mental health services, it has been estimated that between 50%

and 75% of them either never present for treatment or fail to complete treatment (Kazdin,

Mazurick, & Bass, 1993), which is a significant barrier to effective treatment implementation

(Watt & Dadd, 2007). Because mental health issues are pervasive and serious among children, it

is important to explore variables that may help with treatment retention and help produce

positive treatment outcomes.

Children who struggle with mental health problems have difficulties in many different

aspects of their lives (Paster, 1997). They are often disruptive and show behavioral and learning

problems in school. They present symptoms in the home. They often have trouble with the law,

have health issues, are abused or neglected, and are often socially isolated. Additionally, it has

been shown that when children and adolescents struggle with mental health issues, others in the

family can also be negatively affected. For example, research suggests that parents of children

(ages 10 to 15) who are seeking mental health treatment are more likely to report impact on the

family, such as parental well-being, depression, and parental feelings of incompetence (Farmer,

Burns, Angold, & Costello, 1997).

It has been suggested that studying the role of early childhood health, including mental

health, will improve understanding of larger social issues such as social stratification, wage

determination, and intergenerational transmission of inequalities (Palloni, 2006). Palloni also

suggests that “early childhood health matters for achievement of or social accession to adult

social class positions” (p. 587).

It has been argued that families are the central foundation of a civilization’s social

structure (Loveless & Holman, 2007). Supporting families in the act of raising children,

promoting active and meaningful community involvement, and helping families care for social,

physical, and psychological needs should be a central goal of any society (Huffine & Anderson,

2003). It is, therefore, unfortunate, and potentially destructive to society, that so many families

and children do not receive the support and services that they need to flourish.

A recent estimate of the annual cost of mental, emotional, and behavioral disorders in

children and adolescence totaled $247 billion (Eisenberg & Neighbors, 2007). Services related

to the treatment and care of children with mental health needs are in high demand, and there are

struggles related to the decisions about how such resources should be allocated (Stevens, Roberts,

& Shiell, 2010). There is pressure from health care systems and policy makers to only fund

treatment approaches that are evidence based. This, in turn, leads to competition for limited

resources among different mental health professions and psychotherapeutic traditions (for

example, medication vs. talk therapies). Additionally, there is a high level of complexity

apparent in children’s services with multifaceted interventions, multiple agencies and providers

that may have various primary objectives or outcomes (Stevens et al., 2010). When considering

an investment into improved mental health services for children, it is important to explore the

evidence of cost effectiveness. It has been noted that a better understanding of the economic

impacts of interventions, along with the potential gains, can help to make decisions about how

valuable resources can best be utilized (McDaid, Park, Knapp, Losert, & Kilian, 2010). As a

result, it is important for practitioners to be familiar with approaches and modalities that are

effective, efficacious, and cost effective when working with children. The current study explores

the influence of practitioner license type (and associated training), and therapy modality

(individual, family therapy, or a combination of both) on therapy outcomes for children receiving

mental health care services. The design of the current study has a high level of external validity.

The sample is nationally representative and is not based on study recruitment; rather, it contains

data on what type of treatment children and adolescents are currently receiving in the United

States.

Literature Review

A review of literature related to the treatment of childhood mental health issues is here

presented. Some of this literature is indirectly tied to the current research project. It should be

noted that the indirectly related literature is provided as background and supporting information

for the reader.

Differences between children and adults in mental health care. Weisz, Huey, and

Weersing (1998) note some important differences between the mental health treatment of adults

and children. One difference is that fact that children rarely consider themselves as needing

mental health therapy. As a result, most child therapy cases are referred by adults and not by the

child. This creates an interesting distinction of having a child as the identified patient with a

parent or other adult playing the role of the client. One important implication of this is the

inherent systemic inclusion. The child, as an individual, would not likely present to therapy on

her own; the larger system acts to bring the child to therapy. This suggests that child therapy is

inherently systemic. As a result, it is likely that optimal treatment for children in psychotherapy

settings would include various dimensions of larger systemic components.

Another difference between treating adults and children in a mental health setting

involves environmental selection (Weisz, Huey, & Weersing, 1998). Children are more captive

than adults to a larger systemic environment. As a result, the childhood disorder that is being

treated may stem from environmental influences (e.g., school, family) and not from the child. If

this is the case, involving the child in individual therapy alone will likely limit the impact of

interventions. It is likely that a more effective solution would involve others from the child’s

social environment, though there are additional challenges that arise from including others in

treatment.

Individual based treatment. Literature on empirically based individual treatments

suggests various effective interventions for children and adolescents (See Kendall & Beidas,

2007). These include: addressing inaccurate self-perceptions for youth with depression (Stark &

Kendall, 1996); role play exercises to address misattribution of intentionality for youth with

conduct and aggression problems (Lochman, Powell, Whidby, & Fitzgerald, 2006; Nelson, Finch,

& Ghee, 2006); and emotion management and behavior modification for children with

internalizing problems linked to anxiety (Kendall, Hedtke, & Aschenbrand, 2006). Evidence for

empirically supported treatments for children and adolescents has been found for many disorders

including: conduct and aggression problems (Brinkmeyer & Eyberg, 2003; Lochman et al.,

2006); depression (Mufson, Dorta, Wickramaratne, Nomura, Olfson, & Weissman, 2004); and

anxiety disorders (Kendall et al., 1997; Piacentini, March, & Franklin, 2006). One study found

that many children had been involved in several other treatment options before individual

psychotherapy was pursued; in some cases it was reported as a feeling of a last resort (Kam &

Midgley, 2006). Other research suggests that even though certain treatments have been found to

be empirically supported, they are not often used by therapists in regular clinical practice

(Goisman, Warsaw & Keller, 1999). Additionally, when the empirically supported treatments

are used in routine clinical practice, the clinical outcomes are not as positive as seen in the

evaluative studies (Stewart & Chambless, 2009). Though a comprehensive review of empirically

supported, individually based treatments for children and adolescents is outside the scope of the

current project, there are many resources for the interested reader (see Kendall, 2011).

Family based treatment. Children represent a unique clinical population. Typically,

adults seek mental health services for themselves. Children, however, are usually brought to

therapy by parents or a guardian. Parents are also pivotal in deciding the type of treatment that

the children receive, as well as, the frequency and number of sessions that the child attends

(Bannon & McKay, 2005). Children also encounter additional obstacles to obtaining mental

health therapy, such as lack of transportation or lack of child care for siblings (Hahn, 1995;

McKay, McCadam, & Gonzales, 1996). Research has also identified perceived barriers to

obtaining mental health services. These barriers include issues such as not having time to make

it to appointments, a lack of social support, and negative views on potential treatment outcomes

(Kazdin, Holland, & Crowley, 1997; Nock & Kazdin, 2001). Other research suggests that

children and adolescents from two-parent families are more likely to continue with treatment

than those from one-parent families, families with high socio-economic status (SES) are more

likely to continue treatment than those from lower SES, and clients who belong to the majority

social group are more likely to continue treatment than those from minority groups (Armbruster

& Fallon, 1994). As there is a clear systemic influence on children and adolescents receiving

mental health services, it is reasonable to suppose that including the family system in treatment

would lead to improved outcomes and lower dropout rates.

Research has found that when mental health services are matched with parental

preference for service that the number of treatment sessions attended increases significantly

(Bannon & McKay, 2005). These findings suggest that parental involvement in decision making

is particularly important for ensuring that children complete treatment. Research looking at

behavioral interventions targeted at reducing children’s anxiety and increasing compliance prior

to anesthesia has shown that interventions that involve parents in the surgical holding area

significantly improve the compliance rates of the children (MacLaren & Kain, 2008). Children

tend to show less anxiety and distress when a supportive parent is involved in the treatment

process. The same would likely be true when children are receiving mental health treatment.

Evidence also suggests that treating a depressed family member may result in the improvement

of another family member’s depressive symptoms (Hughes & Asarnow, 2011). Studies have

also suggested a correlation between a parent’s and child’s adjustment to chronic illness (Lopez,

Mullins, Wolfe-Christensen, & Bourdeau, 2008). When parents adjust well to the chronic illness,

the child tends to also adjust well. This evidence suggests that increasing family involvement in

the child’s therapeutic process is advantageous to the mental health treatment.

Not only is there evidence that family involvement is advantageous to children’s mental

health treatment, there is also evidence that mental health issues in other family members can

have a negative effect on a child’s mental and physical health (Janicke, Finney, & Riley, 2001).

Research suggests that the parent’s marital cohesion and life satisfaction are significant

predictors of children’s health care use. One study found that lower life satisfaction and higher

reports of martial cohesion were associated with more use of health care services by the child

(Crane, Christenson, Shaw, Fawcett, & Marshall, 2010). One study found that having at least one

parent with depression is related to higher emergency department visits, sick visits, inpatient

services, and specialist visits across all ages of children (Sills, Shetterly, Xu, Magid, & Kempe,

2007). They also found that children (ages 13-17) had a lower rate of well-child visits when at

least one parent suffered from depression. These findings lend evidence to the systemic

influence of mental health in the family. They also suggest that treating the family system, rather

than just the individual child, may lead to better outcomes for treatment of the presenting

problem.

Similarly, literature on family burden discusses the emotional experiences of families that

are coping with acute and long term responsibilities related to inadequate systems of treatment

and community care (Riesser & Schorske, 1994). Burdens include financial costs, disruptions to

family life, worry, a sense of loss, and social isolation. Because the word ‘burdens’ carries more

of a negative tone, some prefer to address these issues in terms of caregiver strain and family

impact (Farmer et al., 1997). In fact, some people report that caring for loved ones with special

needs actually enriches their lives (Yorgason, Booth, & Johnson, 2008; Yatchmenoff, Koren,

Friesen, Gordon, & Kinney, 1998). Enrichment is related to families feeling empowered and

having a sense of competence in their own ability to aid in treatment and deal with the symptoms.

Families who are well informed about the specific needs related to care, are connected to

resources, and have an active role in the treatment of family members with special treatment

needs are likely to experience more family enrichment and cohesion than those who are excluded

from the treatment process. It thus follows that families who are well informed about the

specific needs related to treatment and care of a child’s mental health issue are more likely to

experience greater cohesion and satisfaction with treatment. Full continuums of family oriented

services, including clinical involvement, such as therapy and psychoeducation, as well as non-

clinical involvement, such as education and social support, have been recommended to be

available to serve families of individuals who are struggling with mental illness (Marsh &

Johnson, 1997).

One education-based intervention is parenting training. Parenting training is a well-

known form of family involvement in child centered interventions and is often recommended to

help caregivers improve their parenting skills or basic management techniques (Friesen &

Stephens, 1998). One such program, which has been empirically studied, is the Triple P-Positive

Parenting Program (Sanders, Markie-Dadds, & Turner, 2001; Sanders & Pidgeon, 2005).

Research with participants in the Triple P program suggest that mothers show significant

improvements in parenting, parenting self-esteem, and reductions in stressors related to parenting

(Bodenmann, Cina, Ledermann, & Sanders, 2008). Additional research suggests that parenting

programs can increase confidence, improve relationships with children, and help with the

implementation of behavioral techniques (Patterson, Mockford, & Stewart-Brown, 2005). While

not directly related to the current study, research on parenting programs lends evidence to the

utility of including family members in the treatment of children and adolescents; when parents

and the larger caregiving system are included in the treatment process, children tend to have

better clinical outcomes.

Systems of care that promote family participation and involvement in children’s mental

health treatment are likely to show better clinical outcomes for the children and their families. A

system of care has been defined as “A comprehensive spectrum of mental health and other

necessary services which are organized into a coordinated network to meet the multiple and

changing needs of children and adolescents with serious emotional disturbances and their

families.” (Stroul & Friedman, 1986, p. 3). Families become empowered, better equipped to care

for their children, and increasingly involved in strengthening the family system (McCammon,

Spencer, & Friesen, 2001). Additionally, mental health treatment that involves the family is

likely to show longer lasting results due to the collaborative efforts of the family to modify the

family system rather than changing a single component and then placing it back into the same

system. When addressing the needs of children who are struggling with mental health issues, a

key guiding principle of an effective system of care is that families should participate in all

aspects of treatment planning and delivery (Stroul & Friedman, 1996). Family involvement is a

critical aspect of successful treatment of and recovery from mental health disorders. The term

family typically refers to biological nuclear relations but should also include extended kin

caregivers (i.e. grandparents, aunts & uncles, in-laws, etc.), adopted families, and foster families.

Some researchers have suggested that the degree to which mental health professionals are able to

assist children in reaching their treatment goals depends largely on the amount of true

collaboration between the mental health professional and the family (DeChillo, Koren, &

Schultze, 1994).

As family members are inextricable components of a child’s environment, family

involvement in the treatment of children with behavioral problems becomes crucial to treatment

success (Friesen & Stephens, 1998). Treatment goals and planning should include the relevant

caregivers within the family system. When treating children and adolescents, family members,

in addition to the identified patient, should be considered for involvement in treatment, recovery,

and the overall change process.

Effectiveness of family/systemic based treatments. Research has demonstrated that

systemic based therapy is an effective treatment intervention for various mental health issues and

diagnosis for both adults and children. A review of 20 meta-analyses that explored systemic

based interventions of mental health issues found that families who entered treatment together

showed better functioning after therapy and at follow-up than did 71% of families in control

groups (Shadish & Baldwin, 2003). The average effect size across the meta-analysis studies

was .65 after therapy and .52 at follow-up, which occurred 6-12 months later. This review

provides further evidence that systemic interventions, including family based treatments, are

clinically effective for treating mental health issues. A recent review of literature suggests that

family based interventions are helpful for treating children and adolescents who are struggling

with a variety of disturbances including: mood disorders, anxiety, attention-deficit hyperactivity,

disruptive behavior, pervasive developmental, and eating disorders (Kaslow, Broth, Smith, &

Collins, 2012)

Additional evidence shows that systemic interventions for child-based mental health

issues are an effective form of treatment. Carr (2009b) created a summary of evidence on

systemic treatment for child-focused problems. His review indicates that systemic based

interventions, such as family therapy, are clinically effective, brief – usually less than 20 sessions,

and can be offered by a range of mental health professionals in outpatient settings. The research

shows specific evidence that systemic interventions are effective for treating issues including:

behavioral difficulties, ADHD, drug abuse, delinquency, anxiety, depression, grief, bipolar

disorder, child abuse and neglect, eating disorders, enuresis, encopresis, infant sleep, feeding,

and attachment problems, and poorly controlled asthma and diabetes. Another review of articles

concerning family and systemic based treatment approaches suggests that they can be effective

treatments for adolescent sex offenders, juvenile delinquency, adolescent anorexia nervosa, and

children at risk of out-of-home placement (Carr, 2010).

A study comparing individual psychodynamic psychotherapy and family therapy for the

treatment of childhood depression found that both treatment approaches resulted in significant

reductions in disorder rates (Trowell et al., 2007). Of the cases treated with individual

psychotherapy, 74.3% were no longer clinically depressed and of the cases treated with family

therapy, 75.5% of the cases were no longer clinically depressed. Another study compared the

relative long term benefit of family-focused cognitive behavioral therapy and child-focused

cognitive behavioral therapy for childhood anxiety disorders (Wood, McLeod, Piacentini, &

Sigman, 2009). The results suggest that the children who had been assigned to the family based

therapy had lower anxiety scores from diagnostician and parent report scores at a one year

follow-up. Family therapy has also been shown to be effective for the treatment of adolescents

with anorexia nervosa (Lock & Fitzpatrick, 2007). In a qualitative analysis of why parental

involvement enhances the effectiveness of treatment for anorexia nervosa, researchers found that

parent-to-parent consultations were viewed as intense emotional experiences that helped parents

reflect on changes in the family interactions, feel less isolated, and feel empowered to continue

treatment (Rhodes, Brown, & Madden, 2009).

There is also empirical evidence for specific family based treatment models. Functional

Family Therapy has been identified as an effective, evidence-based intervention by several

reviews (Alexander & Sexton, 2002; Waldron & Turner, 2008). Multisystemic Therapy, a

family based treatment model, has been shown to be effective and efficacious for treating

children and adolescents with studies dating back 35 years (see Henggeler, 2011). Empirical

evidence suggests that it is effective for treating adolescent sex offenders, delinquency, substance

abuse, externalizing symptoms, and out-of-home placements (Curtis, Ronan, Heiblum, & Crellin,

2009; Henggeler et al., 2009; Letourneau et al., 2009) and Dyadic Developmental Psychotherapy

has also been presented as an effective and evidence-based treatment for children and

adolescents (Becker-Weidman & Hughes, 2010). In summary, there is considerable evidence to

suggest that family based treatments are effective for the treatment of children and adolescents.

Combination of individual and family therapy. Studies have also explored the benefits

of combining individual and family sessions over the course of treatment. Combining individual

and family based sessions when working with children and adolescents is a treatment approach

that has received some attention over the years. Feldman (1988) presented an integrative

approach of family and individual based sessions together with clinical examples. Guidelines for

deciding when to use concurrent individual and family therapy as a treatment modality have been

presented by Racusin and Kaslow (1994). Josephson and Serrano (2001) discussed how

individual and family based therapies can be seen as complementary rather than separate and

competing. Another model combines two successive individual sessions, followed by one family

session, for the treatment of sexual behavior problems (Etgar & Shulstain-Elrom, 2009). The

family sessions in this model include the child, both parents, and sometimes siblings. The

authors note the importance of including family therapy for the treatment of children in this

population. Other research concerning attention deficit hyperactivity disorder has suggested the

necessity of using a combination of individual psychotherapy and family based treatment

modalities (Stubbe & Weiss, 2000). While many have suggested the utility of using a

combination of individual and family based psychotherapy approaches to treatment with children,

no studies were found that test the outcomes of a mixed modality compared to individual

psychotherapy or family therapy. When research addresses both modalities in a single paper it is

usually to compare the outcomes of each to the other rather than explore the benefits of

combining both together (see Hughes & Asarnow, 2011). There is a clear need for outcome

research that combines individual and family sessions in the treatment of children and

adolescents.

Differences by license type. In addition to differences in therapy outcomes between

family based and individual based treatments, there is some evidence that the provider license

type may influence treatment outcomes. One study found differences across license types for

dropout rates, recidivism rates, and cost effectiveness for mental health treatment in general

(Crane & Payne, 2011). Specifically, they found that professionals with marriage and family

therapy licenses, which require specific training in family therapy, had lower recidivism rates.

Another study found differences across license types for outcomes of family based treatments

(Moore, Hamilton, Crane & Fawcett, 2011). The results suggest that licensed marriage and

family therapists had lower dropout and recidivism rates, compared to other licenses, when

providing family therapy. Other research suggests that there may be differences by professional

license in terms of the accuracy of diagnosing sexual abuse (Shumaker, 2000). Some of these

differences may be explained by differences in educational training and general approaches to

therapy between the professional licenses. For example, in a content analysis of literature on

clinical licensure programs, it was noted that marriage and family therapists are required to

complete at least three times more family therapy coursework than clinical psychologists,

psychiatrists, psychiatric nurses, professional counselors, and social workers (Crane, Shaw,

Christenson, Larson, Harper, & Feinauer, 2010). They also noted that marriage and family

therapists must complete 16 times more supervised face-to-face therapy hours than any of the

other professions. These are only a few examples of differences in training and license

requirements between mental health professions. It is possible that differences in professional

training and licensing requirements will likely result in some differences in treatment outcomes.

Learning about differences in treatment outcomes between professional license types may help to

discover specific benefits for working with children within the different training approaches.

Drop out and recidivism. Dropout and recidivism are commonly used as measures of

treatment outcome. Therapy dropout, defined as terminating mental health treatment prior to its

completion, is a primary obstacle to providing effective mental health services to children and

families (Nock & Kazdin, 2001). Estimates of dropout rates for child and adolescent treatment

range from 28% to 75% (Kazdin, Mazurick, & Siegel, 1994; Lai, Pang, Wong, Lum, & Lo,

1998). Dropout presents problems related to the quality of mental health care, cost of treatment,

and treatment outcome research. It consumes valuable time and financial resources related to the

intake, assessment, and administrative costs that may be utilized by other children and

adolescents who are in need of these services (Armbruster & Kazdin, 1994; Masi, Miller, &

Olson, 2003; Prinz & Miller, 1994). It also leaves many children and adolescents with untreated

mental health problems. Untreated problems, such as these, leave the children and adolescents

vulnerable to immediate and long-term personal, familial, and social difficulties (Farmer et al.,

1997). Also, children and adolescents who do not complete mental health treatment are less

likely to show improve in their symptoms than clients who complete their treatment (Kazdin,

Mazurick, & Siegel, 1994; Prinz & Miller, 1994). As such, dropout from treatment is related to

the efficacy of the treatment (Johnson, Mellor, & Brann, 2008). Thus, understanding, predicting,

and preventing therapy dropout are important issues for child mental health practitioners and

services.

Researchers have noted that there are some discrepancies and problems with operational

definitions for dropout from mental health treatment (Johnson et al., 2008). The use of different

operational definitions of dropout can make comparisons across studies difficult and confusing.

Studies generally refer to dropout as termination of treatment prior to treatment completion. It is,

however, difficult to define and measure the moment when mental health treatment is complete.

One common operational definition of dropout involves using a cut-off number of sessions.

Clients who attend fewer sessions than the cut-off number are considered as dropouts. Clients

who attend more sessions than the cut-off are considered to complete treatment. Dropout has

also been measured as failing to attend appointments that have been scheduled and not returning

for additional treatment. It has been noted that while this method may be reliable, it does not

consider a client as a dropout if they make their intention to not return explicit by not scheduling

a future appointment (Johnson et al., 2008). Thus, this method does not fully capture the

measurement of dropout.

In addition to differences in operational definitions of treatment dropout, there are also

different reasons for which children, who are receiving mental health services, terminate

treatment. Research on why youth and adolescents dropout of therapy has found that therapeutic

relationship problems accounted for the most variance in a factor analysis (Garcia & Weisz,

2002). Utilizing the 41 item Reasons for Ending Treatment Questionnaire (RETQ), the study

identified that participants reflected concerns that therapists did not appear to be doing the right

things, they were not addressing the right problems, they were not talking enough with family

members, or they were not helping the child. Other factors for dropout included family issues

such as transportation problems or ill family members. Financial issues, time and effort required

to get to appointments, and perception of the need for treatment were also variables that

influenced dropout for children and adolescents. The study identified therapeutic relationship as

the only non-financial related variable that distinguished dropouts from those who completed

therapy.

Other research suggests that parent expectancies for therapy help predict completion of

mental health treatment. A study by Nock and Kazdin (2001) involving 405 children and their

parents looked at the relationship between parent expectancies for therapy and early termination.

They found that parents with lower expectancies of therapy had higher barriers, such as obstacles

to coming to therapy, perceptions that treatment is irrelevant and too demanding, and poor

relationships with the therapist. These barriers often prevented the parents from bringing the

child to therapy. They also found evidence for a curvilinear relationship between expectancies

and therapy attendance; parents with very high and very low expectations for therapy attended

the highest number of sessions and were least likely to drop out of treatment. Research suggests

that therapist’s concern and the fit of treatment were important factors for helping clients stay in

treatment (Allgood & Crane, 1991) Other research related to parental influences on drop out

suggests that parent age and marital status also have an influence on dropout (Kazdin et al.,

1993). Specifically, younger mothers and single mothers of children being treated for conduct

disorder are more likely to have children dropout of treatment.

Research has identified factors related to attrition from treatment of child and adolescent

mental health treatment. These factors include: socio-economic status, referral sources,

geographical distance to services, minority status, pathology attributed to parent figure, previous

treatment attempts, and waiting for services to begin (Armbruster & Kazdin, 1994; Kazdin,

Mazurick, & Bass, 1993; Luk et al., 2001). Another study found that the number of children in a

family influences the likelihood of dropout (Allgood & Crane, 1991). The study found that a

family was more likely to stay in therapy as the number of children in the family increased. The

study also found that people attending conjoint therapy for an individual problem were more

likely to drop out of therapy.

Most dropout research for children and adolescents has explored the phenomenon from

an individual therapy modality. Relatively little research has explored differences in dropout

rates when comparing individual and family therapy modalities. One study found no differences

in dropout rates between three therapy modalities: individual, couple, and family therapy (Masi,

Miller, & Olson, 2003). The study compared differences using three different operational

definitions of dropout: a minimum cut-off number, therapist judgment of clients dropping out,

and treatment ending before the therapeutic goals were met. The results of the study found no

differences in dropout rates by therapy modality. One primary limitation of the study was that it

was a sample of data from only one (training) clinic. The current study will compare dropout

rates for children and adolescents using a large sample with participant data treated by

professionals with different license types and from different treatment facilities across the United

States.

As no one factor has been identified as being sufficient to predict treatment dropout for

children and adolescents, there may be a more complex influence of multiple factors that

contribute to an increased likelihood of premature termination of treatment (Kazdin, Holland, &

Crowley, 1997). Also, though many factors have been associated with dropout from therapy, it is

important to consider multiple possibilities. There is limited information exploring the influence

of provider license and therapy modality on dropout rates for children and adolescents.

The diagnosis that a child receives will also influence their chance of dropping out of

treatment. Individuals with the same diagnosis tend to share specific characteristics related

individual functioning and social interactions. It is reasonable to assume that children who share

the same diagnosis will likely have similar dropout rates and that differences in dropout rates

between diagnoses may be found. Previous research on dropout by diagnosis suggests that

dropout is higher among conduct disorder clients than among individuals with other diagnoses

(Armbruster & Kazdin, 1994). Other research suggests that children with behavioral problems,

such as aggression, antisocial behavior, and delinquency, are more likely to drop out of treatment

(Dickens & Campbell, 2001; Dierker, Nargiso, Wiseman, & Hoff, 2001; Kazdin et al., 1994;

Kazdin & Mazurick, 1994). Another study by Johnson et al. (2008) suggests that children and

adolescents who were diagnosed with eating disorders (71%), conduct disorder (63%), ADHD

(58%), or family problems (62%) have the highest dropout rates when compared to other

diagnoses. Children with anxiety disorders (39%), and no diagnosis (21%), were less likely to

dropout. Despite differences in diagnosis, generally, dropout from mental health services among

youth tended to be high (49%). As differences in dropout rates between diagnoses have been

shown previously, it will be important to explore additional factors that may contribute to

childhood dropout from treatment.

In addition to early termination being a barrier for children in need of mental health

services, recidivism is also an important area of concern for measuring treatment outcome.

Recidivism refers to clients returning for additional treatment following an episode of care where

the clients were treated for the disorder. One underlying assumption of recidivism is that the

longer a client can continue without follow-up treatment, the more efficacious or successful the

previous treatment is considered (Hafemeister & Banks, 1996). Thus, lower recidivism rates are

assumed to be associated with a more effective treatment.

Recidivism is related to issues such as additional costs for treatment, improvement for the

individual client, and additional burdens placed on families, as well as, community resources. As

such, it is important to try to understand what the major influences are for recidivism rates for

outpatient treatment with children and adolescents. In a study exploring the methodological uses

of recidivism rates to assess mental health treatment, Hafemeister and Banks (1996) noted that

there are some complexities related to using recidivism as a treatment outcome measure. One

potential problem is that it is not always possible to identify the exact causal factors that may be

contributing to the recidivism rates. Additional factors should be controlled for where possible.

Another danger that they note is the use of recidivism rates from a single time point without

comparisons from other similar treatments. While there are some potential problems, the authors

noted some potential advantages of using recidivism as an outcome measure for a treatment

program. One is that recidivism rates can provide an accurate description of the performance of

a mental health program or a more general mental health system of care. Recidivism provides a

nice comparison of performance over time to similar treatment options.

Previous research exploring recidivism rates has shown lower recidivism for clients who

were seen by marriage and family therapists (13.4%) compared to nurses (14.2%), professional

counselors (14.4%), medical doctors (14.5%), social workers (15.7%), and psychologists (15.8%)

(Crane & Payne, 2011). The study also found that recidivism rates were lowest for individual

therapy (14.9%), followed by family therapy (15.4%) and a combination of individual and family

therapy sessions (17.6%). These results were comparisons of consumers of therapy in general

and are not specific to one age group or diagnosis. Additional research is needed to understand

the differences in rates specific to children and adolescents.

Dropout and recidivism are treatment outcome measures that may indicate problems in

the treatment process. Previous research literature supports the use of these variables in outcome

studies. While there are differences in how they are operationally defined, these measures can

help provide useful information for mental health treatment delivery.

Cost effectiveness research with children and adolescents. Cost effectiveness refers to

a balance in the cost and effectiveness of a treatment approach. A treatment that costs less than

another but is equally effective is considered to be more cost effective. Similarly, a treatment

approach that costs more but also has a higher level of effectiveness may be considered more

cost effective even though it is more expensive. Finding treatments that are cost effective is

important because they offer the best balance between conserving valuable, limited resources,

alleviating suffering, and providing effective treatment.

It has been noted that cost effectiveness is an ethical concern (Blount, 1987). Often,

more effective treatments have higher delivery costs associated with them. However, decisions

made regarding the treatment of children are usually made in settings where there are limited

resources that must be spread out among many who are in need of services, rather than being

concentrated on a single individual. Thus, it is not plausible that every child will receive the

most efficacious treatment because it would be too taxing on available resources. Also,

practitioners do not necessarily use a treatment just on the basis that it has been found to be

efficacious. Additionally, it is not reasonable to simply treat clients with the least expensive

treatment available because it is available, especially if the treatment is not effective or

efficacious. There are also limits to the generalizability of efficaciousness to non-controlled, real

world settings. The ethical balance involves providing treatment options that provide positive

outcomes and have reasonable costs.

Previous cost effectiveness research with mental health care usage for children suggests

that there are multiple factors that need to be considered. A study exploring cost effective

treatment options for children with ADHD suggests that high costs do not necessarily rule out

cost effectiveness (Foster et al., 2007). Some treatments that have higher costs also have more

long term benefits. For example, including behavioral therapy in the treatment of ADHD in

children may help avoid future costs associated with the disorder such as juvenile justice. The

researchers also suggest that treatments may be more likely to be cost effective when they are

carefully targeted to the individual situation of the child being treated rather than having a

blanket treatment that is cost effective for all situations. These findings are supported by

additional research that suggests that treating depression in children utilizing group cognitive

behavioral therapy (CBT) is more cost effective than individual CBT but less cost effective for

treating drug and alcohol dependence, anxiety, and social phobias (Tucker & Oei, 2007). Other

research suggests that community based multi-systemic therapy is more effective and less

expensive than hospital treatment for children (Sheidow et al., 2004). A study exploring the costs

of including family therapy to the treatment program for youth from a low SES, who struggle

with conduct disorder, found that costs did not increase (Crane, Hillin, & Jakubowski, 2005).

The researchers also found that youth who utilized in-home family therapy utilized fewer

medical services. The results suggest a possible advantage of family based treatment for

lowering overall health care and mental health care costs. Additionally, a study comparing CBT

to family therapy for the treatment of adolescents with eating disorders found a slight advantage

in the immediate effectiveness for CBT (Schmidt et al., 2007). Though no significant

differences were found in terms of cost, the authors concluded that CBT is, because of the

immediate effectiveness advantage, more cost effective than family therapy for treating eating

disorders.

A study exploring cost effectiveness on mental health disorders in general found

significant differences by provider license type and therapy modality (Crane & Payne, 2011).

The results showed that professional counselors were the most cost effective, followed by

marriage and family therapists and medical doctors, then social workers, nurses, and

psychologists. Additionally, in terms of therapy modality, family therapy was found to be the

most cost effective, followed by individual therapy, with a mixed mode being the least cost

effective. Given such evidence, it is important to explore additional treatment provider

characteristics and approaches to treatment that may be more or less cost effective for children

and adolescents.

In a recent review of cost effectiveness research published between 2002 and 2009, it was

noted that the majority of studies focus on treatment for ADHD, conduct disorder, and eating

disorders (Kilian, Losert, Park, McDaid, & Knapp, 2010). Other significant disorders, such as

depression or anxiety, were covered by only a single article. In another recent review of cost

effectiveness for family based treatments for substance abuse it was noted that while many

family based treatments are clinically effective, and while some have been shown to be more

cost effective than other individual based treatments, additional work may be needed to improve

their cost effectiveness, such as distinguishing the weight of high effectiveness versus low cost,

consistency in outcome measures used, and emphasizing systemic influences and costs (Morgan

& Crane, 2010). Thus, there is a need for additional research that explores outcomes, such as

cost effectiveness, for various diagnoses and treatment modalities to help fill this significant gap.

In summary, there are important differences in the treatment of mental health issues

between children and adults. Treatment dropout, recidivism rates, and cost effectiveness are

variables that are often used to study mental health treatment outcome. Additionally, provider

license type and therapy modality may influence treatment outcomes. There is a need for

continued research on treatment outcomes for children with mental health issues. The current

study explores some of these important issues.

The current project. Recent research has noted that evidence on economic issues for

children’s mental health care is often scant and difficult to find (Stevens et al., 2010). In addition,

it can be difficult to interpret the results of studies that have been done. Weisz, Huey, and

Weersing (1998) note some significant limitations of much of the child psychotherapy outcome

research, including non-representative samples and treatment conditions, homogeneous samples,

the fact that participants were often actively recruited for treatment and were not actually

unsolicited clinical cases, and exclusive adherence to a specific treatment technique on the part

of the therapist. The design of the current research project addresses some of these suggested

limitations. For example, the sample is not homogeneous; it includes a wide range of presenting

problems. The sample is sufficiently large and regionally diverse to be considered a

representative sample. Participants in the current study were not recruited based on specific

inclusion criteria; they were all clinical cases form a national health insurer over a six year period.

The mental health professionals in the current study did not adhere to a specific, structured

treatment technique. Thus, the current study provides a high level of external validity to add

effectiveness evidence to existing research.

Previous studies have shown a wide range of dropout rates for youth and adolescents with

rates ranging between 28% and 75% (Armbruster & Schwab-Stone, 1994; Kazdin, Mazurick, &

Siegel, 1994; Lai, Pang, Wong, Lum, & Lo, 1998, Sirles, 1990). These studies may have such

differing dropout rates because of their limited sample demographics. The studies utilized

convenience samples at local outpatient clinics. Also, these studies had dropout data from

significantly fewer participants, ranging between 235 and 555, compared to the 200,210

participants included in the current study. The current study will explore overall dropout rates

for children and youth and compare them to previous results for their peers as well as for adult

populations. The current study will also explore differences in dropout rates between license type

of the practitioner, as well as, the therapy modality used in treatment. Exploring differences

between license types is intended as a means for discovering potential advantages from the

various profession backgrounds rather than to show superiority of professional license type. The

differences in therapy modality are to help provide information about the advantages inherent in

treating children individually or with other family members.

Previous research on child and adolescent mental health dropout has also noted that

dropout rates differ by diagnosis (Johnson, Mellor, & Brann, 2006). Children with family

problems, ADHD, and conduct disorder have higher dropout rates. The current study will

compare child and adolescent dropout by diagnosis to provide additional information in this area

of study. It is predicted that dropout rates will differ by diagnosis and that those diagnoses that

tend to be more taxing on the family will add to the barriers to therapy and thus have higher

dropout rates. As discussed, children and youth face additional barriers to therapy attendance

(Bannon & McKay, 2005; Nock & Kasdin, 2001). As a result, it is predicted that dropout rates

for children and youth will be higher than those that have been found for adults. Previous studies

have not looked specifically at mental health treatment dropout rates for children and adolescents

by therapist license type. The current study will explore these differences. It is hypothesized

that differences will exist for child and adolescent dropout rates when comparing therapist

license type.

Research Questions

The purpose of the current study is to determine the relationship between different types

of professions who provide mental health care services to children in the form of individual and

family therapy. The following research questions will be addressed:

Question 1. What are the differences in outcomes (dropout, recidivism, number of

sessions, cost, cost per session, and cost effectiveness) when comparing license type (medical

doctor, master’s nurses, psychologists, master’s social workers, marriage and family therapists,

and professional counselors)?

Question 2. What are the differences in outcomes when comparing therapy modality

(individual, family, and mixed mode)?

Question 3. What are the differences in outcomes when comparing diagnosis?

Question 4. Is there a relationship between the diagnoses and license type? In other

words, do different professions tend to treat certain diagnoses for children and adolescents more

frequently than other professions?

Question 5. Is there an interaction between diagnosis and license type when considering

therapy outcomes? Because of differences in training, some practitioners may be better prepared

to deal with certain diagnoses or provide family versus individual therapy.

Question 6. What are the differences in outcomes when comparing age and gender for the

treatment of children and adolescents?

Method

Design

The current study is a retrospective design utilizing administrative data from Cigna a

leading health care insurance provider in the United States. Cigna manages hundreds of health

care plans, serving several million patients. Data from six years (2001-2006) of psychotherapy

claims were included in the current study. An entry in the data set represented a single claim by

a mental health care provider. Each entry included the following information: a unique client

identification number, client age, client sex, treatment date, state where service was provided,

current procedural terminology (CPT) code, primary DSM IV diagnosis, therapist license type,

highest degree held by therapist, dollar amount of claim, and number of therapeutic sessions per

claim.

Use of retrospective administrative data for the purposes of providing information for

planning purposes, compiling aggregate statistics, and monitoring trends in the data is allowed

by the Health Insurance Portability and Accountability act of 1996 (HIPPA). Prior to the data

being delivered from Cigna, all patient and provider identification information was removed and

a unique client identification number was assigned for each patient. It was not possible at any

time to identify any subscriber or provider information from the data provided. Research on

retrospective administrative data falls under the exempt status for the Institutional Review Board

(IRB). In compliance with IRB guidelines, exempt status for the current study was confirmed

prior to its completion.

Sample

Participants. Participant data from 200,210 children ages 3 – 18 (M = 12.1, SD = 3.9)

were used in the current study. A power analysis using G*Power 3.1.2 (Faul, Erdfelder, Lang &

Buchner, 2007) revealed that the sample was sufficiently large to find an effect size for both a

chi square analysis and analysis of variance. Because of the difficulty in establishing validity in

diagnosing disorders in infants and young children (Scheeringa, Peebles, Cook, & Zeanah, 2001),

participants under the age of 3 years were excluded from the current study. Participants included

106,374 boys (53.2%) and 93,753 girls (46.8%). Data from all 50 U.S. states were included in

the study.

Procedure

Data cleaning. The raw data set provided by Cigna included all outpatient psychotherapy

claims from 2001 – 2006. The data was cleaned to exclude claims that contained multiple

sessions per claim, claims that reported a refund to Cigna, and claims that were unpaid.

The original data set included 93 different therapeutic licenses. In order to compare

outcomes by license type, professional licenses were sorted into groups. Professions that are not

nationally recognized as independently licensed mental health care providers were excluded

from the data set. Claims that had an unknown license type or that had a license listed as a

general mental health practitioner were also exclude from the final data set. To avoid potential

confounds, providers with multiple license were excluded from the data set. Additionally, clients

who saw therapists of more than one profession type were excluded from the analyses.

Practitioners were grouped into six profession types: medical doctor, master’s nurses,

psychologists, master’s social workers, marriage and family therapists, and professional

counselors. For complete data cleaning procedures see Crane and Payne (2011).

Definitions

Episode of Care (EoC). An EoC is defined by Cigna as a continuous series of services

for the same patient. An EoC begins with the first psychotherapy service and ends after the

patient has had no psychotherapy claims for 90 days. The number of sessions in the first EoC in

the data set ranged from 1 to 326 (M = 5.5, SD = 7.7). Over 84% of the patients completed

therapy within the first EoC, resulting in no recidivism.

Therapy modality. Therapy modality refers to who is included in the treatment process.

The current project includes three modalities: individual, family, and mixed. Individual therapy

is psychotherapy with a single identified patient in the therapy room with the therapist. Family

therapy refers to the inclusion of at least one additional person in the therapy room with the

identified patient and therapist. In the case of childhood issues, this is typically a family member

or legal guardian. Mixed therapy refers to a treatment that includes both individual and family

sessions during a single EoC. The claims for psychotherapy services were classified by providers

using the Current Procedural Terminology (CPT) codes of individual psychotherapy therapy

(90806) or family psychotherapy therapy (90847; American Medical Association, 2006).

Diagnoses. Diagnoses were assigned by providers using criteria from the Diagnostic and

Statistical Manual of Mental Disorders-IV-TR (DSM-IV-American Psychiatric Association,

2000). Diagnoses were sorted into ten categories: adjustment disorders (n = 22,725), anxiety and

PTSD (n = 21,260), disruptive behavior (n = 13,586), dissociative disorders (n = 121), eating

disorders (n = 968), mood disorders (41,832), relational problems (n = 772),

schizophrenic/psychotic (n = 25), substance use and abuse (n = 2166), and other diagnoses (n =

96,755).

Cost. This is the amount in dollars paid by Cigna for therapeutic services. Because

mental health services have great variability in the number of sessions provided for treatment,

cost per session and total cost for treatment are distinguished. Cost per session was calculated as

total cost divided by the total number of sessions in that EoC.

Cost effectiveness. This is an estimation of what psychotherapy costs per patient taking

into account the relative success rates associated with each discipline. Cost effectiveness is

computed as: Estimated cost effectiveness = EoC cost + (Number of sessions in the EoC *

Recidivism rate). The cost effectiveness equation takes into consideration the average number of

sessions in the EoC, the cost of providing the EoC, and the outcome of care in the EoC. This

method of evaluating cost effectiveness has been used in previous research (Crane & Payne,

2011; Moore, Hamilton, Crane, & Fawcett, 2011)

Dropout. For the current study, dropout was operationally defined as not returning for

additional treatment sessions following a single treatment session. Because mixed mode therapy

is defined as attending at least one session of individual therapy and at least one session of

family therapy, dropout for mixed mode therapy was calculated as not returning to treatment

after attending only one session of individual and one session of family therapy. When making

comparisons for dropout rates between therapy modality (individual, family, and mixed), dropout

was calculated as not returning following two sessions.

Types of treatment services. Outpatient therapy is the most commonly utilized mental

health service for children (Kutash & Rivera, 1996). Outpatient therapy is typically conducted

by psychologists, psychiatrists, social workers, counselors, and family therapists and is

performed in a variety of settings including: community mental health centers, private clinics,

and hospitals. A clear advantage to outpatient therapy is that it allows the child to remain in her

home, school, and community while receiving mental health services. This also allows the child

to remain with familial and social support in familiar surroundings. Other treatment options that

are used with children and adolescents include: inpatient/residential treatment, day treatment

facilities, and home based services. Though there is a wealth of research on these additional

treatment options, the current study consists of data only for outpatient therapy.

Recidivism. In medical treatment, recidivism is often referred to as a recurrence of a

disease or relapse to a previous mode of behavior (Mackie et al., 2001; Whitson, Heflin, &

Burchett, 2006). In the current study, recidivism is operationalized as a participant returning to

therapy for additional EoC(s) with the same type of provider (Crane & Payne, 2011; Fawcett &

Crane, In press).

Number of sessions. This variable is defined as the total number of sessions for a patient

during a single EoC.

Analysis. For the independent variables professional license, therapy modality, age, and

diagnosis, the continuous ratio data, including cost, number of sessions, and cost effectiveness,

were analyzed using analysis of variance (ANOVA), and analysis of covariance (ANCOVA). A

two sample t-test was used to analyze the differences by gender. Differences between the

dichotomous dependent variables, including dropout and recidivism, were analyzed using chi-

square test for independence. Analyses involving cost and treatment length were conducted with

dropout cases excluded.

Results

The first research question is what are the differences in outcomes when comparing

license type? Analysis of the first research question showed a statistically significant difference

in percent dropout by professional license type 2 (5, 200210) = 990.3, p < .001. Medical

doctors showed the highest percent dropout, followed by psychologists, nurses, professional

counselors, marriage and family therapists, and social workers. Results can be seen in table 1.

Analysis indicated that all cells in the chi square analysis met the minimum expected count.

Analysis also showed a statistically significant difference in recidivism rates by professional

license type 2 (5, 200210) = 95.6, p < .001. Social workers had the highest percent recidivism,

followed by medical doctors, nurses, psychologists, professional counselors, and marriage and

family therapists. Percentages can be found in Table 1. Results again indicated that all cells in

the chi square analysis met the minimum expected count. Analysis also showed a statistically

significant difference in number of sessions by professional license type F (5, 166273) = 20.3, p

< .001. Tukey post-hoc analysis indicated that, with dropouts excluded, counselors had the

lowest average number of sessions, followed by marriage and family therapists, psychologists,

medical doctors, nurses, and social workers. Post-hoc results can be found in table 3. Analysis

revealed that some of the data was positively skewed, which violates the assumption of

normality for an analysis of variance. The positively skewed variables were log transformed

prior to the statistics being calculated. While the log transformed data were used for the analysis,

both the raw and log transformed data are included in the tables. Also, as practitioners with a

doctorate degree tend to be reimbursed at a higher rate than practitioners with a master’s degree,

the highest earned degree for the practitioner was used as a control variable. Analysis of

covariance also showed a statistically significant difference in treatment cost per session F (5,

166273) = 6155.1, p < .001. Tukey post-hoc analysis showed that professional counselors (M =

$44.65), and social workers (M = $44.78) had statistically lower cost per session (M diff = -.13,

std. error = .09, p = .70), followed by marriage and family therapists (M = $45.32) then nurses

(M = $52.62), psychologists (M = $54.81), and MDs (M = $72.68) cost the most per session.

Complete post-hoc results can be found in table 3. Results from the ANCOVA indicated a

statistically significant difference in total cost of treatment by license type F (6, 166272) = 408.4,

p < .001. Pairwise comparisons showed that counselors (M = 374.73) and marriage and family

therapists (M = 384.15) had the lowest total costs (M diff = -9.42, p = .07), followed by social

workers (M = 405.50), then psychologists (M = 446.50) and nurses (M = 454.01; M diff = 7.51, p

= .62), and finally medical doctors (M = 571.54) had the highest total cost. Results also

indicated a difference in cost effectiveness by profession F (5, 166273) = 247.6, p < .001, with

professional counselors (M = 386.13), marriage and family therapists (M = 395.68; M diff = -9.54,

p = .07) being the most cost effective, followed by social workers (M = 406.13), then nurses (M

= 466.78) and psychologists (M = 458.56; M diff = 7.62, p = .63), and MDs (M = 583.91).

Complete analysis results can be found in table 2.

Table 1 about here

Table 2 about here

Table 3 about here

The second research question is what are the differences in outcomes when comparing

therapy modality? Analysis of the second research question showed a statistically significant

difference in percent dropout by therapy modality 2 (2, 20210) = 19253.6, p < .001. Results of

the second research question can be found in table 1. Family therapy showed the highest percent

dropout, followed by individual therapy, and mixed mode therapy. Results indicated that all cells

in the chi square analysis met the minimum expected count. Analysis also showed a statistically

significant difference in recidivism rates by therapy modality 2 (2, 200210) = 14550.7, p < .001.

Mixed mode therapy had the highest percent recidivism, followed by individual therapy, and

family therapy. Results again indicated that all cells in the chi square analysis met the minimum

expected count. Analysis also showed a statistically significant difference in number of sessions

by therapy modality F (2, 166276) = 5518.1, p < .001. Tukey post-hoc analysis showed that

family therapy (M = 5.61) had the lowest average number of sessions, which was significantly

less than the average number of sessions for individual therapy (M = 7.31; M diff. = 1.70, std.

error = .06, p < .001)., which was significantly less than the averages for mixed mode (M =

11.70; M diff. = -6.09, std. error = .07, p < .001). Analysis of variance showed a statistically

significant difference in total cost of treatment by therapy modality F (2, 166276) = 4366.8, p

< .001. Post hoc analysis showed that, on average, family therapy (M = $281.30) cost less than

individual therapy (M = $361.68; M diff. = 80.37, std. error =3.59, p < .001), which cost less

than mixed mode therapy (M = $584.56; M diff. = 303.26, std. error = 3.91, p < .001). Analysis

of cost per session showed almost no difference between individual therapy (M = $48.64), mixed

mode (M = $49.16), and family therapy (M = $49.78). Results also indicated a difference in cost

effectiveness by profession F (2, 166276) = 4449.5, p < .001, with Tukey post-hoc analysis

showing that family therapy (M = 287.96) was the most cost effective, followed by individual

therapy (M = 371.66; M diff. = 83.71, std. error = .06, p < .001), and mixed mode being the least

cost effective (M = 603.16; M diff. = -231.50, std. error = 2.79, p < .001). Complete analysis

results can be found in table 2.

The third research question is what are the differences in outcomes when comparing

diagnosis? Analysis of the third research question showed a statistically significant difference in

percent dropout by diagnosis group 2 (9, 200210) = 684.3, p < .001. Results of the third

research question can be found in table 4. Relational diagnoses showed the highest percentage of

dropouts, followed by substance use and abuse, other diagnoses, adjustment disorders, disruptive

behavior, schizophrenic/psychotic, dissociative disorders, anxiety disorders, mood disorders, and

eating disorders with the lowest percent dropout. Results indicated that all cells in the chi square

analysis met the minimum expected count. Analysis also showed a statistically significant

difference in recidivism rates by therapy modality 2 (9, 200210) = 350.4, p < .001.

Schizophrenic/psychotic had the highest percent recidivism, followed by eating disorders,

dissociative disorders, mood disorders, anxiety and PTSD, other, disruptive behavior, adjustment

disorders, substance use and abuse and relational problems. Results again indicated that all cells

in the chi square analysis met the minimum expected count. Analysis also showed a statistically

significant difference in number of sessions by diagnosis F (9, 166269) = 133.1, p < .001. Tukey

post hoc findings showed that relational diagnoses had the fewest number of sessions, followed

by substance abuse, adjustment disorders, other, disruptive disorders, dissociative disorders,

anxiety disorders, mood disorders, eating disorders, and schizophrenia. Post hoc analyses for the

third research question can be found in table 5. Analysis of variance showed a statistically

significant difference in total cost of treatment by diagnosis F (9, 166269) = 115.3, p < .001. Post

hoc analysis showed that relational diagnoses had the least total cost, followed by substance

abuse, adjustment disorders, other, disruptive disorders, dissociative disorders, anxiety disorders,

mood disorders, eating disorders, and schizophrenia. Analysis of variance also showed a

difference in treatment cost per session by therapy modality F (9, 166269) = 11.5, p < .001.

Tukey post hoc results showed that, on average, relational diagnoses cost less per session

followed by adjustment disorders, schizophrenia, mood disorders, other, disruptive disorders,

anxiety disorders, substance abuse, eating disorders, and dissociative disorders. Results also

indicated a difference in cost effectiveness by diagnosis F (9, 166269) = 115.6, p < .001, with

Tukey post hoc analysis showing relational problems as being significantly more cost effective

to treat than substance abuse, adjustment disorders, other, disruptive disorders, and dissociative,

which were significantly more cost effective than anxiety disorders, and mood disorders. Eating

disorders and schizophrenia were the least cost effective.

Table 4 about here

Table 5 about here

The fourth research question was related to the relationship between the diagnoses and

license type. Analysis of the relationship between diagnosis and license type suggests that certain

professions tend to treat certain diagnoses more often than others 2 (45, 200210) = 1874.5, p

< .001. Table 6 provides a summary of the proportions. On average, medical doctors and nurses

treat a lower percentage of adjustment disorders than other license types, nurses treat a higher

percentage of eating disorders, psychologists treat a lower percentage of mood disorder cases,

marriage and family therapists treat a higher percentage of relational disorders, and social

workers treat a higher percentage of substace abuse cases. Results can be found in table 6.

Table 6 about here

The fifth research question was related to the possible interaction between diagnosis and

license type when considering therapy outcomes. Results suggest an interaction between

diagnosis and license type when considering therapy outcomes. There were statistically

significant differences in dropout and recidivism percentages, total sessions (F (42, 166279) =

1.87, p < .05) and cost effectiveness (F (42, 166279) = 1.85, p < .05). For example, doctors and

nurses had significantly lower dropout rates when treating disruptive behavior disorders

compared to treating anxiety, adjustment, mood, and substance use/abuse disorders. Also,

marriage and family therapists have a lower percentage of recidivism when treating substance

abuse disorders compared to other diagnoses and other professions. The complete analysis

results for diagnosis by license type can be found in table 7.

Table 7 about here

When analyzing outcomes by license type and family therapy modality, marriage

therapists show a significantly significant lower percentage of recidivism compared to other

license types 2 (5, 16835) = 51.3, p < .001. Nurses, counselors, marriage and family therapists,

and social workers showed lower dropout rates than doctors and psychologists 2 (5, 16835) =

404.6, p < .001. Results indicate a statistically significant difference in cost effectiveness for

family therapy by license type F (5, 16829) = 91.9, p < .001. Tukey post hoc analysis suggest

that social workers (M = 310.4) are most cost effective when utilizing a family therapy modality

followed by marriage therapists (M = 322.8; std. error = 4.98, p = .02) together with counselors

(M = 321.1; std. error = 5.3, p = .17), followed by nurses (M = 364.8; std. error = 15.6, p < .01)

and psychologists (M = 394.0; std. error = 15.10, p = .31), and medical doctors(M = $621.5; std.

error = 10.58, p < .01). There were statistically significant differences in total cost, F (5, 16829)

= 95.2, p < .001. Tukey post hoc comparisons indicated that social workers (M = $302.3) had the

lowest total cost followed by counselors (M = $313.0; std. error = 4.84, p = .02) together with

marriage therapists (M = $314.8; std. error = 5.11, p = .154), followed by nurses (M = $356.7;

std. error = 14.68, p < .01), psychologists (M = $385.8; std. error = 14.66, p < .01), and medical

doctors had the highest mean total cost (M = $611.6; std. error = 10.27, p < .01). There were

also statistically significant differences when analyzing cost per session F (5, 16829) = 993.6, p

< .001. Post hoc analyses indicate that when conducting family therapy, counselors (M = $45.5)

and social workers (M = $44.2; std. error = .09, p = .70) had the lowest cost per session,

followed by marriage and family therapists (M = $45.8; std. error = .13, p < .01), nurses (M =

$54.0; std. error =.40, p < .01), psychologists (M = $55.7; std. error = .39, p < .01), and medical

doctors (M = $79.3; std. error = .27, p < .01). There was not a statistically significant difference

in total number of sessions, F (5, 16829) = 1.3, p = .24. A summary of the results can be found in

table 8.

Table 8 about here

The sixth research question was related to differences in outcomes when comparing age

or gender for the treatment of children and adolescents. Comparisons show a statistically

significant difference in the percentage of dropout by age 2 (15, 200210) = 535.4, p < .001.

Dropout percentages tend to be higher for young children and as the age increases, the percent

dropout tends to decrease, until ages 17 and 18 when the dropout percentage rises. A summary of

the dropout percentages by age can be found in table 9.

There was also a signigicant difference in recidivism percentage by age. Recidivism

tends to be lower for the young children; it rises during the pre-teen and early teenage years, and

then falls again for older teenage children. A summary of the recidivism percentages by age can

be found in table 8. There are also significant differences in total cost (F (15, 166263) = 19.8, p

< .001) and total sessions by age of child F (9, 166263) = 25.8, p < .001. Post hoc analyses

revealed that the youngest and oldest children have the least total treatment costs while children

ages 9 to 15 have the highest. Analysis also showed a similar trend for treatment length.

Summary data for cost of treatment and treatment length can be found in table 9.

Table 9 about here

Analysis showed a statistically significant difference in the percent dropout by gender 2

(1, 200210) = 67.1, p < .001. Results indicate that males (17.6%) are slightly more likely to drop

out of therapy than females (16.2%). Results did not indicate a significant difference in

recidivism rates between males (22.5%) and females (23.1%), 2 (15, 200210) = 11.9, p > .05.

There were also significant differences in total cost (t (166210) = -6.1, p < .001) and total

sessions by gender of child t (166210) = -11.5, p < .001. Analyses revealed that treatment

of.males (M = 408.4, SD = 492.8) cost less than females (M = 423.2, SD = 505.0) and that males

attended fewer sessions (M = 8.1, SD = 8.6) than females (M = 8.6, SD = 9.2).

Discussion

The first research question explored the differences in outcomes by practitioner license

type. The results suggest that medical doctors tend to have about twice as many dropouts when

working with children and adolesents when compared to the other license types. The results also

show social workers to have the lowest dropout rates followed closely by marriage and family

therapists and professional counselors. One reason that medical doctors may show a higher

dropout rate could be related to their tendency to approach tretment from a medical model which

views mental illness as a disturbance of the brain or central nervous system (McCulloch, Ryrie,

Williamson, & St. John, 2005). The medical model tends to focus on identifying the diagnosis

and then prescribing a treatment to fix the problem (Beecher, 2009). This approach to treatment

is likely to result in fewer overall sessions and more cases where the client only comes in for a

single consultation.

While there was a statistically significant difference in recidivism percentages by license

type, with marriage and family therapists showing the lowest recidicism rates, there is only about

a three percent difference in recidivism across license type. Also, the results on recidivism

suggest that nearly one in four children return for additional treatment following a first episode

of care. This suggests that nearly a quarter of children who are treated for mental health issues

do not recieve sufficient treatment and must return for additional treatment at a later date. The

recidivsm percentages in the current study are based on returning to treatment after 90 days at an

outpatient treatment facility, yet they are similar to readmission rates that have been reported

from inpatient treatment centers following the same amount of time (Romansky, Lyons, Lehner,

& West, 2003). Romansky et al. report that 21% of adolescents return for additional treatment

within 3 months of discharge. It is interesting to note that the recidivism percentages in the

current study suggest that children are almost twice as likely to recidivate when compared to the

general population as a whole (see Crane & Payne, 2011). Recidivism percentages by license

type in the current study ranged from 20.5 to 23.8, where Crane and Payne report recidivism

percentages by license type in the general population to be between 13.4 and 15.8. Results on

recidivsm may suggest that treatment of children and adolescents is inherently challenging and

that many children who present to therapy may not achieve desired treatment outcomes during a

single episode of care.

Unfortunately, the current study is unable to assess the reasons for which participants did

or did not return for additional treatment. There are likely multiple influencing factors that bring

children back to therapy. One study explored multiple predictive factors for children returning to

therapy including: age, history of substane dependence, personality traits, family history of

mental illness, history of abuse or neglect, history of sexual abuse, history of self harm, and

accomodation at discharge (Barker, Jairam, Rocca, Goddard, & Matthey, 2010). They found that

none of these factors significantly influenced the likelihood of a child to return to treatment.

Given such evidence, and given the consistent readmission rates for children, which tend to be

higher than those for adults, it is possible that the child is brought back to therapy because of a

characteristic of the caregiver, rather than because of the child’s desire to return. As noted

earlier, Weisz et al. (1998) have stated that children rarely see themselves as needing therapy and

are almost always referred to therapy by an adult. It is likely that there are similar trends when

returning for additional treatment.

Other results in the current study suggest that children and adolesents attend about eight

sessions of therapy on average. These numbers are similar across professional license type.

These results are slightly higher than the average number of sessions for psychotherapy for the

general population (M = 6.95; Crane & Payne, 2011). While children and adolescents appear to

stay in treatment a little longer than adults, outpatient psychotherapy is still relatively brief.

The results on cost of treatment indicate that there are differences by professional license.

The results suggest that professional counselors and marriage and family therapists have the

lowest total treatment costs and that medical doctors have the highest treatment costs when

working with children and adolescents. The analysis of cost per session shows that professional

counselors, social workers, and marriage and family therapists cost significanly less than nurses

and psychologists, who cost less than medical doctors. Analysis of the cost variables indicated

that there were differences by highest degree held. Specifically, those practitioners with a

doctorate degree cost more than those with a master’s degree. Even when degree is used as a

control variable, there are statistically significant difference in total cost of treatment by license

type. On average, professional counselors, marriage and family therapists, and social workers

are more cost effective when treating children than psycholgists, nurses, and medical doctors.

This suggests that treatment with these practitioners costs less when considering the total length

of treatment and the likelihood of returning for additional episodes of care.

The second research question explored differences in treatment outcomes by therapy

modality. There were significant differences in the percentage of dropouts by therapy modality.

Nearly 50% of children who were treated using a family therapy approach dropped out of

treatment, compared to 36% of children who were treated from an individual approach. Those

children treated using a combination of individual sessions and family therapy sessions showed a

very low percentage of dropouts. Less than seven percent of participants dropped out of

treatment when the therapist utilized a mixed mode approach to therapy.

Previous research indicates that there are multiple barriers for children recieving therapy

(Garcia & Weisz, 2002). These include family issues such as: transportation problems, sick

family members, finances, scheduling time, and effort required to get to therapy. Perception

about the need for therapy by different family members also influenced families continuing

treatment. It may be that these family based barriers make continuing a pure family therapy

approach more challenging. There may be a nearly 50% dropout rate for children in family

therapy treatment because of the increased complexity of getting the family to therapy for each

session. Utilizing a mixed modality approach may provide additional flexibility for treatment.

Sometimes the child meets with the therapist with the entire family and sometimes she meets

with the therapist individually. Results in the current study suggest that offering a mixed mode

option for families seeking treatment for children may help increase retention and possibly

completion of treatment. Mixed mode therapy is also associated with greater overall treatment

costs, however, these costs are mostly associated with the number of sessions. There may be

additional benefits for clients staying in treatment longer and recieving an adequate dose of

therapy (Baldwin, Berkeljon, Atkins, Olsen, Neilson, 2009).

Mixed mode therapy also had the highest average number of sessions. Children stayed

in treatment twice as long when utilizing a mixed mode approach when compared to family

therapy alone. They stayed in treatment 40% longer with a mixed mode approach compared to

individual therapy alone. There is research which suggests that as the number of sessions

increases, there is a measurable decrease in negative behaviors (Cotton-Cornelius, 2004).

Utilizing a mixed mode therapy may help increase the total number of sessions that children

attend, which may in turn help decrease negative behaviors.

The analysis also revealed a significant difference in total cost by therapy modality. It

was not surprising that the therapy modality that had the highest average number of sessions was

also the one with the highest average cost. When the number of sessions is considered in the

analysis of cost, the differences between modality fall to within two dollars per session. Results

thus suggest that family therapy, on average, costs less than individual or mixed mode therapy

because clients tend to use fewer therapy sessions. Due to the limitations in the current data set,

it was not possible to assess why family therapy was more brief. As mentioned, it is possible

that it averaged fewer sessions because of the difficulty of getting the family to therapy at the

same time. It is also possible that in a family setting they were able to make quicker progress in

therapy and that the family required fewer sessions to alieviate the presenting problem. Future

research would benefit by comparing client satisfaction with therapy and symptom reduction

when comparing across therapy modality.

Family therapy was shown to be the most cost effective of the three therapy modalities.

When therapists utilize a family therapy modality when working with children, they may be

maximizing impact given the resources. The results of the current study support previous

findings that suggest that systemic therapy is more cost effective than individual therapy when

treating children (Tucker & Oei, 2007). Results of the current study also support previous

research that suggests a difference in cost effectiveness by therapy modality for participants of

all ages and not just for children. Previous research has proposed that utilizing CBT may be more

cost effective than family therapy when treating children (Schmidt et al., 2007). While the results

of the current study cannot address any specific model of therapy, it does suggest that family

based interventions are more cost effective than individual approaches when applied by all

license types.

The results from the third research question suggest that dropout may be influenced by

the diagnosis type. Specifically, those clients who had a relational diagnosis were nearly 30%

more likely to drop out of therapy than clients who were being treated for substance abuse,

which had the next highest percent dropout. Mood disorders and eating disorders showed the

lowest percentage of dropouts for children. The results for recidivism by diagnosis showed an

interesting trend, especially when compared to dropout percentages. Those diagnoses that tend

to have higher percentages of dropouts also tend to have lower recidivism rates. Also, the trend

for total number of sessions by diagnosis matched the trend for dropouts almost exactly. Taken

together, these outcomes could suggest a scale for difficulty of treatment. For example, those

diagnoses that have high dropouts, low recidivism, and fewer average sessions may be

considered less severe or less difficult\complex to treat. Conversely, diagnoses that are more

severe may have lower percentages of dropouts due to the motivation to stay in treatment

because of the severity of the symptoms. These diagnoses are also more likely to experience

recidivism when the client system experiences a relapse in symptom behaviors. Further, difficult

issues are likely to need more total sessions to treat successfully. Results of the current study

may suggest the following order for severity\difficulty of treatment, for diagnoses for children

and adolescents, ranging from more mild to more severe: relational diagnoses, substance use and

abuse, adjustment disorders, other disorders, disruptive behaviors, dissociative disorders, mood

disorders, anxiety disorders, schizophrenia, and eating disorders.

The current study also suggests that professionals with certain license types tend to treat

certain diagnoses more often than others. The percentage of total cases seen by each license type

is similar across most diagnoses. Some notable exceptions include adjustment disorders, mood

disorders, relational disorders, and substance abuse. For example, of all the cases treated by

medical doctors, only 4.6 % were adjustment disorders. The other license types treated

adjustment disorders for 8.2 % to 12.1% of their total cases, with counselors treating the highest

percentage of adjustment cases. These results may indicate that traditional mental health

professionals are more likely to treat adjustment disorders than are traditional medical health

professionals. The results may also indicate that medical doctors and nurses may be less likely

to give an adjustment disorder diagnosis when compared to other mental health professionals. It

is possible that there is a bias within some professions against assigning certain diagnoses or

perhaps less attention may be given to some diagnoses during training. Additionally,

psychologists tend to treat\diagnose mood disorders less often than practitioners with other

license types, while marriage and family therapists tend to treat a larger percentage of relational

diagnoses. The largest discrepancy of percentage of cases seen by diagnosis was substance

abuse. Social workers are more than three times more likely to treat\diagnose substance abuse

than other professions. Nearly 13% of the cases seen by social workers were substance

use\abuse disorders, while only 0.4% of cases seen by nurses were diagnosed with substance

abuse. It is possible that such a trend may be influenced by employment setting. It may be that

social workers are more likely to work in settings that recieve adolesecnt clients seeking

treatment for substance abuse disorders.

There is also an interesting trend in dropout percentages for children by age. Younger

children tend to have the highest percentage of dropouts from therapy. As the age of the child

goes up, the dropout percentage goes down, until late adolescence (ages 17 and 18) when it again

goes up. The recidivism percentages also show that younger and older children are less likely to

return for additional episodes of treatment. It may be that there are more barriers to treatment for

families with young children. Kazdin et al. (1993) found evidence that children with younger

mothers are more likely to drop out of treatment. Also, Allgood and Crane (1991) found that if a

family had more children they were less likely to drop out of treatment. Evidence from the

current study suggests that the age of the child who is being treated is a predictor of dropout.

The evidence also suggests that age is a predictor of treatment length. Children ages

seven to sixteen, on average, tend to stay in treatment longer than younger and older children.

These results are similar to recent research that indicates that 12 to 15 year olds are 90 % more

likely to use mental health services compared to 8 to 11 year olds (Merikangas, He, Brody,

Fisher, Bourdon, & Kortez, 2010). This evidence, together with the dropout percentages, may

suggest that there may be fewer barriers for families to get children in a middle age range to

therapy, compared to younger and older children. There may also be different expectancies from

parents when presenting to therapy with children of different ages. Nock and Kazdin (2001)

suggest that parental expectancies have an influence on therapy attendance. It may be that

parents of very young children do not expect therapy to have much of an influence. It may also

be that parents of older adolescents do not expect their child to change or put sufficient effort

into therapy. Future research could be designed to explore what may be influencing the

differences in dropouts and total number of sessions by age.

Finally, the analysis on outcomes by gender suggest that males are slightly more likely to

dropout of therapy than are females, though recidivsim percentages are not significantly different.

It also suggests that females have a slightly higher mean number of sessions. It appears that

male children utilize fewer total sessions than female children.

Clinical Implications

Results of the current study can potenailly benefit clinicians who are treating children and

adolescents. The current study suggests that children may be more likely than adults to return

for additional treatment. As children do not typically present to therapy on their own, it is likely

the parents or caregivers who are having the children return to therapy. When working with

children, it may be useful to inform parents and caregivers that about one in four children return

for additional treatment later. This may help normalize the parents experience and reduce

potential distress that the caregiver may experience as a result of returning to treatment.

The current study also found that children have much lower dropout and stay in treatment

longer when the therapist utilizes a mixed modality of therapy. Clinicians who are working with

children may consider adopting a mixed mode approach and hold both individual and family

sessions as part of treatment. Systems based approaches to therapy tend to emphasize the

inclusion of family systems in therapy. While the results of the current study suggest that family

therapy has an advantage in terms of cost effectiveness, they also suggest that conducting both

individual and conjoint sessions can help clients remain in treatment. A mixed mode approach

provides for opportunities for the clinician to work with the child’s family system, as well as,

with the child individually. For example, there may be issues about which the child or adolescent

does not feel comfortable discussing in front of other family members and a mixed mode

approach provides opportunities to have private sessions while still involving the larger family

system in treatment. Utilizing a mixed modality approach may also help families overcome some

of scheduling barriers to family based treatments. There would be less pressure on the family to

present to therapy each session; some sessions would only involve the child.

Clinicians may also benefit from the findings on dropout and recidivism by diagnosis.

When a clinician is preparing to work with a child or adolescent who has been identified as

having a diagnosis with a higher dropout percentage, the clinician may want to take additional

steps to help the child complete treatment. This would likely involve developing a strong

therapeutic relationship with the child and the child’s parents or caregiver. It may also be aided

by helping the parents or caregivers have realistic expectations for the course of therapy.

Clinicians can also help parents have realistic expectations related to the duration of therapy,

especially when working with diagnoses that tend to utilize more sessions.

There may also be a benefit to exploring what specific factors are influencing the

differences in outcomes by license type. It may be that certain training approaches have

potential advantages for the treatment of children and adolescents. Identifying what these

advantages are may lead to information that could potentially help professions better prepare

clinicians to work with children and adolescents, both during initial training and through

continuing education.

Limitations and Future Directions

The current study was a retrospective analysis of administrative data. It did not utilize an

experimental design. As such, the results should be interpreted with some degree of caution. As

there is a low level of internal validity, the evidence presented in the current study should not be

considered causal. Participants recieved treatment from various providers utilizing different

therapy modalities, but were not randomly assigned to groups. Participants were likely

selectively referred to providers. Another consideration in the interpretation of the results is that

the distribution between provider types was not uniform. There was not an equal distribution of

participants in each comparison group. Future research exploring therapy outcomes by

profession and therapy type would benefit from an experimental study with random assignment

to group.

Another limitation in the interpretation of the current study is the lack of demographic

information available from participants. Due to the nature of the agreement with Cigna,

information on race, SES, and ethnicity was not available. It was also not possible to assess for

external support systems, level of family distress, and available client resources. It was not

possible to collect assessment information on participants in the current study. We also did not

have data on therapist or client perceived experience of therapy. Future research would

greatley benefit from inclusion of assessment measures for client outcomes.

Despite its limitations, the current study provides some important evidence about the

treatment of children and adolescents. The data represents clients who are being treated in the

real world, independent of the controls of laboratory research. Thus, there is a high level of

external validity. The current study explored differences in child and adolescent therapy

outcomes by license type, therapy modality, diagnosis, age, and gender. Results indicate that

there are differences in dropout, recidivism, cost, and treatment lengths across these variables.

These results provide valuable information about mental health treatment of children and

adolescents. Specifically, utilizing a family based approach may help reduce the total length of

treatment while utilizing a mixed mode approach to therapy may help reduce the risk of dropout

from therapy. Also, some diagnoses appear to be more difficult to treat, with higher percentages

of dropout and requiring more time and money.

References

Alexander, J. F., & Sexton, T. L. (2002). Functional family therapy: A model for treating high-

risk, acting-out youth. In F. W. Kaslow (Ed.), Comprehensive handbook of

psychotherapy: Integrative/eclectic, Vol. 4 (pp. 111-132). Hoboken, NJ: John Wiley &

Sons Inc.

Allgood, S. M., & Crane, D. (1991). Predicting marital therapy dropouts. Journal of Marital and

Family Therapy, 17(1), 73-79. doi:10.1111/j.1752-0606.1991.tb00866.x

American Medical Association. (2006). Current procedural terminology. Chicago, IL: American

Medical Association.

Ani, C, & Garralda, E, (2005). Developing primary mental healthcare for children and

adolescents. Current Opinion in Psychiatry, 18(4), 440-444.

Armbruster, P., & Fallon, T. (1994). Clinical, sociodemographic, and systems risk factors for

attrition in a children's mental health clinic. American Journal of Orthopsychiatry, 64(4),

577-585.

Armbruster, P., & Kazdin, A. (1994). Attrition in child psychotherapy. Advances in Clinical

Child Psychology, 16, 81-108.

Armbruster, P., & Schwab-Stone, M. (1994). Sociodemographic characteristics of dropouts from

a child guidance clinic. Hospital & Community Psychiatry, 45(8), 804-808.

Baldwin, S. A., Berkeljon, A., Atkins, D. C., Olsen, J. A., & Nielsen, S. L. (2009). Rates of

change in naturalistic psychotherapy: Contrasting dose-effect and good-enough level

models of change. Journal of Consulting & Clinical Psychology, 77(2), 203-211.

doi:10.1037/a0015235

Bannon, W. M., & McKay, M. M. (2005). Are barriers to service and parental preference match

for service related to urban child mental health service use. Families in Society, 86(1),

30-34.

Barker, D., Jairam, R., Rocca, A., Goddard, L., & Matthey, S. (2010). Why do adolescents return

to an acute psychiatric unit. Australasian Psychiatry, 18(6), 551-555.

doi:10.3109/10398562.2010.501380

Becker-Weidman, A., & Hughes, D. (2010). Dyadic developmental psychotherapy: An effective

and evidence-based treatment—comments in response to Mercer and Pignotti. Child &

Family Social Work, 15(1), 6-11. doi:10.1111/j.1365-2206.2009.00679.x

Beecher, B. (2009). The medical model, mental health practitioners, and individuals with

schizophrenia and their families. Journal of Social Work Practice, 23(1), 9-20.

doi:10.1080/02650530902723282

Belfer, M. L. (2008). Child and adolescent mental disorders: The magnitude of the problem

across the globe. Journal of Child Psychology and Psychiatry, 49(3), 226-236.

doi:10.1111/j.1469-7610.2007.01855.x

Blount, R. (1987). The dissemination of cost-effective psychosocial programs for children in

health care settings. Children's Health Care, 15(4), 206-213.

Bodenmann, G., Cina, A., Ledermann, T., & Sanders, M. R. (2008). The efficacy of the Triple P-

Positive Parenting Program in improving parenting and child behavior: A comparison

with two other treatment conditions. Behaviour Research and Therapy, 46(4), 411-427.

doi:10.1016/j.brat.2008.01.001

Brannon, A. M., Heflinger, C. A., & Bickman, L. (1997). The caregiver strain questionnaire:

Measuring the impact on the family of living with a child with serious emotional

disturbance. Journal of Emotional and Behavioral Disorders, 5(4), 212-222.

Brinkmeyer, M. Y., & Eyberg, S. M. (2003). Parent-child interaction therapy for oppositional

children. In A. E. Kazdin & J. R. Weisz (Eds.) , Evidence-based psychotherapies for

children and adolescents (pp. 204-223). New York, NY US: Guilford Press.

Carr, A. (2009a). The effectiveness of family therapy and systemic interventions for child-

focused problems. Journal of Family Therapy, 31(1), 3-45.

Carr, A. (2009b). The effectiveness of family therapy and systemic interventions for adult-

focused problems. Journal of Family Therapy, 31(1), 46-74.

Carr, A. (2010). Thematic review of family therapy journals 2009. Journal of Family Therapy,

32(4), 409-427. doi:10.1111/j.1467-6427.2010.00524.x

Chen, H. H., Cohen, P. P., Crawford, T. N., Kasen, S. S., Guan, B. B., & Gorden, K. K. (2009).

Impact of early adolescent psychiatric and personality disorder on long-term physical

health: A 20-year longitudinal follow-up study. Psychological Medicine, 39(5), 865-874.

doi:10.1017/S0033291708004182

Cotton-Cornelius, D. (2004). Change processes and the behavior of foster children as a function

of an increased number of counseling sessions and the therapeutic alliance: A 12–18

month longitudinal analysis. Dissertation Abstracts International: Section B: The

Sciences and Engineering, 65(6-B), 3149.

Crane, D. R. (1995). An introduction to behavioural family therapy for families with young

children. Journal of Family Therapy, 17(2), 229-242.

Crane, D.R., Christenson, J.D., Shaw, A.L., Fawcett, D., & Marshall, E.S. (2010). Predictors of

health care use for children of marriage and family therapy clients. Journal of Couple and

Relationship Therapy, 9(4), 277-292. DOI: 10.1080/15332691.2010.515530

Crane, D. R., Hillin, H. H., & Jakubowski, S. F. (2005). Costs of treating conduct disordered

Medicaid youth with and without family therapy. The American Journal of Family

Therapy, 33(5), 403-413. doi:10.1080/01926180500276810

Crane, D. R., & Payne, S. H. (2011). Individual versus family psychotherapy in managed care:

Comparing the costs of treatments by the mental health professions. Journal of Marital

and Family Therapy, 37(3), 273-289.

Crane, D. R., Shaw, A. L., Christenson, J. D., Larson, J. H., Harper, J. M., & Feinauer, L. L.

(2010). Comparison of the family therapy educational and experience requirements for

licensure or certification in six mental health disciplines. American Journal of Family

Therapy, 38(5), 357-373. doi:10.1080/01926187.2010.513895

Creswell, C., & Waite, P. (2009). The use of CBT with children and adolescents. In P. Waite, &

T. Williams, (Eds.), Obsessive compulsive disorder: Cognitive behaviour therapy with

children and young people (pp. 19-32). New York, NY: Routledge/Taylor & Francis

Group.

Curtis, N. M., Ronan, K. R., Heiblum, N., & Crellin, K. (2009) Dissemination and effectiveness

of multisystemic treatment in New Zealand: A benchmarking study. Journal of Family

Psychology, 23(2), 119–129.

DeChillo, N. (1993). Collaboration between social workers and families of patients with mental

illness. Families in Society, 74(2), 104-114.

DeChillo, N., Koren, P.E., & Schultze, K. H. (1994). From paternalism to partnership: Family

and professional collaboration in children’s mental health. American Journal of

Orthopsychiatry, 64(4), 564-576.

Dickens, G. L., & Campbell, J. J. (2001). Absconding of patients from an independent UK

psychiatric hospital: A 3-year retrospective analysis of events and characteristics of

absconders. Journal of Psychiatric and Mental Health Nursing, 8(6), 543-550.

doi:10.1046/j.1351-0126.2001.00426.x

Dierker, L., Nargiso, J., Wiseman, R., & Hoff, D. (2001). Factors predicting attrition within a

community initiated system of care. Journal of Child and Family Studies, 10(3), 367-383.

doi:10.1023/A:1012581027044

Dorsey, S., Briggs, E. C., & Woods, B. A. (2011). Cognitive-behavioral treatment for

posttraumatic stress disorder in children and adolescents. Child and Adolescent

Psychiatric Clinics of North America, 20(2), 255-269. doi:10.1016/j.chc.2011.01.006

Eisenberg, D., & Neighbors, K. (2007). Economics of preventing mental disorders and substance

abuse among young people. Paper commissioned by the Committee on Prevention of

Mental Disorders and Substance Abuse among Children, Youth, and Young Adults:

National Research Council and the Institute of Medicine, Washington, DC.

Etgar, T., & Shulstain-Elrom, H. (2009). A combined therapy model (individual and family) for

children with sexual behavior problems. International Journal of Offender Therapy and

Comparative Criminology, 53(5), 574-595. doi:10.1177/0306624X08319914

Farmer, E. M. Z., Burns, B. J., Angold, A., & Costello, E. (1997). Impact of children’s mental

health problems on families: Relationships with service use. Journal of Emotional and

Behavioral Disorders, 5(4), 230-238.

Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G*Power 3: A flexible statistical

power analysis program for the social, behavioral, and biomedical sciences. Behavior

Research Methods, 39, 175-191.

Fawcett, D. & Crane, D. R. (In press). The influence of profession and therapy type on the

treatment of sexual dysfunctions. Journal of Sex & Marital Therapy.

Fergusson, D. M., Horwood, L., & Ridder, E. M. (2005). Show me the child at seven: The

consequences of conduct problems in childhood for psychosocial functioning in

adulthood. Journal of Child Psychology and Psychiatry, 46(8), 837-849.

doi:10.1111/j.1469-7610.2004.00387.x

Friesen, B. J. & Stephens, B. (1998). Expanding family roles in the system of care: Research and

practice. In M. H. Epstein, K. Kutash, & A. Duchnowski (Eds.). Outcomes for children

and youth with emotional and behavioral disorders and their families: Programs and

evaluation best practices (pp. 231-259). Austin, TX: PRO-ED.

Foster, E., Jensen, P. S., Schlander, M., Pelham, W. R., Hechtman, L., Arnold, L., & ... Wigal, T.

(2007). Treatment for ADHD: Is more complex treatment cost-effective for more

complex cases. Health Services Research, 42(1), 165-182. doi:10.1111/j.1475-

6773.2006.00599.x

Garrett, A. (1997). Essays in the economics of child mental health. Dissertation Abstracts

International Section A, 57.

Garcia, J., & Weisz, J. R. (2002). When youth mental health care stops: Therapeutic relationship

problems and other reasons for ending youth outpatient treatment. Journal of Consulting

and Clinical Psychology, 70(2), 439-443. doi:10.1037/0022-006X.70.2.439

Goisman, R. M., Warsaw, M. G., & Keller, M. B. (1999). Psychosocial treatment prescriptions

for generalized anxiety disorder, panic disorder and social phobia, 1991–1996. American

Journal of Psychiatry, 156(11), 1819–1821.

Hafemeister, T. L., & Banks, S. M. (1996). Methodological advances in the use of recidivism

rates to assess mental health treatment programs. Journal of Mental Health

Administration, 23(2), 190-206. doi:10.1007/BF02519110

Hahn, E. (1995). Predicting Head Start parent involvement in an alcohol and other drug

prevention program. Nursing Research, 44(1), 45-51.

Hass, S. A. (2007). The long-term effects of poor childhood health: An assessment and

application of retrospective reports. Demography, 44, 113-135.

Henggeler, S. W. (2011). Efficacy studies to large-scale transport: The development and

validation of multisystemic therapy programs. Annual Review of Clinical Psychology,

7351-381. doi:10.1146/annurev-clinpsy-032210-104615

Henggeler, S. W., Schoenwald, S. K., Borduin, C. M., Rowland, M. D., & Cunningham, P. B.

(2009). Multisystemic therapy for antisocial behavior in children and adolescents (2nd

ed.). New York, NY US: Guilford Press.

Huffine, C. & Anderson, D. (2003). Family advocacy development in systems of care. In A. J.

Pumariega & N. C. Winters (Eds.) The handbook of child and adolescent systems of care,

(pp. 35-65). San Francisco, CA: Jossey-Bass.

Hughes, J. L., & Asarnow, J. (2011). Family treatment strategies in adolescent depression.

Psychiatric Annals, 41(4), 235-239. doi:10.3928/00485713-20110325-07

Janicke, D. M., Finney, J. W., & Riley, A. W. (2001). Children’s health care use: A

prospective investigation of factors related to care-seeking. Medical Care, 39(9), 990-

1001.

Johnson, E., Mellor, D., & Brann, P. (2008). Differences in dropout between diagnoses in child

and adolescent mental health services. Clinical Child Psychology and Psychiatry, 13(4),

515-530.

Josephson, A. M., & Serrano, A. (2001). The integration of individual therapy and family

therapy in the treatment of child and adolescent psychiatric disorders. Child and

Adolescent Psychiatric Clinics of North America, 10(3), 431-450.

Kam, S., & Midgley, N. (2006). Exploring 'clinical judgment': How do child and adolescent

mental health professionals decide whether a young person needs individual

psychotherapy?. Clinical Child Psychology and Psychiatry, 11(1), 27-44.

doi:10.1177/1359104506059122

Kaslow, N. J., Broth, M., Smith, C., & Collins, M. H. (2012). Family�based interventions for

child and adolescent disorders. Journal of Marital and Family Therapy, 38(1), 82-100.

doi:10.1111/j.1752-0606.2011.00257.x

Kataoka, S., Zhang, L., & Wells, K. (2002). Unmet need for mental health care among U.S.

children: Variation by ethnicity and insurance status. American Journal of Psychiatry,

159(9), 1548-1555.

Kazdin, A. E., Holland, L., & Crowley, M. (1997). Family experience of barriers to treatment

and premature termination from child therapy. Journal of Consulting and Clinical

Psychology, 65(3), 453-463.

Kazdin, A. E., & Mazurick, J. L. (1994). Dropping out of child psychotherapy: Distinguishing

early and late dropouts over the course of treatment. Journal of Consulting and Clinical

Psychology, 62(5), 1069-1074.

Kazdin, A. E., Mazurick, J. L., & Bass, D. (1993). Risk for attrition in treatment of antisocial

children and families. Journal of Clinical Child Psychology, 22(1), 2-16.

doi:10.1207/s15374424jccp2201_1

Kazdin, A. E., Mazurick, J. L., & Siegel, T. (1994). Treatment outcome among children with

externalizing disorder who terminate prematurely versus those who complete

psychotherapy. Journal of the American Academy of Child & Adolescent Psychiatry,

33(4), 549-557.

Kendall, P. C. (2012). Child and adolescent therapy: Cognitive-behavioral procedures (4th ed.).

New York, NY US: Guilford Press.

Kendall, P. C., & Beidas, R. S. (2007). Smoothing the trail for dissemination of evidence-based

practices for youth: Flexibility within fidelity. Professional Psychology: Research and

Practice, 38(1), 13-20. doi:10.1037/0735-7028.38.1.13

Kendall, P. C., Hedtke, K. A., & Aschenbrand, S. G. (2006). Anxiety disorders. In D. A. Wolfe

& E. J. Mash (Eds.), Behavioral and emotional disorders in adolescents: Nature,

assessment, and treatment (pp. 259-299). New York, NY: Guilford Publications.

Kilian, R., Losert, C., Park, A., McDaid, D., & Knapp, M. (2010). Cost-effectiveness analysis in

child and adolescent mental health problems: An updated review of literature. The

International Journal of Mental Health Promotion, 12(4), 45-57.

Knapp, M., McCrone, P., Fombonne, E., Beecham, J., & Wostear, G. (2002). The Maudsley

long-term follow-up of child and adolescent depression: 3. Impact of comorbid conduct

disorder on service use and costs in adulthood. The British Journal of Psychiatry, 180(1),

19-23. doi:10.1192/bjp.180.1.19

Knizter, J. (1993). Children's mental health policy: Challenging the future. Journal of Emotional

and Behavioral Disorders, 1(1), 8-16.

Kutash, K., & Rivera, V. R., (1996). What works in children’s mental health services?

Uncovering answers to critical questions. Baltimore, MD: Paul H. Brookes Publishing

Co.

Lai, K. C., Pang, A. T., Wong, C. K., Lum, F. F., & Lo, M. K. (1998). Characteristics of

dropouts from a child psychiatry clinic in Hong Kong. Social Psychiatry and Psychiatric

Epidemiology, 33(1), 45-58. doi:10.1007/s001270050021

Letourneau, E. J., Henggeler, S.W., Borduin, C. M., Schewe, P. A., McCart, M. R., Chapman, J.

E. & Saldana, L. (2009). Multisystemic therapy for juvenile sexual offenders: 1-year

results from a randomized effectiveness trial. Journal of Family Psychology, 23, 89–102.

Lochman, J. E., Powell, N. P., Boxmeyer, C. L., & Jimenez-Camargo, L. (2011). Cognitive-

behavioral therapy for externalizing disorders in children and adolescents. Child and

Adolescent Psychiatric Clinics of North America, 20(2), 305-318.

doi:10.1016/j.chc.2011.01.005

Lochman, J. E., Powell, N. P., Whidby, J., & Fitzgerald, D. (2006). Aggressive children:

Cognitive– behavioral assessment and treatment. In P. C. Kendall (Ed.), Child and

adolescent therapy: Cognitive– behavioral procedures 3rd ed., (pp. 33–81). New York:

Guilford Press.

Lock, J., & Fitzpatrick, K. (2007). Evidenced-based treatments for children and adolescents with

eating disorders: Family therapy and family-facilitated cognitive-behavioral therapy.

Journal of Contemporary Psychotherapy, 37(3), 145-155. doi:10.1007/s10879-007-9049-

Lopez, W., Mullins, L., Wolfe-Christensen, C., & Bourdeau, T. (2008). The relation between

parental psychological distress and adolescent anxiety in youths with chronic illnesses:

The mediating effect of perceived child vulnerability. Children's Health Care, 37(3),

171-182.

Loveless, A., & Holman, T. (Eds.). (2007). The family in the new millennium: World voices

supporting the 'natural' clan, Vol 1: The place of family in human society. Westport, CT:

Praeger Publishers/Greenwood Publishing Group.

Luk, E. L., Staiger, P. K., Mathai, J. J., Wong, L. L., Birleson, P. P., & Adler, R. R. (2001).

Children with persistent conduct problems who dropout of treatment. European Child &

Adolescent Psychiatry, 10(1), 28-36. doi:10.1007/s007870170044

Mackie, J., Groves, K., Hoyle, A., Garcia, C., Garcia, R., Gunson, B., & Neuberger, J. (2001).

Orthotopic liver transplantation for alcoholic liver disease: a retrospective analysis of

survival, recidivism, and risk factors predisposing to recidivism. Liver Transplantation,

7(5), 418-427.

MacLaren, J., & Kain, Z. (2008). Development of a brief behavioral intervention for children's

anxiety at anesthesia induction. Children's Health Care, 37(3), 196-209.

Marsh, D. T., & Johnson, D. L. (1997). The family experience of mental illness: Implications for

intervention. Professional Psychology: Research and Practice, 28(3), 229-237.

Masi, M. V., Miller, R. B., & Olson, M. M. (2003). Differences in dropout rates among

individual, couple, and family therapy clients. Contemporary Family Therapy, 25(1), 63-

75. doi:10.1023/A:1022558021512

McCabe, K., Yeh, M., Hough, R. L., Landsverk, J., Hurlburt, M. S., Culver, S., & Reynolds, B.

(1999). Racial/ethnic representation across five public sectors of care for youth. Journal

of Emotional and Behavioral Disorders, 7(2), 72-82. doi:10.1177/106342669900700202

McCammon, S. L., Spencer, S. A., & Friesen, B. J. (2001). Promoting family empowerment

through multiple roles. In D. A. Dosser, D. H. Handron, S. McCammon, & J. Y. Powell

(Eds.), Child mental health: Exploring systems of care in the new millennium. New York:

The Haworth Press, Inc.

McCrone, P., Knapp, M., & Fombonne, E. (2005). The Maudsley long-term follow-up of child

and adolescent depression: Predicting costs in adulthood. European Child & Adolescent

Psychiatry, 14(7), 407-413. doi:10.1007/s00787-005-0491-6

McCulloch, A., Ryrie, I., Williamson, T., & St. John, T. (2005). Has the medical model a

future?. Mental Health Review Journal, 10 (1), 1–11.

McDaid, D., Park, A., Knapp, M., Losert, C., & Kilian, R. (2010). Making the case for investing

in child and adolescent mental health: How can economics help?. The International

Journal of Mental Health Promotion, 12(4), 37-44.

McKay, M., McCadam, K., & Gonzales, J. (1996). Addressing the barriers to mental health

services for inner city children and their caretakers. Community Mental Health Journal,

32(4), 353-361.

Melzer, H., Gatward, R., Goodman, R, & Ford, T. (2000). The mental health of children and

adolescents in Great Britain. London: The Stationery Office.

Merikangas, K.R., He, J.P., Brody, D., Fisher, P.W., Bourdon, K., Koretz, D.S. (2010).

Prevalence and treatment of mental disorders among US children in the 2001-2004

NHANES. Pediatrics, 125(1):75-81.

Meyers, J. (1985). Federal efforts to improve mental health services for children: Breaking a

cycle of failure. Journal of Clinical Child Psychology, 14(3), 182-187.

Moore, A. M., Hamilton, S., Crane, D., & Fawcett, D. (2011). The influence of professional

license type on the outcome of family therapy. American Journal of Family Therapy,

39(2), 149-161. doi:10.1080/01926187.2010.530186

Morgan, T. B., & Crane, D. (2010). Cost-effectiveness of family-based substance abuse

treatment. Journal of Marital and Family Therapy, 36(4), 486-498. doi:10.1111/j.1752-

0606.2010.00195.x

Mufson, L., Dorta, K., Wickramaratne, P., Nomura, Y., Olfson, M., & Weissman, M. M. (2004).

A randomized effectiveness trial of interpersonal psychotherapy for depressed

adolescents. Archives of General Psychiatry, 61(6), 577-584.

doi:10.1001/archpsyc.61.6.577

Nelson, W., Finch, A., Jr., & Ghee, A. (2006). Anger management with children and adolescents:

Cognitive– behavioral therapy. In P. C. Kendall (Ed.), Child and adolescent therapy:

Cognitive– behavioral procedures (3rd ed.) (pp. 114–168). New York: Guilford Press.

Nock, M., & Kazdin, A. (2001). Parent expectancies for child therapy: Assessment and relation

to participation in treatment. Journal of Child and Family Studies, 10(2), 155-180.

Palloni, A. (2006). Reproducing inequalities: Luck, wallets, and the enduring effects of

childhood health. Demography, 43(4), 587-615.

Paster, V. S. (1997). Emerging perspectives in child mental health services. In R. J. Illback, C. T.

Cobb, H. R. Joseph (Eds.), Integrated services for children and families: Opportunities

for psychological practice (pp. 259-279). Washington, DC: American Psychological

Association. doi:10.1037/10236-011

Patterson, J. J., Mockford, C. C., & Stewart-Brown, S. (2005). Parents' perceptions of the value

of the Webster-Stratton Parenting Programme: A qualitative study of a general practice

based initiative. Child: Care, Health and Development, 31(1), 53-64. doi:10.1111/j.1365-

2214.2005.00479.x

Piacentini, J. C., March, J. S., & Franklin, M. E. (2006). Cognitive-behavioral therapy for youth

with obsessive-compulsive disorder. In P. C. Kendall (Ed.), Child and adolescent therapy:

Cognitive-behavioral procedures (3rd ed.) (pp. 297-321). New York, NY US: Guilford

Press.

Prinz, R., & Miller, G. (1994). Family-based treatment for childhood antisocial behavior:

Experimental influences on dropout and engagement. Journal of Consulting and Clinical

Psychology, 62(3), 645-650.

Powers, S. W., Jones, J. S., & Jones, B. A. (2005). Behavioral and cognitive-behavioral

interventions with pediatric populations. Clinical Child Psychology and Psychiatry, 10(1),

65-77. doi:10.1177/1359104505048792

Racusin, G. R., & Kaslow, N. J. (1994). Child and family therapy combined: Indications and

implications. American Journal of Family Therapy, 22(3), 237-246.

doi:10.1080/01926189408251317

Riesser, G., & Schorske, B. J. (1994). Relationships between family caregivers and mental health

professionals: The American experience. In H. P. Lefley & M. Wasow (Eds.), Helping

families cope with mental illness (pp. 3-26). Langhorne, PA England: Harwood

Academic Publishers/Gordon.

Rhodes, P., Brown, J. and Madden, S. (2009) The Maudsley model of family based treatment for

anorexia nervosa: A qualitative evaluation of parent-to parent consultation. Journal of

Marital and Family Therapy, 35(2): 181–192.

Richardson, L., Russo, J., Lozano, P., McCauley, E., & Katon, W. (2008). The effect of

comorbid anxiety and depressive disorders on health care utilization and costs among

adolescents with asthma. General Hospital Psychiatry, 30(5), 398-406.

Romansky, J. B., Lyons, J. S., Lehner, R., & West, C. M. (2003). Factors related to psychiatric

hospital readmission among children and adolescents in state custody. Psychiatric

Services, 54(3), 356-362. doi:10.1176/appi.ps.54.3.356

Sanders, M.R., Markie-Dadds, C., & Turner, K.M.T. (2001). Practitioner’s manual for Standard

Triple P. Milton, Qld: Families International Publishing Pty. Ltd.

Sanders, M.R., Pidgeon, A.M. (2005). Practitioner’s manual for Pathways Triple P. Milton, Qld:

Triple P International Pty. Ltd.

Sburlati, E. S., Schniering, C. A., Lyneham, H. J., & Rapee, R. M. (2011). A model of therapist

competencies for the empirically supported cognitive behavioral treatment of child and

adolescent anxiety and depressive disorders. Clinical Child and Family Psychology

Review, 14(1), 89-109. doi:10.1007/s10567-011-0083-6

Scheeringa, M., Peebles, C., Cook, C., & Zeanah, C. (2001). Toward establishing procedural,

criterion, and discriminant validity for PTSD in early childhood. Journal of the American

Academy of Child & Adolescent Psychiatry, 40(1), 52-60. doi:10.1097/00004583-

200101000-00016.

Schmidt, U., Lee, S., Beecham, J., Perkins, S., Treasure, J., Yi, I., & ... Eisler, I. (2007). A

randomized controlled trial of family therapy and cognitive behavior therapy guided self-

care for adolescents with bulimia nervosa and related disorders. The American Journal of

Psychiatry, 164(4), 591-598. doi:10.1176/appi.ajp.164.4.591

Sexton, T. L., & Alexander, J. F. (2002). Functional family therapy: An empirically supported,

family-based intervention model for at-risk adolescent and their families. In F. Kaslow

(Ed.), Comprehensive handbook of psychotherapy: Cognitive, behavioral and functional

approaches, Vol. 2. (pp. 177–140). Hoboken, NJ: John Wiley & Sons Inc.

Shadish, W., & Baldwin, S. (2003). Meta-analysis of MFT interventions. Journal of Marital and

Family Therapy, 29(4), 547-570.

Sheidow, A. J., Bradford, W., Henggeler, S. W., Rowland, M. D., Halliday-Boykins, C.,

Schoenwald, S. K., & Ward, D. M. (2004). Treatment costs for youths receiving

multisystemic therapy or hospitalization after a psychiatric crisis. Psychiatric Services,

55(5), 548-554. doi:10.1176/appi.ps.55.5.548

Shumaker, K. R. (2000). Measured professional competence between and among different

mental health disciplines when evaluating and making recommendations in cases of

suspected child sexual abuse. Dissertation Abstracts International, 60.

Sills, M., Shetterly, S., Xu, S., Magid, D., & Kempe, A. (2007). Association between parental

depression and children's health care use. Pediatrics, 119(4), e829-e836.

Simpson, G. A., Cohen, R. A., Pastor, P. N., & Reuben, C. A. (2008). Use of mental health

services in the past 12 months by children aged 4-17 years: United States, 2005-2006.

NCHS Data Brief, 8, 1-8.

Sirles, E. (1990). Dropout from intake, diagnostics, and treatment. Community Mental Health

Journal, 26(4), 345-360.

Spielmans, G. I., Pasek, L. F., & Mcfall, J. P. (2007). What are the active ingredients in cognitive

and behavioral psychotherapy for anxious and depressed children? A meta-analytic

review. Clinical Psychology Review, 27(5), 642-654. doi:10.1016/j.cpr.2006.06.001

Stark, K., & Kendall, P. C. (1996). Treating depressed children: Therapist manual for “Taking

ACTION.” Ardmore, PA: Workbook.

Stevens, M., Roberts, H., & Shiell, A. (2010). Research review: Economic evidence for

interventions in children's social care: Revisiting the what works for children project.

Child & Family Social Work, 15(2), 145-154.

Stewart, R. E., & Chambless, D. L. (2009). Cognitive-behavioral therapy for adult anxiety

disorders in clinical practice: A metaanalysis of effectiveness studies. Journal of

Consulting and Clinical Psychology, 77(4), 595–606.

Stroul, B. A., & Friedman, R. M., (1986). A system of care for children and youth with severe

emotional disturbances. Washington DC: Georgetown University Child Development

Center.

Stroul, B. A., & Friedman, R. M. (1996). The system of care concept and philosophy. In B. A.

Stroul (Ed.), Children’s mental health: Creating systems of care in a changing society.

(pp. 3-21). Baltimore, MD: Brookes.

Stubbe, D. E., & Weiss, G. (2000). Psychosocial interventions: Individual psychotherapy with

the child, and family interventions. Child and Adolescent Psychiatric Clinics of North

America, 9(3), 663-670.

Trowell, J., Joffe, I., Campbell, J., Clemente, C., Almqvist, F., Soininen, M., & ... Tsiantis, J.

(2007). Childhood depression: A place for psychotherapy: An outcome study comparing

individual psychodynamic psychotherapy and family therapy. European Child &

Adolescent Psychiatry, 16(3), 157-167. doi:10.1007/s00787-006-0584-x

Tucker, M., & Oei, T. S. (2007). Is group more cost effective than individual cognitive behaviour

therapy? The evidence is not solid yet. Behavioural and Cognitive Psychotherapy, 35(1),

77-91. doi:10.1017/S1352465806003134

U. S. Public Health Service. (2000). Report of the surgeon general’s conference on children’s

mental health: A national action agenda. Washington, DC: U.S. Department of Health

and Human Services.

Waldron, H. B., & Turner, C. W. (2008). Evidence based psychosocial treatments for adolescent

substance abuse. Journal of Clinical Child & Adolescent Psychology, 37(1), 238–261.

Watt, B. D., & Dadds, M. R. (2007). Facilitating treatment attendance in child and adolescent

mental health services: A community study. Clinical Child Psychology and Psychiatry,

12(1), 105-116. doi:10.1177/1359104507071089

Weisz, J. R., Huey, S. J., & Weersing, V. (1998). Psychotherapy outcome research with children

and adolescents: The state of the art. Advances in Clinical Child Psychology, 20, 49-91.

Whitson, H., Heflin, M., & Burchett, B. (2006). Patterns and predictors of smoking cessation in

an elderly cohort. Journal of the American Geriatrics Society, 54(3), 466-471.

Wood, J. J., McLeod, B. D., Piacentini, J. C., & Sigman, M. (2009). One-year follow-up of

family versus child CBT for anxiety disorders: Exploring the roles of child age and

parental intrusiveness. Child Psychiatry and Human Development, 40(2), 301-316.

doi:10.1007/s10578-009-0127-z

Yatchmenoff, D. K., Koren, P. E., Friesen, B. J., Gordon, L. J., & Kinney, R. F. (1998).

Enrichment and stress in families caring for a child with a serious emotional disorder,

Journal of Child and Family Studies, 7(2), 129-145.

Yorgason, J. B., Booth, A., & Johnson, D. (2008). Health, disability, and marital quality: Is the

association different for younger versus older cohorts? Research on Aging, 30(6), 623-

648.

Table 1

Results of statistical analyses for dropout and recidivism by license type and therapy modality

License N Dropout N Recidivism

MD 3672 34.2% 2415 23.4%

Nurse 1405 17.3% 1162 23.3%

Psychologist 74614 18.0% 61200 22.8%

Social worker 64266 15.2% 54510 23.8%

MFT 15404 16.0% 12932 20.5%

Counselor 40849 16.6% 34060 22.0%

Modality

Individual 119797 36.3% 76253 17.7%

Family 32186 47.4% 16835 12.0%

Mixed 48227 6.6% 45059 42.5%

Table 2

Results of statistical analyses for outcomes by license type and therapy modality

Sessions Total Cost Cost / Session

Cost Effectiveness

LN Sessions LN Total Cost

LN Cost Effectiveness

License M SD M SD M SD M SD M SD M SD M SD

MD 8.3 10.7 $571.5 781.7 $72.7 25.2 599.4 797.9 1.69 0.83 5.9 0.90 5.9 0.90

Nurse 8.4 8.3 $454.0 516.2 $52.6 11.1 459.0 532.0 1.76 0.86 5.7 0.87 5.7 0.85

Psychologist 8.3 8.2 $446.6 542.4 $54.8 13.0 473.3 557.5 1.76 0.85 5.7 0.89 5.8 0.85

Social worker 8.6 9.1 $405.5 465.9 $44.8 12.6 406.9 481.3 1.80 0.84 5.6 0.86 5.6 0.87

MFT 8.2 8.8 $384.1 451.8 $45.3 12.1 390.8 466.4 1.75 0.82 5.5 0.85 5.6 0.86

Counselor 8.1 8.4 $374.7 445.2 $44.6 13.0 378.4 459.1 1.74 0.89 5.5 0.88 5.5 0.86

Modality

Individual 7.3 8.0 $361.7 366.2 $48.6 14.3 371.7 391.2 1.65 0.88 5.5 0.87 5.5 0.88

Family 5.6 5.9 $281.3 252.9 $49.8 14.4 288.0 260.2 1.46 0.83 5.3 0.82 5.4 0.82

Mixed 11.7 11.2 $584.6 533.4 $49.2 13.5 603.2 623.1 2.13 0.90 5.9 0.89 6.0 0.90

Notes: LN is the natural log transformation of the data.

Table 3

Tukey post-hoc analyses for comparisons in Table 2.

Variable License (I) License (J) Mean Diff.

(I - J)

Std.

Error

Sig.

Total Sessions Counselor MD -.27 .18 .70

Nurse -.33 .26 .82

Social Worker -.59* .06 .00

MFT -.17 .09 .41

Psychologist -.25* .06 .00

MD Counselor .27 .18 .70

Nurse -.06 .31 1.00

Social Worker -.32 .19 .50

MFT .10 .19 .99

Psychologist .02 .18 1.00

Nurse MD .33 .27 .82

Nurse .06 .32 1.00

Social Worker -.27 .26 .91

MFT .15 .27 .99

Psychologist .07 .26 1.00

Social Worker Counselor .59* .06 .00

MD .32 .18 .50

Nurse .27 .26 .91

MFT .42* .08 .00

Psychologist .34* .05 .00

MFT Counselor .17 .09 .41

MD -.09 .19 .99

Nurse -.15 .27 .99

Social Worker -.41* .08 .00

Psychologist -.08 .08 .93

Psychologist Counselor .25* .06 .00

MD -.01 .18 1.00

Nurse -.07 .26 1.00

Social Worker -.34* .05 .00

MFT .082 .09 .93

Total Cost Counselor MD -220.15* 10.45 .00

Nurse -79.82* 14.81 .00

Social Worker -27.39* 3.43 .00

MFT -12.29 5.12 .16

Psychologist -94.38* 3.35 .00

MD Counselor 220.15* 10.45 .00

Nurse 140.33* 17.73 .00

Social Worker 192.75* 10.32 .00

MFT 207.85* 11.00 .00

Psychologist 125.76* 10.30 .00

Nurse MD 79.82* 14.81 .00

Nurse -140.33* 17.73 .00

Social Worker 52.42* 14.72 .01

MFT 67.52* 15.20 .00

Psychologist -14.55 14.70 .92

Social Worker Counselor 27.39* 3.43 .00

MD -192.75* 10.32 .00

Nurse -52.42* 14.72 .01

MFT 15.10* 4.85 .02

Psychologist -66.98* 2.92 .00

MFT Counselor 12.26 5.12 .15

MD -207.85* 11.00 .00

Nurse -67.52* 15.20 .00

Social Worker -15.10* 4.85 .02

Psychologist -82.08* 4.80 .00

Psychologist Counselor 94.38* 3.35 .00

MD -125.76* 10.30 .00

Nurse 14.56 14.70 .92

Social Worker 66.98* 2.92 .00

MFT 82.08* 4.80 .00

Cost/Session Counselor MD -28.03* .27 .00

Nurse -7.97* .38 .00

Social Worker -.12 .09 .70

MFT -.67* .13 .00

Psychologist -10.16* .08 .00

MD Counselor 28.03 .27 .00

Nurse 20.06* .46 .00

Social Worker 27.90* .27 .00

MFT 27.36* .28 .00

Psychologist 17.87* .27 .00

Nurse MD 7.97* .38 .00

Nurse -20.06* .46 .00

Social Worker 7.84* .38 .00

MFT 7.29* .39 .00

Psychologist -2.19* .38 .00

Social Worker Counselor .12 .09 .70

MD -27.90* .27 .00

Nurse -7.84* .38 .00

MFT -.54* .12 .00

Psychologist -10.03* .07 .00

MFT Counselor .67* .13 .00

MD -27.36* .28 .00

Nurse -7.29* .39 .00

Social Worker .54* .12 .00

Psychologist -9.49* .12 .00

Psychologist Counselor 10.16* .08 .00

MD -17.87* .27 .00

Nurse 2.19* .38 .00

Social Worker 10.03* .07 .00

MFT 9.49* .12 .00

Cost Effective Counselor MD -221.03* 10.77 .00

Nurse -80.58* 15.25 .00

Social Worker -28.49* 3.53 .00

MFT -12.41 5.28 .17

Psychologist -94.95* 3.45 .00

MD Counselor 221.03* 10.77 .00

Nurse 140.44* 18.26 .00

Social Worker 192.53* 10.63 .00

MFT 208.62* 11.33 .00

Psychologist 126.07* 10.61 .00

Nurse MD 80.58* 15.25 .00

Nurse -140.44* 18.26 .00

Social Worker 52.08* 15.16 .01

MFT 68.17* 15.66 .00

Psychologist -14.37 15.14 .93

Social Worker Counselor 28.49* 3.53 .00

MD -192.53* 10.63 .00

Nurse -52.08* 15.16 .01

MFT 16.08* 5.00 .02

Psychologist -66.46* 3.01 .00

MFT Counselor 12.41 5.28 .17

MD -208.62* 11.33 .00

Nurse -68.17* 15.66 .00

Social Worker -16.08* 5.00 .02

Psychologist -82.54* 4.95 .00

Psychologist Counselor 94.95* 3.45 .00

MD -126.07* 10.61 .00

Nurse 14.37 15.14 .93

Social Worker 66.46* 3.01 .00

MFT 82.54* 4.95 .00

Note: * indicates a statistically significant difference at p < .05.

Table 4 Results of statistical analyses for outcomes by diagnosis Sessions Total Cost Cost / session Cost effectiveness

Diagnosis N Dropout N Recidivism M SD M SD M SD M SD

Adjustment 22725 18.1% 18602 20.4% 7.6 8.1 $373.6 444.5 $48.2 13.9 384.4 458.1

Anxiety /PTSD 21260 14.7% 18128 24.2% 9.1 9.5 $457.3 537.1 $49.4 14.6 470.5 553.0

Disruptive behavior

13586 16.2% 11383 21.7% 8.1 8.1 $404.7 447.6 $49.1 13.8 416.2 460.7

Dissociative 121 14.9% 103 26.4% 8.2 8.0 $406.0 382.2 $50.5 17.1 418.0 395.3

Eating disorders 968 9.2% 879 29.0% 12.2 12.2 $620.6 708.6 $49.9 15.5 639.0 729.9

Mood disorders 41832 13.8% 36071 24.6% 9.3 10.0 $465.1 563.2 $48.9 13.8 478.8 580.2

Relational 772 25.8% 573 9.7% 4.9 4.2 $229.5 242.1 $46.4 17.1 235.0 246.7

Schizophrenic/ psychotic

25 16.0% 21 44.0% 12.7 6.4 $650.1 732.2 $48.3 12.0 672.1 755.8

Substance use and abuse

2166 20.4% 1724 15.7% 6.8 6.4 $335.4 348.8 $49.7 17.4 344.2 357.9

Other diagnoses 96755 18.6% 78795 22.6% 7.9 8.4 $395.1 474.5 $49.0 14.3 406.6 488.5

Table 5 Tukey post hoc analyses for comparisons by diagnosis. Variable Diagnosis (I) Diagnosis (J) Mean Diff.

(I - J)

Std.

Error

Sig.

EoC1_Total_Dollars Adjustment Anxiety -83.68* 5.18 .000

Disruptive -31.13* 5.91 .000

Dissociative -32.40 49.11 1.000

Eating -247.00* 17.15 .000

Mood -91.57* 4.48 .000

Relational 144.09* 21.08 .000

Schizophrenia -276.52 108.52 .243

Substance 38.11 12.51 .070

Other -21.55* 4.05 .000

Anxiety Adjustment 83.68* 5.18 .000

Disruptive 52.54* 5.94 .000

Dissociative 51.25 49.11 .990

Eating -163.32* 17.16 .000

Mood -7.85 4.52 .770

Relational 227.77* 21.08 .000

Schizophrenia -192.87 108.52 .750

Substance 121.85* 12.52 .000

Other 62.12* 4.09 .000

Disruptive Adjustment 31.13* 5.91 .000

Anxiety -52.54* 5.94 .000

Dissociative -1.36 49.19 1.000

Eating -215.87* 17.39 .000

Mood -60.43* 5.34 .000

Relational 175.23* 21.28 .000

Schizophrenia -245.4 108.56 .415

Substance 69.31* 12.84 .000

Other 9.59 4.98 .654

Dissociative Adjustment 32.40 49.11 1.000

Anxiety -51.25 49.11 .990

Disruptive 1.36 49.19 1.000

Eating -214.54* 51.76 .001

Mood -59.10 49.04 .972

Relational 176.55* 53.19 .031

Schizophrenia -244.12 119.00 .562

Substance 70.63 50.41 .927

Other 10.90 49.00 1.000

Eating Adjustment 247.01* 17.15 .000

Anxiety 163.329* 17.16 .000

Disruptive 215.87* 17.39 .000

Dissociative 214.54* 51.70 .001

Mood 155.43* 16.96 .000

Relational 391.10* 26.68 .000

Schizophrenia -29.51 109.75 1.000

Substance 285.18* 20.59 .000

Other 225.45* 16.85 .000

Mood Adjustment 91.57* 4.48 .000

Anxiety 7.85 4.52 .770

Disruptive 60.43* 5.34 .000

Dissociative 59.10 49.04 .972

Eating -155.43* 16.96 .000

Relational 235.66* 20.92 .000

Schizophrenia -184.92 108.49 .793

Substance 129.74* 12.25 .000

Other 70.01* 3.15 .000

Relational Adjustment -144.09* 21.08 .000

Anxiety -227.77* 21.08 .000

Disruptive -175.23* 21.28 .000

Dissociative -176.55* 53.19 .031

Eating -391.10* 26.68 .000

Mood -235.66* 20.92 .000

Schizophrenia -420.66* 110.43 .005

Substance -105.92* 23.96 .000

Other -165.65* 20.83 .000

Schizophrenia Adjustment 276.52 108.55 .243

Anxiety 192.87 108.55 .750

Disruptive 245.42 108.56 .415

Dissociative 244.10 119.00 .562

Eating 29.55 109.75 1.000

Mood 184.99 108.49 .793

Relational 420.66* 110.43 .005

Substance 314.74 109.12 .110

Other 255.00 108.47 .356

Substance Adjustment -38.17 12.51 .070

Anxiety -121.85* 12.52 .000

Disruptive -69.31* 12.84 .000

Dissociative -70.63 50.41 .927

Eating -285.18* 20.59 .000

Mood -129.74* 12.25 .000

Relational 105.92* 23.96 .000

Schizophrenia -314.72 109.12 .110

Other -59.73* 12.10 .000

Other Adjustment 21.55* 4.05 .000

Anxiety -62.12* 4.09 .000

Disruptive -9.59 4.98 .654

Dissociative -10.94 49.00 1.000

Eating -225.45* 16.85 .000

Mood -70.01* 3.15 .000

Relational 165.65* 20.83 .000

Schizophrenia -255.06 108.47 .356

Substance 59.73* 12.10 .000

Total Sessions Adjustment Anxiety -1.42* .09 .000

Disruptive -.50* .10 .000

Dissociative -.58 .87 1.000

Eating -4.58* .30 .000

Mood -1.70* .08 .000

Relational 2.76* .37 .000

Schizophrenia -5.03 1.97 .215

Substance .88* .22 .003

Other -.30* .07 .001

Anxiety Adjustment 1.46* .09 .000

Disruptive .95* .10 .000

Dissociative .87 .87 .992

Eating -3.12* .30 .000

Mood -.24 .08 .067

Relational 4.22* .37 .000

Schizophrenia -3.57 1.98 .704

Substance 2.34* .22 .000

Other 1.15* .07 .000

Disruptive Adjustment .50* .10 .000

Anxiety -.95* .10 .000

Dissociative -.08 .87 1.000

Eating -4.08* .31 .000

Mood -1.20* .09 .000

Relational 3.26* .37 .000

Schizophrenia -4.52 1.93 .361

Substance 1.39* .22 .000

Other .19 .08 .437

Dissociative Adjustment .58 .87 1.000

Anxiety -.87 .87 .992

Disruptive .08 .87 1.000

Eating -4.00* .92 .001

Mood -1.12 .87 .958

Relational 3.35* .94 .015

Schizophrenia -4.44 2.13 .529

Substance 1.47 .89 .826

Other .28 .87 1.000

Eating Adjustment 4.58* .30 .000

Anxiety 3.12* .30 .000

Disruptive 4.08* .31 .000

Dissociative 4.00* .92 .001

Mood 2.88* .30 .000

Relational 7.35* .47 .000

Schizophrenia -.44 1.95 1.000

Substance 5.47* .36 .000

Other 4.28* .30 .000

Mood Adjustment 1.70* .08 .000

Anxiety .27 .08 .067

Disruptive 1.20* .09 .000

Dissociative 1.12 .87 .958

Eating -2.88* .30 .000

Relational 4.47* .37 .000

Schizophrenia -3.32 1.95 .783

Substance 2.59* .21 .000

Other 1.40* .05 .000

Relational Adjustment -2.76* .37 .000

Anxiety -4.22* .37 .000

Disruptive -3.26* .37 .000

Dissociative -3.35* .94 .015

Eating -7.35* .47 .000

Mood -4.47* .37 .000

Schizophrenia -7.79* 1.96 .003

Substance -1.87* .42 .000

Other -3.07* .37 .000

Schizophrenia Adjustment 5.03 1.93 .215

Anxiety 3.57 1.93 .704

Disruptive 4.52 1.93 .361

Dissociative 4.44 2.11 .529

Eating .44 1.95 1.000

Mood 3.32 1.93 .783

Relational 7.79* 1.96 .003

Substance 5.92 1.94 .070

Other 4.72 1.93 .298

Substance Adjustment -.88* .22 .003

Anxiety -2.34* .22 .000

Disruptive -1.39* .22 .000

Dissociative -1.47 .89 .826

Eating -5.47* .36 .000

Mood -2.59* .21 .000

Relational 1.87* .42 .000

Schizophrenia -5.92 1.94 .070

Other -1.19* .21 .000

Other Adjustment .30* .07 .001

Anxiety -1.15* .07 .000

Disruptive -.19 .08 .437

Dissociative -.28 .87 1.000

Eating -4.28* .30 .000

Mood -1.40* .05 .000

Relational 3.07* .37 .000

Schizophrenia -4.72 1.93 .298

Substance 1.19* .21 .000

EoC1_CostEffectiveness Adjustment Anxiety -86.09* 5.34 .000

Disruptive -31.80* 6.09 .000

Dissociative -33.67 50.57 1.000

Eating -254.63* 17.66 .000

Mood -94.37* 4.62 .000

Relational 149.36* 21.70 .000

Schizophrenia -287.79 111.75 .229

Substance 40.21 12.88 .057

Other -22.20* 4.17 .000

Anxiety Adjustment 86.09* 5.34 .000

Disruptive 54.29* 6.12 .000

Dissociative 52.43 50.57 .990

Eating -168.54* 17.67 .000

Mood -8.28 4.65 .750

Relational 235.45* 21.71 .000

Schizophrenia -201.67 111.75 .733

Substance 126.30* 12.89 .000

Other 63.88* 4.21 .000

Disruptive Adjustment 31.80* 6.09 .000

Anxiety -54.29* 6.12 .000

Dissociative -1.85 50.65 1.000

Eating -222.83* 17.91 .000

Mood -62.57* 5.50 .000

Relational 181.16* 21.91 .000

Schizophrenia -255.92 111.79 .396

Substance 72.01* 13.22 .000

Other 9.59 5.13 .690

Dissociative Adjustment 33.65 50.57 1.000

Anxiety -52.43 50.57 .990

Disruptive 1.85 50.61 1.000

Eating -220.97* 53.38 .001

Mood -60.71 50.50 .972

Relational 183.02* 54.77 .029

Schizophrenia -254.10 122.54 .546

Substance 73.87 51.91 .920

Other 11.45 50.46 1.000

Eating Adjustment 254.63* 17.66 .000

Anxiety 168.54* 17.67 .000

Disruptive 222.83* 17.91 .000

Dissociative 220.97* 53.30 .001

Mood 160.25* 17.47 .000

Relational 403.99* 27.48 .000

Schizophrenia -33.13 113.01 1.000

Substance 294.84* 21.21 .000

Other 232.42* 17.35 .000

Mood Adjustment 94.37* 4.62 .000

Anxiety 8.28 4.65 .750

Disruptive 62.57* 5.50 .000

Dissociative 60.71 50.50 .972

Eating -160.25* 17.47 .000

Relational 243.74* 21.55 .000

Schizophrenia -193.38 111.72 .778

Substance 134.58* 12.61 .000

Other 72.17* 3.25 .000

Relational Adjustment -149.36* 21.70 .000

Anxiety -235.45* 21.71 .000

Disruptive -181.16* 21.91 .000

Dissociative -183.02* 54.77 .029

Eating -403.99* 27.48 .000

Mood -243.74* 21.55 .000

Schizophrenia -437.12* 113.71 .005

Substance -109.15* 24.68 .000

Other -171.52* 21.45 .000

Schizophrenia Adjustment 287.76 111.75 .229

Anxiety 201.67 111.75 .733

Disruptive 255.96 111.79 .396

Dissociative 254.10 122.54 .546

Eating 33.13 113.01 1.000

Mood 193.38 111.72 .778

Relational 437.12* 113.71 .005

Substance 327.97 112.36 .100

Other 265.55 111.70 .339

Substance Adjustment -40.21 12.88 .057

Anxiety -126.30* 12.89 .000

Disruptive -72.01* 13.22 .000

Dissociative -73.87 51.91 .920

Eating -294.84* 21.21 .000

Mood -134.58* 12.56 .000

Relational 109.15* 24.68 .000

Schizophrenia -327.97 112.36 .100

Other -62.41* 12.46 .000

Other Adjustment 22.20* 4.17 .000

Anxiety -63.88* 4.21 .000

Disruptive -9.59 5.13 .690

Dissociative -11.45 50.46 1.000

Eating -232.42* 17.35 .000

Mood -72.17* 3.25 .000

Relational 171.57* 21.45 .000

Schizophrenia -265.55 111.70 .339

Substance 62.41* 12.46 .000

EoC1_CostPerSession Adjustment Anxiety -1.20* .14 .000

Disruptive -.94* .16 .000

Dissociative -2.32 1.40 .818

Eating -1.69* .48 .019

Mood -.73* .12 .000

Relational 1.79 .60 .084

Schizophrenia -.11 3.09 1.000

Substance -1.54* .35 .001

Other -.81* .11 .000

Anxiety Adjustment 1.20* .14 .000

Disruptive .26 .16 .872

Dissociative -1.11 1.40 .999

Eating -.48 .48 .993

Mood .47* .12 .009

Relational 3.00* .60 .000

Schizophrenia 1.09 3.09 1.000

Substance -.33 .35 .995

Other .39* .11 .029

Disruptive Adjustment .94* .16 .000

Anxiety -.26 .16 .872

Dissociative -1.38 1.40 .993

Eating -.75 .49 .888

Mood .21 .15 .930

Relational 2.74* .60 .000

Schizophrenia .83 3.09 1.000

Substance -.596 .36 .831

Other .122 .14 .997

Dissociative Adjustment 2.32 1.40 .818

Anxiety 1.11 1.40 .999

Disruptive 1.38 1.40 .993

Eating .63 1.47 1.000

Mood 1.59 1.39 .981

Relational 4.12 1.51 .167

Schizophrenia 2.21 3.39 1.000

Substance .78 1.43 1.000

Other 1.50 1.34 .987

Eating Adjustment 1.69* .44 .019

Anxiety .48 .48 .993

Disruptive .75 .49 .888

Dissociative -.63 1.47 1.000

Mood .96 .48 .608

Relational 3.49* .76 .000

Schizophrenia 1.50 3.13 1.000

Substance .15 .58 1.000

Other .87 .48 .719

Mood Adjustment .73* .12 .000

Anxiety -.47* .12 .009

Disruptive -.21 .15 .930

Dissociative -1.59 1.35 .981

Eating -.96 .48 .608

Relational 2.52* .59 .001

Schizophrenia .62 3.09 1.000

Substance -.81 .34 .376

Other -.08 .09 .995

Relational Adjustment -1.79 .60 .084

Anxiety -3.00* .60 .000

Disruptive -2.74* .60 .000

Dissociative -4.12 1.51 .167

Eating -3.49* .76 .000

Mood -2.52* .59 .001

Schizophrenia -1.90 3.15 1.000

Substance -3.33* .68 .000

Other -2.61* .59 .000

100

Schizophrenia Adjustment .11 3.09 1.000

Anxiety -1.09 3.09 1.000

Disruptive -.83 3.08 1.000

Dissociative -2.21 3.39 1.000

Eating -1.58 3.13 1.000

Mood -.62 3.09 1.000

Relational 1.90 3.15 1.000

Substance -1.43 3.11 1.000

Other -.70 3.09 1.000

Substance Adjustment 1.54* .35 .001

Anxiety .33 .35 .995

Disruptive .59 .36 .831

Dissociative -.78 1.43 1.000

Eating -.15 .58 1.000

Mood .81 .34 .376

Relational 3.33* .68 .000

101

Schizophrenia 1.43 3.11 1.000

Other .72 .34 .524

Other Adjustment .81* .13 .000

Anxiety -.39* .11 .029

Disruptive -.12 .14 .997

Dissociative -1.50 1.39 .987

Eating -.87 .48 .719

Mood .08 .09 .995

Relational 2.617* .59 .000

Schizophrenia .70 3.09 1.000

Substance -.72 .34 .524

102

Table 6

A summary of the percentage of cases treated by each license type.

Diagnosis Group MD Nurse Psychologist Social Worker MFT Counselors

Adjustment 4.6% 8.2% 11.7% 11.1% 10.7% 12.1%

Anxiety/PTSD 11.5% 12.0% 11.2% 10.3% 10.6% 10.0%

Disruptive 4.7% 5.7% 7.1% 6.6% 7.3% 6.5%

Dissociative 0.1% 0.1% 0.1% 0.001% 0.1% .004%

Eating Disorder 0.5% 1.2% 0.4% 0.6% 0.5% 0.5%

Mood Disorder 23.7% 25.3% 18.0% 23.6% 22.7% 20.8%

Relational 0.0% 0.0% 0.2% 0.5% 0.7% 0.5%

Schizophrenia 0.0% 0.0% 0.001% 0.0% 0.001% 0.001%

Substance Abuse 3.6% 0.4% 0.5% 13.% 1.0% 1.6%

Other 51.2% 47.1% 50.8% 46.0% 46.4% 48.0%

103

Table 7 Analysis results by diagnosis and license type Sessions Total Cost Cost effectiveness

Diagnosis License N Dropout N Recidivism M SD M SD M SD

Adjustment MD 170 28.8% 121 18.8% 8.7 8.2 604.0 707.1 616.1 719.7

Nurse 115 15.7% 97 23.5% 8.2 8.2 420.1 451.7 432.4 467.8

Psychologist 8745 19.1% 7072 21.4% 7.7 7.4 423.1 506.1 434.2 520.5

Social worker

7101 17.2% 5887 20.7% 7.8 7.9 348.8 427.9 360.0 442.6

MFT 1642 18.3% 1342 17.2% 7.2 7.5 319.0 360.1 328.6 372.3

Counselor 4946 17.4% 4083 19.3% 7.5 7.1 333.4 348.5 343.6 359.6

Anxiety MD 423 29.8% 297 25.8% 8.7 9.5 633.2 790.0 646.1 805.9

Nurse 168 16.1% 141 23.2% 9.2 8.9 503.8 564.2 517.3 579.8

Psychologist 8351 15.1% 7089 24.1% 9.0 9.2 508.0 595.0 521.1 611.0

Social worker

6607 13.1% 5744 25.3% 9.6 9.1 434.6 494.7 448.6 511.7

MFT 1637 16.6% 1365 21.3% 8.6 8.1 405.4 503.5 417.5 518.9

Counselor 4074 14.3% 3492 23.9% 8.7 8.9 394.9 445.3 407.4 459.9

Disruptive Behavior

MD 171 21.6% 134 20.5% 7.3 7.6 348.3 378.3 493.8 426.5

104

Nurse 80 10.0% 72 23.8% 7.8 8.0 377.0 432.4 388.0 448.8

Psychologist 5291 16.7% 4408 23.0% 8.4 8.6 464.8 509.5 476.7 523.4

Social worker

4272 15.7% 3601 22.1% 8.3 8.6 378.5 420.0 390.3 433.3

MFT 1128 13.9% 971 18.6% 7.7 7.8 347.9 353.9 358.4 365.7

Counselor 2644 16.9% 2197 19.5% 7.7 7.9 348.3 378.3 358.9 390.4

Dissociative MD 4 0.0% 4 0.0% 3.8 3.2 298.1 226.1 301.9 229.0

Nurse 1 0.0% 1 100% 3 -- 189.0 -- 195.0 --

Psychologist 59 20.3% 47 27.1% 8.8 8.4 490.6 407.8 503.4 419.6

Social worker

31 9.7% 28 29.0% 7.0 6.7 288.3 271.4 298.9 282.1

MFT 8 25.0% 6 25.0% 4.9 4.3 191.8 132.9 198.7 138.4

Counselor 18 5.6% 17 22.2% 9.9 10.1 479.8 489.4 496.1 512.3

Eating Disorders

MD 18 5.6% 17 11.1% 9.0 8.7 611.3 526.5 622.0 533.9

Nurse 17 5.9% 16 47.1% 14.7 13.9 834.7 751.3 859.0 773.2

Psychologist 284 9.8% 315 27.3% 12.4 12.6 709.7 798.7 728.3 821.2

Social worker

355 8.7% 324 29.6% 12.2 12.4 576.1 635.4 594.7 656.4

MFT 75 12.0% 66 32.0% 12.2 11.8 556.4 529.8 575.3 548.5

105

Counselor 188 8.5% 172 29.8% 12.0 12.8 562.9 742.6 580.9 764.9

Mood Disorders

MD 872 31.8% 595 22.8% 9.1 11.2 655.7 813.5 669.0 830.4

Nurse 356 16.3% 298 26.1% 9.2 8.8 500.4 566.0 514.2 583.0

Psychologist 13447 12.8% 11720 25.0% 9.5 9.1 528.7 620.5 542.7 638.6

Social worker

15165 13.1% 13180 25.5% 9.5 9.7 436.4 500.0 450.3 516.5

MFT 3493 13.1% 3035 22.6% 9.2 9.1 431.1 531.2 444.3 548.7

Counselor 8499 14.8% 7243 23.3% 8.9 9.4 411.8 549.4 424.5 565.8

Relational MD 1 0.0% 1 100.0% 5.0 -- 483.8 -- 493.0 --

Nurse -- -- -- -- -- -- -- -- -- --

Psychologist 142 33.1% 95 7.7% 5.0 6.1 277.0 276.7 282.8 272.9

Social worker

323 21.1% 255 9.6% 5.1 5.0 229.5 261.9 235.0 265.9

MFT 115 27.0% 84 15.7% 4.5 4.2 208.5 198.9 214.0 205.4

Counselor 191 27.7% 138 7.3% 4.9 5.6 207.6 204.0 213.0 208.4

Schizophrenia MD -- -- -- -- -- -- -- -- -- --

Nurse -- -- -- -- -- -- -- -- -- --

Psychologist 6 33.3% 4 50.0% 9.8 10.2 606.3 404.8 623.0 415.5

Social 5 0.0% 4 40.0% 12.2 12.3 494.6 453.3 515.4 478.9

106

worker

MFT 5 20.0% 4 20.0% 5.8 6.2 231.5 259.4 241.3 274.0

Counselor 9 11.1% 8 55.6% 17.9 16.2 978.6 1036 1010.1 1067.3

Substance use/ abuse

MD 134 26.1% 99 15.7% 6.5 7.1 458.4 386.9 467.1 394.5

Nurse 6 16.7% 5 50.0% 4.8 3.2 233.6 80.3 421.4 82.4

Psychologist 380 18.4% 310 16.1% 7.1 7.5 398.4 438.8 407.4 448.3

Social worker

827 20.3% 659 15.7% 6.7 7.0 318.2 321.5 326.9 330.4

MFT 150 18.0% 123 10.0% 7.2 6.8 320.6 309.4 329.6 320.8

Counselor 669 21.1% 528 15.7% 6.6 6.9 301.2 314.2 309.8 323.9

Other MD 1879 39.0% 1147 24.2% 8.0 9.2 561.8 829.0 574.0 846.3

Nurse 662 19.6% 532 20.7% 7.7 7.9 406.7 481.4 417.5 496.0

Psychologist 37878 20.3% 30171 22.1% 7.8 8.1 431.5 503.4 442.7 517.1

Social worker

29574 16.1% 24827 23.9% 8.3 8.9 377.2 454.4 389.2 469.2

MFT 7151 17.0% 5936 20.5% 8.0 8.3 367.4 428.2 378.6 441.9

Counselor 19611 17.5% 16182 22.4% 7.8 8.2 352.2 420.2 363.9 433.7

107

Table 8 Results of statistical analyses for outcomes of family therapy by license type Sessions Total Cost Cost / session Cost effectiveness

License N Dropout N Recidivism M SD M SD M SD M SD

MD 1192 51.9% 573 28.6% 7.6 7.7 611.6 665.7 79.3 27.0 621.5 674.9

Nurse 247 24.3% 187 16.3% 6.6 6.1 356.7 252.5 54.0 10.1 364.8 259.2

Psychologist 12333 32.1% 8379 16.2% 6.9 6.7 385.8 367.1 55.7 13.1 394.0 374.5

Social worker 9112 27.3% 6624 15.4% 6.8 6.2 302.3 272.5 44.2 11.6 310.4 279.3

MFT 2784 26.4% 2050 14.0% 6.9 6.1 314.8 267.9 45.8 10.5 322.8 275.1

Counselor 6518 25.9% 4832 15.7% 6.8 6.6 313.0 299.8 45.5 11.5 321.1 307.9

108

Table 9

Treatment outcomes by age.

Age Dropout

Recidivism

Total Cost $

Total Sessions M

Cost per Session $

Cost Effective

3 25.6% 16.9% 338.7 6.8 49.3 348.0

4 22.7% 20.6% 370.6 7.5 48.9 381.2

5 20.4% 20.4% 372.6 7.6 48.7 383.4

6 19.5% 20.9% 378.8 7.7 49.1 389.6

7 18.6% 22.5% 405.8 8.2 49.0 417.5

8 17.3% 22.9% 405.8 8.1 49.2 417.5

9 17.0% 24.0% 417.4 8.3 49.2 429.6

10 16.6% 24.3% 414.8 8.3 49.2 427.0

11 15.8% 24.6% 427.1 8.6 49.1 439.6

12 16.4% 25.7% 433.2 8.7 49.0 446.1

13 15.3% 25.4% 434.4 8.8 48.8 447.3

14 15.4% 25.1% 437.8 8.8 48.5 450.9

15 15.2% 24.3% 441.5 8.9 48.8 454.4

16 15.5% 22.1% 413.6 8.4 48.7 425.5

17 17.5% 18.1% 395.0 7.9 49.0 405.8

18 20.0% 16.7% 394.8 7.8 49.6 405.5

Title Page
Abstract
Acknowledgments
Table of Contents
List of Tables
Introduction

Literature Review
Research Questions

Method

Design
Sample
Procedure
Definitions

Results
Discussion

Clinical Implications
Limitations and Future Directions

References
Table 1
Table 2
Table 3
Table 4
Table 5
Table 6
Table 7
Table 8
Table 9