Week 4 Appendix A Article
Nursing Case Management, Peer Coaching, and Hepatitis A and B Vaccine Completion Among Homeless Men Recently Released on Parole: Randomized Clinical Trial
Adeline Nyamathi, ANP, PhD, FAAN [Distinguished Professor], University of California, Los Angeles, School of Nursing
Benissa E. Salem, RN, MSN, PhD [Project Director], University of California, Los Angeles, School of Nursing
Sheldon Zhang, PhD [Research Sociologist], San Diego State University, Department of Sociology
David Farabee, PhD [Professor], University of California, Los Angeles, Integrated Substance Abuse Programs
Betsy Hall, PhD [Professor], University of California, Los Angeles, Integrated Substance Abuse Programs
Farinaz Khalilifard, MA, MFT [Project Director], and University of California, Los Angeles, School of Nursing
Barbara Leake, PhD [Senior Statistician] University of California, Los Angeles, School of Nursing
Abstract
Background—Although hepatitis A virus (HAV) and hepatitis B virus (HBV) infections are vaccine-preventable diseases, few homeless parolees coming out of prisons and jails have received
the hepatitis A and B vaccination series.
Objectives—The study focused on completion of the HAV and HBV vaccine series among homeless men on parole. The efficacy of three levels of peer coaching and nurse-delivered
interventions was compared at 12-month follow up: (a) intensive peer coaching and nurse case
management (PC-NCM); (b) intensive peer coaching (PC) intervention condition, with minimal
nurse involvement; and a (c) usual care (UC) intervention condition, which included minimal PC
Corresponding Author: Adeline Nyamathi, ANP, Ph.D., FAAN, UCLA, School of Nursing, Room 2-250, Factor Building, Los Angeles, CA 90095-1702. (anyamath@sonnet.ucla.edu). Adeline Nyamathi, ANP, PhD, FAAN, is Distinguished Professor, University of California, Los Angeles, School of Nursing. Benissa E. Salem, RN, MSN, PhD, is Project Director, University of California, Los Angeles, School of Nursing. Sheldon Zhang, PhD, is Research Sociologist, San Diego State University, Department of Sociology. David Farabee, PhD, is Professor, University of California, Los Angeles, Integrated Substance Abuse Programs. Betsy Hall, PhD, is Professor, University of California, Los Angeles, Integrated Substance Abuse Programs Farinaz Khalilifard, MA, MFT, is Project Director, University of California, Los Angeles, School of Nursing. Barbara Leake, PhD, is Senior Statistician, University of California, Los Angeles, School of Nursing.
The authors have no conflicts of interest to report.
HHS Public Access Author manuscript Nurs Res. Author manuscript; available in PMC 2016 May 01.
Published in final edited form as: Nurs Res. 2015 ; 64(3): 177–189. doi:10.1097/NNR.0000000000000083.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
and nurse involvement. Further, we assessed predictors of vaccine completion among this targeted
sample.
Methods—A randomized control trial was conducted with 600 recently paroled men to assess the impact of the three intervention conditions (PC-NCM vs. PC vs. UC) on reducing drug use and
recidivism; of these, 345 seronegative, vaccine-eligible subjects were included in this analysis of
completion of the Twinrix HAV/HAB vaccine. Logistic regression was added to assess predictors
of completion of the HAV/HBV vaccine series and chi-squared analysis to compare completion
rates across the three levels of intervention.
Results—Vaccine completion rate for the intervention conditions were 75.4% (PC-NCM), 71.8% (PC), and 71.9% (UC) (p =. 78). Predictors of vaccine noncompletion included being Asian
and Pacific Islander, experiencing high levels of hostility, positive social support, reporting a
history of injection drug use, being released early from California prisons, and being admitted for
psychiatric illness. Predictors of vaccine series completion included reporting six or more friends,
recent cocaine use, and staying in drug treatment for at least 90 days.
Discussion—Findings allow greater understanding of factors affecting vaccination completion in order to design more effective programs among the high-risk population of men recently
released from prison and on parole.
Keywords
accelerated Twinrix hepatitis A/B vaccine; ex-offenders; homelessness; parolees; prisoners; substance abuse
With 1.6 million men and women behind bars, the United States (U.S.) has one of the largest
numbers of incarcerated persons when compared to other nations (Pew Charitable Trusts,
2008). In California, over 130,000 are in custody and over 54,000 are on parole (California
Department of Corrections and Rehabilitation, 2013b). Incarcerated populations are at
significant risk for homelessness. When compared to the general population, those who were
in jail were more likely to be homeless (Greenberg & Rosenheck, 2008). In one study,
homeless inmates were more likely to have past criminal justice system involvement for
both nonviolent and violent offenses, to have mental health and substance abuse problems,
and a lack of personal assets (Greenberg & Rosenheck, 2008).
Globally, incarcerated populations encounter a host of public health care issues; two such
issues—hepatitis A virus (HAV) and hepatitis B virus (HBV) diseases—are vaccine
preventable. In addition, viral hepatitis disproportionately impacts the homeless due to
increased risky sexual behaviors and drug use (Stein, Andersen, Robertson, & Gelberg,
2012), along with substandard living conditions (Hennessey, Bangsberg, Weinbaum, &
Hahn, 2009). Other risk factors include, but are not limited to, injection drug use (IDU),
alcohol use and older age, which place the population at risk for being seropositive (Stein et
al., 2012).
As a member of the hepatovirus family, HAV is primarily transmitted via the fecal-oral
route (Zuckerman, 1996). The rate of acute hepatitis in the US is 0.5 per 100,000 (Centers
for Disease Control and Prevention, 2010). While the rate among paroled populations is hard
Nyamathi et al. Page 2
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
to ascertain, data suggest that HAV infection is related to unsanitary living conditions, i.e.,
poor water sanitation (World Health Organization, 2014), for which homeless populations
are at risk.
A member of the hapdnavirus family, HBV (Immunization Action Coalition, 2013;
Zuckerman, 1996) disproportionately burdens homeless (Nyamathi, Liu, et al., 2009;
Nyamathi, Sinha, Greengold, Cohen, & Marfisee, 2010) and incarcerated populations
(Immunization Action Coalition, 2013; Khan et al., 2005), leading to fulminant liver failure,
chronic liver disease, hepatocellular carcinoma, and death (Rich et al., 2003). HBV can be
transmitted through unprotected sexual activity, needle sharing, IDU (Diamond et al., 2003;
Maher, Chant, Jalaludin, & Sargent, 2004), and percutaneous blood exposure. National
prevalence statistics indicate that HBV affects between 13% to 47% of U.S. prison inmates
(Centers for Disease Control and Prevention, 2004). Illicit drug use is a major contributor to
incarceration and homelessness among ex-offenders (McNeil & Guirguis-Younger, 2012;
Tsai, Kasprow, & Rosenheck, 2014), placing ex-offenders who use drugs at high risk for
HBV infection.
Despite the availability of the HBV vaccine, there has been a low rate of completion for the
three-dose core of the accelerated vaccine series (Centers for Disease Control and
Prevention, 2012). Among incarcerated populations, HBV vaccine coverage is low; in a
study among jail inmates, 19% had past HBV infection, and 12% completed the HBV
vaccination series (Hennessey et al., 2009). While HBV vaccination is well accepted behind
bars—due to a lack of funding and focus on prevention as a core in the prison system—few
inmates may complete the series (Weinbaum, Sabin, & Santibanez, 2005). In addition,
prevention may not be a priority for those who are struggling with managing mental health,
drug use, and dependency issues, along with the need to meet basic necessities (Nyamathi,
Shoptaw, et al., 2010). Authors contend that while the HBV vaccine is cost effective, it is
underutilized among high-risk (Rich et al., 2003) and incarcerated populations (Hunt &
Saab, 2009).
For homeless men on parole, vaccination completion may be affected by level of custody;
generally, the higher the level of custody, the higher the risk an inmate poses. In addition,
various contract types, such as drug treatment-related, and length of time in residential drug
treatment (RDT)—for those with drug histories—may also affect completion of the vaccine
series. For those transitioning into the community, stress, family reunification issues, and the
potential for relapse and recidivism may represent real challenges (Seiter & Kadela, 2003),
and may influence vaccine completion.
Until 1981, the HBV vaccine was not licensed in the U.S. (Centers for Disease Control and
Prevention, 2012). Twenty years later, in 2001, a combination of the HAV and HBV
vaccine, Twinrix, was developed by GlaxoSmithKline and approved by the Food and Drug
Administration (FDA) (Centers for Disease Control and Prevention, 2012). The standard
dosing for this regimen is 0, 1, and 6 months. An alternative dosing schedule (core doses at
0, 7, and 21–31 days and a booster dose 12 months) was approved by the FDA in 2007
(Centers for Disease Control and Prevention, 2012). Thus, many individuals, particularly
older individuals, may not have been vaccinated.
Nyamathi et al. Page 3
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
One strategy to improve vaccination for HAV and HBV among high-risk populations has
been to utilize the accelerated Twinrix HAV/HBV vaccination which provides the core
doses at 0 days, 7 days, and 21–30 days (Nyamathi, Liu, et al., 2009). The Twinrix
recombinant vaccination is administered intramuscularly (GlaxoSmithKline, 2011) by a
licensed nurse. In a randomized controlled trial (RCT) comparing vaccination completion
among incarcerated IDUs in Denmark—using the accelerated versus a standard vaccine
schedule (0, 1 and 6 months)—63% completed the three accelerated dose series compared to
20% of those who received the nonaccelerated series (Christensen et al., 2004). In another
RCT conducted among 297 homeless adults with a history of incarceration, findings
revealed that 50% completed the Twinrix vaccine series. Logistic regression analysis
revealed that those who were more likely to complete the HBV vaccination were over 40
years of age (p = .02), partnered (p = .02), homeless more than one year (p = .025), recent
binge drinkers (p = .03), and had attended recent alcohol anonymous or narcotic anonymous
meetings (p = .006) (Nyamathi, Marlow, Branson, Marfisee, & Nandy, 2012). In another
RCT focused on improving HAV/HBV vaccine completion among 256 homeless adults who
were on methadone maintenance, a greater percentage of participants who completed the
vaccine series also reduced their alcohol consumption by 50% as compared to those who
were unsuccessful in reducing their alcohol consumption (74.4% vs. 64.1%) (Nyamathi,
Shoptaw, et al., 2010).
Finally, in a larger, three-group RCT with 865 homeless adults in shelters located in Los
Angeles, individuals were randomly assigned to one of three groups: (a) nurse case managed
sessions plus hepatitis education, incentives, and tracking; (b) standard hepatitis education
plus incentives and tracking; and (c) standard hepatitis education and incentives only.
Findings reveal that those who were in the nurse case management education, incentives,
and tracking program were significantly more likely to complete a standard three-series
Twinrix vaccination or core of the accelerated dosing schedule (68% vs. 61% vs. 54%,
respectively; p = .01) compared to those who were in the other two programs (Nyamathi,
Liu, et al., 2009). While accelerated vaccination programs have shown success in RCT
studies, including those utilizing nurse case management, little is known about vaccine
completion among an ex-offender population using varying intensities of nurse case
management and peer coaches.
Theoretical Framework
The comprehensive health seeking and coping paradigm (CHSCP) (Nyamathi, 1989),
adapted from a coping model (Lazarus & Folkman, 1984) and the health seeking and coping
paradigm (Schlotfeldt, 1981) guided this study and the variables selected (See Figure 1.).
The CHSCP has been successfully applied by our team to improve our understanding of
HIV and HBV/HCV protective behaviors and health outcomes among homeless adults
(Nyamathi, Liu, et al., 2009)—many of whom had been incarcerated (Nyamathi et al.,
2012).
In this model, a number of factors are thought to relate to the outcome variable, completion
of the HAV/HBV vaccine series. These factors include sociodemographic factors,
situational, personal, and social factors, and health seeking and coping responses.
Nyamathi et al. Page 4
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
Sociodemographic factors that might relate to completion of the vaccine series among
incarcerated populations include age, education, race/ethnicity, and marital and parental
status (Hennessey, Kim, et al., 2009; Salem et al., 2013). Situational factors such as being
homeless (Nyamathi et al., 2012), history of criminal activities, and severity of criminal
history (level of custody and contract type) may likewise influence interest in completing a
vaccination series. Similarly personal factors, such as history of psychiatric and drug use
problems (Hennessey, Kim, et al., 2009; Salem et al., 2013), having hostile tendencies
(Nyamathi et al., 2014), or dealing with physical and mental health problems (Nyamathi et
al., 2011) may interfere with health protective strategies while having social factors present,
such as social support, may facilitate health promotion. Finally, health seeking and coping
strategies may also be known to impact health promotion (Nyamathi, Stein, Dixon,
Longshore, & Galaif, 2003) and compliance with hepatitis vaccine completion.
Purpose
Despite knowledge of awareness of risk factors for HBV infection, intervention programs
designed to enhance completion of the three-series Twinrix HAV/HBV vaccine and
identification of prognostic factors for vaccine completion have not been widely studied.
The purpose of this study was to first assess whether seronegative parolees previously
randomized to any one of three intervention conditions were more likely to complete the
vaccine series, as well as to identify the predictors of HAV/HBV vaccine completion.
Methods
Design
An RCT where 600 male parolees from prison or jail and participating in a RDT program
were randomized into one of three intervention conditions aimed at assessing program
efficacy on reducing drug use and recidivism at six and 12 months, as well as vaccine
completion in eligible subjects: (a) six-month intensive peer coaching and nurse case
management (PC-NCM) intervention condition; (b) an intensive peer coaching (PC)
intervention condition, with minimal nurse involvement; and (c) the usual care (UC)
intervention condition, which had minimal PC and nurse involvement. Of these 600, 345
were eligible for the vaccine (seronegative) and constitute the sample for this report. Data
were collected from February 2010 to January 2013. The study was approved by the
University of California, Los Angeles Institutional Review Board and registered with
Clinical Trials.gov (NCT01844414).
Sample and Site
There were four inclusion criteria for recruitment purposes in assessing program efficacy on
reducing drug use and recidivism: (a) history of drug use prior to their latest incarceration;
(b) between ages of 18 and 60; (c) residing in the participating RDT program; and (d)
designated as homeless as noted on the prison or jail discharge form. A homeless individual
was defined as one who does not have a fixed, regular, and adequate nighttime residence
(National Health Care for the Homeless Council, 2014). Exclusion criteria included: (a)
monolingual speakers of languages other than English or Spanish; and (b) persons judged to
Nyamathi et al. Page 5
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
be cognitively impaired by the research staff. A total of 42 men were screened out due to the
following reasons: age, not being on parole, had not been released from jail or prison within
six months prior to entering the study, or had not used drugs 12 months prior to their most
recent incarceration. Eligibility for receiving the HAV/HBV vaccine series was not
considered an inclusion criterion regarding drug use and recidivism. Among those eligible
and interested, urn randomization (Stout, Wirtz, Carbonari, & Del Boca, 1994) was used to
allocate participants. The variables used in the urn randomization included: age (18–29 and
30 and over), level of custody (1–2 vs. 3–4), HBV vaccine eligibility (HBV seronegative or
seropositive), and level of substance use prior to prison time (low vs. moderate/high
severity). For the present analysis, only vaccine-eligible subjects were included.
Amistad De Los Angeles (Amity) served as the main research site. For the last three
decades, Amity, a nonprofit organization located in California, Arizona, and New Mexico
has been focused on substance abuse treatment, and works with individuals and families
(Amity Foundation, 2014) utilizing a therapeutic environment.
The State of California Assembly passed criminal justice realignment legislation (AB109)
on October 1, 2011 allowing low-level offenders (non-violent, non-serious and non-sex
offenders) to serve their sentence in county jails instead of state prisons (California
Department of Corrections and Rehabilitation, 2011). Post-realignment offenders were more
likely to be convicted of a felony for drug and property crimes (California Department of
Corrections and Rehabilitation, 2013a).
Power analysis—With at least 114 men in each intervention condition, there was 80% power to detect differences of 15 to 20 percentage points (for example, 50% vs. 70%, 75%
vs. 90%) for vaccine completion between either of the two intervention conditions, and the
usual care intervention condition at p = .05.
Vaccine eligibility—Vaccine eligibility included being HBV seronegative and no absolute contraindications (having an allergy to yeast or neomycin, history of neurological disease
[e.g., Guillian-Barre]), prior anaphylactic reaction to HAV/HBV vaccine, a fever of over
100.5 degrees Fahrenheit, and reporting any moderate or severe acute illness beyond mild
cold symptoms (e.g., nonproductive cough, rhinorrhea, or other upper respiratory
symptoms). Of the total sample of 600 study participants, 345 men were eligible for the
HAV/HBV vaccine. Figure 2 (CONSORT diagram) reflects both the larger sample and the
subsample of vaccine eligible participants.
Interventions
Building upon previous studies, we developed varying levels of peer-coached and nurse-led
programs designed to improve HAV/HBV vaccine receptivity at 12-month follow up among
homeless offenders recently released to parole.
Peer coaching–nurse case management (PC-NCM)—The peer coach interacted weekly for about 45 minutes with their assigned participants in person, and for those who
left the facility, interacted by phone. Their focus was on building effective coping skills,
personal assertiveness, self-management, therapeutic nonviolent communication (NVC), and
Nyamathi et al. Page 6
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
self-esteem building. Attention was given to supporting avoidance of health-risk behaviors,
increasing access to medical and psychiatric treatment and improving compliance with
medications, skill-building, and personal empowerment. Discussions also centered on
strategies to assist in seeking support and assistance from community agencies as parolees
prepare for completion of the drug treatment program. Integrated throughout, skill building
in communication and negotiation and issues of empowerment were highlighted.
Peer coaches were also trained to deliver nonviolent communication (NVC), the goal of
which was to increase participants’ mastery of empathic communication skills via a specific
process. The intervention comprised a series of interactive exercises and role playing, based
on conflict in social situations, as identified by the participants. In our study, peer coaches
were former parolees who successfully completed a similar RDT program; as
paraprofessionals, they were positive role models with whom the parolees could identify and
have successfully reintegrated into society. The peer coachers were selected based on having
excellent social skills and found joy helping recent parolees to be successful. The assigned
coach worked with up to 15 parolees at any given time. The coaches in the PC-NCM and PC
intervention conditions were trained in: (a) understanding the needs and challenges faced by
parolees discharged to the community; (b) gaining information about the resources that are
available in the community; and (c) normalizing parolee experiences, setting realistic
expectations and helping the parolee to problem solve with day-to-day events and build on
strengths. The training period for coaches took about one month and consisted of mock role
plays of coaching sessions—with many simulations of problematic and challenging
participants and situations.
Case management, provided by a dedicated nurse (about 20 minutes) was delivered in a
culturally competent manner weekly over eight consecutive weeks. Case management
focused on health promotion, completion of drug treatment, vaccination compliance, and
reduction of risky drug and sexual behaviors. Furthermore, the nurse engaged participants in
role-playing exercises to help them identify potential barriers to appointment keeping, and
asked them to identify personal risk triggers that may hinder vaccine series completion, and
successful HAV, HBV, HCV, and HIV risk reduction. Nurses were trained by experts in
nurse case management, hepatitis infection and transmission, and barriers which impede
HAV/HBV vaccination.
Peer coaching (PC)—Participants assigned to the PC intervention condition received weekly peer coaching interaction similar to the PC component of the NCM-PC intervention
condition. However, while nurse case management was not included, an intervention-
specific nurse encouraged the HAV/HBV vaccination and provided a brief 20-minute
education session on hepatitis and HIV risk reduction.
Usual care (UC)—Participants assigned to the UC intervention condition received the encouragement by a nurse to complete the three series HAV/HBV vaccine. In addition, they
received a brief 20-minute session by a peer counselor about health promotion. They did not
receive any intensive peer coaching sessions or nurse case management sessions.
Nyamathi et al. Page 7
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
At the RDT site, all participants received recovery and rehabilitation services traditionally
delivered for the parolee population, such as residential substance abuse services, assistance
with independent living skills, job skills assistance, literacy, individual, group (small and
large) and family counseling, and coordinated discharge planning. Residents receive highly
structured curriculum and aftercare services in this generally six-month, 24-hour per day,
and seven-day per week community. All coordination for services took place through the
efforts of the in-prison treatment staff, RDT community-based staff, and the parole office.
Procedure
This RCT was conducted in a setting close to the one participating RDT program from
which all participants were enrolled. Posted flyers announced the study to all incoming
residents, and research staff visited the RDT frequently to respond to questions and provide
information in group sessions and individually to those interested in a private location in the
RDT setting. Among interested participants, an informed consent was signed that allowed
the research staff to administer a brief screening questionnaire to assess eligibility criteria.
Among participants who met eligibility criteria, a second informed consent allowed
administration of a baseline questionnaire; a detailed locator guide allows participants to fill
out contact information, addresses and phone numbers in order for research staff to follow
up.
Vaccine administration—After pretest counseling, the research nurses collected serum for testing HBV, HCV, and HIV (hepatitis B core antibody, hepatitis B surface antibody,
hepatitis C antibody, and human immunodeficiency virus antibodies) and provided test
results one week later. Based upon the HBV test result, participants were educated regarding
the timeline for the HAV/HBV vaccine series, provided consent regarding administration,
were inoculated intramuscularly using three doses of the Standard Twinrix (Hepatitis A
Inactivated and Hepatitis B Recombinant Vaccines) for the accelerated dosing schedule of 0
days, 7 days, and 21–30 days. The recommended series of three intramuscular injections of
1.0 ml of Twinrix was administered in the deltoid muscle of the nondominant arm. All
eligible study participants were encouraged to accept the HAV/HBV vaccine; however, this
was not coercive. The nurse documented refusal for vaccination.
Vaccine tracking—On a weekly basis, the research nurse or peer coach reviewed the vaccine dosing and tracked progress. In order to encourage participants to complete the
vaccine series, participants were reminded regarding their next dose by the nurse or peer
coach and provided appointment cards. Further, they were called if not present any longer at
the RDT facility as a reminder. A detailed locator guide, completed by the participant and
interviewer, supported follow up to be successful. Information included contact information
to be used by the research staff for vaccine scheduling, as well as administration of
structured questionnaires at six- and 12-month follow up.
Measures
Vaccine completion—Receipt of three core doses on the accelerated schedule was considered completion. This was assessed by the vaccine tracking system.
Nyamathi et al. Page 8
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
Sociodemographics—Sociodemographic information was collected by a structured questionnaire assessing age, education, race/ethnicity, marital status, and parental status.
Situational factors—included being homeless, history of criminal activity and severity, of criminal history such as level of custody and contract type. Contract type was measured
by asking participants whether they were in custody drug treatment program (ICDTP),
residential multiservice center (RMSC), or parolee substance abuse program (SAP). Time in
RDT was assessed by the total time participants resided at the RDT study site after discharge
from jail/prison to RDT placement. RDT site was dichotomized at the median of 90 days for
analysis.
Personal factors—Personal factors included drug, alcohol, and tobacco use. A modified version of the Texas Christian University (TCU) Drug History form (Simpson & Chatham,
1995) was used to measure use six months preceding the latest incarceration. Information
regarding the frequency of use of alcohol, tobacco, and seven other drugs was collected,
allowing us to review the use of these drugs and selected combinations of these drugs in
terms of use by injection and orally, as well as to extract information about lifetime drug and
alcohol use.Anglin et al. (1996) have verified the reliability and validity of this format.
History of hospitalization for psychiatric and substance use problems, and past treatment for
alcohol or drug problems (number of times in formal treatment for alcohol and for drugs)
was also obtained.
General health was assessed by a single item which asked participants to rate their overall
health on a five-point scale (Stewart, Hays, & Ware, 1988). Responses included poor, fair,
good, very good, and excellent—with a higher score indicating better perceived health.
General health was dichotomized at fair/poor versus good/very good/excellent.
Hostility was measured by the five-item hostility subscale of the Brief Symptom Inventory
(Derogatis & Melisaratos, 1983), in which participants rated the extent to which they have
been bothered (0 = not at all to 4 = extremely) by selected issues. Cronbach’s alpha for the
hostility scale in this sample was .81. The cut-point for hostility was the upper quartile of 2.
Depressive symptoms were assessed by the 10-item, short form of the Center for
Epidemiological Studies Depression scale (CES-D; Radloff, 1977), which was previously
used to assess depressive symptoms in homeless populations (Nyamathi, Christiani, Nahid,
Gregerson, & Leake, 2006; Nyamathi et al., 2008). The 10-item, self-report CES-D
questionnaire measures the frequency of 10 depressive symptoms in the past week on a 4-
point response scale from 0 = rarely or none of the time (less than 1 day) to 3 = all of the
time (5–7 days). Scale scores range from 0 to 30, with higher scores indicating greater
severity of depressive symptoms. Reliability in this sample was .80.
Social factors—Social factors included ever having been removed from their parents as children and having spent time in juvenile hall. In addition, social support was measured by
the Medical Outcomes Study (MOS) Social Support Survey (Sherbourne & Stewart, 1991).
This 18-item scale includes four subscales: emotional support (eight items, reliability in this
sample .95), tangible support (three items, reliability .88), positive support (three items,
reliability .89), and affective support (three items, reliability .90). Items had 5-point, Likert-
Nyamathi et al. Page 9
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
type response options ranging from 1 = none of the time to 5 = all of the time. Responses
were summed for subscale formation with higher scores indicating more support.
Respondents were also asked how many close friends they had outside of prison, which was
dichotomized at the upper quartile of 6 for analysis.
Health seeking and coping were captured by history of drug use and treatment style, as well
as coping. The Carver Brief Cope instrument (Carver, 1997) was used to measure six
dimensions. Coping were assessed with two items for each; planning, instrumental support,
religious, disengagement, denial, and self-blame. Item responses ranged from 1 = I do not do
this at all to 4 = I do this a lot. Coping subscales were dichotomized at their medians for
analysis.
Data Analysis
Sample characteristics were described with frequencies and percentages or means, and
standard deviations and continuous variables were evaluated for normality. Due to highly
skewed distributions that were not resolved by transformations, some variables had to be
categorized for analysis. Associations of sample characteristics with intervention condition
and vaccine noncompletion were assessed with chi-squared tests or analysis of variance
(ANOVA) and two-sample tests. Since IDU may have confounded the relationship between
intervention condition and vaccine noncompletion, we examined the impact of intervention
condition on vaccine noncompletion controlling for IDU using multiple logistic regression
analysis. The model contained IDU and dummy variables for each intervention condition;
the only significant predictor of noncompletion was IDU (p-values for the PC-NCM and PC
intervention conditions were .70 and .79, respectively).
In examining other potential predictors of vaccine noncompletion, we emphasized
noncompletion since individuals who did not complete the vaccine series are the ones who
need to be targeted for future interventions. Variables that were related to vaccine
noncompletion at the .10 level in unadjusted analyses were used as predictors in multiple
logistic regression modeling of noncompliance. Although the overall significance level for
race/ethnicity did not meet this inclusion criterion, it was included in the modeling because
subgroupings (African American, “‘other’ race/ethnicity”) did so. Predictors that were not
significant at the .10 level were removed one by one in descending order of significance.
The final model was checked for multicollinearity and the Hosmer-Lemeshow test was used
to assess model goodness of fit.
Results
In terms of sociodemographic characteristics, the 345 participants who were eligible for the
HAV/HBV vaccine reported a mean age of 42.0 (SD = 9.5) and were predominantly African
American (51%) or Latino (31%), as shown in Table 1. The small subsample of men from
“other” ethnicities comprised mostly Asian-Americans and Pacific Islanders. The mean
education was 11.6 (SD = 1.4). Over half of the participants had never been married (59%).
The distribution of participant characteristics was similar across the three intervention
conditions.
Nyamathi et al. Page 10
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
Vaccine Completion Rates by Intervention Condition
In total, there were 345 individuals who were eligible for the Twinrix recombinant vaccine
(NCM-PC: n = 114; PC: n = 117; and UC: n = 114). The vaccine completion rate for three
or more doses was 73% among all three intervention conditions. Using chi-squared tests
(group × vaccine completion), findings revealed no differences in vaccine completion across
groups (p = .780); PC-NCM: n = 86 [75.4%]; PC: n = 84 [71.8%]; and UC: n = 82 [71.9%]).
Associations With Vaccine Noncompletion
A number of social, personal, coping, and situational factors were found to be related to
vaccine noncompletion (Table 2). In particular, having six or more friends and high
instrumental coping were related to vaccine completion, while having been taken away from
parents or spending time in juvenile hall were related to noncompletion. A history of alcohol
treatment was associated with vaccine completion while having been hospitalized for mental
health problems was related to noncompletion. In terms of drug use, cocaine use within six
months prior to the last incarceration was associated with vaccine completion, while the
opposite was true for IDU ever. Being HCV positive was also associated with not
completing the vaccine series. No association was found between vaccine noncompletion
and childhood physical abuse, whereas a very weak association was found with childhood
sexual abuse.
Finally, those who were released following prison realignment and those tested positive for
HCV at baseline were both related to vaccine noncompletion. Those who spent 90 days or
more in a residential drug treatment facilities following release were more likely to complete
the vaccine series. On the other hand, incarceration location (prison vs. jail) and contract
type had no relationship with vaccine completion, as shown in Table 2.
Table 3 presents the findings of logistic regression analysis. Asian/Pacific Islander ethnicity
(compared to White), higher levels of hostility, higher levels of positive social support, and
history of IDU were related to vaccine noncompletion. Moreover, having been admitted for
a psychiatric illness was related to noncompletion of the HAV/HBV vaccine. Alternatively,
reporting six or more friends was a protective factor. Any alcohol treatment in the past and
recent cocaine use were also found to be related to vaccine completion. Being part of post-
realignment was related to vaccine noncompletion, whereas having been in RDT for at least
90 days was a strong predictor of completion. Although there were no multicollinearity
problems, and the zero-order correlation between having six or more friends and positive
social support was low (.23), we performed sensitivity analyses alternatively dropping one
and then the other variable from the regression model. The direction of the effect of the
social support variable that remained in the model did not change, but the significance was
no longer below the p < .05 level.
Discussion
Although homeless men on parole from California jails and prisons are at high risk for
Hepatitis A and B infection (Weinbaum et al., 2005), few studies have focused on improving
HAV/HBV vaccination completion for this population. This paper presents findings of
Nyamathi et al. Page 11
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
varying levels of peer coaching and nurse-delivered intervention that encouraged all
participants—regardless of intervention condition assignment—to complete the three-series
HAV/HBV accelerated Twinrix vaccine among those eligible. While no treatment
differences were found in terms of vaccine completion rates—due to the bundled nature of
the programs—it is not possible to say whether the peer coaching or nurse-delivered
intervention resulted in the overall successful 73% completion rate of the three series
vaccine. Clearly, an intensive nurse case management approach did not necessarily result in
a greater vaccine completion rate for the PC-NCM intervention condition. Further,
regardless of level of interaction by peer coaches or nurses, encouragement of vaccine
completion was helpful across all intervention conditions (PC-NCM vs. PC vs. UC).
However, we must acknowledge that more than one quarter (27%) did not complete the
vaccine series, despite being informed of their risk for HBV infection.
The fact that Asian American/Pacific Islander (AA/PI) ethnicity was found to be related to
noncompletion of the HAV/HBV vaccine is novel. Minimal work has been done
understanding vaccination compliance among various races and ethnicities within homeless
populations. AA/PIs are a large umbrella group composed of many subgroups; thus, it is
somewhat challenging to decipher why AA/PIs had a higher level of noncompletion.
However, in one study focused on ethnic-specific influences and barriers among AA/PI
children, speaking limited English at home, length of time in the U.S., and not discussing
HBV vaccination with a healthcare provider were found to be barriers to vaccination
(Pulido, Alvarado, Berger, Nelson, & Todoroff, 2001). Despite these findings, the authors
contended that greater understanding of nuances between groups is necessary to understand
barriers (Pulido et al., 2001).
Interestingly, this was not the case for African Americans or Hispanics. In one study,
understanding psychosocial predictors of HAV/HBV vaccination among young African
American men in the south (N = 143), data reveal that increased vaccination was related to
decreased barrier perception, increased perceived medical severity, and perceived barriers of
HBV infection (Rhodes & Diclemente, 2003).
High levels of hostility and having a history of psychiatric hospitalization were likewise
related to noncompletion of the HAV/HBV vaccine series. Adequate assessment of
psychiatric comorbidity may be necessary to improve HAV/HBV vaccine completion by
helping individuals to contend with hostility. Further, adequate mental health referral may
enable homeless ex-offenders to improve vaccine receipt. Future intervention work should
focus on reducing hostility by providing additional group sessions that may aid in managing
the hostility and, ultimately, increasing vaccine receptivity. Furthermore, anger management
has been shown to likewise result in improved outcomes such as sustained reduction in
feeling of anger and physical aggression (Wilson et al., 2013), and improved behavioral and
cognitive coping mechanisms (Tang, 2001).
A history of IDU was also related to vaccine noncompliance. For those struggling with drug
and alcohol addiction, prevention of infection may not be a high priority as meeting the
challenges of overcoming addiction becomes paramount. Despite these findings, recent
cocaine use was found to be related to vaccine completion. It may be that cocaine was not
Nyamathi et al. Page 12
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
used heavily or that it served as a proxy for unmeasured variables associated with vaccine
completion. Daily crack users were less likely to initiate the HBV vaccine series (Ompad et
al., 2004). In this study, however, men who refused the vaccine were counted as not having
completed it.
Increased social support in terms of self-report of having six or more friends was a
protective factor for noncompletion, while the positive social support subscale predicted
noncompletion. Another study found that partner support was predictive of vaccine
completion (Nyamathi et al., 2012); therefore, social support does appear to play a role in
vaccine compliance. When either six or more friends or positive social support was dropped
from the model, the effect of the remaining measure was reduced. Thus, more information
related to the individuals providing social support and the nature of their support is needed
to understand how social support influences HAV/HBV vaccine completion. However, it
seems likely that vaccine completion would be enhanced by interventions aimed at
improving positive social support networks. There was also a trend for those who had any
alcohol treatment to be more likely to complete the vaccine series, perhaps due to increased
access to health education and care. However, drug treatment was unrelated.
Length of time at the RDT site was positively associated with vaccine completion. In fact, in
our sample, homeless men on parole who spent at least three months in RDT programs were
far more likely to complete the vaccination series. Other studies have found that those who
complete RDT are less likely to relapse and use drugs; in addition, they may be less likely to
recidivate (Condelli & Hubbard, 1994; Conner, Hampton, Hunter, & Urada, 2011).
Preventive care, such as vaccination, may be further improved by RDT sites with access to
healthcare practitioners such as public health nurses.
Policies enacted in the California state prison system, in particular, realignment (or reducing
state prison population by transferring inmates to county jails), may affect vaccination
completion. Realignment has shifted responsibility for the custody, treatment, and
supervision of individuals convicted of nonviolent, nonserious, nonsex crimes from the state
to counties (California Realignment, 2013). Our study sample included individual’s pre- and
post- realignment and our findings show that following realignment, vaccination completion
dropped markedly. As this is a relatively new policy enacted in California, it is challenging
to ascertain the possible causes; however, contract types may have been altered for some
individuals at the RDT site, while others may have been shifted from RDT to community
supervision. Thus, the long-term impact of realignment will need to be assessed in the near
future. Findings in this study point to the need for greater understanding of the ramifications
of major criminal justice policies and their effect on preventive care.
This study provides preliminary evidence of the need to incorporate public health nurses
along with peer coaches at RDT sites to improve health promotion, education, and
prevention; and in particular, HAV/HBV vaccination. In fact, RDT facilities are in a prime
position to address the healthcare needs of homeless ex-offenders who are exiting prison and
jail. Partnering with nurses may improve HAV/HBV vaccination rates, but may also
promote health in general. In particular, it would be important for nurses to understand
Nyamathi et al. Page 13
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
predictors of vaccine completion in this targeted population, and to promote greater attention
and focus in the screening process to those individuals less likely to complete.
Equally important, future studies need to incorporate more therapeutic resources and
medical resources for a population which emerges from penitentiaries having experienced
abuse, victimization, and a history of drug use and dependency issues. This study points to
the need for a greater awareness of the needs of IDUs and of the efficacy of tailored
programs focused on these issues. Likewise we propose that more effort be spent on
understanding the thought process of IDU users regarding their beliefs of HAV/HBV
prevention.
Limitations
Homeless men on parole constitute a population with unique health concerns and life issues
affected by the laws and penal practices in their areas. The degree to which findings from
Los Angeles County generalize to other jurisdictions is unknown. Further, self-report is
liable to distortion and impression management. To enhance the vaccination efforts of ex-
offenders, more research is needed to better understand how homeless men on parole
perceive their health, report their health behaviors, and access health care.
Conclusion
Vaccine completion rates were similar to those reported by others and did not differ
according to level of intervention delivered. Asian/Pacific Islander ethnicity, having been
admitted for a psychiatric illness, having higher levels of hostility, higher levels of positive
social support, having a history of IDU, and being part of post realignment were
independently associated with noncompletion, while recent cocaine use, having six or more
friends, and RDT stay of at least 90 days were predictive of completion. Findings advocate
for special attention to screening and enhanced intervention focused among these high-risk
individuals.
Acknowledgments
The authors acknowledge this study was funded by the National Institute on Drug Abuse (NIDA): 1R01DA27213-01. This protocol was registered at ClinicalTrials.gov: NCT 01844414.
References
Amity Foundation. Amity Experience and Expertise. 2014. Retrieved from http://www.amityfdn.org/ About%20Amity/index.php
Anglin, MD.; Longshore, D.; Turner, S.; McBride, D.; Inciardi, JA.; Prendergast, ML. Studies of the Functioning and Effectiveness of Treatment Alternatives to Street Crime (TASC) programs. Los Angeles, CA: UCLA Drug Abuse Research Center; 1996. https://www.ncjrs.gov/App/Publications/ abstract.aspx?ID=169780
Assem. Bill 109. Chapter 15, 2011 Cal. Stat. 2011. http://www.leginfo.ca.gov/pub/11-12/bill/asm/ ab_0101-0150/ab_109_bill_20110404_chaptered.html
California Department of Corrections and Rehabilitation. Public Safety Realignment. 2011. Retrieved from http://www.cdcr.ca.gov/About_CDCR/docs/Realignment-Fact-Sheet.pdf.
California Department of Corrections and Rehabilitation. Realignment Report: A One-Year Examination of Offenders Released from State Prison in the First Six Months of Public Safety
Nyamathi et al. Page 14
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
Realignment. Sacramento, CA: 2013a. Retrieved from http://www.cdcr.ca.gov/realignment/docs/ Realignment%206%20Month%20Report%20Final_5%2016%2013%20v1.pdf.
California Department of Corrections and Rehabilitation. Spring 2013: Adult Population Projections Fiscal Years 2012/13–2017/18. Sacramento, CA: 2013b. Retrieved from http://www.cdcr.ca.gov/ Reports_Research/Offender_Information_Services_Branch/Projections/S13Pub.pdf
California Realignment. About California Realignment.org. Sacramento, CA: 2013. Retrieved from http://www.safeandjust.org/CalRealignment/About-Realignment
Carver CS. You want to measure coping but your protocol’s too long: consider the brief COPE. International Journal of Behavioral Medicine. 1997; 4:92–100. [PubMed: 16250744]
Centers for Disease Control and Prevention. Transmission of Hepatitis B Virus in Correctional Facilities—Georgia January 1999–June 2002. Atlanta, GA: Morbidity and Mortality Weekly Report; 2004. Retrieved from http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5330a2.htm
Centers for Disease Control and Prevention. Table 2.1 reported cases of acute hepatitis A by state— United States, 2006–2010. Atlanta, GA: 2010. Retrieved from http://www.cdc.gov/hepatitis/ Statistics/2010Surveillance/Table2.1.htm
Centers for Disease Control and Prevention. Hepatitis B Epidemiology and Prevention of Vaccine- Preventable Diseases. The Pink Book: Course Textbook—12th Edition. 2012. Retrieved from http://www.cdc.gov/vaccines/pubs/pinkbook/hepb.html
Christensen PB, Fisker N, Krarup HB, Liebert E, Jaroslavtsev N, Christensen K, Georgsen J. Hepatitis B vaccination in prison with a 3-week schedule is more efficient than the standard 6-month schedule. Vaccine. 2004; 22:3897–3901. [PubMed: 15364437]
Condelli WS, Hubbard RL. Relationship between time spent in treatment and client outcomes from therapeutic communities. Journal of Substance Abuse Treatment. 1994; 11:25–33. [PubMed: 8201630]
Conner BT, Hampton AS, Hunter J, Urada D. Treating opioid use under California's Proposition 36: Differential outcomes by treatment modality. Journal of Psychoactive Drugs. 2011; 43:77–83. [PubMed: 22185042]
Derogatis LR, Melisaratos N. The Brief Symptom Inventory: An introductory report. Psychological Medicine. 1983; 13:595–605. [PubMed: 6622612]
Diamond C, Thiede H, Perdue T, Secura GM, Valleroy L, Mackellar D. Seattle Young Men's Survey Team. Viral hepatitis among young men who have sex with men: Prevalence of infection, risk behaviors, and vaccination. Sexually Transmitted Diseases. 2003; 30:425–432. [PubMed: 12916134]
GlaxoSmithKline. Highlights of prescribing information. 2011. https://www.gsksource.com/gskprm/ htdocs/documents/TWINRIX.PDF
Greenberg GA, Rosenheck RA. Jail incarceration, homelessness, and mental health: A national study. Psychiatric Services. 2008; 59:170–177. [PubMed: 18245159]
Hennessey KA, Bangsberg DR, Weinbaum C, Hahn JA. Hepatitis A seroprevalence and risk factors among homeless adults in San Francisco: Should homelessness be included in the risk-based strategy for vaccination? Public Health Reports. 2009; 124:813–817. [PubMed: 19894423]
Hennessey KA, Kim AA, Griffin V, Collins NT, Weinbaum CM, Sabin K. Prevalence of infection with hepatitis B and C viruses and co-infection with HIV in three jails: A case for viral hepatitis prevention in jails in the United States. Journal of Urban Health. 2009; 86:93–105. [PubMed: 18622707]
Hunt DR, Saab S. Viral hepatitis in incarcerated adults: A medical and public health concern. American Journal of Gastroenterology. 2009; 104:1024–1031. [PubMed: 19240708]
Immunization Action Coalition. Vaccine-Preventable Diseases: Hepatitis B. Saint Paul, MN: 2013. Retrieved from http://www.vaccineinformation.org/hepatitis-b
Khan AJ, Simard EP, Bower WA, Wurtzel HL, Khristova M, Wagner KD, Bell BP. Ongoing transmission of hepatitis B virus infection among inmates at a state correctional facility. American Journal of Public Health. 2005; 95:1793–1799. [PubMed: 16186457]
Lazarus, RS.; Folkman, S. Stress, appraisal, and coping. New York, NY: Springer; 1984.
Nyamathi et al. Page 15
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
Maher L, Chant K, Jalaludin B, Sargent P. Risk behaviors and antibody hepatitis B and C prevalence among injecting drug users in south-western Sydney, Australia. Journal of Gastroenterology Hepatology. 2004; 19:1114–1120. [PubMed: 15377287]
McNeil R, Guirguis-Younger M. Illicit drug use as a challenge to the delivery of end-of-life care services to homeless persons: Perceptions of health and social services professionals. Palliative Medicine. 2012; 26:350–359. [PubMed: 21464120]
National Health Care for the Homeless Council. What is the official definition of homelessness?. Nashville, TN: 2014. Retrieved from http://www.nhchc.org/faq/official-definition-homelessness/
National Vaccine Information Center. Vaccine Laws. 2014. Retrieved from http://www.nvic.org/ vaccine-laws.aspx#
Nyamathi A. Comprehensive health seeking and coping paradigm. Journal of Advanced Nursing. 1989; 14:281–290. [PubMed: 2661620]
Nyamathi A, Christiani A, Nahid P, Gregerson P, Leake B. A randomized controlled trial of two treatment programs for homeless adults with latent tuberculosis infection. International Journal of Tuberculosis and Lung Disease. 2006; 10:775–782. [PubMed: 16848340]
Nyamathi A, Leake B, Albarran C, Zhang S, Hall E, Farabee D, Faucette M. Correlates of depressive symptoms among homeless men on parole. Issues in Mental Health Nursing. 2011; 32:501–511. [PubMed: 21767252]
Nyamathi A, Liu Y, Marfisee M, Shoptaw S, Gregerson P, Saab S, Gelberg L. Effects of a nurse- managed program on hepatitis A and B vaccine completion among homeless adults. Nursing Research. 2009; 58:13–22. [PubMed: 19092551]
Nyamathi A, Marlow E, Branson C, Marfisee M, Nandy K. Hepatitis A/B vaccine completion among homeless adults with a history of incarceration. Journal of Forensic Nursing. 2012; 8:13–22. [PubMed: 22372394]
Nyamathi A, Nahid P, Berg J, Burrage J, Christiani A, Aqtash S, Leake B. Efficacy of nurse case- managed intervention for latent tuberculosis among homeless subsamples. Nursing Research. 2008; 57:33–39. [PubMed: 18091290]
Nyamathi A, Salem BES, Farabee D, Hall E, Zhang S, Khalilifard F, Leake B. Predictors of high level of hostility among homeless men on parole. Journal of Offender Rehabilitation. 2014; 53:95–115. [PubMed: 25083121]
Nyamathi A, Shoptaw S, Cohen A, Greengold B, Nyamathi K, Marfisee M, Leake B. Effect of motivational interviewing on reduction of alcohol use. Drug & Alcohol Dependence. 2010; 107:23–30. [PubMed: 19836904]
Nyamathi A, Sinha K, Greengold B, Cohen A, Marfisee M. Predictors of HAV/HBV vaccination completion among methadone maintenance clients. Research in Nursing & Health. 2010; 33:120– 132. [PubMed: 20143328]
Nyamathi AM, Stein JA, Dixon E, Longshore D, Galaif E. Predicting positive attitudes about quitting drug and alcohol use among homeless women. Psychology of Addictive Behaviors. 2003; 17:32– 41. [PubMed: 12665079]
Ompad DC, Galea S, Wu Y, Fuller CM, Latka MH, Koblin B, Vlahov D. Acceptance and completion of hepatitis B vaccination among drug users in New York City. Communicable Disease and Public Health. 2004; 7:294–300. http://deepblue.lib.umich.edu/bitstream/handle/2027.42/40386/ Ompad_Acceptance%20and%20Completion%20of%20Hepatitis%20B %20Vaccination_2004.pdf?sequence=1&isAllowed=y. [PubMed: 15779793]
Pew Charitable Trusts. One in 100: Behind Bars in America in 2008. Washington, DC: 2008. Retrieved from http://www.pewstates.org/research/reports/one-in-100-85899374411
Pulido MJ, Alvarado EA, Berger W, Nelson A, Todoroff C. Vaccinating Asian Pacific Islander children against hepatitis B: Ethnic-specific influences and barriers. Asian American and Pacific Islander Journal of Health. 2001; 9:211–220. [PubMed: 11846367]
Radloff LS. The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement. 1977; 1:385–401.
Rhodes SD, Diclemente RJ. Psychosocial predictors of hepatitis B vaccination among young African- American gay men in the deep south. Sexually Transmitted Diseases. 2003; 30:449–454. [PubMed: 12916138]
Nyamathi et al. Page 16
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
Rich JD, Ching CG, Lally MA, Gaitanis MM, Schwartzapfel B, Charuvastra A, Flanigan TP. A review of the case for hepatitis B vaccination of high-risk adults. American Journal of Medicine. 2003; 114:316–318. [PubMed: 12681460]
Salem BE, Nyamathi A, Keenan C, Zhang S, Marlow E, Khalilifard F, Marfisee M. Correlates of risky alcohol and methamphetamine use among currently homeless male parolees. Journal of Addictive Diseases. 2013; 32:365–376. [PubMed: 24325770]
Schlotfeldt R. Nursing in the future. Nursing Outlook. 1981; 29:295–301. [PubMed: 6908691]
Seiter RP, Kadela KR. Prisoner reentry: What works, what does not, and what is promising. Crime & Delinquency. 2003; 49:360–388.
Sherbourne CD, Stewart AL. The MOS social support survey. Social Science & Medicine. 1991; 32:705–714. [PubMed: 2035047]
Simpson, D.; Chatham, L. TCU/DATAR forms manual. Ft. Worth, TX: Institute of Behavioral Research, Texas Christian University; 1995.
Stein JA, Andersen RM, Robertson M, Gelberg L. Impact of hepatitis B and C infection on health services utilization in homeless adults: A test of the Gelberg-Andersen behavioral model for vulnerable populations. Health Psychology. 2012; 31:20–30. [PubMed: 21574705]
Stewart AL, Hays RD, Ware JE Jr. The MOS short-form general health survey. Reliability and validity in a patient population. Medical Care. 1988; 26:724–735. [PubMed: 3393032]
Stout RL, Wirtz PW, Carbonari JP, Del Boca FK. Ensuring balanced distribution of prognostic factors in treatment outcome research. Journal of Studies on Alcohol and Drugs. 1994; 12:70.
Tang M. Clinical outcome and client satisfaction of an anger management group program. Canadian Journal of Occupational Therapy. 2001; 68:228–236.
Tsai J, Kasprow WJ, Rosenheck RA. Alcohol and drug use disorders among homeless veterans: Prevalence and association with supported housing outcomes. Addictive Behaviors. 2014; 39:455– 460. [PubMed: 23490136]
Weinbaum CM, Sabin KM, Santibanez SS. Hepatitis B, hepatitis C, and HIV in correctional populations: A review of epidemiology and prevention. AIDS. 2005; 19:S41–S46. [PubMed: 16251827]
Wilson C, Gandolfi S, Dudley A, Thomas B, Tapp J, Moore E. Evaluation of anger management groups in a high-security hospital. Criminal Behavior and Mental Health. 2013; 23:356–371.
World Health Organization. Hepatitis A: Key facts. Geneva, Switzerland: 2014. Retrieved from http:// www.who.int/mediacentre/factsheets/fs328/en/
Zuckerman, AJ. Chapter 70: Hepatitis viruses. In: Baron, S., editor. Medical microbiology. 4th ed. Galveston, TX: University of Texas Medical Branch; 1996.
Nyamathi et al. Page 17
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
FIGURE 1. Comprehensive health seeking and coping paradigm. Note for Production/Karina: FETCH
FROM EM
Nyamathi et al. Page 18
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
FIGURE 2. CONSORT diagram. PC = peer coaching; PC-NCM = peer coaching-nursing case
management; UC = usual care.
NOTE for Production/Karina: additional arrows added fro, PC-NCM and UC to flow into
vaccine eligible. This version of the figure should be used (not version in EM). (Objects are
not grouped.)
Nyamathi et al. Page 19
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
Nyamathi et al. Page 20
T A
B L
E 1
D em
og ra
ph ic
, S oc
ia l,
S it
ua ti
on al
, C op
in g
an d
P er
so na
l C
ha ra
ct er
is ti
cs b
y In
te rv
en ti
on C
on di
ti on
A ll
(N =
3 45
) P
C -N
C M
(n =
1 14
) P
C (n
= 1
20 )
U C
(n =
1 11
)
T yp
e C
h ar
ac te
ri st
ic M
(S D
) M
(S D
) M
(S D
) M
(S D
) p
D em
og ra
ph ic
A ge
42 .0
(9 .5
) 41
.3 (1
0. 1)
42 .3
(9 .4
) 42
.6 (8
.9 )
.5 7
n (%
) n
(% )
n (%
) n
(% )
R ac
e/ et
hn ic
it y
.3 0
A fr
ic an
A m
er ic
an 17
5 (5
0. 7)
51 (4
4. 7)
70 (5
9. 8)
54 (4
7. 4)
L at
in o
10 7
(3 1.
0) 40
(3 5.
1) 29
(2 4.
8) 38
(3 3.
3)
A si
an /P
ac if
ic I
sl an
de r
16 (4
.6 )
6 (5
.3 )
3 (2
.6 )
7 (6
.1 )
W hi
te 47
(1 3.
6) 17
(1 4.
9) 15
(1 2.
8) 15
(1 3.
2)
M ar
it al
s ta
tu s
(n ev
er )
20 4
(5 9.
1) 66
(5 7.
9) 71
(6 0.
7) 67
(5 9.
3) .9
1
S oc
ia l
P ar
tn er
s (≥
2 =
y es
) 19
9 (5
7. 7)
64 (5
6. 1)
76 (6
5. 0)
59 (5
1. 8)
.1 2
R em
ov ed
f ro
m p
ar en
ts (
ye s)
18 9
(5 4.
8) 64
(5 6.
1) 69
(5 9.
0) 56
(4 9.
1) .3
0
C hi
ld ho
od s
ex ua
l ab
us e
(y es
) 56
(1 6.
2) 14
(1 2.
3) 22
(1 8.
8) 20
(1 7.
5) .3
6
C hi
ld ho
od p
hy si
ca l
ab us
e (y
es )
12 0
(3 4.
8) 38
(3 3.
3) 44
(3 7.
6) 38
(3 3.
3) .7
3
F ri
en ds
( ≥
6 =
y es
)a 10
1 (2
9. 3)
33 (2
9. 0)
41 (3
5. 0)
27 (2
3. 7)
.1 7
In st
ru m
en ta
l co
pi ng
( hi
gh )b
11 2
(3 2.
5) 42
(3 6.
8) 40
(3 4.
2) 30
(2 6.
3) .2
1
R el
ig io
us c
op in
g (h
ig h)
b 13
8 (4
0. 0)
45 (3
9. 5)
55 (4
7. 0)
38 (3
3. 3)
.1 0
Ju ve
ni le
h al
l (a
ny t
im e)
19 2
(5 5.
7) 67
(5 8.
8) 69
(5 9.
0) 56
(4 9.
2) .2
3
S it
ua ti
on al
D is
ch ar
ge d
fr om
j ai
l 19
3 (5
5. 9)
59 (5
2. 2)
64 (5
4. 7)
70 (6
1. 4)
.3 5
D is
ch ar
ge d
fr om
p ri
so n
15 1
(4 3.
8) 54
(4 7.
8) 53
(4 5.
3) 44
(3 8.
6)
C on
tr ac
t .4
9
IC D
T P
10 5
(3 0.
4) 32
(2 8.
1) 31
(2 6.
5) 42
(3 6.
8)
R M
S C
20 3
(5 8.
8) 71
(6 2.
3) 72
(6 1.
5) 60
(5 2.
6)
S A
P 35
(1 0.
1) 11
(9 .7
) 12
(1 0.
3) 12
(1 0.
5)
P re
/p os
t re
al ig
nm en
t (y
es )
15 8
(4 5.
8) 52
(4 5.
6) 55
(4 7.
0) 51
(4 4.
7) .9
4
R D
T t
im e
(d ay
s) .6
7
1– 49
86 (2
4. 9)
30 (2
6. 3)
34 (2
9. 1)
22 (1
9. 3)
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
Nyamathi et al. Page 21
A ll
(N =
3 45
) P
C -N
C M
(n =
1 14
) P
C (n
= 1
20 )
U C
(n =
1 11
)
T yp
e C
h ar
ac te
ri st
ic M
(S D
) M
(S D
) M
(S D
) M
(S D
) p
50 –8
9 93
(2 7.
0) 27
(2 3.
7) 31
(2 6.
5) 35
(3 0.
7)
90 –1
78 67
(1 9.
4) 23
(2 0.
2) 20
(1 7.
1) 24
(2 1.
1)
≥ 17
9 99
(2 8.
7) 34
(2 9.
8) 32
(2 7.
4) 33
(2 9.
0)
C op
in g
A lc
oh ol
t re
at m
en t
(a ny
) 10
3 (2
9. 9)
31 (2
7. 2)
38 (3
2. 5)
34 (2
9. 8)
.6 8
D ru
g tr
ea tm
en t
(a ny
) 29
0 (8
4. 1)
94 (8
2. 5)
10 1
(8 6.
3) 95
(8 3.
3) .7
0
C ra
ck u
se (
re ce
nt )c
15 3
(4 4.
4) 44
(3 8.
6) 54
(4 6.
2) 55
(4 8.
3) .3
0
C oc
ai ne
u se
( re
ce nt
)c 94
(2 7.
3) 26
(2 2.
8) 35
(2 9.
9) 33
(2 9.
0) .4
2
B in
ge d
ri nk
in g
(r ec
en t)
c 13
2 (3
8. 3)
43 (3
7. 7)
48 (4
1. 0)
41 (3
6. 0)
.7 2
P er
so na
l H
ea lt
h (f
ai r/
po or
) 99
(2 8.
7) 34
(2 9.
8) 23
(2 0.
0) 42
(3 7.
2) .0
1
H os
ti li
ty (
hi gh
)a 67
(1 9.
4) 28
(2 4.
6) 21
(1 8.
0) 18
(1 5.
8) .2
2
In je
ct io
n dr
ug u
se (
ev er
) 11
2 (3
2. 5)
33 (2
9. 0)
33 (2
8. 2)
46 (4
0. 4)
.0 9
M et
ha m
ph et
am in
e us
e (e
ve r)
17 1
(4 9.
6) 61
(5 4.
0) 53
(4 5.
7) 57
(5 0.
4) .4
5
P sy
ch ia
tr ic
h os
pi ta
li za
ti on
( ev
er )
63 (1
8. 3)
18 (1
5. 8)
27 (2
3. 1)
18 (1
5. 8)
.2 5
H IV
( po
si ti
ve )
7 (2
.0 )
0 (0
) 4
(3 .9
) 3
(3 .2
) .1
5
H C
V (
ye s)
97 (2
8. 1)
32 (8
.1 )
30 (2
5. 6)
35 (3
0. 7)
.6 9
M (S
D )
M (S
D )
M S
D M
S D
C E
S -D
( to
ta l)
20 .8
(1 4.
2) 9.
0 (6
.6 )
8. 7
(5 .4
) 9.
2 (6
.5 )
.8 5
P os
it iv
e so
ci al
s up
po rt
24 .2
(1 4.
3) 10
.5 (9
.6 )
10 .5
(3 .6
) 9.
7 (3
.6 )
.1 2
N o te
. N =
3 45
. C E
S -D
= C
en te
r fo
r E
pi de
m io
lo gi
ca l
S tu
di es
-D ep
re ss
io n;
H C
V =
h ep
at it
is v
ir us
; H
IV =
h um
an i
m m
un od
ef ic
ie nc
y vi
ru s;
I C
D T
P =
i n
cu st
od y
dr ug
t re
at m
en t
pr og
ra m
; P
C =
p ee
r co
ac hi
ng ;
P C
-N C
M =
p ee
r co
ac hi
ng -n
ur si
ng c
as e
m an
ag em
en t;
R D
T =
r es
id en
ti al
d ru
g tr
ea tm
en t;
R M
S C
= r
es id
en ti
al m
ul ti
se rv
ic e
se rv
ic e
ce nt
er ;
S A
P =
s ub
st an
ce a
bu se
p ro
gr am
; S D
= s
ta nd
ar d
de vi
at io
n; U
C =
us
ua l
ca re
.
a U
pp er
q ua
rt il
e.
b S
co re
a bo
ve m
ed ia
n.
c W it
hi n
6 m
on th
s pr
io r
to m
os t
re ce
nt i
nc ar
ce ra
ti on
.
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
Nyamathi et al. Page 22
T A
B L
E 2
A ss
oc ia
ti on
s B
et w
ee n
H A
V /H
B V
V ac
ci ne
C om
pl et
io n
S ta
tu s
an d
S el
ec te
d V
ar ia
bl es
N on
co m
p le
te rs
(n =
9 3)
C om
p le
te rs
(n =
2 52
)
T yp
e C
h ar
ac te
ri st
ic M
(S D
) M
(S D
) p
D em
og ra
ph ic
A ge
40 .8
(9 .9
) 42
.5 (9
.3 )
.1 3
E du
ca ti
on 11
.4 (1
.4 )
11 .7
(1 .4
) .0
9
n (%
) n
(% )
R ac
e/ E
th ni
ci ty
.1 6
A fr
ic an
-A m
er ic
an 39
(4 1.
9) 13
6 (5
4. 0)
L at
in o
33 (3
5. 5)
74 (2
9. 4)
W hi
te 14
(1 5.
1) 33
(1 3.
1)
A si
an /P
ac if
ic I
sl an
de r
7 (7
.5 )
9 (3
.6 )
In te
rv en
ti on
P ee
r co
ac h-
nu rs
e ca
se m
an ag
em en
t 28
(3 0.
1) 86
(3 4.
1) .7
8
P ee
r co
ac h
33 (3
5. 5)
84 (3
3. 3)
U su
al c
ar e
32 (3
4. 4)
82 (3
2. 5)
S oc
ia l
P ar
tn er
s (≥
2 o
r <
2 )
46 (4
9. 5)
15 3
(6 0.
7) .0
6
R em
ov ed
f ro
m p
ar en
ts (
ye s
or n
o) 59
(6 3.
4) 13
0 (5
1. 6)
.0 5
Ju ve
ni le
h al
l (a
ny t
im e
or n
ev er
) 60
(6 4.
5) 13
2 (5
2. 4)
.0 4
F ri
en ds
( ≥
6 o
r <
6 )
18 (1
9. 4)
83 (3
2. 9)
.0 1
In st
ru m
en ta
l co
pi ng
( hi
gh o
r lo
w )
19 (2
0. 4)
88 (3
4. 9)
.0 1
R el
ig io
us c
op in
g (h
ig h
or l
ow )
27 (2
9. 0)
10 1
(4 0.
1) .0
6
S it
ua ti
on al
D is
ch ar
ge d
fr om
j ai
l 59
(6 3.
4) 13
3 (5
3. 2)
.0 9
D is
ch ar
ge d
fr om
p ri
so n
34 (2
2. 5)
11 7
(7 7.
5)
C on
tr ac
t T
yp e
.3 8
IC D
T P
30 (3
2. 3)
76 (3
0. 4)
R M
S C
57 (6
1. 3)
14 5
(5 8.
0)
S A
P 6
(6 .5
) 29
(1 1.
6)
P os
t re
al ig
nm en
ta 60
(6 4.
5) 97
(3 8.
8) .0
01
R D
T T
im e
≥ 90
b 10
(1 0.
8) 15
5 (6
2. 0)
.0 01
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
Nyamathi et al. Page 23
N on
co m
p le
te rs
(n =
9 3)
C om
p le
te rs
(n =
2 52
)
T yp
e C
h ar
ac te
ri st
ic M
(S D
) M
(S D
) p
H C
V (
po si
ti ve
) 34
(3 6.
6) 63
(2 5.
2) .0
3
H IV
( po
si ti
ve )c
1 (1
.2 )
6 (2
.8 )
.6 8d
C op
in g
A lc
oh ol
t re
at m
en t
(a ny
o r
no ne
) 19
(2 0.
4) 84
(3 3.
3) .0
2
D ru
g tr
ea tm
en t
(a ny
o r
no ne
) 80
(8 6.
0) 21
0 (8
3. 3)
.5 5
C ra
ck u
se (
re ce
nt o
r no
t) 34
(3 6.
6) 11
9 (4
7. 2)
.0 8
C oc
ai ne
u se
( re
ce nt
o r
no t)
15 (1
6. 1)
79 (3
1. 4)
.0 05
B in
ge d
ri nk
in g
(r ec
en t
or n
ot )
41 (4
4. 1)
91 (3
6. 1)
.1 8
P er
so na
l H
os ti
li ty
( hi
gh )
27 (2
9. 0)
51 (2
0. 2)
.0 8
In je
ct io
n dr
ug u
se (
ev er
o r
ne ve
r) 39
(4 1.
9) 73
(2 9.
0) .0
2
M et
ha m
ph et
am in
e us
e (e
ve r
or n
ev er
) 51
(5 4.
8) 12
0 (4
8. 2)
.2 7
P sy
ch ia
tr ic
h os
pi ta
li za
ti on
( ye
s) 25
(2 6.
9) 38
(1 5.
1) .0
1
M (S
D )
M (S
D )
C E
S -D
( to
ta l)
9. 87
(6 .4
) 8.
62 (6
.1 )
.0 9
P os
it iv
e so
ci al
s up
po rt
10 .8
2 (3
.3 )
10 .0
1 (3
.6 )
.0 6
N o te
. H C
V =
h ep
at it
is v
ir us
; H
IV =
h um
an i
m m
un od
ef ic
ie nc
y vi
ru s;
I C
D T
P =
i n-
cu st
od y
dr ug
t re
at m
en t
pr og
ra m
; R
D T
= r
es id
en ti
al d
ru g
tr ea
tm en
t; R
M S
C =
r es
id en
ti al
m ul
ti -s
er vi
ce c
en te
r; S
A P
=
su bs
ta nc
e ab
us e
pr og
ra m
; S D
= s
ta nd
ar d
de vi
at io
n.
a O
ct ob
er 1
, 2 01
1.
b T
im e
in R
D T
p ro
gr am
( da
ys ).
c B as
ed o
n 29
8 m
en .
d F
is he
r’ s
ex ac
t te
st .
Nurs Res. Author manuscript; available in PMC 2016 May 01.
A u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t A
u th
o r M
a n u scrip
t
Nyamathi et al. Page 24
TABLE 3
Logistic Regression Model for Non-Completion of HAV/HBV Vaccine Series
Type Predictor Adjusted OR 95% CI p
Interventiona PC-NCM 0.59 [0.27, 1.28] .18
PC 0.83 [0.39, 1.76] .63
Demographics Raceb
African American 1.80 [0.68, 4.78] .24
Latino 2.33 [0.87, 6.21] .09
Asian/Pacific Islander 5.86 [1.23, 27.92] .03
Social Friends (≥ 6 = yes) 0.46 [0.22, 0.95] .04
Situational Post realignment (yes) 2.21 [1.19, 4.09] .01
RDT stay (at least 90 days) 0.06 [0.03, 0.13] .001
Coping Alcohol treatment (any) 0.50 [0.24, 1.03] .06
Cocaine use (any) 0.34 [0.16, 0.73] .006
Personal Hostility (high) 2.24 [1.06, 4.73] .04
Injection drug use (ever) 2.19 [1.07, 4.47] .03
Psychiatric hospitalization (any) 2.58 [1.22, 5.46] .01
Positive social support (yes) 1.10 [1.00, 1.21] .04
Note. N = 345. CI = confidence interval; OR = odds ratio; PC = peer coaching; PC-NCM = peer coaching-nursing case management; RDT = residential drug treatment.
a Reference class is usual care.
b Reference class is White.
Nurs Res. Author manuscript; available in PMC 2016 May 01.