Child Abuse

Journal ofFamily Violence, Vol. 21, No . 3, April 2006 (© 2006) DOI: 10.1007/s10896-006-9019-l

The Effects of Childhood Physical Abuse or Childhood Sexual Abuse in Battered Women's Coping Mechanisms: Obsessive-Compulsive Tendencies and Severe Depression

Debra K. Miller1,2

Published online: 31 August 2006

What role does childhood abuse have on the coping choices made by a battered woman? Ancillary to a depression study (Bailey, 1996) in 79 battered women from a Houston area women's shelter were compared for past abuse experiences and how the women were coping with abuse in adulthood. This study compared coping styles between two groups of battered women: those who experienced childhood physically abuse (CPA) (n = 35), and those who did not experience childhood physically abuse (NCPA) in childhood (n = 44). All of the women filled out a battery of questionnaires in­ cluding The BriefSymptom Inventory (BS/), and a scale for learned helplessness. At-test conducted on obsessive-compulsive tendencies (OCT) scale of the BSI found that women who were NPPA had significantly lower BSI-OCT scores t(77) = - 2.05, p < .05 than women who were PPA. No statistically significant differences were found between groups for learned helplessness. Out of the 35 battered women who reported physical abuse in childhood were more likely to report sexual abuse as girls than battered women who were not physically abused, t(77) = - 3.40, p < .001. Battered women who had been physically and sexually abused in childhood were more severely depressed. Battered women who were not abused in childhood had more obsessive compulsive tendencies. The ramifications of these findings for therapeutic treatment are discussed.

KEY WORDS: battered women; coping mechanisms; learned helplessness; obsessive-compulsive tendencies.

Traditionally, the principal characteristics associated with battered women have been learned helplessness and posttraumatic stress disorder (PTSD). The battered woman syndrome was a term used by Lenore Walker for the assortment of symptoms endured by battered women--consisting primarily of learned helplessness and PTSD (Walker, 1984). Her theory was that battered women, parallel to the dogs in Seligman's (1 974) classic investigation, have to relearn the feeling of having control and that they have the ability to escape unbearable cir­ cumstances. Walker's hypothesis of learned helplessness as an explanation of why battered women stay in an abusive home was not supported by statistical analysis.

Seligman (1974) was first to find empirical support for learned helplessness. In his original research, dogs

1University of Houston, Houston, Texas. 2To whom correspondence should be addressed at 1452 Waseca, Hous­ ton, TX 77055; e-mail: [email protected].

185

were placed into cages-the dogs that were studied were unable to escape mild electrical shocks. The dogs became unmotivated to search for escape. When the opportunity for escape was presented the animals did not immediately attempt escape or take advantage of the opportunity. If learned helplessness was a result of battering the indi­ cators of susceptibility, according to the battered woman syndrome, included seven factors that indicate suscep­ tibility to becoming a battered woman: social learning (gender roles taught to children), learned helplessness in childhood, early and repeated sexual molestation, phys­ ical assault, a high level of family of origin violence, critical events over which the child has no control, and those who are at high risk for depression. Hotaling and Sugarman (1986) performed a meta-analysis of risk mark­ ers for women who would likely experience battering in adult relationships. The most prevalent risk marker was witnessing violence in their family of origin as children. This study looked at the antecedent factors of childhood

0885-7482/06/0400-0185/0 © 2006 Springer Science+Business Media, Inc.

186

physical abuse and childhood sexual abuse as determi­ nants of coping styles in battered women. Does a woman who experienced physical abuse in childhood exhibit pre­ dictable reactions to abuse in adulthood? Does sexual abuse in childhood have a predictable influence on the coping choices made by a battered woman?

Recent research has shown multiple reactions to bat­ tering. Talbot et al. (2000) measured personality traits in adult women who experienced incidence of sexual abuse in childhood-the women were much more intro­ verted and much less open to new experiences than other women. Sexually abused girls have a tendency to become re-victimized as battered or sexually assaulted women (Bleiberg, 2000). Incest survivors have a higher preva­ lence of dissociation and greater general distress (Brown, 1998). The women in Brown's study also had increased body image distortion. Barber (1998) noted a frequent oc­ currence ofdissociation in incest victims and children who experienced traumatic events. The impact of sexual abuse dominated the list of antecedent events resulting in greater use of dissociation. Other antecedent variables include loss, greater frequency of traumatic events, and younger age of onset. Depression and anxiety are pervasive in women who endured sexual abuse during childhood. And they have longer duration of depressive symptoms and greater frequency of suicide attempts.

Survivors of incest are often diagnosed with post­ traumatic stress disorder. One relevant study by Courtois (2000) supports the presence of"complex PTSD"-a pro­ posed diagnostic spectrum that includes long-term reper­ cussions of severe childhood sexual abuse. The diagno­ sis encompasses dissociation along with other disorders and a deficit of normal developmental skills. Kessler and Bieschke (1999) examined the relationship between incest survivors and adult victimization. They found that shame was a significant predictor of those who were abused in childhood as well as adulthood. Smith (1999) disclosed four major characteristics of women with a history of childhood sexual abuse: nightmares, hopelessness, feeling "like a failure", and hypersexual behavior. Mothers who experienced sexual abuse in their own childhood tended to have delayed reactive symptoms when they discovered sexual abuse in their own child (Green et al., 1995). The symptoms included acute schizophrenic reaction, depres­ sion, insomnia, anorexia, or panic attacks. In addition, the symptoms included reliving their own experience­ described in diagnostic criteria of PTSD as acting or feel­ ing as if the traumatic event was reoccurring (flashback episode, hallucinations).

Adults who disclose abuse in their childhood tend to be socially stigmatized by acquaintances as well as their immediate family. Tunick (2000) substantiated this fact

Miller

in an exploration of adult survivors of incest and phys­ ical abuse. Abuse survivors described past abuse, which was reviewed by 1375 professionals-including psychia­ trists, psychologists, undergraduate psychology students, and social workers. The incest survivors were significantly stigmatized over the physical abuse group and the control group-even among psychological professionals.

The emotional coping skills used during childhood by those who were sexually abused in childhood were withdrawal, distraction, or dismissing-avoidant attach­ ment behaviors. They also reported elevated levels of fear, sadness, anger, guilt, and shame. Depression and anxiety were related to lower perceived support from nonoffend­ ing parents in sexually abused girls (aged 11-18 years) as noted by Spaccarelli and Fuchs (1997). The girls relied on cognitive avoidance coping-tending to deal with social problems by self-distraction.

Gleason (1993) tested 62 battered women for per­ sonality characteristics. He found that 99% of the battered women had anxiety symptoms: obsessive-compulsive disorder, generalized anxiety disorder, PTSD, major depression, and substance abuse. Obsessive-compulsive tendencies were prevalent in the group of battered women who were living in the home with their assailant.

ACTIVE OR PASSIVE COPING MECHANISM

How a woman reacts to battering varies depending on her cognitive schema and her emotional development. These two elements are profoundly influenced by abuse in childhood and by what type of abuse she endured. Some battered women may try to do everything they can to forestall a battering incident; conversely, some battered women may feel that there is nothing within their power that she can do to stop the battering.

The Diagnostic and Statistical Manual of Mental Disorders (American Psychiatric Association, 2000; DSM-IV-TR) outlined defense mechanisms stating that defensive functioning deals with internal or external stressors by action or withdrawal. To differentiate between active and passive coping mechanisms we can define coping as physical or mental effort put forth to effect a change in circumstances or environment in one's own favor. Examples of active coping mechanisms are acting out, altruism, devaluation, displacement, reaction formations, and self-assertion. Passive (withdrawal) coping mechanisms are exemplified by little or no effort put forth to effect a favorable change--efforts might only be toward escape or denial. Examples of passive means of coping are anticipation, denial, dissociation, repression, and suppression. Obsessive-compulsive tendencies and

Obsessive-Compulsive Tendencies in Battered Women

learned helplessness can be understood in the framework of active or passive coping mechanisms.

Obsessive-compulsive tendencies are symptomatic expression of active coping means. They are ritualized behaviors that reduce anxiety and impact circumstances in a beneficial way (hopefully lessening the chances of a bat­ tering incident). Obsessive-compulsive tendencies (OCT) are differentiated from obsessive-compulsive disorder and obsessive-compulsive personality disorder by the reduced pathology associated with a tendency versus a disorder. Obsessive-compulsive tendencies are coping responses to stress--characterized by an impulse to take action to re­ duce stress. Obsessive-compulsive tendencies are directed toward a specific purpose-that of alleviating a distressing situation or avoiding a distressing situation in the near fu­ ture. Dumont (1996) describes the extreme reaction many obsessive-compulsive disorder sufferers have to feelings of anxiety. Feelings of panic invoke worries of having a heart attack in obsessive-compulsive sufferers, who have a strong belief in the danger of their racing heart, tightness in the chest, and shortness of breath indicates the onset of imminent danger. Obsessive-compulsive disorders are also characterized by magical thinking-the performance of ritual to extinguish threatening thoughts and feelings. The belief is that if the ritual is performed correctly, then the source of their stress will be removed or altered. To insure this result, someone with an obsessive-compulsive disorder must repeat the ritual over and over again. Very often she will avoid situations that create anxious feelings and the subsequent compulsion to perform rituals. There is some awareness and embarrassment over the illogical thinking that fuels compulsions, but she is not able to re­ sist following through with the ritual that will be rewarded by alleviating anxiety. In a battered women, however, the rewards can be very real-the prevention of a battering incident or even saving her life.

Belief that one's actions can prevent a negative event proves to be an insidious means of living life. Preven­ tion calls for supreme diligence. If the battered woman feels she can reliably predict an outcome to her obsessive­ compulsive action then she feels more confident in per­ forming a preventative action. For example, if she knows that if the batterer consumes alcohol, then there is a high probability that the evening will end in a battering inci­ dent. The prevention of consuming alcohol is a concrete and reasonable preventative action to take-a reasonable active coping mechanism. If, on the other hand, she can never predict whether a particular circumstance or event will cause the batterer to get angry, such as the possibility that dust on the furniture will initiate a battering incident, then she has to continually dust the furniture to prevent a battering incident. Escalation into formidable levels of

187

cleanliness is inevitable. Ironically no matter what efforts she puts forth the batterer will eventually get angered-if not about dust, then angry about something else. Realiza­ tion that the dust does not having anything to do with the level of her batterer's anger is hard to come by. If there is even a small chance that removal of the dust will prevent pain and injury, she will be compelled to clean.

Case Study: An Obsessive-Compulsive Response to a Battered Woman's Anxiety

Examining data from an extreme case study can help highlight what otherwise might be overlooked in a multi­ symptomatic case or a patient that has a milder form of a symptom or coping mechanism. In Walker's book Terrify­ ing Love (Walker, 1989); she cites a case study of a woman called "Agnes." Agnes married a man called "Charlie" while still in her teens. During the 30 years they were married, what began as occasional psychological abuse by Charlie increased in frequency and intensity. Charlie's threats became more frequent and more vicious-leading to murderous threats and proportionately increased anx­ iety in Agnes. Eventually Agnes developed a belief that Charlie would kill her by poisoning her food or drink. In response to her belief, Agnes developed rituals around preparing and serving food, she became afraid of germs, and began to wear gloves to keep from touching anything that Charlie had touched. Her vigilance for poisoning spread to wearing a surgical mask in the house for fear of poisonous fumes spread through the ventilation system. Another protective measure Agnes took was to hide a gun in the kitchen in the eventuality that Charlie attempted to carry through with his threats to kill her. Agnes was un­ able to concentrate on driving and, therefore, discontinued excursions out of the home. Soon the only place Agnes felt safe was in her own room. Eventually that failed as a refuge because Agnes did not know what Charlie was doing while she was locked in her room.

Agnes developed obsessive-compulsive tendencies, as evidenced by her ritualistic behaviors of wearing masks and gloves, attention to food handling, and isolation. These tendencies were specific to her relationship to the batterer, and perceived as a means of coping with his threats to kill her. During an argument one day, Agnes felt her life was threatened and retrieved the gun from its hid­ ing place. Charlie attempted to grab the gun from Agnes ' hands and, in the ensuing struggle; the gun went off killing Charlie. After Charlie's death, Agnes' obsessive­ compulsive tendencies abated. A court ordered psychia­ trist examined Agnes and determined that she no longer had signs of obsessive-compulsive behaviors and ruled out psychosis.

188

Walker mentioned obsessive-compulsive tendencies (OCT), but did not include the tendency in her description of the battered woman syndrome. Walker stated that the battered woman, " .. . may appear to be passive [LH], but the truth is that she can be strong, often manipulating the people and objects in her environment extensively [OCT], at least enough to avoid being killed." (Walker, 1989) This is a description of two different tendencies; the woman is usually either predominantly passive as in learned help­ lessness, or more active as shown in obsessive-compulsive tendencies.

Walker (1984) also found that battered women have a higher than normal locus of control, indicating that bat­ tered women feel they have control over their own actions and life. Brockner and Rubin (1985) report that internal locus of control is a significant variable in the search for entrapment-prone personality (analogous to passive cop­ ing or learned helplessness). Entrapment was defined as the extent to which someone perceives her investment in an undesirable situation. A decision maker might persist in a failing course of action in order to justify prior in­ vestment or commitments, even if she is full of conflict about the decision. Brockner and Rubin found a correla­ tion between locus of control and persistence behaviors. Those who had greater internal locus of control scores re­ ported more feelings ofentrapment than those with greater external locus of control scores. Internal locus of control is the idea that one has the ability to take responsibility for their own actions and have an effect on the environment, the situation, or other people in order to produce a desir­ able outcome for oneself. Success of failure is generally attributed to internal sources. External locus of control is the notion that the environment, the situation, or other people determine your destiny. Success or failure is gen­ erally attributed to external sources. The women who had greater perception of ability to change their environment, their situation, or their batterer also had more feelings of entrapment. They were motivated to make changes and felt it was within their power to effect a change for the better. Internal locus of control-or the belief a woman can effect positive change in her life-has similar char­ acteristics of obsessive-compulsive tendencies. Someone with internal locus of control believes they have an ability to make positive changes in their life. Brockner and Rubin (1985) found that those who exhibited entrapment tenden­ cies consistently reported "concentrating on the task to the exclusion of any other thoughts."

For the purposes of this study "Learned Helpless­ ness" is defined as the tendency to attribute events as (1) personalized internally: i.e., explains adverse events have internal causes; good events have external causes; depen­ dence on the help of others to accomplish something are

Miller

due to her personal deficits; (2) universally pervasive: be­ lieves that she will fail in all areas ofher life when she fails in one area of her life, "I failed the exam, and it's no use trying to pass any exam;" does not consider the problem may be specific to the situation; and (3) permanent: thinks of things in terms of always and never; "I'll never amount to anything;" instead of looking at the possibility it may be a temporary situation such as ''I'm just exhausted right now" (Seligman, 1974).

The operational definition of obsessive-compulsive tendencies in this study use the same definition used in the Brief Symptom Inventory questionnaire (Derogatis & Melisaratos, 1983). The definition states, "The focus of this dimension is on thoughts and actions that are expe­ rienced as umemitting and irresistible by the patients but are of an ego-alien nature."

Hypotheses

Statistical analysis of the women in this sample will compare coping mechanisms of two categories of battered women: (1) Childhood Physical Abuse (CPA), and (2) No Childhood Physical Abuse (NCPA). It was predicted that there is a dichotomous nature between passive and active coping styles. Abuse in childhood indicates abuse by a primary caregiver.

Hypothesis 1: Battered women who experienced childhood physical abused (CPA) will have a more pro­ nounced tendency toward passive coping, e.g. learned helplessness, as compared to those who were not pre­ viously physically abused (NCPA).

Hypothesis 2: Battered women who were not physi­ cally abused (NCPA) will exhibit an active coping mech­ anism, obsessive-compulsive tendencies, as compared to battered women who were previously physically abused by their family of origin (CPA).

METHODS

Participants

Seventy-nine battered women from the Houston Area Women's Center volunteered to participate in an investi­ gational study of the effectiveness of a self-management therapy for depression between the years 1992-1996 (Bailey, 1996). Some of the women were still living with their batterer and seeking council through the shelter and some had left their batterer and were living in a battered women's shelter. The women in this study identified them­ selves as either physically abused in childhood or not

Obsessive-Compulsive Tendencies in Battered Women

physically abused in childhood on a demographic ques­ tionnaire filled out at their first session. The participants signed informed consent forms at the outset of data col­ lection and depression treatment. The data reported herein is an extension of that study and conforms to the terms of consent.

Thirty-five women identified themselves as hav­ ing experienced physical abused as children. Forty-four women identified themselves as not having experienced physical abuse as children. During on their initial visit, participants were given a battery of demographic forms and questionnaires measuring mood, attitudes, social ad­ justment, and self-control qualities as described below. The women were asked to use their subjective understand­ ing of the definition of terms requested on the forms-no operational definitions were given to them.

Measures

The Brief Symptom Inventory (Derogatis & Spencer, 1982; Derogatis & Melisaratos, 1983) is a self-report mea­ sure developed to assess clinical patient presenting symp­ toms. Questions fall within nine symptom categories: Somatization, obsessive-compulsive tendencies, interper­ sonal sensitivity, depression, anxiety, hostility, phobic anxiety, and paranoid ideation, and psychoticisim.

The Dysfunctional Attitude Scale (Dobson & Shaw, 1986; Oliver & Baumgart, 1987) contains forty statements that were developed to measure the attitudes that exacer­ bate depression based on research on clinically depressed individuals.

The Social Adjustment Scale (Weissman & Bothwell, 1976) was designed to measure social adjust­ ment in areas including work, social activity, leisure time, family life, intimate relationships, and finances . The scale includes 54 questions rated on a Likert-type scale.

The Self-Control Schedule/ The Self-Control Ques­ tionnaire (Rehm et al., 1984) consists of forty questions designed to assess self-control behaviors and depressive cognitions. The measure was designed to assess the effec­ tiveness of Rehm's Self-Management Therapy program for depression.

Procedure

Each participant answered a battery of question­ naires at the outset of treatment-each item on each ques­ tionnaire was evaluated for summation onto a singular scale for LH and a singular scale for OCT. One of the measures-the DAS-has a scale of 1-7 whereas the oth­ ers all have Likert-type scales from 1-5, therefore the DAS was converted to a 1-5 scale.

189

The Miller Scale for Learned Helplessness is a compilation of questions from other measures. Symp­ toms of learned helplessness were compared to Martin Seligman's description of the syndrome, whose focus was on cognitive attributions of the participant's subjective ex­ perience(s) (Seligman, 1974). Ten items that theoretically loaded onto the project designed learned helplessness con­ struct were summed into a learned helplessness scale. For example, one cognitive aspect indicative of learned help­ lessness is a feeling of inability to change one's situation. The evidence of that tendency was measured in this study by looking at the SCS/SCQ questions like question num­ ber 49: "I depend heavily on other people's opinions to evaluate objectively what I do" and number 56 "It's no use trying to change most of the things that make me miserable."

There is not a separate definition of obsessive­ compulsive tendencies, as differentiated from obsessive­ compulsive disorder and obsessive-compulsive personal­ ity disorder, in the Diagnostic and Statistical Manual for Mental Disorders. For purposes of this study, the Brief Symptom Inventory definition of obsessive-compulsive tendencies was used. The BSI identified OCT by focus­ ing on cognitions and actions that were experienced as unremitting and irresistible by the patients, and of an ego­ alien nature. The BSI-OCT scale consists of six items in­ dicating obsessive-compulsive tendencies. Ten items that theoretically loaded onto the project designed obsessive­ compulsive tendency construct were summed into the Miller Obsessive-Compulsive Tendency scale.

RESULTS

A demographic analysis was performed based on forms each participant filled out during their first treat­ ment session. The forms requested information about the participant, her batterer, and her family of origin-if she had witnessed abuse between her parents and if she had experienced emotional, physical, or sexual abused by her primary caregivers as a child (see Table I). Each group was reviewed for statistical differences in demographic data such as the age upon entry into the program, the number of children she had, and the number of battering relationships she had been in.

Seventy-four percent of the 35 women who identi­ fied themselves as physically abused in childhood (CPA) reported they were also emotionally abused and 40% of this group reported they were sexually abused in child­ hood. Fifty-one percent witnessed inter-parental emo­ tional abuse and 54% witnessed interparental physical abuse. Fifty-two percent of the women who reported no

190

Table I. Type of Abuse in Childhood

CPA NCPA Two-tailed

Mean SD Mean SD 1(77) significance

Emotional 1.43 0.88 1.30 1.27 -0.53 .60ns Sexual 0.71 0.83 0.23 0.42 -3.40 .01

Note. Of the 79 participants, 22 gave no answer as to whether they had experienced childhood emotionally abused and two gave no answer as to whether they had experienced childhood sexual abuse. Women who experienced previous physical abused also had significantly higher scores for childhood sexual abuse.

physical abuse in childhood (NCPA) reported emotional abuse in childhood. Seventy-seven percent of the women in the NCPA group said they were not sexually abused in childhood. In the NCPA group 41 % did not answer the question if they had witnessed inter-parental emotional abuse, 36% said they had not witnessed interparental phys­ ical abuse.

Women in the CPA group sought help later in life than those who were not physically abused as children, t(77) = .55, p < .05. It took longer for the women who were previously physically abused to seek treatment, on average a full decade longer. Twenty-three percent of NCPA sought help between 31 and 35 years of age, whereas 29% CPA women sought help between 41 and 45 years of age. The youngest in the CPA group is 25 and the oldest is 59. The youngest in the NCPA group was 20 and the oldest 64.

The women who were physically abused as children (CPA) had less income than the women who were not (NCPA) t(77) = - .70, p < .05. The mean difference be­ tween groups was $97 per month. Battered women in the CPA group also had more children, t(77) = .64, p < .05, number= 21, than the NCPA group who had 17 children.

More of the CPA group were still in the battering relationship, t(77) = - 2.32, p < .05. All of the partici­ pants said they were currently being emotionally abused. Of the women who were physically abused as children (CPA), 83% had been in multiple abusive relationships, of the women in the NCPA group, 55% had been in more than one abusive relationship as adults.

-Test Results

Two t-tests were conducted. The independent vari­ able was CPA (n = 35) and NCPA (n = 44) and the de­ pendant variables were either Learned Helplessness or Obsessive-Compulsive tendencies. Two separate t-tests were conducted on Obsessive-Compulsive Tendencies­ one for the Miller Scale and the other for the Brief

Miller

Table II. I-Test Results-Physical Abuse

CPA NCPA p-Value

Mean SD Mean SD 1(77) (two-tailed)

MillerLH 27.09 6.98 27.39 7.88 0.18 .86ns Miller OCT 26.23 5.27 28.68 6.37 1.83 .07 BSIOCT 2.61 0.91 2.19 0.94 -2.05* .04

Note. Brief Symptom Inventory (BSI); Obsessive-Compulsive Tenden­ cies subscale. *p < .05. p < .10.

Symptom Inventory OCT scale. The analysis for the BSI-OCT scale found significant differences between groups, t(77) = - 2.05, p < .05 (see Table II). The main effect for obsessive-compulsive tendencies (as described in the Miller Scale for Obsessive-Compulsive Tendencies) are not significant: however, there is a trend toward signif­ icance at t(77) = 1. 83, p < . 08. Frequencies and univariate statistics were investigated for outlying subject scores on the t-test OCT scores. A normal distribution was found.

No significance was found for learned helplessness in t-tests or a separate AN OVA analysis. A check of the Beck Depression Inventory (BDI) scores [analogous to learned helplessness] against the two subject variable groups was run, as well as the BSI-OCT scores for each group. The main effect of BDI depression scores was not significant.

Since there was a high incidence of sexual abuse among those who were physically abused in childhood t(77) = .83, p < 0.01, another statistical analysis was per­ formed separating the battered women were separated into groups of women who identified themselves as experienc­ ing childhood sexual abuse and those who said they did not experience childhood sexual abuse.

Hypotheses

Statistical analysis of two new categories of the bat­ tered women in this study anticipates that there will be more prevalence of learned helplessness (a passive cop­ ing style) and depression among women who were sex­ ually abused by family members and those who were not sexually abused will have more obsessive-compulsive tendencies (an active coping style). The two categories are defined as (1) Previously Sexually Abused (PSA), and (2) No Childhood Sexual Abuse (NCSA). Sexual abuse in childhood indicates abuse by a primary caregiver.

Hypothesis 1: Battered women who were sexually abused in childhood (CSA) will have a more pronounced tendency toward passive coping, e.g. learned helplessness

†

†

t

Obsessive-Compulsive Tendencies in Battered Women

and depression, as compared to those who were not sexu­ ally abused by their family of origin (NCSA).

Hypothesis 2: Battered women who were not sexu­ ally abused (NCSA) in their family of origin will exhibit an active coping mechanism, obsessive-compulsive tenden­ cies, as compared to battered women who were sexually abused in childhood (CSA).

Results

For the most part, the two categories of battered women (CSA or NCSA) had the same demographic profile as CPA-NCPA, but with a few notable exceptions. Ninety­ eight percent of battered women were sexually abused as children were also physically assaulted as children. Bat­ tered women who were sexually abused as children also had decreased earnings as compared to battered women who were not sexually assaulted as children, p < .04. The mean income level for CSA was $620, whereas the mean income level for NCSA was $1200.

-Test Results

Statistical results comparing those who were and those who were not sexually abused as children showed significant differences in severity of depression scores, t(73) = - 1.99, p < .05. The independent variable was CSA (n = 25) and NCSA (n = 50), and the dependent vari­ able was learned helplessness, obsessive-compulsive ten­ dency, or depression. Two scales measured OCT: the BSI­ OCT scale and the Miller Scale for Obsessive-Compulsive Tendencies. There was no main effect of obsessive­ compulsive tendencies found in these groups. The main effect for learned helplessness was nonsignificant, how­ ever a trend was detected at p < .10 ( see Table III).

A comparison was made between the previous study on the differences between those who were physically abused or not physically abused and this study comparing those who had been sexually abused or not in childhood. At the outset, it was hypothesized that prior abuse his­ tory had an effect on the means used by adult women in battering relationships. The hypothesis was that physical abuse would be an indicator of learned helplessness, and subsequently a more passive style of coping with abuse. Additionally, it was thought that a lifetime of abuse would lead to learned helplessness-it is first the lack of control, and then the perceived lack of control, over one's envi­ ronment that leads to learned helplessness. No substantial statistical evidence was found for learned helplessness

191

Table m. t-Test Results Battered Women and Childhood Sexual Abuse

CPA NCPA p-Value

Mean SD Mean SD t(73) (two-tailed)

MillerLH 27.09 6.98 27.39 7.88 1.73 .09 Miller OCT 27.98 6.20 26.36 5.84 1.09 .28 BSIOCT 2.29 0.95 2.48 0.93 -0.84 .40 BDI 32.12 11.87 26.64 10.90 -1.99 .05*

Note. Brief Symptom Inventory (BSI); Obsessive-Compulsive Tenden­ cies subscale: Beck Depression Inventory (BDI); Miller (LH) Learned Helplessness. *p < .05.

tendencies in women who were abused as children, and later abused in adulthood.

Results of the comparison found that those who were previously physically abused exhibited obsessive­ compulsive tendencies t(77) = - 2.05, p < .05. In con­ trast, those who were previously sexually abused exhibited severe depression scores on the Beck Depression Inven­ tory scale t(73) = - 1.99, p < .05.

In order to clarify the results of the two studies, it is observable that there are statistically significantly more obsessive-compulsive tendencies in the prior physi­ cal abuse groups. Conversely, there is considerably more severe depression, and a trend toward learned helplessness tendencies in those who experienced prior sexual abuse (see Table IV).

DISCUSSION

Battered women who reported physical abuse in childhood were more likely to report sexual abuse as girls than battered women who were not physically abused. A perpetrator of sexual abuse will coerce or threaten his victim into silence about her abuse. For some children, the only way to comply with the perpetrators demand of silence is to repress the experience. The victim learns to distrust her own thoughts, instincts, and emotions. As a result she is vulnerable to a complex variety of reactions including dissociative disorders-a result of the split be­ tween their emotions, cognitions, and instincts.

Because the social stigma of incest in America, vic­ tims underreport it. Improving the rate of report will only happen when social stigma has been improved. The effect of the taboo nature of incest is to play into the abuser's hands-if we tum our backs on an ugly secret, if we refuse to acknowledge a problem exists, it breeds the ability to perpetrate and cover up incidences of incest. In the book Conspiracy of Silence, Butler (1978) notes, "Incest vic­ tims initially are betrayed by the adults in their families

t

192 Miller

Table Iv. Combined t-Test Results-Previous Physical Abuse and Previous Sexual Abuse

Mean (SD)

CPA NCPA CSA NCSA p-Value (two-tailed)

Miller OCT 26.23 (5.27) 28.68 (6.37) .ITT BSIOCT 2.61 (0.91) 2.19 (.94) .04* BDI 32.12 (11.87) 26.64 (10.90) .05* MillerLH 27.09 (6.98) 27.39 (7.88) .09

Note. Miller Obsessive-Compulsive Tendencies Scale: Brief Symptom Inventory (BSI); Obsessive­ Compulsive Tendencies subscale: Beck Depression Inventory (BDI); Miller (LH) Learned Helplessness. *p < .05.

who fail to provide them with emotional, physical, and sexual safety, and they are further victimized by a so­ ciety that shuts its doors and its eyes at the mention of sexual abuse." Americans have a social responsibility to discontinue ignoring incest.

Statistical significance was found for girls who wit­ nessed physical violence between her parents. Violence in the home sends a clear message to the child that it is a normal and unavoidable happening within the family. It leaves a strong imprint on the child's emotional develop­ ment and cognitive schema. There is a profound effect on a child who is witness to violence in the family and when it is combined with sexual abuse the adverse effects are compounded.

Obsessive-Compulsive Tendencies

A woman who has never experienced physical abuse is stunned when she is assaulted for the first time. She might have cognitive denial; she cannot believe that he hit her. And he apologizes so sincerely that even though still hurt and confused she forgives him. Soon thereafter, tensions rise again and another assault ensues. The cycle has begun: an explosive attack, a period of peace, rising tension that eventually leads to another attack. What kind of reaction would a woman have to physical assault if she has not been trained to expect assault from her "loved ones?" Would she cower and take whatever assaults and apologies come her way, or would she assume that there is something she is doing wrong? Is it something she can correct? Obsessive-compulsive tendencies (OCT) can be adaptive to her situation; it is a concerted effort to change her environment to prevent further assault.

Results of physical trauma generally manifests in one or more of the following symptoms, (1) Posttraumatic Stress Disorder, (2) she learns to cope in some way with living with her assailant, and, (3) anxiety and somatic reactions such as additional adrenaline, which instead of

lethargy creates excess energy, sweating, and a general state of alarm. This state of alarm propels her to action­ a situation that is not very likely to accommodate LH behaviors, just the opposite in fact. By motivating her to action, in whatever form the action takes (such as "re­ pairing" herself or the environment to accommodate her batterer's assumed wants or explicit demands), or to make the necessary plans to leave him.

Between the two groups of women who had experi­ enced physical abuse as adults, CPA were predicted to have more learned helplessness behaviors, and NCPA would be more likely to exhibit more OCT behaviors. The underlying assumption traditionally has been that learned helplessness belongs in the repertoire of diag­ noses applicable to battered women, however no statisti­ cal evidence has been found to support the hypothesis that battered women stay in the home due to learned helpless­ ness. There is, however, statistical support for obsessive­ compulsive tendencies.

There was a statistically significant main effect for BSI-OCT scores of. As a result of the significance found for OCT in this study, there is an indication of a need to address this issue in therapeutic settings. The form it takes can be dependent on whether treatment is based on somatic, cognitive, behavioral, or client-centered thera­ pies: a combination of cognitive therapy to deal with the trauma of assault and of somatic treatment to help reduce the strength of her anxiety reaction to stress. A battered woman will have to relearn her response that triggers a pounding heart and sweaty palms. For example, a door slam may override her knowledge that she is now in a safe place and she has all of the same physical symptoms as if she was in immediate danger of a battering incident. A positive that OCT brings to the therapeutic situation is that treatment can be more effective by encouraging the woman to utilize the tendency to help herself.

Using the energy inherent in anxiety and teaching a recovering battered woman how to convert obsessive­ compulsive tendencies into an impulse to take action to

Obsessive-Compulsive Tendencies in Battered Women

secure her own safety will expedite positive results in psychotherapy. Assessment measures for severe depres­ sion and LH tendencies and OC tendencies could help direct the therapist working with a woman in distress. Cognitive therapy might be different for those who ex­ hibit more depressive-LH tendencies than OC tendencies, isolating the schemata that influence depression, anxiety, and learned helplessness tendencies. Learned helplessness was described by Seligman (1990) as permanence, perva­ siveness, and stability of personally negative attributions for unfavorable circumstances.

Depression

The prevalence of depression has higher percentage for women than men, and even higher for battered women. In a comparison between women who were severely bat­ tered and those who were not battered at all, the for­ mer group experienced four times the rate of depression (Straus & Gelles, 1990). Hilberman and Munson (1977) described battered women as turning their anger toward themselves, evidenced by chronic depression, passive or active suicidal behavior, and in some cases even self­ mutilation. They all exhibited impaired self-concept, as well as high levels of anxiety.

Psychotherapeutic treatments for depressive and anxious disorder types have some common characteris­ tics. However, they are less apparent when the disorder has taken on a more severe form. For instance, in the case of severe phobic anxiety reaction, desensitization by repeated exposure (a form of rumination) works well, but severe depression responds very well to an opposite remedy-cognitive-based therapy that teaches antirumi­ nation skills.

Rumination as a thought pattern is similar to obsessive-compulsive ritual-the pattern or ritual is re­ peated over and over again until a reduction of anxiety is achieved. Seligman (1 990) has concluded that, "rumi­ nation combined with pessimistic explanatory style is the recipe for severe depression." Women, he said, have a tendency to analyze and contemplate their problems, and this exacerbates their depression. They also take more (than warranted) responsibility for adverse events. Cog­ nitive therapy has been designed to counteract the pro­ cess of rumination and the negative attributional style so often associated with depression. Depressed persons of­ ten view negative events as failures of personal control. Seligman found depression to be associated with thought patterns that assigned blame to oneself, that the fault was a pervasive one, and tending to think of the adverse sit­ uation as permanent. Cognitive treatment is designed to

193

counteract the depressive effects of adverse thought pat­ terns, and train the client to think in terms that are more temporary, specific to the situation, and explaining events as not occurring as an internal problem, but has a source outside oneself. Depression is thought to have an inher­ ited component, possibly from chromosome number 11, creating an imbalance in neurochemicals. Nevertheless, children who do not have chromosomal traits that lead to depression can still become depressed by their social and familial circumstances. Abusive parents can bring about depression in their children by physically assault, incest, or by emotional abuse.

Wetzel (1984) notes that children who have experi­ enced incest are prone to depression, "Incest victims are ten times more likely to be girls than boys. It has been estimated that one out of four women have been victims of childhood sexual assault in the family." The perpe­ trator tends to be domineering, both with his wife and with his children. Dissonance and depression is created in the child when he or she feels powerless yet forced to deal with extreme anxiety. She has to somehow de­ velop situation assessment ability and maturity beyond her years to survive abuse. She is told by the offending parent any number of things to make incest (or physical abuse) seem okay, but the child's inability to cognitively rectify the event with their feelings can be corrosive to their emotional and cognitive development. Children are completely dependent upon their parent, whose influence is pervasive in scope; the child can view the parent as god­ like. The developmental processes ofmastery are hindered by abuse, as a result self-confidence wanes, the ability to cope distorted into whatever means the child can think of at the time. Unfortunately, if whatever means the child comes up with does work at the time, they will tend to come back to that solution whenever they are under stress or duress later. Obsessive-compulsive tendencies are commendable if they provide a solution to a prob­ lem of physical battering, but it becomes problematic if it tends to keep the woman in the battering relationship. She becomes stuck trying to alter herself and her environ­ ment to please her batterer. Obsessive action is a physical manifestation of rumination; sometimes it does not solve the problem, only keeps her "hopelessly" involved in the same situation. This can be characterized as a lack of mastery over present circumstances, just as she had a lack of mastery over her childhood circumstances. The long­ term effects of early sexual abuse from three-years-old to adolescence have repercussions in cognitive and so­ cial development, such as behavior problems, anger, and depression. Victimization, especially after sexual assault, and spousal abuse also increase the probability of suicide (Snyder, 1994).

194

Learned helplessness (LH) was not statistically sup­ ported for any of the groups of battered women in this study. Perhaps LH would have been more accurately de­ scribed as a form of depression-it was meant to be an answer to the perplexing question of why battered women stay in the home, but has failed to adequately explain a complex decision based on economics, children, and cognitions of the battered woman. Depression, with ac­ companying low energy and slowed cognitive-processing ability in severe cases, might be a more plausible explana­ tion for staying. However, it still remains as only a singular component of a multifaceted syndrome.

An overall look at both studies would conclude that battered women who had been physically abused as chil­ dren (PPA) developed obsessive-compulsive tendencies, and battered women who were previously sexually abused as children (PSA) developed more severe symptoms of depression. No statisictical significance was found for learned helplessness in this data set of battered women. However, for sexually abused battered women, there was a trend toward learned-helplessness behaviors.

Rather than emphasizing just one or two character­ istics or symptoms of a battered woman, it would be best to widen the scope to include all of the symptoms experienced by these women. It is not just a feeling of helplessness experienced by some women, and posttrau­ matic stress syndrome, but also anxiety, depression and obsessive-compulsive tendencies that comprise a complex reaction to battering. Assault in the home is a pervasive and complex situation that taxes every resource the woman has.

A model was formulated based on statistical re­ sults reported herein. Previous hypothesis predicted a di­ chotomy between groups as on extreme of a continuum consisting ofactive coping means for battered women who had been not physically abused in childhood (exhibited by obsessive-compulsive tendencies), and the other extreme consisting of passive coping means for battered women who had been physically abused in childhood (exhibited by learned helplessness) (see Table V). This model ex­ plains the effects of different forms of abuse in childhood in reaction to adult battering-that it was women who had experienced prior sexual abuse that were more passive, exhibited by depression-and prior physical abuse was indicative of obsessive-compulsive tendencies.

Future research should focus on different therapeutic methods for different types ofbattered women. Inherent in the concept of passive or active tendencies is the idea that those propensities will carry over into reaction to and cop­ ing with therapy. A battered woman exhibiting obsessive­ compulsive tendencies would have a better prognosis for recovery-at a more rapid rate than those who are

Miller

Table V. Model: Battered Women's Abuse History and Coping With Assault

Passive coping

Childhood sexual abuse* Severe depression

Active coping

No childhood physical/sexual abuse Obsessive-compulsive tendencies

Note. *Childhood sexual abuse and other physical abuse coincide, p < .01.

depressed. A battered woman who has experienced prior sexual and physical abuse would benefit from a different method of therapy than the previously mentioned group, perhaps one that centered on the cognitive effects of de­ pression, as well as alternative coping methods than those she has used for battering during childhood or adulthood.

REFERENCES

American Psychiatric Association (2000). Diagnostic and Statistical Manual of Mental Disorders, 4th edn. Washington, DC: American Psychiatric Association.

Bailey, S. (1996). Self-Management Therapy Depression in Battered Women. Unpublished master's thesis, University of Houston, Hous­ ton, TX.

Barber, J. S. (1998). A log-linear analysis of the variables related to dissociation in women with histories of incest. Dissert. Abstr. Int. 59: 733.

Bleiberg, K. L. (2000). The disclosure of childhood sexual abuse and post-traumatic stress disorder and related symptoms of cognitive and affective impairment. Dissert. Abstr. Int. 61 : 1072.

Brockner, J., and Rubin, J. Z. (1985). Entrapment in Escalating Con­ flicts : A Social Psychological Analysis. New York: Springer-Verlag.

Brown, L. E. (1998). A comparison of coping styles and body image of abused and non-abused women. Dissert. Abstr. Int. 59: 1842.

Butler, S. (1978). Conspiracy of Silence: The Trauma of Incest. San Francisco: Volcano Press.

Courtois, C. A. (2000). The aftermath of child sexual abuse: The treatment of complex posttraumatic stress reactions. In Szuchman, L. T., and Muscarella, F. (eds.), Psychological Perspectives on Human Sexuality. New York: Wiley.

Derogatis, L. R., and Melisaratos, N. (1983). The Brief Symptom Inven­ tory: An introductory report. Psycho/. Med. 13: 596-605.

Dumont, R. (1996). The Sky is Falling: Understanding and Coping with Phobias, Panic, and Obsessive-Compulsive Disorders. New York: Norton.

Gleason, W. J. (1993). Mental disorders in battered women: An empirical study. Violence Viet. 8(1): 53-08.

Green, A. H., Coupe, P. Fernandez, R., and Stevens, B. (1995). Incest revisited: Delayed post-traumatic stress disorder in mothers following the sexual abuse of their children. Child Abuse Neg/. 19(10): 1275- 1282.

Hilberman, E., and Munson, K. (1977-1978). Sixty battered women. Victimol: Int. J. 2(3/4): 460-470.

Hoteling, G. T., and Sugarman, D. B. (1986). An analysis of risk mark­ ers in husband and wife violence: The current state of knowledge. Violence Viet. 1(2): 101-124.

Kessler, B. L., and Bieschke, K. J. (1999). A retrospective analysis of shame, dissociation, and adult victimization in survivors ofchildhood sexual abuse. J. Counsel. Psycho/. 46(3): 335-341.

Oliver, J. M., and Baumgart, E. P. (1987). A psychometric comparison of the Beck Depression Inventory and the Inventory for Diagnosing Depression in college population. Assessment 2: 167-172.

Obsessive-Compulsive Tendencies in Battered Women

Relun, L. P. (1984). Self-control questionnaire. In Herson, M., and Bellack, A. S. (eds.), Dictionary of Behavioral Assessment Tech­ niques. New York: Pergamon.

Seligman, M. E. P. (1974). Depression and learned helplessness. In Friedman, R. J., and Katz, M. M. (eds.), The Psychology ofDepres­ sion: Contemporary Theory andResearch (pp. 83-113). Washington, DC:Wiley.

Seligman, M. E. P. (1990). Learned Optimism. New York: Knopf. Smith, G. R. (1999). Psychotherapy experiences of women sur­

vivors of severe childhood sexual abuse. Dissert. Abst. Int. 60: 1873.

Spaccarelli, S., and Fuchs, C. (1997). Variability in symptom expres­ sion among sexually abused girls: Developing multivariate models. J. Clin. Child Psycho/. 26(1): 24-35.

Snyder, C. R. (1994). The Psychology of Hope. New York: Free Press.

195

Straus, M. A., and Gelles, R. J. (1990). Physical Violence in American Families: RiskFactors andAdaptations to Violence in 8,145 Families. New Brunswick: Transaction.

Talbot, N. L., Duberstein, P. R., King, D. A., Cox, C., and Giles, D. E. (2000). Personality traits of women with a history of childhood sexual abuse. Comp. Psycho/. 41(2): 130-136.

Tunick, W. S. (2000). Stigmatization of and willingness to discriminate against adult incest survivors. Dissert. Abstr. Int. 60: 5840.

Walker, L. E. (1984). The Battered Woman Syndrome. New York: Springer.

Walker, L. E. (1989). Terrifying Love-Why Battered Women Kill and How Society Responds. New York: Harper & Row.

Weissman, M. M., and Bothwell, S. (1976). Assessment of social adjust­ ment by patient self-report. Arch. Gen. Psychiatry 33: 1111-1115.

Wetzel, J. W. (1984). Clinical Handbook of Depression. New York: Gardner.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.