RSS Feeds
Brief Communication: Physical Abuse of Boys and Possible Associations with Poor Adult Outcomes
- William C. Holmes, MD, MSCE; and
- Mary D. Sammel, ScD
- From Philadelphia Veterans Affairs Medical Center and University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania.
Abstract
Background: Men's childhood physical abuse experiences are understudied.
Objective: To obtain descriptions about men's personal childhood physical abuse histories and estimate their association with adult outcomes.
Design: Population-based telephone survey.
Setting: Urban areas with high frequency of domestic violence against girls and women.
Participants: 298 men recruited through random-digit dialing.
Measurements: 6 Conflict Tactics Scale items and psychiatric, sexual, and legal history questions.
Results: One hundred of 197 (51%) participants had a history of childhood physical abuse. Most (73%) participants were abused by a parent. Childhood physical abuse history was associated with depression symptoms (P = 0.003), post-traumatic stress disorder symptoms (P < 0.001), number of lifetime sexual partners (P = 0.035), legal troubles (P = 0.002), and incarceration (P = 0.007) in unadjusted analyses and with depression symptoms (P = 0.015) and post-traumatic stress disorder symptoms (P = 0.003) in adjusted analyses.
Limitations: There may have been inaccurate recall of past events. Lack of exposure time data disallowed direct comparison of abuse perpetration by mothers versus fathers. Other unmeasured variables related to childhood physical abuse might better explain poor adult outcomes.
Conclusions: The high frequency of childhood physical abuse histories in this population-based male sample, coupled with the high proportion of parent perpetrators and the association between childhood physical abuse and adult outcomes that are often associated with perpetration of violence, argues for more study of and clinical attentiveness to potential adult outcomes of men's own childhood physical abuse histories.
Editors' Notes
Context
-
Boys are sometimes victims of physical abuse, and fathers are often identified as the perpetrators. Whether abused boys become abusing men is conjectural.
Content
-
Interviewers surveyed a randomly selected sample of 197 men from neighborhoods with a high prevalence of HIV infection. Of the men, 51% had experienced childhood physical abuse, which was severe in 57%. Compared with men who had not experienced childhood physical abuse, men who did had more symptoms of depression and post-traumatic stress syndrome as adults.
Caution
-
Unidentified conditions during childhood may lead to both physical abuse and poor adult outcomes.
Conclusions
-
Domestic physical abuse of boys was common in this sample of men. Being abused is associated with psychiatric illness in adult life.
—The Editors
Publications about childhood physical abuse focus predominantly on girls' and women's histories. What we know from male samples, however, indicates that childhood physical abuse prevalence in men may be high—28% in male college students, 34% in Canadian men, and 51% in active-duty male soldiers from the U.S. Army (1-3). Furthermore, the National Child Abuse and Neglect Data System (NCANDS) identifies parents (most often mothers) as the most frequent abusers of boys (4). We hypothesize that when boys personally experience physical abuse as a resolution to conflict in their childhood home, they may learn that perpetrating domestic or intimate partner violence (both called “domestic violence” hereafter) as men is an acceptable way to resolve conflict in their adult home.
Claims that men's childhood physical abuse histories might affect their likelihood of becoming perpetrators of domestic violence are conjectural. Thus, our hypothesis-generating study sought to 1) assess how many men from a population-based urban sample reported having been physically abused as boys, 2) assess what proportion of perpetrators were parents, and 3) estimate the association between childhood physical abuse and adult outcomes that are often associated with perpetration of violence (5-8).
Methods
Participants and Procedures
We recruited participants by random-digit dialing after approval by the University of Pennsylvania Institutional Review Boards. Another study aim required use of telephone exchanges for Philadelphia County ZIP codes of areas with a high incidence of AIDS. These areas also have high rates of domestic violence against girls and women (9, 10). The interviewers screened households to identify and recruit men into 1 of 3 age groups: 18 to 29 years of age, 30 to 39 years of age, and 40 to 49 years of age. Appendix Figure 1 summarizes screening and interviewing methods and numbers.
Variables
Interviewers asked 6 items from the validated Conflict Tactics Scale's (CTS) Physical Assault dimension (Appendix Figure 2). These items have been used previously to identify childhood physical abuse histories in men and are similar to items used to identify domestic violence in girls and women (2, 11-14). We considered childhood physical abuse to have occurred if responses to questions 1, 2, or 4 were “sometimes” or “often” or if responses to questions 3, 5, or 6 were “rarely,” “sometimes,” or “often.” We considered severe childhood physical abuse to have occurred if responses to questions 3 or 4 were “often” or if responses to questions 5 or 6 were “rarely,” “sometimes,” or “often.” Participants identified perpetrators for all instances of childhood physical abuse.
Interviewers asked participants whom they had lived with during most of their childhood and then administered the Parental Bonding Instrument (PBI), instructing participants to answer by thinking “about what the parents and/or guardians most influential in your life were like during the time before you turned 18 years old” (15). We used the responses to create 2 dimensions—care and overprotection—that define 4 parenting categories through cross-classification (16). These categories are (in descending order of preferred parenting style) optimal (high care and low overprotection), affectionate constraint (high care and high overprotection), affectionless control (low care and high overprotection), and neglectful (low care and low overprotection).
Interviewers assessed the number and frequency of symptoms of depression and post-traumatic stress disorder with the Center for Epidemiologic Studies Depression Scale (CES-D) (17) and Posttraumatic Stress Diagnostic Scale (16), respectively.
Statistical Analyses
We performed group comparisons by using 2-tailed t-tests for continuous variables and chi-square methods (or the Fisher exact test when expected frequency for 1 subgroup was < 5) for categorical variables. We completed a forced-entry multivariable linear regression in which we used all variables that either were associated with childhood physical abuse (P < 0.100) or were related conceptually to the association between childhood physical abuse and psychiatric or risk behavior outcomes to adjust for continuous outcomes (we used dummy variables for included ordinal data). We used SPSS 12.0 for Windows (SPSS Inc., Chicago, Illinois) to manage and analyze data.
Role of the Funding Source
This study was supported by a grant from the National Institute of Drug Abuse (DA015635). The funding source had no role in the design, conduct, or reporting of the study or in the decision to submit the manuscript for publication.
Results
Participants
Of 298 recruited men, 197 (66%) participated. Participants' mean age was 34 years, 68% were nonwhite, 7% were homosexual or bisexual, 45% had a high school education or less, and 27% had an annual income of $20 000 or less.
Childhood Physical Abuse Histories
Table 1 enumerates types of abuse (as assessed by CTS items): Fifty-five men were hit with an object; 51 men were kicked, bit, or punched; 44 men were pushed, grabbed, or shoved; 41 men were physically attacked; 18 men had an object thrown at them; and 13 men were choked, burned, or scalded. Parents were the most frequent perpetrators.
Of the 197 participants, 100 (51%) men experienced at least 1 form of abuse that meets the definition of childhood physical abuse. Fifty-seven of these men experienced at least 1 form of abuse that meets the definition of severe childhood physical abuse. Table 2 indicates that educational attainment (P = 0.002) and parental bonding (P = 0.054) were associated with childhood physical abuse prevalence. While childhood physical abuse prevalence seemed to differ by age and race or ethnicity, these relationships were not statistically significant (P = 0.061 and P = 0.075, respectively). No variable in Table 2 was associated with severe childhood physical abuse.
Although a larger percentage of childhood physical abuse was attributed to mothers than fathers (Table 1), the amount of time spent with mothers and fathers is not known. Table 3 provides an indication of abuse within different living arrangements, which is an indirect measure of exposure to potential perpetrators of domestic violence. Table 3 also clarifies the extent to which men had been abused by parents. Of the 100 men with childhood physical abuse histories, 73 reported that a mother or father had perpetrated at least 1 form of violence that meets the definition of childhood physical abuse.
Potential Outcomes Associated with Childhood Physical Abuse Histories
Unadjusted analyses (Table 4) indicate that a childhood physical abuse history was associated with the number or frequency of depression symptoms (P = 0.003) and post-traumatic stress disorder symptoms (P < 0.001), with the number of lifetime sexual partners (P = 0.035), and with legal troubles (P = 0.002) and incarceration (P = 0.007). After adjustment of continuous outcomes for all variables that either were associated with childhood physical abuse (P < 0.100) or were related conceptually to the association between childhood physical abuse and psychiatric or risk behavior outcomes, however, a childhood physical abuse history was associated only with depression and post-traumatic stress disorder symptoms (P = 0.015 and P = 0.003, respectively).
Discussion
We recruited this population-based sample of men from an urban locale that is known to have a high prevalence of domestic violence against girls and women. More than half of the sample reported that they had been physically abused as boys. This high prevalence of childhood physical abuse underscores the importance of paying greater attention to childhood physical abuse histories personally experienced by men when they were boys (1-3). Coupled with the high reported frequency of a parent as the perpetrator, this high childhood physical abuse prevalence also suggests that we need to understand what boys' childhood physical abuse experiences may teach them about domestic conflict resolution.
We could not test the hypothesis that men with childhood physical abuse histories—particularly if perpetrated by parents—may be more likely to perpetrate violence against their own intimate partners, children, or both. If we had assessed domestic violence perpetration by our respondents, we would have been mandated by law to report perpetrators. If we had disclosed this requirement when we sought informed consent, we would have biased our study sample or responses. Instead, we asked participants about outcomes, such as depression, sexual risk, substance abuse, and hostility, which previous studies have reported to be associated with dating violence, domestic violence, and other violent criminal behavior (5-8). These outcomes were substantially more common in men with childhood physical abuse histories, providing a circumstantial case that boys who are victims of domestic violence may be more likely to perpetrate domestic violence as adults.
Three major limitations curb our certainty about this circumstantial case. First, although the sample was population-based, it was also nonaffluent and was largely minority- and urban-based. Thus, results cannot be extended beyond this population. Second, we based our approach to determining the presence of a childhood physical abuse history on responses to CTS questions about acts that did not have equivalent levels of potential injury. Furthermore, we did not assess whether actual bodily harm occurred. Current versions of the CTS allow better characterization of severity (based on likelihood for injury) and chronicity (18). Finally, adjustment for childhood physical abuse subgroup differences attenuated the associations between childhood physical abuse and depression symptoms, post-traumatic stress disorder symptoms, and number of sexual partners. We had too few participants to adjust for dichotomous outcomes. Explanations other than childhood physical abuse histories may, in fact, explain associations with adult outcomes that are associated with violent behavior.
Childhood physical abuse may be a marker for other conditions that are the primary explanatory variables for poor adult outcomes. For instance, the living arrangement differences reported in Table 3 highlight that one potentially explanatory but unmeasured variable may be a chaotic childhood social environment (which may continue into adulthood). Other candidate variables could include age, sex, or educational attainment of parent or parents; alcohol or drug use in the home or neighborhood; number and identity of people living in the primary residence, as well as presence of siblings or a trusted adult; or involvement of child protective services.
Future studies using a case–control design—where cases are men with known perpetration histories—might be conducted to assess cycle-of-violence hypotheses more directly. These and other epidemiologic studies of childhood physical abuse in men must adjust for a fuller set of potential confounders and effect modifiers.
Until research gives us a broader knowledge about men with childhood physical abuse histories, clinicians may be the only resource available to many affected men and boys. Clinicians who are aware that some populations of boys and men may have a high frequency of childhood physical abuse histories have an opportunity to identify these histories when clinical situations suggest their presence. Clinicians also have an obligation to advocate measures to assure the safety of abused boys (as rigorous as those recommended for girls and women), as well as to be alert that adult male patients with childhood physical abuse histories may be experiencing abuse-related outcomes that have negative health consequences (19, 20).
Article and Author Information
-
Grant Support: Dr. Holmes was funded by a grant from the National Institute of Drug Abuse (DA015635).
-
Potential Financial Conflicts of Interest: None disclosed.
-
Requests for Single Reprints: William C. Holmes, MD, MSCE, Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania School of Medicine, 733 Blockley Hall, 423 Guardian Drive, Philadelphia, PA 19104-6021; e-mail, holmeswc{at}mail.med.upenn.edu.
-
Current Author Addresses: Drs. Holmes and Sammel: Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania School of Medicine, 733 Blockley Hall, 423 Guardian Drive, Philadelphia, PA 19104-6021.
-
Author Contributions: Conception and design: W.C. Holmes.
-
Analysis and interpretation of the data: W.C. Holmes, M.D. Sammel.
-
Drafting of the article: W.C. Holmes.
-
Critical revision of the article for important intellectual content: W.C. Holmes, M.D. Sammel.
-
Final approval of the article: W.C. Holmes, M.D. Sammel.
-
Provision of study materials or patients: W.C. Holmes.
-
Statistical expertise: M.D. Sammel.
-
Obtaining of funding: W.C. Holmes.
-
Administrative, technical, or logistic support: W.C. Holmes.
-
Collection and assembly of data: W.C. Holmes.
Related Clinical Content
