Reprint requests should be addressed to David M. Fergusson, Christchurch ...... rather than deliberate maltreatment (EPOCH, 1992; Ritchie & Ritchie, 1993), it is ...
Child Abuse & Neglect, Vol. 21, No. 7, pp. 617-630, 1997 Copyright © 1997 ElsevierScience Ltd Printed in the USA. All rights reserved 0145-2134/97 $17.00 + .00
Pergamon
PII S0145-2134(97) 00021-5
PHYSICAL PUNISHMENT/MALTREATMENT DURING CHILDHOOD AND ADJUSTMENT IN YOUNG ADULTHOOD DAVID M. FERGUSSON
AND MICHAEL T. LYNSKEY
Christchurch Health & Development Study, Department of Psychological Medicine, Christchurch School of Medicine, Christchurch, New Zealand
ABSTRACT Objective: To study the relationships between retrospective reports of physical punishment/maltreatment and rates of adjustment difficulties at age 18 in a birth cohort of New Zealand subjects. Method: Data were gathered over the course of an 18 year longitudinal study of a birth cohort of 1,265 New Zealand born children. At age 18 retrospective reports of exposure to physical punishment/maltreatment were obtained. At this time the cohort was also assessed on measures of psychosocial adjustment including juvenile offending, substance abuse behaviors, and psychiatric disorder. Results: Young people reporting exposure to harsh or abusive treatment during childhood had elevated rates of juvenile offending, substance abuse, and mental health problems. However, subsequent analysis using logistic regression methods showed that much of the elevated risk shown by this group was explained by social and contextual factors that were associated with patterns of childhood punishment/maltreatment. Nonetheless, even after control for confounding factors those reporting harsh or abusive childhood experiences were at increased risks of violent offending, suicide attempts, being a victim of violence, and alcohol abuse. Conclusion: This study leads to three major conclusions: (1) Those exposed to harsh or abusive treatment during childhood are an at-risk population for juvenile offending, substance abuse, and mental health problems; (2) Much of this elevated risk arises from the social context within which harsh or abusive treatment occurs; (3) Nonetheless, exposure to abuse appears to increase risks of involvement in violent behavior and alcohol abuse. © 1997 Elsevier Science Ltd
Key Words--Physicalpunishment/maltreatment, Psychiatric illness, Substance abuse/dependence, Criminal offending, Longitudinal study.
INTRODUCTION IN RECENT YEARS there have been increasing concerns expressed about the physical maltreatment of children and the long term consequences of such maltreatment for individual adjustment (Kempe, Silverman, Steele, Droegemueller, & Silver, 1962; Malinosky-Rummell & Hansen, 1993; Wissow, 1995). In general, prevalence studies have suggested that while maltreatment leading to serious physical injury or death is uncommon (Kotch, Chalmers, Fanslow, Marshall, & Langley, 1993; McClain, Sacks, Froehlke, & Ewigman, 1993), a substantial number of children are exposed to parental maltreatment and discipline practices that are sufficient to lead to physical injury (Mullen, Martin, Anderson, Romans, & Herbison, 1996; This research was funded by grants from the Health Research Council of New Zealand, the National Child Health Research Foundation, Lottery Health, and the Canterbury Medical Research Foundation. Received for publication October 8, 1996; final revision received December 9, 1996; accepted December 9, 1996. Reprint requests should be addressed to David M. Fergusson, Christchurch Health & Development Study, Department of Psychological Medicine, Christchurch School of Medicine, P.O. Box 4345, Christchurch, New Zealand. 617
618
D.M. Fergusson and M. T. Lynskey
Wolfner & Gelles, 1993 ). Research in this area has suggested that children exposed to physical abuse during childhood are at increased risks of a wide range of later outcomes (for an overview of this literature see Malinosky-Rummell & Hansen, 1993). These include increased risks of later violent behaviors and offending (Briere & Runtz, 1990; Cavaiola & Schiff, 1988; Lewis, Shanok, Pincus, & Glaser, 1979; Pollock et al., 1990), increased risks of psychiatric disorder (Briere & Runtz, 1988; Bryer, Nelson, Miller, & Krol, 1987; Chu & Dill, 1990), increased risks of substance use behaviors (Brown & Anderson, 1991; Cavaiola & Schiff, 1988; Schaefer, Sobieraj, & Hollyfield, 1988) and increased risks of suicidal behaviors (Briere & Runtz, 1988; Cavaiola & Schiff, 1988; Deykin, Alpert, & McNamara, 1985). There are, however, a number of difficulties in interpreting the available literature. Many studies of the consequences of exposure to physical abuse have been based on samples of individuals who have already been identified as having some form of psychosocial problem (Brown & Anderson, 1991; Bryer et al., 1987; Cavaiola & Schiff, 1988; Chu & Dill, 1990; Deykin et al., 1985; Lewis et al., 1979; Pollock et al., 1990; Schaefer et al., 1988). There are two ways in which such studies are limited. First, analysis of those exposed to severe childhood physical abuse may not reflect the spectrum of abusive behaviors to which children are exposed and may examine the outcomes of only those exposed to extremes of violent behaviors. Second, clinical and agency samples may suffer from sample selection biases in which the processes by which subjects come to attention may lead to artifactual relationships between exposure to child abuse and outcome risks (Bertolli, Morgenstern, & Sorenson, 1995; Ellenberg, 1994). Secondly, a majority of studies have failed to consider the extent to which apparent associations between exposure to child abuse and later outcomes may have arisen from social and contextual factors that were related to both risks of exposure to child abuse and outcome risks. In particular, the evidence on risk factors associated with physical abuse suggests that those exposed to abuse tend to come disproportionately from family and social environments characterized by multiple disadvantageous features including poverty, limited parental education, generally impaired parenting skills, stress, and a variety of similar factors (Bousha & Twentyman, 1984; Connelly & Straus, 1992; Dubowitz, Hampton, Bithoney, & Newberger, 1987; Martin & Walters, 1982; Trickett & Susman, 1986; Whipple & Webster-Stratton, 1991). It may be argued that many of the apparent associations between childhood physical abuse and later adjustment reflect the social and family context within which abuse occurs rather than the direct traumatic effects of abuse on individual adjustment (Mullen et al., 1996). In addition to these problems of sampling and confounding, research in this area has been subject to threats to validity from a series of measurement issues centering around the accuracy and validity of measures of abuse (Carlin et al., 1994; Plunket & Oates, 1990). The best way of addressing many of these issues would be through a longitudinal study in which childhood abuse experiences were assessed throughout childhood concurrently with measures of social and contextual factors with exposure to child abuse being related to measures of adjustment in later life (Bertolli et al., 1995). This design, however, faces some insurmountable practical difficulties centering around the adequate assessment of physical abuse during childhood. Since physical child abuse is generally intra-familial, there are clear difficulties in assessing abuse on the basis of reports of family members or children. Furthermore, even if such reports could be collected in a valid way, disclosure of physical abuse during childhood would pose severe ethical problems since investigators would be obliged to report or intervene to reduce physical abuse. A weaker but more practical design would be to follow a birth cohort of children to the point of young adulthood measuring relevant social, family, and related circumstances throughout this period. At the point of young adulthood cohort members could be questioned about their exposure to physical child abuse in a way that would avoid many of the problems associated with the prospective assessment of child abuse. In this paper we report on a study of exposure
Physical abuse
619
to physical abuse during childhood and its consequences based on this design. Specifically, the study involved a birth cohort of children who have been studied to the age of 18 with regular assessments of family, social, and related circumstances being made at annual intervals throughout childhood. At the age of 18 cohort members were then asked to provide retrospective accounts of childhood exposure to physical maltreatment and, concurrently with this, a series of measures of personal adjustment were obtained. This design makes it possible to examine the relationships between retrospectively reported physical abuse and adjustment at the age of 18, taking into account prospectively measured childhood, family, and related circumstances. The specific aims of the present analysis were: 1. To examine the relationships between reports of physical ill treatment and abuse during childhood and measures of adjustment at age 18, including psychiatric disorders, suicidal ideation, substance abuse/dependence, juvenile offending, and being a victim of violent crime. 2. To examine the extent to which exposure to child abuse was associated with other adverse family factors including economic factors, social background, parenting behaviors, parental characteristics, and exposure to childhood sexual abuse. 3. To estimate the associations between reports of physical abuse during childhood and outcomes at age 18 taking into account social, familial, and contextual factors that were associated with reports of physical abuse during childhood.
METHOD The data reported here were collected during the course of the Christchurch Health and Development Study. The Christchurch Health and Development Study is a longitudinal study of a birth cohort of 1,265 children born in the Christchurch (New Zealand) urban region during mid 1977. These children were studied at birth, 4 months, and at annual intervals to the age of 16 years and again at age 18 years. An overview of the study design has been given previously (Fergusson, Horwood, Shannon, & Lawton, 1989). The data analyzed in this report were measured in the following ways.
The Assessment of Exposure to Physical Punishment~Maltreatment At the age of 18, sample members were asked to report the extent to which their parent figures used physical methods of punishment. Reports were made on a five-point scale ranging from "parent never used physical punishment" to "parent treated me in a harsh and abusive w a y . " Separate ratings were obtained for the child's mother figure and father figure (if available). The ratings for both parents were combined into a single rating by classifying the young person's reported exposure to physical child abuse into a composite four-point scale based on the greatest exposure to physical punishment reported by the respondent. This classification revealed that 10.8% of the sample reported that both parents never used physical punishment; 77.7% of respondents reported that both parents seldom used physical punishment; 7.6% of respondents reported that at least one parent used physical punishment methods regularly; 2.0% of respondents reported that at least one parent used physical punishment methods too often and too severely; and 1.9% of respondents described at least one parent as treating them in a harsh and abusive way. Due to the small numbers in the last two categories, these were combined in the analysis into a single category representing those sample members who reported either overly frequent and severe physical punishment or harsh and abusive treatment.
620
D . M . Fergusson and M. T. Lynskey
To provide cross-validation of the overall classification, the rate at which subjects reported specific abusive experiences in childhood was related to the overall classification. Reports of specific abusive experiences spanned a range of measures describing physical punishment and maltreatment during childhood. These measures were based on the self-report measures of punitive experiences described by Berger, Knutson, Mehm, and Perkins (1988), augmented by a number of other measures. Table 1 shows the associations between the global rating of physical punishment/maltreatment described above and specific reports of abusive experiences in childhood. For each of the four groups to be analyzed the Table shows: (a) the frequency with which each item was reported; (b) the association between the specific measure and the overall classification based on the phi coefficient; and (c) the statistical significance of the association based on the Mantel-Haenszel chi square test of linearity. The Table shows clear gradients in the extent to which reports of abusive experiences were made. The trends in this Table can be seen from the mean number of punishment or abusive experiences reported for each group: While those reporting that their parents seldom used punishment only reported a mean of .24 punishment or abuse experiences, those reporting frequent/severe punishment or being treated in a harsh or abusive way reported a mean of 4.55 such experiences. These results support the view that the overall rating of exposure to abuse during childhood was generally consistent with the individual reports of specific punishment and abuse experiences in childhood.
Mental Health, Substance Use, and Criminal Off'ending at Age 18 Years Psychiatric disorder and attempted suicide. Concurrently with the assessment of physical abuse, subjects were questioned about their psychiatric symptoms over the period from 1618 years using a questionnaire based on the Composite International Diagnostic Interview (CIDI; World Health Organization, 1993), supplemented by an instrument based on the SelfReport Delinquency Inventory (SRDI; Elliott & Huizinga, 1989). The CIDI items were used to assess mood disorders, anxiety disorders, and substance use disorders in the sample while the SRDI was used to assess conduct disorder.
Table 1. Associations Between Evaluation of the Extent of Physical Punishment and Reports of Various Forms of Physical Punishment and Injury During Childhood (0-16 years) Extent of Physical Punishment
l None % Reporting Frequent Smacking % Reporting Being Hit Around Head or Body with Fists % Reporting Being Frequently Hit on the Bottom with a Cane, Strap or Similar Object % Reporting Being Hit Around Head or Body with a Cane, Strap or Similar Object % Reporting Receiving a Severe Beating % Reporting Being Kicked % Reporting Being Choked or Throttled % Reporting Being Locked in a Cupboard or Shed % Reporting Being Burnt % Reporting Being Injured as a Result of Physical Punishment Mean Number of Acts N
2 Seldom
3 Regular
4 Severe/Harsh
d~
P
11.0
3.8
56.4
60.0
.59
