Alcoholism in the family statistics Alcoholism in the family what you can do Alcoholism in the family system Alcoholism in the family article Alcoholism in the family essay Alcoholism runs in the family Alcoholism in the family family roles Alcoholism eff. SPECIFICITY OF DISRUPTED PROCESSES IN FAMILIES OF ADULT CHILDREN OF ALCOHOLICS, Alcohol and Alcoholism, Oxford Academic


Children of alcoholics (COAs) have been characterized as an at-risk population in part, because of the dysfunctional family environments that disrupt psychosocial development among offspring exposed to parental alcoholism. This study examined the specificity of problematic family environments to children of alcoholics vs children exposed to other significant family stressors that included parental death, unemployment, separation, divorce, or major illness. University students completed self-report measures of family stressors, family relationship problems, family communication quality, family conflict, and relationship with parents. Based on a family stressor checklist, 20 students exposed only to parental alcoholism and no other family stressors were compared to several other groups exposed to specific family stressors, and to 50 control subjects who had reported no family stressors. Results showed that students from families where a parent is or was an alcoholic, and where there were no other family stressors, recall disturbed family relations no more commonly than students specifically exposed to other family stressors such as parental divorce, death, or major illness, and no more commonly than those who reported no family stressors. Further analyses suggest that the disruption of COAs' family environments may be explained by their increased likelihood of experiencing additional family stressors that can have a disruptive effect, such as parental separation, divorce and unemployment. Alcoholism in the family.


Children of alcoholics (COAs) are a population at risk for developmental and interpersonal problems ( West and Prinz, 1987 ; Woodside 1988 ; Fisher et al., 1992 ). Perhaps the most common explanation for the adverse effects of parental alcoholism has been characterized as the general environmental mechanism ( Velleman, 1992 ). According to this explanation, parental alcoholism is thought to produce disturbed family relationships and dynamics that impact negatively on the psychosocial well-being of children who grow up in such environments.

Indeed, results of a large sample cross-sectional study indicate that COAs exhibited significant differences from non-COAs in greater involvement in alcohol use, more drug dependence, more depression, agoraphobia, social phobia and generalized anxiety, less behavioural control, lower self-esteem, lower scores on tests of verbal ability, and lower academic achievement ( Sher et al., 1991 ). Other findings indicate that COAs are more depressed, less satisfied with their marriages, and more likely to drink for coping purposes than non-COAs ( Domenico and Windle, 1993 ).

Notwithstanding some of the significant problems that appear to be associated with being COAs, a significant body of literature is emerging that questions the distinctness and at-risk status of the COA population. For example, studies have found no differences between COAs and non-COAs in alcohol-related problems, suicidal ideation, personal control, perceived social support ( Wright and Heppner, 1991 ), state anxiety, mood ( Clair and Genest, 1987, 1992 ), social skills ( Segrin and Menees, 1996 ), social maladjustment ( Dinning and Berk, 1989 ), personality traits of expressiveness, alienation, and defensiveness ( Havey and Dodd, 1993 ), object relations deficits, compulsive behaviour ( Hadley et al., 1993 ), or depression ( Reich et al., 1993 ). The list of such studies is too long, and the breadth of dependent variables too extensive to dismiss these findings to sampling error or other artefacts.

Inconsistencies in conceptual and methodological criteria in this literature have been identified as a potential source of heterogeneity in study findings (e.g. Velleman, 1992; Henderson et al., 1994). In the typical study, a group of COAs is identified and then compared to a group of non-COAs. Such methods fail to account for the fact that parental alcoholism is associated with numerous other family stressors ( West and Prinz, 1987 ). It is entirely reasonable to assume that other family stressors, such as divorce, major illness, and unemployment, do not vary randomly with problem parental drinking. These stressors can, themselves, disrupt the family dynamics that often are identified as the source of COAs' increased incidence of psychosocial problems. Thus problems that may appear to result from the stress of parental alcoholism may actually be the result of other stressors that co-occur with parental alcoholism.

A number of researchers have recently sought to address this issue by including distressed parental controls in the designs of studies that evaluate the well-being of COAs (e.g. Jacob and Leonard, 1986; Giunta and Compas, 1994; Senchak et al., 1996). These studies indicate that some of the problems often attributed to being a COA, such as externalizing, behaviour problems, fear of intimacy, and problematic social skills, are not specific to the offspring of alcoholics. Children from families with other major stressors often exhibit comparable outcomes or problems as COAs.

Even in those investigations with a distressed control group, other family stressors that co-vary with parental alcoholism are rarely assessed, much less excluded. This has the potential to cloud the conclusions that can be drawn about the family environments and subsequent ill-effects of being reared in a family with a problem drinker. The question of specificity of disrupted family environments among families of COAs has consequential theoretical implications for understanding the nature of the causal processes that may lead to psychosocial problems among COAs ( Garber and Hollon, 1991 ). It has been assumed, at least implicitly, that parental alcoholism disrupts family processes that in turn are a sufficient and proximal cause of a host of adjustment problems among offspring of alcoholics. The specificity of these problems for the COA has clearly been called into question by recent research findings. However, much less is known about the specificity of dysfunctional environments in the families of origin of COAs, vs children of other stressed families. Throughout this paper, we use the phrase ‘family of origin’ to refer to the family environment in which participants lived during their youth, recognizing that in many cases this involved a step-parent, rather than two biological parents.

Alcoholism in the family statistics Alcoholism in the family what you can do Alcoholism in the family system Alcoholism in the family article Alcoholism in the family essay Alcoholism runs in the family Alcoholism in the family family roles Alcoholism eff

The purpose of the present investigation was to examine the specificity of disrupted family environments to families of adult children of alcoholics. In particular, we sought to answer the following research question: Are disrupted family environments and relations experienced specifically by COAs, or are they also experienced by children exposed to other common family stressors? Toward that end, the family functioning (i.e. family relationship problems, family communication, family conflict, and relationship with parents) of COAs was compared to that of a number of other groups with significant family stressors that include death of a parent, parental unemployment, separation, divorce, or parental illness, and a group of controls whose families of origin were not exposed to any of these stressors. A variety of stressors were chosen for comparison, because reaction to stress is not necessarily a generic phenomenon ( Rutter, 1981 ; Burr and Klein, 1994 ). Group definition in this investigation required specific or ‘pure’ exposure to one of the above-mentioned stressors, without which a valid test of specificity is impossible. Hence, COA status in this investigation is not confounded with parental divorce, unemployment, major illness, etc., nor is parental divorce confounded with parental unemployment, major illness, etc. All cases of multiple family stress exposure were screened out of the analyses that compared groups.

Because the incidence of specific or ‘pure’ exposure to a particular family stressor, such as parental alcoholism, is not particularly high in the population, we chose to employ a survey method that would allow for assessments of a large number of potential respondents. Family environment variables were selected for analysis based on their potential for disruption as a result of family stressors, and the availability of well-established measures for these variables.



Initial assessments were conducted on 684 undergraduate students enrolled in introductory communication courses at a large American university who participated in fulfilment of a course requirement or received extra credit. The sample contained 60% females and 40% males with a mean age (±SD) of 20.48 ± 3.47 years. The ethnic background of the participants was 88% White, 3.8% Black, 3.1% Asian, 2.5% Hispanic, 2.6% other. Information on biological parents indicated that 70.2% of the sample had biological parents who were married, with 25.4% divorced and 4.4% widowed.

From the total sample of 684, seven comparison groups were constructed based on subjects' responses to a checklist of stressors. Past research indicates that single-item methods of identifying COAs are stable over time, correlate with clinical assessments of parental alcoholism ( DiCicco et al., 1984 ) and produce prevalence rates similar to more detailed diagnostic instruments ( Claydon, 1987 ; Berkowitz and Perkins, 1988 ). In order to qualify for membership in one of the groups, participants must have experienced one and only one of the following stressors: parental alcoholism (n = 20), parental death (n = 12), parental unemployment (n = 28), parental separation (n = 12), parental divorce (n = 65), or parental major illness (n = 38). A non-stressed control group was created by including those participants who did not experience any of these stressors (n = 338, of which 50 were randomly sampled for inclusion in subsequent analyses). A total of 50 non-stressed controls were included since that number is still in range with the numbers for the other stress groups, while being sufficiently large to contribute appreciably to the statistical power of subsequent tests of group differences. Of the 684 in the original sample, 171 were dropped from further analyses due to experiencing multiple family stressors. Demographic statistics for each of the comparison groups appear in Table 1.

The only significant group difference for all of the family environment variables involved the CAF scale. However, this effect did not involve COAs. Rather, the significant contrasts, as determined by the Dunnett C test (a test that does not require homogeneity of variance among groups) indicated that participants who experienced parental divorce had significantly less positive attitudes toward their fathers than no stress controls, or those who experienced parental unemployment. COAs did not differ significantly from any of the groups on any of the seven family variables under investigation. Especially striking is the lack of any significant differences between the COAs and non-stressed controls.

To appropriately test the specificity research question, the incidence of co-occurrence among family stressors must be reduced to zero. Consequently, any participant with an alcoholic parent, who also was exposed to parental divorce, separation, unemployment, etc. had to be dropped from the analyses. One might legitimately question the representativeness of a group of COAs whose parents are not divorced, separated, have never been unemployed, seriously ill, and so on. In fact, the definition of COA status in most investigations is simply based on the presence of parental alcoholism, without regard to other family stressors. We re-analysed the group comparisons, this time defining COA status regardless of other family stressors. We then compared this group of COAs to a group of stressed controls, who had been exposed to one or more of the family stressors under investigation, and to the group of no stress controls. This operational procedure yielded 77 COAs, 259 stressed controls, and 348 no stress controls. A series of one-way ANOVAs were conducted to test for differences between these three groups. Where significant effects were attained, either Tukey HSD or Dunnett C tests were conducted, respectively depending upon the homogeneity or heterogeneity of variances within the groups, to specify the exact nature of the group differences. These tests had power in excess of 0.99 to detect a medium-sized effect of f = 0.25, or η

Alcoholism in the family statistics Alcoholism in the family what you can do Alcoholism in the family system Alcoholism in the family article Alcoholism in the family essay Alcoholism runs in the family Alcoholism in the family family roles Alcoholism eff

= 0.06 ( Cohen, 1988 ). Results of these analyses appear in Table 3, there were significant differences between the three groups for every family environment variable under investigation. Furthermore, the pattern of group differences was remarkably consistent over these dependent measures: COAs and stressed controls have less functional family environments than no stress controls, but they do not differ significantly from each other. For two family environment variables, the communication and affective responsiveness subscales of the FAD, this pattern was evident in the magnitude of the means, but did not emerge through the post-hoc tests. The only family variable whose disturbance appears somewhat specific to COAs was family problem solving, as assessed by a subscale of the FAD. COAs scored significantly lower than no stress and stressed controls, who did not differ from each other.

A comparison of the results in Tables 2 and 3 indicates that, when ‘pure’ COAs were compared to no stress controls, there were no significant group differences. However, when COAs, with or without other family stressors, were compared to no stress controls, significant differences were evident. What might explain this different pattern of results? Aside from the increased statistical power of the three-group tests in Table 3 shows that COAs exposed to no other family stressors consistently exhibited higher standard deviations on all family environment measures than their non-stressed counterparts. This suggests that, on average, COAs' families may appear similar to other stressed and even non-stressed families. What may distinguish COAs' families of origin is their greater variability in functioning. Perhaps those COAs whose family environments exist on the lower end of the distribution, a distribution that drops lower than that of non-stressed families, are the ones at greatest risk for dysfunction.

Results of this study highlight the impact of different methods for defining COA group status. When children exposed only to parental alcoholism, in the absence of other family stressors, are compared to no stress controls, their families of origin appear to be similar. However, when cases are defined by exposure to parental alcoholism, regardless of the presence of other family stressors, children of alcoholics appear to have more disrupted families of origin, relative to no stress controls. The odds ratio analyses clearly indicated that a host of other family stressors co-occur with parental alcoholism. Therefore, one might conceptualize parental alcoholism as existing in a nomological network of family stress variables that include separation, divorce, unemployment, and major illness, to name but a few possibilities. It is most likely that the amalgamation of these other family stressors working in concert with parental alcoholism, not parental alcoholism per se, creates the disrupted family environments that have been documented among COAs. Stated differently, parental alcoholism may exert its effects on family members through related disturbances in family life, such as unemployment, divorce, and illness. This conclusion is consistent with the findings from a series of recent carefully controlled studies on large samples of adult COAs (e.g. Velleman and Orford, 1993a,b; Orford and Velleman, 1995). Velleman and Orford's studies documented heightened family disharmony associated with parental problem drinking, but at the same time noted that family discord was a better predictor than parental drinking status of the adult adjustment of COAs.

It is important to consider that the non-specificity of disrupted family processes documented in this study centre on interpersonal outcomes, such as family conflict and family communication. Other dysfunctional outcomes such as alcoholism itself, which is common among COAs ( Pollock et al., 1987 ), may be more specific to the family with a parental problem drinker. The role of genes that influence neurochemical responses to alcohol may make the transmission of alcoholism from parent to child more alcohol-specific, rather than the result of more general family disturbances.

One important limitation of the present study concerns the extent to which a full range of family functioning was captured by our sampling method. The different groups not only scored similarly on the family measures, but they tended to report positive family environments. It is plausible that these children who have experienced various family stressors and have made it to college are a special, higher functioning and resilient portion of the population. Perhaps their positive family environments are precisely what helped them to reach college. Family stress research suggests that a positive family environment can be helpful in the face of numerous family stressors ( McCubbin et al., 1980 ; Rutter, 1987 ). Consequently, the COAs and other stressed groups in this study may not be entirely representative of those in the general population.

Another limitation of this research is the definition of stress that was employed. The mere existence of an event (parental alcoholism, divorce, chronic illness, etc.) was sufficient for its characterization as a stressor in participants' lives. This could be problematic, given that researchers have suggested that the family's view or assessment of the stressor has an impact on how they adapt to that stressor ( McCubbin, et al., 1980 ). For parental alcoholism in particular, it would be valuable to assess the severity of the alcoholism in addition to the family reaction to the drinker. Such measures may be more predictive of the family consequences associated with parental alcoholism (e.g. Orford et al., 1976). On a related point, the creation of ‘pure’ stress groups for comparison may have resulted in a bias toward capturing more functional instances of such family stressors. For example, by eliminating cases of parental divorce from cases of parental alcoholism, the resultant parental alcoholism group would be biased toward including the offspring of higher quality parental marriages. Also, creation of the single stressor groups may enhance the likelihood of detecting no significant differences from the no stress control group. Families may be far better able to cope with and adjust to a single stressor, vs multiple stressors. Indeed, when multiple stressor groups were compared with no stress controls, differences in the family environments were more evident. Unfortunately, the elimination of co-morbid stressors is a scientific necessity for adequately addressing questions of specificity, yet it comes at the cost of producing what are sometimes atypical groups for analysis.

Alcoholism in the family statistics Alcoholism in the family what you can do Alcoholism in the family system Alcoholism in the family article Alcoholism in the family essay Alcoholism runs in the family Alcoholism in the family family roles Alcoholism eff

It should be noted that no assessments of the degree of exposure to the family stressor were made. For example, the actual chronicity of the stressor or age at which the participant left the parental home could greatly influence the impact of the stressor on the participant and his or her family environment more generally. Although one might expect that such factors are randomly distributed across the different stress groups, these factors represent an important co-variate for future investigations of the effects of family stressors.

The assessment of family stressors in this study was not exhaustive. A number of childhood adversities that have been associated with parental alcoholism, such as parental violence and abuse, neglect, and parental criminality were not assessed. It may be those COAs who also experience these childhood adversities that exhibit the most disrupted psychosocial adjustment.

In conclusion, this study indicates that there is not a deterministic relationship between parental alcoholism and disrupted family environments. In this sample, the environment in COAs' families of origin were no different than those of other family stress groups, nor a non-stressed control group (bearing in mind that parental alcoholism was not confounded with any other family stressor in this sample). A beneficial route for further research may not entail comparing COAs' families with those of other groups, but rather to determine the exact combination of factors that translates parental alcoholism into dysfunctional family environments, and subsequent psychosocial problems among offspring.


Posted by at 02:26PM

Tags: alcoholism in the family statistics alcoholism in the family what you can do alcoholism in the family system alcoholism in the family article alcoholism in the family essay alcoholism runs in the family alcoholism in the family family roles alcoholism eff

Overall rating page: 3.14/5 left 67897 people.


There are no comments for this post "SPECIFICITY OF DISRUPTED PROCESSES IN FAMILIES OF ADULT CHILDREN OF ALCOHOLICS, Alcohol and Alcoholism, Oxford Academic". Be the first to comment...

Add Comment