Resilient Children of Parents Affected by a Dependency

Conclusions and Recommendations

In the Area of Research

Methodological Aspects

The key contribution of research into resilience is the attention paid to the factors which contribute to success despite the presence of risk factors. However, research on resilience has often ignored resilience or protective factors (limited to factors exerting a moderating effect, as discussed earlier) or confused them with beneficial factors which exert a compensatory effect with regard to risk factors. Criticisms have also been leveled against the weakness of research designs used: on this issue, research into resilience is condemned to produce biased results and uncertain conclusions if it avoids longitudinal designs which track subjects from an early age, or if it fails to use reliable measurement instruments, appropriate statistical analysis, or population samples which are sufficiently large or composed of groups matched on the basis of relevant control variables. The research into resilience has also been criticized for the murkiness which surrounds the concept of successful adaptation in resilient children, as well as the failure to integrate various levels of analysis which might serve to explain the mechanisms of resilience (Luthar and Cicchetti, 2000a; Masten, 2001).

One of the important challenges facing research into resilience is clearly to rigorously define successful adaptation, rather than view the absence of adjustment problems as the sole criterion for resilience (Jessor, Van Den Bos, Vanderryn, Costa and Turbin, 1995; Rutter, 1999). It should be said, however, that some authors have added a second criterion, namely the acquisition of competencies (Masten and Coastworth, 1998). These two criteria are difficult to reconcile, since an individual can demonstrate competencies in certain areas of functioning while experiencing problems in other areas. Moreover, the concept of successful adaptation is dependent upon social norms which vary according to ethnicity, cultural habits and geographic location (Werner, 1993). Resilience criteria must also be adjusted according to the area of functioning and to the specific period of development. They do not apply uniformly to all areas of functioning, nor to all periods of development, particularly when norms vary. It may well be that children are resilient in certain facets of their personalities, in certain aspects of their experience, or during certain periods of their development only (Cicchetti et al., 1993; Luthar et al., 2000). While some studies have demonstrated that resilient children and adolescents tend to develop, as a general rule, into well-adjusted adults, the continuity of resilience cannot be taken for granted (Masten and Curtis, 2000; Werner, 1995). As some authors have found (Cicchetti, 2003), it is important to view resilience or protective factors not as the absence of adjustment problems at a given point in life, but rather at several sequences spread over a lengthy period of development (infancy, childhood, adolescence, early adulthood, and so on). These various evaluation periods could serve to establish developmental profiles or trajectories for determining success or lack thereof during a complete stage of development (for illustration, see Vitaro, Carbonneau, Tremblay and Gosselin, in press). As such, it has been proposed to establish such development profiles for the various relevant areas of functioning and at every stage of development. Particular attention should be paid to transitional periods between two stages of development, since the new challenges that arise at these times can often lead to problems of adaptation.

As mentioned earlier, with respect to the research into resilience, problems arise with the equivalency criteria of risk factors to which were exposed resilient and non-resilient study groups (Rutter, 2000). These factors are not always clearly defined and ignoring or neglecting the perceptions of the individuals concerned or the mechanisms and other parameters related to their mode of action (Luthar and Cushing, 1999). Furthermore, the same risk factors emanating from the socio-familial environment do not always have the equivalent harmful effect on the adaptation process, since some children are genetically better equipped to deal with them (Caspi et al., 2002, 2003). It may be, for example, that being genetically spared certain vulnerability factors can completely or partially block the psychosocial chain reactions which are the precursors of adjustment problems.

Multiple Levels of Analysis

Future research into resilience should integrate multiple levels of analysis, the first being genotype analysis, as informed by molecular and quantitative genetics. This level of analysis would make it possible to detect elements of vulnerability and, when necessary to intervene at the level of the phenotypical manifestations which are likely to become exacerbated under the effect of stressful social experiences.

The second level of analysis should focus on the risk factors to which the child is exposed. Obviously, this level of analysis includes the factors associated with alcoholism and drug dependency of parents, but it should also take into account the phenomenon of comorbidity, which is to say the presence of other factors which can increase stress, such as parental psychopathology, deficient parenting skills and other similar variables (marital conflict, difficult economic situation, abuse, neglect etc.). This level must equally take into account the neurophysiological effects of prenatal postnatal alcohol, drug and tobacco use by parents.

A third level of analysis relates to the positive and negative social experiences to which children are exposed in the course of their development. This extends to the adults in extended families and with significant adults outside the family, as well as with siblings and peers. Cognitive and behavioural learning as well as the perceptions derived from these experiences are also important. Finally, this third level of analysis should also address the cognitive and socio-cognitive dispositions and personality traits of the child, all of which should be taken into account as early as possible and throughout the child's development.

The fourth level of analysis is that of the processes which distinguish resilient children from others when all else is equal. This involves the microanalysis of transactional processes between the child and the socialization agents or experiences to which he or she is exposed (for example, the manner in which academic or social challenges are faced, or the socio-cognitive or affective messages derived from this experience in terms of personal competence and self-esteem). The fifth and final level of analysis considers how the child functions in various domains of experience in order to identify areas of successful, moderately successful, and unsuccessful adaptation at various stages in life, from early childhood to adulthood, with particular emphasis on the transitional periods between stages. This kind of follow-up should also consider new positive or negative experiences to which the child continues to be exposed to, as these can play as great a role as past experiences in influencing subsequent adaptation. In fact, a number of studies have demonstrated that individuals exposed to severe stress can nonetheless recover under favourable conditions. An eloquent example of this is the case of Romanian orphans adopted by functional families (Rutter, in press). Having said this, child rehabilitation is subject to strong individual variability; in some cases, after-effects of the accumulated delay experienced by children persist despite the high quality of the new environment.

Another factor which must not be overlooked in this type of study is gender. In fact, it would appear that boys are more vulnerable to the effects of socio-familial stress than are girls (Earls, 1987), although the results of some studies are rather equivocal on this score (Zaslow, 1988, 1989). Similarly, the age at which the child is exposed to risk or stress factors also warrants consideration. Although few studies have shed light on this aspect, children would appear to be particularly vulnerable between the ages of 2 and 6 years (Rutter, 1979). To gain a better understanding of this dimension, one would need to examine children of alcohol- or drug-dependent parents at different ages (including children of parents who have undergone rehabilitation).

In the Area of Intervention

Earlier, we proposed a generic prevention model which would help vulnerable children to acquire as many personal skills as possible and to integrate as many resilience-building experiences as possible, preferably within the family circle, as well as beyond that circle. However, as Rutter (in press) reminds us, resilience does not always manifest itself in every aspect of life. It is important then, to identify the specific areas in which a child's vulnerability expresses itself, in order to propose appropriate and effective strategies that will promote resilience in these areas. The necessary strategies and learning experiences can be put in place before, during or even after periods of exposure to risk factors, and their efficacy should be evaluated on an ongoing basis.

A rigorous evaluation in an experimental context seems crucial from a research and a clinical standpoint, for at least two reasons. First, if the acquisition of favourable competency skills conducive to resilience translates into better adaptive outcomes for children in the experimental group, as compared to those in an initially equivalent control group, it then becomes possible to:

objectively observe the desired changes (provided that different evaluation sources and valid measurement instruments are used over several measurement periods); and to
attribute these changes to the intervention strategies rather than to the usual sources of internal invalidity which include, among others, the passage of time, the repetition of measurement, and positive expectations.

Second, assessing the acquisition of favourable competencies and the integration of experiences which are conducive to resilience in the experimental group, as compared to the control group, constitutes the best and only means of verifying the causal role of these skills and experiences in terms of the subsequent adjustment of at-risk children (of alcohol- and drug-dependent parents). After randomly distributing at-risk children into an experimental group and a control group, the surest way of demonstrating the causal power of a resilience factor is to embed that factor within the experimental group and then to compare the level of adaptation of the children in the experimental group with that of the children in the control group. All else being equal, a better level of adaptation in the experimental group may be viewed as resulting from the manipulation of the resilience factor. On the other hand, the absence of differences between the experimental group and the control group would suggest that the presumed resilience factor is either factitious or that the intervention strategy used to embed it has been ineffective. Obviously, it may be necessary to "manipulate" several resilience factors concomitantly in order to achieve maximum efficacy. Under such circumstances, it would be impossible to determine which specific resilience factor is responsible for the positive effects. The risk of failure or of iatrogenic effects should constitute sufficient reason to use a control group while remaining mindful of ethical considerations (for as long as it takes to clearly establish the efficacy of the intervention strategies under consideration).

A number of authors advocate the use of prevention programs considered to be "experimental manipulations" of resilience variables, in order to determine the causal coefficient of these variables with respect to the chain of psychosocial developments which naturally lead to successful adaptation (Weersing and Weisz, 2002). By using prevention studies to verify whether presumed resilience (or compensation) variables effectively play a causal role in protecting vulnerable children from problems of adaptation, a complete cycle is created, from research to intervention and back again, and the benefits of prevention efforts are maximized.

Without funding for first-class research teams, as well as the active cooperation of practice setting (Treatment Centres, CLSC, schools) and without a clear and steadfast political will, these goals will not be achievable.

Summary table

• The next generation of studies on resilience should adopt a longitudinal perspective and integrate multiple levels of analysis.
• Workers dealing with adult drug users and adults with an alcohol dependency should also concern themselves with other family members, particularly children, who are at risk of becoming their future clients. CLSC workers, schools and other stakeholders should establish prevention programs based on the research into resilience and prevention, or order to intervene early on behalf of their clients' children. Many of these children will already display early signs of adjustment problems which may emerge in adolescence or adulthood.
• The presence of an alcohol- or drug-dependent parent should not be viewed as the sole risk factor when identifying vulnerable children. An appropriate assessment of children's psychophysiological, behavioural and cognitive makeup is also needed.
• The implementation and evaluation of prevention programs whose targets and strategies are derived from the research into resilience would help to break the cycle of intergenerational transmission of adjustment problems in alcohol- and drug-dependent families. It would also provide a means of testing the causal role of the resilience factors which lead, in principle, to the development of the programs. At present, all of our knowledge about risk and resilience rests on correlational studies.