Kaminski, Valle, Filene, and Boyle (2008
), in a meta-analysis of 128 parent training programs, evaluated a variety of content, delivery settings, and delivery techniques that targeted a variety of parents who were either experiencing behavioral problems with their children or at risk of experiencing behavioral problems. They asked what components of these parent training programs were most effective in helping parents be more successful with their children and found that the “program components consistently associated with larger effects included increasing positive parent-child interactions and emotional communication skills” (p. 567) as well as teaching specific parenting tools and having parents practice during class what they were learning. To increase positive parent-child interactions, particularly if the goal is to develop prevention programs, it is helpful to understand the mechanisms of how childhood problems develop (Frick & Morris, 2004
). Research is also needed to clarify what mediators and moderators influence parent-child interactions when children develop childhood disorders (Eyberg, Nelson, & Boggs, 2008
; Silverman & Hinshaw, 2008
Unfortunately, large-scale intervention programs designed to improve positive parent-child interactions would be prohibitively expensive if they used professionals for in-person assessment and intervention. However, intervening from a childhood temperament perspective offers a possible solution to this problem, because childhood temperament can be measured with parental questionnaires and parents can receive computer-generated feedback about how to respond more effectively to their child's specific temperament-related behaviors. In that regard, the present study examines the utility and outcomes of a written temperament-based anticipatory guidance program for parents.
Four decades of evidence, including two longitudinal studies, document a robust relationship between the innate behavioral patterns of childhood temperament, parenting, and subsequent childhood behavioral problems (Cameron, 1978
; Guerin, Gottfried, Oliver, & Thomas, 2003
; Laucht, Schmidt, & Esser, 2008
; Prior, Sanson, Smart, & Oberklaid, 2000
; Thomas, Chess, & Birch, 1968
). In addition, research shows that childhood temperament and parenting influence each other, both as distinct influences and through interactive effects on a child's behavior (Putnam, Sanson, & Rothbart, 2002
). Therefore, if parents have a clearer understanding of their child's temperament, the parent-child relationship and the child's well-being should improve.
More recent research has linked temperament to specific childhood psychological diagnoses. This includes the externalizing diagnoses of attention deficit hyperactive disorder (ADHD; Bussing, Lehninger, & Eyberg, 2006
) and oppositional defiant disorder or conduct disorder (Campbell, Shaw, & Gilliom, 2000
; Rettew, Copeland, Stanger, & Hudziak, 2004
), as well as the internalizing diagnoses of anxiety (Goodwin, Fergusson, & Horwood, 2004
; Inca, Benga, & Fox, 2006
) and depression (Pitzer, Jennen-Steinmetz, Esser, Schmidt, & Laucht, 2011). Though this is known, Muris and Ollendick (2005
) have argued that “research on … the predictive value of temperament on the development of psychological problems [is] urgently needed” (p. 284). Frick (2004
) and Shiner and Caspi (2003
) have pointed to the lack of research integrating temperament and childhood psychopathology and the importance of understanding the different ways temperament influences psychopathology, to develop more effective treatment programs for children.
When measuring temperament with questionnaires, parent and observer ratings do not necessarily agree (Stifter, Willoughby, & Towe-Goodman, 2008
). While this has led some to question the validity of parent ratings, “it would be more realistic to take a components of variance approach. Parents, as well as observers, vary in the degree to which subjective and objective factors contribute to their ratings” (Stifter et al., p. 423). Rothbart and Bates (1998
) concluded that parent ratings “provide a useful perspective on the personality of children … and … have established a fair degree of objective validity” (p. 126). Given that parent perceptions of childhood temperament affect parent-child interactions (Rothbart & Bates, 1998
), parents’ ratings of their child's temperament have clinical value in a prevention program.
Beginning in 1979, we conducted a series of studies to investigate if providing anticipatory guidance to parents could improve the parent-child relationship beyond the efforts of normal pediatric care, which may not include advice based on the child's temperament. We generated advice for parents based on the nine temperament scales developed by Chess and Thomas (Thomas, Chess, & Birch, 1968
). We anticipated that guidance tailored to each child's unique temperament could help parents understand and cope with temperament-related behaviors better than either general parenting advice or generalized information about temperament.
In a previous study we found that parents reported a greater understanding of temperament-related behaviors when they received anticipatory guidance. This advice was also rated as more useful by parents who reported a higher incidence of problem behaviors (Cameron & Rice, 1986
). However, questions still remained whether such guidance would actually benefit children, by either improving the parent-child relationship or reducing the occurrence of problematic temperament-related behaviors. To further evaluate the feasibility of providing effective temperament-related anticipatory guidance, we conducted the present study and included (a) more detailed parent evaluation measures and (b) the frequency of children's visits with internalizing and externalizing diagnoses to a large health maintenance organization (HMO).
We expected effective anticipatory guidance would improve the parent-child relationship by making parents sensitive to temperament and to temperament's influence on their child's behavior. Therefore, our first hypothesis is that the parents in the intervention group will report the guidance helped them see their child more clearly, understand how temperament contributes to their child's behavior, and change their expectations of what is normal for their child's behavior. We also expected intervention parents would rate the support positively and recommend the program to others.
Given an improvement in the parent-child relationship, we expected intervention parents’ children would have a decrease in visits to their health maintenance organization (HMO) associated with temperament-related behaviors, either because the guidance helped parents more effectively manage those behaviors or because an improved parent-child relationship led to fewer temperament-related behavioral problems. However, a decrease in visits to an HMO should not be due to parent's becoming discouraged from seeking needed treatments for their children. Therefore, our second hypothesis is that children in the intervention group will have fewer behavior-related visits than the control group, and the intervention group parents will report no changes in the amount of advice sought from their HMO compared with the control.
Parents who report children with harder-to-manage temperaments are more likely to have trouble understanding and managing their child (Crockenberg, 2008
) and may think there is something wrong with their child or their parenting. We anticipated that our individualized guidance would have a greater effect for these parents. Therefore, our third hypothesis is that the effectiveness of the intervention will be greater for children with harder-to-manage temperaments.