Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
2 Approach to Research and Its Evaluation SUMMARY Research Challenges in Confronting Depression in Parents â¢ The challenges for researchers, clinicians, and policy makers in attempting to address the problems associated with the care of depression in parents include the integration of knowledge, the application of a developmental framework, conceptualizing the problems in a two-generation nature, and acknowledging the pres- ence of the constellation of risk factors, context, and correlates associated with depression. Issues Considered in Searching the Literature â¢ To fully understand the linkages among depression, parenting, and the child health outcomes, researchers should consider issues sur- rounding (but not limited to) the definition and measurement of depression and parenting, the etiology of depression, timing and use of appropriate screening interventions, the process of risk and resilience in children of depressed parents, correlates of depression, and developmental processes and time points. 43
44 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN Challenges in Evaluating the Literature â¢ Researchers face multiple methodological challenges studying de- pression in parents and its effects on parenting practices and child health outcomes that need to be addressed in order to provide recommendations for the development of future research, inter- ventions, and policyâincluding conceptual frameworks, sampling designs, data analysis, and integration of research findings across literatures. ____________________ In this chapter, the committee describes their approach to the litera- ture on the effects of parental depression on parenting practices and child outcomes and its evaluation. The chapter is organized in three sections, relating to the challenges that researchers face in confronting the problem of parental depression, the wide range of issues that we considered relevant, and standards of evidence and methodological issues that are important to keep in mind in reading this report. Some topics are addressed in more than one section, but they are focused on different aspects of the topic. For example, in the section on research challenges, we show that a conceptual framework relating to the effects of parental depression on families should be guided by a developmental psychopathology perspective. Later, in the section on research standards, we mention what the literature has shown in this regard and that research relating to any psychopathology should ad- dress questions âacross generations and across timeâ (Hinshaw, 2008). The type of evidence and criteria used to judge the importance of that evidence vary from area to area. This chapter does not attempt to explicitly summarize the specific criteria used for the evaluation of the evidence in each area, but instead offers a guideline of the general areas of interest and inquiry that the committee used when the committee searched and evalu- ated the literature. For example, studies of screening for parental depres- sion are different from studies of treatment and intervention, and these are different from studies of prevention programs. These are also different from inquiries relating to changes in policy at the macro level or the avail- able studies on the effects of parental depression. Thus, in this overview on standards of evidence and methodology, we present general guidelines that the reader should apply when appropriate in the subsequent chapters. Recommendations based on the evaluation of the evidence in each area are presented.
APPROACH TO RESEARCH AND ITS EVALUATION 45 RESEARCH CHALLENGES IN CONFRONTING DEPRESSION IN PARENTS Four themes emerged from the committeeâs review of the research litera- ture and discussions with service providers, policy makers, and stakeholder organizations: (1) the integration of knowledge regarding the dynamics of parental depression, parenting practices, and child outcomes; (2) the need to recognize the multigenerational dimensions of the effects of depression in a parent; (3) the application of a developmental framework in the study and evaluation of the effects of parental depression; and (4) the need to acknowledge the presence of the constellation of risk factors, context, and correlates of parental depression. Each of these is discussed in turn. Integration of Knowledge Depression is a complex disorder that is affected by biological, psycho- logical, behavioral, interpersonal, and social contextual processes. Examin- ing the effects of depression on parenting and the development of children requires the integration of knowledge from research, practice, and service delivery from multiple scientific, health care, and social service disciplines. Major scientific and integrative advances have emerged that point the way for future research and practice, but the challenges presented by the integra- tion of these widely varied disciplines are enormous. Similarly, addressing parental depression as a public health problem requires multiple points of access for delivery of services and multiple types of intervention and care. The approach we have taken to the problem of depression, parenting, and the healthy development of children has been interdisciplinary and transdisciplinary from the outset. The composition of the committee reflects multiple disciplines, including pediatrics, psychiatry, psychology, nursing, pharmacology, biostatistics, obstetrics, public health, behavioral health, and community and family medicine. We sought input and information from experts in a wide range of practice and community settings to consider the broad context in which services are delivered to parents with depression and their children. The range of professionals and settings potentially concerned with de- pression in parents and their children is vast. Among health care services, it includes mental health professions (psychiatry, psychology, psychiatric nursing, social work), primary care, family practice, obstetrics, and pediat- rics. Among family and community support systems, it includes extended families, religious institutions, and formal and informal community sup- ports. Among policy makers, it includes program administrators running home visiting programs, Medicaid officials, mental health administrators, and others with responsibility for fiscal resources that could be reallocated. Much like the divisions that characterize scientific disciplines, we have iden-
46 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN tified barriers between the delivery (formal or informal) of interventions to parents with depression and delivery of interventions for children and families. Our agenda included the identification of ways to develop new strategies to bridge the different cultures of health care delivery, community and family support services, and policy makers. Transdisciplinary Approaches to Research, Training, and Service Delivery Although the need for integration and synthesis across these disci- plines is great, the breadth of knowledge and practice that is relevant to understanding and addressing depression and parenting is a double-edged sword. On one hand, depression is a complex disorder that is affected by biological, psychological, behavioral, interpersonal, and social contextual processes. Major scientific advances have been made in all of these areas, and integrative perspectives are emerging to guide the synthesis of current findings and point the way for future research and practice. Similarly, ad- dressing the public health problem of parental depression requires multiple points of access for delivery of services and multiple types of intervention and care. On the other hand, the challenges presented by the integration of these widely varied disciplines are enormous. Much of the work to date has been carried out in separate scientific entities, divided by different meth- odologies, levels of analysis, terminology, and agendas. The committee has purposefully attempted to break down these barriers in our approach by looking for examples of research that have integrated multiple disciplines and opportunities to move forward with new transdisciplinary work. Over the course of our work, we began to understand that depression in parents is a topic with no homeâit is everywhere and nowhere. This is partly a consequence of the many disciplines and fields that are concerned with de- pression in parents. With this perspective in mind, we undertook the task to begin to build a âhomeâ with this report. As outlined by the National Institutes of Health, a transdisciplinary approach involves integrative science that spans basic and clinical research from multiple disciplines (e.g., genetics, medicine, behavioral science) and can generate rapid new developments in conceptual models to initiate breakthroughs and speed the translation of findings in basic research into practice and policy. This involves the translation of research from basic science into the development and testing of novel intervention approaches and, conversely, the translation of intervention findings back into basic sci- ence studies that can identify and model mechanisms of risk and etiology. Translational and transdisciplinary research is bidirectional in nature, with basic science informing interventions and intervention findings stimulating further basic research inquiry. A number of recent examples of translational research are relevant to
APPROACH TO RESEARCH AND ITS EVALUATION 47 the understanding of depression, parenting, and children. For example, there is growing recognition of the role of multiple biological processes that are relevant to the etiology and course of depression and to the comorbidity of depression and medical conditions, most notably coronary heart disease (Miller and Blackwell, 2006). This work has come from the synthesis of relatively independent lines of research on the role of stress in proinflam- matory processes, coronary heart disease, and depression. Similarly, the integration of methodologies and findings from research on genetics and brain structure and function has led to significant advances in knowledge of the underlying neurobiology of depression (Hariri and Brown, 2006). The translation of basic behavioral research on parenting processes and their effects on childrenâs development has provided a foundation for un- derstanding the ways in which parenting may be disrupted by depression and the subsequent impact on children (Bornstein, 2006). Linking Research and Practice Throughout this volume we emphasize the importance of the pipeline from the discovery of underlying mechanisms of risk in children of de- pressed parents to the delivery of services and to the development of policy (Abrams, 2006). This includes, as Abrams notes, âthe science of dissemi- nation along with improving the dissemination of evidence-based scienceâ (p. 515). The dissemination of research findings to practitioners and, con- versely, the application of the experience and observations of practitioners to research are significant challenges in mental health research and practice (Barlow, 1996; Weisz, Jensen, and McLeod, 2005). This complex process is even more complicated in the case of depression in parents as it spans research and practice with adults (parents), children (from infancy through adolescence), and families. Establishing links between research and practice involves the interactions among the full range of health care services and family and community support systems that may intervene with depressed parents or their children. A Two-Generation Approach The evidence of the effects of depression on the parentsâ parenting skills and on their childrenâs health and development makes a compelling case for fostering this integration through such a two-generation approach (parent and child) at every level, including systems that provide care and services, efforts to more fully understand the epidemiology and effects of parental depression, and future research on intervention approaches. We make refer- ence to this two-generation approach throughout the report. We also recognized the frequent need for a three-generation perspec-
48 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN tive, taking into account grandparents who may be thrust into the child rearing role. However, we determined that, despite their importance in understanding and working with parental depression, the three-generation approach as well as concerns about sibling relationships was outside the scope of this report. We were able to glean useful information on the role of marital relationships in the occurrence and treatment of depression (some of which is mentioned later in this chapter) and in the effects of parental depression on children (included in the summaries in Chapter 4). Developing an explicit focus on the nature and quality of behaviors and relationships between parents and children challenges the traditional division of responsibilities in medicine and mental health practices and in many service settings. Both facilities and professionals traditionally special- ize in either adults or children (exceptions to this practice include family medicine and, in some settings, combined medicine and pediatrics). Primary care, mental health, and community service systems that see adults with depression do not routinely consider whether those adults have children, nor do they provide links to services that focus on parenting skills and on the needs of their children. Although increased attention to postpartum depression has led to some changes for women and infant children, most systems that provide care for children are not well equipped to identify parents with depression or at risk for it or to refer them to appropriate services for adequate treatment. A vast research literature on depression examines its epidemiology, etiology, correlates, and treatments among adults, but, for the most part, this body of work does not identify whether or not these adults are parents. Thus, the literature and national surveys on etiology, prevalence, screening, prevention, and treatment of depression in adults are often not specific to depression in parents and do not include measures of its effects on their children. This limitation makes it difficult to gauge the true extent of the problem. An extensive body of research on parenting practices and parent and child relationships highlights effective interventions that support positive outcomes for children. Yet with regard to depression in parents, only a few intervention studies have adopted such a two-generation approach, which focuses on the relationships and interactions between parents with depression and their children. In addition, interventions with good evidence for improving child outcomes rarely consider the influence of parental de- pression on their effectiveness and even more rarely examine the potential effects of including program components that identify and refer or treat parents with depression. When this has been done, improvements in effec- tiveness have been noted, such as in the work of Kazdin and Sanders and colleagues (Kazdin, 2005; Sanders et al., 2000). In our work, we noted increasing interest in understanding how and
APPROACH TO RESEARCH AND ITS EVALUATION 49 when improvements in parentsâ depression lead to improvements in par- enting and, even further, in child functioning. We include an overview of studies of treatment for depression in parents as a preventive intervention for their children (Chapter 7). Equally interesting, although less often studied, is how and when inter- ventions designed to improve qualities of parenting or to improve couplesâ relationship issues also lead to improvements in the parentâs depression and potential benefits for their children. We recognized these approaches as reflecting the two-generation perspective, which we judged to be a very promising approach to the problem of depression in parents. In terms of parenting interventions, a recent review of parenting train- ing interventions noted consistent evidence that depression in parents im- proves as a secondary benefit of such training, even when the depression per se is not a target of the intervention (Kaminski et al., 2008). Although the authors note that some of the studiesâ designs did not always allow for ruling out that the depression may have improved even without the inter- vention, the findings are promising. Other studies targeted mothers with depression and intervened directly in their qualities of parenting (Cicchetti, Rogosch, and Toth, 2000; Gelfand et al., 1996; Lyons-Ruth et al., 1990). In terms of couplesâ relationship problems and discord, behavioral ap- proaches to marital therapy have been found to be more effective for the treatment of depression than standard cognitive behavioral therapy target- ing the depression per se (Baucom et al., 1998). We also found promising the new generation of marital therapies being designed to treat depression based on knowledge of the associations between depression and marital problems (Beach, Fincham, and Katz, 1998). We also noted a recent meta- analysis that found inconclusive evidence for the efficacy of couple therapy as a treatment for depression yet concluded that couple therapy may still be the treatment of choice when relational distress is predominant (Barbato and DâAvanzo, 2008) A Developmental Framework The implementation of a two-generation approach involves many chal- lenges. Childhood is a period of rapid change and development, and the presence or absence of supportive parental behaviors can have lasting ef- fects if they interfere with the children getting their stage-salient needs met. The impact and consequences of depression in parents can vary by both the age of the children in the family as well as the length, severity, and history of exposure to parental depression. In the course of our work, we recognized the value of a developmental approach. That is, we understood that multiple aspects of concern about depression in parents would vary by the age of the children in the family at
50 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN any particular point in time of studying these families. We also recognized the importance, and the challenge, of understanding any childâs history of exposure to depression in a parent. We expected that the age of the children at the times the parent was depressed would influence the particular risks to their development, the resiliencies that they would be able to bring to bear, the mechanisms by which those risks might be transmitted, the challenges to parenting with depression, the approach to screening for depression, and the nature of the preventive or other intervention that might be found to be effective. Although we reviewed literature that suggests reasons to be concerned and to intervene regardless of the childrenâs ages, we also noted a compelling case for the youngest children. Most important was the un- derstanding that development needs to be taken into account. A Developmental Psychopathology Perspective We were also guided by a developmental psychopathology perspective, which conceptualizes risk for the development of psychopathology in terms of processes that extend through time and are considered in the context of normal developmental processes (Cicchetti and Schneider-Rosen, 1984; Sroufe and Rutter, 1984). Understanding the mechanisms of risk in chil- dren of depressed parents requires understanding the processes underlying individual patterns of adaptation and the consequences of the individual patterns for the development of depression or other problems. Early in their developmental pathways, children of depressed mothers may develop vulnerabilities for depression, which, in turn, increase the likelihood of developing depression. Given the stage-salient needs of infants, of particu- lar concern have been limits on depressed mothersâ abilities to provide the sensitive, responsive care needed for the development of healthy attachment relationships (Egeland and Farber, 1984) and emotional self-regulation (Tronick and Gianino, 1986). We recognized the value of knowledge of childrenâs potential vulnerabilities and developmental needs to inform screening, prevention, treatment, and policy. Depression and Parenting Children Across Development Even when one takes a developmental perspective, it is often ignored that, among women with depression, it occurs during pregnancy with at least as high a rate as in the postpartum period (Evans et al., 2001). Ante- natal depression may be an early life stress that alters fetal development of stress-related biological systems (especially hypothalamic-pituitary-adrenal functioning), retards fetal growth, constricts the length of gestation, and increases risk of obstetrical complications (Van den Bergh et al., 2005). Concerns about the fetus suggest potential long-term implications for the
APPROACH TO RESEARCH AND ITS EVALUATION 51 development of problems in cognitive and social-emotional functioning. We therefore sought to expand on the typical perspective of parenting with depression to include the prenatal period. Infants and young children are dependent on sensitive, responsive care- giving in order to develop skills in emotion regulation, interpersonal skills, and stress response mechanisms, problems in any of which may disrupt the earliest stages of development, predisposing children to the later develop- ment of depression. Consistent with these understandings, a meta-analytic review identified a larger body of research suggesting that depression is even more strongly associated with parenting problems in parents of infants compared with parents of older children (Lovejoy et al., 2000). Another literature shows that brain development continues at a rapid pace at least for several years after birth (Chugani and Phelps, 1986), and its course of development can be strongly influenced by early life stressors, such as might be associated with having a mother with depression. Thus we were particu- larly concerned about depression in parents of young children. Despite compelling reasons for concern about pregnancy and infancy and early childhood, we recognized that they are not the only developmen- tal periods of concern in relation to depression in parents. For example, depressive symptoms and depressive disorder interfere with the tasks of par- enting in parents of school-age children and adolescents (e.g., Du Rocher, Schudlich, and Cummings, 2007; Gerdes et al., 2007). Disruptions in par- enting associated with depression in parents (e.g., parental withdrawal, in- trusiveness, and irritability) are related to higher levels of internalizing and externalizing problems in adolescents (e.g., Jaser et al., 2005). Furthermore, certain aspects of brain development, particularly in the prefrontal regions responsible for executive functions and emotion regulation, continue on into young adulthood and are vulnerable to the effects of chronic stress throughout childhood and adolescence (e.g., Nelson et al., 2005; Romeo and McEwen, 2006). These lines of research suggest that the risks associ- ated with depression in a parent extend well beyond early infancy and early childhood. Depression Does Not Exist in a Vacuum: Bringing Comorbidities, Correlates, and Context to Light Depression rarely occurs in isolation. Depressed adults often face other health problems and psychopathologies (such as substance use disorders, anxiety disorders) and stressful social environments (such as poverty, be- reavement, social isolation). Those risk factors may be essential to under- standing the problems in children of depressed parents. The associated features also may be essential in identifying approaches to treating parentsâ depression and also improving the quality of parenting and childrenâs
52 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN outcomes. Such recognition might involve, at a minimum, assessing the constellation of risk factors and also evaluating the relative contribution of the multiple disorders to the problems with parenting and with child functioning. These risk factors may also affect how and where services can best be provided. A long history of research supports the theory that depression is both exacerbated by problems in interpersonal relationships and also contributes to them (Joiner, Coyne, and Blalock, 1999). This includes the finding that living with a parent with depression is stressful, not only in relation to problems with parenting but also in relation to the parentâs problems with other relationships (Hammen, 2002), and children struggle to cope with these stressors (Compas et al., 2002). Attention to the context of the lives of women with depression suggests concern for the children in terms of their exposure to a wide range of stressors (Hammen, 2002). More broadly, the methodological issue this raises is that depression in parents may be an essential marker of risk for problems in children but not necessarily a âmain effect.â Whereas depression is known to be comorbid with anxiety disorders and substance abuse, in particular, and these are addressed in this report, we also addressed the potential importance of other correlates and contexts, including intrafamily violence, child abuse, and neglect. This report shows that the comorbidities, correlates, and con- texts may function as independent risk factors for adverse child outcomes or as modifying factors, which exacerbate the effects of parental depres- sion. Yet research on the effects of parental depression on parenting or child functioning rarely provides adequate tests of the role of co-occurring conditions. Integrating the Comorbidities, Correlates, and Context into Approaches to Treating or Preventing Depression in Parents Understanding the dynamics of how parental depression and its comor- bidities, correlates, and context impact child outcomes are important to the design and implementation of successful future service delivery models for families affected by parental depression. We recognized the tendency of interventions to focus on depression in the parent as a risk factor for the development of psychopathology in the children without considering that the families might best be characterized by the broader set of risk and pro- tective factors we bring to light here. Moreover, if depression is secondary to other overriding issues in the family, such as poverty, violence, marital or couple distress, issues related to immigration, acculturation, etc., then the implications for screening, prevention, and treatment are quite differ- ent than if one were addressing depression as the primary problem. At a minimum, one would need to assess co-occurring conditions and also
APPROACH TO RESEARCH AND ITS EVALUATION 53 evaluate the relative contribution of the multiple conditions to the problems with parenting and with child functioning. Once carefully assessed, the as- sociated features are likely to suggest different approaches to treating the parentâs depression, improving the quality of parenting, and in preventing negative outcomes in the children, relative to the situation in which depres- sion is the primary problem. We also recognized that comorbidities may affect how and where services can best be provided. Integrating Services in Different Settings for Diverse Populations The methodological issues raised by comorbidities, correlates, and context also have implications for where depressed parents might receive services. In our approach to the problem of depression in parents, we rec- ognized that the services for comorbidities, such as substance abuse, or cor- relates and contextual variables, such as family violence or issues related to immigration, are typically offered in settings other than where mainstream mental health services are offered. We are also aware that many of these problems related to depression in parents first come to light in primary care settings, such as family practice, obstetrics, and pediatrics. Thus, another issue concerns strategies that can bridge service cultures and institutional networks. Needs of Vulnerable Populations As this understanding of the context in which depression reveals, cer- tain populations are particularly at risk. The committee thus worked to determine that we were including literature that addresses the needs of a wide range of populations, including vulnerable populations, for example, first- and second-generation immigrants and refugees and others, espe- cially those who are known to be at risk, including those who live in pov- erty. Learning about correlates of psychiatric disorders among vulnerable populations is of critical importance to inform clinical practice and guide program development. Although current research has not yet covered the specific topic of parental depression among vulnerable populationsâwith the exception of a few qualitative studies (e.g., Lazear et al., 2008)âfor the purpose of this report it is important to point to major studies of mental illness in specific racial and ethnic populations that may provide significant insights into prevalence rate differences, help-seeking behaviors, and pat- terns of use of mental health services. We identify below some of the largest and most recent psychiatric epidemiological studies that have been used in studies described throughout the report in order to help provide insight into how specific racial and ethnic populations use and need mental health services in Box 2-1.
54 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN BOX 2-1 Summary of Large Psychiatric Epidemiological Studies in Racial and Ethnic Minority Populations â¢ Mexican American Prevalence and Service Survey U.S.-born Mexican Americans and Mexican immigrants Residents of Fresno, California Ages 18â59 National Survey of American Life frican Americans, black immigrants from the Caribbean (recent and sec- A ond-generation), non-Hispanic whites Prevalence of mental health disorders, use of mental health services â¢ National Latino and Asian American Study Hispanics, Asian Americans, further stratified by ethnic subgroups Ages 18 and older Similarities and differences in mental illness and service use â¢ National Epidemiologic Survey on Alcohol and Related Conditions .S.-born Mexican Americans, Mexican immigrants residing in the United U States, U.S.-born non-Hispanic whites, and non-Hispanic white immigrants residing in the United States Considers immigration status in conjunction with psychiatric morbidity Doorways into the Domain of Depression in Parents Research has highlighted two key facts relevant to our task of evaluat- ing how best to approach the problem of depression in parents. First, most adults with depression do not get treated for it (Kessler et al., 1999; Kessler, Merikangas, and Wang, 2007; Narrow et al., 1993; Regier et al., 1993). The large-scale, psychiatric, epidemiological study by Kessler et al. revealed that less than one-third of adults with major depression or dysthymia used either general medical or specialty outpatient mental health services in the previous year. Second, the one-third of adults of parenting age with depres- sion who do get treated uses a wide range of alternative points of contact. Yet given the number and multiplicity of depressive disordersâ comorbid mental and physical health problems (e.g., substance abuse, chronic pain), impairment (e.g., problems getting to work), and co-occurring conditions (e.g., poverty), one promising avenue to the identification of parents with depression is to focus on those who are seeking treatment or assistance for those associated conditions. These alternative avenues to identifying parents with depression include a wide range of formal and informal service settings, which offer important points for screening and identification. Also, the children themselves may be
APPROACH TO RESEARCH AND ITS EVALUATION 55 an important avenue for identifying parents with depression. For example, in Finland, a national program involves educating children and adolescents about mental health problems in adults and how children can cope with depression in their parents (Solantaus and Toikka, 2006). More broadly, the committee recognized that children being screened for developmental delays or evaluated or treated for emotional or behavior problems, school behavior problems, or even chronic illnesses or injuries present a typically neglected opportunity to screen for depression in parents. Finally, in the realm of policy relating to parental depression and its effects, we conducted interviews with states leading initiatives on mater- nal depression (Illinois, Iowa, New Jersey), seeking evaluation data when they were available, drawing on primary data from two other studies conducted by the National Center for Children in Poverty (Cooper et al., 2008; Knitzer, Theberge, and Johnson, 2008). We also reviewed other lit- erature on program initiatives, as well as the policy literature, to extract, whenever possible, the policy implications. To assess the feasibility of the recommendations, we asked four basic policy-relevant questions (Kingdon, 1995): Will it promote efficiency? Will it promote equity? It is politically feasible? Does it seem doable from a cost perspective? Finally, we relied on our own understanding of how policy processes work and what kinds of recommendations are likely to be persuasive. ISSUES CONSIDERED IN SEARCHING THE LITERATURE To fully understand the linkages among depression, parenting, and chil- drenâs development, we considered a wide range of research. This includes research on but not limited to the definition and measurement of depres- sion and parenting practices, the etiology of depression, timing and use of appropriate screening interventions, the processes of risk and resilience in children of depressed parents, correlates of depression, and developmental processes and time points. Definition and Measurement of Depression and Parenting Depression presents a number of challenges in its definition and mea- surement. Foremost of these is the distinction between depression as a categorical psychiatric diagnosis and dimensional approaches that consider symptoms of depression as a matter of degree along a continuum. Diagnoses of depression are derived from psychiatric interviews, whereas dimensional approaches are assessed through symptom inventories and questionnaires. We have considered both in our review and have offered suggestions for the integration of what are often seen as contrasting, rather than comple- mentary, approaches. Furthermore, depression is a heterogeneous disorder
56 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN with varying patterns of symptom presentation, chronicity, duration, and severity. Similar issues were encountered in our review of parenting. The central dimensions of parenting (e.g., warmth, structure) are addressed, and we provide consideration of the characteristics of effective versus impaired parenting. However, there is far less consensus regarding the measurement of parenting than the measurement of depression. At the same time, it is important to recognize that âparental depres- sion,â âmaternal depression,â and even âpostpartum depressionâ are not distinct diagnostic entities from depression as it occurs in any adult. Perina- tal depression is not a separate diagnosis according to the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV) (American Psychiatric Association, 1994). Typically, clinically significant perinatal depression is operationalized as major depression, a diagnosis that requires meeting a set of symptom, duration, and impairment criteria as defined by the DSM-IV. Under Depressive Disorders in the DSM-IV, pregnancy is not given any special consideration; however, âpostpartumâ is an added quali- fier to refer to the timing of a major depressive episode. It is thus important to recognize that when we refer to parental, maternal, or paternal depres- sion, we refer to depression defined in the same ways as depression that occurs at other times in peopleâs lives. Etiology of Depression We have examined the broad literature on the etiology of depression in parents and in their children. This includes genetic, biological, and interper- sonal processes that contribute to depression in parents as well as in their children. Recent research has begun to illuminate the complex interactions and combinations of biological and social processes that are associated with increased risk for depression. Examples of integrative research on the etiology of depression include studies of the interaction of stress and genetic liabilities, such as mutations of the genes responsible for serotonin transport in the brain (e.g., Caspi et al., 2003) and research showing that traumatic stress early in development can lead to a cascade of biological processes that increase the risk for depression (Gilespie and Nemerooff, 2007). Screening for Depression in Parents The challenge in evaluating the literature is that screening measures are often used without validation by other methods. While very brief and longer symptom checklist measures have been shown to perform similarly as screening tools to detect those at risk of major depression (Whooley et al., 1997), their positive predictive value for a major depressive disorder varies according to the prevalence of depression in that population. Screen-
APPROACH TO RESEARCH AND ITS EVALUATION 57 ing measures typically detect 75 percent with any depressive disorder and 25â40 percent with a major depressive disorder (Kroenke, 2006; U.S. Pre- ventive Services Task Force, 2002). Thus, the widespread practice of only using a screening tool to define depression, typically maternal depression, underestimates the prevalence. Similarly, the use of symptom levels without an assessment of functional impact is likely to overestimate the occurrence of maternal depression. The development of a specific screening tool for the postpartum period, the Edinburgh Postpartum Depression Scale, has promoted screening of women in the perinatal period. However, the poor sensitivity of this measure to detect major or minor depression, varying cut scores in different studies, and the infrequent validation with a clinical diagnostic interview (Gaynes et al., 2005) make it difficult to determine the magnitude and severity of postpartum depression and mental health service needs of mothers postpartum. Box 2-2 lists the common tools and approaches available for the screening and diagnosis of depression. Effective interventions depend on identification of who, when, and what pattern. The type of intervention to use may be affected by the dura- tion of depression and the intensity of its symptoms. Postpartum depres- sion may be limited to the first few months or may persist throughout the first year or beyond. A parent may have chronic unremitting depression, relapsing bouts of depression, or milder depression in which functioning BOX 2-2 Common Tools and Approaches for Screening and Diagnosis of Depression Tools Symptom checklist â¢ Center for Epidemiologic Studies Depression Scale (CESD) â¢ Edinburgh Postnatal Depression Scale (EPDS) â¢ Beck Depression Inventory-II (BDI-II) â¢ Patient Health Questionnaire-2 (PHQ-2) Diagnostic interviews â¢ Hamilton Rating Scale for Depression (HRSD) â¢ Patient Health Questionnaire-9 (PHQ-9) Approaches â¢ tructural Clinical Interview for Diagnostic and Statistical Manual of Mental S Disorders-IVa (SCI-D) â¢ Informal or formal inquiry by a clinician aClinical diagnostic criteria.
58 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN as a parent is maintained; all of these may be considered together in many studies. These complexities of depression are important in determining who is most impacted. For screening programs, when an adult is identified as suffering from depression, it is important to identify if that adult is a parent, if the spouse/ partner is also depressed, or if one or both are at risk for depression, has been diagnosed with depression, if both parents have been diagnosed with depression, or if one or both parents have been diagnosed with other co- morbid conditions (i.e., substance use, trauma). Identification of these issues in both parents is necessary. It is also important to note when depression or symptoms of depression were identified in the parents: prenatal, postpar- tum, the first year of the childâs life, the first 3 years of the childâs life, the first 5 years of the childâs life, or throughout childhood into adolescence. However, the overriding concern with screening is the lack of available and effective resources and knowledge needed after an adult is identified with depression as a result of a screening procedure. Processes of Risk and Resilience in Children of Depressed Parents Closely related to etiology is consideration of factors that are associ- ated with increased risk, as contrasted with resilience, among children of depressed parents. Kraemer and colleagues (1997) refer to risk as a predis- position of an individual or a population to a negative outcome. In the field of mental health, these negative outcomes can be measured in terms of the onset, severity, and duration of a disorder or the frequency and intensity of individual symptoms or clusters of symptoms. Individuals and populations at risk have an increased likelihood of developing symptoms or more severe pathology than the population as a whole (Rutter, 1987). The magnitude, or degree, of risk is measured by the probability of a specific negative out- come in the presence of a risk factor versus the probability when the risk factor is absent. The odds of adolescents developing depression are greater for those who are exposed to a parent with a history of depression com- pared with those whose parents have no such psychopathology. A risk factor is a feature of an individual or the environment that increases the probability of the occurrence of a negative outcome. Risk factors, however, do not explain the processes through which these fac- tors influence the likelihood of an undesirable outcome. In contrast, risk mechanisms or processes describe the intervening paths that connect risk conditions with specific dysfunctional outcomes (Garber, 2006). In the case of parental depression, it is not merely the presence of psychopathology in the parent that leads to psychopathology in the children, but the processes that result from the disorder (such as a biological predisposition, including
APPROACH TO RESEARCH AND ITS EVALUATION 59 genetic mechanisms, negative parenting, or stressful parent-child interac- tions) linking the risk factor with unfavorable outcomes. Even in the face of significant risk, not all individuals do poorlyânot all children and adolescents of depressed parents develop psychopathol- ogy. Although children and adolescents of affectively ill parents are at an increased risk for depression and other forms of psychopathology, a number of studies have demonstrated that many of them actually do quite well. These individuals identified as âresilientâ exhibit the ability to respond positively to significant adversity. Masten and colleagues (1999) conceptual- ize resilience as âphenomena involving successful adaptation in the context of significant threats to developmentâ (p. 143). Although ambiguity exists in the definition of resilience in the literature, researchers have converged on two critical conditions inherent in this concept: (1) exposure to significant threat or adversity and (2) the achievement of positive adaptation despite challenging life circumstances (Luthar, Cicchetti, and Becker, 2000; Luthar and Zigler, 1991; Masten, Best, and Garmezy, 1990). Research on risk and resilience related to depression is vast and grow- ing, encompassing biological, environmental, and psychological processes and their interface. Examples include psychobiological processes that are related to resilience to stress (Charney, 2004; Southwick, Vythilingam, and Charney, 2005); the developmental neurobiology of stress hormones, including cortisol, that are related to depression (Gunnar and Quevedo, 2007); the primate model of stress inoculation based on experiences in early development (Parker et al., 2005); and oxytocin and protection from depression (Scantamburlo et al., 2007). Correlates of Depression As noted above, depression is characterized by high rates of comorbid- ity with other psychiatric disorders and with medical illnesses. Adults with depression, including those who are parents, have an increased risk of a number of other medical and psychiatric conditions. This presents a chal- lenge to researchers, as it is difficult to disentangle the effects of parental depression on children from the effects of these other comorbid conditions. However, these comorbidities also offer a number of opportunities for the identification of parents with depression and contexts for the delivery of interventions to prevent adverse outcomes in their children. Depression in parents is also likely to be accompanied by other prob- lems in families and communities. For example, parents with depression are likely to have spouses with psychiatric disorders that further increase the risk of mental health problems in their children. Some depressed parents and their families live in poverty, with limited access to health and mental health care, further compounding the risk to children in these families.
60 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN The criteria for major depressive disorder as outlined in the DSM-IV include significant impairment in social functioning, interpersonal relation- ships, and work. Significant impairment in multiple domains is also associ- ated with high but subthreshold levels of depressive symptoms. Overall, both depressive disorder and depressive symptoms in parents are accom- panied by comorbid disorders, problems in families and communities, and disruptions in work, marital relationships, and parenting. Developmental Approach We take an explicitly developmental approach in our perspective on parental depression. The effects of depression may be dramatically differ- ent when a parent experiences an episode of depression when pregnant or while a child is in infancy, middle childhood, or adolescence. The different effects of parental depression may be due in part to the different devel- opmental tasks that are central as children move through childhood and adolescence. Moreover, the characteristics of effective parenting change as children develop, and depression may affect different aspects of parenting. For example, the formation of fundamental attachment relationships may be impaired when a parent of an infant experiences an episode of depres- sion. In contrast, depression may interfere with the provision of structure and monitoring when a parent of an adolescent is affected. The effects of depression may be greater when a parent experiences multiple episodes that occur throughout a childâs lifetime and disrupts parenting at multiple points in development. Summary of Our Approach Searching the Literature The committee reviewed multiple types of research or lines of investi- gation on depression in parents. First, we examined the literature on the effects of depression in parents on children. In this literature, we looked both at effects on physical health and health-related outcomes, such as use of health services, and also on childrenâs psychological outcomes. The latter included not only evidence of emotional or behavioral problems in association with depression in parents, but also problems in psychological functioning that might be early signs of psychopathology or vulnerabilities to the later development of psychopathology. Such vulnerabilities include behavior, psychophysiology, and neuroendocrine functioning. We took into account the full developmental spectrum of infancy through adolescence and also considered fetal development. Given the understanding that not all children of depressed parents would be expected to develop problems, we made special note of studies that examined moderatorsâthat is, factors that may affect the strength of
APPROACH TO RESEARCH AND ITS EVALUATION 61 the relationship between depression in parents and outcomes in children. Tests of moderators of the associations between depression in parents and outcomes in children reveal the characteristics of parents and children who are more or less likely to have adverse outcomes. Moderators that have been examined include child characteristics (e.g., age, sex), characteristics of the parent with depression (e.g., severity, chronicity), and characteris- tics of the coparent (e.g., mental health and involvement of fathers when mother is depressed), and broader social context qualities (e.g., poverty). Second, we reviewed the literature on effects of depression in parents on their parenting practices. This literature was of particular interest for two reasons: the well-known reliance of children on good-quality parenting for healthy child development (Collins et al., 2003) and a long-standing understanding that depression is likely to interfere with good-quality par- enting (Weissman and Paykel, 1974). Thus, we were particularly interested in research on the mediational role of parenting in the association between depression in parents and adverse outcomes in children. By mediating role, we mean the extent to which depression has negative influences on parent- ing and, subsequently, the extent to which those changes in parenting ac- count for associations between parentsâ depression and the outcomes in the children. In general, a mediating variable is one that at least partly explains the association between one variable and another. Thus, we examined evi- dence that the associations researchers have found between depression in parents and adverse outcomes in children are at least partially explained by parenting practices. Third, to take into consideration other possible mediators of associa- tions between parentsâ depression and child outcomes, in addition to par- enting practices, we noted studies of the effects of depression in parents on the children that also considered the potential role of marital relationships. However, it was beyond the scope of this report to review the literature on effects of depression in parents on their relationships with their spouse/part- ner, extended family relationships, or work, although we acknowledge that these are likely important aspects of a broad model of how depression in parents is related to the outcomes in their children. Fourth, we reviewed evidence for the effectiveness of screening, pre- vention, and treatment of depression in the parent. In these literatures, we searched broadly for evidence of effectiveness of a range of interventions designed to address the effects of parental depression on children. CHALLENGES IN THE EVALUATION OF THE LITERATURE We now present an overview of the many methodological challenges faced by researchers studying parental depression and its effects on par- enting practices and child outcomes. We focus particularly on issues that
62 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN make recommendations possible for the development of new research, interventions, and policy. Such issues involve all aspects of research, from conceptual frameworks, to study sampling designs, to data analysis and the integration of research findings across literatures. As noted previously, these do not specifically address each area of inquiry but instead provide the reader with a set of issues to consider when reading this report. In each sub- sequent chapter, the committee considered these issues when evaluating the available literature to help guide their conclusions and recommendations. Balance Hypothesis-Driven and Real-World Challenges The integration of the large and growing literature on parental depres- sion has presented a number of problems and challenges. First and foremost is the need to balance hypothesis-driven research with real-world challenges in addressing complex health and behavioral disorders. As noted above, the goal of translational research is to move from basic to applied research, from bench to bedside. However, disparities between research conducted under carefully controlled conditions and research in real-world contexts have been widely noted. For example, Weisz and colleagues have shown that psychological treatments for children can be effective when delivered and evaluated under optimal conditions but often have negligible or no ef- fect in community settings (e.g., Weersing and Weisz, 2002). Theories on which interventions on parental depression are based must be tested in the field and modified, if necessary, to increase the likelihood of success when implemented on a large scale. How parental depression affects parenting and childrenâs outcomes is a question that will be best answered using trans- lational research methods that meet the standards that we now outline. Integrative Models in Research Scientific research in general is often incremental in its development, and interrelated factors are not always examined properly in the same study. Research in the area of parental depression and its effects on parent- ing and child development has rarely been conducted to address hypotheses from an integrative model. The use of theory-based models in the design of studies helps to provide a framework for success in ultimate treatment effectiveness that is replicable across locations and time and that can be sustained over time. For example, not accounting for the effects of parental substance abuse, violence, and depression or other psychiatric distress in families may result in an overestimation of the individual effects of these factors on the development of the child. The relative lack of generalizable, integrative research and interventions on the scope and nature of the prob-
APPROACH TO RESEARCH AND ITS EVALUATION 63 lems caused by parental depression has provided the impetus for the work of our committee. A particular interest of the committee was to identify potential un- tapped sources of information on the scope and dynamics of parental depression on families available in large-scale databases that have been developed in recent years. As a result of this interest, the committee has un- dertaken analyses of the National Comorbidity Survey-Replication sample and the Medical Expenditure Panel Survey, two large surveys generalizable to the U.S. population with data on family mental health issues, the results of which are included in this report. Also of note, studies in families have largely not examined the genetic component and its potential interactions, something that is addressed in a paper prepared for the committee as input into this report. Despite being acknowledged as one of the four primary mechanisms by which depression in parents is related to the development of depression and other problems in children (Goodman and Gotlib, 1999; See Figure 4-1 in Chapter 4), few researchers have used genetically informed designs to address ques- tions related to the effect of parental depression on parenting and on child functioning. A Developmental Perspective Studies on the effects of parental depression on parenting and child development would benefit from taking a developmental perspective, al- though we found such an approach to be rare. As described earlier, a de- velopmental perspective considers developmental processes (e.g., the timing of exposure), takes into account roles of early experience of development, integrates multiple transactional influences on development, and considers the multiple alternative pathways through development. For example, as- suming that risk factors have effects or interventions that work the same among families with infants and toddlers, school-age children, adolescents, and young adults ignores the vast evidence in the area of child development. A recent editorial by Hinshaw (2008) notes that this is a concern in general with respect to studies of developmental psychopathology and that studies need to address questions âacross generations and across timeâ (p. 361). A Shared Framework Across Literature Compounding the challenge of developing shared integrative mod- els, researchers in these areas often have not used common theoretical frameworks, study designs, or measures or had a common focus on the parent-child relationship as the fundamental dynamic to be addressed. This
64 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN integration has been described as âmethodological pluralismâ (Hinshaw, 2008, p. 361). For example, depending on the nature of the outcomes of interest and often the field of expertise of the research team (e.g., social sci- ence, medicine), the quantification of effects can be varied. Social scientists often use standardized measures of âeffect sizeâ (Cohen, 1988) to allow for comparison from study to study. Cutoffs are used to describe these ef- fects as âsmallâ (0.20), âmediumâ (0.50), or âlargeâ (0.80). The drawback to these measures is that they are unitless and difficult to interpret from a clinical perspective. In contrast, medical researchers often prefer to examine unstandardized differences, âeffects,â which make between-study comparisons difficult if the same measures are not used, or they focus on categorical outcomes, such as âdepressedâ versus ânot depressedâ on the basis of established cri- teria, such as the DSM-IV. The effects in studies employing such outcomes are often summarized in the form of odds ratios or risk ratios, which can vary between 0 and 1 (denoting negative association), 1 (no association), or greater than 1 (to infinity, denoting positive association). The size of such ratio-based measures could be interpreted, for example, with respect to positive associations, as âsmallâ (greater than 1.0 to 1.5), âmoderateâ (1.5 to 2.0), or âlargeâ (2.0 or greater). These must be interpreted cautiously, however, as their size can be dependent on the scale of the independent variable in question. Moreover, the interpretation of the sizes of effects should also take into consideration what has been previously observed in the relevant field in order to provide proper context. Choice of Target Population of Study and Its Impact on Generalizability Another issue that can adversely affect the interpretability of research findings in this area is the lack of generalizability of the samples used. This can result, for example, from a studyâs being based on a population referred to treatment or a population of parents of children with special needs. Re- sults from such studies clearly cannot be generalized to the population of families at risk because of the varying characteristics of each population. Another way in which generalizability can be reduced is by the inadvertent restriction to a subgroup of the population of interest, which can result in studies with small numbers of subjects. Generalizability is enhanced when random sampling is implemented in a design that samples from all subgroups of interest. Examples of such studies can be found in many so- called complex sample surveys funded by the federal government, such as the National Comorbidity Survey (see http://www.icpsr.umich.edu/CPES/).
APPROACH TO RESEARCH AND ITS EVALUATION 65 Randomized Treatment and Intervention Studies In comparative intervention studies, which compare a given interven- tion strategy to another, the lack of randomization to study groups can adversely affect the validity of the study findings. With randomization implemented with large sample sizes, one can minimize the likelihood of the misattribution of effects to the intervention that could otherwise be ex- plained by other factors. In the area of parental depression, the relative lack of randomized studies can be explained by the difficulty of implementation because of ethical issues, such as withholding a potentially effective treat- ment, or scientific concerns, such as avoiding contamination of intervention groups in behavioral studies. However, randomized clinical trials represent the optimal design for evaluating the efficacy of interventions to prevent adverse outcomes in children of depressed parents. Sample Size and Its Implications In general, studies that include small numbers of subjects can produce findings that are suggestive but do not provide evidence that is statistically significant. Such studies are said to lack âstatistical power.â In these cases, results that are clinically relevant may lack precision and thus confidence in their interpretability is reduced. Researchers often gauge the utility of the sample size of a study by the size of effect that it can minimally detect with 80 percent statistical power and a two-sided alpha (type I error rate) of 0.05. This âsizeâ of effect is often quantified by the standardized effect size (Cohen, 1988) in which an effect of 0.20 is considered âsmall,â 0.50 âmedium,â and 0.80 âlarge.â To achieve 80 percent power, the 0.20 effect size requires approximately 800 subjects overall, 0.50 requires approxi- mately 120, and 0.80 requires approximately 50. From another view, one can evaluate power for selected sample sizes of interest, for example, 100, 150, 200, 300, and 500, the latter two being in the range of what many would consider large. These sample sizes would require minimal effect size respectively of 0.57, 0.46, 0.40, 0.33, and 0.25. The sizes of the samples employed in research on the effects of parental depression have often not been large, thus limiting the conclusions that can be drawn from the results and consequently preventing the kind of confi- dent interpretation of results that is required to make policy changes. It is not surprising that large effects of individual risk or protective factors have not been observed given the inherent complexity of the factors at play with respect to the relationship of parental depression, parenting, and child de- velopment. Specifically in the context of intervention studies, the difficulty in impacting the multiple functional pathways that define this problem is reflected in the frequently observed small-size intervention effects. As we
66 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN show in reviewing the literature, many published studies have only been able to detect small effects as statistically significant given the sizes of their samples. And large-scale research projects tied specifically to dissemination and implementation strategies are needed to help motivate changes in poli- cies and systems. Consideration of Possible Mediation of Effects Integrative models of parental depression and its effects on parenting and child development need to consider the potential mediation of effects, which hold promise for explaining how depression in parents has its effects on children. Tests of mediation require a more complex analytic approach, such as path analysis, structural equations models, or other causal frame- works than simply adjusting out effects as covariates. The examination of possible mediation effects is an important element in understanding the underlying mechanisms for the effects of parental depression, and thus it has implications for the design and implementation of interventions. Potential problems arise, however, when mediational effects are assessed in solely cross-sectional analyses (Cole and Maxwell, 2003). We show that many studies of the effects of parental depression have not employed longitudinal designs and thus have not properly been able to examine how its effects are mediated through other factors, such as parenting practices on child outcomes. Assessment of Effect-Modifying Factors In many studies in the area of parental depression as a risk factor, moderators of effect have not been measured, such as exposure to violence. We would expect that when the constellation of factors associated with depression in a parent are measured, effects for the subgroup with multiple risk factors, such as depression and violence exposure, could be significantly higher than for those who have depression and no other factors. More broadly, a moderator is variable that increases or decreases the strength of association between two other variable. With the effect of depression in parents in their children, we considered the circumstances under which the effects may be stronger or weaker, including characteristics of the depres- sion in the parents, characteristics of the children, and so forth. Such combinations of risk factors in synergy (i.e., multiplicative) may be what actually result in poorer outcomes and not individual effects op- erating in a purely additive fashion. Such modifying effects may be present even in a randomized trial. Although the randomization aims to balance across the treatment or intervention groups on all potential confounding variables, measures or unmeasured, this does not eliminate the presence of
APPROACH TO RESEARCH AND ITS EVALUATION 67 effect modification. For example, if gender is an effect modifier, it is pos- sible that the intervention could work differently in female compared with male subjects. Genetics is likely to play a moderating role in at least one of several possible patterns (Rutter, 2007). Thus, not recognizing the modera- tion of effects may prevent the targeting for intervention of subgroups of those who are at particularly increased risk. Developmental models of psychopathology emphasize the importance of cumulative risk or multiple risk factors (Rutter, 2000; Rutter et al., 2001). The action of risk factors is cumulative in the sense that the presence of more risk factors is related to a higher certainty of negative outcomes (Seifer et al., 1992). Cumulative models reflect the natural covariation of many risk factors. These models assume that the confluence of many risk factors, rather than any single risk, is what leads to dysfunction. This is at least partly explained by the multiple risk factors overwhelming the adap- tive capacities of the organism. For example, it has been shown that the magnitude of effect of any one family risk factor associated with psychoso- cial adjustment of youth is relatively small (Reid and Crisafulli, 1990) and that an increase of adjustment difficulties occurs with an increase in the number of risk factors (Forehand, Biggar, and Kotchick, 1998). Effect of Measurement Error and Other Measurement Issues An issue that has not been addressed in the interpretation of findings in this area is the difficulty in measuring depression and depressive symptoms, exposure to violence, and degree of substance use problems (because of the possible illicit nature of the behavior or risk of the loss of the child to state custody that might be imposed by identification of use by the authorities). Either the overestimation or underestimation of risk factor or intervention effects is possible in the presence of measurement error. Replication of Findings With a lack of large-scale, generalizable trials, researchers are forced to rely on the replication of findings of quasi-experimental and nonexperi- mental studies with similar consideration of the multiple, intercorrelated factors suspected to impact family function and child developmentâthat is, an integrated model that links these factors together. Such replication has generally not been found in studies of parental depression, parenting, and child development. We consider strategies to reduce the negative ef- fects on parental depression on parental practices and child outcomes that have been efficacious in two or more peer-reviewed studies. We thus con- sider such strategies to be ready for wider dissemination and evaluated for their effectiveness. Approaches that have not met this level of evidence are
68 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN described as âpromisingâ and should be tested in future studies to demon- strate their efficacy. CONCLUSION In this chapter, we have presented the committeeâs approach to the evaluation of the extensive literature on the relationship of parental de- pression, parenting practices, and child outcomes. Our ultimate aim is to set the bar for level of evidence to show (1) associations between parental depression, parenting practices, and child outcomes and (2) the efficacy of screening, prevention, treatments, and policies on the negative effects of parental depression on children. More generally, our aim is to identify the challenges posed by researchers, clinicians, and policy makers in attempting to address problems caused by parental depression. Four themes emerge from the committeeâs approach: (1) the integration of knowledge regarding the dynamics of parental depression, parenting practices, and child outcomes; (2) the application of a developmental frame- work in the study and evaluation of the effects of parental depression; (3) the need to conceptualize the problem as two generation in nature; and (4) the need to acknowledge the presence of the constellation of risk factors, context, and correlates of parental depression. In the chapters that follow, attention is paid to each of these themes, which is essential to understanding the complexity of the problems associated with parental depression. The report shows that, although much is known, there are still many critical questions needing to be addressed with further research and many promis- ing approaches to screening, prevention, treatment, and policy. REFERENCES Abrams, D.B. (2006). Applying transdisciplinary research strategies to understanding and eliminating health disparities. Health Education and Behavior, 33, 515â531. American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental Dis- orders (4th ed.). Washington, DC: Author. Barbato, A., and DâAvanzo, B. (2008). Efficacy of couple therapy as a treatment for depres- sion: A meta-analysis. Psychiatric Quarterly, 79, 121â132. Barlow, D.H. (1996). Health care policy, psychotherapy research, and the future of psycho- therapy. American Psychologist, 51, 1050â1058. Baucom, D.H., Shoham, V., Mueser, K.T., Daiuto, A.D., and Stickle, T.R. (1998). Empirically supported couple and family interventions for marital distress and adult mental health problems. Journal of Consulting and Clinical Psychology, 66, 53â88. Beach, S.R.H., Fincham, F.D., and Katz, J. (1998). Marital therapy in the treatment of depres- sion: Toward a third generation of therapy and research. Clinical Psychology Review, 18, 635â661. Bornstein, M.H. (2006). Parenting Science and Practice. Hoboken, NJ: John Wiley & Sons.
APPROACH TO RESEARCH AND ITS EVALUATION 69 Caspi, A., Sugden, K., Moffitt, T.E., Taylor, A., Craig, I.W., Harrington, H., McClay, J., Mill, J., Martin, J., Braithwait, A., and Poulton, R. (2003). Influence of life stress on depres- sion: Moderation by a polymorphism in the 5-HTT gene. Science, 301, 386â389. Charney, D.S. (2004). Psychobiological mechanisms of resilience and vulnerability: Implica- tions for successful adaptation to extreme stress. American Journal of Psychiatry, 161, 195â216. Chugani, H.T., and Phelps, M.E. (1986). Maturational changes in cerebral function in infants determined by 18FDG positron emission tomography. Science, 231, 840â843. Cicchetti, D., and Schneider-Rosen, K. (1984). Toward a transactional model of childhood depression. New Directions for Child Development, 26, 5â27. Cicchetti, D., Rogosch, F.A., and Toth, S.L. (2000). The efficacy of toddler-parent psycho- therapy for fostering cognitive development in offspring of depressed mothers. Journal of Abnormal Child Psychology, 28, 135â148. Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates. Cole, D.A., and Maxwell, S.E. (2003). Testing mediational models with longitudinal data: Questions and tips in the use of structural equation modeling. Journal of Abnormal Psychology, 112, 558â577. Collins, W.A., Maccoby, E.E., Steinberg, L., Hetherington, E.M., and Bornstein, M.H. (2003). Contemporary research on parenting: The case for nature and nurture. In M.E. Hertzig and E.A. Farber (Eds.), Annual Progress in Child Psychiatry and Child Development: 2000-2001 (pp. 113â140). New York: Brunner-Routledge. Compas, B.E., Langrock, A.M., Keller, G., Merchant, M.J., and Copeland, M.E. (2002). Chil- dren coping with parental depression: Processes of adaptation to family stress. In S.H. Goodman and I.H. Gotlib (Eds.), Children of Depressed Parents: Mechanisms of Risk and Implications for Treatment (pp. 227â252). Washington, DC: American Psychologi- cal Association. Cooper, J.L., Aratani, Y., Knitzer, J., Douglas-Hall, A., Masi, R., Banghart, P., and Dababnah, S. (2008). Unclaimed Children Revisited: The Status of Childrenâs Mental Health Policy in the United States. New York: National Center for Children in Poverty. Du Rocher Schudlich, T.D., and Cummings, E.M. (2007). Parental dysphoria and childrenâs adjustment: Marital conflict styles, childrenâs emotional security, and parenting as media- tors of risk. Journal of Abnormal Child Psychology, 35, 627â639. Egeland, B., and Farber, E.A. (1984). Infant-mother attachment: Factors related to its develop- ment and changes over time. Child Development, 55, 753â771. Evans, J., Heron, J., Francomb, H., Oke, S., Golding, J., and Avon Longitudinal Study of Parents and Children Study Team. (2001). Cohort study of depressed mood during pregnancy and after childbirth. British Medical Journal, 323, 257â260. Forehand, R., Biggar, H., and Kotchick, B.A. (1998). Cumulative risk across family stressors: Short- and long-term effects for adolescents. Journal of Abnormal Child Psychology, 26, 119â128. Garber, J. (2006). Depression in children and adolescents: Linking risk research and preven- tion. American Journal of Preventative Medicine, 31, S104âS125. Gaynes, B.N., Gavin, N., Meltzer-Brody S., Lohr, K.N., Swinson, T., Gartlehner, G., Brody, S., and Miller, W.C. (2005). Perinatal Depression: Prevalence, Screening Accuracy, and Screening Outcomes. (Evidence Report/Technology Assessment No. 119, AHRQ Publica- tion No. 05-E006â2.) Rockville, MD: Agency for Healthcare Research and Quality. Gelfand, D.M., Teti, D.M., Seiner, S.A., and Jameson, P.B. (1996). Helping mothers fight depression: Evaluation of a home-based intervention for depressed mothers and their infants. Journal of Clinical Child Psychology, 24, 406â422.
70 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN Gerdes, A.C., Hoza, B., Arnold, L.E., Pelham, W.E., Swanson, J.M., Wigal, T., and Jensen, P.S. (2007). Maternal depressive symptomatology and parenting behavior: Exploration of possible mediators. Journal of Abnormal Child Psychology, 35, 705â714. Gilespie, C.F., and Nemerooff, C.B. (2007). Corticotropin-releasing factor and the psychobiol- ogy of early-life stress. Current Directions in Psychological Science, 16, 85â89. Goodman, S.H., and Gotlib, I.H. (1999). Risk for psychopathology in the children of de- pressed mothers: A developmental model for understanding mechanisms of transmission. Psychological Review, 106, 458â490. Gunnar, M.R., and Quevedo, K. (2007). The neurobiology of stress and development. In S.T. Fiske, A.E. Kazdin, and D.L. Schacter (Eds.), Annual Review of Psychology, 58, 145â174. Hammen, C. (2002). Context of stress in families of children with depressed parents. In S.H. Goodman and I.H. Gotlib (Eds.), Children of Depressed Parents: Mechanisms of Risk and Implications for Treatment (pp. 175â202). Washington, DC: American Psychologi- cal Association. Hariri, A.R., and Brown, S.M. (2006). Serotonin. American Journal of Psychiatry, 163, 12. Hinshaw, S.P. (2008). Lessons from research on the developmental psychopathology of girls and women. Journal of the American Academy of Child and Adolescent Psychiatry, 47, 359â361. Jaser, S.S., Langrock A.M., Keller G., Merchant, M.J., Benson, M.A., Reeslund, K., Champion, J.E., and Compas, B.E. (2005). Coping with the stress of parental depression II: Ado- lescent and parent reports of coping and adjustment. Journal of Child and Adolescent Psychology, 34, 193â205. Joiner, T.E., Coyne, J.C., and Blalock, J. (1999). On the interpersonal nature of depression: Overview and synthesis. In T. Joiner and J.C. Coyne (Eds.), The Interactional Nature of Depression: Advances in Interpersonal Approaches (pp. 3â20). Washington, DC: Ameri- can Psychological Association. Kaminski, J.W., Valle, L.A., Filene, J.H., and Boyle, C.L. (2008). A meta-analytic review of components associated with parent training effectiveness. Journal of Abnormal Child Psychology, 36, 567â589. Kazdin, A.E. (2005). Parent Management Training: Treatment for Oppositional, Aggressive, and Antisocial Behavior in Children and Adolescents. New York: Oxford University Press. Kessler, R.C., Merikangas, K.R., and Wang, P.S. (2007). Prevalence, comorbidity, and service utilization for mood disorders in the United States at the beginning of the twenty-first century. Annual Review of Clinical Psychology, 3, 137â158. Kessler, R.C., Zhao, S., Katz, S.J., Kouzis, A.C., Frank, R.G., Edlund, M., and Leaf, P. (1999). Past-year use of outpatient services for psychiatric problems in the National Comorbidity Survey. American Journal of Psychiatry, 156, 115â123. Kingdon, J.W. (1995). Agendas, Alternatives and Public Policies. New York: Addison-Wesley Educational Publishers. Knitzer, J., Theberge, S., and Johnson, K. (2008). Reducing Maternal Depression and its Im- pact on Young Children: Toward a Responsive Early Childhood Policy Framework. New York: Columbia University, National Center for Children in Poverty. Kraemer, H.C., Kazdin, A.E., Offord, D.R., Kessler, R.C., Jensen, P.S., and Kupfer, D.J. (1997). Coming to terms with the terms of risks. Archives of General Psychiatry, 54, 337â343. Kroenke, K. (2006). Minor depression: Midway between major depression and euthymia. Annals of Internal Medicine, 144, 528â530. Lazear, K.J., Pires, S.A., Isaacs, M.R., Chaulk, P., and Huang, L. (2008). Depression among low-income women of color: Qualitative findings from cross-cultural focus groups. Jour- nal of Immigrant Minority Health, 10, 127â133.
APPROACH TO RESEARCH AND ITS EVALUATION 71 Lovejoy, M.C., Graczyk, P.A., OâHare, E., and Neuman, G. (2000). Maternal depression and parenting behavior: A meta-analytic review. Clinical Psychology Review, 20, 561â592. Luthar, S.S., and Zigler, E. (1991). Vulnerability and competence: A review of research on resilience in childhood. American Journal of Orthopsychiatry, 61, 6â22. Luthar, S.S., Cicchetti, D., and Becker, B. (2000). The construct of resilience: A critical evalu- ation and guidelines for future work. Child Development, 71, 543â562. Lyons-Ruth, K., Connell, D.B., Grunebaum, H.U., and Botein, S. (1990). Infants at social risk: Maternal depression and family support as mediators of infant development and security of attachment. Child Development, 61, 85â98. Masten, A.S., Best, K.M., and Garmezy, N. (1990). Resilience and development: Contributions from the study of children who overcome adversity. Development and Psychopathology, 2, 425â444. Masten, A.S., Hubbard, J.J., Gest, S.D., Tellegen, A., Garmezy, N., and Ramirez, M. (1999). Competence in the context of adversity: Pathways to resilience and maladaptation from childhood to late adolescence. Development and Psychopathology, 11, 143â169. Miller, G.E., and Blackwell, E. (2006) Turning up the heat: Inflammation as a mechanism linking chronic stress, depression, and heart disease. Current Directions in Psychological Science, 15, 269â272. Narrow, W.E., Regier, D.A., Rae, D.S., Manderscheid, R.W., and Locke, B.Z. (1993). Use of services by persons with mental and addictive disorders: Findings from the National Institute of Mental Health Epidemiologic Catchment Area Program. Archives of General Psychiatry, 50, 95â107. Nelson, E.E., Leibenluft, E., McClure, E.B., and Pine, D.S. (2005). The social re-orientation of adolescence: A neuroscience perspective on the process and its relation to psychopathol- ogy. Psychological Medicine, 35, 163â174. Parker, K.J., Buckmaster, C.L., Justus, K.R., Schatzberg, A.F., and Lyons, D.M. (2005). Mild early stress enhances prefrontal-dependent response inhibition in monkeys. Biological Psychiatry, 57, 848â855. Regier, D.A., Narrow, W.E., Rae, D.S., Manderscheid, R.W., Locke, B.Z., and Goodwin, F.K. (1993). The de facto U.S. mental and addictive disorders service system: Epidemiologic catchment area prospective 1-year prevalence rates of disorders and services. Archives of General Psychiatry, 50, 85â94. Reid, W.J., and Crisafulli, A. (1990). Marital discord and child behavior problems: A meta- analysis. Journal of Abnormal Child Psychology, 18, 105â117. Romeo, R.D., and McEwen, B.S. (2006). Stress and the adolescent brain. Annals of the New York Academy of Science, 1094, 202â214. Rutter, M. (1987). Psychosocial resilience and protective mechanisms. American Journal of Orthopsychiatry, 57, 316â331. Rutter, M. (2000). Psychosocial influences: Critiques, findings, and research needs. Develop- ment and Psychopathology, 12, 375â405. Rutter, M. (2007). Gene-environment interdependence. Developmental Science, 10, 12â18. Rutter, M., Pickles, A., Murray, R., and Eaves, L. (2001). Testing hypotheses on specific envi- ronmental causal effects on behavior. Psychological Bulletin, 127, 291â324. Sanders, M.R., Markie-Dadds, C., Tully, L.A., and Bor, W. (2000). The Triple P-Positive Parenting Program: A comparison of enhanced, standard, and self-directed behavioral family intervention for parents of children with early onset conduct problems. Journal of Consulting and Clinical Psychology, 68, 624â640. Scantamburlo, G., Hansenne, M., Fuchs, S., Pitchot, W., MarÃ©chal, P., Pequeux, C., Ansseau, M., and Legros, J.J. (2007). Plasma oxytocin levels and anxiety in patients with major depression. Psychoneuroendocrinology, 32, 407â410.
72 DEPRESSION IN PARENTS, PARENTING, AND CHILDREN Seifer, R., Sameroff, A.J., Baldwin, C.P., and Baldwin, A. (1992). Child and family factors that ameliorate risk between 4 and 13 years of age. Journal of the American Academy of Child and Adolescent Psychiatry, 31, 893â903. Solantaus, T., and Toikka, S. (2006). The Effective Family Programme: Preventative services for the children of mentally ill parents in Finland. International Journal of Mental Health Promotion, 8, 37â43. Southwick, S.M., Vythilingam, M., and Charney D.S. (2005). The psychobiology of depres- sion and resilience to stress: Implications for prevention and treatment. Annual Review of Clinical Psychology, 1, 255â291. Sroufe, L.A., and Rutter, M. (1984). The domain of developmental psychopathology. Child Development, 55, 17â29. Tronick, E.Z., and Gianino, A.F. (1986). The transmission of maternal disturbance to the infant. In E.Z. Tronick and T. Field (Eds.), Maternal Depression and Infant Disturbance (pp. 5â12). San Francisco: Jossey-Bass. U.S. Preventive Services Task Force. (2002). Screening for depression: recommendations and rationale. Annals of Internal Medicine, 136, 760â764. Van den Bergh, B.R., Mulder, E.J.H., Mennes, M., and Glover, V. (2005). Antenatal mater- nal anxiety and stress and the neurobehavioural development of the fetus and child: Links and possible mechanisms. A review. Neuroscience and Biobehavioral Reviews, 29, 237â258. Weersing, V.R., and Weisz, J.R. (2002). Community clinic treatment of depressed youth: Benchmarking usual care against CBT clinical trials. Journal of Consulting and Clinical Psychology, 70, 299â310. Weissman, M.M., and Paykel, E.S. (1974). The Depressed Woman: A Study of Social Relation- ships. Chicago: University of Chicago Press. Weisz, J.R., Jensen, A.L., and McLeod, B.D. (2005). Milestones and methods in the develop- ment and dissemination of child and adolescent psychotherapies: Review, commentary, and a new deployment-focused model. In E.D. Hibbs and P.S. Jensen (Eds.), Psychosocial Treatments for Child and Adolescent Disorders: Empirically Based Strategies for Clinical Practice (2nd ed., pp. 9â39). Washington, DC: American Psychological Association. Whooley, M.A., Avins, A.L., Miranda, J., and Browner, W.S. (1997). Case-finding instru- ments for depression. Two questions are as good as many. Journal of General Internal Medicine, 12, 439â445.