Routine Assessment of Family and Community Health Risks: Parent Views and What They Receive
Objective. To examine the prevalence of parent–provider discussions of family and community health risks during well-child visits and the gaps between which issues are discussed and which issues parents would like to discuss.
Methods. Data came from the National Survey of Early Childhood Health, a nationally representative sample of parents of 2068 children aged 4 to 35 months. The outcome measures were 1) the reported discussions with pediatric clinicians about 7 family and community health risks and 2) whether the parent believes that pediatric clinicians should ask parents about each risk.
Results. Most parents believe that pediatric providers should discuss topics such as smoking in the household, financial difficulties, and emotional support available to the parent. However, with the exception of “household smoking,” fewer than half of parents have been asked about these topics by their child’s clinician. Parents of black and Hispanic children were more likely than parents of white children to be asked about several of these issues, as were parents of the youngest children and those with publicly financed health insurance. The greatest gap between parents’ views and their reports of discussion with the clinician occur for parents of white children and older children. Among parents who hold the view that a topic should be discussed, parents of white and older children are less likely than others to report discussing some or all family and community health risks.
Conclusion. The low frequency of discussions for many topics indicates potential unmet need. More universal surveillance of parents with young children might ensure that needs are not missed, particularly given that strong majorities of parents view family and community topics, with the exception of community violence, as appropriate for discussion in clinic visits.
The quality of family relationships and the impact of community contexts are increasingly recognized as important influences on the health and development of children. The recent National Academy of Sciences report From Neurons to Neighborhoods synthesizes the scientific evidence of how the quality of the child’s environment, especially the quality of caregiving relationships, influences the development of young children and their long-term outcomes.1 Similar evidence has accumulated about the risk created by specific family behaviors and conditions, such as exposure to household smoking or to violence in their communities.2 Other studies have shown that experiencing social risks such as food insufficiency can significantly affect a child’s school performance and his or her level of risk-taking behaviors.3
More than 30 years ago, Roghmann and Haggerty4 called for a reframing of threats to child health using the term “new morbidities” and codified the importance of considering a broad range of family, social, and community factors that can influence children’s health and development. Since that time, successive editions of the American Academy of Pediatrics (AAP) guidelines for well-child care have increasingly emphasized anticipatory guidance regarding a range of family and community influences on children.5 Bright Futures guidelines for health supervision go even further by suggesting that routine health promotion provide more focus on parent–child interactions, constructive relationships with adults and peers, and community interactions.6
Parents’ demand for anticipatory guidance about issues regarding their child’s behavior, discipline, and social interactions with peers is actually high. The Commonwealth Survey of Families With Young Children found that parents want additional information about child-rearing issues including how to minimize risks to their child’s healthy development. In fact, this study found that parents would be willing to pay more for pediatric care if such services were more readily available.7
The routine assessment of family and community risks to children’s development has not become a regular part of pediatric care despite the growing recognition of the important influence of such factors on child health and development and the increased identification of psychosocial issues in some practices in the United States.8,9 Although the AAP has recently made additional recommendations about screening for a variety of psychosocial risks, including domestic violence,5 significant barriers persist. In 1 recent study conducted in Cincinnati, the vast majority of practitioners in the surrounding tristate area were unaware of domestic violence recommendations, were not routinely screening for domestic violence, felt ill prepared to conduct such screenings, and believed that they had neither the time nor office support to provide such screenings.10
Although pediatric providers have historically focused on the child and family relationship, until recently, there were very few practice-based assessment tools that providers could use to conduct systematic and routine assessments of family and community risks to child health and development. Recently developed assessment instruments have increased the tools available to pediatric providers, but a recent study suggests that pediatricians are even less likely to conduct formal assessments of family risks and community exposures than they are to assess a child’s development.11 Pediatricians identify multiple barriers to conducting family and community risk assessments as part of routine pediatric care. Barriers identified in national surveys of pediatricians include a lack of time to fit these issues into a busy practice agenda, the lack of training in assessing psychosocial issues, the lack of reimbursement for such assessments, and a lack of potential resources for referral once a problem such as maternal depression is identified.11,12
Even child health experts who contend that an expanded model of pediatric care is essential to the future of child health supervision warn about the difficulty of addressing child and family risk issues as part of routine pediatric care.12 Some have also argued that use of psychosocial risk assessment tools about sensitive issues such as parental drug use, family relationship problems, and domestic violence can actually be a deterrent to developing the kind of trusting relationships between parent and pediatric provider that are necessary to have open and effective communication.10
The objective of this study was to assess which family and community issues are routinely addressed in primary pediatric care for young children and to better understand parent views about the appropriateness of asking about these potentially sensitive issues. These include topics that have been long accepted as directly affecting children’s health (eg, smoking) as well as topics that have been added more recently to AAP and Bright Futures recommendations for health supervision (eg, financial well-being, emotional support available to the parent).
The National Survey of Early Childhood Health (NSECH) is a national survey of 2068 parents of children between 4 and 35 months of age that was conducted during 2000. The data were collected as part of the National Center for Health Statistics’ State and Local Area Integrated Telephone Survey, a stratified random-digit-dial survey used to provide state-level estimates of immunization coverage for children. After completion of the State and Local Area Integrated Telephone Survey interview, the parent or guardian in the household who is responsible for the child’s medical care completed the NSECH interview. Black and Hispanic children were oversampled. Child-level sampling weights were developed to adjust for nonresponse, oversampling, and nontelephone coverage.
The term “parent” is used throughout this analysis for simplification, although not all respondents were parents. Approximately 98% of respondents were parents (87% were mothers and 11% fathers), with 2% of respondents being a relative or other kind of caregiver.
This analysis describes parent views about what pediatric clinicians should address, and whether they have been asked about the topics by their own child’s provider. The outcome measures in this analysis are drawn from 2 questions. Parents first were asked, “In the last 12 months (or since birth if child younger than 12 months), has [your child’s] doctor or health provider ever asked you about [topic]?” They were then asked, “Should a child’s doctors or health providers discuss [topic] with parents?” The 7 topics were parents’ physical health, smoking in the household, drinking alcohol and/or using drugs in the household, having someone to turn to for emotional support, having a spouse or partner who is supportive of their parenting, financial difficulties paying for a child’s basic needs, and violence in the community. Items that ask about emotional support and having a supportive partner in parenting are combined into 1 measure of emotional support because these items stem from the same construct. We combine the parent’s view on the appropriateness of discussing a topic with their report of having discussed the topic with their own child’s provider to categorize parents into 4 groups: 1) asked about the topic by their child’s provider and believe that pediatric clinicians should ask about it, 2) asked about the topic but believe that clinicians should not ask about it, 3) not asked about the topic and believe that clinicians should not ask about it, and 4) not asked about the topic but believe that clinicians should ask about it. The multivariate analysis evaluates predictors of the fourth group because these parents are potentially reporting an unmet need.
Sociodemographic variables used to identify patterns in parent views and in topics discussed include maternal education, maternal race/ethnicity, maternal age, marital status, child’s age, household income, and region of the country.
Parents were asked whether their child has a particular doctor or other health care provider for well-child care. Parents who reported having a particular provider (46% of children) were asked about the type of provider. Frequency of provider type in this analysis includes categories of pediatrician (35%), family practitioner (8%), nonphysician (pediatric nurse practitioner, physician’s assistant, and other were combined owing to small numbers; 3%), and no particular provider (54%).
The child was classified as having private insurance when he or she was covered, at the time of the interview, by insurance obtained through employment or unions or purchased directly. The child was classified as being currently covered by public insurance when covered by 1 or more public programs: Medicaid, the state Title V program, the State Children’s Health Insurance Program, military coverage, or Medicare. The 2 other insurance categories are having both private and public insurance and having no health insurance.
The child’s usual setting for check-ups and shots includes the following categories: doctor or nurse practitioner in a private or group practice; community health center or public clinic; or hospital clinic, urgent care/walk-in clinic, or emergency department. The small number of children with no one place or with an “other” setting are excluded from the analysis.
The data presentation includes 3 sections. We first present the frequency of discussion between clinician and parent for each of the family and community topics and parent views on whether each of the topics should be discussed. Frequencies of the measure that combines views with actual discussion are presented. Second, bivariate analysis compares maternal and child characteristics first with reports of discussion and second with parent views. Third, multivariate logistic regression is used to identify patterns of discussion. The first logistic model identifies predictors of parent discussion for each topic. The second model identifies predictors of potential unmet need, which is defined as the group of parents who believe that clinicians should ask about the topic but who themselves have not discussed the topic with their child’s clinician (group 4). This model predicts lack of discussion among all parents who believe that a topic should be discussed (groups 1 and 4). In the bivariate analyses, χ2 tests are used to test the association of each of the family and community topics with sociodemographics (maternal education, maternal race/ethnicity, maternal age, marital status, household income, region of the country, and the child’s age) and with health care factors (health care coverage, setting of well-child care, and type of provider). All bivariate factors are conceptually related to parents’ views and discussions and are included in the multivariate models. The multivariate analyses provide adjusted odds ratios (ORs) and 95% confidence intervals using all of the factors used in bivariate analysis. Analyses include adjustments for nonresponse and are conducted using the Survey Data Analysis (SUDAAN) software program using weighting to make estimates representative of the 10.7 million US children who were aged 4 to 35 months in 2000.13
Most parents stated that pediatric clinicians should discuss each of the family and community topics. Nearly all parents concurred that smoking and alcohol/drug use should be discussed, with fewer stating that the topics of parent health and community violence should be discussed. The percentage ranges from 56% for community violence to 94% for smoking in the household. The frequency of clinician discussion varied significantly across the 6 topics (Fig 1). Relatively few parents reported that their child’s providers have asked about community violence (10%) or financial difficulties (12%), whereas more parents have been asked about parent health (39%), household drug or alcohol use (44%), emotional support (47%), or smoking in the household (77%).
Table 1 shows frequency of parent discussion with their own child’s clinician according to their view on whether clinicians generally should discuss the topic with parents. Most (74%) parents believed that smoking should be addressed and had themselves discussed it. For other topics, discussions among parents who viewed a topic as appropriate for discussion occurred less frequently. For example, the proportion who believed that a topic should be discussed and also had discussed the topic ranges from 9% of all parents on community violence to 47% on parent emotional support. Conversely, the proportion who believed that clinicians should discuss a topic but had not discussed the topic with their own child’s clinician is low for smoking (20%) but higher for less traditional health supervision topics such as financial difficulties (63%) and community violence (47%).
Some parents stated that pediatric clinicians should not discuss certain family and community topics. Summing columns 3 and 4 in Table 1 shows that ∼44% of parents said that parents should not be asked about issues of community violence, whereas fewer reported that providers should not ask about parent health (27%) or financial difficulties (25%). Large majorities of parents who reported discussing financial difficulties and community violence stated that clinicians should not ask about these topics. For example, ∼36% of all parents had been asked about financial difficulties, and more than half of these parents stated that this topic should not be discussed. Approximately 51% of all parents had been asked about community violence, and 82% of these parents stated that this topic should not be discussed.
Table 2 shows the bivariate associations between sociodemographic factors and parent views of what clinicians should discuss. Child age, race/ethnicity, marital status, household income, health insurance, setting of well-child care, and type of well child care provider are associated with parent views on at least 3 of the 6 topics (P < .05). Having a particular provider is associated with only 1 of the 6 topics. Table 3 examines the bivariate associations between sociodemographic factors and actual discussion of the topics. Nearly all child and family factors examined are associated with discussion of most topics, including maternal education, race/ethnicity, maternal age, child age, marital status, household income, health insurance status, provider type, and region of the country. However, having a particular provider is associated with only 1 of the 6 topics.
Multivariate logistic regression results are shown in Tables 4 and 5. Table 4 shows that parents of black children have greater odds than parents of white children of discussing parental drug or alcohol use, health, and emotional support. Hispanic parents have greater odds than white parents of discussing drug and alcohol use and community violence. Parents of older children have lower odds of reporting a recent discussion of most topics. Lower household income is associated with greater odds of discussion about household smoking and financial difficulties. Geographic region was not associated with topic discussion with the exception that more parents in the West region than in other regions discuss household drug or alcohol use. In comparison with children without a particular clinician for well-child care, children who see a family practitioner have twice the odds of discussing parental health, those with a nonphysician clinician have greater odds of discussing household smoking, and those who receive care from a pediatrician have lower odds of discussing household drug or alcohol use.
Table 5 shows predictors of no discussion among the subgroup of parents who reported that a given topic should be discussed. Determinants of no discussion despite parent views that discussion is appropriate are generally consistent with determinants of discussion. Black parents who reported that a topic should be discussed had lower odds of not discussing the topic than white parents who held the same views for 3 of the 6 topics: household drug or alcohol use, emotional support, and community violence. In general, increasing age of the child is associated with greater odds of no discussion in the past year. The exception is financial difficulties, for which odds of discussion among those who believed that discussion is appropriate are no different for parents of children aged 10 to 18 months and children aged 19 to 35 months compared with parents of children aged 4 to 9 months. Maternal age does not have a consistent association with the discussion of topics. Older mothers have higher odds than the youngest mothers of believing that financial difficulties should be discussed but not having such a discussion with their own child’s provider. Maternal education and marital status are generally not associated with lack of discussion. Setting of health care is associated only with discussion of community violence; compared with parents of children in community health centers, odds of not being asked about community violence are higher among parents of children in private or group practice (OR: 2.21) and urgent care or hospital outpatient facilities (OR: 2.81). Odds of no discussion among parents who believed that this topic should be discussed are lower for publicly insured children (OR: 0.22) and uninsured children (OR: 0.32) compared with privately insured children. For only 1 other topic (household drug or alcohol use) is insurance status associated with no discussion. Odds of no discussion among parents who believed that a topic should be discussed are lower for publicly insured children (OR: 0.58) compared with privately insured children.
This nationally representative study of early childhood health indicates that most parents believe that pediatric clinicians should address family and community influences on child health and development. Most parents said that any given topic should be discussed, ranging between 56% and 94% on the topics examined. Parents are more equivocal about the appropriateness of discussing less traditional topics such as community violence, parental health, and financial difficulties, although more than half support discussion of these topics as well. The relatively low frequency of discussion of many topics in this study, similar to previous findings,14,15 suggests that some parents are not given the opportunity to discuss topics that they believe are appropriate for pediatric providers to ask about.
Certain topics are clearly less frequently discussed despite parents’ views that these are appropriate topics for pediatric providers to discuss with parents. Only on household smoking are the majority of parents who said it is appropriate to discuss the topic actually asked about it. In the areas of parent health, household drug or alcohol use, and emotional support, approximately half of the parents who believed that clinicians should ask about these topics reported that their clinician did so. On the topics of financial difficulties and community violence, fewer than 20% of the respondents who believed that these are appropriate topics for discussion were actually asked about these topics. These findings indicate that pediatric providers may be underestimating risk or at least declining to raise topics among parents who may not object to being asked. The topic of community violence, however, provides a counter example. For this topic, the majority (82%) of parents who had been asked about community violence themselves believed that pediatric clinicians should not be asking about this topic.
That fewer parents of toddlers than of infants reported discussing most topics within the past year could be attributable to pediatric providers’ raising these issues initially but not readdressing the issue as the child grows, yet it may be particularly important to target parents of toddlers given that sources of emotional and parenting stress may change over time. The finding that parents of older children are equally likely to consider the topics appropriate to discuss adds support for ongoing assessment of family and community issues as children grow.
The findings of racial/ethnic and insurance differences are consistent with patterns from other studies on advice received during prenatal care.16 Clinicians, perhaps because of limited time, may target psychosocial assessments to subgroups of parents whom they perceive to be at higher risk. The need to tailor health supervision to the experiences and resources of the individual family is recognized in the professional guidelines.5 The increasing number of topics recommended for health supervision creates a growing pressure on how to use time in pediatric practice. It was recently estimated that even without addressing family and community issues, a typical pediatrician would have to spend 2.2 hours a day satisfying the minimum recommendations of the 1996 US Preventive Services Task Force Guide to Clinical Preventive Services,17 approximately twice the amount of time currently spent on preventive services.18 Nevertheless, the findings in this study clearly show that even groups that are not readily perceived to be higher risk (eg, those with greater incomes, when considering the risk factor of financial difficulties) believe that clinicians should be addressing these topics with parents in general.
Overall, rates of discussion fall short of AAP and Bright Futures recommendations, with only half of the parents of young children reporting discussion of topics such as alcohol/drug use or emotional support for the parent. Health care setting and pediatric provider type are not consistently associated with discussion of child and community topic, suggesting the need for greater surveillance among parents of all young children. Other studies show that addressing topics of interest to parents may yield greater satisfaction with the care received.7 Addressing topics that are salient to parents may influence their overall satisfaction and quality of care ratings. A large proportion of parents of young children in NSECH reported not receiving information about such issues as toilet training, sleep patterns, injury prevention, and discipline,19 but when these issues are discussed, parents give higher ratings of quality and satisfaction with care.20
There are certain limitations to this study. Parent memory of clinician encounters may not be entirely accurate, although research suggests that parents remember encounters better than total visits.21 However, if parents do not remember discussions with clinicians about these family and community topics, then it is unlikely that meaningful guidance and a positive change for the family resulted from the discussion. It is not known how frequently and intensively such topics need to be revisited to create ongoing and lasting impact. The effectiveness of discussion among parents who were asked about these topics is also unknown. We do not know how many parents would have found a discussion helpful had the topic been raised with them, because parents were not asked specifically whether they themselves would have benefited from a discussion. Provider type is based on parent report and thus may not always represent the actual specialty of the well-child care provider. Finally, because of small sample size, we could not describe the views and the care received for other racial/ethnic minorities such as Asians and American Indians.
In summary, the majority of parents believed that clinicians should raise family and community issues. Fewer endorsed the topic of community violence than other topics; whereas at least 75% of parents said that 5 of 6 studied topics should be discussed, fewer parents (55%) said that community violence should be asked about. Most parents are asked about the common environmental health issue of smoking, whereas fewer are discussing topics that are associated with the “new morbidities,” such as violence and financial difficulties, which have not received as much focus in traditional medical training. In general, parents from all ethnic and socioeconomic groups seem to support more extensive screening and discussion on the topics evaluated here. Clinicians may underestimate health risks in certain clinical situations, and this possibility argues for more universal assessment. That parents of toddlers are less likely than parents of infants to have discussed family and community topics, while believing that they are important, points to a need for ongoing screening of family and community risks as the child grows.
Some of these topics may be difficult for clinicians and parents to discuss comfortably. Training to increase the comfort level of pediatric clinicians with discussion of these topics may require not only additional knowledge about the topic area but also skills in addressing these issues in a responsive manner and the capacity to connect families to additional resources when needed. In a recent AAP periodic survey of its members, lack of reimbursement and inadequate training about psychosocial issues were reported to be significant barriers to addressing these issues.11 Greater familiarity of clinicians with available community services and supports, such as parenting support groups, could help clinicians feel more comfortable with addressing family- and community-based psychosocial risks. The reports from parents in this study indicate that they may be hesitant to discuss some topics related to family and community stress. Clinicians may be hesitant to raise topics because they know of few resources in their community to which families can turn when problems are identified, yet we know that, under the right circumstances, pediatric providers can refer families to community resources and to programs for which low-income families are eligible, programs that can provide emotional support in parenting, and programs that can help address financial difficulties.22 This study suggests that future research should investigate improvements in satisfaction and family outcomes that may occur through more universal assessment of family and community psychosocial risks. This study also shows that parents believe that assessments of family and community psychosocial risks deserve a place in the care of their children. What remains to be seen is how these assessments will be incorporated into routine pediatric care, how they can be financed, and what long-term effects they will have.
This research was made possible by funding from The Gerber Foundation, the American Academy of Pediatrics Friends of Children Fund, the Maternal and Child Health Bureau in the Health Resources and Services Administration (5-U05MC-00010200), and The Commonwealth Fund.
The opinions expressed in this article are solely those of the authors and do not reflect the views of the organizations with which they are affiliated.
- ↵Shonkoff JP, Phillips DA, eds. From Neurons to Neighborhoods. Washington, DC: National Academy Press; 2000
- ↵Alaimo K, Olson CM, Frongillo EA Jr. Food insufficiency and American school-aged children’s cognitive, academic, and psychosocial development. Pediatrics.2001;108 :44– 53
- ↵Roghmann KJ, Haggerty RJ. Family stress and the use of health services. Int J Epidemiol.1972;1 :279– 286
- ↵American Academy of Pediatrics, Committee on Psychosocial Aspects of Child and Family Health. Guidelines for Health Supervision III. Elk Grove Village, IL: American Academy of Pediatrics; 1997 (updated 2002)
- ↵Green M, Palfrey JS, eds. Bright Futures: Guidelines for Health Supervision of Infants, Children, and Adolescents. 2nd ed., rev. Arlington, VA: National Center for Education in Maternal and Child Health; 2002
- ↵Kelleher KJ, McInerny TK, Gardner WP, Childs GE, Wasserman RC. Increasing identification of psychosocial problems: 1979–1996. Pediatrics.2000;105 :1313– 1321
- ↵Erickson MJ, Hill TD, Siegel RM. Barriers to domestic violence screening in the pediatric setting. Pediatrics.2001;108 :98– 102
- ↵Olson AL, Kemper KJ, Kelleher KJ, Hammond CS, Zuckerman BS, Dietrich AJ. Primary care pediatricians’ roles and perceived responsibilities in the identification and management of maternal depression. Pediatrics.2002;110 :1169– 1176
- ↵Blumberg SJ, Olson L, Osborn L, et al. Design and operation of the National Survey of Early Childhood Health, 2000. Vital Health Stat 1.2002;(40)
- ↵Bethell C, Peck C, Schor E. Assessing health system provision of well-child care: the Promoting Healthy Development Survey. Pediatrics.2001;107 :1084– 1094
- ↵Olson LM, Inkelas M, Halfon N, Schuster MA, O’Connor KG, Mistry R. Overview of the content of health supervision for young children: reports from parents and pediatricians. Pediatrics.2004;113(suppl) :1907– 1916
- ↵Halfon N, Inkelas M, Olson LM, Mistry R, Bethell CD, Schuster MA. Measuring satisfaction with well child care: technical and interpersonal quality. Pediatr Res.2002;51 :161a
- Copyright © 2004 by the American Academy of Pediatrics