OBJECTIVES. Our objectives were to (a) estimate the prevalence of children's mental health problems, (b) assess family functioning, and (c) investigate the relationship between children's mental health and family functioning in Rhode Island.
METHODS. From the 2003 National Survey of Children's Health, Rhode Island data for children 6 to 17 years of age were used for the analyses (N = 1326). Two aspects of family functioning measures, parental stress and parental involvement, were constructed and were examined by children's mental health problems, as well as other child and family characteristics (child's age, gender, race/ethnicity, special needs, parent's education, income, employment, family structure, number of children, and mother's general and mental health). Bivariate analyses and multivariate logistic regression were used to investigate the relationship.
RESULTS. Among Rhode Island children, nearly 1 (19.0%) in 5 had mental health problems, 1 (15.6%) in 6 lived with a highly stressed parent, and one third (32.7%) had parents with low involvement. Bivariate analyses showed that high parental stress and low parental involvement were higher among parents of children with mental health problems than parents of children without those problems (33.2% vs 11.0% and 41.0% vs 30.3%, respectively). In multivariate logistic regression, parents of children with mental health problems had nearly 4 times the odds of high stress compared with parents of children without those problems. When children's mental health problems were severe, the odds of high parental stress were elevated. However, children's mental health was not associated with parental involvement.
CONCLUSIONS. Children's mental health was strongly associated with parental stress, but it was not associated with parental involvement. The findings indicate that when examining the mental health issues of children, parental mental health and stress must be considered.
- children’s mental health
- family functioning
- parental stress
- parental involvement
- National Survey of Children's Health
- Rhode Island
At least 1 in 5 children in the United States has a mental health disorder, and 1 in 10 has a serious emotional disturbance.1 These problems are a source of stress for the child, as well as the family, school, community, and larger society.1–3
It is important to understand the burden of children's mental health problems and their impact on family functioning to assess the needs and gaps in mental health services for children and their parents.4,5 Previous research has shown that parents of children with mental or behavioral health problems experience higher levels of stress in their parenting roles.6–9 Parents of these children have higher rates of mental health problems, as well as higher marital, employment, and financial problems.10–15 In addition, parents of these children are less likely to be involved or have a close relationship with their children because of stressful life circumstances.15
In Rhode Island, studies have indicated that there are inadequate resources (eg, child psychiatrists) and fragmented systems of care in mental health services for children and their families.16,17 However, there have been no population-based studies in the state that estimate the burden of children's mental health problems and its relationship with family functioning.
Our objectives were to (a) estimate the prevalence of children's mental health problems, (b) assess the level of family functioning, and (c) investigate the relationship between children's mental health and family functioning in Rhode Island. We hypothesized that parents of children with mental health problems would be more likely to have high stress levels and less likely to be involved with their children.
The 2003 National Survey of Children's Health's (NSCH) Rhode Island data were used for the analyses. The NSCH is a random-digit-dial telephone survey of a sample of parents with children <18 years of age from each of the 50 states and the District of Columbia, using a state-specific sampling frame. In Rhode Island, a total of 2019 interviews were completed, with the weighted overall response rate of 57.1%. Among them, children 6 to 17 years of age (N = 1326) were used for this study, because the measures of mental health status and family functioning for children <6 years of age were less reliable. Human subjects review was not required for this study. The survey design is described briefly in the article by Kogan and Newacheck in this issue18; more in-depth information can be found elsewhere.19
Two aspects of family functioning were assessed: parental stress and parental involvement with their child.
Parental stress was measured by using the Aggravation in Parenting Scale (APS), which was derived from the Parenting Stress Index.20 The APS is an index of stress in the parenting role measured by administering a 4-item sequence of questions that have been used in many other studies.21–23 In accordance with previous studies, we used those 4 questions to measure the level of parental stress. The 4 questions were: “During the past month, how often have you felt [sample child] is much harder to care for than most children [his/her] age?”; “During the past month, how often have you felt [he/she] does things that really bother you a lot?”; “During the past month, how often have you felt you are giving up more of your life to meet [sample child]'s needs than you ever expected?”; and “During the past month, how often have you felt angry with [him/her]?” Respondents were asked to choose 1 answer among 4 categories: never, sometimes, usually, or always. Following previous research, we summed respondents' answers to create scores ranging from 4 to 16: a score of ≤11 was defined as high parental stress.21–23
Parental involvement was measured by using 2 questions: “Is your relationship with [sample child] very close, somewhat close, not very close, not close at all?”; and “How well can you and [sample child] share ideas or talk about things that really matter? Would you say very well, somewhat well, not very well, or not well at all?” Parents who reported “very close” and “very well” to both questions were defined as having high parental involvement.
The measures of parental stress and parental involvement were coded as a dichotomous variable (high versus low) for the analyses.
Four questions were used to identify whether a child had a mental health problem: “Does [sample child] have any kind of emotional, developmental, or behavioral problem for which [he/she] needs treatment or counseling?”; “Has a doctor or health professional ever told you that [sample child] has: Attention Deficit Disorder (ADD) or Attention Deficit Hyperactive Disorder (ADHD)?; depression or anxiety problems?; behavioral or conduct problems?” A child is classified as having a mental health problem when a parent answered “yes” to any of these 4 questions. Therefore, when we refer to “child's mental health problem” in this article, it refers to a broader array of children's mental, emotional, developmental, and behavioral problems. The degree of the child's mental health problem was further classified by using the number of questions that were answered “yes” by parents, with categories of no problem, 1 problem, and 2 or more problems. We considered 2 or more problems as a severe problem in this article.
Other covariates include child's characteristics (age, gender, race/ethnicity, and special care needs), household or parental characteristics (education, household income, family structure, employment status, number of children in the household, and mother's general health and mental health). These variables were considered as confounding factors when examining the association of children's mental health with the dependent variables in the logistic regression analyses.
Analyses were conducted by using the Survey Data Analysis (SUDAAN 9.0) statistical software to account for the weights and the complex survey design.24 Univariate descriptive statistics were used to describe the prevalence of our independent and dependent variables: children's mental health problems, high level of parental stress, and low level of parental involvement. Bivariate analyses were employed, by using the χ2 tests, to examine the estimated prevalence in relation to selected child/family/parent characteristics. Multivariate logistic regression was used to determine the association of children's mental health status with parental stress and parental involvement. A logistic regression model was constructed with variables that were significantly associated with at least 1 of our dependent variables in the bivariate analysis. Because mother's general health was highly correlated with mother's mental heath (correlation coefficient ρ = 0.42), we excluded mother's general health from the final model. Separate logistic regression models were constructed for parental stress and parental involvement. Responses of “don't know” and “refuse to answer” were excluded from the analyses. The results presented here are weighted data to reflect the 165778 Rhode Island children 6 to 17 years of age, except for sample sizes (N) that were unweighted.
Table 1 describes the distribution of the study population by selected child/family/parent characteristics used in the analyses. The characteristics of the children included: more than half (59%) of the children were aged 6 to 12 years; 51% were male; the majority were white (86%), non-Hispanic (87%), and did not need special services or care (86%). The characteristics of the household or parents included: the majority had more than a high school education (anyone in the household who had the highest level of education) (66%); one third (33%) of the children lived in households with <200% of the federal poverty level; 61% of the children were living with 2 parents (biological/adopted); 89% were living in households with anyone who was employed ≥50 weeks in the past year; 62% of the household had 1 or 2 children. The majority of the mothers of these children had “good, very good or excellent” mental health (91%) and “good, very good or excellent” general health (90%).
Univariate and Bivariate Analyses
Table 2 shows the prevalence of children with a mental health problem, high parental stress, and low parental involvement, and their relation to selected characteristics.
Mental Health Problem
Nearly 1 in 5 (19.0%; 95% confidence interval [CI]: 16.5–21.9) children 6 to 17 years of age in Rhode Island had mental health problems by our definition. The proportion of children in Rhode Island that had a severe mental health problem was 9.8% (95% CI: 8.0–12.1; data not shown). The prevalence of children's mental health problems varied by child and parent social/health/demographic characteristics. Compared with their counterparts, the rate of mental health problems was significantly higher among teens (22.8%); males (22.7%); children with special health care needs (56.6%); those living below 200% of the poverty level (25.3%); those living with single mothers (25.7%) or in stepfamilies (26.6%); those living in the families with no adults who worked full-time (30.2%); and children whose mothers had “poor or fair” mental health (47.0%) or “poor or fair” general health (36.0%).
Approximately 1 (15.6%) in 6 (95% CI: 13.3–18.1) Rhode Island children lived with a parent who felt highly stressed. Parents who had children with mental health problems were 3 times more likely to be stressed than parents of children without mental health problems (33.2% vs 11.0%). The prevalence of high parental stress was higher among parents of teens (18.6%), Hispanics (32.9%), and children who needed special care (26.5%). Parental stress was higher among parents with less than a high school education (26.0%), household incomes <200% of federal poverty (21.8%), single mothers (23.0%), and those not working full-time (26.4%). Mothers with “poor or fair” mental health were more likely to be stressed (34.5%).
Nearly one third (32.7%; 95% CI: 29.6–35.9) of parents had a low level of involvement with their child. Parents who had children with mental health problems were more likely to have low involvement with their child than parents who had children without mental health problems (41.0% vs 30.3%). Low parental involvement was more prevalent among parents of teens (44.7%), parents with 3 or more children (37.4%), and among mothers who have “poor or fair” mental health (55.5%) and “poor or fair” general health (43.0%). Parents with incomes of 200% to 400% of poverty level (38.6%), single mothers (37.2%), or stepparents (45.4%) were more likely to have low involvement compared with their counterparts.
Our final model of multivariate logistic regression, based on our bivariate analysis, included child's mental health status, child's age, race/ethnicity, whether or not the child needed special care, parent's education, poverty level, family structure, employment status, number of children in the household, and mother's mental health (Table 3).
The results of logistic regression revealed that children's mental health was strongly associated with parental stress. After controlling for all variables in the model, parents of children with mental health problems had nearly 4 times the odds of high stress (adjusted odds ratio [aOR]: 3.9; 95% CI: 2.3–6.6), compared with parents of children without mental health problems. When a child's mental health problems were severe, the odds of high parental stress were elevated (aOR: 5.7; 95% CI: 2.8–11.6; data not shown). In addition to children's mental health status, factors significantly associated with high parental stress were race/ethnicity, parental education, mother's mental health status, and whether or not the child needed special care. Blacks (aOR: 2.9; 95% CI: 1.1–7.8), Hispanics (aOR: 3.6; 95% CI: 1.5–8.4), parents of children who needed special care (aOR: 2.0; 95% CI: 1.1–3.7), parents with a high school education (aOR: 2.2; 95% CI: 1.1–4.3), and mothers with poor mental health (aOR: 2.3; 95% CI: 1.1–4.8) had significantly increased odds of high parental stress compared with their reference groups.
After controlling for child and parent characteristics in the logistic regression model, children's mental health was not associated with parental involvement (aOR: 1.2; 95% CI: 0.7–1.9). In this model, child's age, parental employment status, and mother's mental health were significantly associated with the level of parental involvement. Low parental involvement was significantly higher among mothers of teens (aOR: 2.6; 95% CI: 1.8–3.7), families with no one employed full-time (aOR: 2.3; 95% CI: 1.2–4.6), and mothers with poor mental health (aOR: 2.7; 95% CI: 1.4–5.3) compared with their reference groups.
This is the first population-based study conducted in Rhode Island that allows us to estimate the prevalence of children's mental health problems and to investigate the relationship between children's mental health and family functioning. According to our study, nearly 1 (19.0%) in 5 Rhode Island children 6 to 17 years of age, or an estimated 31500 children, had a mental health problem. Approximately 1 (9.8%) in 10, an estimated 16200 children, had a severe problem. Using the same definition and data source, the prevalence of children with mental health problems for the United States was 15.5% (95% CI: 15.0–15.9) and those with a severe problem was 7.7% (95% CI: 7.3–8.1). This result is consistent with the national report that ∼20% of children and adolescents have a mental health disorder, and 10% have a serious emotional disturbance.1 Among those children who had mental health problems, Approximately one half (48.7%) of Rhode Island children reported that they received mental health care or counseling 12 months before the survey, compared with 42.2% of the US children.
Rhode Island family functioning level, measured by parental stress and parental involvement, was lower than the United States as a whole. Our study revealed that 15.6% of Rhode Island parents felt high stress, and 32.7% of parents did not have strong involvement with their children. The figures for the United States were 13.5% (95% CI: 13.1–14.0) and 29.6% (95% CI: 29.0–30.3), respectively.
The parental stress and the parental involvement seemed to be quite different aspects of family functioning (correlation coefficient ρ = −0.14). Accordingly, factors associated with parental stress and parental involvement were quite different in the results of the logistic regression. For example, a child's mental health problem and/or a child's need of special care were strongly associated with high parental stress, although these factors were not associated with parental involvement. In other words, a child's poor physical or mental health may be a stressor for the parents, but these factors do not affect parental involvement with their child. In addition, minority parents (black and Hispanic) were found to have significantly higher stress levels, whereas minority status was not associated with parental involvement. Parental involvement was significantly lower among those parents who had teenagers and parents who were not fully employed; however, these factors were not associated with parental stress. Mother's mental health status was found to be the only factor that was significantly associated with both parental stress and parental involvement. If a mother had poor mental health, the family functioning level was significantly lower.
Several limitations of this study should be noted. First, our study found that children's mental health was strongly associated with parental stress. Because this is a cross-sectional study, we could not determine the direction of the association. Children's mental health certainly can impact parental stress, but high parental stress can also be a risk factor of children's mental health problems.13,14 Second, there is no standard definition of children's mental health problems in the survey questions, making it difficult to compare with other sources of data. By asking a parent whether a doctor diagnosed their child with certain symptoms of mental health problems, we may exclude those children who experienced the symptoms but were not diagnosed by a doctor. Therefore, the prevalence of children with mental health problems may be underestimated. Third, previous research revealed that the measure of parental stress using 4 questions has limited cultural validity among Spanish-speaking Hispanic parents.23 The cultural validity of this measure was not verified in this study. Lastly, by using state-specific data, some groups have small sample sizes that make it difficult to produce meaningful results (eg, blacks, multiple and other races, those with less than a high school education, etc).
When examining the mental health issues and needs of children, parental mental health and stress must also be considered. Identification of families with high stress levels and providing appropriate support services to address the needs of both parents and children is critical. Health care providers, schools, public and private insurers, and social service programs can play a key role in early identification and counseling. It is important that these services and programs be accessible, family-centered, continuous, comprehensive, compassionate, culturally effective, and coordinated via the medical home concept.
- Accepted September 15, 2006.
- Address correspondence to Hyun (Hanna) K. Kim, PhD, Division of Family Health, Rhode Island Department of Health, 3 Capitol Hill, Room 302, Providence, RI 02908-5097. E-mail:
The authors have indicated they have no financial interests relevant to this article to disclose.
- ↵US Department of Health and Human Services. Mental Health: A Report of the Surgeon General. Rockville, MD: US Department of Health and Human Services; 1999
- Cross Creek Family Counseling. Behavioral problems. Available at: www.crosscreekcounseling.com/behavior.html. Accessed March 26, 2006
- ↵President's New Freedom Commission on Mental Health. Policy options: Subcommittee on Children and Family—promoting, preserving and restoring children's mental Health. February 5, 2003. Available at: www.mentalhealthcommission.gov/subcommittee/children_family020703.doc. Accessed April 30, 2006
- ↵Brauner CB, Stephen, CB. Estimating the prevalence of early childhood serious emotional/behavioral disorders: challenges and recommendations. Public Health Rep.2006;121 :303– 310
- ↵Waxman HA. Improving the care of children with mental illness: a challenge for public health and the federal government. Public Health Rep.2006;121 :299– 302
- Cameron SJ, Orr RR. Stress in families of school-aged children with delayed mental development. Can J Rehabil.1989;2 :137– 144
- ↵University of Washington. Research shows link between child and parental mental-health problems. Available at: www.uwnews.org/article.asp?articleID=2150. Accessed April 22, 2006
- ↵Washington Kids Count. Family matters: mental health of children and parents. Available at: www.hspc.org/wkc. Accessed April 15, 2006
- ↵Moore KA, Vandivere S. Stressful Family Lives: Child and Parent Well-being. Washington, DC: Urban Institute, New Federalism; 2000:Series B (17)
- ↵The Rhode Island SHAPE Foundation. Difficult Seas: Navigating Behavioral Healthcare in Rhode Island. Providence, RI: Rhode Island SHAPE Foundation; 2005
- ↵Rhode Island Kids Count. Children's Mental Health. Providence, RI: Rhode Island Kids Count Factbook; 2006
- ↵Kogan MD, Newacheck PW. Introduction to the volume on articles from the National Survey of Children's Health. Pediatrics.2007;119(suppl 1) :S1– S3
- ↵Blumberg SJ, Olson L, Frankel M, Osborn L, Srinath KP, Giambo P. Design and operation of the National Survey of Children's Health, 2003. Vital Health Stat 1.2005;(43) :1– 124
- ↵Abidin RR. Parenting Stress Index: a measure of the parent-child system. In Zalaquett CP, Wood RJ, eds. Evaluating Stress: A Book of Resources. Lanham, MD: Scarecrow Press, Inc; 1997:277–291
- ↵Macomber JE, Moore KA, Brown BV. Adults' environment and behavior: parental aggravation. Available at: www.urban.org/publications/900872.html. Accessed February 25, 2006
- The Urban Institute. 1997 Snapshots of American families: definitions of indicators. Available at: www.urban.org/url.cfm?ID=900878. Accessed February 27, 2006
- ↵Oster H, Morales L, Anderson RM, et al. The Aggravation in Parenting Scale: Language application issues among Spanish-speaking Latino families living in Los Angeles county. Presented at: Academy of Health Annual Research Meeting; June 23–25, 2002;Washington, DC
- ↵Research Triangle Institute. SUDAAN Language Manual: Release 9.0. Research Triangle Park, NC: Research Triangle Institute; 2004
- Copyright © 2007 by the American Academy of Pediatrics