This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Web of Science (14)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Mansour, M. E.
Right arrow Articles by DeWitt, T. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mansour, M. E.
Right arrow Articles by DeWitt, T. G.
Related Collections
Right arrow Office Practice
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Facebook   Add to Reddit   Add to Technorati   Add to Twitter  
What's this?
PEDIATRICS Vol. 111 No. 6 June 2003, pp. 1372-1381

Health-Related Quality of Life in Urban Elementary Schoolchildren

Mona E. Mansour, MD, MS*, Uma Kotagal, MD, MSc{ddagger},§, Barbara Rose, MPH§, Mona Ho, MS, David Brewer, Ashwini Roy-Chaudhury, MPH*, Richard W. Hornung, DrPH, Terrance J. Wade, PhD and Thomas G. DeWitt, MD*

* Divisions of General and Community Pediatrics
{ddagger} Neonatology
§ Clinical Effectiveness, Children’s Hospital Medical Center, Cincinnati, Ohio
Child Policy Research Center, Children’s Hospital Medical Center and the Institute for Health Policy and Health Services Research, University of Cincinnati, Cincinnati, Ohio


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Background. Health disparities between children from urban minority backgrounds and children from more affluent backgrounds are well-recognized. Few studies specifically address urban children’s perceptions of their health-related quality of life (HRQOL) or the factors that contribute to these perceptions. Since schools are pivotal to children’s intellectual, social, and emotional development, school connectedness may be a factor that contributes to their perception of HRQOL.

Objective. To examine children’s perceptions of HRQOL in an elementary school-based population of urban children.

Methods. The study population consisted of 2nd, 3rd, and 5th graders from 6 urban kindergarten to 8th grade schools and their parents. Children completed a survey that included questions on HRQOL and school connectedness. Parents completed a telephone survey that assessed demographics, the child’s health, health care usage, and parental health status. Data on school absences and mobility from the computerized school database were linked to survey data. Bivariate analyses were used to evaluate the association between child report of HRQOL and collected variables, including school connectedness. Multivariable linear regression was conducted to identify the factors best predicting HRQOL in these urban children.

Results. Of the 1150 eligible students, parent and child survey data were available for 525 (45.6%). Fifty-one percent of students were male and 89% were black. Ninety-four percent of parents were female, 29% were married, and 62% had family incomes below $20 000 per year. The mean total score for HRQOL was 67.2, with a possible range of 0 to 100 (higher scores reflecting better HRQOL). In the multivariable analysis, child grade, the relationship of the " parent" to the child, employment, family income, type/presence of insurance, and school connectedness were significantly associated with the HRQOL total score.

Conclusions. Young urban children self-report low HRQOL scores and do so as early as the 2nd grade. These low scores, which reflect children’s own perceptions of impaired psychological and physical health, have potential implications for the success of urban children in their learning environments. The association between HRQOL and school connectedness might suggest that health and educational programs that improve a child’s attachment to school could result in improved perceptions of health by urban children.

Key Words: school health • underserved populations • health status

Abbreviations: HRQOL, health-related quality of life • CSDHP, Cincinnati School Health Demonstration Project • SF, short form


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Health disparities between children from urban minority backgrounds and those from more affluent backgrounds are well-recognized.1–6 In contrast, there are less data regarding health-related quality of life (HRQOL) for urban children and families.7,8 In addition, many assessments of child health and health status have focused on physical aspects of health, and have disregarded broader definitions of health that include other aspects of a child’s psychosocial health and well-being.

For young children, schools play a pivotal role in their intellectual, social, and emotional development. Children’s health is intricately intertwined with their ability to learn. Children’s perception of closeness to school personnel and the school environment, "school connectedness," may be a factor that contributes to their perception of HRQOL. School connectedness has been identified in older children and adolescents to be associated with less involvement in risk-taking behaviors and with better perceived health status.9–14 Social bonding or attachment-based interventions have been effective in decreasing social alienation and risk-taking behavior in youth.15–18 One meta-analysis of school-based prevention programs and interventions targeted at problem behaviors in adolescents revealed positive changes in attachment to school are consistently accompanied by reductions in problem behavior.19 Understanding the relationship between HRQOL and school connectedness in younger children may identify 1 potential mechanism for improving health and educational outcomes for at-risk children.

Many of the tools developed to examine HRQOL in pediatric populations focus on specific diseases or conditions such as juvenile rheumatoid arthritis, cancer, asthma, and diabetes, out of the need to more comprehensively assess the impact of chronic illness on children.20–27 However, there is a need for tools that can be used across various pediatric populations to measure quality of care and clinical effectiveness of medical interventions or programs aimed at improving health outcomes of all children.28–33

The purpose of this study is to examine urban children’s own perceptions of their HRQOL and the factors that contribute to this perception of health in a school-based population. The measurement of HRQOL was conducted as part of a 3-year longitudinal project, the Cincinnati School Health Demonstration Project (CSHDP). The CSHDP is a collaboration between the Cincinnati Public Schools, the Cincinnati Health Department, and the Children’s Hospital Medical Center to examine health and educational outcomes of 3 different models of school-linked health care services over a 3-year period. All data are from the baseline assessment of children in the schools before any health service intervention.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Sample
The target population was 2nd, 3rd, and 5th grade students from 6 kindergarten through 8th grade public elementary schools, and the person who makes the health care decisions for the student, referred to in this study as the parent. Six project schools were selected from a list of Cincinnati public elementary kindergarten through 8th grade schools with the following characteristics: high minority population (>70%) and high percentage of children in federal school lunch program (>60%). Letters were mailed to parents describing the CSHDP and informing them they may be contacted to complete a survey regarding their child and their child’s health.

Telephone surveys were conducted with parents in November and December of 1999. At the end of the parent telephone survey, parents were asked to provide verbal consent to administer a survey to the child while at school. Parents could decline participation for themselves and/or their child and this would not affect receipt of services provided by the project. Parents unable to be reached by phone were sent written consent forms for their child’s participation in the survey, as well as letters asking them to contact project staff if they were interested in participation. The child survey was administered to children on site at the individual project schools between November 1999 and February 2000.

Attempts were made to contact all parents of 1228 students identified in the 2nd, 3rd, and 5th grades at the 6 project schools as indicated by the centralized school computer database. Of the 1228 students, 78 were ineligible because children had either changed grades or were not enrolled at 1 of the project schools at the time the parent survey was conducted. Of the remaining 1150 children, 340 (29.6%) were not contacted because of no telephone service or wrong number, 70 (6.1%) had no answer after multiple attempts, and 46 (4.0%) could not be contacted for other reasons. Of the 694 parents contacted, 63 refused to participate, giving a response rate of 91% for those contacted (631/694) and 54.8% (631/1150) of those presumed to be eligible.

Among the 1150 eligible parents, 633 were contacted by phone and 21 were contacted by mail to ask for consent for child participation. Parents who refused to participate in the parent survey may have terminated the telephone conversation before initiation of the consent process for the child. Consent to administer the child portion of the survey was given for a total of 560 children resulting in a response rate of 85.6%. Eighty-nine parents contacted (13.6%) refused participation for their child; 5 "parents" were not the legal guardian and thus could not give consent for participation. Because of high mobility in this population, 33 (5.9%) children for whom consent was obtained were no longer in the school at the time of the child survey. In addition, 2 consented children were not developmentally capable of completing the survey, leaving 525 completed child surveys.

The analysis was performed on the 525 parent and child pairs for whom we had a completed child and parent survey.

Data on enrollment, absences, and school mobility were provided on individual students in electronic form from the computerized school database. Information on school health service use was collected through a computerized school health information system, Welligent, 34 that is populated with demographic data from the central school database. Data from the surveys were linked to absence, enrollment, and school service data through the use of a unique identifier, the school identification number. Once data from the different sources were linked, personal identifiers were removed to maintain confidentiality. The study protocol was approved by the Institutional Review Boards of the Children’s Hospital Medical Center, the Cincinnati Health Department, and the review board at the Cincinnati Public Schools.

Measures
Parent Survey
The parent survey included questions on demographics, the target child’s health including overall perception of child’s health, the presence/type of chronic health conditions, health care usage, parental health status, and the parent’s perception of their child’s school connectedness.

Parent perception of child health status was assessed with the following question, "In general, how would you rate [child’s name] health ... excellent, very good, good, fair, or poor?" Chronic illness data were obtained from 2 sources: questions on the parent survey and data collected at the school in the computerized health record, Welligent. The parent survey asked about 8 specific chronic illnesses (diabetes, attention-deficit/hyperactivity disorder, learning disability, developmental delay/mental retardation, sickle cell disease, seizure disorder, asthma, and headaches) and an open-ended question about any other chronic health conditions diagnosed by a doctor or other health care professional. These 8 conditions were selected because of their suspected high frequency in the general pediatric population and/or the target population. A child was defined as having a chronic illness if any source or individual question indicated a positive response for chronic illness. Health care usage questions addressed presence of a primary care provider, type/location of this provider, and number of ill and well health care visits by the child to their primary care provider in the past 12 months.

Parental health status was assessed using the short form (SF)-12 that results in a physical and mental component summary scale. Both scales are transformed to have a mean of 50 and a standard deviation of 10.35 The SF-12 is a shortened version of the SF-36, and has undergone reliability and validity testing for use in varied populations. School connectedness was measured using 7 questions modified for the CSHDP from the National Longitudinal Study of Adolescent Health.9 Five questions have a score of 1 (strongly disagree) to 5 points (strongly agree), and 2 questions have a score of 1 (everyday) to 5 points (never) with the overall school connectedness score ranging from 7 to 35 (see Table 3 for individual questions). Higher scores reflect higher levels of school connectedness. Reliability and validity testing has been performed with the school connectedness questions in adolescent populations, but not with younger, elementary-aged children.9,36


View this table:
[in this window]
[in a new window]
  		TABLE 3. Distribution of Child Responses to School Connectedness Questions (N = 525)

 
Child Survey
The child survey contained questions on HRQOL and school connectedness. The child assessment of HRQOL was done using the PedsQL Version 4.0, a 23-item measure of pediatric HRQOL. Each item has 5 response selections that range from "never" to "almost always." The PedsQL is scored from 0 to 100, with higher scores reflecting better HRQOL. The PedsQL has a total score and 2 subscale scores, physical and psychosocial. The psychosocial subscale has questions that assess the domains of emotional, social, and school function. The PedsQL has age-appropriate versions for child self-report for ages 5 to 7, 8 to 12, and 13 to 18. For this study, the 8- to 12-year tool was self-administered by 5th graders, but was interviewer administered to both 2nd and 3rd graders because of lower reading capabilities in this urban population as well as for consistency in tools used across years of the CSHDP. During pilot testing, 8-year-old children with similar racial and socioeconomic backgrounds of study subjects consistently had difficulty answering questions without the interviewer reading certain segments aloud. The same questions modified from the National Longitudinal Study of Adolescent Health contained in the parent survey were asked of students.

Other Data Sources
School absences were defined as the total number of days out of school, including verified, unverified, excused, and unexcused absences. School mobility was defined as a continuous variable, the number of school moves within the 1999–2000 academic school year.

Data Analysis
The primary outcome measure was pediatric HRQOL using the total score of the PedsQL. Analyses were also done using the physical and psychosocial subscales as outcome measures.

Data regarding the frequency of responses and the distribution of school connectedness are presented, as this was 1 of the key independent variables of interest. Because reliability and validity testing of the connectedness questions have not been performed with elementary-aged students, we assessed internal consistency using Kronbach’s {alpha}, and examined construct validity by examining school connectedness in children with low, middle, and high levels of school absences. Children with more school absences and less opportunity to interface with school personnel and other students were expected to have lower school connectedness scores.

Descriptive statistics were computed for all outcome variables and for all covariates. Means and standard deviations were calculated for all continuous variables and proportions were computed for all categorical variables (Tables 1 and 2). To investigate relationships between outcome variables and the set of covariates listed in Tables 1 and 2, a series of analyses was conducted. Student’s t tests were used to assess the difference in HRQOL for categorical variables with 2 levels. Analysis of variance was used for categorical variables with >2 levels, and linear regression was used for continuous variables. Residuals were tested to determine if any transformations were necessary to satisfy normality assumptions and none were needed.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Characteristics of Children in the Sample (N = 525)

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. Characteristics of "Parents" in the Sample (N = 525)

 
Bivariate relationships were examined between HRQOL and the following variables: parental health status, parental perception of child health status, grade, race, gender of the child, presence of chronic illness in the child, type of health insurance for the child, age, gender, marital status, employment status, educational level of the parent, family income, family size, school absences, school mobility, school connectedness, presence of primary care provider, type of primary care provider, location of provider, and number of visits to the primary care provider for well and ill care.

All variables with P values < .2 for any HRQOL scale, as well as variables with literature support (parental health status, 37,38 marital status,39,40 educational level,1,5,38 health insurance,3,4,41 and school absences42,43) for having a relationship with HRQOL or health status were considered as candidates for multivariable linear regression analysis to identify the set of variables that best explain HRQOL in these children. Variables were entered in the model using a modified stepwise procedure to assure that variables considered a priori as potentially important were given the highest priority for inclusion. After the models were developed, regression diagnostics were run to ascertain whether the set of independent variables in the model resulted in collinearity or if a small number of observations exercised too much influence on the model. Final models (reported in Table 5) contain all factors found to be significant at the .05 {alpha} level. Because of missing data for certain variables, the number of children/parent pairs included in the multivariate analysis for total, physical, and psychosocial HRQOL were 413, 426, and 412, respectively.


View this table:
[in this window]
[in a new window]
  		TABLE 5. Multivariate Analysis of Child Perception of HRQOL

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Demographics
Of the 525 child subjects, 51% were male, 89% were black, and there was an equal distribution of grades (35% from the 2nd grade, 30% from the 3rd grade, and 35% from the 5th grade). Forty-four percent of children had public forms of health insurance and ~18% had no insurance (Table 1). Of the 525 parents, 94% were female, 89% were the birth parent, 29% were currently married, and the mean age was 34.6 years. Sixty-two percent of families had household incomes below $20 000/year. (Table 2)

Children of parents who responded were not significantly different from children of parents who did not respond in gender, race, or presence of chronic illness. However, children of parents who did not respond were more likely to have moved schools 1 or more times during the school year (P < .0001) and were more likely to be absent from school (P = .005).

Child HRQOL
The mean for the total score of HRQOL was 67.2 + 16.2, range 0 to 100. Mean psychosocial and physical subscale scores were 64.6 + 17.2 and 71.9 + 18.6, respectively. Selected frequencies for individual items are reported to demonstrate the responses of children. Almost 18% of children reported it was "almost always" or "often" hard to do sports activities, and 12% reported it was "almost always" or "often" hard to walk >1 block. Almost 20% of children answer with these same 2 response categories to the statement "I feel angry," and 26.5% respond similarly to the statement " I worry what will happen to me." Nineteen percent of children report that "almost always" or "often" other kids do not want to be my friend and 16% report that they miss school for not feeling well "often" or "almost always."

School Connectedness
The mean score for child perception of school connectedness was 26.7 + 4.7, range 7 to 35, with higher scores reflecting higher levels of school connectedness. Fifteen percent of children reported that they "disagree" or "strongly disagree" with the statement "I feel safe in my school" and a similar percent (16%) "disagree" or "strongly disagree" with the statement "I feel close to people in this school." Twenty-one percent report that they have trouble getting along with other students "almost everyday" or "everyday," with an additional 10% having trouble "about once a week" (Table 3). The internal consistency, as measured by Kronbach’s {alpha}, for the 7 school connectedness was .62. Supporting the construct validity of the school connectedness measure, children with more absent days had statistically significantly lower mean school connectedness scores compared with children with fewer school absences (P = .003). The mean school connectedness scores for the low, middle, and high absence groups were 27.6, 26.3, and 26.1, respectively.

Bivariate Analysis
Child grade, parental age, family income, and child’s perception of school connectedness were positively associated with the child’s perception of HRQOL total score (Table 4). The number of health care visits to the primary care provider for well-child care and the number of visits to the primary care provider for ill child care were negatively associated with HRQOL. Full-time employment of the "parent" was associated with better HRQOL compared with "parents" with part-time or no employment. Male children reported better HRQOL than female children. When the "parent," the person who made the health care decisions for the child, was the birth parent, the child reported poorer HRQP than when the "parent" was a non-birth parent.


View this table:
[in this window]
[in a new window]
  		TABLE 4. Bivariate Associations With Child’s Perception of HRQOL (PedsQL)

 
Variables significantly associated with the physical subscale of HRQOL include those for the total scale. In addition, parental perception of the child’s HRQOL was positively associated with the physical subscale score. No additional variables were significantly associated with the psychosocial subscale, and gender and number of visits for well child care were no longer significant.

Multivariable Analysis
In linear regression models, variables that were included in the final model for the total score of child HRQOL were child grade, child gender, family income, relationship of "parent" to the child, parent employment, insurance, and school connectedness (Table 5). The adjusted R2 for the final model was .30. Child grade, family income, and school connectedness were positively associated with the total score of HRQOL. Every point increase on the school connectedness scale is associated with a 1.38-point increase in the self-report of the total score of the PedsQL. A clinically significant increase in HRQOL as measured by the PedsQL is 5 points. Therefore, relatively small changes in connectedness are associated with clinically significant differences in HRQOL.

Children living with non-birth parents had total HRQOL scores 9.3% better than those living with birth parents. Children with a full-time working parent had 8.0% better self-reported total HRQOL compared with those with a part-time working parent. Contrary to expectations, children with private insurance fared worse than children with public insurance (Medicaid, Medicare, Children’s Health Insurance Program). Children with private insurance had 6.6% lower total PedsQL scores than children with public health insurance.

Variables significant in the model for physical health were child grade, school mobility, school absences, school connectedness, parental rating of child’s HRQOL, and whether the "parent" was or was not a biological parent. The same direction of relationship was demonstrated for child grade and school connectedness as was present for the total scale. In addition, there was a positive relationship between the physical subscale and school absences as well as parental ratings of the child’s HRQOL. School mobility was negatively correlated with physical HRQOL. Children living with their birth parent had worse HRQOL compared with those children living with nonbiological "parents."

Variables included in the model for psychosocial health were child grade, child gender, family income, parent age, parent employment, insurance, and school connectedness. The direction of the relationship of these variables to the psychosocial subscale of HRQOL were the same as for total and physical health scores. "Parent" age was positively associated with better psychosocial HRQOL. The adjusted R2 for the physical and psychosocial models were .24 and .27, respectively.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
This project provides some of the first published data that focuses on the HRQOL of urban children. As early as the second grade, these young urban children rate their HRQOL below that of children with known chronic physical health conditions in populations tested by Varni,44 the developer of the tool. Healthy children in Varni’s study had mean total scores of 83, and children with acute and chronic health conditions had means of 79 and 77, respectively.44 These low scores that reflect urban children’s own perceptions of their impaired psychological and physical health have potential implications for their success in learning environments.

These lower scores may simply reflect that when health is more broadly defined to include social and emotional aspects of health that urban children fare poorly. Previous studies have documented that children with known chronic health conditions rate their HRQOL and psychological adjustment below that of "healthy" children.32,45,46 In our study, the trend in mean scores for students with a chronic health condition were lower than mean scores for children without chronic illness, but were not statistically significant. It is possible that the psychosocial and emotional well-being of urban children is so poor that as a group they function similarly to children with chronic illness.

It is also plausible that the broad definition of chronic illness masked the ability to detect differences in HRQOL between children with and without chronic illness. In the Stein et al45 study, both analyses that included and excluded children with cognitive and emotional chronic health conditions demonstrated poorer psychological adjustment when compared with healthy peers. However, mean scores of psychological adjustment were lower when mental health conditions were included, and there were fewer differences between healthy children and children with chronic conditions when children with mental health conditions were excluded. To further explore this issue, posthoc analyses were conducted examining HRQOL scores for children with a specific physical health condition, asthma, compared with children without a chronic health condition. In t tests comparing mean scores, children with asthma did not have significantly lower scores on the PedsQL scales compared with children without a chronic illness (ptotal = .8, ppsych = .7, pphysical = .3). However, the mean PedsQL scores were consistently lower for all 3 HRQOL scales for children with asthma.

The relationship between HRQOL and school connectedness is both interesting and encouraging. School connectedness has been documented to be associated with decreased risk behaviors and perception of health status in older, adolescent populations.9–14 Results from our elementary-aged population reveal that school connectedness scores are higher than scores for older, urban, adolescent children from the same geographic area.47 In the Bonny study,47 mean scores were 15.7 + 4.7 on a scale using only 5 of the 7 school connectedness questions (range: 5–25). The mean of these 5 questions in our study was 19.0 + 3.7. There may be a natural decline in school connectedness as children get older. Possibly school-based health interventions initiated at the elementary level may allow better, earlier connection to the school and prevent or postpone a child’s involvement in risk-taking behaviors.

Data in the literature suggest lower school quality is associated with poor health outcomes.48 Higher quality schools as rated by parents have been associated with decreased adolescent risk-taking behavior even when characteristics of families are controlled. Quality of schools as defined in this study includes concepts such as safety of the school and how much teachers care for students, similar to concepts inherent to the measure of school connectedness. In addition, there is literature that suggests positive changes in attachment and commitment to school resulting from preventative interventions is consistently associated with positive changes in problem behaviors.19 Longitudinal data from subsequent years of this study will allow further examination of the relationship between school connectedness, HRQOL, and the impact of school-based services on these outcomes.

Most of the associations between HRQOL and measured variables identified in this high-risk urban population were not surprising. Prior studies have demonstrated health status and/or health access varying as a function of income or poverty, parent age, parental employment, parental education, child gender, and insurance status.4,14, 38,41,49,50 Although there was a significant association between HRQOL and parental employment, the same relationship was not demonstrated for parental educational status. However, the mean scores for HRQOL did increase with each higher level of parental education. Although these measures are often used interchangeably as markers for socioeconomic status, educational level may impact parental health-seeking behaviors and access to care to a greater degree than it does on actual measures of health status.51 Limited variability in educational status within our sample (only 3.4% who completed college) may also have contributed to inability to detect a significant difference.

More difficult to explain is the finding that children with private insurance had poorer HRQOL than children with public forms of insurance, including Medicaid, Medicare, and the Children’s Health Insurance Program. The geographic area in which the study was conducted is not highly penetrated by Medicaid managed care. This minimizes potential misclassification of subjects, ie, the likelihood that families may have erroneously responded that their "public" insurance was "private." Welfare reform that requires return to work may be placing these families into jobs with substandard health insurance for children. It may be possible that the private insurance these families have offered through their employment may have limited coverage for child and/or preventative health that limits the use of this as an indicator of health access in the traditional sense. It is also plausible that the types of jobs in which parents work limit parents’ abilities to miss work for health care for children, making the type of employment more important than the insurance provided through that employment.41

The relationship between HRQOL and child grade has not been previously described for this age group. Studies in adolescents have demonstrated that HRQOL decreases, not increases, with age.14 It is possible that children in the 2nd grade, many of whom were 7, who were administered the 8- to 12-year-old tool by the project staff did not understand the differences in meanings of the responses to the questions. When patterns of response were examined for individual questions, the full range of responses was used by all grades, suggesting this was not the case. It is also plausible that as children get older, they may respond to questions in more socially acceptable, but inaccurate ways. It may be more difficult as one gets older to admit to feeling angry, sad, or unsafe.

In bivariate comparisons, children who lived with non-birth parents reported better total, physical, and psychosocial health. However, in multivariable analysis, living with a non-birth parent was an independent predictor for self-reported total and physical health. This may contradict logical thinking that children would fare better if living with their biological parents. In this high-risk urban population, living with a non-birth parent may be a marker for having increased social or family support in situations where children cannot live with birth parents. In the Solomon52 study, children living in grandparent-headed households had less behavioral health problems and better health vulnerability scores when compared with households with 1 or 2 biological parents.

There has been significant debate in the literature regarding parent proxy report versus child self-report of HRQOL.25,53–56 Many advocate for usage of both measures when possible, as children are more likely to accurately represent internal measures of health than parents, but parents’ perception of health is often what leads to health-seeking behavior or health care usage. There is also concern that parental perception of the parent’s own health may alter the parent’s perception of the child’s health.31,38,57,58 In our study, parent’s perception of the child’s overall health was only correlated with the physical subscale of the child’s self-report of HRQOL and not with the total or psychosocial scores. This may be related to the lack of sensitivity of a single-question measure for assessment of parental perception of health status, but is more likely to reflect the ability of the parent to "observe" the child for response to physical questions versus those that address emotional and social functions.30,59 We also found no relationship between parental health status as measured by the mental summary score or the physical summary score of the SF-12, and the child’s perception of HRQOL. Addition of the parent-proxy report of the PedsQL to subsequent years of the longitudinal study should provide further insight into this issue for urban children.

This study also provides insight into some of the methodological issues of using HRQOL tools in urban populations. Literacy was certainly an issue with the child subjects in this urban study. The need to interviewer-administer the tool to younger children in this study suggests that studies to compare modes of administration may need to be performed specifically with this population to make sure there are no response biases that occur with alternate methods of administration. Personal communications during the project with the tool developer indicated that comparisons of mode of administration with subjects from varied socioeconomic backgrounds did not yield significantly different response patterns (J. Varni, personal communication).

There are a number of limitations of the data presented. Foremost, the study is observational in nature; therefore, the data only support an association between measured variables and HRQOL, not causality. For example, it is equally plausible that children that perceive themselves in better health are more connected to their school as it is that children that are more attached to their schools perceive themselves in better health. Variables that were not measured or controlled for items such as family structure, parental involvement in school, and maternal/caretaker depression could be responsible for the association between school connectedness and a child’s perception of their health. In addition, caution needs to be exercised when generalizing the findings to other urban populations of children. This study was conducted in 1 urban city with children from 6 elementary schools. The schools that participated in the study are similar to the majority of the other inner-city schools in the district with respect to percentage of children in federal school lunch program, percentage of minorities, and mobility of students. The number of eligible parents and children we were able to contact was lower than desired; therefore, the children and parents surveyed may not accurately represent the urban youth in this community. Specifically, the use of telephone surveys may have resulted in a selection bias for families with less troubled or less chaotic households. Given nonrespondents had higher rates of mobility and school absence, it is plausible that the children we did not reach would have even lower scores on self-report of HRQOL and school connectedness.

Because our sample of children differed in race-ethnic composition and had mean HRQOL scores significantly below those of the sample used for reliability and validity testing of the PedsQL,44 we performed analyses for range of measurement (ceiling, floor effects), internal consistency, and factor structure. Results of these analyses support the use of the tool in this population and are available from the primary author on request. We also provide some data on the reliability and validity of the school connectedness tool in elementary-aged children to address concerns regarding lack of previous reliability and validity testing in this age group. Although further testing on this age group needs to be conducted, our data suggest the use of this tool in the age group is appropriate.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
This study provides some of the first data available regarding HRQOL of urban children. The low scores measured for children’s self-reported HRQOL suggest that urban children need access to programs and services that address their mental and physical health needs. Although many of the predictors of HRQOL identified by this study are not easily changed, school connectedness is a potentially modifiable factor. Health and educational programs and school health services provided through school-linked and school-based health centers or programs may be 1 mechanism to improve a child’s attachment to their school with concurrent reduction in risk-taking behaviors.


   ACKNOWLEDGMENTS
 
This study was funded by the Health Foundation of Greater Cincinnati.

We thank the project staff from the Cincinnati Health Department and the Children’s Hospital Medical Center, as well as the children, parents, teachers, and other Cincinnati Public School staff, whose participation and contributions have made this project a success.


   FOOTNOTES
 
Received for publication Mar 25, 2002; Accepted Nov 21, 2002.

Reprint requests to (M.E.M.) Assistant Professor of Pediatrics, Division of General and Community Pediatrics, Children’s Hospital Medical Center, 3333 Burnet Ave, Cincinnati, Ohio 45229-3039. E-mail: Mona.mansour{at}chmcc.org


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 

  1. Flores G, Bauchner H, Feinstein AR, Nguyen US. The impact of ethnicity, family income, and parental education on children’s health and use of health services. Am J Public Health.1999; 89 :1066 –1071[Abstract/Free Full Text]
  2. Lieu TA, Newacheck PW, McManus MA. Race, ethnicity, and access to ambulatory care among US adolescents. Am J Public Health.1993; 83 :960 –965[Abstract/Free Full Text]
  3. St Peter RF, Newacheck PW, Halfon N. Access to care for poor children: separate and unequal? JAMA.1992; 267 :2760 –2764[Abstract/Free Full Text]
  4. Newacheck PW, Hughes DC, Stoddard JJ. Children’s access to primary care: differences by race, income, and insurance status. Pediatrics.1996; 97 :26 –32[Abstract/Free Full Text]
  5. Ronsaville DS, Hakim RB. Well child care in the United States: racial differences in compliance with guidelines. Am J Public Health.2000; 90 :1436 –1443[Abstract/Free Full Text]
  6. Shi L, Starfield B. Primary care, income inequality, and self-rated health in the United States: a mixed-level analysis. Int J Health Serv.2000; 30 :541 –555[CrossRef][Web of Science][Medline]
  7. Collier J, MacKinlay D, Phillips D. Norm values for the generic children’s quality of life measure (GCQ) from a large school-based sample. Qual Life Res.2000; 9 :617 –623[CrossRef][Web of Science][Medline]
  8. Woolf SH, Rothemich SF, Johnson RE, Marsland DW. The functional status of inner-city primary care patients: diminished function in a family practice population and its potential determinants. J Fam Pract.1998; 47 :312 –315[Web of Science][Medline]
  9. Resnick MD, Harris LJ, Blum RW. The impact of caring and connectedness on adolescent health and well-being. J Paediatr Child Health.1993; 29(suppl 1) :S3 –S9
  10. Resnick MD, Bearman PS, Blum RW, et al. Protecting adolescents from harm: findings from the national longitudinal study on adolescent health. JAMA.1997; 278 :823 –832[Abstract/Free Full Text]
  11. Wade TJ, Pevalin DJ, Vingilis E. Revisiting student self-rated physical health. J Adolesc.2000; 23 :785 –791[CrossRef][Web of Science][Medline]
  12. Wade TJ, Pevalin DJ, Brannigan A. The clustering of severe behavioral, health and educational deficits in Canadian children: preliminary evidence from the National Longitudinal Survey of Children and Youth. Can J Public Health.1999; 90 :253 –259[Web of Science][Medline]
  13. Vingilis E, Wade TJ, Adlaf E. What factors predict student self-rated physical health? J Adolesc.1998; 21 :83 –97[CrossRef][Web of Science][Medline]
  14. Wade TJ, Vingilis E. The development of self-rated health during adolescence: an exploration of inter- and intra-cohort effects. Can J Public Health.1999; 90 :90 –94[Web of Science][Medline]
  15. Hawkins JD, Doueck HJ, Lishner DM. Changing teaching practices in mainstream classrooms to improve bonding and behavior of low achievers. Am Educ Res J.1988; 25 :31 –50[Abstract/Free Full Text]
  16. Pentz MA, Dwyer JH, MacKinnon DP, et al. A multicommunity trial for primary prevention of adolescent drug abuse. JAMA.1989; 262 :3259 –3266
  17. Tobler NS. Meta-analysis of 143 adolescent drug prevention programs: quantitative outcome results of program participants compared to a control or comparison group. J Drug Issues.1986; 16 :537 –567[Web of Science]
  18. McBride C, Curry SJ, Cheadle A, et al. School-level application of a social bonding model to adolescent risk-taking behavior. J School Health.1995; 65 :63 –68[Web of Science][Medline]
  19. Najaka SS, Gottfredson DC, Wilson DB. A meta-analytic inquiry into the relationship between selected risk factors and problem behavior. Prev Sci.2001; 2 :257 –271[CrossRef][Medline]
  20. Singh G, Athreya BH, Fries JF, Goldsmith DP. Measurement of health status in children with juvenile rheumatoid arthritis. Arthritis Rheum.1994; 37 :1761 –1769[Web of Science][Medline]
  21. Seid M, Varni JW, Rode CA, Katz ER. The pediatric cancer quality of life inventory: a modular approach to measuring health-related quality of life in children with cancer. Int J Cancer Suppl.1999; 12 :71 –76[CrossRef][Medline]
  22. Varni JW, Seid M, Rode CA. The PedsQL: measurement model for the pediatric quality of life inventory. Med Care.1999; 37 :126 –139[CrossRef][Web of Science][Medline]
  23. Juniper EF, Guyatt GH, Feeny DH, Ferrie PJ, Griffith LE, Townsend M. Measuring quality of life in children with asthma. Qual Life Res.1996; 5 :35 –46[CrossRef][Web of Science][Medline]
  24. Guyatt GH, Juniper EF, Griffith LE, Feeny DH, Ferrie PJ. Children and adult perceptions of childhood asthma. Pediatrics.1997; 99 :165 –168[Abstract/Free Full Text]
  25. Parsons SK, Barlow SE, Levy SL, Supran SE, Kalpan SH. Health-related quality of life in pediatric bone marrow transplant survivors: according to whom? Int J Cancer Suppl.1999; 12 :46 –51[CrossRef][Medline]
  26. Vogels T, Verrips GHW, Verloove-Vanhorick SP, et al. Measuring health-related quality of life in children: the development of the TACQOL parent form. Qual Life Res.1998; 7 :457 –465[CrossRef][Web of Science][Medline]
  27. Bender BG. Measurement of quality of life in pediatric asthma clinical trials. Ann Allergy Asthma Immunol.1996; 77 :438 –447[Web of Science][Medline]
  28. Guyatt GH, Feeny DH, Patrick DL. Measuring health-related quality of life. Ann Intern Med.1993; 118 :622 –629[Abstract/Free Full Text]
  29. Eiser C. Children’s quality of life measures. Arch Dis Child.1997; 77 :350 –354[Free Full Text]
  30. Pal DK. Quality of life assessment in children: a review of conceptual and methodological issues in multidimensional health status measures. J Epidemiol Community Health.1996; 50 :391 –396[Abstract/Free Full Text]
  31. Starfield B, Bergner M, Ensminger M, et al. Adolescent health status measurement: development of the child health and illness profile. Pediatrics.1993; 91 :430 –435[Abstract/Free Full Text]
  32. Starfield B, Forrest CB, Ryan SA, Riley AW, Ensminger ME, Green BF. Health status of well vs ill adolescents. Arch Pediatr Adolesc Med.1996; 150 :1249 –1256[Abstract/Free Full Text]
  33. Seid M, Varni JW, Kurtin PS. Measuring quality of care for vulnerable children: challenges and conceptualization of a pediatric outcome measure of quality. Am J Med Qual.2000; 15 :182 –188[Abstract/Free Full Text]
  34. Welligent: Intelligent Student Health Management. The Center for Pediatric Research, Norfolk, VA: The Children’s Hospital of the King’s Daughters and Eastern Virginia Medical School; 1999
  35. Ware JE, Kosinski M, Keller SD. SF-12: How to Score the SF-12 Physical and Mental Health Summary Scales. 2nd Ed. Boston, MA: The Health Institute, New England Medical Center; 1995
  36. Resnick MD, Harris LJ, Rosenwinkel K, Blum RW. Technical Report on the Sampling Design, Methodology, and Psychometric Properties of the Adolescent Health Survey. Minneapolis, MN: University of Minnesota Adolescent Health Program; 1991
  37. Kahn RS, Zuckerman B, Bauchner H, Charles JH, Wise PH. Women’s health after pregnancy and child outcomes at age 3 years: a prospective cohort study. Am J Pub Health.2002; 92 :1312 –1318[Abstract/Free Full Text]
  38. McGauhey PJ, Starfield B. Child health and the social environment of white and black children. Soc Sci Med.1993; 36 :867 –874
  39. McMunn AM, Nazroo JY, Marmot MG, Boreham R, Goodman R. Children’s emotional and behavioural well-being and the family environment: findings from the Health Survey for England. Soc Sci Med.2001; 53 :423 –440
  40. Weinick RE, Monheit AC. Children’s health insurance coverage and family structure, 1977–1996. Med Care Res Rev.1999; 56 :55 –73[Abstract/Free Full Text]
  41. Guendelman S, Wyn R, Tsai YW. Children of working low-income families in California: does parental work benefit children’s insurance status, access, and utilization of primary health care? Health Serv Res.2000; 35 :417 –441[Web of Science][Medline]
  42. Weitzman M, Walker DK, Gortmaker S. Chronic illness, psychosocial problems, and school absences: results of a survey of one county. Clin Pediatr (Phila).1986; 25 :137 –141
  43. Stein RE, Jessop DJ. Relationship between health status and psychological adjustment among children with chronic conditions. Pediatrics.1984; 73 :169 –174[Abstract/Free Full Text]
  44. Varni JW, Seid M, Kurtin PS. The PedsQL: reliability and validity of the Pediatric Quality of Life Inventory version 4.0 generic core scales in healthy and patient populations. Med Care.2001; 39 :800 –812[CrossRef][Web of Science][Medline]
  45. Stein RE, Westbrook LE, Silver EJ. Comparison of adjustment of school-age children with and without chronic conditions: results from community-based samples. J Dev Behav Pediatr.1998; 19 :267 –272[Web of Science][Medline]
  46. Forrest CB, Starfield B, Riley AW, Kang M. The impact of asthma on the health status of adolescents. Pediatrics.1997; 99(2) . Available at www.pediatrics.org//cgi/content/full/99/2/e1
  47. Bonny AE, Britto MT, Klostermann BK, Hornung RW, Slap GB. School disconnectedness: identifying adolescents at risk. Pediatrics.2000; 106 :1017 –1021[Abstract/Free Full Text]
  48. Kowaleski-Jones L. Staying out of trouble: community resources and problem behavior among high-risk adolescents. J Marriage Fam.2000; 62 :449 –464[CrossRef][Web of Science]
  49. Vital and Health Statistics—Access to Health Care Part 1: Children, US From the Centers for Disease Control and Prevention/National Center for Health Statistics. July 1997, Series 10, No. 196
  50. Shi L, Starfield B, Kennedy B, Kawachi I. Income inequality, primary care, and health indicators. J Fam Pract.1999; 48 :275 –284[Web of Science][Medline]
  51. Desai S, Alva S. Maternal education and child health: is there a strong causal relationship? Demography.1998; 35 :71 –81[Web of Science][Medline]
  52. Solomon JC, Marx J. "To grandmother’s house we go": health and school adjustment of children raised solely by grandparents. Gerontologist.1995; 35 :386 –394[Abstract]
  53. Levi RB, Drotar D. Health-related quality of life in childhood cancer: discrepancy in parent-child reports. Int J Cancer Suppl.1999; 12 :58 –64[CrossRef][Medline]
  54. Theunissen NCM, Vogels TGC, Koopman HM. The proxy problem: child report versus parent report in health-related quality of life research. Qual Life Res.1998; 7 :387 –397[CrossRef][Web of Science][Medline]
  55. Sweeting H, West P. Health at age 11: reports from schoolchildren and their parents. Arch Dis Child.1998; 78 :427 –434[Abstract/Free Full Text]
  56. Waters E, Doyle J, Wolfe R, Wright M, Wake M, Salmon L. Influence of parental gender and self-reported health and illness on parent-reported child health. Pediatrics.2000; 106 :1422 –1428[Abstract/Free Full Text]
  57. Renouf AG, Kovacs M. Concordance between mothers’ reports and children’s self-reports of depressive symptoms: a longitudinal study. J Am Acad Child Adolesc Psychiatry.1994; 33 :208 –216[Web of Science][Medline]
  58. Perrin EC, Shapiro E. Health locus of control beliefs of healthy children, children with a chronic physical illness, and their mothers. J Pediatr. 185:107:627 –633
  59. Pantell RH, Lewis CC. Measuring the impact of medical care on children. J Chronic Dis.1987; 40(suppl 1) :99S –115S
PEDIATRICS (ISSN 1098-4275). ©2003 by the American Academy of Pediatrics

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Facebook Facebook   Add to Reddit Reddit   Add to Technorati Technorati   Add to Twitter Twitter    What's this?


This article has been cited by other articles:


Home page
 PediatricsHome page
C. Arnaud, M. White-Koning, S. I. Michelsen, J. Parkes, K. Parkinson, U. Thyen, E. Beckung, H. O. Dickinson, J. Fauconnier, M. Marcelli, et al.
Parent-Reported Quality of Life of Children With Cerebral Palsy in Europe
Pediatrics, January 1, 2008; 121(1): 54 - 64.
[Abstract] [Full Text] [PDF]

Home page
 PediatricsHome page
A. E. Simon, K. S. Chan, and C. B. Forrest
Assessment of Children's Health-Related Quality of Life in the United States With a Multidimensional Index
Pediatrics, January 1, 2008; 121(1): e118 - e126.
[Abstract] [Full Text] [PDF]

Home page
 Arch Pediatr Adolesc MedHome page
P. H. Casey, K. L. Szeto, J. M. Robbins, J. E. Stuff, C. Connell, J. M. Gossett, and P. M. Simpson
Child Health-Related Quality of Life and Household Food Security
Arch Pediatr Adolesc Med, January 1, 2005; 159(1): 51 - 56.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Web of Science (14)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Mansour, M. E.
Right arrow Articles by DeWitt, T. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Mansour, M. E.
Right arrow Articles by DeWitt, T. G.
Related Collections
Right arrow Office Practice
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Facebook   Add to Reddit   Add to Technorati   Add to Twitter  
What's this?