Objective. To determine the factors associated with addressing the transition from pediatric to adult-oriented health care among US adolescents with special health care needs.
Methods. Data for 4332 adolescents, 14 to 17 years of age, from the 2000–2001 National Survey of Children With Special Health Care Needs were used. The adequacy of transition services was determined by parent self-report. Explanatory variables, including parental education, family poverty status, race/ethnicity, measures of the severity and complexity of conditions, health insurance status, having a personal doctor, and the quality of the parent's relationship with the adolescent's doctor, were entered into a regression model.
Results. Overall, 50.2% of parents reported that they had discussed transition issues with their adolescent's doctor and 16.4% had discussed and developed a plan for addressing those needs. In a multivariate regression analysis, correlates of the adequacy of transition services included older age, female gender, complexity of health care needs, and higher quality of the parent-doctor relationship.
Conclusions. Among adolescents with special health care needs, those who were older and those with more complicated needs were more likely to have addressed the transition from a pediatric to adult-oriented system of care. Furthermore, this analysis demonstrated a strong association between a high-quality parent-provider relationship and the extent to which transition issues were addressed. The importance of transition services for adolescents with less complex needs and the overall impact of health care transition services were not assessed in this study and remain important questions for future investigations.
An estimated 21% of US adolescents today have special health care needs (SHCN),1,2 as a result of improving survival rates for children with chronic illnesses and increasing rates of childhood disability. These youths are identified by the Maternal and Child Health Bureau as “those who have or are at increased risk for a chronic physical, developmental, behavioral, or emotional condition and who also require health and related services of a type or amount beyond that required by children generally.”3 Each year, 750000 of these adolescents cross the threshold into adulthood. Many adolescents with SHCN navigate their way into adulthood and adult-oriented medical care much like their healthy peers. For others, the added complexities of SHCN limit these developmental achievements.2,4–6 Therefore, the success of the child-oriented health care system has created a need to facilitate the transition from a child-focused to an adult-oriented system of care, to ensure uninterrupted, high-quality care. This need to facilitate health care transition (HCT) received national attention with the 1989 Surgeon General's report Growing Up and Getting Medical Care: Youth With Special Health Care Needs.7 Since then, other organizations have recognized the need to support HCT.8–10 The recently published consensus statement of the American Academy of Pediatrics, American Academy of Family Physicians, American College of Physicians, and American Society of Internal Medicine11 on HCT stated that all adolescents with SHCN should have a written HCT plan by the age of 14 years, aligning the policies of the endorsing organizations with the Individuals with Disabilities Education Act and the Maternal and Child Health Bureau goal for the year 2010, namely, “All youth with special health care needs will receive the services necessary to make transitions to all aspects of adult life, including adult health care, work, and independence.”12 To date, it seems that clinically based transition services are not generally available.13
The extent to which adolescents with SHCN receive HCT services and an understanding of the factors that influence the adequacy of such services are important gaps in the research. Approaches to understanding the determinants of unmet health care needs and health care service utilization may provide a useful framework for understanding HCT services. The health services utilization model14,15 is an empirically validated model for studying health care access and utilization for vulnerable populations. This framework has demonstrated utility in categorizing and describing the factors that limit or promote utilization of services for children with SHCN (CSHCN).16–18 Factors are categorized as predisposing (sociodemographic characteristics), enabling (personal and community resources), or needs (health status measures). For example, predisposing factors such as minority race and ethnicity have been identified as predicting lower use of mental health care services for youths with SHCN,18 as well as higher rates of foregone care17 and unmet health care needs16 among all adolescents. Problems generated by activity limitations predict increased use of health care services,18 as well as a greater number of unmet needs.19 Having a usual source of care is associated with fewer unmet needs.16 Although it is promoted as a best practice,20 little is known about the impact that a good relationship between parents and providers has on unmet needs generally or on HCT services specifically.
The purpose of this study was to delineate the opportunities for and barriers to addressing the transition from child-centered health care to adult-oriented health care for adolescents with SHCN. We based our approach on a theoretically and empirically validated model for understanding the determinants of unmet health care needs. Specifically, we sought to determine the adequacy of HCT services for US adolescents with SHCN, as well as to determine the sociodemographic (predisposing), condition-specific (needs), and health care system (enabling) factors associated with receiving such services.
Data Source and Sample
This study used data from the 2000–2001 National Survey of Children With Special Health Care Needs conducted by the National Center for Health Statistics (NCHS) for the Maternal and Child Health Bureau.21 This survey used a random-digit-dial method to generate a sample for making state- and nation-level estimates of the prevalence of SHCN among children up to the age of 18 years, as well as assessing their health care needs and service utilization. Between October 2000 and December 2001, 196888 households were screened for inclusion, until complete interview data were collected for ≥750 CSHCN from each of the 50 states and Washington, DC. Eligibility for inclusion was established with the CSHCN Screener,22 which operationalizes the Maternal and Child Health Bureau definition of CSHCN.3 Complete data were collected from an adult respondent for 38866 CSHCN <18 years of age. The response rate for the SHCN interview was 61.0%. Sample weights provided adjustment for family characteristics (number of children in the home, race/ethnicity, mother's education, income, and telephone status). Together, the sampling design and weighting adjustments provided a sample that is, as much as possible, nationally representative of CSHCN.23 Because current consensus recommends that transition planning begin by the age of 14 years,11 this analysis was restricted to data for the 4332 youths who were 14 to 17 years of age; it has no missing data on the constituent items of the dependent variable. Sample characteristics are presented in Table 1.
Variable Selection and Analytic Plan
Three questions explored the extent to which HCT had been addressed. The first question was asked of all respondents: “Have doctors or other providers talked about how health care needs might change when (he/she) becomes an adult?” The subsequent 2 questions were asked only of those who responded affirmatively to the first question: “Has a plan for addressing these changing needs been developed with the doctor or health care providers?” and “Have doctors or other health care providers discussed eventually seeing a doctor who treats adults?” A composite variable representing the adequacy of addressing HCT was created as the sum of the affirmative answers to these 3 questions. Subjects (n = 185) with missing responses to any of the 3 questions were excluded from all analyses. Scores for this variable ranged from 0 to 3. A negative response to the first stem question received a score of 0 and represented not addressing HCT with the health care provider. Affirmative responses to all 3 questions represented addressing HCT in the most comprehensive manner and received a score of 3. The distribution of this variable was skewed, with a median of 0 and a mean of 1.00. Responses to the 3 questions are presented in Table 2.
Explanatory variables were selected on the basis of the theoretical model and the empirically derived determinants of unmet health care needs and health service utilization. Predisposing factors included age, gender, race/ethnicity, poverty status, and maternal education. Race/ethnicity was defined by self-report and was categorized as white (non-Hispanic), black (non-Hispanic), Hispanic, or other. Self-report of Hispanic overrode a self-report of white, black, or other. Poverty status for each subject was categorized by the NCHS as a percentage of calculated poverty guidelines, accounting for the respondent's state of residence, household income, and household size, and was dichotomized as below versus at or above the poverty level. Maternal educational achievement was dichotomized as graduation from high school or less versus education beyond high school. Factors potentially affecting HCT services included measures of severity of impact and complexity of needs.24,25 The severity of impact was determined from the response to the question, “Overall, how would you rank the severity of the child's condition, from 0 (lowest) to 10 (highest)?” The complexity of health care needs was characterized as the sum of the number of services needed by the adolescent, from a list of 14 items (potential range: 0–14). Enabling factors included having had continuous insurance coverage during the past 12 months versus having been uninsured at any point during that time, identifying a single provider as a personal doctor, and reporting care that reflected a high-quality parent-provider interaction. The parent-provider interaction was assessed with 5 questions regarding the interaction (how often the provider spent enough time, listened carefully, was sensitive to family values, provided enough information, and made the parent feel like a partner in care), each with a possible response of never, sometimes, usually, or always. Responses to the 5 items were summed, with the scale oriented so that a higher value represented a higher quality of parent-provider interaction. A Cronbach's α of .86 demonstrated very good internal consistency among the 5 items.
Bivariate analyses were undertaken to examine the relationship between each of the predisposing needs and enabling factors and the dependent variable HCT services. Comparisons between the mean HCT scores within categories of the explanatory variables were made with t tests for the dichotomous variables and with correlation coefficients for the continuous explanatory variables. Linear regression analyses were used to determine the independent contribution that each explanatory variable made to the adequacy of HCT services. All analyses were completed with SAS version 8.02 software (SAS Institute, Cary, NC), with procedures that used Taylor series approximation to calculate variance estimates that accurately accounted for the complex sampling design. Sample weights supplied by the NCHS were applied, to allow the sample to be reflective of the US population. An α equal to .05 was used to determine statistical significance.
Overall, 50.23% of respondents reported having discussed with their health care provider how adolescent health care needs might change in adulthood. Only 30.08% had developed a plan for addressing those needs, and 29.93% had discussed seeing a doctor who treats adults. HCT had been addressed in a comprehensive manner, as indicated by an affirmative response to each of the 3 questions, for only a few adolescents (16.35%) (Table 2).
Bivariate analyses assessing the relationship between the theoretically derived explanatory variables and the adequacy of addressing HCT are presented in Table 3. Older age, white race (versus Hispanic), greater number of needs, and higher quality of parent-provider relationship were associated with significantly higher scores on the HCT scale.
Results from the linear regression models are presented in Table 4. In the regression model, the strongest predictors of the adequacy of HCT services were a high-quality parent-provider interaction, a greater number of needed services, female gender, and older age. Hispanic ethnicity, which was a significant predictor of HCT services in the bivariate analysis, was not a statistically significant predictor in the regression analysis.
Results from this analysis begin to elucidate the complex array of factors influencing services that facilitate the transition to adulthood for youths with SHCN. Predisposing, needs, and enabling factors each contribute to the adequacy of services received; however, 2 factors stand out as having significant impact, namely, condition complexity (greater number of health care services needed) and high-quality relationships with health care providers. Beyond providing insights into the complex transaction, results from this analysis provide evidence to support current activities promoting HCT services and raise questions for future research.
As we hypothesized, it seems that, overall, few youths and their families are receiving HCT services from their providers. One in 6 youths reported having discussed changing needs and had a plan to address them. Although the proportion of youths who received adequate services increased at the age of 17 years, the vast majority of youths on the cusp of adulthood had not yet discussed changing needs or developed a plan for addressing them. The National Survey of Children With Special Health Care Needs provides the first view of the state of HCT services for a nationally representative sample of US adolescents with special needs. Because this survey is a first assessment, there are no data for comparison. It may be reasonable to assume that, with increasing attention to the issue of HCT in the past decade, the number of youths who are receiving such services has been increasing. Follow-up studies to the National Survey of Children With Special Health Care Needs will be an important means of assessing progress in addressing HCT.
This analysis demonstrates that adolescents with more complex needs are more likely to address HCT issues. Through this analysis we are unable to answer the question of why this is so. What remains unknown is the necessity and utility of providing services for adolescents with fewer needs and less complex conditions. Do these youths need but not receive HCT services? How should these services be organized and who should be responsible for ensuring that they are provided?
The literature regarding the care of youths with SHCN emphasizes the importance of a family-centered approach to care, one that is based on a high-quality relationship between parents and providers. Donabedian26 described the importance of the interpersonal aspect of care as “the vehicle by which technical care is implemented and on which its success depends.” Garwick et al27 identified the need, among families with children with chronic conditions, for providers to adapt services to fit family needs. Findings from this analysis support an increased emphasis on high-quality parent-provider relationships. The Medical Home training project, a collaboration between the American Academy of Pediatrics, Family Voices, the federal Maternal and Child Health Bureau, the National Association of Children's Hospitals and Related Institutions, and Shriners Hospitals for Children, serves as a national model for teaching family-centered care. This effort brings training to physicians, families, and those who provide care to CSHCN. Medical student and resident training programs that emphasize the development of collaborative relationships with families through home and community site visits are innovative approaches likely to produce long-term rewards.28,29
We have described several correlates of parent reports of HCT services, on the basis of the model described by Aday and Awe14 and Anderson and Newman.15 Certainly other factors at the individual and health care delivery levels, which are not measured in this survey, contribute to this complex transaction. It may be that clinic-based support systems such as those that trigger and support screening and prevention efforts may play an important role. Furthermore, we know little about how the attitudes and beliefs of providers, families, and adolescents influence HCT.
Because this was an analysis of cross-sectional data, there are several limitations to the conclusions that can be drawn. First, as in all cross-sectional data analysis, regression analysis describes associations and does not demonstrate causation. However, our independent variable selection was based on a well-accepted theoretical model and the empirical literature, and our findings are consistent with that literature. These points add to the probability that our findings represent meaningful associations, and they reduce the likelihood of type I error. Second, the relatively low survey response rate prompts consideration of how the results would be different if the nearly 40% nonresponse rate could be reduced. It might be expected that the effect of nonresponse would be to bias the results toward over-representing those who are more connected to the health care system. Therefore, our estimate of the proportion of the population that had addressed HCT might be overstated. Third, our dependent variable was a composite developed from 3 questions that described the adequacy of HCT services on the basis of key elements from consensus editorials and program descriptions. To date, we can find no outcome studies on transition interventions in the published scientific literature to substantiate the value of HCT services. Furthermore, these 3 items relied on parent recall of conversations with the adolescent's health care provider. Geenen et al30 pointed out that there might be considerable discordance between provider and parent recall over interpretation of the content of clinical encounters regarding transition issues. Therefore, it might be meaningful to interpret these results as factors influencing parents' perceptions of the adequacy of transition services. Fourth, nearly 43% of the adolescents in this analysis had a general practitioner as their personal doctor and the content of transition discussions might be different for these adolescents than it is for adolescents with pediatric or child-oriented personal doctors or health care providers. In particular, it may be that adolescents who see family physicians do not need to discuss eventually seeing a doctor who treats adults, which may result in a lower score for the HCT variable. A separate regression analysis was conducted with a dependent variable that eliminated the question about discussing the need to change providers. The only change in the significance or direction among predictor variables, compared with those presented in this report, was that age no longer predicted HCT services significantly.
The American Academy of Pediatrics and the Maternal and Child Health Bureau have made improving the delivery of HCT services to youths with SHCN a priority. It seems that youths with more complex conditions are receiving HCT services more frequently than are youths with less complex conditions, although the overall delivery rate shows substantial room for improvement. Promoting high-quality parent-physician interactions may be one means to improve the delivery of HCT.
This work was completed while Dr Scal was a Fellow in the Division of Health Services Research and Policy, School of Public Health, University of Minnesota, funded by the Agency for Healthcare Research and Quality through a National Research Scientist Award and by a grant from the Viking's Children's Fund of the Department of Pediatrics, University of Minnesota.
- Accepted October 4, 2004.
- Address correspondence to Peter Scal, MD, MPH, Division of General Pediatrics and Adolescent Health, University of Minnesota, 200 Oak St SE, Suite 160, Minneapolis, MN 55455. E-mail:
No conflict of interest declared.
- ↵Newacheck PW, Strickland B, Shonkoff JP, et al. An epidemiologic profile of children with special health care needs. Pediatrics.1998;102 :117– 123
- ↵McPherson M, Arango P, Fox H, et al. A new definition of children with special health care needs. Pediatrics.1998;102 :137– 140
- ↵Scal P, Larson S, Ireland M, Blum R. Young adults with childhood onset disability making their way into adulthood. Pediatr Res.2003;53 (4 suppl)
- Fiorentino L, Datta D, Gentle S, et al. Transition from school to adult life for physically disabled young people. Arch Dis Child.1998;79 :306– 311
- ↵Gortmaker S, Perrin J, Weitzman M. An unexpected success story: transition to adulthood in youth with chronic physical health conditions. J Res Adolesc.1993;3 :317– 336
- ↵Koop C. Executive summary. In: McGrab P, ed. Growing Up and Getting Medical Care: Youth With Special Health Care Needs. Jekyll Island, GA: US Public Health Service; 1989
- ↵Maternal and Child Health Bureau. Moving On: Transition From Child-Centered to Adult Health Care for Youth With Disabilities. Washington, DC: Department of Health and Human Services, Health Resources and Services Administration, Maternal and Child Health Bureau; 1992
- American Academy of Pediatrics, Committee on Children With Disabilities, Committee on Adolescence. Transition of care provided for adolescents with special health care needs. Pediatrics.1996;98 :1203– 1206
- ↵Maternal and Child Health Bureau. Achieving Success for All Children and Youth With Special Health Care Needs: A 10-Year Action Plan to Accompany Healthy People 2010. Rockville, MD: Maternal and Child Health Bureau; 2001
- ↵Aday L, Awe W. Health services utilization models. In: Gochman D, ed. Handbook of Health Behavior Research. New York, NY: Plenum Press; 1997:153– 172
- ↵Weller WE, Minkovitz CS, Anderson GF. Utilization of medical and health-related services among school-age children and adolescents with special health care needs (1994 National Health Interview Survey on Disability baseline data). Pediatrics.2003;112 :593– 603
- ↵Newacheck PW, McManus M, Fox HB, Hung Y-Y, Halfon N. Access to health care for children with special health care needs. Pediatrics.2000;105 :760– 766
- ↵American Academy of Pediatrics, Medical Home Initiatives for Chil-dren With Special Needs Project Advisory Committee. The medical home. Pediatrics.2002;110 :184– 186
- ↵Blumberg S, Olson L, Frankel M. Design and operation of the National Survey of Children With Special Health Care Needs. Vital Health Stat 1.2003;(41) :1– 136
- ↵Parent to Parent of Vermont. Teaching Family-Centered Care to Medical Students and Pediatric Residents at the University of Vermont. Williston, VT: Parent to Parent of Vermont. Available at: www.partoparvt.org/ffi.html. Accessed February 23, 2004
- Copyright © 2005 by the American Academy of Pediatrics