Preparing for adulthood: Health care transition counseling for youth with arthritis

Abstract

Objective

To determine the proportion of adolescents with arthritis who receive health care transition services and to compare the rates with those reported for adolescents with other special health care needs and adolescents with diabetes.

Methods

We used data from the 2005–2006 National Survey of Children with Special Health Care Needs. A parent/guardian identified youth ages 12–17 years with arthritis (n = 1,052), diabetes (n = 389), and special health care needs (n = 18,189). Four questions examined the extent to which providers discussed health care transition issues, including 1) transfer of care to adult providers, 2) health care needs of adults, 3) acquiring health insurance, and 4) encouraging self-care responsibility. Bivariate comparisons assessed the associations between sociodemographic characteristics and health care transition services, and multivariate regression models compared outcomes between conditions.

Results

Many adolescents with arthritis are being encouraged to assume self-care responsibilities (74.8%); fewer discussed how health needs will change in adulthood (52.1%), acquiring insurance (22.5%), or transferring care to a provider who sees adults (19.0%). These results are similar to youth with other special health care needs, but behind youth with diabetes.

Conclusion

Among this sample of US adolescents, many report discussions about health care needs and self-management, but few are addressing critical aspects of the transition to adult-oriented health care.

INTRODUCTION

Supporting youth during the transition from adolescence to adulthood and from child- to adult-focused health care is a national priority in the US (1, 2). Research describing the preferences for and barriers to getting health care transition services is emerging across many chronic conditions, including arthritis (3–6). Evaluations of clinical programming for youth with arthritis illustrate the positive impact that a planned and purposeful approach to health care transition can have for youth (7, 8).

Broadly defined, health care transition is the process by which youth with chronic conditions develop the skills and secure the resources to assure that their health care needs are met as they transition from adolescence to adulthood (9). The goal of health care transition is to assure optimal functioning in adulthood (10, 11). Many health care transition programs focus on the transfer of care between pediatric and adult providers; for instance, between a rheumatologist trained in pediatrics and one trained in adult medicine (3, 12). However, youth perceive their needs to be much broader in scope. For instance, Shaw et al (13) note that youth with arthritis in the UK identify transition needs as encompassing not just their physical health, but also their social, psychological, and vocational experiences. Canadian youth with arthritis report similarly broad health care transition needs (5). From a patient-centered perspective, a comprehensive health care transition program would address the transfer of care and health care self-management, as well as a broad array of issues that prepare youth for adult life.

The extent to which youth with special health care needs receive health care transition services is quite low (14, 15). The degree to which youth with arthritis actually receive health care transition services in the US has not been previously reported. The purpose of this study is to describe the extent to which youth with arthritis in the US receive health care transition services and to compare the rate with youth with all other special health care needs and with youth with diabetes. We use recently released nationally representative survey data in order to generate estimates of the current state of health care transition services for youth with arthritis.

MATERIALS AND METHODS

Data and sample.

Data came from the 2005–2006 National Survey of Children with Special Health Care Needs, which was sponsored by the Maternal and Child Health Bureau of the Health Resources and Services Administration and conducted by the National Center for Health Statistics (NCHS) at the Centers for Disease Control and Prevention. The purpose of the survey was to determine the prevalence of special health care needs, rates of service use, and needs (16). A similar survey was completed in 2001–2002, although that survey did not include questions to identify youth with arthritis (17). Data for the current analysis were collected between April 2005 and February 2007 via a random-digit-dial methodology and a sampling strategy to produce a sample suitable for national and state level estimates. A total of 192,083 households with a total of 363,183 children ages 0–18 years were screened for the presence of a special health care need using the Children with Special Health Care Needs Screener (18). For the 50,014 children determined to have a special health care need, full interviews were completed by a parent or guardian (n = 40,465). Data were therefore collected via a proxy; in some cases, the proxy respondent may not be the most knowledgeable respondent for that question. The overall response rate, which combined the response rates to the household screener, the child level screener, and the full child interview, was 54.2%.

Respondents for adolescents ages 12–17 years were asked questions pertinent to the transition to adulthood and adult-oriented health care. Adolescents with arthritis were identified as those with an affirmative answer to the question, “To the best of your knowledge, does (child's name) have arthritis or joint problems?” We excluded from this group those who also responded affirmatively to questions regarding cerebral palsy and muscular dystrophy to more accurately identify youth with arthritis. The analytic sample included 1,052 youth with arthritis. Samples for comparisons included adolescents identified as having diabetes and using insulin via similar questions (n = 389), as well as adolescents with special health care needs, but excluding arthritis (n = 17,137). We chose comparisons with youth with diabetes because of the similar needs for self-management skills and for the transfer of care to an adult-focused system.

Measures.

Four survey questions from the 2005–2006 National Survey of Children with Special Health Care Needs specifically addressed components of health care transition that were our primary outcomes of interest. These items examined whether a physician or health care provider had discussed each of the following: 1) seeing a provider who treats adult patients (restricted only to adolescents who reported seeing a provider whose practice is limited to children and adolescents only), 2) health care needs as the adolescent becomes an adult, and 3) how to keep or obtain health insurance in adulthood. The fourth question asked how often the adolescent is encouraged to take responsibility for health care needs, such as taking medication, understanding diagnosis, or following medical advice. The first 3 questions required a yes/no response. The fourth question, about self-care responsibility, allowed for responses of never, sometimes, often, and always. For ease of interpretation of results, we dichotomized the responses to often and always verses never and sometimes.

Explanatory variables included the sociodemographic characteristics of age (categorized as 12–14 and 15–17 years), sex, race (white only, African American only, and other as characterized by the NCHS), Hispanic ethnicity, poverty status (relative to the federal poverty level [FPL]), insurance status, and 3 measures of condition impact (number of doctor visits, number of school days missed in the past 12 months, and respondent-rated severity of difficulties caused by the adolescent's health problems). For the portion of the sample where the family poverty status was missing (n = 70), our analyses used the value imputed by the NCHS.

Statistical analysis.

We prepared descriptive statistics in the form of proportions for sample characteristics. We calculated the proportion of the sample reporting each outcome measure at each level of the sample characteristic measures and tested for differences using t-tests. Logistic regression was used to test whether youth with arthritis were more or less likely to report receiving each of the health care transition services than were youth with either all other special health care needs or youth with diabetes, controlling for potentially confounding sociodemographic factors. All analyses were conducted using STATA, version 9.1 for Windows (StataCorp, College Station, TX), sample weights were applied, and sample design variables were used in the calculation of variance estimates to adjust for the complex sampling design.

RESULTS

Sample characteristics for youth with arthritis ages 12–17 years (n = 1,052) are shown in Table 1. In this sample, the rate of arthritis was higher among the older subjects, nearly equally distributed among sexes, and a substantial proportion were African American or Hispanic. Over half of the sample live below 200% of the FPL and >45% were covered at least in part by public health insurance. Nearly 1 in 5 had >15 doctor visits in the past year and >1 in 4 had missed >15 days of school.

Table 1. Sample characteristics of youth with arthritis (n = 1,052)*

	%
* Data source: Centers for Disease Control and Prevention, National Center for Health Statistics, and National Survey of Children with Special Health Care Needs, 2005–2006. FPL = federal poverty level.
Age, years
12–14	43.1
15–17	56.9
Sex
Male	47.4
Female	52.4
Race/ethnicity
White	67.2
African American	19.2
Other	13.7
Hispanic
Yes	7.7
No	92.0
Poverty status, % of FPL
≤200	54.8
>200–300	14.9
>300–400	11.0
>400	19.3
Insurance status
Private	46.8
Public	37.6
Both	8.2
Other	1.5
Uninsured	5.9
Doctor visits in past 12 months
0–3	28.1
4–6	25.5
7–15	27.3
>15	19.2
School days missed in past 12 months
0–2	25.8
3–6	25.3
7–15	22.9
>15	26.0
Severity of difficulties caused by condition
Mild	29.0
Moderate	48.8
Severe	22.2

Rates of transition-related counseling are reported (Table 2) for the sample of youth with arthritis overall, as well as among this group by sociodemographic characteristics and measures of condition impact. Overall rates are also reported for youth with other special health care needs (n = 18,198), as well as for youth with diabetes (n = 389). Nearly 75% of youth with arthritis had usually or always been encouraged to take on self-care responsibility, 52.1% had discussed changing health care needs in adulthood, 22.5% had discussed acquiring insurance in adulthood, and 19.0% had discussed transferring care to an adult-focused provider. For only a few of the sociodemographic factors or condition impact measures, there was a statistically different rate of reporting each of the health care transition services. For example, the rate of reporting counseling regarding seeing providers who treat adults was 9.3 percentage points higher for older youth. The rate of reporting that providers encouraged self-care responsibility among youth living in households with income >400% of poverty was 17.3 percentage points higher than youth living at <200% of poverty.

Table 2. Transition-related counseling for youth with arthritis*

	Discussed seeing providers who treat adults	Discussed health care needs of adults	Discussed acquiring insurance as an adult	Providers encourage self-care responsibility, usually or always
* Values are the percentage. Data source: Centers for Disease Control and Prevention, National Center for Health Statistics, and National Survey of Children with Special Health Care Needs, 2005–2006. FPL = federal poverty level. † Comparisons made between value and the first row/referent value for that variable are significant at P ≤ 0.05. ‡ Sample size is too small for reliable estimate (standard error of the estimate is >30% of point estimate).
Youth with arthritis	19.0	52.1	22.5	74.8
Age, years
12–14	14.2	49.3	23.3	75.7
15–17	23.5†	54.2	21.8	74.1
Sex
Male	20.6	51.7	19.1	72.2
Female	17.6	52.3	25.4	77.1
Race/ethnicity
White	17.7	52.9	18.9	75.0
African American	20.8	57.8	34.3†	70.9
Other	21.9	40.3	23.2	78.8
Hispanic
No	18.8	53.6	22.2	75.9
Yes	21.6	33.7†	25.9	61.2
Poverty status, % of FPL
≤200	19.9	52.3	22.7	70.3
>200–300	18.9	45.5	22.4	70.5
>300–400	21.9	58.3	18.9	80.1
>400	15.5	53.0	23.7	87.6†
Insurance status
Private	17.5	54.0	23.0	82.0
Public	22.7	52.5	22.7	70.7†
Both	20.1	41.6	22.6	61.4†
Other‡
Uninsured	11.9	38.3	22.7	53.3†
Doctor visits
0–3	16.7	53.2	26.7	71.3
4–6	20.2	58.1	22.5	74.4
7–15	22.4	46.6	16.5	74.9
>15	16.6	52.3	23.9	80.0
School days missed
0–2	16.8	45.6	16.5	71.1
3–6	17.7	54.2	22.1	81.4
7–15	18.9	57.5	29.8†	71.2
>15	20.2	52.7	20.8	76.2
Severity
Mild	17.1	46.9	18.3	80.6
Moderate	21.5	55.8	26.6	77.3
Severe	17.9	49.8	17.2	60.4†
Youth with diabetes	29.0	66.5	18.8	96.6
Youth with special health care needs	19.4	46.2	21.3	78.0

Overall, the rates of transition-related counseling for youth with arthritis were similar to the rates reported for youth with other special health care needs. However, youth with arthritis are notably less likely than youth with diabetes to report 3 of the 4 transition services. These findings were confirmed in regression analyses, controlling for sociodemographic characteristics (Table 3). The odds of youth with arthritis discussing seeing providers who treat adults was 0.52 compared with that of youth with diabetes, but not statistically different from the odds of youth with all other special health care needs (odds ratio 0.88). The odds of youth with arthritis reporting that providers encourage self-care responsibility was 0.09 compared with that of youth with diabetes. These results are robust to various model specifications, in particular including various measures of severity and impact of condition.

Table 3. Odds of reporting each transition-related topic*

	Comparing youth with arthritis with all other youth with special health care needs, OR (95% CI)†	Comparing youth with arthritis with youth with diabetes, OR (95% CI)†
* Data source: Centers for Disease Control and Prevention, National Center for Health Statistics, and National Survey of Children with Special Health Care Needs, 2005–2006. Sample size for regression models comparing youth with arthritis with youth with other special health care needs ranged from 10,124–17,904, and for comparing with youth with diabetes, from 865–1,411. † Odds ratios (ORs) and 95% confidence intervals (95% CIs) are adjusted for age, sex, race, ethnicity, poverty status, and insurance coverage.
Discussed seeing providers who treat adults	0.88 (0.66–1.18)	0.52 (0.3–0.89)
Discussed health care needs of adults	1.21 (0.97–1.51)	0.55 (0.35–0.86)
Discussed acquiring insurance as an adult	1.04 (0.8–1.36)	1.18 (0.72–1.94)
Providers encourage self-care responsibility (reporting sometimes or always)	0.89 (0.7–1.13)	0.09 (0.04–0.21)

DISCUSSION

Among this US sample of youth with arthritis, nearly three-quarters were encouraged to take responsibility for their health care needs, and ∼50% discussed how their health care needs might change when they become an adult. However, only 1 of 5 received counseling around 2 critical health care transition issues: 1) the need for the transfer of care to adult-oriented physicians and 2) how to obtain health insurance as an adult. The state of transition services for youth with arthritis was similar to that of youth with special health care needs nationally, but lags behind what was reported for youth with diabetes.

Health care transition is a complex set of tasks that are embedded within a complex developmental period and a complex health care system. It is not surprising, then, that the development and evaluation of services to facilitate health care transition has been slow (19). However, with a systematic approach to understanding the needs of stakeholders, the British Society of Paediatric and Adolescent Rheumatology (online at: www.bspar.org.uk) has developed a comprehensive health care transition program for youth with arthritis that appears to have a positive impact (20). This program builds upon research examining the needs of key stakeholders in the transition process (youth, families, and providers). Although the program is delivered at the local level, it is supported by policy that sets a national standard for care around the issue of the preparation for adulthood.

We compare the experience of youth with arthritis with youth with other special health care needs in general, and specifically with youth with diabetes. Across conditions it is assumed that a knowledgeable, activated patient will lead to optimal disease control (21, 22). The basis for the non-categorical approach to identifying youth with special health care needs is based in part on the similarities in needs and the potential systems-wide approaches to meeting those needs (23, 24). Although we make a comparison between youth with arthritis and diabetes, in many ways the conditions are different. Youth with arthritis may experience pain and limitations in physical activities leading to reduced social engagement and social support. Furthermore, arthritis differs from diabetes in that for some youth with arthritis, their symptoms may not persist into adulthood and the perceived need for ongoing care may be reduced. Although it is unclear why there is a difference between youth with arthritis and diabetes in their health care transition experiences, the findings of this study present across 3 of the 4 measures of health care transition. There is extensive empiric literature regarding the management of diabetes in children and adolescents; however, there is considerably less related to arthritis. In the UK, there is a substantial discrepancy between diabetes and arthritis centers in their provision of transition programming (25, 26). Perhaps greater research efforts related to youth with arthritis may help improve the delivery of services.

There are several limitations to the conclusions that can be drawn from these analyses. First, we identify youth with arthritis using the parent report of arthritis or other joint problems. Although we exclude youth with cerebral palsy or muscular dystrophy, the subjects in our sample are likely to have a broad range of conditions with a broad range of severity. We controlled for severity of condition in our comparison with youth with diabetes and found no impact of severity on report of outcomes of interest. Unfortunately, it is not possible in this survey to be more specific about the particular diagnoses. Second, we rely on parent report of interactions with health providers. It may be that during adolescents' visits with their provider, they discuss these health care transition–related issues without the parents present. Therefore, the parent report may actually underreport the level of counseling. In addition, it is clear from other research that there may be a discrepancy between youth and parent report of pain, disability, and health-related quality of life (27). Third, the questions asked in this survey narrowly focus on the health care–related issues, which are only one component of what youth report that they want and need in transition services. Certainly the impact of arthritis on emotional well-being and social opportunity are other areas that need to be addressed in transition programming, given the profound need (28). It is unknown from this survey data about how these youth would describe the comprehensive nature and satisfaction with the services they are currently receiving.

In summary, there appears to be much room for improvement in the current health care system to assist youth with arthritis in achieving successful health care transition. More research is needed to understand youth's perspectives regarding their health care transition needs that should include items related specifically to health care transition, as well as how health care transition needs intersect with social, educational, and vocational concerns. From a clinical perspective, there seems to be an opportunity to provide education to health care providers serving youth with arthritis to encourage them to engage their patients in discussions about the transfer of care to adult providers and how to acquire insurance as an adult. Such exchanges would not necessarily require a great deal of time, but may have a large impact on the ease of health care transition for youth with arthritis.

AUTHOR CONTRIBUTIONS

Dr. Scal had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Scal, Garwick.

Acquisition of data. Scal.

Analysis and interpretation of data. Scal, Horvath, Garwick.

Manuscript preparation. Scal, Horvath, Garwick.

Statistical analysis. Scal.