OBJECTIVE: Growing numbers of children with severe chronic illnesses are surviving to adulthood. Little is known about what primary care physicians perceive as the resources for and barriers to providing primary care services for young adults who transfer care from pediatric to adult medicine practitioners. The objective of this study was to describe primary care physicians' resources for and barriers to caring for young adults with childhood-onset chronic diseases.
METHODS: We conducted a national mailed survey of general internists and pediatricians to assess their office and specialty resources, attitudes toward, and barriers faced in treating young adult patients with childhood-onset chronic diseases. We then analyzed how these factors were associated with overall perceived quality of chronic illness care delivery.
RESULTS: Overall response rate was 53% (1289 of 2434). Only half of general internists viewed themselves as readily able to provide for the primary care needs of young adults with childhood-onset chronic diseases. Half of the internists and 62% of pediatricians thought that it would be difficult for these young adults to find an adult-focused primary care provider. Both specialties reported lack of time and reimbursement as major barriers in providing primary care to transitioning patients. Good office systems for coordinating patient care and improved coordination with subspecialty resources were both associated with improved provider perception of providing high-quality chronic illness care.
CONCLUSIONS: General pediatricians and internists report multiple barriers to providing care for adults with childhood-onset chronic diseases. Improvements in office-based support seem to be most associated with perceived quality of care delivery.
- transitional care
- cystic fibrosis
- sickle cell disease
- young adults
- children with special health care needs
- primary care
WHAT'S KNOWN ON THIS SUBJECT:
Young adults with childhood-onset chronic diseases have difficulty with transitioning to adult health care providers.
WHAT THIS STUDY ADDS:
We describe specific barriers and facilitators that general internists and pediatricians face in providing primary care to young adults with childhood-onset chronic conditions.
The chronic disease and disability burden among children has increased significantly in the past 20 years.1 Between 18% and 31% of children and adolescents report at least 1 chronic condition.2,–,4 Notably, individuals with diseases such as cystic fibrosis (CF), complex congenital heart disease, and sickle cell disease (SCD), all of which in the past were often lethal in early childhood, now have life expectancies that reach 40 years and beyond.5,–,7 Traditionally, many of these patients continued to be treated as adults by pediatric generalists and subspecialists in pediatrics; however, in response to the growing number of adult survivors of serious childhood-onset chronic diseases, multiple medical societies argue that the pediatric health care setting is an inappropriate place for the care of these adult patients and advocate transitioning young adults to adult-centered care.8,–,11
Despite these recommendations, little is known about the barriers that adult-focused primary care providers face to providing for the primary care needs of young adults with childhood-onset diseases. Barriers to care that have been anecdotally reported include lack of training and lack of resources, but which factors actually play a role in the care for these young adults remain uncertain.12 For effectively transferring care to adult settings, it is important to identify barriers to and the extent to which these identified barriers affect high-quality primary care for these young adult patients. Moreover, we need to evaluate factors that may be remediable through policy or changes in resource allocation to improve chronic illness management for this growing population of young adults with chronic diseases.
Accordingly, we sought to evaluate primary care providers' attitudes toward, available resources for, and major barriers faced in treating young adults with childhood-onset chronic diseases. We hypothesized that higher disease-specific knowledge—generally found in pediatrics—and resources would facilitate a physician's ability to care for young adults with childhood-onset chronic diseases. We also thought that disease-specific barriers would be a greater obstacle to care delivery than general disease management barriers for general internists in treating young adults with childhood-onset chronic conditions. Disease-specific resources often are accompanied by improved overall resources and infrastructure to provide chronic illness care; therefore, we hypothesized that improved disease-specific resources would be associated with a higher perceived ability to deliver high-quality chronic illness care to all patients with chronic conditions.
Between October 2005 and March 2006, we conducted a national mailed survey of a random sample of general pediatricians and general internists identified through the American Medical Association (AMA) Masterfile. We sampled physicians who reported their primary profession as being a general internist or pediatrician and reported providing “direct patient care.” A total of 3000 surveys (1500 to pediatricians and 1500 to internists) were sent in the first mailing, together with a $5 cash incentive. We followed this with 2 additional survey mailings. We also telephoned the offices of persistent nonrespondents to verify the information included in the AMA Masterfile. Additional details of the study design are reported elsewhere.13 Approval was received through the University of Michigan Medical School Institutional Review Board for Human Subject Research.
Views on Primary Care Delivery
We measured views of physicians on the quality of primary care delivery to young adults with childhood-onset chronic diseases. We asked whether it was easy for young adults to find a general internist to take care of their primary care needs and whether it was easy for a pediatric and adult care provider to communicate on transfers of care. Respondents answered these questions on a 6-point Likert scale from “strongly disagree” to “strongly agree.” To capture agree versus disagree, we dichotomized the variable between “somewhat agree,” “agree,” and “strongly agree” versus “somewhat disagree,” “disagree,” and “strongly disagree.”
We measured a variety of practice attributes that are associated with care coordination and practice organization and considered important in primary care delivery for patients with chronic diseases.14,–,16 Specifically, we asked providers to give us their views about their practice processes and outpatient care for patients with chronic illnesses. Respondents rated their agreement with availability of each of these attributes in their own practices from “strongly disagree” to “strongly agree” on a 6-point Likert scale in the following questions: ′There is effective communication between subspecialists and me,“ ”My office structure is effective in managing the care coordination needs of my patients,“ ”I receive adequate reimbursement for the care I provide to patients,“ and, :All patients with chronic illness in my practice are assigned to 1 physician or clinical staff member who serves as the primary coordinator of medical care.”17 For our analyses, we dichotomized these variables to “agree” and “strongly agree” versus all other responses, on the basis of the distribution of answers. All dichotomizations described in this article were based on distribution, and sensitivity analyses were done to ensure appropriateness of dichotomization points.
Available Resources and Training
Along with views about general office structure, we asked physicians whether they had various specialty-based services that would be needed to care for young adults with complex chronic diseases.18 We asked about availability of specialized disease centers, mental health support, rehabilitation medicine, and social work, all of which have been advocated in the literature as important components of chronic illness care.9,10,14,15,19,–,25 We then investigated whether they had access to subspecialty referral and care coordination by asking whether these services were “easily available,” “available,” “not available,” or “unknown.” When coded as unknown, we coded this as “not available” to the provider. For the purposes of this article, we report whether the resource was either available or not available. We used CF and SCD centers as model centers because SCD and CF represent 2 serious chronic childhood-onset diseases that have had marked increases in survival.26 These 2 diseases require significant health care coordination and management and are relatively new diseases to adult health care providers. These centers often follow children through adulthood and provide many of the specialized services needed for the care of both adults and children.25 We then asked the primary care physicians whether these specialized centers were “unavailable,” “available but difficult to access,” or “easy to access.” Those who answered “do not know” to any resource question were considered not to have the resource. Physicians were also asked whether they needed additional training to provide for the care needs of this population of patients. Respondents answered these questions on a 6-point Likert scale from “strongly disagree” to “strongly agree.” The variable was then dichotomized between “agree” and “strongly agree” and all other responses.
We evaluated factors that may impede the primary care delivery to young adults with chronic childhood-onset diseases by asking, “To what extent do the factors listed below limit your ability to provide high-quality primary care for young adults with chronic childhood-onset disease?” The individual factors are listed in Table 1. These questions were answered on a 6-point Likert-type scale from “severe limitation” to “no limitation.” The variables were then dichotomized as being a severe or significant limitation versus not.
Views on Quality of Care
Finally, we evaluated whether the measured practice resources and barriers that affect young adults are associated with provider views of the overall quality of care delivery in their practice. To determine this, we created our dependent variable by asking whether the provider was able to provide high-quality care for chronically ill patients by asking respondents to rate their agreement with the statement, “I am able to provide high quality care to all my patients with chronic disease.” We dichotomized this variable to “agree” and “strongly agree” versus all other responses.
We compared characteristics and views among general internists and pediatricians of specific barriers and facilitators relating to how chronic illness care is delivered by using bivariate statistics. To examine how these factors were associated with quality of care delivery, we ran a multivariable logistic regression model with perceived quality of care as the dependent variable and independent variables of years in practice, gender, office type, coordination, subspecialty communication, and availability as well as the individual barriers and resources as described in detail already. Because nearly all physicians had access to a hematologist and a pulmonologist, we did not include those variables in the model. We report both unadjusted and adjusted odds ratios. All analyses were performed by using Stata 9.1 (Stata Corp, College Station, TX).
Response Rate and Respondent Characteristics
Of the 3000 surveys mailed, 421 physicians were found to be ineligible (nonworking address, died, retired, not practicing, or subspecialty with no primary care). We were unable to verify eligibility for 470 of the nonrespondents. Assuming that the proportion of eligible physicians was the same among those whom we were and were not able to contact,27 our final eligible sample size was 2434, 1289 of whom responded, leading to an overall response rate of 53%. Most questions had ∼1% to 4% missing responses, with 1.7% missing in our main dependent variables. All variation in the total sample sizes listed in the tables generated from the survey compared with the overall response rate generated from the AMA Masterfile is attributed to missing values for those particular questions.
Among internists, there was no difference in years in practice, age, or board certification between respondents and nonrespondents on the basis of the data provided by the AMA Masterfile (Table 2). Women had a higher response rate than men among internists (48% vs 41%; P = .02). Among pediatricians, respondents were slightly younger than nonrespondents (mean age: 49.1 vs 50.4 years; P < .01) and had fewer years in training (18 vs 21 years; P < .001). In comparing internists' and pediatricians' responses from self-reported answers from the administered survey, pediatricians had a higher response rate (n = 751 [58%]) than internists (n = 515 [42%]; P < .001; Table 3). General internists were more likely to be men and to be in solo practice than pediatricians. Among respondents, internists practiced in urban, non–inner-city–based practices as compared with pediatricians, whereas more pediatricians practiced in suburban-based practices in this survey. Otherwise, there were no differences between internists and pediatrician respondents in rates of board certification, years in practice, or age.
Views About Care Delivery and Transferring Young Adult Patients to Adult-centered Care
Only 38% of general pediatricians and 51% of general internists agreed to the statement, “It would be easy for young adult patients with a childhood-onset chronic disease to find a general internist who would be willing to care for their primary care needs” (Table 2). General pediatricians and internists were similar in their views in their ability to communicate with each other. A total of 57% of the general pediatricians reported that it was easy to communicate with an adult provider to transition their young adult patients to adult-based care, and 62% of internists reported that it was easy communicate with a pediatric provider about transitioning (Table 2).
Practice Structure and Resources
General internists and pediatricians varied little in their type of practice (Table 3). The majority of general pediatricians and internists believed that they did not receive enough reimbursement to care for young adults with chronic illnesses (Table 1). Only half of the providers found that their subspecialty communication and care coordination were sufficient to care for these patients' needs. More general internists assigned a single provider to a patient with chronic diseases than the pediatricians (71% internists, 54% pediatricians; P < .001; Table 1).
In evaluating available resources, general internists reported significantly less access to specialized disease centers such as CF and SCD than their general pediatrician colleagues (Table 1). Pain clinics, a resource that is important in the care of many adults with chronic diseases, were much more readily available for general internists than for pediatricians.
Barriers to Care
Significantly more general internists reported that their lack of training was a barrier to the care of these young adults with chronic diseases as compared with pediatricians (24% internists vs 12% pediatricians P < .001; Table 1). In general, a greater proportion of pediatricians reported being affected by a lack of resources in caring for young adults with childhood chronic diseases compared with general internists.
Factors Related to Perceived Quality of Care Delivery
In the multiple regression analyses, there were slight differences between internists and pediatricians in the relative importance of various practice features for quality of care delivery for patients with chronic diseases (Table 4). For both internists and pediatricians, having an office structure that is effective in care coordination had the greatest positive association on the perceived quality of care, and lack of subspecialty access and care coordination access had independently negative associations on the perceived quality of care for these patients. For pediatricians, subspecialty support was significantly associated with higher perception of quality of care delivery. For general internists, effective communication between subspecialists was associated with higher perceptions of the quality of chronic care delivery. Other potential barriers to chronic illness delivery, such as lack of training, were not associated with perceived quality of chronic illness delivery in respondents' practices.
Primary care, as opposed to a predominant reliance on subspecialty-based care, is essential to providing continuity of care and high-quality care to patients with chronic diseases.28,–,30 Generalist providers play a critical role in care coordination and continuity of care for patients with chronic diseases.31 It is concerning that the majority of both adult and pediatric primary care providers reported that these young adults with childhood-onset chronic diseases would have difficulty finding an adult-focused primary care provider. Previous reports in the pediatric literature suggested that pediatricians view their adult care provider counterparts as largely unwilling to care for their patients who age out of their pediatric services.32
High-quality chronic illness care depends on effective provider-level, practice-level, and system-level integration of services. For instance, many of the chronic disease models emphasize the need for improvements in communication and care coordination of patients with chronic illness.14,17,21,22,29 We found multiple provider-level and practice-level barriers in this study. As we had initially hypothesized, disease-specific factors would affect internists more than pediatricians. This was the case in regard to training. Surprising, however, only one quarter of general internists viewed lack of training as a major barrier in treating young adults with childhood-onset chronic diseases. Although we hypothesized otherwise, our study showed that clinicians perceived general system-level factors as far more significant barriers than individual-level training. Fortunately, the majority of the physicians believed that their office structure was conducive to managing chronic illnesses, although many still reported deficiencies and barriers in their ability to care for this population.
Consistent with our initial hypothesis, physicians perceived poorer general chronic illness quality of care delivery when physicians lacked care coordination or subspecialty support. Many of these barriers would be amenable to changes in practice-level improvements in resource allocation. For instance, system-based improvements such as the medical home14,17,22,33 and the chronic care model16,29,30 have been developed to improve health care delivery to patients with chronic illness. Ideally, these systems improve care coordination and decrease the time required by a physician to oversee coordination of care. The American Academy of Pediatrics has advocated for chronic illness care within a medical home model to improve transitions to youth and young adults with chronic illness. Time and reimbursement seem to be major barriers to clinicians. Initiating reimbursement strategies such as reimbursement for primary care services that are delivered within a medical home may help primary care providers who care for these youth and young adults with complex chronic illnesses. It would also be important to know how the views and specific barriers that were found in this study translate into the quality of actual care that is delivered to young adults with childhood-onset chronic diseases.
This is one of the largest nationwide surveys to assess general internists' and pediatricians' barriers and practice characteristics for treating young adults with chronic diseases of childhood origin. Several limitations should be noted. First, we examined opinions of general internists and pediatricians only and did not assess resources of family physicians, nurse practitioners, or subspecialists. Second, in our measure of office structure, we evaluated overall process and perceived effectiveness of care coordination and existence of a care coordinator, rather than many specific structural components (eg, information technology, ancillary services). Third, our estimate of level of perceived barriers could be biased if there were a systematic tendency to report more or fewer barriers among nonrespondents. Although it was not possible to compare all characteristics of respondents and nonrespondents, we saw no variation in outcome measures between responses to the initial survey and those from later survey rounds; however, respondents and nonrespondents did not differ in age, years in practice, and certification for internists, and our response rate of 53%, although low, compares favorably with other national surveys of physicians.34
Despite current recommendations that adults with childhood-onset chronic illnesses find adult-focused primary care physicians to help coordinate their care,11,35 most general internists and pediatricians believed that it would be difficult for a young adult to find an adult-focused provider who is able to care for their needs. Our study further found significant system-level and policy-level barriers to treating young adults with a childhood-onset chronic disease, such as poor mental health care support and reimbursement for chronic illness care. These general health care delivery factors are important across nearly all disease conditions in providing high-quality care for patients with chronic diseases; therefore, having systems in place for promoting high-quality chronic illness care for young adults with childhood-onset chronic diseases will likely benefit all patients with chronic diseases in a practice.
This work was funded by the Robert Wood Johnson Clinical Scholars Program and the Division of Pediatrics, University of California, San Francisco. Dr Okumura is funded through an Agency for Healthcare Research and Quality Career Development Award K08 HS017716. Dr Heisler is funded through a VA Health Services Research and Development Career Development Award.
- Accepted November 6, 2009.
- Address correspondence to Megumi J. Okumura, MD, University of California, San Francisco, 3333 California St, Suite 245, San Francisco, CA 94118. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
Funded by the National Institutes of Health (NIH).
- AMA =
- American Medical Association •
- CF =
- cystic fibrosis •
- SCD =
- sickle cell disease
- Lakdawalla DN,
- Bhattacharya J,
- Goldman DP
- Newacheck PW,
- Strickland B,
- Shonkoff JP,
- et al
- Elborn JS,
- Shale DJ,
- Britton JR
- 11.↵American Academy of Pediatric, American Academy of Family Physicians, American College of Physicians-American Society of Internal Medicine. A consensus statement on health care transitions for young adults with special health care needs. Pediatrics. 2002; 110(6 pt 2): 1304–1306
- Wagner EH
- Wagner EH,
- Austin BT,
- Davis C,
- Hindmarsh M,
- Schaefer J,
- Bonomi A
- 17.↵Medical Home Initiatives for Children With Special Needs Project Advisory Committee. American Academy of Pediatrics. The medical home. Pediatrics. 2002; 110(1 pt 1): 184–186
- Ziring PR,
- Brazdziunas D,
- Cooley WC,
- et al
- 22.↵American Academy of Pediatrics Medical Home Initiatives for Children With Special Needs Project Advisory Committee. Policy statement: organizational principles to guide and define the child health care system and/or improve the health of all children. Pediatrics. 2004; 113(5 suppl): 1545–1547
- Lotstein DS,
- McPherson M,
- Strickland B,
- Newacheck PW
- Frankel LR
- Rosen DS,
- Blum RW,
- Britto M,
- Sawyer SM,
- Siegel DM
- Copyright © 2010 by the American Academy of Pediatrics