|
 |
|
|||||||
|
|||||||||
© 2005 American Public Health Association
DOI: 10.2105/AJPH.2004.059089
RESEARCH AND PRACTICE |
The authors are with the Nemours Clinical Management Program, Orlando, FL.
Correspondence: Requests for reprints should be sent to Ian Nathanson, MD, Nemours Clinical Management Program, 496 South Delaney Avenue, Orlando, FL 32801 (e-mail: inathans{at}nemours.org).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Objective. The evolution of managed care has raised concerns about differential access to care for children with cystic fibrosis (CF). We tested the hypothesis that children with CF with managed care attended CF centers less frequently compared with children with non-managed care.
Methods. We conducted a prospective cohort study by telephone every 4 months to measure access to care at CF centers among 630 randomly selected patients aged 6–18 years from 15 US CF centers. We analyzed data with unconditional logistic regression and generalized estimating equations.
Results. Attendance at CF centers was significantly reduced among children with managed care (odds ratio [OR] = 0.74; 95% confidence interval [CI] = 0.57, 0.98; P=0.03) and among girls (OR=0.68; 95% CI=0.48, 0.97; P=0.04).
Conclusions. Children with CF with managed care attended CF centers significantly less frequently than those with non-managed care. These findings suggest that children with CF with managed care may not have equal access to experts in CF as children with non-managed care. Families should consider this when selecting their medical insurance plan. These findings may apply to other children with special health care needs.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Children with chronic illnesses rely on access to experts for many of their health care needs. These needs include frequent visits to specialists, expensive prescription medications, and other state-of-the-art interventions. Although traditional indemnity or fee-for-service plans allow patients unrestricted access to care, the evolution of managed care as a primary payer for services has caused patients and providers alike to voice concerns that this form of insurance may adversely affect access to care.
Despite well-established guidelines for immunizations, infants in an ambulatory setting who were enrolled in Medicaid managed care did not have improved rates of immunizations when compared with children who had Medicaid fee-for-service.1 Given that immunization schedules are well delineated, questions arise as to the effectiveness of managed care in meeting guidelines for care of patients with chronic illnesses. Konstan et al. reported that 40% of cystic fibrosis (CF) patients did not meet the recommended criteria for routine evaluation and monitoring and that managed care might have restricted access to specialty care (CF centers).2
Compared with healthy children, costs for medical care for children with chronic illnesses are higher. Newacheck et al. showed that children with chronic illnesses use hospital and outpatient services 10-fold more than well children.3 Ireys et al. reported that 8 chronic illnesses termed "tracer conditions" were responsible for disproportionate consumption of available health care dollars in a Medicaid program.4 Fowler and Anderson demonstrated that regardless of insurance type, costs of services for healthy children were predictable, whereas the costs of services for children with chronic illnesses were underestimated.5 Although it is logical that a greater economic burden on the payer, family, or both, may affect access to care among chronically ill children, other factors related to socioeconomic status (e.g., caregiver’s education and employment status) may also play a role.6–9
CF is an example of this type of chronic illness. CF is a common genetic disorder characterized by tenacious secretions in multiple organs, most notably the lung, pancreas, and liver, with symptoms typically appearing early in life.10 Formerly, children died at an early age, but adherence to treatment guidelines developed over the past 2 decades and the use of comprehensive centralized care have increased the median survival rate to 31.6 years.11,12
Approximately 85% of all CF patients in the United States receive care from multidisciplinary CF centers that are part of a network accredited by the Cystic Fibrosis Foundation. CF centers use guidelines to deliver care and submit individual patient data to a national patient registry. CF centers also collaborate in educational programs and basic and clinical research. Such a well-organized network provides a good sampling frame and makes CF an attractive model to evaluate the effect of health insurance coverage on access to health care among patients with a chronic disease.
We hypothesized that managed care limited use of CF centers by children with managed care compared with children with non-managed care. The outcome was whether the child attended the CF center in the past 4 months, as called for in the national guidelines.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
Study Design
This was a prospective cohort study designed to assess the impact of insurance coverage on the cost and use of services by children with CF. Trained interviewers collected data by telephone every 4 months over a 2.5-year period for a total of 7 interviews. In an effort to minimize recall bias, the study methods called for each interview to be completed within 120 to 150 days of the previous one. We collected data for this study from three sources: (1) the child’s primary care-giver; (2) representatives of insurance companies; and (3) the Cystic Fibrosis Foundation Patient Registry.
Cohort Selection
The sampling frame consisted of 11492 children aged between 6 and 18 years followed at more than 100 CF centers in the United States that provided care to children. To ensure geographic diversity, 16 CF centers that provided care to a minimum of 75 patients in this age range and experienced or anticipated approximately 50% managed care penetration by the end of the study were invited to participate. Fifteen CF centers agreed to participate. Children from these CF centers were eligible for selection if (1) they were aged 6–18 years as of January 1, 1998, (2) they were examined by a physician at a selected CF center in 1996, and (3) the Cystic Fibrosis Foundation Patient Registry contained data to determine disease severity. Children were not eligible for selection if they were (1) aged younger than 6 years because of the lack of reliable pulmonary function data, (2) pregnant, or (3) the younger affected sibling of a sibling pair. Patients aged 19 years or older were also excluded because a separate, unvalidated questionnaire directed to the patient rather than to the caregiver would have been required. In addition, patients reaching age 21 before the completion of the study would have faced a different set of issues, including changing eligibility for continuation of their insurance coverage or state-assisted programs.
Because disease severity impacts rates of medical service use, patients were stratified by disease severity as having mild disease or advanced disease. Although several scoring systems exist for the clinical setting,13–16 they all required some information that was unavailable to us. Because spirometry alone does not reflect all clinical characteristics of children with CF with advanced disease (e.g., children with nutritional failure), we convened a panel of CF center directors and representatives from the Cystic Fibrosis Foundation and devised a severity classification using critical data elements obtainable from the national patient registry. These data consisted of pulmonary function status, frequency of intravenous antibiotics, and nutritional status, which included weight, use of caloric supplements, and use of parenteral or enteral feedings. As shown in Table 1
, patients were classified as having "mild disease" if they had all of the characteristics listed. Patients were classified as having "advanced disease" if they had 1 or more criteria listed.
|
To do subgroup analyses by disease severity, we oversampled the advanced disease group. Thus, the initial sample size estimate of 735 consisted of 535 children with advanced disease and 200 with mild disease. The sample size was based on a power of 0.80, an expected 70% response rate at baseline, and a 25% dropout rate by the end of the follow-up period (including a 2% annual mortality rate).
Recruitment
For 8 of the CF centers, central recruitment was required. For these centers, the study coordinating center (SCC) randomly selected potential participants by their unique Cystic Fibrosis Foundation Patient Registry identification number. The identification numbers were forwarded to these CF centers for verification that those patients met eligibility criteria. Afterward, the SCC recruited the patients. The remaining 7 CF centers recruited locally. For these centers, the SCC provided a randomly selected list of eligible identification numbers to the local CF center director, who was responsible for recruiting participants and obtaining informed consent. To avoid recruitment bias, the CF center directors were only permitted to enroll patients on the list provided by the SCC. In both central and local recruitment, all selected identification numbers were accounted for by the SCC and placed in 1 of the following categories: (1) participant; (2) refusal; or (3) unable to contact (after 4 attempts).
Data Collection–Caregiver Survey Instrument
We used a survey based on a previously pilot-tested instrument17 to assess the effect of insurance on access to care rendered by CF centers. At entry into the study, information was collected about the previous 12 months. The 6 follow-up surveys inquired about the 4-month intervals between surveys. We collected the following data elements to evaluate insurance status and use of services: health insurance (policy benefits, co-payments, deductibles, and out-of-pocket health insurance premium costs); perceived barriers to care (receipt of appropriate health care, including distance to a usual source of medical care, transportation difficulties, financial barriers, distance to a CF center, knowledge of available services, and limitations of insurance coverage); use of health care services (use of health care services in the past year and in the past 4 months, including emergency room visits, inpatient hospital stays, use of a physician, and other outpatient services); use of other services (use and payment for home health care, social services, child care, respite care, mental health services, transportation services, medical supplies, respiratory therapy, and caregiver training); use of medications, supplements, and other therapies (use of prescription drugs, enteral and parenteral feedings, home oxygen, surgical consultation, and other therapies); health status characteristics of the child (child’s age at diagnosis, comorbid conditions, and duration of CF); sociodemographic characteristics of the parent or care-giver (education, marital status, relationship to CF child, household income, burden of care, and employment status); and sociodemographic characteristics of the child (age, gender, race, and education).
Definition of Managed Care
During their interviews, caregivers were asked the following questions about the child with CF: (1) Do you need to go to your primary care physician first for all medical care? (2) Can you choose to go to a specialist without a referral?
Children were classified as having managed care if the caregiver answered yes to the first question and no to the second question. Otherwise, children were classified as having non-managed care. Because 49 care-givers did not answer either or both questions, their children were not classified. This narrow definition minimizes misclassification of managed care and is consistent with other studies that use self-report health insurance data.18–20 However, some plans not meeting our definition still had restrictive characteristics, yet were included in non-managed care. For example, we found that 73.9% of care-givers of children in non-managed care reported that they paid less for medical care if they used physicians on a list provided by their insurance companies.
Analytical Strategy
We evaluated differential attendance at CF centers at entry into the study with a fitted unconditional logistic regression model. A longitudinal data model was fitted for subsequent interviews using generalized estimating equations with an exchangeable correlation matrix to predict the probability of using routine care at the CF center in the interval 4 months. This technique was selected because we wanted to estimate population parameters to compare the impact of managed care.21 To select potential confounders to be included in the model, we used the change-in-estimate method.22 The models included an outcome variable (was the child seen at a CF center in last 4 months: yes or no) and predictor variables that included managed care (yes or no), disease severity at baseline (advanced or mild), the CF child’s age (aged younger than 13 or aged 13 or more years), the CF child’s gender, the primary caregiver’s highest education (grade 12 or lower including vocational/trade school or higher than grade 12), the primary caregiver’s employment status (working full- or part-time or not working), and annual family income (< $40 000 or 
$40 000).
The analytical strategy involved performing univariate analyses with contingency tables to evaluate for confounders, logit plots to visually examine the relationship between the outcome and each predictor variable, and model fitting with the PROC LOGISTIC and PROC GENMOD procedures in SAS version 8.02 (SAS Institute, Inc., Cary, NC). A full model with all possible 2-way interactions was fitted first.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
All CF identification numbers in the sampling frame were accounted for, and 630 of 1152 (54.7%) eligible caregivers agreed to participate. These participants represented 85.7% of the prerecruitment estimate of 735. Of the 522 not participating, 383 (73.4%) refused. The 2 most common indications for refusal given were "no interest" and "no time." Of the remaining non-participants, 89 (17.0%) were unable to be contacted, and 50 (9.6%) initially agreed to participate but did not return their informed consent. The overall attrition rate was 16%, which was lower than the initially expected 25% dropout rate. At baseline, 63.5% of the cohort was in the non-managed care category and 36.5% had managed care. Although some individuals changed insurance type between interviews 2 and 7, the proportion of participants in the non-managed care category was relatively constant ranging from 62.4% to 68.1%.
As shown in Table 2, the participating children with CF were not significantly different from those who were eligible but did not participate. As intended, the cohort had more cases of advanced disease than the population in the Cystic Fibrosis Foundation Patient Registry. The average age of the 630 children with CF was 12.6 ±3.3 (mean ±SD) years, and 314 (49.8%) were male. A total of 421 (66.8%) children were classified as having advanced disease at baseline. As shown in Table 3, there were no differences in baseline characteristics by managed care, except for the child’s gender.
|
|
At baseline, logistic regression showed that there were no significant differences in attendance at the CF center by children with managed care (odds ratio [OR] = 0.64; 95% confidence interval [CI] = 0.32, 2.30; P = 0.21) in a model adjusting for the child’s age, gender, and disease severity; the care-giver’s education and employment status; and family income. The model also included the following significant 2-way interactions: (1) disease severity x child’s age; (2) care-giver’s education x employment status; and (3) caregiver’s education x child’s age.
In the longitudinal analysis, the interview number was included along with predictor variables and the most significant 2-way interaction term, which was child’s age and interview number. As shown in Table 4, the 2 variables that were found to be significantly associated with decreased attendance at the CF center were managed care (OR = 0.74; 95% CI = 0.57, 0.98; P = 0.03) and female gender (OR = 0.68; 95% CI = 0.48, 0.97; P= 0.04).
|
To assess why children did not visit the CF center, caregivers were asked whether not visiting was related to any of the following: (1) unable to obtain a referral from primary care physician or health insurance denied visit; (2) personal financial reasons (co-payments, deductibles, co-insurance, etc.); (3) transportation difficulties; (4) parent did not feel routine visit was needed; (5) CF center physician did not recommend routine visit; and (6) other. There were no significant differences by management of care. The most frequently reported reason was "other" (e.g., physician on vacation, child was well, child could not miss school).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
This was a prospective cohort study that evaluated the effect of managed care on attendance at the CF centers by children aged 6–18 years. To our knowledge, this is the first report describing decreased use to care for children with CF with managed care.
The data showed that children with managed care were significantly less likely to attend CF centers compared with those who had non-managed care. These differences persisted after adjusting for the child’s disease severity, age, and gender; the caregiver’s education and employment status; and family income. The cohort was not significantly different from children who chose not to participate (Table 2
). Additionally, the cohort was not different from children in the Cystic Fibrosis Foundation Patient Registry with regard to age and gender. By design, there was over-sampling of children with advanced disease, which explains the significant differences in disease severity. Given this, the findings should be generalizable to the CF population as a whole.
Our analysis of baseline data indicated that group characteristics were comparable between managed care and non-managed care, with the exception of the child’s gender (Table 3). Also, baseline data showed reduced attendance for children with managed care, although the relationship was not statistically significant. By contrast, the longitudinal analysis showed significantly decreased odds of visiting the CF center in the previous 4-month interval (Table 4). Whereas baseline data were cross-sectional and more prone to recall bias, longitudinal data were collected every 4 months, more accurately capturing changes in insurance type and features and changes in health care needs.
Our data showed that girls were less likely to attend CF centers. The reasons for this are unclear. Other reports have suggested that socioeconomic status may play a role in differential access to care.1,23 In our cohort, there were no significant differences by gender with respect to our socioeconomic markers, which were caregiver education/employment status and family income. However, decreased use may have clinical implications, because girls have more rapid progression of disease and decreased survival,24–26 suggesting that they have a greater need for access to expert CF care.
We found that 20% of caregivers and 10% of insurance company representatives stated that they did not know whether the child’s plan type was regarded as managed care. In view of potential limitations related to self-report information,18,27,28 we also validated insurance plan characteristics with the insurance customer service representatives listed on the child’s insurance card. The 
statistics to assess the level of agreement about various benefits and characteristics of plans ranged from 0.10 to 0.40, indicating poor-to-fair agreement between subscriber and insurance company representative.29 These findings suggest that the data obtained from insurance company representatives may not be a "gold standard" and should be used with caution. It also raises concerns about the accuracy of information given to caregivers by customer service representatives when caregivers inquire about plan characteristics.
We defined managed care, on the basis of self-report information, as a plan that requires a gatekeeper and does not cover use of a specialist without a referral. Although this is a narrow definition of managed care, consistent with previous reports,18–20 it will likely underestimate the influence of managed care because it only captures the major restrictive components of managed care that would prevent a child from accessing the CF center.
Our findings support concerns that managed care may limit access to providers or interventions that are state-of-the-art for CF. We found no differences between Medicaid managed care and private managed care with respect to visiting the CF center. This suggests that it is the plan characteristic and not the funding source that might limit access. Plans that require approval from a gatekeeper have been used to oversee the delivery of care to children with special health care needs.30 The effect of this system on clinical outcomes is uncertain, although it has been demonstrated that providers are not equally adept at managing chronic illnesses31 and denying the recommendations of CF center specialists or directing patients to less experienced providers may jeopardize the health of a child with CF.12
We found that the odds of attending the CF center were 26% lower for children in managed care. Families should consider this when selecting their medical insurance plan. In addition, insurance companies should recognize that managed care might limit access to specialists at the CF center. It remains to be seen whether managed care also imposes restrictions to other interventions, medications, and therapies recommended by national guidelines. The myriad of plan characteristics in an evolving marketplace necessitates a standard definition of managed care to assess its full impact. Until this occurs, comparing findings across studies will be hampered, and we recommend that the insurance industry, the medical community, and policymakers work together to address this issue. Although our study focused on CF, these issues may be applicable to other chronic illnesses (e.g., cancer, muscular dystrophy, spina bifida) that require a multidisciplinary approach to care.
|
Acknowledgments |
|---|
This work was supported by the Cystic Fibrosis Foundation (grant ZUCCON7A0) and the Nemours Foundation.
We thank Mary Clancy, MSN, and the following cystic fibrosis center directors: B. Rosenstein, MD (Baltimore, MD); H. Dorkin, MD (Boston, MA); D. Borowitz, MD (Buffalo, NY); G. W. Fernald, MD, G. Retsch-Bogart, MD, and M. Leigh, MD (Chapel Hill, NC); R. Wilmott, MD (Cincinnati, OH); C. Doershuk, MD, and M. Konstan, MD (Cleveland, OH); J. Cunningham, MD (Fort Worth, TX); P. Hiatt, MD (Houston, TX); H. Eigen, MD (Indianapolis, IN); M. Bowman, MD, PhD, and M. Woo, MD (Los Angeles, CA); M. Rock, MD (Madison, WI); L. Varlotta, MD (Philadelphia, PA); M. Wall, MD (Portland, OR); G. B. Mallory, MD, and T. Ferkol, MD (St. Louis, MO); D. Nielson, MD, PhD, and B. Chatfield, MD (Salt Lake City, UT).
Human Participant Protection
This study was approved by the institutional review board at each participating institution.
|
Footnotes |
|---|
Peer Reviewed
Contributors
I. Nathanson served as principal investigator responsible for the concept and overall design and directed the study. G. Ramírez-Garnica assisted with data collection and study coordination. S. Armatti Wiltrout served as data manager. All authors participated in analysis and interpretation of data.
Accepted for publication March 21, 2004.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONReferences |
|---|
1. Alessandrini EA, Shaw KN, Bilker WB, Schwarz DF, Bell LM. Effects of Medicaid managed care on quality: childhood immunizations. Pediatrics.2001;107:1335–1342.
2. Konstan MW, Butler SM, Schidlow DW, Morgan WJ, Julius JR, Johnson CA. Patterns of medical practice in cystic fibrosis: part I. Evaluation and monitoring of health status of patients. Pediatr Pulmonol.1999;28:242–247.[CrossRef][Web of Science][Medline]
3. Newacheck PW, McManus MA, Fox HB. Prevalence and impact of chronic illness among adolescents. Am J Dis Child.1991;145:1367–1373.
4. Ireys HT, Anderson GF, Shaffer TJ, Neff JM. Expenditures for care of children with chronic illnesses enrolled in the Washington state Medicaid program, fiscal year 1993. Pediatrics.1997;100:197–204.
5. Fowler EJ, Anderson GF. Capitation adjustment for pediatric populations. Pediatrics.1996;98:10–17.
6. Schechter MS, Shelton BJ, Margolis PA, FitzSimmons SC. The association of socioeconomic status with outcomes in cystic fibrosis patients in the United States. Am J Respir Crit Care Med.2001;163:1331–1337.
7. Seid M, Steven GD, Varni JW. Parents’ perceptions of pediatric primary care quality: effects of race/ethnicity, language, and access. Health Serv Res.2003;38:1009–1031.[CrossRef][Web of Science][Medline]
8. Cooper H, Smaje C, Arber S. Use of health services by children and young people according to ethnicity and social class: secondary analysis of a national survey. BMJ.1998;317:1047–1051.
9. Saxena S, Majeed A, Jones M. Socioeconomic differences in childhood consultation rates in general practice in England and Wales: prospective cohort study. BMJ.1999;318:642–646.
10. Wood R, Boat T, Doershuck C. Cystic fibrosis. Am Rev Respir Dis.1976;113:833–878.[Web of Science][Medline]
11. Patient Registry 2002 Annual Report. Bethesda, MD: Cystic Fibrosis Foundation; 2003.
12. Johnson C, Butler SM, Konstan MW, Morgan W, Wohl MEB. Factors influencing outcomes in cystic fibrosis: a center-based analysis. Chest.2003;123:20–27.
13. Cabrera ME, Lough MD, Doershuk CF, Salvator AE. An expanded scoring system including an index of nutritional status for patients with cystic fibrosis. Pediatr Pulmonol.1994;18:199–205.[Web of Science][Medline]
14. Schwachman H, Redmond A, Khaw K-T. Studies in cystic fibrosis: report of 130 diagnosed under 3 months of age over a 20-year period. Pediatrics.1970; 46:335–343.
15. Taussig LM, Kattwinkel J, Friedwald WT, di Sant’Agnese PA. A new prognostic score and clinical evaluation system for cystic fibrosis. J Pediatr.1973;82:380–390.[CrossRef][Web of Science][Medline]
16. Sockrider MM, Swank PR, Seilheimer DK, Schidlow DV. Measuring clinical status in cystic fibrosis: internal validity and reliability of a modified NIH score. Pediatr Pulmonol.1994;17:86–96.[Web of Science][Medline]
17. Zucconi SL, Nathanson IT, Jacobson LP, Clancy MK, Denier M. Validation of caregiver reports of insurance coverage for children with chronic illnesses. Paper presented at: Annual Meeting of the American Public Health Association; November 1996; New York, NY.
18. Nelson D, Thompson BL, Davenport NJ, Penaloza LJ. What people really know about their health insurance: a comparison of information obtained from individuals and their insurers. Am J Public Health.2000;90:924–928.
19. Blendon RJ, Brodie M, Benson JM, et al. Understanding the managed care backlash. Health Aff (Millwood).1998;17:80–94.[Abstract]
20. Hargraves JL, Cunningham PJ, Hughes RG. Racial and ethnic differences in access to medical care in managed care plans. Health Serv Res.2001;36:853–868.[Web of Science][Medline]
21. Diggle PJ, Liang K, Zeger SL. Analysis of Longitudinal Data. New York, NY: Oxford University Press; 1994.
22. Maldonado G, Greenland S. Simulation study of confounder selection strategies. Am J Epidemiol.1993; 138:932–936.
23. Moore P, Fenlon N, Hepworth JT. Indicators of differences in immunization rates of Mexican American and white non-Hispanic infants in a Medicaid managed care system. Public Health Nurs.1996;13:21–30.[Web of Science][Medline]
24. Rosenfeld M, Davis R, FitzSimmons S, Pepe M, Ramsey B. Gender gap in cystic fibrosis mortality. Am J Epidemiol.1997;145:794–803.
25. Liou TG, Adler FR, FitzSimmons SC, Cahill BC, Hibbs JR, Marshall BC. Predictive 5-year survivorship model of cystic fibrosis. Am J Epidemiol.2001;153:345–352.
26. Fogarty A, Hubbard R, Britton J. International comparison of median age at death from cystic fibrosis. Chest.2000;117:1656–1660.
27. Alessandrini EA, Shaw KN, Bilker WB, Perry KA, Baker MD, Schwartz DF. Effects of Medicaid managed care on health care use: infant emergency department and ambulatory services. Pediatrics.2001;108:103–110.
28. Guendelman S, Wyn R, Yi-Wen T. 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]
29. Armatti SR, Ramírez-Garnica G, Nathanson IT. Understanding and classifying insurance benefits of children with CF. Pediatr Pulmonol.2001;32(suppl 22): 359.
30. Ireys HT, Grason HA, Guyer B. Assuring quality of care for children with special needs in managed care organizations: roles for pediatricians. Pediatrics.1996; 98:178–185.
31. Powe NR, Weiner JP, Starfield B, Stuart M, Baker A, Steinwachs DM. Systemwide provider performance in a Medicaid program: profiling the care of patients with chronic illnesses. Med Care.1996;34:798–810.[CrossRef][Web of Science][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |