Despite improved diagnosis and management of

The Impact of Follow-up Physician Visits on Emergency Readmissions for Patients with Asthma and Chronic Obstructive Pulmonary Disease: A Population-based Study Don D. Sin, MD, MPH, Neil R. Bell, MD, MSc, Lawrence W. Svenson, BSc, S. F. Paul Man, MD PURPOSE: To examine the relation between follow-up office visits after emergency discharge and the risk of emergency readmissions in patients with asthma or chronic obstructive pulmonary disease (COPD). SUBJECTS AND METHODS: We used population-based data to identify all patients in Alberta, Canada, who had at least one emergency visit for asthma or COPD between April 1, 1996, and March 31, 1997 (N 25 256). A Cox proportional hazards model was used to estimate the adjusted relative risk (RR) of a repeat visit to an emergency department within 90 days of an initial emergency visit in patients who did or did not have an office follow-up within the first 30 days. RESULTS: There were 7829 patients (31%) who had an office visit during the 30 days after their initial emergency encounter. Follow-up visits were associated with a significant reduction in the 90-day risk of an emergency readmission (RR 0.79; 95% confidence interval [CI]: 0.73 to 0.86). Sensitivity analyses showed that a follow-up visit was inversely associated with a repeat emergency visit after adjusting for age, sex, area of residence, and income. CONCLUSION: Although these data should be interpreted with caution because of missing information on factors such as quality of care, they suggest that follow-up office visits are effective in reducing early relapses in patients who have been recently treated in emergency departments for asthma or COPD. Am J Med. 2002;112:120 125. 2002 by Excerpta Medica, Inc. From the Departments of Medicine, Pulmonary Division (DDS, SFPM), Family Medicine (NRB), and Public Health Sciences (NRB, LWS), University of Alberta, Edmonton, Alberta, Canada; and the Health Surveillance Branch, Alberta Health and Wellness (LWS), Alberta, Canada. Requests for reprints should be addressed to Don D. Sin, MD, MPH, 2E4.39 Walter C. Mackenzie Centre, University of Alberta, Edmonton, Alberta T6G 2B7, Canada. Manuscript submitted April 19, 2001, and accepted in revised form October 29, 2001. Despite improved diagnosis and management of asthma and chronic obstructive pulmonary disease (COPD), these conditions still have considerable effect on morbidity, mortality, and health care costs in the United States and other developed countries (1,2). For instance, the prevalence of asthma and COPD increased by 75% in the United States between 1980 and 1994 (3), while age-adjusted mortality increased by 71% between 1966 and 1995 (2,4). The overall direct costs for care were an estimated $23 billion in the United States in 1993 (5) and over $1.3 billion in Canada (6). Emergency departments provide a substantial proportion of the overall care for patients with asthma and COPD. During 1994, there were an estimated 3 million emergency department admissions for asthma and COPD in the United States (4), many of which were due to relapse, or were repeat admissions within days to weeks after the initial discharge. Recent studies report relapse rates of 17% within 14 days (7) and up to 45% at 8 weeks after discharge (8), similar to those found in earlier studies, suggesting that relapse rates have not changed substantially during the past 20 years (9 13). Patients with frequent relapses requiring admission to emergency departments often have more severe disease and use a disproportionately higher share of health care resources (14 16). They may have a history of prior emergency department or urgent clinic visits, or a longer duration of symptoms; be exposed to multiple environmental triggers; report difficulty performing work or daily activities in the 4 weeks before admission; or use home nebulizers (7,8). Given that the value of pulmonary function testing in predicting emergency admissions due to relapse is unclear (7 9,17), follow-up care after discharge from emergency departments may be an effective means of preventing such readmissions (17,18). Since there is a paucity of data to support this hypothesis, we sought to examine the relation between follow-up care with physicians and the risk of emergency readmissions in patients with asthma or COPD. MATERIAL AND METHODS Study Design and Patient Data We conducted a population-based cohort study of all patients in Alberta, Canada, who were admitted to emergency rooms with asthma or COPD as their principal diagnosis. In accordance with the Canada Health Act, Alberta has a single-payer, publicly funded health care system that provides universal health care coverage to all 120 2002 by Excerpta Medica, Inc. 0002-9343/02/$ see front matter All rights reserved. PII S0002-9343(01)01079-8

residents. Patient encounter data were obtained from medical record registries maintained by Alberta Health and Wellness. From the physicians claims registry, we identified all patients who had at least one emergency visit during April 1, 1996, to March 31, 1997. We defined the initial emergency visit as the patient s first admission to an emergency department during this period. All subsequent admissions to emergency departments for these conditions during the 3 months after discharge were considered readmissions. To be reimbursed for services rendered to their patients, physicians must have submitted a claim to Alberta Health and Wellness. A claims submission includes the type, date, and location of services rendered, as well as the specialty of the physician. It also lists the primary diagnosis for each of the encounters using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes (19). Separate fee codes exist for emergency department and office visits, allowing the identification of the different types of visits. Asthma was defined by ICD-9-CM code 493.x, and COPD was defined by ICD-9-CM codes 490.x, 491.x, 492.x, and 496.x. We linked this data with the Alberta Health Care Insurance Plan registry, which contains demographic information on each person residing in Alberta, including age, sex, income, and area of residence. To ensure patient confidentiality, all patient information was linked using an encrypted common patient health care number prior to release from Alberta Health and Wellness. To determine the number of emergency department and office visits for asthma and COPD in 1995 (the fiscal year before the study year), we linked patient data with the 1995 physicians claims registry, and used similar emergency and office billing codes. The study was approved by the Health Research Ethics Board of the University of Alberta. Statistical Analysis Continuous variables are expressed as means ( SD) and were compared with the Student t test. Nonnormal distributions were compared using the Wilcoxon rank sum test. Ordinal and binary variables were compared using the chi-squared test. We used a Cox proportional hazards model to compare the 90-day risk of emergency readmission for asthma or COPD in patients who did or did not have a follow-up physician visit within 30 days of the initial emergency visit. We chose 90 days because previous studies have demonstrated that the risk of readmission and relapse was greatest within the first 3 months after discharge (7 9). This is also the time during which follow-up visits with physicians were likely to affect emergency readmissions. Patients were followed from the time of discharge from their initial emergency visit until the date of readmission, 90 days after discharge, or March 31, 1997, whichever was first. Emergency visits occurring after the first repeat emergency visit were censored. The mean duration of follow-up was 73 29 days. In the final model, we adjusted for age, sex, area of residence (metropolitan vs. nonmetropolitan), and low income. There were five mutually exclusive categories for age: 20 years, 20 to 39 years, 40 to 59 years, 60 to 79 years, and 80 years. We defined metropolitan regions as areas with 500 000 inhabitants. Low income was defined as an adjusted annual individual income of less than $7560 (Canadian dollars) (20). In secondary analyses, we evaluated the association between follow-up visits and the risk of emergency readmissions in selected subgroups of patients to determine the sensitivity of our findings. We hypothesized that if the inverse association between follow-up visits and emergency readmissions was true, then the protective effect of follow-up visits should be evident in all subgroups. A two-tailed P value of 0.05 was considered statistically significant. All analyses were conducted using SAS software version 7.01 (Cary, North Carolina). RESULTS Of the 25 256 persons who were admitted to an emergency department for asthma or COPD during the study period, 2610 (10%) had a repeat emergency visit, and 7829 (31%) had at least one follow-up visit with a physician during the month following discharge from their initial emergency visit. Of the patients who had a follow-up physician visit, 6643 (85%) had a visit with a general practitioner, while 1186 (15%) had a visit with a specialist. Table 1 shows the demographic and clinical characteristics of patients in the study. Female patients, those residing in metropolitan areas, or those who had low income were more likely to have had an office visit following their initial emergency admission. The number of emergency or office visits in the year before the initial emergency admission was similar between those who did and did not have a subsequent follow-up visit. A follow-up physician visit was associated with a lower risk of emergency readmissions (RR 0.82; 95% CI: 0.76 to 0.86), which was unchanged after adjusting for age, sex, area of residence, and income (RR 0.79; 95% CI: 0.73 to 0.86) (Figure). Similar associations were observed for both asthma and COPD. When analyzed separately, the relative risk of readmission for asthma among patients who had had a follow-up visit was 0.75 (95% CI: 0.67 to 0.83), whereas the relative risk for COPD was 0.77 (95% CI: 0.63 to 0.94). Factors associated with an increased risk of emergency readmissions following the initial emergency visit included older age ( 60 years), male sex, low income, and residence within a metropolitan center (Table 2). February 1, 2002 THE AMERICAN JOURNAL OF MEDICINE Volume 112 121

Table 1. Characteristics of Patients Who Did and Did Not Have a Follow-up Office Visit following an Emergency Admission Characteristic Office Visit (n 7829) No Office Visit (n 17 427) P Value Number (%) or Mean SD Age (years) 31 26 31 30 0.10 Female sex 4083 (52) 8619 (50) 0.005 Metropolitan residence 3878 (50) 7070 (41) 0.01 Low income 2260 (29) 4704 (27) 0.01 Prior emergency admissions* 0.4 1.3 0.4 1.4 0.71 Prior office visits* 1.3 2.4 1.3 2.6 0.99 * Number of emergency admissions or office visits in 1995 (the fiscal year before the study year). A sensitivity analysis was undertaken to examine the effect of demographic and geographic factors on the association between follow-up visits and the risk of emergency readmissions for patients with asthma or COPD (Table 3). Across all age, sex, income, and residential subgroups, a follow-up visit was associated with a protective effect against emergency readmissions, suggesting that our findings were not confounded by these variables. This association did not change when emergency cases were removed where secondary diagnostic fields were substituted with diagnoses other than asthma or COPD (RR 0.67; 95% CI: 0.61 to 0.73), nor when follow-up visits were restricted to only those with a primary code of asthma or COPD (RR 0.81; 95% CI: 0.74 to 0.89). The exclusion or inclusion of emergency visits associated with a hospital admission also had little effect on overall results. We observed a 27% reduction in the risk of an emergency readmission (RR 0.73; 95% CI: 0.64 to 0.83) among patients whose initial emergency encounter did not result in hospitalization, as well as a relative risk reduction of 18% in favor of follow-up visits among those whose initial emergency encounter resulted in hospitalization (RR 0.82; 95% CI: 0.71 to 0.94). DISCUSSION We found that an early follow-up visit after an emergency encounter was associated with a significant reduction in the risk of emergency readmissions among patients with asthma and COPD. When asthma and COPD are analyzed together, office visits within the first 30 days of the initial admission to the emergency department were associated with 21% fewer readmissions during the first 90 days after discharge. In patients with asthma, follow-up Figure. The cumulative risk of emergency department readmissions for patients with asthma and chronic obstructive pulmonary disease who did and did not have an office follow-up visit. The curves are significantly different from each other (P 0.005). Table 2. Risk Factors Independently Associated with Repeat Emergency Admissions Characteristic Adjusted Relative Risk (95% Confidence Interval) for Emergency Readmission* 1 Physician visit 0.79 (0.73-0.86) Age (years) 20 to 39 0.90 (0.81 1.00) 40 to 59 1.09 (0.96 1.23) 60 to 79 1.43 (1.28 1.59) 80 1.22 (1.04 1.43) Female sex 0.92 (0.85 0.99) Annual income ($) 7560 1.18 (1.08 1.29) Metropolitan area of residence 1.41 (1.30 1.53) Number of prior emergency admissions 1 to 2 2.18 (2.01 2.35) 3 4.29 (3.88 4.73) * Adjusted for all variables listed (see Methods section). In Canadian dollars. 122 February 1, 2002 THE AMERICAN JOURNAL OF MEDICINE Volume 112

Table 3. Association between Follow-up Physician Visits and the Risk of Emergency Readmissions in Selected Subgroups of Patients Subgroup Number of Patients Risk of Repeat Emergency Visit Number (%) Adjusted Relative Risk (95% Confidence Interval)* Age (years) 0 to 19 10 940 1067 (10) 0.76 (0.66 to 0.87) 20 to 39 5992 532 (9) 0.89 (0.73 to 1.07) 40 to 59 3142 349 (10) 0.83 (0.66 to 1.05) 60 to 79 3365 474 (14) 0.76 (0.63 to 0.93) 80 1468 188 (13) 0.65 (0.45 to 0.95) Sex Male 12 554 1358 (11) 0.82 (0.73 to 0.93) Female 1252 12 702 (10) 0.75 (0.66 to 0.85) Annual income ($) 7560 18 292 1767 (10) 0.83 (0.74 to 0.92) 7560 6964 843 (12) 0.72 (0.62 to 0.84) Area of residence Nonmetropolitan 14 308 1257 (9) 0.78 (0.68 to 0.88) Metropolitan 10 948 1353 (12) 0.80 (0.72 to 0.90) * Adjusted for all variables listed (see Methods section). In Canadian dollars. visits were associated with 25% fewer emergency readmissions, while for those with COPD, a 23% reduction in risk was observed. These findings, however, should be interpreted with caution because some important information, including that on quality of care provided by emergency departments during the initial encounter, was missing. Nevertheless, these data suggest that follow-up visits are effective in lowering the risk of clinical relapses following emergency rescue care, and provide an impetus for future research on continuity of care. We chose to follow our cohort for 90 days following the initial emergency encounter because most relapses occur during this period, as do interventions that are likely to be most effective in reducing the number of relapses and repeat emergency visits. In a study by Naylor et al. (21), comprehensive discharge planning and intense follow-up were associated with 45% fewer readmissions to hospitals for elderly patients who had been previously hospitalized. However, the beneficial effect of an intense follow-up decreased after 90 days postdischarge. Similarly, the effect of an intense follow-up was shown to be greatest during the first 90 days postdischarge (22). Despite the few financial barriers to physician access in the Alberta health care system, only 35% of patients with asthma and COPD had a follow-up visit with a physician within 30 days of their initial emergency visit, suggesting that other barriers to care exist and that more effective primary care could further reduce emergency admissions for asthma and COPD (23 25). Additional studies to identify these barriers would allow the development of new strategies that would reduce emergency readmissions (26). In this study, we could not determine the specific practice characteristics and management strategies of physicians that may have reduced the number of emergency readmissions. Physicians can provide patient care in an environment that encourages continuity of care and the development of long-term physician-patient relationships (27,28). This pattern of care would facilitate patient education, counseling, medication adjustments, and referrals to specialists and other health care professionals, as well as improve patient satisfaction, increase compliance with medical care, and provide the opportunity to manage acute disease exacerbations on an outpatient basis, which should reduce the risk of readmissions to emergency departments (29,30). Previous studies that examined the relation between access to primary care and subsequent care needs have had conflicting results. Lieu et al. (31) demonstrated that having a regular physician provider reduced the risk of hospitalization due to asthma by over 30%, compared with not having a regular provider (RR 0.64; 95% CI: 0.43 to 0.95). Similarly, Emerman and Cydulka reported that 91% of relapses occurred in patients without a follow-up visit to a physician, emphasizing continuity of care (32 34). Our findings corroborate these data by showing that follow-up physician visits are associated with reduced risks of emergency readmissions across all age and sex groups for both asthma and COPD. In contrast to these studies, Weinberger et al. (35) found that improved access to primary care following hospital discharges for older patients with diabetes, COPD, or congestive heart failure was associated with increased rehospitalization. One explanation may be that February 1, 2002 THE AMERICAN JOURNAL OF MEDICINE Volume 112 123

follow-up care might not have been effective in reducing the number of hospitalizations or other adverse outcomes in elderly patients who had advanced disease. It is also possible that by assigning patients to primary care physicians with whom patients had had no prior relationship, factors present in long-standing relationships, such as trust and confidence, might have been lacking, thus possibly affecting the treatment of very ill, elderly patients in an ambulatory setting (36 38). There are several limitations to our study. First, because the study was observational in nature, our findings may have been affected by unknown differences in the characteristics of patients who did or did not have follow-up visits with physicians. We therefore restricted our analyses to patients who had an emergency visit, ensuring that their disease severity was similar. We also adjusted for age, sex, area of residence, and income to improve comparisons between the two groups. The crude and adjusted risk estimates were similar, providing some evidence that the two groups were comparable. Moreover, the association between follow-up visits with physicians and the reduction in risk for emergency readmissions was similar regardless of how the data were stratified for age, sex, location, or income. Second, because the databases did not contain any information on therapies received by patients while in the emergency departments, we could not compare the suitability of treatment between those who did and did not undergo follow-up visits. Third, although clinicians distinguish between asthma and COPD in theory, diagnostic misclassification can occur in practice, especially in older patients (39). However, when we analyzed the data for asthma and COPD visits combined, as well as separately, we found that results were similar in all analyses. While not definitive, our data provide some assurances for practicing clinicians and other allied health care workers that follow-up care should be encouraged after an exacerbation of acute asthma or COPD. Future randomized trials are required to confirm our findings and to determine the factors involved in follow-up care that improve outcomes in these patients. ACKNOWLEDGMENT We thank Thomas Marrie, MD, and Finlay McAlister, MD, for their helpful review of the manuscript. REFERENCES 1. Grant EN, Wagner R, Weiss KB. Observations on emerging patterns of asthma in our society. J Allergy Clin Immunol. 1999; 104(suppl):S1 S9. 2. Hurd S. The impact of COPD on lung health worldwide: epidemiology and incidence. Chest. 2000;117(suppl):1S 4S. 3. Mannion DM, Homa DM, Pertowski CA, et al. Surveillance for asthma, United States, 1960 1995. MMWR CDC Surveill Summ. 1998;47:1 27. 4. NHLBI Morbidity and Mortality Chartbook, 1998. Available at: http://www.nhlbi.nih.gov/resources/docs/cht-book.htm. 5. Sullivan SD, Ramsey SD, Lee TA. The economic burden of COPD. Chest. 2000;117(suppl):5S 9S. 6. Health Canada. Economic Burden of Illness, Canada. 1993. Available at: http://www.hc-sc.gc.ca/hpb/lcdc/publicat/burden/burd4_e. html#burdn1. 7. Emerman CL, Woodruff PG, Cydulka RK, et al. Prospective multicenter study of repeat following treatment for acute asthma among adults presenting to the emergency department. Chest. 1999;115: 919 927. 8. McCarren M, McDermott MF, Zalenski RJ, et al. Prediction of repeat within eight weeks after an acute asthma exacerbation in adults. J Clin Epidemiol. 1998;51:107 118. 9. Emerman CL, Cydulka RK. Factors associated with repeat after emergency department treatment of acute asthma. Ann Emerg Med. 1995;26:6 11. 10. Ducharme FM, Kramer MS. Relapse following emergency treatment for acute asthma: can it be predicted or prevented? J Clin Epidemiol. 1996;46:1395 1402. 11. Kelsen SG, Kelsen DP, Fleegler BF, et al. Emergency room assessment and treatment of patients with acute asthma: adequacy of conventional approach. Am J Med. 1978;64:622 628. 12. Fischl MA, Pitchenik A, Gardner LB. An index predicting repeat and need for hospitalization inpatients with acute bronchial asthma. N Engl J Med. 1981;305:783 789. 13. Lederle FA, Pluhar RE, Joseph AM, Niewoehner DE. Tapering of corticosteroid therapy following exacerbation of asthma: a randomized double-blind, placebo-controlled trial. Arch Intern Med. 1987;147:2201 2203. 14. Sullivan S, Elixhauser A, Buist AS, et al. National Asthma Education and Prevention Program working group report on the cost effectiveness of asthma care. Am J Respir Crit Care Med. 1996;154(3 pt 2):584 585. 15. Smith DH, Malone DC, Lawson KA, et al. A national estimate of the economic costs of asthma. Am J Respir Crit Care Med. 1997;156: 787 793. 16. Stanford R, McLaughlin T, Okamoto LJ. The cost of asthma in the emergency department and hospital. Am J Respir Crit Care Med. 1999;160:211 215. 17. Varon J, Fromm RE. Emergency department care of the asthma patient: predicting bounce-back patients. Chest. 1999;115:909 911. 18. Steigman DM. Is it urban or is it asthma? Lancet. 1966;348:143 144. 19. The International Classification of Diseases, Ninth Revision, Clinical Modification: ICD-9-CM. Ann Arbor: Commission on Professional and Hospital Activities; 1992. 20. Alberta Health and Wellness: Premium and Rates, 2000. Available at: http://www.health.gov.ab.ca/ahcip/premiums.htm. 21. Naylor MD, Brooten D, Campbell, et al. Comprehensive discharge planning and home follow-up of hospitalized elders. A randomized clinical trial. JAMA. 1999;281:613 620. 22. Rich MW, Beckham V, Wittenberg C, et al. A multidisciplinary intervention to prevent the readmission of elderly patients with congestive heart failure. N Engl J Med. 1995;333:1190 1195. 23. Hartert TV, Windom HH, Peebles RS, et al. Inadequate outpatient medical therapy for patients with asthma admitted to two urban hospitals. Am J Med. 1966;100:386 394. 24. Wolle JM, Cwi JC. Physicians prevention-related practice behaviors in treating adult patients with asthma: results of a national survey. J Asthma. 1995;32:309 318. 25. Roghmann M, Sexton M. Adherence to asthma guidelines in general practices. J Asthma. 1999;36:381 387. 124 February 1, 2002 THE AMERICAN JOURNAL OF MEDICINE Volume 112

26. Neville RG, Higgins B. Providing better asthma care: what is there left to do? Thorax. 1999;54:813 817. 27. Neuberger J. Primary care: core values. Patient priorities. BMJ. 1998;317:260 262. 28. Olesen F, Dickinson J, Hjortdahl P. General practice time for a new definition. BMJ. 2000;320:354 357. 29. McWhinney IR. Primary care: core values. Core values in a changing world. BMJ. 1998;316:1807 1809. 30. Hjortdahl P, Laerum E. Continuity of care in general practice: effect on patient satisfaction. BMJ. 1992;304:1287 1290. 31. Lieu TA, Quesenberry CPA, Sorel ME, et al. Computer-based models to identify high-risk children with asthma. Am J Respir Crit Care Med. 1998;157:1173 1180. 32. Emerman CL, Cydulka RK. Factors associated with relapse after emergency department treatment for acute asthma. Ann Emerg Med. 1995;26:6 11. 33. Berwick DM. Quality of health care. Part 5: payment by capitation and the quality of care. N Engl J Med. 1996;335:1227 1231. 34. Ettner SL. The relationship between continuity of care and the health behaviors of patients: does having a usual physician make a difference? Med Care. 1999;37:547 555. 35. Weinberger M, Oddone EZ, Henderson WG. Does increased access to primary care reduce hospital readmission? N Engl J Med. 1996; 334:1441 1447. 36. Henley DE. A trial of increased access to primary care. N Engl J Med. 1996;335:895. 37. Starfield B, Parrino TA. A trial of increased access to primary care. N Engl J Med. 1996;335:895 896. 38. Rich MW. A trial of increased access to primary care. N Engl J Med. 1996;335:896. 39. Kesten S, Chapman KR. Physician perception and management of COPD. Chest. 1993;104:254 258. February 1, 2002 THE AMERICAN JOURNAL OF MEDICINE Volume 112 125