COPD Readmissions and Hospital Primary Care Physicians

Chronic Obstructive Pulmonary Disease, more commonly known as COPD, is a prevalent chronic condition affecting about 10% of individuals worldwide and 15 million individuals within the United States.(1) COPD can be defined as a multi-dimensional disease characterized by the presence of “poorly reversible airflow limitation” within patients.(2)

Currently, COPD is the third leading cause of death in the United States, with significant economic implications resulting in annual direct costs equating to $50 billion.(1) Populations that appear to be at risk of COPD and 30-day readmissions are current and past smokers, individuals working in occupationally hazardous conditions, the medically indigent, and individuals with a lower socioeconomic status.(3)

Established by the National Heart, Lung & Blood Institute (NHLBI) and World Health Organization (WHO) in 2001, the Global Strategy for the Diagnosis and Prevention of COPD (GOLD) has helped bring the highest level of standardized practices for treatment and diagnosis within the United States.(4) Within standard diagnosis practices, patients suspected of being symptomatic of COPD take a survey administered by a primary care provider (PCP).(5) Upon hospital or ED admission for an acute exacerbation, typically due to chronic comorbidities, a patient is attended to by a general practicing physician or pulmonologist.(6) Generally, a patient’s symptoms are the main focal points of treatment, with pneumonia and respiratory failure being responsible for 22.3% and 21.4% of all COPD-related hospitalizations, respectively.(4)

There is sparse evidence to guide healthcare facilities in developing successful care and quality programs that improve COPD readmissions. A 2014 study found inconsistencies between case definitions of COPD and study populations identified as having COPD.(7) Furthermore, many programs seem to assume that the majority of readmissions occurring early in the post-discharge period are avoidable, but it is more likely that only a small number of early hospital readmissions are potentially modifiable.(8) In fact, to date no evidence-based intervention has shown to be effective in reducing all-cause readmission risk.(8-10)

Following discharges, approximately 1 in 5 individuals are readmitted within 30 days.(11) These readmissions are catalyzed by poor care provided, premature discharge, suboptimal medication reconciliation, lack of disease knowledge, and a disconnect between outpatient physicians and the patient.(12)

Post-discharge, 10%–20% of COPD patients are readmitted to a healthcare facility within 30 days.(11,13) A 2017 U.S.-based study demonstrated that approximately 22.8% of COPD readmissions were potentially preventable, while a 2020 Australian study demonstrated 39% were potentially preventable.(14,15) Current methods of care have yielded similar and undesirable results, with the 30-day readmission rate staying at a constant ~20% for most instances.

The introduction of PCP-led care could be a viable alternative to the current methods of care. In many instances, patients are not advised to schedule a post-discharge follow-up with a PCP or pulmonary specialist 48–72 hours after discharge.(12) PCP engagement in chronic condition management is critical for positive outcomes, and COPD is not an outlier in that regard.

Studies have proven that PCP engagement for in-patient care and post-discharge can lead to better patient education, a better understanding of treatment goals, and improvements in medication management, all of which have been inferred to lead to reductions in 30-day readmissions(16); therefore, this article aims to examine the overall relationship between hospital PCP and COPD 30-day readmissions.

Methodology

Data Source and Study Population

We used data from the Centers for Medicare & Medicaid Services (CMS), the American Hospital Association (AHA), and the Area Health Resource Files (AHRF) for our study period (2010–2019) to examine the relationship between primary care physicians in an inpatient setting and hospitals’ readmission rate for COPD. Specifically, we used the 2014–2019 CMS HRRP,(1) which contains risk-adjusted hospital inpatient readmission for 3,288 hospitals from July 1 through December 30.

The CMS HRRP data examine hospital inpatient readmission using a three-year rolling score between July 1 and December 30. For example, the 2014 CMS HRRP data examine hospital inpatient readmission from July 1, 2011, through December 30, 2014. Data for 2015 CMS HRRP examine hospital inpatient readmission from July 1, 2012, through December 30, 2015, etc.

Rates were risk-adjusted for patients’ medical history, comorbidities, age, and sex. CMS conducted standardization by using hierarchical linear modeling to estimate expected and predicted ratios for COPD readmissions and multiplied it by the unadjusted overall readmission and mortality rates.(2)

To report hospital characteristics, we aligned hospital characteristics years with HRRP readmission years and eight years of AHA annual survey data accounting for a one-year data collection lag.(4) We used the AHRF to augment the data with county characteristics. To generate a strongly balanced sample, we limited our sample to non-specialty acute care facilities within the 50 states and the District of Columbia that reported data on all eight years. We linked these datasets using each hospital’s Medicare provider number.

Our final sample included 1,527 hospitals that reported data on all eight annual surveys, participated in CMS HRRP, and included county information in the AHRF.

Main Measures

The primary dependent variable for this study was readmission associated with COPD. The primary independent variable was the percentage of primary care physicians in a hospital calculated by the total number of privileged primary care physicians divided by the total number of physicians privileged to practice at the hospital and multiplied by a hundred. The AHA survey defines primary care physicians to be physicians in general practice, general internal medicine, family practice, general pediatrics, and geriatrics.(17)

Control Variables

Hospital characteristic variables were included to control for confounding factors for the study period (2010–2019). We controlled for hospital size ownership, teaching affiliation, system status, and proportion of Medicare and Medicaid discharges. Hospital size is defined as the total number of staffed hospital beds. Size may influence hospitals’ performance 30-day readmission rate and is a predictor of hospital quality of care and available resources.(18,19) U.S. hospital bed size may range from very few beds to more than 2,000; therefore, we used traditionally defined size categories of hospital beds: small (fewer than 100 beds), medium (100–399 beds), and large (400 beds or more).

Hospital ownership is categorized as not-for-profit, for-profit, and government-owned. Ownership status has been shown in the literature to be a predictor of quality performance.(20) Teaching affiliation was operationalized as a categorical variable and is an indicator of hospital safety and readmission rates.(21) System status has been shown as a predictor of resource availability and therefore market power.(22) Medicare and Medicaid percentages were used as a predictor of a hospital’s financial stability.(23)

Lastly, we included hospital location; hospitals located in rural counties, as opposed to urban counties, may not be exposed to demands of powerful constituents, and they may be able to provide highly efficient care.(24)

To control for community-related confounding factors, we included in our study the average median age of county residents, recognizing that age is a strong predictor of respiratory-related chronic conditions.(25) Lastly, we used the Herfindahl-Hirschman Index (HHI) as a continuous variable to assess a county’s hospital market competition. An HHI of 0 is indicative of a competitive market.(26,27)

Statistical Approach

Using Stata 17 MP (StataCorp 2021: Release 17) we conducted descriptive and panel analyses to examine the relationship between PCP and COPD 30-day readmission. Furthermore, we conducted bivariate analysis to explore the association between our outcome variable, independent variable, and control variables.

For our analysis, panel data were balanced short and wide, which allowed us to examine a large number of hospitals over a brief period. The statistical analysis was conducted by estimating a multilevel Poisson panel regression model with variance components — random effects accounting for county population size as an exposure term. As Wooldridge(28) recommended, we conducted the standard Hausman test and its alternative robust formulation, which resulted in failing to reject the random-effects model.

Our random-effects model examines the relationship between hospital COPD 30-day readmission rate and PCP percent, allowing for individual effects for market characteristics and hospital characteristics. We conducted a robustness check by analyzing panel data of only large hospitals that are most likely to have a higher percentage of PCP. The Institutional Review Board (IRB) categorized the research as exempt since the study analyzed secondary data that is publicly available.

Results

Across all six years, the average COPD 30-day readmission rate was 20.71 with a 1.36 standard deviation. The average percent of PCP was 25.65% (see Table 1). About 60% of hospitals in our sample were medium-sized hospitals, and 26% were large. About 77% of hospitals were not-for-profit, 79% were part of larger health systems, 69% were teaching hospitals, and less than 10% of hospitals were located in rural counties.

An average of approximately 50% of hospital inpatient days were Medicare while only 20% were Medicaid days. Across all counties in our sample, the average median age was found to be about 38 years. Population per 100,000 was 9.72 and HHI was 0.61, suggesting a low level of competition.

Our final analysis model, presented in Table 2, considered both hospital characteristics and county-level demographics. PCP percent was associated with 30-day readmissions following COPD hospitalization, with an increase in PCP percent being associated with a 1.0016 (95% CI: 1.0006, 1.0026) incidence rate increase. Furthermore, compared to urban hospitals, rural hospitals were at an increased incident rate (IRR: 1.2221, 95% CI: 1.1569, 1.2910) of having a higher COPD 30-day readmission rate.

However, other hospital characteristics were associated with decreased COPD 30-day readmission rates. For example, we found that compared to small-sized hospitals, both medium- (IRR: 0.8892, 95% CI: 0.8410, 0.9402) and large-sized hospitals (IRR: 0.8230, 95% CI: 0.7680,0.8818) were more likely to have lower COPD 30-day readmission rates. Hospitals that are part of a system (IRR: 0.9341, 95% CI: 0.9011, 0.9682), teaching (IRR: 0.9472, 95% CI: 0.9211, 0.9740), and have higher Medicaid percentage (IRR: 0.9967, 95% CI: 0.9948, 0.9985) were also more likely to have lower COPD 30-day readmission rates; however, it is important to note the small effect size.

We found that all of our county characteristics were associated with an increased COPD 30-day readmission rate. Increase in county median age (IRR: 1.0138, 95% CI: 1.0050, 1.0227) and HHI (IRR: 1.7605, 95% CI: 1.5940,1.9443) were associated with an increased COPD 30-day readmission rate.

Discussion

Reducing COPD readmissions remains a priority for stakeholders throughout the U.S. healthcare system. Our study findings provide insight into organizational characteristics associated with those readmissions.

The primary focus of this work was to determine whether having an increased number of PCPs within the facility would contribute to reducing COPD readmission rates. Our results demonstrate a slight positive association. Hospitals in rural areas and those in less-competitive environments are associated with increased COPD readmission rates. On the other hand, hospitals that are part of a system, provide medical education, or are larger tend to have lower readmission rates.

While our results demonstrate a significant relationship between the increased number of PCPs and increased rates of COPD, the difference (effect size) is small enough that we question its relevance.(29) That is, the direction and strength of the association are so low, they round to 1, indicating no difference.

Despite this, it is still important to identify why additional PCPs might be related to readmissions. On one hand, it makes sense that an increased number of PCPs could be associated with an increased readmission rate due to improved access to care.(30,31) The increased availability of PCPs provides the opportunity for patients to interact with physicians more quickly and communicate more effectively; it also improves continuity of care.(32) While early treatment might promote a reduction in readmissions, PCPs may also err on the side of caution when patients express respiratory-related concerns. As a result, readmissions may increase.

On the other hand, while primary care COPD in-patient treatment seems to be a solution to reducing 30-day readmission, PCP capacity may limit the widespread implementation of this solution.(33) Multiple studies have identified current and impending shortages of primary care physicians throughout the United States and their associated impact on health outcome.(34-37) As a result, capacity limits could be true even in hospitals that have a higher PCP percentage. Additionally, the demand for PCPs continues to rise due to an aging population and an increasing need for chronic condition care management in the United States.(33)

Care fragmentation and the patient composition of the PCP’s patient panel have been associated with reduced quality of care.(38,39) These aspects suggest (1) a widespread application of PCP coordinated care for hospitalized COPD patients may not currently be feasible due to capacity constraints already affecting PCP care, and (2) even in a facility with a greater percentage of PCPs, the ability to specifically care for COPD patients may be limited by competing demands and care fragmentation.(40,41)

Ultimately, it would seem some COPD readmission-related issues are associated with coordination of or access to care. A recent commission focusing on challenges of COPD care delivery in the United States identified uncoordinated care, gaps in access, poor guideline implementation, and lack of proven methods to reduce COPD admissions as key challenges.(42) Additionally, previous studies have indicated associations between rural residency, minorities, low-income status, and COPD readmissions or mortality.(3,43,44)

Concerning access, physician shortages are an obvious limitation, but so is the availability of and access to proven services for the management of COPD and COPD symptoms. For instance, pulmonary rehabilitation (PR) has been established to be an effective alternative to traditional medication-based treatment following a diagnosis and/ or discharge from hospitalization.(45) While PR does provide a significant benefit, access to PR remains poor, with services generally being more expensive than traditional alternatives and an overall lack of providers in areas of high disease prevalence.(43)

Our findings also suggest that hospitals that are larger, part of a system, or provide teaching services are associated with reduced COPD readmissions. Previous research supported that these organizational components provide greater access to resources and promote more standardized care and, as a result, are associated with increased quality.(5,18,46) Additionally, organizations with these characteristics tend to provide improved access to care for the surrounding communities, have more physicians, and provide more external services.(6,47) These hospital characteristics may indicate additional access to services such as PR or the ability to provide greater coordination of care.

It is important to note, however, that the locations and contexts in which these organizations exist likely influence COPD readmissions.(16) Our results support these assertions; hospitals in rural areas and those in more monopolistic environments have associated increased COPD readmission rates. More specifically, while it is not clear the exact influence these factors have on patients with COPD, it is clear that differences in community risk do influence COPD rates and that the presence of hospitals influences community risk(19-22); therefore, readmission reduction interventions should incorporate specific clinical and environmental needs of the patient.

Limitations

This study has several limitations. First, we could not distinguish whether a particular PCP treated patients, nor were any specific characteristics of the PCP or their patient panel available. Additionally, it is possible that the physician’s role could influence COPD readmissions.

For example, a PCP who follows the patient both in and out of the hospital would provide more continuity of care compared to a PCP or even a hospitalist who followed the patient only within the hospital setting. Data limitations restrict the ability to define these roles, and future research should seek to more fully articulate how to separate PCP roles and patient engagement influence COPD readmissions.

Next, the ability to assess processes and outcomes of care in a retrospective study that uses administrative claims data is limited. For example, the PCP ratio is based upon reported privileged PCP and the total numbers of privileged physicians. If this data is inaccurate or only includes active rosters, the results could be biased. Similarly, the fact that the severity of COPD is not measured can affect the outcome of care.

Additionally, results are based on ICD-9 and ICD-10 coding. While it is common to use these codes for research, known issues associated with under- and over-coding by organizations may bias the results. Due to the study’s cross-sectional nature, we were unable to measure the impact of change in PCP percentage over time.

We were limited by the aggregated hospital-level data and were not able to present the outcomes of COPD patients specifically cared for by PCPs. Complex patients post-discharge are more likely to benefit from a shared model of care to improve COPD management.

In addition, the results of the study are only generally applicable to patients 65 and older. With a dataset that contains 1,527 hospitals, we did restrict the analysis to facilities that had data from each dataset used. As a result, generalizability may be reduced.

Conclusion

Reducing readmissions has become a requirement for hospitals, emphasized by public reporting and financial penalties. In summary, compared to hospitals with fewer PCPs, hospitals with a greater percentage of privileged PCPs experienced higher COPD 30-day readmissions; therefore, future research should examine how incorporating PCPs in a multifaced intervention could impact COPD readmission rates and quality of care.

References

1. Shah T, Press VG, Huisingh-Scheetz M, White SR. COPD Readmissions: Addressing COPD in the Era of Value-based Health Care. Chest. 2016; 150(4):916–926.

2. Agusti A, Calverley PM, Celli B, et al. Characterisation of COPD Heterogeneity in the ECLIPSE Cohort. Respir Res. 2010;11:122.

3. Nastars DR, Rojas JD, Ottenbacher KJ, Graham JE. Race/Ethnicity and 30-Day Readmission Rates in Medicare Beneficiaries with COPD. Respir Care. 2019;64(8):931–936.

4. Goto T, Faridi MK, Gibo K, et al. Trends in 30-day Readmission Rates After COPD Hospitalization, 2006–2012. Respir Med. 2017;130:92–97.

5. Guarascio AJ, Ray SM, Finch CK, Self TH. The Clinical and Economic Burden of Chronic Obstructive Pulmonary Disease in the USA. Clinicoecon Outcomes Res. 2013;5:235–245.

6. Hamm N, Loria K. How to Reduce Hospital Readmissions: The Ultimate Guidebook on Everything You Need To Know About Reducing Hospital Readmissions, Beginning with Heart Failure. Managed Healthcare Executive. 2019;29(9):4–7.

7. Prieto-Centurion V, Rolle AJ, Au DH, et al. Multicenter Study Comparing Case Definitions Used To Identify Patients with Chronic Obstructive Pulmonary Disease. Am J Respir Crit Care Med. 2014;190(9):989–995.

8. Van Walraven C, Bennett C, Jennings A, Austin PC, Forster AJ. Proportion of Hospital Readmissions Deemed Avoidable: A Systematic Review. CMAJ. 2011;183(7):E391–E402.

9. Feemster LC, Au DH. Penalizing Hospitals for Chronic Obstructive Pulmonary Disease Readmissions. Am J Respir Crit Care Med. 2014;189(6):634–639.

10. Hansen LO, Young RS, Hinami K, Leung A, Williams MV. Interventions to Reduce 30-day Rehospitalization: A Systematic Review. Ann Intern Med. 2011;155(8):520–528.

11. Shah T, Churpek MM, Coca Perraillon M, Konetzka RT. Understanding Why Patients With COPD Get Readmitted: A Large National Study to Delineate the Medicare Population for the Readmissions Penalty Expansion. Chest. 2015;147(5):1219–1226.

12. Auerbach AD, Kripalani S, Vasilevskis EE, et al. Preventability and Causes of Readmissions in a National Cohort of General Medicine Patients. JAMA Internal Medicine. 2016;176(4):484–493.

13. Simmering JE, Polgreen LA, Comellas AP, Cavanaugh JE, Polgreen PM. Identifying Patients with COPD at High Risk of Readmission. Chronic Obstructive Pulmonary Diseases. 2016;3(4):729.

14. Burke RE, Schnipper JL, Williams MV, et al. The HOSPITAL Score Predicts Potentially Preventable 30-Day Readmissions in Conditions Targeted by the Hospital Readmissions Reduction Program. Medical Care. 2017;55(3):285–290.

15. Johnston J, Longman J, Ewald D, King J, Das S, Passey M. Study of Potentially Preventable Hospitalisations (PPH) for Chronic Conditions: What Proportion Are Preventable and What Factors Are Associated with Preventable PPH? BMJ Open. 2020;10(11):e038415.

16. Garvey C, Kaplan AG. Optimizing Adherence to Improve Clinical Outcomes in Patients with Chronic Obstructive Pulmonary Disease. J Fam Pract. 2018;67(10 Suppl):S40–S45.

17. American Hospital Association. 2019 AHA Annual Survey https://s3-us-west-2.amazonaws.com/aha-report-download/3373775721/2019%20AHA%20Annual%20Survey_FINAL.pdf . Published 2021. Accessed August 23, 2021.

18. Foster D, Zrull L. 100 Top Hospitals: Hospital Performance Differences by Size and Teaching Status. Truven Health Analytics. 2013.

19. Sosunov EA, Egorova NN, Lin HM, et al. The Impact of Hospital Size on CMS Hospital Profiling. Med Care. 2016;54(4):373–379.

20. Hamadi H, Apatu E, Spaulding A. Does Hospital Ownership Influence Hospital Referral Region Health Rankings in the United States. Int J Health Plann Manage. 2018;33(1):e168–e180.

21. Chen AS, Revere L, Ratanatawan A, Beck CL, Allo JA. A Comparative Analysis of Academic and Nonacademic Hospitals on Outcome Measures and Patient Satisfaction. Am J Med Qual. 2018:1062860618800586.

22. Cuellar AE, Gertler PJ. How The Expansion of Hospital Systems Has Affected Consumers. Health Affairs. 2005;24(1):213–219.

23. Bazzoli GJ, Chen HF, Zhao M, Lindrooth RC. Hospital Financial Condition and the Quality of Patient Care. Health Econ. 2008;17(8):977–995.

24. Younis MZ. A Comparison Study of Urban and Small Rural Hospitals’ Financial and Economic Performance. Online Journal of Rural Nursing and Health Care. 2012;3(1):38–48.

25. Lakomkin N, Graffeo CS, Hadjipanayis CG. Specific Causes and Predictors of Readmissions Following Acute and Chronic Subdural Hematoma Evacuation. J Clin Neurosci. 2020;75:35–39.

26. Hamadi H, Spaulding A, Haley DR, Zhao M, Tafili A, Zakari N. Does Value-Based Purchasing Affect US Hospital Utilization Pattern: A Comparative Study. Int J Health Manage. 2017:1–7.

27. Wong HS, Zhan C, Mutter R. Do Different Measures of Hospital Competition Matter in Empirical Investigations of Hospital Behavior. Rev Ind Organ. 2005;26(1):27–60.

28. Wooldridge JM. Econometric Analysis of Cross Section and Panel Data. Boston:MIT Press; 2010.

29. Ialongo C. Understanding the Effect Size and Its Measures. Biochem Med (Zagreb). 2016;26(2):150–163.

30. Herrin J, St Andre J, Kenward K, Joshi MS, Audet A-MJ, Hines SC. Community Factors and Hospital Readmission Rates. Health Serv Res. 2015;50(1):20–39.

31. Weinberger M, Oddone EZ, Henderson WG. Does Increased Access to Primary Care Reduce Hospital Readmissions? NEJM. 1996;334(22):1441–1447.

32. Bodenheimer T, Pham HH. Primary Care: Current Problems and Proposed Solutions. Health Aff. 2010;29(5):799–805.

33. Agarwal A, Zhang W, Kuo Y, Sharma G. Process and Outcome Measures among COPD Patients with a Hospitalization Cared for by an Advance Practice Provider or Primary Care Physician. PLoS One. 2016;11(2):e0148522.

34. Basu S, Berkowitz SA, Phillips RL, Bitton A, Landon BE, Phillips RS. Association of Primary Care Physician Supply with Population Mortality in the United States, 2005–2015. JAMA Internal Medicine. 2019;179(4):506–514.

35. Gemelas JC. Post-ACA Trends in the US Primary Care Physician Shortage with Index of Relative Rurality. J Rural Health. 2021;37(4):700–704.

36. MacArthur KR, Royer E, Hawkins DN. Addressing the Enduring Primary Care Physician Shortage in The United States: The Direct and Indirect Effects of Gender on the Medical Specialty Decision-Making Process. Int J Med Students. 2018;6(3):91–97.

37. Zhang X, Lin D, Pforsich H, Lin VW. Physician Workforce in the United States of America: Forecasting Nationwide Shortages. Human Resources for Health. 2020;18(1):8.

38. Frandsen BR, Joynt KE, Rebitzer JB, Jha AK. Care Fragmentation, Quality, and Costs Among Chronically Ill Patients. Am J Manag Care. 2015;21(5):355–362.

39. Hong CS, Atlas SJ, Chang Y, et al. Relationship Between Patient Panel Characteristics and Primary Care Physician Clinical Performance Rankings. JAMA. 2010;304(10):1107–1113.

40. Bodenheimer TS, Smith MD. Primary Care: Proposed Solutions to the Physician Shortage Without Training More Physicians. Health Aff. 2013;32(11):1881–1886.

41. Holman GT, Beasley JW, Karsh B-T, Stone JA, Smith PD, Wetterneck TB. The Myth of Standardized Workflow in Primary Care. J Am Med Inform Assoc. 2015;23(1):29–37.

42. Fidahussein SS, Croghan IT, Cha SS, Klocke DL. Posthospital Follow-Up Visits and 30-Day Readmission Rates in Chronic Obstructive Pulmonary Disease. Risk Manag Healthc Policy. 2014;7:105–112.

43. Bhatt SP, Patel SB, Anderson EM, et al. Video Telehealth Pulmonary Rehabilitation Intervention in Chronic Obstructive Pulmonary Disease Reduces 30-Day Readmissions. Am J Respir Crit Care Med. 2019;200(4):511–513.

44. Wier LM, Elixhauser A, Pfuntner A, Au DH. Overview of Hospitalizations Among Patients with COPD, 2008: Statistical Brief# 106. 2011.

45. Spruit MA, Singh SJ, Garvey C, et al. An Official American Thoracic Society/European Respiratory Society Statement: Key Concepts and Advances in Pulmonary Rehabilitation. Am J Respir Crit Care Med. 2013;188(8):e13–64.

46. Foster JA, Yawn BP, Maziar A, Jenkins T, Rennard SI, Casebeer L. Enhancing COPD Management in Primary Care Settings. Medscape General Medicine. 2007;9(3):24.

47. Skolnik N, Kaplan AG, Lake T, Hayden ML, Hanania NA. COPD Management in the Primary Care Setting. J Fam Pract. 2018;67(10 Suppl):S27–S32.