Abstract
Heart failure (HF) is the leading cause for 30-day all-cause readmission in older Medicare beneficiaries and 30-day all-cause readmission is associated with a higher risk of mortality. In the current analysis we examined if that association varied by timing of 30-day all-cause readmission. Of the 8049 Medicare beneficiaries hospitalized for HF, 1688 had 30-day all-cause readmissions, of whom 1519 were alive at 30 days. Of these, 626 (41%) had early (first 10 days) 30-day readmission. Propensity scores for early 30-day readmission, estimated for all 1519 patients, were used to assemble a matched cohort of 596 pairs of patients with early versus late (after the first 10 days) all-cause readmission, balanced on 34 baseline characteristics. Two-year all-cause mortality occurred in 51% and 57% of matched patients with early vs. late 30-day all-cause readmissions, respectively (hazard ratio {HR} associated with early 30-day readmission, 1.21; 95% CI, 1.04–1.42; p=0.014). This association was not observed in the subset of 436 patients whose 30-day all-cause readmission was due to HF (HR, 1.01; 95% CI, 0.79–1.28; p=0.963), but was observed in the subset of 756 patients whose 30-day all-cause readmission was not due to HF (HR, 1.37; 95% CI, 1.12–1.67; p=0.002; p for interaction, 0.057). In conclusion, in a high-risk subset of older hospitalized HF patients readmitted within 30 days, readmission within 11 to 30 (vs. 1–10) days was associated with a higher risk of death and this association appeared to be more pronounced among those readmitted for non-HF related reasons.
Keywords: 1-10-Day All-Cause Readmission, 11-30-Day All-Cause Readmission, Heart Failure, All-Cause Mortality
Heart failure (HF) is the leading cause of 30-day all-cause readmission, which has been identified in the Affordable Care Act as a target for reduction of Medicare cost.1,2 Although the clinical relevance of the cost-driven metric of short-term readmission at 30 days has been questioned,3 it has been shown to be associated with a significantly higher risk of subsequent mortality.4 However, it remains unclear whether timing of 30-day readmission is associated with mortality. It has been suggested that early 30-day readmissions are potentially more preventable and thus markers of poor quality of care.5 For hospitalized HF patients, it is also not clear if the association between an early 30-day readmissions and mortality would vary by the cause of 30-day readmission. The objective of our study was to examine the association between early vs. late 30-day readmission, and if this association was modified by cause of 30-day readmission.
Methods
We used data from the Alabama Heart Failure Project, the details of which have been previously described.6 Briefly, the cohort included 8555 Medicare beneficiaries hospitalized for acute HF in 106 Alabama hospitals during 1998–2001, of whom 8049 were discharged alive. Of these, 1688 (21%) patients had a 30-day all-cause readmission, of whom 1519 were alive at 30 days. Of these, 626 (41%) were readmitted within 10 days. We matched 596 these patients with 596 patients who were readmitted during 11 to 30 days, but had similar propensity scores for 10-day readmission, thus assembling a matched cohort of 1192 patients (Figure 1).7–12 Post-match balance on measured baseline characteristics was checked by estimating absolute standardized difference.13–16 Data on death and time to death were obtained from Medicare administrative data.
Figure 1.
Flow chart displaying assembly of matched cohort of patients with heart failure (HF) who had a 30-day all-cause readmission
Descriptive analyses comparing baseline characteristics between the two groups were performed using Pearson’s Chi-square and Wilcoxon rank-sum tests as appropriate. Cox regression models were used to estimate hazard ratios (HR) and 95% confidence intervals (CI) associated with 10-day readmission (versus 11–30-day readmission) among matched patients. Kaplan-Meier survival analyses were used to generate plots for all-cause mortality during about 2 years of follow-up, by time and cause of 30-day all-cause readmission. To assess nonlinearity in the relationship between days to readmission as a continuous variable and all-cause 2-year mortality, we fitted restricted cubic spline models with 4 knots at 5, 10 (reference), 20 and 25 days, in the matched data as well as in the pre-match data adjusting for propensity scores. Overall, 548 (36%) patients were readmitted for HF. We conducted a subgroup analysis to examine if the association between early 30-day readmission and mortality varied between patients readmitted for HF vs. for other reasons. All statistical tests were two-tailed with a p-value <0.05 considered significant. IBM SPSS Statistics for Windows, Version 23.0. Armonk, NY: IBM Corp. and SAS version 9.4 (SAS Institute Inc., Cary, NC, USA) were used for data analyses.
Results
The 1192 matched patients had a mean age of 75 years, 56% were women and 23% were African American. Except for rural hospital, before matching patients were balanced on all key measured baseline characteristics. All 34 measured baseline characteristics were balanced in the matched cohort (Table 1). Two-year all-cause mortality occurred in 51% and 57% of matched patients with early vs. late 30-day all-cause readmissions, respectively (hazard ratio {HR} associated with early 30-day readmission, 1.22; 95% CI, 1.04–1.40; p=0.014; Table 2 and Figure 2). Our restricted cubic spline analysis showed no evidence of a nonlinear relationship between days to readmission and all-cause mortality for either the pre-match data or the matched data (p > 0.20 for test for non-linearity, for both; Figure 3).
Table 1.
Baseline characteristics of patients with heart failure who had a 30-day all-cause readmission, by timing of readmission
| Before propensity score matching | After propensity score matching | |||||
|---|---|---|---|---|---|---|
| Variable | 30-day readmission timing (n=1519) |
30-day readmission timing (n=1192) |
||||
| 1–10 days (n=626) |
11–30 days (n=893) |
p-value | 1–10 days (n=596) |
11–30 days (n=596) |
p-value | |
| Age (years) | 75 (±11) | 74 (±11) | 0.106 | 75 (±11) | 75 (±11) | 0.700 |
| Women | 350 (56%) | 500 (56%) | 0.975 | 334 (56%) | 338 (57%) | 0.815 |
| Black | 145 (23%) | 211 (24%) | 0.833 | 140 (23%) | 133 (22%) | 0.629 |
| Admission from nursing home | 45 (7%) | 47 (5%) | 0.122 | 32 (5%) | 34 (6%) | 0.800 |
| Smoking history | 72 (12%) | 122 (14%) | 0.214 | 70 (12%) | 70 (12%) | 1.000 |
| Left ventricular ejection fraction | ||||||
| <45% | 236 (38%) | 337 (38%) | 225 (38%) | 222 (37%) | ||
| ≥45% | 182(29%) | 272 (30%) | 0.792 | 175 (29%) | 172 (29%) | 0.934 |
| Unknown | 208 (33%) | 284 (32%) | 196 (33%) | 202 (34%) | ||
| Heart failure | 478 (76%) | 677 (76%) | 0.806 | 454 (76%) | 452 (76%) | 0.892 |
| Hypertension | 435 (70%) | 644 (72%) | 0.266 | 419(70%) | 415 (70%) | 0.800 |
| Coronary artery disease | 387 (62%) | 530 (59%) | 0.333 | 367 (62%) | 364 (61%) | 0.858 |
| Atrial fibrillation | 167 (27%) | 257 (29%) | 0.369 | 162 (27%) | 153 (26%) | 0.554 |
| Left bundle branch block | 80 (13%) | 113 (13%) | 0.942 | 76 (13%) | 80 (13%) | 0.731 |
| Diabetes mellitus | 290 (46%) | 436 (49%) | 0.337 | 281 (47%) | 279 (47%) | 0.908 |
| Stroke | 139 (22%) | 191 (21%) | 0.704 | 130 (22%) | 125 (21%) | 0.724 |
| Chronic obstructive pulmonary disease | 252 (40%) | 329 (37%) | 0.178 | 235 (39%) | 231 (39%) | 0.812 |
| Dementia | 50 (8%) | 68 (8%) | 0.790 | 46 (8%) | 45 (8%) | 0.913 |
| Cancer | 20 (3%) | 25 (3%) | 0.655 | 18 (3%) | 16 (3%) | 0.728 |
| Pulse (beats per minute) | 88 (±22) | 89 (±22) | 0.297 | 89 (±22) | 88 (±22) | 0.431 |
| Systolic blood pressure (mmHg) | 147 (±33) | 149 (±33) | 0.427 | 148 (±33) | 150 (±32) | 0.196 |
| Pulmonary edema by chest x-ray | 423 (68%) | 617 (69%) | 0.530 | 401 (67%) | 404 (68%) | 0.853 |
| Serum creatinine (mEq/L) | 1.7 (±1.2) | 1.8 (±1.5) | 0.193 | 1.7 (±1.3) | 1.7 (±1.4) | 0.957 |
| In hospital events | ||||||
| Pneumonia | 173 (28%) | 232 (26%) | 0.473 | 160 (27%) | 157 (26%) | 0.844 |
| Acute myocardial infarction | 39 (6%) | 47 (5%) | 0.422 | 32 (5%) | 38 (6%) | 0.460 |
| Pressure ulcer | 47 (8%) | 87 (10%) | 0.131 | 47 (8%) | 45 (8%) | 0.828 |
| Hospital and care characteristics | ||||||
| Rural hospital | 225 (36%) | 272 (30%) | 0.022 | 205 (34%) | 202 (34%) | 0.855 |
| Cardiology care | 327 (52%) | 465 (52%) | 0.950 | 313 (53%) | 309 (52%) | 0.817 |
| Intensive care | 31 (5%) | 56 (6%) | 0.276 | 31 (5%) | 29 (5%) | 0.791 |
| In hospital length of stay | 7.0 (±5.1%) | 7.2 (±5.3) | 0.448 | 7.1 (±5.2) | 6.8 (±5.6) | 0.245 |
| Discharge medications | ||||||
| ACE inhibitors or ARBs | 333 (53%) | 485 (54%) | 0.667 | 323 (54%) | 322 (54%) | 0.954 |
| Beta blockers | 187 (30%) | 268 (30%) | 0.954 | 179 (30%) | 178(30%) | 0.950 |
| Loop diuretics | 487 (78%) | 713 (80%) | 0.335 | 471 (79%) | 471 (79%) | 1.000 |
| Digoxin | 261 (42%) | 362 (41%) | 0.652 | 249 (42%) | 252 (42%) | 0.860 |
| Potassium sparing diuretics | 98 (16%) | 131 (15%) | 0.597 | 89 (15%) | 94 (16%) | 0.688 |
| Potassium supplements | 274 (44%) | 374 (42%) | 0.464 | 259 (43%) | 247 (41%) | 0.482 |
| Opioids | 53 (6%) | 40 (6%) | 0.716 | 37 (6%) | 37 (6%) | 1.000 |
Table 2.
All-cause mortality in 1192 propensity score-matched patients with heart failure who had a 30-day all-cause readmission, by early (1–10 days) versus late (11–30 days) readmission
| Mortality | Events (%) | Hazard ratio* (95% CI); p-value | |
|---|---|---|---|
| 1–10 days (n=596) | 11–30 days (n=596) | ||
| 1-year | 218 (37%) | 259 (44%) | 1.25 (1.05–1.49); p=0.013 |
| 2-year | 303 (51%) | 342 (57%) | 1.22 (1.04–1.40); p=0.014 |
Hazard ratios when 1-to-10-day readmission was compared with 11-to-30-day readmission
Figure 2.
Kaplan Meier plot for all-cause mortality in a propensity-matched cohort of 1192 older patients with heart failure with a 30-day all-cause readmission, by timing of readmission (HR=hazard ratio; CI=confidence interval)
Figure 3.
Hazard ratios (HR) and 95% confidence intervals (CI) for 2-year all-cause mortality by days to 30-day all-cause readmission according to restricted cubic spline regression models using four knots at 5, 10 (reference), 20 and 25 days. Solid black lines represent HRs and shaded areas represent 95% CIs. Plots on the left panel (A) are based on 1519 pre-match patients, adjusting for propensity scores, and those on the right panel (B) are based on 1192 propensity score-matched patients balanced on 34 baseline characteristics
Among the subset of 436 patients whose 30-day all-cause readmission was due to HF, two-year all-cause mortality occurred in 60% and 59% of patients with early vs. late 30-day all-cause readmissions, respectively (HR associated with early readmission, 1.01; 95% CI, 0.79–1.32; p=0.963). Among the subset of 756 patients whose 30-day all-cause readmission was not due to HF, two-year all-cause mortality occurred in 45% and 56% of patients with early vs. late 30-day all-cause readmissions, respectively (HR associated with early readmission, 1.37; 95% CI, 1.11–1.67; p=0.002). This differential association between 11–30-day readmission and 2-year mortality between patients with HF and non-HF readmission was of borderline significance (p=0.057).
Discussion
Findings from the current study demonstrate that among older patients hospitalized for HF who had a readmission within 30 days after discharge from the index hospitalization, readmission during days 11–30 was independently associated with a significantly higher risk of all-cause mortality at 2 years. We also observe that this association between early readmission and 2-year mortality appeared to be more pronounced among the subset of patients whose 30-day readmission were due to reasons other than HF. To the best of our knowledge this is the first study to examine the differential association of timing of 30-day all-cause readmission on mortality in hospitalized older patients with HF. These findings suggest that timing of readmissions may be used as a marker to further risk-stratify HF patients who are readmitted within 30 days of discharge.
It is tempting to speculate that HF patients who are readmitted within the first 10 days after hospital discharge had suboptimal inpatient HF care and early decompensation. It is possible that more immediate readmissions are due to reasons that are more readily preventable and manageable, such as discharge without achieving euvolemia and/or with inadequate dose of diuretics.5 However, an early 30-day readmission may also be a marker of a higher socioeconomic status and/or health literacy as these patients may recognize early warning symptoms. They are also more likely to better comprehend and adhere to discharge instructions about diet and medications and have timely outpatient follow-up. Interestingly, among patients whose 30-day readmission was due to HF, timing of 30-day readmission had no association with outcomes. It is not clear why among patients whose 30-day readmission was not due to HF, those readmitted during days 11–30 (vs. the first 10 days) would have worse outcomes.
Hospitalized older adults with HF who are readmitted within 30 days of discharge constitute a high-risk subset with poor prognosis which include a higher rate of all-cause mortality, subsequent readmission, and greater long-term cost.4 Findings from our current study suggest that timing and cause of 30-day all-cause readmission may identify those even at a higher risk among this already high-risk subset. If these findings can be replicated, these metrics could be used to further risk stratify patients with HF who have a 30-day all-cause readmission. Future studies need to develop and test interventions that may improve outcomes in these patients.
There are several limitations of our study. As in any observational study, findings of our study are subject to bias and despite our use of propensity score matching, residual bias or bias due to an unmeasured confounder is possible. Importantly, we had no markers on frailty including body mass index. Our analysis was restricted to fee-for-service Medicare beneficiaries from a single state from an earlier era of heart failure management and may not be generalizable. In conclusion, among hospitalized patients with HF who had a 30-day all-cause readmission, a late readmission after the first 10 days, is associated with a higher risk of all-cause mortality, especially among patients whose 30-day readmissions are not due to HF.
Funding:
Dr. Ahmed was in part supported by the National Institutes of Health through grants (R01-HL085561, R01-HL085561-S and R01-HL097047) from the National Heart, Lung, and Blood Institute.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of interest: None of the authors reported conflicts of interest related to this work. Dr. Deepak L. Bhatt discloses the following relationships - Advisory Board: Cardax, Elsevier Practice Update Cardiology, Medscape Cardiology, Regado Biosciences; Board of Directors: Boston VA Research Institute, Society of Cardiovascular Patient Care, TobeSoft; Chair: American Heart Association Quality Oversight Committee; Data Monitoring Committees: Baim Institute for Clinical Research (formerly Harvard Clinical Research Institute, for the PORTICO trial, funded by St. Jude Medical, now Abbott), Cleveland Clinic, Duke Clinical Research Institute, Mayo Clinic, Mount Sinai School of Medicine (for the ENVISAGE trial, funded by Daiichi Sankyo), Population Health Research Institute; Honoraria: American College of Cardiology (Senior Associate Editor, Clinical Trials and News, ACC.org; Vice-Chair, ACC Accreditation Committee), Baim Institute for Clinical Research (formerly Harvard Clinical Research Institute; RE-DUAL PCI clinical trial steering committee funded by Boehringer Ingelheim), Belvoir Publications (Editor in Chief, Harvard Heart Letter), Duke Clinical Research Institute (clinical trial steering committees), HMP Global (Editor in Chief, Journal of Invasive Cardiology), Journal of the American College of Cardiology (Guest Editor; Associate Editor), Population Health Research Institute (for the COMPASS operations committee, publications committee, steering committee, and USA national co-leader, funded by Bayer), Slack Publications (Chief Medical Editor, Cardiology Today’s Intervention), Society of Cardiovascular Patient Care (Secretary/Treasurer), WebMD (CME steering committees); Other: Clinical Cardiology (Deputy Editor), NCDR-ACTION Registry Steering Committee (Chair), VA CART Research and Publications Committee (Chair); Research Funding: Abbott, Amarin, Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Chiesi, Eisai, Ethicon, Forest Laboratories, Idorsia, Ironwood, Ischemix, Lilly, Medtronic, PhaseBio, Pfizer, Regeneron, Roche, Sanofi Aventis, Synaptic, The Medicines Company; Royalties: Elsevier (Editor, Cardiovascular Intervention: A Companion to Braunwald’s Heart Disease); Site Co-Investigator: Biotronik, Boston Scientific, St. Jude Medical (now Abbott), Svelte; Trustee: American College of Cardiology; Unfunded Research: FlowCo, Merck, Novo Nordisk, PLx Pharma, Takeda.
References
- 1.Jencks SF, Williams MV, Coleman EA. Rehospitalizations among patients in the Medicare fee-for-service program. N Engl J Med 2009;360:1418–1428. [DOI] [PubMed] [Google Scholar]
- 2.Rau J Hospitals Face Pressure to Avert Readmissions The New York Times November 16, 2012. http://www.nytimes.com/2012/11/27/health/hospitals-face-pressure-from-medicare-to-avert-readmissions.html. Accessed July 10, 2016.
- 3.Konstam MA, Upshaw J. Sisyphus and 30-Day Heart Failure Readmissions: Futility in Predicting a Flawed Outcome Metric. JACC Heart Fail 2016;4:21–23. [DOI] [PubMed] [Google Scholar]
- 4.Arundel C, Lam PH, Khosla R, Blackman MR, Fonarow GC, Morgan C, Zeng Q, Fletcher RD, Butler J, Wu WC, Deedwania P, Love TE, White M, Aronow WS, Anker SD, Allman RM, Ahmed A. Association of 30-Day All-Cause Readmission with Long-Term Outcomes in Hospitalized Older Medicare Beneficiaries with Heart Failure. Am J Med 2016;129:1178–1184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Graham KL, Dike O, Doctoroff L, Jupiter M, Vanka A, Davis RB, Marcantonio ER. Preventability of early vs. late readmissions in an academic medical center. PLoS One 2017;12:e0178718. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Feller MA, Mujib M, Zhang Y, Ekundayo OJ, Aban IB, Fonarow GC, Allman RM, Ahmed A. Baseline characteristics, quality of care, and outcomes of younger and older Medicare beneficiaries hospitalized with heart failure: findings from the Alabama Heart Failure Project. Int J Cardiol 2012;162:39–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Mujib M, Rahman AA, Desai RV, Ahmed MI, Feller MA, Aban I, Love TE, White M, Deedwania P, Aronow WS, Fonarow G, Ahmed A. Warfarin use and outcomes in patients with advanced chronic systolic heart failure without atrial fibrillation, prior thromboembolic events, or prosthetic valves. Am J Cardiol 2011;107:552–557. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Roy B, Desai RV, Mujib M, Epstein AE, Zhang Y, Guichard J, Jones LG, Feller MA, Ahmed MI, Aban IB, Love TE, Levesque R, White M, Aronow WS, Fonarow GC, Ahmed A. Effect of warfarin on outcomes in septuagenarian patients with atrial fibrillation. Am J Cardiol 2012;109:370–377. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Inampudi C, Parvataneni S, Morgan CJ, Deedwania P, Fonarow GC, Sanders PW, Prabhu SD, Butler J, Forman DE, Aronow WS, Allman RM, Ahmed A. Spironolactone use and higher hospital readmission for Medicare beneficiaries with heart failure, left ventricular ejection fraction <45%, and estimated glomerular filtration rate <45 ml/min/1.73 m(2.). Am J Cardiol 2014;114:79–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Filippatos GS, Adamopoulos C, Sui X, Love TE, Pullicino PM, Lubsen J, Bakris G, Anker SD, Howard G, Kremastinos DT, Ahmed A. A propensity-matched study of hypertension and increased stroke-related hospitalization in chronic heart failure. Am J Cardiol 2008;101:1772–1776. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Aronow WS, Ahmed MI, Ekundayo OJ, Allman RM, Ahmed A. A propensity-matched study of the association of peripheral arterial disease with cardiovascular outcomes in community-dwelling older adults. Am J Cardiol 2009;103:130–135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Desai RV, Banach M, Ahmed MI, Mujib M, Aban I, Love TE, White M, Fonarow G, Deedwania P, Aronow WS, Ahmed A. Impact of baseline systolic blood pressure on long-term outcomes in patients with advanced chronic systolic heart failure (insights from the BEST trial). Am J Cardiol 2010;106:221–227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Tsimploulis A, Lam PH, Arundel C, Singh SN, Morgan CJ, Faselis C, Deedwania P, Butler J, Aronow WS, Yancy CW, Fonarow GC, Ahmed A. Systolic Blood Pressure and Outcomes in Patients With Heart Failure With Preserved Ejection Fraction. JAMA Cardiol 2018;3:288–297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lam PH, Gupta N, Dooley DJ, Singh S, Deedwania P, Zile MR, Bhatt DL, Morgan CJ, Pitt B, Fonarow GC, Ahmed A. Role of High-Dose Beta-Blockers in Patients with Heart Failure with Preserved Ejection Fraction and Elevated Heart Rate. Am J Med 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lam PH, Bhyan P, Arundel C, Dooley DJ, Sheriff HM, Mohammed SF, Fonarow GC, Morgan CJ, Aronow WS, Allman RM, Waagstein F, Ahmed A. Digoxin use and lower risk of 30-day all-cause readmission in older patients with heart failure and reduced ejection fraction receiving beta-blockers. Clin Cardiol 2018;41:406–412. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Bayoumi E, Lam PH, Dooley DJ, Singh S, Faselis C, Morgan CJ, Patel S, Sheriff HM, Mohammed SF, Palant CE, Pitt B, Fonarow GC, Ahmed A. Spironolactone and Outcomes in Older Patients with Heart Failure and Reduced Ejection Fraction. Am J Med Sep 19 pii: S0002–9343(18)30878–7 doi: 101016/jamjmed201809011 [Epub ahead of print] 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]