Literature Review 7 pages

Journal of the American Heart Association

J Am Heart Assoc. 2020;9:e014907. DOI: 10.1161/JAHA.119.014907 1

ORIGINAL RESEARCH

Association Between Subsequent Hospitalizations and Recurrent Acute Myocardial Infarction Within 1 Year After Acute Myocardial Infarction Yun Wang, PhD; Erica Leifheit, PhD; Sharon-Lise T. Normand, PhD; Harlan M. Krumholz , MD, SM

BACKGROUND: Patients who survive acute myocardial infarction (AMI) are at high risk for recurrence. We determined whether rehospitalizations after AMI further increased risk of recurrent AMI.

METHODS AND RESULTS: The study included Medicare fee- for- service patients aged ≥65 years discharged alive after AMI from acute- care hospitals in fiscal years 2009–2014. The outcome was recurrent AMI within 1 year of the index AMI. The Clinical Classifications Software (CCS) was used to classify rehospitalizations into disease categories. A Cox regression model was fit accounting for CCS- specific hospitalizations as time- varying variables and patient characteristics at discharge for the index AMI, adjusting for the competing risk of death. The rate of 1- year recurrent AMI was 5.3% (95% CI, 5.27%–5.41%), and median (interquartile range) time from discharge to recurrent AMI was 115 (34–230) days. Eleven disease categories (diabetes mellitus, anemia, hypertension, coronary atherosclerosis, chest pain, heart failure, pneumonia, chronic obstructive pulmonary disease, gastrointestinal hemorrhage, renal failure, complication of implant or graft) were associated with increased risk of recurrent AMI. Septicemia was associated with lower recurrence risk. Hazard ratios ranged from 1.6 (95% CI, 1.55–1.70, heart failure) to 1.1 (95% CI, 1.04–1.25, pneumonia) to 0.6 (95% CI, 0.58–0.71, septicemia).

CONCLUSIONS: Patient risk of recurrent AMI changed based on the occurrence of hospitalizations after the index AMI. Improving post–acute care to prevent unplanned rehospitalizations, especially rehospitalizations for chronic diseases, and extending the focus of outcomes measures to condition- specific rehospitalizations within 30 days and beyond is important for the second- ary prevention of AMI.

Key Words: cardiovascular prevention ■ myocardial infarction ■ rehospitalization

D espite the marked decline in the rate of recurrent AMI over recent decades,1–5 recurrence remains a significant threat to AMI survivors. Classifying

patients according to their risk of recurrent AMI may be helpful in efforts to prevent the next AMI. Although patient baseline characteristics, postdischarge life- style, quality of care, and medication adherence are associated with risk of recurrent AMI,6–10 subsequent hospitalizations after the initial AMI may also increase the risk of recurrence and, therefore, may be important

when considering a patient’s risk. Identifying subse- quent hospitalizations associated with recurrent AMI could better inform patients, families, and physicians about any further increases in the risk of recurrent AMI and help ensure intensive follow- up and risk modifica- tion behaviors.

Efforts to identify risk markers and risk factors for long- term outcomes after AMI have focused on patient char- acteristics available at the time of the initial hospitalization for AMI.8,11,12 Identification of postdischarge events as

Correspondence to: Harlan M. Krumholz, MD, SM, 1 Church Street, Suite 200, New Haven, CT, 06510. E-mail: [email protected]

Supplementary material for this article is available at https://www.ahajo urnals.org/doi/suppl/ 10.1161/JAHA.119.014907

For Sources of Funding and Disclosures, see page 12.

© 2020 The Authors. Published on behalf of the American Heart Association, Inc., by Wiley. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non- commercial and no modifications or adaptations are made.

JAHA is available at: www.ahajournals.org/journal/jaha

D ow

nloaded from http://ahajournals.org by on June 13, 2020

J Am Heart Assoc. 2020;9:e014907. DOI: 10.1161/JAHA.119.014907 2

Wang et al Rehospitalization Associated with Recurrent AMI

risk markers for adverse outcomes is limited, and many of the available data focus on mortality.13–16 There is scant information on the association between subsequent hos- pitalizations and recurrent AMI. A comprehensive, con- temporary, national evaluation of such rehospitalizations could provide important information for the prevention of recurrent AMI, particularly among Medicare beneficia- ries who are a high- risk population for AMI.

Accordingly, we used national Medicare inpatient claims data to assess the association between subse- quent hospitalizations and recurrent AMI within 1 year after an initial AMI and identify clinically important hos- pitalizations that increased the risk of recurrence. This study, which was based on 100% national data and detailed follow- up information about patients with AMI, is ideally positioned to generate information to update risk stratification for recurrent AMI in the year after hos- pital discharge.

METHODS Restricted by our Data Use Agreement with the Centers for Medicare & Medicaid Services (CMS), the Medicare data used for this study cannot be made publicly available to other researchers for purposes of reproducing the results or replicating the procedure. However, Medicare data are available from the Centers for Medicare & Medicaid Services upon request (https://www.cms.gov/Resea rch-Stati stics-Data-and- Syste ms/Files-for-Order/ Data-Discl osures-Data-Agree ments/ DUA_-_Forms.html).

Study Sample We used the Centers for Medicare & Medicaid Services Medicare denominator files to identify all beneficiaries aged 65  years or older enrolled in the fee- for- service program for at least 12 months in fiscal years (FY) 2009–2014 (October 1, 2008 to September 31, 2014), a period in which all diagnosis codes were classified according to the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD- 9- CM). We linked these enrollment data to Medicare fee- for- service inpatient claims to identify beneficiar- ies who were discharged alive after hospitalization for AMI at an acute- care hospital in the United States. This was designated the index AMI hospitalization. If a patient had >1 AMI hospitalization during the study period, we selected the first AMI during the study pe- riod as the index AMI. Data from FY 2008 were used to identify patients who were rehospitalized with AMI in FY 2009; FY 2015 data were used to ensure 1 year of follow- up for patients hospitalized with AMI during FY 2014.

AMI was defined as an ICD- 9- CM principal dis- charge diagnosis code of 410.xx. We excluded pa- tients with ICD- 9- CM codes 410.x2 because the codes represent subsequent episodes of care related to the index AMI. We also excluded patients who had a length of stay ≤1 day (because these patients were unlikely to have had an AMI), had conflicting dates of death and hospitalization, or were subsequently transferred to an- other acute- care hospital for continuing care after the initial AMI.

CLINICAL PERSPECTIVE

What Is New? • This study addresses how subsequent hospi-

talizations after an acute myocardial infarction may influence the risk of a recurrent acute myo- cardial infarction.

• We show that many types of subsequent hos- pitalizations can increase the risk of a recurrent acute myocardial infarction.

• We are introducing the idea of dynamic risk pre- diction, in which subsequent events influence the likelihood of a future cardiovascular event.

What Are the Clinical Implications? • Risk-stratification after an acute myocardial

infarction is necessary to inform the choice of clinical strategies.

• We are showing that risk prediction should be updated as new information becomes available. In this case, subsequent hospitalizations modify risk.

• This study indicates the need to develop dy- namic risk calculators that can be updated over time.

Nonstandard Abbreviations and Acronyms

AMI acute myocardial infarction CABG coronary artery bypass grafting CAD coronary artery disease CATH cardiac catheterization CCS Clinical Classifications Software COPD chronic obstructive pulmonary disease DM diabetes mellitus FY fiscal years HR hazard ratio HTN hypertension ICD- 9- CM International Classification of Diseases,

Ninth Revision, Clinical Modification IQR interquartile range PCI percutaneous coronary intervention UTI urinary tract infection

D ow

nloaded from http://ahajournals.org by on June 13, 2020

J Am Heart Assoc. 2020;9:e014907. DOI: 10.1161/JAHA.119.014907 3

Wang et al Rehospitalization Associated with Recurrent AMI

Patient Baseline Characteristics Patient baseline characteristics included age (con- tinuous), sex, race (white, black, other), and clinical comorbidities identified using the method employed by the Centers for Medicare & Medicaid Services to profile hospital 30- day mortality measures for AMI.17 We determined comorbidities from secondary diag- nosis codes for the index AMI hospitalization as well as the principal and secondary diagnosis codes from all hospitalizations during the 12  months before the index AMI. Because the maximum number of diagno- sis codes in Medicare data increased from 10 to 25 in 2011,18 we restricted the 2011–2015 data to the first 10 diagnosis codes to calculate comorbidities.

Outcome The primary outcome was recurrent AMI within 1 year of discharge for the initial AMI. For patients with >1 re- current AMI, the first recurrence was selected. Deaths during the 1- year follow- up period without a recurrent AMI hospitalization were treated as competing risks in the analysis. Secondary outcomes included 30- day all- cause mortality, 30- day all- cause readmission, and 1- year all- cause mortality using the index AMI discharge as the time zero. Mean length of stay and mean Medicare payment for the index AMI hospitaliza- tion were also assessed.

Subsequent Hospitalizations We identified all subsequent hospitalizations within 1  year after discharge for the initial AMI. For patients with a recurrent AMI, subsequent hospitalizations were restricted to the period prior to the recurrent AMI. Because of the large volume of individual ICD- 9- CM codes, we used the Clinical Classifications Software (CCS),19 a diagnosis and procedure categorization algorithm developed by the Agency for Healthcare Research and Quality, to characterize the subsequent hospitalizations. Using the CCS single- level diagnosis- specific algorithm, we collapsed >14  000 individual ICD- 9- CM principal discharge diagnosis codes into 285 clinically homogeneous, meaningful, and mutu- ally exclusive disease categories (Table S1). If a patient had >1 hospitalization for the same disease category, the first one was selected. We excluded CCS hospi- talizations that occurred at a frequency <1% to avoid counting hospitalizations for less frequent diseases in the Medicare population.

Statistical Analysis We divided patients into 2 samples, 2009–2011 and 2012–2014. We used the first sample to conduct the main analysis and the second to confirm the find- ings. We compared baseline characteristics between

patients who had a recurrent AMI and those who did not have a recurrent AMI using the chi- squared test for categorical variables and the t test for continuous varia- bles. Using the 2009–2011 sample, for each patient, we estimated the baseline risk at the time of discharge of having a recurrent AMI within 1 year after discharge by fitting a Cox proportional hazards model with Markov Chain Monte Carlo simulations that modeled time to first recurrent AMI as a function of a patient’s baseline characteristics described above. The model also in- cluded in- hospital treatments (percutaneous coronary intervention, coronary artery bypass grafting, and car- diac catheterization), length of stay, and discharge to home (yes/no) because these variables may be associ- ated with the outcome. We retained a variable in the model if the posterior probability of its nonzero coef- ficient was >0.95. We used the regression coefficients estimated from this model to calculate a baseline risk score for recurrent AMI for each patient. We standard- ized the score through the Z score method and strati- fied patients into 1 of 3 risk groups based on the risk score distribution: low (<10th percentile), average (10th– 90th percentile), and high (>90th percentile). The base- line risk group represented a patient’s risk of recurrent AMI at discharge. We used variables selected using the 2009–2011 data to calculate the score for patients in the 2012–2014 data as well.

We fit a single- variable Cox regression model to describe the observed relationship between 1- year recurrent AMI and a CCS- specific condition- related subsequent hospitalization, without accounting for patient baseline risk of recurrence. The time a hos- pitalization occurred was used as a time- varying variable in the analysis. We repeated this analysis for each of the potential subsequent hospitalizations. To further assess the association between rehospi- talizations and 1- year recurrent AMI, we fit the Cox regression model with Markov Chain Monte Carlo simulations that modeled recurrent AMI as a function of all potential subsequent hospitalizations (event, yes/no, and time) as time- varying variables, adjusted for the patient baseline risk score for recurrent AMI. We retained a rehospitalization in the model if the posterior probability of its nonzero coefficient was >0.95. To further assess the change in risk of recur- rent AMI between patients with and without at least 1 subsequent rehospitalization, we fit the Cox model with a binary time- varying indicator (1=had ≤1 sub- sequent CCS- specific condition- related rehospital- izations; 0=no rehospitalization), stratified by baseline risk group and further by age group. If a patient had >1 CCS- specific condition- related rehospitalization, the time that the first event occurred was used for the model.

Analyses were conducted using SAS version 9.4, 64- bit Windows (SAS Institute Inc., Cary, NC). As

D ow

nloaded from http://ahajournals.org by on June 13, 2020

J Am Heart Assoc. 2020;9:e014907. DOI: 10.1161/JAHA.119.014907 4

Wang et al Rehospitalization Associated with Recurrent AMI

Table 1. Patient Baseline Characteristics by Study Sample

Fiscal Year 2009–2011 Fiscal Year 2012–2014

Aggregated (n=447 690)

Without Recurrent AMI

(n=426 426)

With Recurrent

AMI (n=21 264)

Aggregated (n=437 241)

Without Recurrent AMI

(n=419 120)

With Recurrent

AMI (n=18 121)

Demographics, n (%)

Age, mean (SD) 78.3 (8.6) 78.2 (8.5) 79.1 (9.0) 77.8 (8.6) 77.8 (8.6) 78.3 (9.0)

Female 218 034 (48.7) 207 463 (48.7) 10 571 (49.7) 204 361 (46.7) 195 846 (46.7) 8515 (47.0)

White 391 029 (87.3) 373 133 (87.5) 17 896 (84.2) 378 338 (86.5) 363 327 (86.7) 15 011 (82.8)

Black 35 049 (7.8) 32 874 (7.7) 2175 (10.2) 35 158 (8.0) 33 234 (7.9) 1924 (10.6)

Other 21 612 (4.8) 20 419 (4.8) 1193 (5.6) 23 745 (5.4) 22 559 (5.4) 1186 (6.5)

Prior cardiovascular events, n (%)

Heart failure 57 891 (12.9) 53 063 (12.4) 4828 (22.7) 52 783 (12.1) 48 970 (11.7) 3813 (21.0)

AMI 18 926 (4.2) 16 607 (3.9) 2319 (10.9) 17 917 (4.1) 15 946 (3.8) 1971 (10.9)

Unstable angina 11 499 (2.6) 10 219 (2.4) 1280 (6.0) 10 356 (2.4) 9375 (2.2) 981 (5.4)

Chronic atherosclerosis 326 924 (73.0) 310 783 (72.9) 16 141 (75.9) 322 038 (73.7) 308 135 (73.5) 13 903 (76.7)

Cardiopulmonary respiratory failure or shock

20 232 (4.5) 18 717 (4.4) 1515 (7.1) 21 444 (4.9) 20 102 (4.8) 1342 (7.4)

Anterior MI (ICD- 9 410.00–410.19) 40 964 (9.2) 39 685 (9.3) 1279 (6.0) 36 106 (8.3) 35 157 (8.4) 949 (5.2)

Inferior/lateral/posterior MI (ICD- 9 410.20–410.69)

60 244 (13.5) 58 556 (13.7) 1688 (7.9) 56 186 (12.9) 54 903 (13.1) 1283 (7.1)

Comorbidities, n (%)

Hypertension 299 244 (66.8) 284 380 (66.7) 14 864 (69.9) 301 284 (68.9) 288 320 (68.8) 12 964 (71.5)

Stroke 17 964 (1.8) 16 798 (1.8) 1166 (2.3) 16 651 (1.7) 15 600 (1.7) 1051 (2.3)

Cerebrovascular disease 57 891 (4.0) 53 063 (3.9) 4828 (5.5) 52 783 (3.8) 48 970 (3.7) 3813 (5.8)

Renal failure 52 288 (11.7) 47 971 (11.2) 4317 (20.3) 54 285 (12.4) 50 429 (12.0) 3856 (21.3)

COPD 88 711 (19.8) 83 582 (19.6) 5129 (24.1) 85 379 (19.5) 81 060 (19.3) 4319 (23.8)

Pneumonia 64 817 (14.5) 61 108 (14.3) 3709 (17.4) 58 620 (13.4) 55 605 (13.3) 3015 (16.6)

Protein- calorie malnutrition 21 010 (4.7) 20 090 (4.7) 920 (4.3) 21 924 (5.0) 21 057 (5.0) 867 (4.8)

Dementia 48 997 (10.9) 46 521 (10.9) 2476 (11.6) 24 691 (5.6) 23 583 (5.6) 1108 (6.1)

Functional disability 10 927 (2.4) 10 182 (2.4) 745 (3.5) 10 552 (2.4) 9908 (2.4) 644 (3.6)

Peripheral vascular disease 27 609 (6.2) 25 358 (5.9) 2251 (10.6) 24 431 (5.6) 22 712 (5.4) 1719 (9.5)

Metastatic cancer 29 141 (6.5) 27 664 (6.5) 1477 (6.9) 27 341 (6.3) 26 094 (6.2) 1247 (6.9)

Major trauma in past year 26 039 (5.8) 24 762 (5.8) 1277 (6.0) 22 871 (5.2) 21 951 (5.2) 920 (5.1)

Major psychiatric disorder 9417 (2.1) 8867 (2.1) 550 (2.6) 9679 (2.2) 9210 (2.2) 469 (2.6)

Chronic liver disease 3009 (0.7) 2835 (0.7) 174 (0.8) 3446 (0.8) 3282 (0.8) 164 (0.9)

Depression 25 098 (5.6) 23 821 (5.6) 1277 (6.0) 26 353 (6.0) 25 187 (6.0) 1166 (6.4)

Diabetes mellitus 139 047 (31.1) 130 064 (30.5) 8983 (42.2) 144 787 (33.1) 136 443 (32.6) 8344 (46.0)

Parkinson or Huntington disease 6186 (1.4) 5845 (1.4) 341 (1.6) 6045 (1.4) 5771 (1.4) 274 (1.5)

Anemia 11 0791 (24.7) 104 398 (24.5) 6393 (30.1) 112 477 (25.7) 106 785 (25.5) 5692 (31.4)

Asthma 10 699 (2.4) 10 178 (2.4) 521 (2.5) 11 153 (2.6) 10 681 (2.5) 472 (2.6)

In- hospital procedures, n (%)

Percutaneous coronary intervention 181 125 (40.5) 174 586 (40.9) 6539 (30.8) 192 090 (43.9) 185 794 (44.3) 6296 (34.7)

Coronary artery bypass grafting 41 365 (9.2) 40 676 (9.5) 689 (3.2) 39 558 (9.0) 38 948 (9.3) 610 (3.4)

Cardiac catheterization 260 515 (58.2) 250 689 (58.8) 9826 (46.2) 263 534 (60.3) 254 605 (60.7) 8929 (49.3)

Discharge disposition, n (%)

Home 253 528 (56.6) 241 745 (56.7) 11 783 (55.4) 255 527 (58.4) 245 028 (58.5) 10 499 (57.9)

Home with care 69 802 (15.6) 65 712 (15.4) 4090 (19.2) 66 388 (15.2) 63 034 (15.0) 3354 (18.5)

Skilled nursing facility or intermediate care facility

84 997 (19.0) 80 767 (18.9) 4230 (19.9) 76 120 (17.4) 72 897 (17.4) 3223 (17.8)

(Continued)

D ow

nloaded from http://ahajournals.org by on June 13, 2020

J Am Heart Assoc. 2020;9:e014907. DOI: 10.1161/JAHA.119.014907 5

Wang et al Rehospitalization Associated with Recurrent AMI

of 2019, the data were 5  years old. Analyses were repeated using the 2012–2014 data. Deaths before recurrent AMI were addressed using the Fine and Gray20 method for competing risks. The Lee, Wei, and Amato method21 of robust sandwich variance matrix estimation was used to adjust for within- hospital clustering of patients. All statistical testing was 2- sided, and P<0.05 was considered statisti- cally significant. The study followed the guidelines for cohort studies described in the Strengthening the Reporting of Observational Studies in Epidemiology Statement: Guidelines for Reporting Observational Studies.22 The Yale University Institutional Review Board reviewed the study protocol and granted a waiver of informed consent for the use of the deiden- tified database.

RESULTS Study Sample and Patient Baseline Characteristics The study included 884 931 (447 690 in 2009–2011 and 437 241 in 2012–2014) unique patients who were discharged alive after AMI, were not transferred to another acute- care hospital, and were hospitalized for >1 day during their index admission. Overall, pa- tients had a mean age of 78.0 (SD, 8.6) years, and 47.7% were female. The most common comorbidi- ties were chronic atherosclerosis (73.3%), hyperten- sion (67.9%), diabetes mellitus (32.1%), and anemia (25.2%). During the index AMI hospitalization, 42.2% of patients had a percutaneous coronary interven- tion, 9.1% underwent coronary artery bypass graft- ing, and 58.2% had cardiac catheterization. The median length of stay was 4 (interquartile range [IQR], 2–7) days, and 57.5% of patients were discharged to home. Patient characteristics were no different

between the 2009–2011 and 2012–2014 samples (Table 1).

Outcome For the 2009–2011 and 2012–2014 samples, the rates of 1- year recurrent AMI were 5.3% (95% CI, 5.27–5.41) and 4.6% (95% CI, 4.54–4.67), respectively (P<0.001). Among these patients who had a recurrent AMI, the median (IQR) days from discharge to a recurrent AMI was 115 (34–230) for the 2009–2011 sample and 106 (31–217) for the 2012–2014 sample. In the 2009–2011 and 2012–2014 samples, respectively, the median (IQR) survived days among patients who died within 1  year without a recurrent AMI were 56 (20–128) and 53 (19–121), and the median (IQR) survived days among patients who died with a recurrent AMI were 105 (49–188) and 97 (48–179).

All- cause mortality rates after the index AMI and before a recurrent AMI were 17.5% (95% CI, 17.4–17.6) and 15.7% (95% CI, 15.6–15.8) for the 2009–2011 and 2012–2014 samples, respectively. Compared with pa- tients without a recurrent AMI, patients with a recur- rent AMI had a higher 30- day postdischarge all- cause mortality rate (5.3% versus 2.7%; P<0.001), higher 30- day all- cause readmission rate (38.9% versus 18.1%; P<0.001), higher 1- year all- cause mortality rate (23.3% versus 19.2%; P<0.001), lower median Medicare pay- ment ($10 300 versus $10 500), and shorter mean (SD) length of stay (5 [4.6] days versus 6 [5.5] days). These observed outcomes were similar in the 2009–2011 and 2012–2014 samples (Table 1).

Association Between Patient Baseline Characteristics and Recurrent AMI In the 2009–2011 sample, the 5 baseline char- acteristics most strongly associated with 1- year

Fiscal Year 2009–2011 Fiscal Year 2012–2014

Aggregated (n=447 690)

Without Recurrent AMI

(n=426 426)

With Recurrent

AMI (n=21 264)

Aggregated (n=437 241)

Without Recurrent AMI

(n=419 120)

With Recurrent

AMI (n=18 121)

Outcome

Length of stay, mean (SD) days 6 (5.4) 6 (5.5) 5 (4.6) 5 (5.1) 5 (5.1) 5 (4.2)

Medicare payment, median (IQR), $1000

10.8 (8.3–16.1) 10.8 (8.4–16.2) 10.3 (7.1–14.2) 11.1 (8.5–16.7) 11.1 (8.5–16.8) 10.4 (7.1–14.8)

30- day mortality after discharge, n (%) 23 061 (5.2) 22 487 (5.3) 574 (2.7) 21 743 (5.0) 21 240 (5.1) 503 (2.8)

1- year mortality after discharge, n (%) 86 692 (19.4) 81 729 (19.2) 4963 (23.3) 76 542 (17.5) 72 639 (17.3) 3903 (21.5)

30- day all- cause readmission after discharge, n (%)

85 234 (19.0) 76 967 (18.1) 8270 (38.9) 72 910 (16.7) 65 913 (15.7) 6997 (38.6)

AMI indicates acute myocardial infarction; COPD, chronic obstructive pulmonary disease; ICD- 9, International Classification of Diseases, Ninth Revision; IQR, interquartile range; and MI, myocardial infarction.

Table 1. Continued

D ow

nloaded from http://ahajournals.org by on June 13, 2020

J Am Heart Assoc. 2020;9:e014907. DOI: 10.1161/JAHA.119.014907 6

Wang et al Rehospitalization Associated with Recurrent AMI

recurrent AMI were AMI before the index admission (hazard ratio [HR], 1.8 [95% CI, 1.76–1.93]), unsta- ble angina (HR, 1.5 [95% CI, 1.41–1.59]), diabe- tes mellitus (HR, 1.4 [95% CI, 1.39–1.47]), chronic atherosclerosis (HR, 1.3 [95% CI, 1.26–1.35]), and renal failure (HR, 1.2 [95% CI, 1.14–1.23]; Figure 1). Patients with a standardized risk <−1.2 times the SD, between −1.2 and 1.2, and >1.2 were stratified into low- , average- , and high- risk groups, respec- tively (Figure  2). The mean (SD) estimated rates of 1- year recurrent AMI were 1.7% (0.47) for the low- risk group, 5.1% (2.03) for average risk, and 14.3% (5.31) for high risk. Within a risk group, the rate of recurrence increased with patient age (Figure  3, top panel). Results were similar for the 2012–2014 sample (Figure 3, bottom panel).

Subsequent Hospitalizations Among 285 CCS- specific conditions, 19 occurred in at least 1% of patients and were included in the model for the 2009–2011 sample (Table 2). The median (IQR) tetrachoric correlations among parts of these condi- tions were low (0.11 [95% CI, 0.08–0.16] in 2009–2011 and 0.12 [95% CI, 0.09–0.18] in 2012–2014). The high- est correlation occurred between chronic obstructive pulmonary disease and bronchiectasis (CCS- 127) and respiratory failure/insufficiency/arrest (CCS- 131), which was 0.40 in 2009–2011 and 0.41 in 2012–2014. Among 447  690 patients in the 2009–2011 sample, 36.0% (n=161  327) had at least 1 rehospitalization for 1 of these 19 CCS- specific conditions before a recurrent AMI within 1 year. The 5 most common rehospitaliza- tion events were congestive heart failure (CCS- 108,

Figure 1. Patient baseline characteristics associated with recurrent AMI within 1 year after the initial AMI. AMI indicates acute myocardial infarction; CABG, coronary artery bypass grafting; CATH, cardiac catheterization; COPD, chronic obstructive pulmonary disease; and PCI, percutaneous coronary intervention.

D ow

nloaded from http://ahajournals.org by on June 13, 2020

J Am Heart Assoc. 2020;9:e014907. DOI: 10.1161/JAHA.119.014907 7

Wang et al Rehospitalization Associated with Recurrent AMI

9.8%), coronary atherosclerosis and other heart dis- ease (CCS- 101, 6.6%), septicemia (CCS- 2, 4.0%), car- diac dysrhythmias (CCS- 106, 3.7%), and pneumonia (CCS- 122, 3.4%). The 5 events that occurred soonest

after discharge were complications of surgical proce- dures or medical care (CCS- 238; median, 53 [IQR, 12–165] days), coronary atherosclerosis and other heart disease (CCS- 101; median, 54 [IQR, 19–154]

Figure 2. Distribution of baseline risk of recurrent acute myocardial infarction within 1 year after discharge for AMI. AMI indicates acute myocardial infarction.

Figure  3. Baseline risk groups of recurrent acute myocardial infarction within 1  year after discharge for AMI. AMI indicates acute myocardial infarction.

D ow

nloaded from http://ahajournals.org by on June 13, 2020

J Am Heart Assoc. 2020;9:e014907. DOI: 10.1161/JAHA.119.014907 8

Wang et al Rehospitalization Associated with Recurrent AMI

days), congestive heart failure (CCS- 108; median, 62 [IQR, 18–161] days), respiratory failure/insufficiency/ arrest (CCS- 131; median, 88 [IQR, 28–198] days), and cardiac dysrhythmias (CCS- 106; median, 90 [IQR, 25–204] days; Figure 4, top panel). The findings were similar for the 2012–2014 sample (Figure  4, bottom panel). In- hospital mortality for these CCS- specific hospitalizations ranged from 0.3% (nonspecific chest pain, CCS- 102) to 21.9% (septicemia, CCS- 2).

Association Between Subsequent Hospitalizations and Recurrent AMI Most of the 19 subsequent hospitalizations were associated with an increased recurrent AMI risk in the descriptive analysis without accounting for pa- tient baseline risk of recurrent AMI (Table  3). The Cox model based on the 2009–2011 data identified 12 CCS- specific subsequent hospitalizations sig- nificantly associated to recurrent AMI risk (Figure  5). These hospitalizations were septicemia (CCS- 2), diabetes mellitus with complications (CCS- 50), de- ficiency and other anemia (CCS- 59), hypertension with complications and secondary hypertension (CCS- 99), coronary atherosclerosis and other heart disease (CCS- 101), nonspecific chest pain (CCS- 102), congestive heart failure (CCS- 108), pneumonia (CCS- 122), chronic obstructive pulmonary disease and bronchiectasis (CCS- 127), gastrointestinal hemor- rhage (CCS- 153), acute and unspecified renal failure

(CCS- 157), and complication of device (implant or graft; CCS- 237). All these rehospitalizations except septicemia (CCS- 2) were associated with increased risk of recurrent AMI; septicemia (CCS- 2) was associ- ated with a lower risk of recurrent AMI (Figure 5). The HRs ranged from 1.6 (95% CI, 1.55–1.70, heart fail- ure [CCS- 108]) to 1.1 (95% CI, 1.04–1.25, pneumonia [CCS- 122]); the HR for septicemia (CCS- 2) was 0.6 (95% CI, 0.58–0.71; Figure 5).

Overall, 26.9% of patients in 2009–2011 and 22.5% of patients in 2012–2014 had at least 1 of the identified subsequent CCS- specific hospitalizations significantly associated with increased risk of recurrent AMI. For the low- , average- , and high- risk groups in the 2009–2011 sample, having at least 1 CCS- specific hospitalization was associated with an increase in the risk of recur- rent AMI by 210% (95% CI, 77%–149%), 73% (95% CI 66%–79%), and 43% (95% CI 34%–52%), respectively. The younger age group (65–74 years) in the average- risk strata was most likely to have a recurrence with at least 1 CCS-specific hospitalization (Figure  6, left panel). The findings were similar for the 2012–2014 co- hort (Figure 6, right panel).

DISCUSSION In this study, we demonstrated that hospitalizations after AMI were associated with the risk of a subse- quent AMI. Although patient baseline characteristics

Table 2. Occurrence of 19 Targeted Types of CCS- Specific Condition- Related Subsequent Hospitalizations

Rehospitalization, N (%)

FY 2009–2011 FY 2012–2014

N=447,690 N=437,241

Septicemia (CCS- 2; except in labor) 18 582 (4.2) 19 333 (4.4)

Diabetes mellitus with complications (CCS- 50) 5445 (1.2) 4462 (1.0)

Fluid and electrolyte disorders (CCS- 55) 7210 (1.6) 5407 (1.2)

Deficiency and other anemia (CCS- 59) 5159 (1.2) 3971 (0.9)

Hypertension with complications and secondary hypertension (CCS- 99) 6305 (1.4) 6216 (1.4)

Coronary atherosclerosis and other heart disease (CCS- 101) 31 034 (6.9) 21 222 (4.9)

Nonspecific chest pain (CCS- 102) 10 046 (2.2) 6855 (1.6)

Cardiac dysrhythmias (CCS- 106) 17 167 (3.8) 14 521 (3.3)

Congestive heart failure (CCS- 108; non- hypertensive) 46 340 (10.4) 37 766 (8.6)

Acute cerebrovascular disease (CCS- 109) 8825 (2.0) 7837 (1.8)

Pneumonia (CCS- 122; except that caused by tuberculosis or sexually transmitted disease) 15 762 (3.5) 12 688 (2.9)

Chronic obstructive pulmonary disease and bronchiectasis (CCS- 127) 9949 (2.2) 8012 (1.8)

Respiratory failure; insufficiency; arrest (CCS- 131; adult) 8397 (1.9) 7288 (1.7)

Gastrointestinal hemorrhage (CCS- 153) 9592 (2.1) 8825 (2.0)

Acute and unspecified renal failure (CCS- 157) 11 172 (2.5) 10 684 (2.4)

Urinary tract infections (CCS- 159) 9389 (2.1) 7232 (1.7)

Fracture of neck of femur (CCS- 226; hip) 4804 (1.1) 4002 (0.9)

Complication of device; implant or graft (CCS- 237) 10 109 (2.3) 8191 (1.9)

Complications of surgical procedures or medical care (CCS- 238) 7680 (1.7) 6306 (1.4)

AMI indicates acute myocardial infarction; CCS, Clinical Classifications Software; and FY, fiscal years.

D ow

nloaded from http://ahajournals.org by on June 13, 2020

J Am Heart Assoc. 2020;9:e014907. DOI: 10.1161/JAHA.119.014907 9

Wang et al Rehospitalization Associated with Recurrent AMI

were also associated with the risk of a recurrent AMI, we showed that patient risk of recurrence was in- fluenced by hospitalizations that occurred after dis- charge. Among the 12 rehospitalization categories identified in this study, 11 were associated with in- creased risk of recurrent AMI, with the increase in risk ranging from 14% (pneumonia) to 62% (heart failure). We found that patients who survived a hos- pitalization for septicemia had a lower risk of a recur- rent AMI.

There are several potential explanations for the associations between subsequent hospitalizations and increased patient risk of recurrent AMI. It is pos- sible that the hospitalization is a marker for the pres- ence and severity of comorbidities. Sick patients tend to have more comorbidities23 and are more likely to be rehospitalized after AMI.24 We adjusted for baseline comorbidities, but information about the severity does not reside within the administrative codes in our Medicare database. Additionally, the hospitalizations may be a marker for postdischarge

quality of care. Postdischarge care factors, such as continuity of care, type of care, and care providers, could impact AMI patient outcomes.25–27 Studies have identified associations between poor postdis- charge care and subsequent hospitalizations,28–30 including recurrent AMI. Many subsequent hospital- izations identified by our study, including those for diabetes mellitus, anemia, hypertension, coronary atherosclerosis, chest pain, heart failure, pneumo- nia, chronic obstructive pulmonary disease and bronchiectasis, respiratory failure, gastrointestinal hemorrhage, renal failure, and complications of an implant or graft, have been individually identified as potential risk markers for recurrent AMI or major car- diovascular events in previous studies.6,8,31 Another possible explanation for our findings is that the hos- pitalization itself increased the risk. The reason for the hospitalization may have been associated with inflammation, a known contributor to AMI risk, or to other factors associated with AMI, such as stress or depression. It is also possible that the hospitalization

Figure 4. Median (interquartile range [IQR]) days to subsequent rehospitalizations within 1 year after discharge for index AMI. The median (IQR) days to recurrent AMI were 115 (34–230) in the 2009–2011 sample and 106 (31–217) in the 2012 to 2014 sample. AMI indicates acute myocardial infarction; CAD, coronary artery disease; CCS_101, Coronary atherosclerosis and other heart disease; CCS_102, Nonspecific chest pain; CCS_106, Cardiac dysrhythmias; CCS_108, Congestive heart failure; CCS_109, Acute cerebrovascular disease; CCS_122, Pneumonia; CCS_131, Respiratory failure; insufficiency; arrest; CCS_153, Gastrointestinal hemorrhage; CCS_157, Acute and unspecified renal failure; CCS_159, Urinary tract infections; CCS_2, Septicemia; CCS_226, Fracture of neck of femur; CCS_237, Complication of device; implant or graft; CCS_238, Complications of surgical procedures or medical care; CCS_50, Diabetes mellitus with complications; CCS_55, Fluid and electrolyte disorders; CCS_59, Deficiency and other anemia; CCS_99, Hypertension with complications and secondary hypertension; CCS, Clinical Classifications Software; COPD, chronic obstructive pulmonary disease; DM, diabetes mellitus; HTN, hypertension; and UTI, urinary tract infection

D ow

nloaded from http://ahajournals.org by on June 13, 2020

J Am Heart Assoc. 2020;9:e014907. DOI: 10.1161/JAHA.119.014907 10

Wang et al Rehospitalization Associated with Recurrent AMI

led to the discontinuation of secondary preventive medications or the addition of medications to a pa- tient’s regimen, which may have resulted in nonad- herence to the regimen.

The negative association of septicemia with recur- rent AMI may represent a survivorship bias because patients who survived sepsis may have been healthier than those who died with sepsis and were therefore

Table 3. Observed Association Between a Targeted CCS- Specific Condition- Related Subsequent Hospitalization and 1- Year Recurrent AMI Based on a Single- Variable Cox Regression Model

Rehospitalization

FY 2009–2011 FY 2012–2014

HR (95% CI) HR (95% CI)

Septicemia (CCS- 2; except in labor) 0.80 (0.72–0.89) 0.87 (0.78–0.97)

Diabetes mellitus with complications (CCS- 50) 2.10 (1.85–2.37) 2.50 (2.18–2.86)

Fluid and electrolyte disorders (CCS- 55) 1.49 (1.31–1.68) 1.74 (1.51–2.01)

Deficiency and other anemia (CCS- 59) 1.87 (1.64–2.14) 1.97 (1.68–2.31)

Hypertension with complications and secondary hypertension (CCS- 99) 2.35 (2.11–2.62) 2.39 (2.13–2.69)

Coronary atherosclerosis and other heart disease (CCS- 101) 1.70 (1.61–1.80) 1.94 (1.82–2.08)

Nonspecific chest pain (CCS- 102) 1.79 (1.63–1.97) 1.89 (1.68–2.13)

Cardiac dysrhythmias (CCS- 106) 1.26 (1.16–1.37) 1.25 (1.13–1.38)

Congestive heart failure (CCS- 108; non- hypertensive) 2.12 (2.03–2.21) 2.00 (1.90–2.11)

Acute cerebrovascular disease (CCS- 109) 1.03 (0.90–1.18) 1.00 (0.86–1.17)

Pneumonia (CCS- 122; except that caused by tuberculosis or sexually transmitted disease) 1.46 (1.33–1.59) 1.57 (1.42–1.73)

Chronic obstructive pulmonary disease and bronchiectasis (CCS- 127) 1.93 (1.75–2.13) 1.75 (1.55–1.97)

Respiratory failure; insufficiency; arrest (CCS- 131; adult) 1.54 (1.38–1.73) 1.28 (1.11–1.47)

Gastrointestinal hemorrhage (CCS- 153) 1.42 (1.28–1.58) 1.48 (1.32–1.67)

Acute and unspecified renal failure (CCS- 157) 1.60 (1.45–1.76) 1.47 (1.31–1.63)

Urinary tract infections (CCS- 159) 1.37 (1.22–1.54) 1.35 (1.17–1.56)

Fracture of neck of femur (CCS- 226; hip) 1.40 (1.18–1.66) 1.26 (1.02–1.56)

Complication of device; implant or graft (CCS- 237) 1.98 (1.80–2.17) 2.13 (1.91–2.37)

Complications of surgical procedures or medical care (CCS- 238) 0.90 (0.79–1.04) 1.03 (0.88–1.20)

AMI indicates acute myocardial infarction; CCS, Clinical Classifications Software; FY, fiscal years; and HR, hazard ratio.

Figure 5. Association between subsequent rehospitalizations and recurrent AMI after discharge for index AMI, accounting for baseline risk of recurrence. AMI indicates acute myocardial infarction; CAD, coronary artery disease; CCS, Clinical Classifications Software; COPD, chronic obstructive pulmonary disease; DM, diabetes mellitus; and HTN, hypertension.

D ow

nloaded from http://ahajournals.org by on June 13, 2020

J Am Heart Assoc. 2020;9:e014907. DOI: 10.1161/JAHA.119.014907 11

Wang et al Rehospitalization Associated with Recurrent AMI

less likely to have a recurrent AMI. Our data showed that approximately 22% of AMI survivors rehospital- ized for sepsis died during the sepsis hospitalization.

Our study, based on real- world data, has several important characteristics. We focused on the first year after an initial AMI, a period that has the highest risk of recurrent AMI.4 Our findings provide real- world em- pirical evidence of the importance of accounting for postdischarge rehospitalizations to help ensure bet- ter long- term outcomes. We showed that patient risk stratification for recurrent AMI was a dynamic mea- sure that could change immediately after discharge. The CCS categories allowed the grouping of similar medical conditions to provide hospitals and physi- cians a parsimonious, clinically meaningful, and prac- tically useful composite measure of rehospitalizations. Such a composite measure could be used more easily than a traditional approach based on individual ICD diagnosis codes. The 12 subsequent hospitalizations identified in the study were based on the 285 CCS cat- egories, which represent all principal diagnosis codes for rehospitalizations. These CCS categories are easy to collect and readily available at the time of discharge for the rehospitalization.

A model that combines rehospitalizations with pa- tient baseline characteristics would allow hospitals and physicians to reevaluate patient risk for recurrent AMI throughout the first year and may help patients under- stand that their risk of recurrence depends not only on their baseline characteristics but also the sequence of rehospitalizations that occur after their initial AMI.

The ability to identify individuals with the highest risk of recurrent AMI after a rehospitalization may aid in the provision of targeted, intensive, and higher- quality longitudinal care after discharge. Additionally, insight regarding the long- term risk of subsequent hospital- izations associated with recurrent AMI is important from a patient perspective as educating patients re- garding their long- term risk might provide an even stronger incentive to follow- up and adhere to medica- tions. Our study also provides evidence that hospitals and primary care physicians caring for patients with a history of AMI should be aware that subsequent hos- pitalizations can change patient risk of recurrent AMI.

Our study has several limitations. We considered only the first recurrent AMI and acknowledge that patients may experience multiple recurrent events, in which a recurrent event model can be fit. The sub- sequent hospitalizations identified in our study were based on the CCS categories, which represent mul- tiple principal discharge diagnosis codes, while an individual rehospitalization only represents a single principal diagnosis code that could be more clinically important. We accounted for only inpatient rehospi- talizations and did not consider outpatient care, ob- servation stays, or emergency department visits. We treated subsequent rehospitalizations independently and acknowledge that some hospitalizations may have been related. Nevertheless, we found that the median tetrachoric correlation among these rehospi- talizations was not high, indicating these rehospital- izations were not strongly related to each other. We

Figure 6. Association between at least 1 subsequent hospitalization and risk of recurrent AMI by patient baseline risk and age groups. AMI indicates acute myocardial infarction.

D ow

nloaded from http://ahajournals.org by on June 13, 2020

J Am Heart Assoc. 2020;9:e014907. DOI: 10.1161/JAHA.119.014907 12

Wang et al Rehospitalization Associated with Recurrent AMI

did not address whether the association between a subsequent hospitalization and a recurrent AMI de- pended on the hospitalization- free duration from an index AMI discharge to a rehospitalization, which could be clinically important. We restricted the 2011– 2015 data to the first 10 diagnosis codes to align with the 2009–2010 data, which only contained 10 diagnosis codes. Accordingly, we may have missed some comorbidity information carried by the addi- tional codes. Our study was limited by the availability of data resources, and therefore it did not incorporate information on medication adherence, nursing home stays, and home health services, which were asso- ciated with rehospitalizations and recurrent AMI in prior work.32–34 Moreover, we used comorbidity infor- mation from administrative data. These data lack de- tailed clinical information on patient functional status, left ventricular function, non–ST- segment–elevation myocardial infarction, and ST- segment elevation myocardial infarction, which could be important for assessing risk of recurrence and reducing measure- ment error.

In conclusion, patient risk of recurrent AMI changed on the basis of the occurrence of subsequent hospi- talizations. Improving post–acute care to prevent un- planned rehospitalizations, especially those for chronic diseases, and extending the current focus on all- cause 30- day rehospitalizations to condition- specific rehos- pitalizations beyond the 30- day period are important for the secondary prevention of AMI. Moreover, there should be strong efforts to ensure that patients who experience these events have optimal secondary pre- vention strategies.

ARTICLE INFORMATION Received October 8, 2019; accepted January 31, 2020.

Affiliations From the Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA (Y.W., S.-L.T.N.); Center for Outcomes Research and Evaluation, Yale-New Haven Hospital, New Haven, CT (Y.W., H.M.K.); Departments of Chronic Disease Epidemiology (E.L.), and Health Policy and Management (H.M.K.), Yale School of Public Health, New Haven, CT; Department of Health Care Policy, Harvard Medical School, Boston, MA (S.-L.T.N.); Section of Cardiovascular Medicine, Department of Internal Medicine, Yale School of Medicine, New Haven, CT (H.M.K.).

Disclosures Dr Krumholz was a recipient of a research grant, through Yale, from Medtronic and the US Food and Drug Administration to develop methods for postmarket surveillance of medical devices; is a recipient of research agreements with Medtronic and Johnson & Johnson (Janssen), through Yale, to develop methods of clinical trial data sharing; was a recipient of a research agreement, through Yale, from the Shenzhen Center for Health Information for work to advance intelligent disease prevention and health promotion; collaborates with the National Center for Cardiovascular Diseases in Beijing; chairs a Cardiac Scientific Advisory Board for UnitedHealth; is a participant/ participant representative of the IBM Watson Health Life Sciences Board; is a member of the Advisory Board for Element Science and the Physician Advisory Board for Aetna; received payment from the Arnold & Porter Law Firm for work related to the Sanofi clopidogrel litigation and from the Ben

C. Martin Law Firm for work related to the Cook inferior vena cava filter litigation; and is the founder of Hugo, a personal health information platform. Drs Krumholz and Normand work under contract to the Centers for Medicare & Medicaid Services to develop and maintain performance measures that are publicly reported. The remaining authors have no disclosures to report.

Supplementary Material Table S1

REFERENCES 1. Brown TM, Deng L, Becker DJ, Bittner V, Levitan EB, Rosenson RS,

Safford MM, Muntner P. Trends in mortality and recurrent coronary heart disease events after an acute myocardial infarction among Medicare beneficiaries, 2001–2009. Am Heart J. 2015;170:249–255.

2. Buch P, Rasmussen S, Gislason GH, Rasmussen JN, Kober L, Gadsboll N, Stender S, Madsen M, Torp-Pedersen C, Abildstrom SZ. Temporal decline in the prognostic impact of a recurrent acute myocardial infarc- tion 1985 to 2002. Heart. 2007;93:210–215.

3. Chaudhry SI, Khan RF, Chen J, Dharmarajan K, Dodson JA, Masoudi FA, Wang Y, Krumholz HM. National trends in recurrent AMI hospital- izations 1 year after acute myocardial infarction in Medicare beneficia- ries: 1999–2010. J Am Heart Assoc. 2014;3:e001197. DOI: 10.1161/ JAHA.114.001197.

4. Smolina K, Wright FL, Rayner M, Goldacre MJ. Long- term survival and recurrence after acute myocardial infarction in England, 2004 to 2010. Circ Cardiovasc Qual Outcomes. 2012;5:532–540.

5. Krumholz HM, Normand S-LT, Wang Y. Twenty- year trends in outcomes for older adults with acute myocardial infarction in the United States. JAMA Netw Open. 2019;2:e191938.

6. Gilpin E, Ricou F, Dittrich H, Nicod P, Henning H, Ross J. Factors asso- ciated with recurrent myocardial infarction within one year after acute myocardial infarction. Am Heart J. 1991;121:457–465.

7. McCormick D, Gurwitz JH, Lessard D, Yarzebski J, Gore JM, Goldberg RJ. Use of aspirin, β- blockers, and lipid- lowering medications be- fore recurrent acute myocardial infarction. Arch Intern Med. 1999; 159:561.

8. Wang Y, Li J, Zheng X, Jiang Z, Hu S, Wadhera RK, Bai X, Lu J, Wang Q, Li Y, et al. Risk factors associated with major cardiovascu- lar events 1  year after acute myocardial infarction. JAMA Netw Open. 2018;1:e181079.

9. Shang P, Liu GG, Zheng X, Ho PM, Hu S, Li J, Jiang Z, Li X, Bai X, Gao Y, et al. Association between medication adherence and 1- year major car- diovascular adverse events after acute myocardial infarction in China. J Am Heart Assoc. 2019;8:011793. DOI: 10.1161/JAHA.118.011793.

10. Shore S, Jones PG, Maddox TM, Bradley SM, Stolker JM, Arnold SV, Parashar S, Peterson P, Bhatt DL, Spertus J, et al. Longitudinal per- sistence with secondary prevention therapies relative to patient risk after myocardial infarction. Heart. 2015;101:800–807.

11. Arnold SV, Smolderen KG, Buchanan DM, Li Y, Spertus JA. Perceived stress in myocardial infarction: long- term mortality and health status outcomes. J Am Coll Cardiol. 2012;60:1756–1763.

12. Roe MT, Chen AY, Thomas L, Wang TY, Alexander KP, Hammill BG, Gibler WB, Ohman EM, Peterson ED. Predicting long- term mortality in older patients after non- ST- segment elevation myocardial infarction: the CRUSADE long- term mortality model and risk score. Am Heart J. 2011;162:875–883.e1.

13. Ketchum ES, Dickstein K, Kjekshus J, Pitt B, Wong MF, Linker DT, Levy WC. The Seattle Post Myocardial Infarction Model (SPIM): prediction of mortality after acute myocardial infarction with left ventricular dysfunc- tion. Eur Heart J Acute Cardiovasc Care. 2014;3:46–55.

14. Plakht Y, Shiyovich A, Gilutz H. Predictors of long- term (10- year) mor- tality postmyocardial infarction: age- related differences. Soroka Acute Myocardial Infarction (SAMI) Project. J Cardiol. 2015;65:216–223.

15. Smolderen KG, Buchanan DM, Gosch K, Whooley M, Chan PS, Vaccarino V, Parashar S, Shah AJ, Ho PM, Spertus JA. Depression treatment and 1- year mortality after acute myocardial infarction: in- sights from the TRIUMPH Registry (Translational Research Investigating Underlying Disparities in Acute Myocardial Infarction Patients’ Health Status). Circulation. 2017;135:1681–1689.

16. Tu JV, Austin PC, Walld R, Roos L, Agras J, McDonald KM. Development and validation of the Ontario acute myocardial infarction mortality pre- diction rules. J Am Coll Cardiol. 2001;37:992–997.

D ow

nloaded from http://ahajournals.org by on June 13, 2020

J Am Heart Assoc. 2020;9:e014907. DOI: 10.1161/JAHA.119.014907 13

Wang et al Rehospitalization Associated with Recurrent AMI

17. Krumholz HM, Wang Y, Mattera JA, Wang Y, Han LF, Ingber MJ, Roman S, Normand S-LT. An administrative claims model suitable for profiling hospital performance based on 30- day mortality rates among patients with an acute myocardial infarction. Circulation. 2006;113:1683–1692.

18. Centers for Medicare & Medicaid Services. 5010 implementation— Processing additional International Classification of Diseases, 9th Revision—Clinical Modification (ICD-9-CM) diagnosis and proce- dure codes in Pricer, Grouper, and the Medicare Code Editor (MCE). Pub 100-04. 2010. Available at: https://www.cms.gov/Regulations- and-Guidance/Guidance/ Transmit tals/2010 -Transmit tals-Items/ CMS1237956.html. Accessed January 16, 2020.

19. Elixhauser A, Steiner C, Palmer L. Clinical Classifications Software (CCS), 2015. U.S. Agency for Healthcare Research and Quality. Available at: https://www.hcup-us.ahrq.gov/toolssoftware/ccs/CCSUsersGuide. pdf. Accessed January 16, 2020.

20. Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999;94:496–509.

21. Lee EW, Wei LJ, Amato DA, Leurgans S. Cox-type regression analysis for large numbers of small groups of correlated failure time observa- tions. In: Klein JP, Goel PK, eds. Survival Analysis: State of the Art. Dordrecht: Kluwer Academic; 1992:237–247.

22. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. PLoS Med. 2007;4:e296.

23. Radovanovic D, Maurer L, Bertel O, Witassek F, Urban P, Stauffer JC, Pedrazzini G, Erne P. Treatment and outcomes of patients with recur- rent myocardial infarction: a prospective observational cohort study. J Cardiol. 2016;68:498–503.

24. Ahmedani BK, Solberg LI, Copeland LA, Fang-Hollingsworth Y, Stewart C, Hu J, Nerenz DR, Williams LK, Cassidy-Bushrow AE, Waxmonsky J, et al. Psychiatric comorbidity and 30- day readmissions after hos- pitalization for heart failure, AMI, and pneumonia. Psychiatric Serv. 2015;66:134–140.

25. van Walraven C, Mamdani M, Fang J, Austin PC. Continuity of care and pa- tient outcomes after hospital discharge. J Gen Intern Med. 2004;19:624–631.

26. Van Walraven C, Oake N, Jennings A, Forster AJ. The association be- tween continuity of care and outcomes: a systematic and critical review. J Eval Clin Pract. 2010;16:947–956.

27. Pereira Gray DJ, Sidaway-Lee K, White E, Thorne A, Evans PH. Continuity of care with doctors—a matter of life and death? A systematic review of continuity of care and mortality. BMJ Open. 2018;8:e021161.

28. Greenblatt DY, Weber SM, OʼConnor ES, LoConte NK, Liou J-I, Smith MA. Readmission after colectomy for cancer predicts one- year mortal- ity. Ann Surg. 2010;251:659–669.

29. Lum HD, Studenski SA, Degenholtz HB, Hardy SE. Early hospital read- mission is a predictor of one- year mortality in community- dwelling older Medicare beneficiaries. J Gen Intern Med. 2012;27:1467–1474.

30. Stein LK, Kwon CS, Agarwal P, Jette N, Dhamoon MS. Readmission to another hospital following acute stroke admission is associated with worse outcomes: nationally representative data. Abstract m127. Ann Neurol. 2018;84:S149.

31. Qintar M, Spertus JA, Tang Y, Buchanan DM, Chan PS, Amin AP, Salisbury AC. Noncardiac chest pain after acute myocardial infarction: frequency and association with health status outcomes. Am Heart J. 2017;186:1–11.

32. Di Bartolomeo S, Marino M, Guastaroba P, Valent F, De Palma R. Self- controlled case- series study to verify the effect of adherence to beta- blockers in secondary prevention of myocardial infarction. J Am Heart Assoc. 2015;4:e001575. DOI: 10.1161/JAHA.114.001575.

33. Ortolani P, Di Bartolomeo S, Marino M, Vagnarelli F, Guastaroba P, Rapezzi C, De Palma R. Adherence to agents acting on the renin–an- giotensin system in secondary prevention of non- fatal myocardial in- farction: a self- controlled case- series study. Eur Heart J Cardiovasc Pharmacother. 2015;1:254–259.

34. Wang Y, Pandolfi MM, Fine J, Metersky ML, Wang C, Ho S-Y, Galusha D, Nuti SV, Murugiah K, Spenard A, et al. Community- level association between home health and nursing home performance on quality and hospital 30- day readmissions for Medicare patients. Home Health Care Manag Pract. 2016;28:201–208.

D ow

nloaded from http://ahajournals.org by on June 13, 2020

SUPPLEMENTAL MATERIAL

D ow

nloaded from http://ahajournals.org by on June 13, 2020

Table S1. Clinical Classifications Software categories.

CCS ID Disease category

1 Tuberculosis

2 Septicemia (except in labor)

3 Bacterial infection; unspecified site

4 Mycoses

5 HIV infection

6 Hepatitis

7 Viral infection

8 Other infections; including parasitic

9 Sexually transmitted infections (not HIV or hepatitis)

10 Immunizations and screening for infectious disease

11 Cancer of head and neck

12 Cancer of esophagus

13 Cancer of stomach

14 Cancer of colon

15 Cancer of rectum and anus

16 Cancer of liver and intrahepatic bile duct

17 Cancer of pancreas

18 Cancer of other GI organs; peritoneum

19 Cancer of bronchus; lung

20 Cancer; other respiratory and intrathoracic

21 Cancer of bone and connective tissue

22 Melanomas of skin

23 Other non-epithelial cancer of skin

24 Cancer of breast

25 Cancer of uterus

26 Cancer of cervix

27 Cancer of ovary

28 Cancer of other female genital organs

29 Cancer of prostate

30 Cancer of testis

31 Cancer of other male genital organs

32 Cancer of bladder

33 Cancer of kidney and renal pelvis

34 Cancer of other urinary organs

35 Cancer of brain and nervous system

36 Cancer of thyroid

37 Hodgkin`s disease

38 Non-Hodgkin`s lymphoma

39 Leukemias

40 Multiple myeloma

41 Cancer; other and unspecified primary

42 Secondary malignancies

43 Malignant neoplasm without specification of site

D ow

nloaded from http://ahajournals.org by on June 13, 2020

44 Neoplasms of unspecified nature or uncertain behavior

45 Maintenance chemotherapy; radiotherapy

46 Benign neoplasm of uterus

47 Other and unspecified benign neoplasm

48 Thyroid disorders

49 Diabetes mellitus without complication

50 Diabetes mellitus with complications

51 Other endocrine disorders

52 Nutritional deficiencies

53 Disorders of lipid metabolism

54 Gout and other crystal arthropathies

55 Fluid and electrolyte disorders

56 Cystic fibrosis

57 Immunity disorders

58 Other nutritional; endocrine; and metabolic disorders

59 Deficiency and other anemia

60 Acute posthemorrhagic anemia

61 Sickle cell anemia

62 Coagulation and hemorrhagic disorders

63 Diseases of white blood cells

64 Other hematologic conditions

650 Adjustment disorders

651 Anxiety disorders

652 Attention-deficit, conduct, and disruptive behavior disorders

653 Delirium, dementia, and amnestic and other cognitive disorders

654 Developmental disorders

655 Disorders usually diagnosed in infancy, childhood, or adolescence

656 Impulse control disorders, NEC

657 Mood disorders

658 Personality disorders

659 Schizophrenia and other psychotic disorders

660 Alcohol-related disorders

661 Substance-related disorders

662 Suicide and intentional self-inflicted injury

663 Screening and history of mental health and substance abuse codes

670 Miscellaneous disorders

76 Meningitis (except that caused by tuberculosis or sexually transmitted disease)

77 Encephalitis (except that caused by tuberculosis or sexually transmitted disease)

78 Other CNS infection and poliomyelitis

79 Parkinson`s disease

80 Multiple sclerosis

81 Other hereditary and degenerative nervous system conditions

82 Paralysis

83 Epilepsy; convulsions

84 Headache; including migraine

85 Coma; stupor; and brain damage

D ow

nloaded from http://ahajournals.org by on June 13, 2020

86 Cataract

87 Retinal detachments; defects; vascular occlusion; and retinopathy

88 Glaucoma

89 Blindness and vision defects

90

Inflammation; infection of eye (except that caused by tuberculosis or sexually

transmitted disease)

91 Other eye disorders

92 Otitis media and related conditions

93 Conditions associated with dizziness or vertigo

94 Other ear and sense organ disorders

95 Other nervous system disorders

96 Heart valve disorders

97

Peri-; endo-; and myocarditis; cardiomyopathy (except that caused by tuberculosis

or sexually transmitted disease)

98 Essential hypertension

99 Hypertension with complications and secondary hypertension

100 Acute myocardial infarction

101 Coronary atherosclerosis and other heart disease

102 Nonspecific chest pain

103 Pulmonary heart disease

104 Other and ill-defined heart disease

105 Conduction disorders

106 Cardiac dysrhythmias

107 Cardiac arrest and ventricular fibrillation

108 Congestive heart failure; nonhypertensive

109 Acute cerebrovascular disease

110 Occlusion or stenosis of precerebral arteries

111 Other and ill-defined cerebrovascular disease

112 Transient cerebral ischemia

113 Late effects of cerebrovascular disease

114 Peripheral and visceral atherosclerosis

115 Aortic; peripheral; and visceral artery aneurysms

116 Aortic and peripheral arterial embolism or thrombosis

117 Other circulatory disease

118 Phlebitis; thrombophlebitis and thromboembolism

119 Varicose veins of lower extremity

120 Hemorrhoids

121 Other diseases of veins and lymphatics

122 Pneumonia (except that caused by tuberculosis or sexually transmitted disease)

123 Influenza

124 Acute and chronic tonsillitis

125 Acute bronchitis

126 Other upper respiratory infections

127 Chronic obstructive pulmonary disease and bronchiectasis

128 Asthma

129 Aspiration pneumonitis; food/vomitus

D ow

nloaded from http://ahajournals.org by on June 13, 2020

130 Pleurisy; pneumothorax; pulmonary collapse

131 Respiratory failure; insufficiency; arrest (adult)

132 Lung disease due to external agents

133 Other lower respiratory disease

134 Other upper respiratory disease

135 Intestinal infection

136 Disorders of teeth and jaw

137 Diseases of mouth; excluding dental

138 Esophageal disorders

139 Gastroduodenal ulcer (except hemorrhage)

140 Gastritis and duodenitis

141 Other disorders of stomach and duodenum

142 Appendicitis and other appendiceal conditions

143 Abdominal hernia

144 Regional enteritis and ulcerative colitis

145 Intestinal obstruction without hernia

146 Diverticulosis and diverticulitis

147 Anal and rectal conditions

148 Peritonitis and intestinal abscess

149 Biliary tract disease

150 Liver disease; alcohol-related

151 Other liver diseases

152 Pancreatic disorders (not diabetes)

153 Gastrointestinal hemorrhage

154 Noninfectious gastroenteritis

155 Other gastrointestinal disorders

156 Nephritis; nephrosis; renal sclerosis

157 Acute and unspecified renal failure

158 Chronic kidney disease

159 Urinary tract infections

160 Calculus of urinary tract

161 Other diseases of kidney and ureters

162 Other diseases of bladder and urethra

163 Genitourinary symptoms and ill-defined conditions

164 Hyperplasia of prostate

165 Inflammatory conditions of male genital organs

166 Other male genital disorders

167 Nonmalignant breast conditions

168 Inflammatory diseases of female pelvic organs

169 Endometriosis

170 Prolapse of female genital organs

171 Menstrual disorders

172 Ovarian cyst

173 Menopausal disorders

174 Female infertility

175 Other female genital disorders

D ow

nloaded from http://ahajournals.org by on June 13, 2020

176 Contraceptive and procreative management

177 Spontaneous abortion

178 Induced abortion

179 Postabortion complications

180 Ectopic pregnancy

181 Other complications of pregnancy

182 Hemorrhage during pregnancy; abruptio placenta; placenta previa

183 Hypertension complicating pregnancy; childbirth and the puerperium

184 Early or threatened labor

185 Prolonged pregnancy

186

Diabetes or abnormal glucose tolerance complicating pregnancy; childbirth; or the

puerperium

187 Malposition; malpresentation

188 Fetopelvic disproportion; obstruction

189 Previous C-section

190 Fetal distress and abnormal forces of labor

191 Polyhydramnios and other problems of amniotic cavity

192 Umbilical cord complication

193 OB-related trauma to perineum and vulva

194 Forceps delivery

195 Other complications of birth; puerperium affecting management of mother

196 Normal pregnancy and/or delivery

197 Skin and subcutaneous tissue infections

198 Other inflammatory condition of skin

199 Chronic ulcer of skin

200 Other skin disorders

201

Infective arthritis and osteomyelitis (except that caused by tuberculosis or sexually

transmitted disease)

202 Rheumatoid arthritis and related disease

203 Osteoarthritis

204 Other non-traumatic joint disorders

205 Spondylosis; intervertebral disc disorders; other back problems

206 Osteoporosis

207 Pathological fracture

208 Acquired foot deformities

209 Other acquired deformities

210 Systemic lupus erythematosus and connective tissue disorders

211 Other connective tissue disease

212 Other bone disease and musculoskeletal deformities

213 Cardiac and circulatory congenital anomalies

214 Digestive congenital anomalies

215 Genitourinary congenital anomalies

216 Nervous system congenital anomalies

217 Other congenital anomalies

218 Liveborn

219 Short gestation; low birth weight; and fetal growth retardation

D ow

nloaded from http://ahajournals.org by on June 13, 2020

220 Intrauterine hypoxia and birth asphyxia

221 Respiratory distress syndrome

222 Hemolytic jaundice and perinatal jaundice

223 Birth trauma

224 Other perinatal conditions

225 Joint disorders and dislocations; trauma-related

226 Fracture of neck of femur (hip)

227 Spinal cord injury

228 Skull and face fractures

229 Fracture of upper limb

230 Fracture of lower limb

231 Other fractures

232 Sprains and strains

233 Intracranial injury

234 Crushing injury or internal injury

235 Open wounds of head; neck; and trunk

236 Open wounds of extremities

237 Complication of device; implant or graft

238 Complications of surgical procedures or medical care

239 Superficial injury; contusion

240 Burns

241 Poisoning by psychotropic agents

242 Poisoning by other medications and drugs

243 Poisoning by nonmedicinal substances

244 Other injuries and conditions due to external causes

245 Syncope

246 Fever of unknown origin

247 Lymphadenitis

248 Gangrene

249 Shock

250 Nausea and vomiting

251 Abdominal pain

252 Malaise and fatigue

253 Allergic reactions

254 Rehabilitation care; fitting of prostheses; and adjustment of devices

255 Administrative/social admission

256 Medical examination/evaluation

257 Other aftercare

258 Other screening for suspected conditions (not mental disorders or infectious disease)

259 Residual codes; unclassified

260 E Codes: All (external causes of injury and poisoning)

2601 E Codes: Cut/pierceb

2602 E Codes: Drowning/submersion

2603 E Codes: Fall

2604 E Codes: Fire/burn

2605 E Codes: Firearm

D ow

nloaded from http://ahajournals.org by on June 13, 2020

2606 E Codes: Machinery

2607 E Codes: Motor vehicle traffic (MVT)

2608 E Codes: Pedal cyclist; not MVT

2609 E Codes: Pedestrian; not MVT

2610 E Codes: Transport; not MVT

2611 E Codes: Natural/environment

2612 E Codes: Overexertion

2613 E Codes: Poisoning

2614 E Codes: Struck by; against

2615 E Codes: Suffocation

2616 E Codes: Adverse effects of medical care

2617 E Codes: Adverse effects of medical drugs

2618 E Codes: Other specified and classifiable

2619 E Codes: Other specified; NEC

2620 E Codes: Unspecified

2621 E Codes: Place of occurrence

D ow

nloaded from http://ahajournals.org by on June 13, 2020