Elsevier

American Heart Journal

Volume 160, Issue 5, November 2010, Pages 885-892
American Heart Journal

Clinical Investigation
Congestive Heart Failure
Generalizability and longitudinal outcomes of a national heart failure clinical registry: Comparison of Acute Decompensated Heart Failure National Registry (ADHERE) and non-ADHERE Medicare beneficiaries

https://doi.org/10.1016/j.ahj.2010.07.020Get rights and content

Background

Clinical registries are used increasingly to analyze quality and outcomes, but the generalizability of findings from registries is unclear.

Methods

We linked data from the Acute Decompensated Heart Failure National Registry (ADHERE) to 100% fee-for-service Medicare claims data. We compared patient characteristics and inpatient mortality of linked and unlinked ADHERE hospitalizations; patient characteristics, readmission, and postdischarge mortality of linked ADHERE patients to a random 20% sample of Medicare beneficiaries hospitalized for heart failure; and characteristics of Medicare sites participating and not participating in ADHERE.

Results

Among 135,667 ADHERE records for eligible patients ≥65 years, we matched 104,808 (77.3%) records to fee-for-service Medicare claims, representing 82,074 patients. Linked hospitalizations were more likely than unlinked hospitalizations to involve women and white patients; there were no meaningful differences in other patient characteristics. In-hospital mortality was identical for linked and unlinked hospitalizations. In Medicare, ADHERE patients had slightly lower unadjusted mortality (4.4% vs 4.9% in-hospital, 11.2% vs 12.2% at 30 days, 36.0% vs 38.3% at 1 year [P < .001]) and all-cause readmission (22.1% vs 23.7% at 30 days, 65.8% vs 67.9% at 1 year [P < .001]). After risk adjustment, modest but statistically significant differences remained. ADHERE hospitals were more likely than non-ADHERE hospitals to be teaching hospitals, have higher volumes of heart failure discharges, and offer advanced cardiac services.

Conclusion

Elderly patients in ADHERE are similar to Medicare beneficiaries hospitalized with heart failure. Differences related to selective enrollment in ADHERE hospitals and self-selection of participating hospitals are modest.

Section snippets

Data sources

We accessed the ADHERE-Core registry and the 100% inpatient Medicare claims files to identify patients for this study. The institutional review board of the Duke University Health System (Durham, NC) approved the study. The design and objectives of ADHERE have been reported previously.2, 5 In brief, ADHERE was established to evaluate characteristics, treatments, and temporal trends in therapy and outcomes among patients hospitalized with acute decompensated heart failure. More than 300

Results

There were 187,138 unique records in the ADHERE registry, each representing a single hospitalization. Of those, 135,674 (72.5%) were hospitalizations of patients ≥65 years. Seven patients were not eligible for linking with Medicare claims data because they were missing data in critical fields. The 135,667 records eligible for linking with Medicare claims data were submitted by 306 hospitals. Of these hospitals, 279 (91.2%) were identified in the Medicare data. Most of the unidentified hospitals

Discussion

By using indirect identifiers to link ADHERE records to Medicare claims, we were able to evaluate postdischarge outcomes of fee-for-service Medicare beneficiaries in ADHERE. The analysis yielded several important findings. First, linking a unique, large clinical registry to Medicare data is feasible. Second, in the ADHERE registry, linked patients were similar to unlinked patients with respect to comorbid conditions and outcomes. Third, the ADHERE cohort has external validity (and hence,

Conclusion

ADHERE, a large hospital-based heart failure registry, can be linked to fee-for-service Medicare claims data. The ADHERE Medicare population is similar with respect to demographic characteristics and comorbid conditions to the general Medicare heart failure population, with the exception of slightly lower mortality in unadjusted analysis. The ADHERE Medicare population is similar to the elderly ADHERE population as a whole with respect to comorbid conditions and outcomes, with the exception of

Acknowledgements

Damon M. Seils, MA, Duke University, provided editorial assistance and prepared the manuscript. Mr Seils did not receive compensation for his assistance apart from his employment at the institution where the study was conducted.

References (22)

  • CurtisL. et al.

    Representativeness of a national heart failure quality-of-care registry: comparison of OPTIMIZE-HF and non-OPTIMIZE-HF Medicare patients

    Circ Cardiovasc Qual Outcomes

    (2009)
  • Cited by (0)

    Trial registration: clinicaltrials.gov Identifier: NCT00366639.

    View full text