3M Potentially Preventable Readmissions (3M-PPR) software does not effectively distinguish differences in case-level quality of care among patients with acute myocardial infarction (MI) or heart failure (HF), according to research published in Circulation: Cardiovascular Quality and Outcomes.
In a retrospective, cross-sectional study, investigators from the department of health policy and management at the Boston University School of Public Health and the Center for Healthcare Organization and Implementation Research (CHOIR) at the Bedford VAMC Campus analyzed hospital discharge data and outpatient visit diagnoses collected by the Department of Veterans Affairs (VA) between 2006 and 2010, in an effort to assess quality of care.
All discharges were categorized with a principal diagnosis of either acute MI or HF and linked to VA readmissions within 30 days. Readmissions were defined by the researchers as “the first VA acute care hospitalization occurring within 30 days following index discharge.” Cases were flagged as either PPR-Yes (readmission was potentially preventable) or PPR-No (readmission was potentially not preventable).
Explicit process criteria were developed to represent the standards of care for both acute MI and HF; the criteria are based on the RAND/UCLA Appropriateness Method. Using clinical investigator input, candidate criteria were divided into 4 sections: admission work-up, in-hospital evaluation and treatment, discharge readiness (clinical stability at discharge) and planning, and postdischarge period. An expert panel worked to refine the criteria by rating individual items on a 7-point scale based on how well the experts felt that these items represented the standard of acute MI or HF care.
In order to assess care quality, the researchers collected medical records to develop a sample size of 100 cases per condition; 77 of 100 acute MI cases and 86 of 100 HF cases were flagged as PPR-Yes.
Both PPR-Yes and PPR-No cases were compared using the same electronic medical record (EMR)-abstracted comorbidities and variables. The researchers found that time to readmission was shorter in PPR-Yes cases for both acute MI and HF, although the difference was only significant in acute MI. Administrative data were used to identify a nonsignificant trend toward multiple comorbidities in the PPR-Yes cohort. The investigators also noted that while all patients within the acute MI group received cardiology care, 34% of PPR-Yes HF patients received the same care.
PPR-No acute MI cases exhibited higher achievement rates than PPR-Yes cases on 34 of 83 process criteria (41%); PPR-Yes cases performed better on 27 criteria (33%). Among HF cases, PPR-Yes cases exhibited higher achievement rates on 42% of process criteria compared to 37% of PPR-No cases.
“Among VA readmissions after either an acute MI or HF discharge, we found no significant association between quality of care, measured by adherence to selected processes of care received during the index admission and postdischarge, and potential preventability of readmission, as designated by 3M-PPR [software],” the researchers wrote. “Across study cohorts, we found no significant statistical differences between cases flagged as PPRs and nonflagged cases.”
As conducted, the study presented several limitations; primarily, the sample size was too small to allow the researchers to detect small differences in the quality score as statistically significant. Additionally, it is unclear whether the low postdischarge scores indicate poor EMR documentation of received care or a lack of VA care, among other limitations.
“Our findings suggest that although the 3M-PPR [software] is an attractive alternative to CMS’ all-cause readmission measure, it is no different than determining which individual cases are preventable,” the investigators concluded. “PPR classification … was not associated with adherence to process reflecting quality of care received during the index admission or postdischarge period among acute MI and HF readmissions.”
Disclosures: Dr Bhatt has received research funding from Amarin, AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Forest Laboratories, Ischemix, Medtronic, Pfizer, Roche, Sanofi, and The Medicines Company, among others.
Borzecki AM, Chen Q, Mull HJ, et al. Do acute myocardial infarction and heart failure readmissions flagged as potentially preventable by the 3M Potentially Preventable Readmissions Software have more process of care problems? Circ Cardiovasc Qual Outcomes. 2016;9; doi: 10.1161/circoutcomes.115.002509.