Correlation of the Agency for Health Care Policy and Research congestive heart failure admission guideline with mortality: Peer Review Organization Voluntary Hospital Association Initiative to Decrease Events (PROVIDE) for congestive heart failure

Study objective: We quantify patient risk as related to the presence or absence of the Agency for Health Care Policy and Research (AHCPR) congestive heart failure (CHF) hospital admission criteria. Methods: This was a retrospective observational cohort study at 12 acute care hospitals examining consecutive patients with the final primary diagnosis of CHF. Trained record abstractors blinded to outcome extracted 386 data elements, including 6 AHCPR admission criteria: (1) pulmonary edema (determined by radiograph) or severe respiratory distress (respiration >40 breaths/min), (2) hypoxia (oxygen saturation <90%) not caused by pulmonary disease, (3) significant edema (greater than or equal to+2) or anasarca, (4) symptomatic hypotension (<90 mm Hg systolic blood pressure) or syncope, (5) CHF of recent onset, and (6) clinical evidence (chest pain) of myocardial ischemia. The association between admission criteria and mortality rate (30 days, 6 months, and 1 year) was quantified and risk adjusted by stepwise logistic regression analysis. Results: Of the 1,674 patients with CHF, 1,340 (80%) were admitted to the hospital. Patients not admitted had a lower mortality rate than admitted patients (30-day mortality rate, 2.1% [95% confidence interval [CI] 0.6 to 3.6] versus 11.5% [95% CI 9.8 to 13.2]; odds ratio 0.20 [95% CI 0.09 to 0.45]). Two of the admission criteria did not correlate with a higher mortality rate: CHF of recent onset and myocardial ischemia. Excluding those 2 criteria, the number of admission criteria present correlated with the patient's probability of hospital admission (P < .001), length of hospital stay (P = .014), and 30-day mortality rate (P < .0001). When zero or 1 admission criteria was present, physician clinical judgment did distinguish patients less likely to die in the subsequent 30 days (1.5% [95% Cl 0.2 to 2.8] sent home Versus 10.2% [95% CI 8.5 to 11.9] admitted). When 2 or more admission criteria were present, physician clinical judgment did not distinguish patients less likely to die in the subsequent 30 days (18.2% [95% CI 0 to 42.0] sent home versus 19.4% [95% CI 13.6 to 25.2] admitted). Conclusion: Selected criteria of the AHCPR CHF admission guideline correlate with mortality rate. Combined with physician clinical judgment, they may be useful in the risk stratification of patients with CHF. Selected low-risk patients with CHF identified by the admission criteria who are presently managed in the acute care hospital may be candidates for outpatient management.