Purpose: To compare quality of care, provider morale, and patient/relative satisfaction in one high- and one low-intensity academic medical center’s (AMC) medical intensive care unit.
Method: We conducted a mixed-methods case study of two U.S. AMCs in the same state and health care system to hold regulatory/financial factors constant. We measured quality using JCAHO protocol evaluation of compliance with process measures for the prophylaxis of ventilator-associated pneumonia (VAP), stress ulcer disease (SUD), and deep venous thrombosis (DVT). We measured provider morale using the Maslach Burnout Inventory (MBI) and Areas of Worklife (AoW) survey, and patient/relative satisfaction using the Guyatt satisfaction with treatment intensity and involvement in decision-making instrument.
Results: We observed treatment of 80 patients at the low-intensity 16-bed unit and 81 patients at the high-intensity 24-bed unit over one month at each AMC. Among the 25 patients at the low-intensity and 58 patients at the high-intensity AMC assessed during spot-checks of process measure compliance, VAP prophylaxis compliance was lower at the low-intensity AMC (76% vs. 100%, p<.001), but SUD (100% vs. 100%) and DVT (100% and 100%) prophylaxis did not differ. 132/228 (58%) of ICU providers from the low-intensity AMC and 50/128 (39%) from the high-intensity AMC completed a survey. Providers from the low-intensity AMC scored higher on the MBI emotional exhaustion subscale (23.4 vs. 19.8, p=.62), but did not differ on the depersonalization (7.5 vs. 6.5, p=.31) and personal accomplishment (35.5 vs. 34.5, p=.044) subscales. Provider scores did not differ on the AoW workload (3.4 vs. 3.3, p=.64), control (3.3 vs. 3.4, p=.28), rewards (3.2 vs. 3.2, p=.08), community (3.5 vs. 3.4, p=.32), fairness (2.9 vs. 3.0, p=.86), and values (3.3 v. 3.4, p=.64) subscales. Surveys were completed on behalf of 9/15 (60%) eligible patients at the low-intensity and 8/26 (31%) at the high-intensity AMC. Competent patients (115 vs. 111.3, p=.82), relatives of competent patients (172.3 vs. 189.7, p=.82), and relatives of incompetent patients (139 vs. 100.3, p=.51) did not differ in their satisfaction with treatment intensity and decision-making involvement.
Conclusion: We did not find support for the hypotheses that hospitals with higher end-of-life treatment intensity result in worse compliance with process of care quality or provider burnout or perceptions regarding their workplace. Comparisons of patient/relative satisfaction are inconclusive due to insufficient power.
Candidate for the Lee B. Lusted Student Prize Competition