Abstract & Commentary
Source: Stelfox HT, et al. Safety of patients isolated for infection control. JAMA 2003;290:1899-1905.
Patient safety and medical errors have become a major focus of the provision of health care during the last decade. Critics of transmission-based precautions question whether isolation affects the quality of care provided to these patients. This study examined whether the isolation of patients to prevent nosocomial transmission of disease leads to patient neglect and medical errors.
This retrospective, case-controlled study examined two separate cohorts of consecutive patients isolated for methicillin-resistant Staphylococcus aureus (MRSA) colonization or infection: 1) a general cohort of patients admitted for any diagnosis; and 2) a disease-specific cohort of patients admitted for congestive heart failure (CHF). The investigators then selected two matched controls for each isolated patient. The MRSA isolation policies used at the study institutions included private rooms, gloves and gowns for all visitors, dedicated equipment, and restriction of the patient to the room except for essential movement. The main outcome measures included several measures of quality of care, patient outcomes, and patient satisfaction.
In the general cohort, 78 isolated patients and 156 control patients were enrolled. In the disease-specific CHF cohort, 72 isolated patients and 144 control patients were enrolled. The cases and controls were well-matched, with very similar baseline characteristics.
The isolated patients had more days without vital signs recorded (5% vs 1%, p = 0.02) or with vital signs not recorded as ordered (51% vs 31%, p < 0.01). Isolated patients also were more likely to have days without physician progress notes (26% vs 13%, p < 0.01). Isolated patients with CHF were far less likely to have a stress test or angiogram if they had angina (14% vs 45%, p < 0.01). Isolated patients with CHF were less likely to have documentation of CHF education (29% vs 51%, p < 0.01) and to have timely follow-up appointments scheduled (24% vs 46%).
Isolated patients were twice as likely as controls to experience adverse events (31 vs 15 events per 1000 days, p < 0.01), with this difference reflecting a greater rate of preventable adverse events (20 vs 3 per 1000 days, p < 0.01), but no difference in non-preventable adverse events (11 vs 12 per 1000 days). Isolated patients were eight times more likely to experience supportive care failures such as pressure ulcers, falls, and fluid or electrolyte disturbances. No difference in total hospital mortality was observed.
In addition, isolated patients expressed greater dissatisfaction with their treatment than controls. More isolated patients lodged unsolicited formal complaints to the hospital than controls (8% vs 1%). These complaints were related to negative perceptions of treatment, access to staff, and communication.
The authors conclude that these results demonstrate a strong relationship between patient isolation and shortfalls in processes, outcomes, and satisfaction. They suggest that isolation policies could be examined to keep the components most important for infection control, and to perhaps discard those most deleterious to patient safety.
Commentary by Jacob W. Ufberg, MD
Perhaps the one positive note in this study for emergency physicians is the fact that in the CHF group, ED treatment of cases and controls was identical. No differences were noted in the rates of patients receiving ECGs, chest radiographs, general bloodwork, and cardiac enzyme measurement.
The decision to admit a patient to an isolation bed frequently is made by the emergency physician, or is made by an admitting physician based on the fact that the ED isolated the patient. For emergency physicians, there are two major take-home points in this study: 1) We should be sure to check on our isolated patients in the ED frequently; and 2) we should be careful to isolate only the patients who truly need it, as isolation clearly carries risk.
Dr. Ufberg, Assistant Professor of Emergency Medicine, Assistant Residency Director, Department of Emergency Medicine, Temple University School of Medicine, Philadelphia, PA, is on the Editorial Board of Emergency Medicine Alert.