Postdischarge efforts fail to reduce readmissions

Two separate projects to reduce patients' hospital readmissions within 30 days of discharge by providing more postdischarge support reported negative results last week.

In the first study, researchers at a safety net hospital in Northern California randomized 700 low-income patients age 55 years and older who were being discharged in the community to either usual care or an intervention. The intervention included in-hospital, individual, self-management education by a registered nurse speaking the patient's language (English, Spanish, or Chinese) with postdischarge telephone follow-up by a nurse practitioner. Results were published in the Oct. 7 Annals of Internal Medicine.

The intervention didn't reduce the combined outcome of ED visits and readmissions within 30, 90, or 180 days after discharge. There was a trend toward more ED visits in the intervention group, although the difference was not statistically significant. The study may have been limited by the fact that the hospital's patients already had good access to primary care and relatively low readmission rates or that the study nurses were not integrated with the patients' other clinicians. The project was an adaptation of an intervention that had worked successfully elsewhere, so the negative findings “should give pause to hospitals adopting interventions shown to work in dissimilar populations and settings without evaluating their effect,” the study authors concluded.

The author of an accompanying editorial noted that conducting an improvement project as a randomized, controlled trial prevents researchers from changing any parts of the intervention that aren't effective, and he suggested that concentrating on high-risk patients, providing different educational materials to health-illiterate patients, and involving patients' families could have led to better results.

In the second study, 1,923 adult patients at a Toronto hospital who were seen as high risk for readmission were randomized to usual care or what the study called a “virtual ward” intervention. The intervention included several weeks of postdischarge care coordination and direct care (via telephone, home, and clinic visits) by an interdisciplinary team. Outcomes were readmission, death, ED visits, or nursing home admission within 30 days, 90 days, 6 months, and 1 year. Results were published in the Oct. 1 Journal of the American Medical Association.

There were no significant differences between the groups on any of the outcomes, although the intervention appeared to be associated with a small benefit at 30 days. Based on the results, the researchers concluded the intervention is highly unlikely to be an efficient use of health care resources. They noted that their effort was limited by a lack of integration and communication between the study personnel and the patients' outpatient and inpatient clinicians. It is possible that the model could work in a more integrated system, they noted. An accompanying editorial said that the data are useful for developing new approaches and that elements of this model could be included in future readmission prevention efforts.