Most interventions have a positive effect on patient handoffs from hospital discharge back into primary care, but trials examining them are of such mixed sizes and types that it is extremely difficult to compare them, researchers say.
A review of randomized trials of transition-of-care interventions found great variability in methodologies, patient populations, outcome measures, and the interventions themselves. It is impossible, at least at present, to identify which approaches are the most effective, wrote Gijs Hesselink of the Scientific Institute for Quality of Healthcare, Radboud University, Nijmegen, the Netherlands, and his associates. They reported their findings in the Sept.17 issue of Annals of Internal Medicine.
©claudiobaba/iStockPhoto.com
It is very difficult to narrow down which interventions have the most impact on patient handoffs.
"There is no strong evidence that a single intervention is regularly associated with positive effects on a specific outcome measure," they noted.
Mr. Hesselink and his colleagues performed a systematic review of 36 randomized clinical trials performed between 1990 and 2011 that assessed interventions aimed at improving adult patients’ transitions from the hospital to primary care. They excluded studies of psychiatric patients and pregnant women and included studies of patients with general medical, surgical, heart failure, geriatric, stroke, or breast cancer diagnoses.
The studies included patients who had stayed in urban, secondary, tertiary, teaching, and university-affiliated hospitals. The sample sizes ranged from 20 subjects to 1,098 subjects.
The following factors were assessed in these clinical trials: the quality of the information exchanged between the hospital and primary care providers, including the completeness, accuracy, and clarity of that information; the quality of the coordination of health care, including patient assessment, care planning, and organization of follow-up services; and the quality of communication, including personal contact, accessibility, and time lines.
"Because of heterogeneity of the study designs, participants, and outcome measures, meta-analysis was not possible," the researchers said.
Overall, the quality of the 36 trials was judged to be "relatively high." However, 12 of the studies were not blinded, and the blinding status was unclear in another 10. And, in 10 studies, the intervention group was not similar to the control group at baseline.
The interventions were found to be significantly more effective than usual care for at least 1 outcome in 25 of the 36 trials, Mr. Hesselink and his associates said (Ann. Intern. Med. 2012;157;417-28).
In particular, discharge interventions that focused on structuring and reconciling discharge information, coordinating follow-up care, and direct, timely communication between hospital and primary providers were effective at reducing rehospitalization and emergency department visits and at improving patient satisfaction and quality of life.
A total of 22 trials examined interventions aimed at improving the quality of information exchanged at hospital discharge, and 14 of them showed a statistically significant improvement in quality of care, errors, and adverse events. Similarly, 27 trials examined interventions aimed at improving coordination of care, and 20 of them showed a significant improvement in continuity of care, early assessment of patient needs and resources, and postdischarge contact with the patient.
A total of 31 trials examined interventions aimed at improving communication, such as use of liaison nurses, liaison pharmacists, faxes, or e-mail to quickly transmit discharge summaries, plans, and other information. And 22 of them showed a significant improvement in patient status, rehospitalization, and appropriate use of postdischarge primary care.
"More attention should be directed to developing standardized measures of continuity of care for a better evaluation of, and comparison between, discharge interventions," the investigators said.
This study was supported by the European Union. No industry-related conflicts of interest were reported.