Four RCTs (858 participants) were included. Three pre-intervention post-intervention studies (number of participants not given) were found.
RCTs.
The readmission rates for moderate-risk patients in one study were 42.2% lower in the intervention group than in the control (P=0.01), whilst in high-risk patients no differences were seen between the comparison groups. However, the time to readmission was shorter with the intervention. In another study, the overall readmission rates were reduced by 44.5% in the treatment group compared with the control (P=0.02), and admissions for heart failure were reduced by 56.2% in the treatment group (P=0.01). In a third study, the readmission rates for medical patients were significantly lower with the intervention at 2 and 6 weeks, but not at 12 weeks. For surgical patients in the same study, there was no significant effect of the intervention. In a fourth study of medical patients, those in the intervention group were less likely to be readmitted two or more times (P=0.05).
In one study, LOS was reduced by 36.6% in the treatment group (P=0.04). In a second study, the average LOS was significantly lower for medical patients at 2 and 6 weeks, but not at 12 weeks. There was no significant effect of the intervention in surgical patients. In a third study, for medical patients, the hospital days per patient were lower in the treatment group (2.3) than for the control (5.0), (P=0.05). There was no significant effect for surgical patients in this study. A fourth study found no significant effect of the intervention on the average LOS.
In one study the quality of life scores were higher in the treatment group than in the control: 22.1 out of 28 versus 11.3 out of 28, respectively (P=0.001).
The effect on mortality in all studies was limited by the short follow-up times.
Pre-intervention post-intervention studies.
One study reported an 85% reduction in hospital admissions (P<0.0001), whilst a second showed a reduced LOS with the intervention (P=0.001). A third study showed that the LOS did not decrease substantially with the intervention.