This systematic review demonstrates that a self-management intervention in patients with heart failure decreases hospital readmissions, both all-cause and HF-related. There was no significant effect on mortality or on health related quality of life. However, health behaviors, such as regular weighing and monitoring of symptoms, increased in patients who were enrolled in self-management programs. The decrease in all-cause readmission may be partly due to the decrease in HF-readmissions. Also, as self-management interventions in this study encouraged beneficial health practices, these interventions may have had the auxiliary effect of reducing readmissions for causes other than heart failure.
Results of the individual studies seem to indicate that the self-management interventions with more intensive education components, such as that in Krumholz et al, are more likely to show benefits than those with less intensive education sessions. In the study which showed the least benefit in reducing all-cause readmission [23
], the patients were given educational software and a messaging system but were not otherwise involved in structured follow-up or nurse-led education. In this study, the readmission rates in the intervention group did not significantly differ from those in the usual care group. The studies also involved different degree of contact with medical staff. It is difficult to assess from the available data to what degree, if any, communication frequency during follow up affected the primary outcomes, because the studies that involved more frequent follow up also involved more education. Moreover, it is not clear to whether the follow-up by medical professionals affected patients' health status. Although we excluded the studies in which doctors or nurses assessed patients' health or altered their health regimen, it is possible that contact with medical professionals influenced patients' health outcomes. Further research is needed to conclusively determine the effects of the quantity of self-management education, the method of delivery, and the duration and nature of follow up on health outcomes.
The results of the study by Ross et al. indicate that patients who communicated with medical staff through messaging software were as satisfied with the communication as were the patients undergoing standard care [23
]. These results must be interpreted with caution, since nearly all the participants in this study had access to a computer at home and had previous experience using the Internet, compared to about 50% of those who declined to participate. Future studies should examine the effectiveness of electronic communication in broader patient populations.
This study adds to the current body of literature in that it analyses the effect of self-management interventions on health outcomes of patients with heart failure. The findings of this review are consistent with results from systematic reviews of disease management strategies for patients with heart failure. Multi-disciplinary heart failure management programs that involve specialized follow-up significantly decreased hospital readmission but did not affect mortality rates [6
]. Similarly, a systematic review reported that telemonitoring interventions reported may decrease the number of readmissions as well as reduce mortality and morbidity [25
]. As in the case of the self-management interventions, multidisciplinary heart failure management programs and telemonitoring programs can include several components, and it is difficult to evaluate to what extent they contribute to the overall effect on patients' health.
As with any systematic review, the strength of the results and the extent of the analysis that is possible depend on the strength of and the data reported in the individual studies. Results for heart failure and mortality were each reported in three of the six studies, and thus data was available for less than one half of the patients in the included studies, which may have affected the strength of the results. And, not all studies described details about blinding [22
]. Moreover, the included studies did not consistently report statistics on race or socioeconomic factors such as the level of education. Social conditions such as level of education have been found to be more powerful predictors of health status than many of the risk factors associated with cardiovascular disease [24
]. Insufficient data was also provided for other predictors such as ethnicity and smoking [24
]. Unavailability of such information could influence the results of individual studies as well as the combined result. In the study by Jaarsma et al., the patients who dropped out were significantly older, more frequently lived in the nursing home, were diagnosed with hypertension, and had cardiomyopathy as the underlying cause of heart failure. Differences in attrition between the control and intervention group could influence the results of this study [20