This review concluded that telemonitoring appeared to be an acceptable method of monitoring patients with heart failure. However, the reliability of this conclusion is unclear and it should be interpreted cautiously given the risk of missing data, varied types of studies included and risk of bias within the data.

To assess the effectiveness of telemonitoring for patients with heart failure.

MEDLINE was searched up to the end of August 2007 for studies written in English. Search terms were reported. Reference lists of included studies were checked for further studies.

Primary studies that evaluated a telemonitoring technology or device used for assisting patients with heart failure in self monitoring were eligible for inclusion in the review. Self monitoring was defined as monitoring of signs/symptoms and/or weight over time. Studies were excluded if they assessed implantable devices such as pacemakers, used single telephone calls to monitor patients or included mixed populations of patients and did not report data for patients with heart failure separately.

The included studies assessed the following types of interventions: telemonitoring devices (28%); telephone touch pads (21%); video consultation (5%); websites (9%); and combinations of these interventions (37%). Where reported, patients in the included studies varied in terms of their functional class of heart failure ranging from class I to IV; most were between II and IV. Study duration ranged from one to 18 months. The most frequently monitored parameters were weight, blood pressure and heart rate.

The authors did not state how papers were selected for review.

The authors did not state that they assessed validity.

The authors did not state how data were extracted for the review. A summary of the overall effects was extracted for each study.

Studies were grouped by intervention type and data synthesised using narrative methods.

Fifty-six articles (42 interventions) were included in the review. Sample sizes ranged from six to 461 and were not specified in 10 articles (eight interventions).

Device-based technologies (16 articles, 14 interventions, at least 1,022 participants): Four randomised controlled trials (RCTs), five non-randomised controlled trials and seven pre/post studies were included. Intervention effects included increased hospitalisations in eight studies and improved quality of life in five studies. Six studies reported no significant differences between study groups that used these measures.

Telephone touch-pads (12 articles, seven interventions, at least 1,928 participants): Three RCTs, one pre/post study and three studies of unknown design were included. One study reported fewer rehospitalisations and combined events. One study reported that it took a shorter period of time to achieve the target dose of carvedilol.

Video-consultation-based studies (three articles, two studies, 54 participants): One RCT reported no significant differences between study groups. One pre/post study reported both positive and negative effects in mood ratings for the intervention group depending on the severity of heart failure.

Website-based telemonitoring (five articles, four interventions, 116 participants): One RCT, two pre/post studies and one non-randomised trial were included. The RCT reported a reduction in hospital days for the intervention; one pre/post study reported a corresponding increase. The other pre/post study reported beneficial differences in quality of life post intervention. The study of unknown design failed to report any results.

Combination modalities (21 articles, 15 interventions, at least 1,053 participants): Eight RCTs, three non-randomised controlled studies, four pre/post studies and one feasibility study were included. Seven studies reported that interventions were associated with decreased hospital admissions. Six studies reported that healthcare costs (hospitalisation, emergency care) were reduced in the intervention group.

A reduction in costs/charges associated with the telemonitoring intervention were reported for three device-based technologies, two telephone touch-pad systems and five studies that used multimodality interventions.

Telemonitoring appeared to be an acceptable method of monitoring patients with heart failure.

This review assessed a clear research question. Types of study design were defined only briefly and covered a wide range of varying designs. Searches for relevant data were confined to only one database; the authors also checked reference lists of included studies, but relevant data may have been missed. Inclusion of studies was limited to those written in English, which may have increased risk of missing relevant studies. Risk of publication bias was difficult to assess as the authors did not report whether unpublished studies were eligible for inclusion. However, given the types of studies sought and the lack of specific attempts to locate unpublished studies, publication bias seemed likely. Risk of reviewer error and bias was unclear as the reviewers did not report their methods clearly for study selection and data extraction. Reliability of included data was unclear as no assessment of methodological quality was reported. However, given the broad range of studies designs included in the review (some of which failed to use a control, did not report what type of control was used or used a retrospective control group) there appeared to be a risk of bias within studies. Sample sizes were small, which increased risk of bias. Differences between studies extended beyond differences in study design to include differences in study populations, outcome measurement and interventions. The authors' use of narrative synthesis methods appeared appropriate, but variation between studies made clear assessment of effectiveness of telemonitoring difficult. The lack of reported effect sizes and 95% confidence intervals/p values made it difficult to assess the significance of reported effects. Therefore, the reliability of the authors conclusion is unclear and it should be interpreted cautiously given the risk of missing data, varied types of studies included and risk of bias within the data.

Practice: The authors did not state any implications for practice.

Research: The authors stated that further large randomised controlled trials were required to compare different modalities for telemonitoring and evaluate that success and feasibility as part of routine care for patients with heart failure. Future studies should in particular assess hospitalisations, quality of life, functional capacity, self-monitoring ability and cost-effectiveness.

Michael Smith Foundation for Health Research and Canadian Institutes for Health Research.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.