Managing chronic illnesses with remote patient monitoring in primary health care: an overview of overviews
Systematic review
|Published
We summarized systematic reviews of a specific type of RPM that the Norwegian Directorate of Health is most interested in: RPM that is occurring in primary health services, in which pro-vider feedback is included, and not including technologies based on internet, mobile, or tablet applications.
Key message
Remote patient monitoring (RPM) allows for the real-time transmission of health data, evaluation of this data, and appropriate follow-up. This allows providers to monitor the health status of chronically ill patients and quickly adjust treatment regimes, without requiring that patients continually visit providers’ offices.
We summarized systematic reviews of a specific type of RPM that the Norwegian Directorate of Health is most interested in: RPM that is occurring in primary health services, in which provider feedback is included, and not including technologies based on internet, mobile, or tablet applications.
We included 11 randomized controlled trials of patients with diabetes and/or hypertension, from four systematic reviews. Patients were on average in their 50s, 60s, or 70s, and roughly one to two of every 20 patients had at least one additional multi-morbidity.
Based on summaries of each outcome and our assessment of the certainty of the evidence, we have drawn the following conclusions:
- RPM probably makes little to no difference on HbA1c in diabetic patients (types I and II) and on systolic blood pressure in hypertensive patients.
- RPM probably has a small negative effect on the physical component of health-related quality of life.
- RPM may make little to no difference to diastolic blood pressure, cholesterol, number of patients needing hospitalizations or emergency stays, and the mental health component of health-related quality of life.
The specific type of RPM we examined in this review does not appear commonly implemented among people with chronic conditions other than diabetes or hypertension. Evidence of its clinical and health care utilization effectiveness is weak.
Summary
Background
The proportion of Norwegians with chronic conditions is increasing, as is the amount of years they will survive with these conditions. The health care system must move away from a curative perspective and towards a chronic care model: how best can it help patients manage daily life with one or more chronic conditions? How can patients maintain optimal functioning and as good a quality of life for as many years as possible?
Ideally, patient data could be collected unobtrusively and sent frequently to providers, to allow for continuous monitoring and the provision of care before patients’ conditions deteriorate. One technique is remote patient monitoring (RPM), a broad term referring to the remote transmission and evaluation of patient data that provides health personnel with real-time or frequently collected information about a patient’s health condition. This broad term has been fine-tuned by the Norwegian Directorate of Health for this review to refer to interventions occurring within the primary health services and requiring the involvement of providers (as opposed to fully-automated processes).
Evidence of the efficacy of RPM and related strategies has increased significantly, yet we do not know whether previous strategies describe the specific type of RPM in which the Directorate is most interested. A systematic review that assesses both the evidence and the types of strategies used is therefore needed.
Objective
This overview of systematic reviews sought to measure the effectiveness of RPM on clinical and health care utilization outcomes among chronic disease patients.
Method
We conducted an overview of systematic reviews. We systematically searched the literature for systematic reviews and overviews that conducted their own searches in 2015 or more recently. Reviews of randomized controlled trials (RCTs) that included adult patients with cardiovascular disease, diabetes, hypertension, chronic lung diseases, cancer, mental disorders, chronic musculoskeletal disorders, osteoporosis, or impaired vision/hearing were included if they examined the effectiveness of RPM according to our definition of RPM, and reported clinical or health utilization outcomes. Two researchers screened 3373 records at the title and abstract level, and included reviews that contained at least one RCT that met the inclusion criteria. As all systematic reviews included both RCTs that were eligible for our review and RCTs that were ineligible, we included only eligible RCTs from the systematic reviews for further inclusion. We summarized results and displayed these in forest plots, but we conducted no meta-analyses, as our overview of systematic reviews was not a comprehensive identification of all existing RCTs. Our certainty in the primary outcomes was assessed using the Grading of Recommendations Assessment, Development, and Evaluation approach (GRADE).
Results
We included four systematic reviews that together reported on 11 RCTs that met our definition of RPM. Only patients with diabetes (types I and II) and/or hypertension were captured by these RCTs, with average ages from 51 to 73 years. Roughly one to two of every twenty patients had a multi-morbidity, among the RCTs that reported these.
The RCTs lasted from 6 to 12 months, and while all met our definition of RPM, they were heterogeneous with regards to how data was transmitted (from commercial telehealth units to patients’ existing landlines) and who assessed it (providers, monitoring centers, or the devices themselves). In most cases, patients were only followed up with if data values were of concern, so that patients without an assessed need for further medical attention would not be contacted by providers. It was difficult to determine whether the follow-up they received was akin to usual care or was more enhanced, because most RCTs scarcely described usual care. In the most conservative interpretation, RPM patients received the same follow-up as usual care patients but more often (if needed); in the most generous interpretation, RPM patients received not only more contact with providers but also enhanced treatment.
Among our eight primary outcomes, only three were affected by RPM. RPM probably makes little to no difference on HbA1c levels in diabetic patients. Similarly, RPM probably leads to a slight reduction in systolic blood pressure, with questionable clinical meaningfulness. RPM probably has a small negative effect on the physical component of health-related quality of life; the clinical significance of this reduction is again uncertain. We have low confidence in the findings that RPM makes no difference to the remaining five primary outcomes: diastolic blood pressure, cholesterol, number of patients needing hospitalizations or emergency stays, the mental health component of health-related quality of life, and Hospital Anxiety and Depression Scale scores. RPM also showed no effect in 22 of the remaining 23 outcomes.
Discussion
Many of our findings are consistent with reviews of other, broader definitions of RPM. The clinically insignificant reduction to HbA1c may be explained by our RCTs mainly utilizing single-component interventions instead of multi-component interventions. However, if RPM itself is more of a mechanism to facilitate contact with providers at the cusp of patient deterioration, it may be that increased contact is insufficient. Patients with diabetes and/or hypertension may need treatment that focuses on behavioral change in order to improve clinical outcomes such as HbA1c, blood pressure, and cholesterol, and increased contact may not be enough to change behavior.
One identified gap in the research is an understanding of why RPM has a negative effect on quality on life, a finding that has also been reported by previous reviews. Qualitative methods are likely the best tools to explore this question. Other research gaps include the effects of RPM on patients with both chronic physical conditions and psychiatric conditions, who are most often excluded from clinical trials, and the effects of RPM on patients with impaired vision/hearing. The inherent innovative nature of technologies included in RPM, and particularly their ability to be tailored to patients’ capacities and limitations, make RPM seem uniquely able to address the needs of these patient groups – yet we found no RCTs including these patients.
Conclusion
The type of RPM examined in this review is neither particularly commonly implemented nor particularly effective for patients with diabetes and/or hypertension.
