According to a report conducted by medics working at Cedars Saini in Los Angeles, U.S., many commercially available devices lack high-quality research to support the claims made in terms of accuracy, in relation to healthcare.
The wearables market is very lucrative, with positive growth as shipments grew 10.3 percent year over year (based on 2017 data) to reach $26.3 million during the second quarter of 2017 alone, according to data produced by the International Data Corporation (IDC) Worldwide Quarterly Wearable Device Tracker survey.
The new research has taken account of wearable biosensors, including wristbands and watches designed to count steps or to track the quality of sleep. Some devices make claims about vital signs, such a blood pressure, or provide data that users apply in terms of weight loss programs. Biosensors have been integrated into a variety of platforms, including watches, wristbands, skin patches, shoes, belts, textiles, and smartphones. Many wearables transmit data to a web portal or mobile app.
Mobile health has great potential. The devices are unimpeded by geographical boundaries and they are built around real-time data streams. This can lead to automated clinical decision–support tools which can enhance healthcare.
The data can also be used for patient self-monitoring or health provider assessment. Some of the information collected is used by insurance companies (in the competitive U.S. insurance sector) to calculate premiums. Data is also used by medical experts to assess the effectiveness of new treatment plans or medical devices. But what is the integrity of the data and how reliable is it?
According to lead researcher Dr. Brennan Spiegel: “As of now, we don’t have enough evidence that they consistently change clinical outcomes in a meaningful way.” This uncertainty has triggered the latest investigation.
The review conducted by Dr. Spiegel revealed that remote patient monitoring with conventional biosensors had no statistically significant impact. This was assessed by examining six clinical outcomes. These were: body mass index, weight, waist circumference, body fat percentage, systolic and diastolic blood pressure.
However, the analysis conducted discovered that the devices do have the potential for improving certain conditions like obstructive pulmonary disease, Parkinson’s disease, hypertension and low back pain.
The researchers conclude that how significant these outcomes are depends on why are user has purchased a device and what she or he wants to do with it. For instance, there is a conceptual difference between using sensors to track sleep in order to improve the quality of one’s sleeping patterns to gathering data for clinical decisions.
To gather the data the researchers used a statistical tool and conducted a review of 27 independent studies undertaken across 13 countries, between 2000 and 2016. The devices examined included physical activity trackers, blood pressure monitors, electrocardiograms, electronic weight scales, accelerometers and pulse oximeters (oxygen saturation detectors). The devices took the form of watches, belts, skin patches and textiles.
The analysis outcome was that such remote patient monitoring led to no significant impact on any measured clinical outcome. The primary reason was the absence of meaningful data and a lack of research of sufficient quality from many device manufacturers. This was shown by the low number of randomized controlled trials that had been conducted.
Wearable technology is continuing to improve; however, to gain more robust data users of the devices need to agree to take part in studies and there are an insufficient pool of volunteers to help make this happen, a conundrum pointed out by Healthcare Analytics News.
The research has been published in a new journal called npj Digital Medicine. The research paper is titled “Impact of remote patient monitoring on clinical outcomes: an updated meta-analysis of randomized controlled trials.”
Essential Science
This article is part of Digital Journal’s regular Essential Science columns. Each week Tim Sandle explores a topical and important scientific issue. Data-driven science, the interdisciplinary field of scientific methods, is increasingly being used by governments and businesses to develop policy and strategy. Last week we looked at some examples of data science in action.
The week before we looked at hydrogen as a power source. Hydrogen can be produced at a low cost; it is relatively efficient to other fuels; and it is low polluting.
