TL;DR
This study surveyed primary care physicians and advanced practice registered nurses (APRNs) in Michigan about their perceptions of consumer-grade health tracking devices and apps. The findings reveal that while providers generally see benefits in these technologies for patient engagement and data review, they also acknowledge significant patient-level barriers like access and usability. Notably, providers who personally use these technologies are far more likely to recommend them to their patients.
What Problem Does This Paper Address?
The proliferation of wearable devices and health-tracking apps has created a wealth of patient-generated health data. While consumer adoption is high, the integration and perception of this data within clinical settings remain underexplored. Specifically, this paper addresses the gap in understanding how healthcare providers, particularly those in primary care, view these technologies. Do they trust the data? Do they see value in it for patient care? What are their concerns, and how does their own use of these technologies influence their recommendations to patients? This understanding is crucial for bridging the divide between personal health tracking and clinical practice.
Methods
The researchers employed a cross-sectional mailed survey design. They sampled 300 primary care physicians and 300 APRNs from Michigan. The survey collected data on:
- Provider use of tracking technologies: Whether they personally used these devices/apps.
- Provider recommendation of technologies: Whether they recommended them to patients for physical activity, diet, or sleep tracking.
- Perceptions of benefits and barriers: Using a 5-point Likert scale, providers responded to statements about data review usefulness, provider trust, patient interest, security/privacy, liability, and perceived patient barriers (age, literacy, access).
- Demographic and organizational characteristics.
Statistical analyses included descriptive statistics, exploratory factor analysis to identify underlying perception constructs, reliability analysis, paired sample t-tests, and multinomial logistic regression to examine relationships between variables.
Datasets
Not discussed in the paper. The study utilized a survey administered to healthcare providers.
Key Insights and Ideas
- Provider adoption drives recommendation: The most striking finding is the strong correlation between a provider's personal use of health tracking technologies and their likelihood of recommending them to patients. This suggests that firsthand experience is a powerful influencer.
- Trust in data is present, but patient barriers loom large: Providers generally trust the data from these devices and see its utility in clinical encounters. However, their primary concerns shift to the patient's side, focusing on access, usability, and technical literacy.
- Privacy is not a major provider concern (yet): Interestingly, providers in this study did not express significant concerns about data privacy or liability when recommending these technologies. This might change if the data were more formally integrated into electronic health records.
- APRNs and physicians are largely aligned: The study found no significant differences in perceptions between physicians and APRNs, suggesting a unified perspective on these technologies within primary care.
Key Results
- Response Rate: 189 providers completed the survey, yielding a response rate of 33.6%.
- Provider Demographics: The sample was predominantly female (70.4%), Caucasian (89.1%), and split between physicians (43.4%) and APRNs (56.6%).
- Provider Use: Nearly half (49.2%) of respondents reported owning or using behavioral tracking technologies themselves.
- Recommendation Likelihood: Providers who used these technologies were significantly more likely to recommend them:
- 6.4 times more likely for physical activity tracking.
- 6.5 times more likely for diet tracking.
- 3.0 times more likely for sleep tracking.
- Perceived Usefulness: Providers found these technologies most useful for tracking physical activity (55% very/extremely useful), diet (48%), and vital signs (43.9%). They were perceived as less useful for sleep (16.9%), smoking (18%), mental states (11.8%), and alcohol/drug use (14.7%).
- Perception Factors: Five key perception factors emerged: data review usefulness, provider trust, patient interest, security/liability, and perceived patient barriers. All were significantly different from a neutral stance.
Limitations and Future Work
The authors acknowledge several limitations:
- Recall Bias: As with any survey, participants' recall might be imperfect.
- Geographic Scope: The study was limited to Michigan, and a national sample would provide broader generalizability.
- Response Rate: The 33.6% response rate, while noted as a trend in healthcare provider surveys, is still a limitation.
- Sample Demographics: The sample was predominantly White and female, which may influence findings as these groups tend to participate more in surveys.
Future work suggested includes:
- Replication with a national sample: To increase generalizability.
- Exploring differences within organizations: To understand how organizational factors might influence perceptions.
- Investigating the effectiveness of sleep tracking: Given the lower perceived usefulness, further research into why this function is less valued could be beneficial.
- Examining the impact of EMR integration: The study hints that standardized data integration into EMRs could increase provider concerns about privacy and liability. Future research could explore this.
- Direct-to-provider marketing: The authors suggest device makers consider advertising directly to providers, given their gatekeeper role.
Summary updated