We used bivariate statistics and logistic regression to compare patients who shared any PFT data vs. patients who did not.
A patient portal was used to invite 79,953 registered portal users to share their data. Of 66,105 users included in our analysis, 499 (0.8%) uploaded data during an initial 37-day study period. Bivariate and regression analysis showed that early adopters were more likely than non-adopters to be younger, male, white, health system employees, and to have higher BMIs. Neither comorbidities nor utilization predicted adoption.
Our results demonstrate that patients had little intrinsic desire to share PFT data with their providers, and suggest that patients most at risk for poor health outcomes are least likely to share PFT data. Marketing, incentives, and/or cultural change may be needed to induce such data-sharing.
Citation: Pevnick JM, Fuller G, Duncan R, Spiegel BMR (2016) A Large-Scale Initiative Inviting Patients to Share Personal Fitness Tracker Data with Their Providers: Initial Results. PLoS ONE 11(11): e0165908. https://doi.org/10.1371/journal.pone.0165908
Editor: Frank van Rijnsoever, Utrecht University, NETHERLANDS
Received: April 19, 2016; Accepted: October 19, 2016; Published: November 15, 2016
Copyright: © 2016 Pevnick et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data described were aggregated from operational sources for this analysis with the approval of the Cedars-Sinai Medical Center Institutional Review Board (CSMC IRB). Although we used de-identified data, we are hesitant to post it publicly due to concerns that it could be re-identified. We will nonetheless entertain specific requests for data extracts sent to Josh Pevnick at firstname.lastname@example.org. Such access would need separate approval by the CSMC IRB.
Funding: Joshua Pevnick was supported by the National Institute On Aging and the National Center for Advancing Translational Science of the National Institutes of Health under awards K23AG049181 and UCLA CTSI KL2TR000122. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. There were no other relevant funding sources, and the authors have no conflicts of interest with this work. The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Background and Significance
In late 2014, 64% of US adults owned smartphones. Most new smartphones can be used with or without complementary PFTs to help automate acquisition of biometric data, including step and stair counts, body weight, heart rate, and blood pressure. Furthermore, standalone PFTs are increasingly used by the general public to track biometric data. Such data has incredible potential for characterizing, tracking, and ultimately improving patients’ health. However, one small study suggested that people with chronic illnesses were much less likely to use such devices, as compared to healthy people. We sought to evaluate the willingness of large populations to share PFT data with healthcare providers via their electronic health records (EHR), which might allow providers to leverage this data to provide better healthcare. After inviting 79,953 patients to sync their PFTs to an EHR for continuous uploading of PFT data, we compared patients who synced their PFTs with those who did not. To develop a priori hypotheses and guide our comparison of demographic, health, and socioeconomic characteristics of both groups of patients, we consulted the Extended Unified Theory of Acceptance and Use of Technology  and existing literature on health information technology and patient portal adoption.
This study utilized the patient portal of CS-Link, Cedars-Sinai Health System’s (CSHS) branding of the EpicCare enterprise EHR product (Epic Systems Corporation, Verona, WI). CS-Link is used at a large, non-profit hospital and at many local associated provider organizations both within and outside CSHS. These organizations provide multidisciplinary care across the care continuum to a socioeconomically diverse population.
To move towards an organizational goal of driving patient engagement and use of an existing patient portal, CSHS began inviting patients to sync PFTs on April 25, 2015. All registered patient portal users received a message entitled “Syncing your Wearable Devices With My CS-Link” stating “You can now sync data from your own personal health devices with your Cedars-Sinai medical record using My CS-LinkTM. Click HERE to learn more.” Users also received an email indicating a new portal message. All applications and devices compatible with the Apple HealthKit, Fitbit, and Withings software frameworks were supported.
We began our analysis by reviewing the literature to generate a priori hypotheses regarding which patient characteristics would predict device syncing.[4–11] Because we found few studies of PFT adoption, we also extrapolated from study of health information technology in general, and of patient portals specifically. Our review suggested that compared to non-adopters, patients choosing to share PFT data would be more likely to be younger, white, non-Hispanic, English-speaking, insured, higher income, employed, sicker, and higher utilizers of healthcare.
Next, we determined which patients had uploaded PFT data via synced devices by June 1, 2015. We queried the EHR and other electronic sources of administrative data to obtain the necessary data elements. We excluded patients without any portal logins during January 1, 2015–April 24, 2015, patients not 18–89 years old, and patients with at least one missing or clearly erroneous data element. We used bivariate statistics to compare characteristics of patients who uploaded any data vs. patients who did not. We then constructed a logistic regression model using complete case analysis (Stata, College Station, TX). The analysis received no explicit external funding. The CSHS Institutional Review Board approved this analysis with a waiver of informed consent based on its use of de-identified data.
Of 79,953 registered portal users invited to sync devices over the 37 days of our study period, we excluded: 8019 users without any portal logins during January 1, 2015–April 24, 2015; 513 users over 89 years old; 2809 users under 18 years old; and 2507 users with missing or clearly erroneous data. Of 66,105 remaining users, 499 (0.8%) uploaded data during the first 37 days of this initiative. Bivariate analysis showed that these early adopters were more likely than non-adopters to be younger, male, and white, lower income, CSHS employees, and to have higher body mass indices (BMI). Each aforementioned association except income was robust to multivariable regression, which also showed early adopters were more likely to be insured and non-Hispanic (Table 1). Neither Charlson comorbidity index nor CSHS utilization (as measured by cost, hospitalizations, or outpatient encounters) predicted adoption.
In this large-scale initiative inviting patient portal users to sync PFTs with an EHR, only 0.8% of users uploaded data. This low uptake was likely related to the use of only one invitation via patient portal. Indeed, greater uploading among organizational employees suggests that greater awareness of the initiative would have induced more adoption. To be sure, this might also reflect a better understanding of the potential utility of such data or greater confidence in the ability of the institution to hold such data private. Nonetheless, the low adoption rate demonstrates that there was not substantial intrinsic desire to share PFT data, such that marketing, incentives, and/or cultural change may be needed to increase syncing. Although population-level data suggests that the majority of the patient portal user population likely had access to a compatible device,  addressing users’ willingness and ability to sync devices would likely increase data sharing.
Regarding health status, early adopters were younger and without greater comorbidity or resource utilization, but had higher BMIs. If these higher BMIs were due to adipose tissue stores rather than muscle mass, these higher BMIs may reflect greater potential for health benefit among early adopters. However, our other results, except for the higher adoption rate among males, suggest that those most at risk for poor health outcomes were least likely to share PFT data with providers. A systematic review of patient portals also found decreased use among racial and ethnic minorities. However, it found that higher portal use was associated with worse health status, which contrasts with our finding of less PFT data sharing among those with worse health status.
The Extended Unified Theory of Acceptance and Use of Technology could help to explain such differences, in that hedonic motivation may be responsible for early adoption of PFT data sharing among younger, healthier men. In contrast, effort expectancy, facilitating conditions, social influence, and/or performance expectancy may drive longer term use of patient portals among sicker patients. If this is the case, we should expect to see increased sharing of PFT data among sicker patients as this technology matures. Future research should test this hypothesis by studying data sharing over a longer time period. Other improvements might involve more active recruitment strategies, which could involve multiple communications and specific information regarding the value of sharing PFT data with providers. Researchers should also seek to understand patients’ reluctance to share PFT data, to increase such sharing, and to measure its benefits, especially among those at risk for poor health outcomes.
- Conceptualization: JMP GF RD BMRS.
- Data curation: JMP RD GF.
- Formal analysis: JMP GF.
- Investigation: JMP GF RD.
- Methodology: JMP GF RD BMRS.
- Project administration: JMP.
- Resources: RD.
- Software: RD.
- Supervision: JMP.
- Validation: JMP RD GF.
- Visualization: JMP.
- Writing – original draft: JMP BMRS.
- Writing – review & editing: JMP GF RD BMRS.
- 1. Pew Research Center, April, 2015, “The Smartphone Difference” Available at: http://www.pewinternet.org/2015/04/01/us-smartphone-use-in-2015/ Accessed February 25, 2016.
- 2. Patel MS, Asch DA, Volpp KG. Wearable Devices as Facilitators, Not Drivers, of Health Behavior Change. JAMA. 2015;313(5):459–460. pmid:25569175
- 3. Shaw RJ, Steinberg DM, Bonnet J, et al. Mobile health devices: will patients actually use them? Journal of the American Medical Informatics Association 2016 pmid:26911820
- 4. Venkatesh V, Thong JYL, Xu X. Consumer acceptance and use of information technology: extending the unified theory of acceptance and use of technology. MIS Q. 2012;36(1):157–78
- 5. Ancker JS, Barrón Y, Rockoff ML, et al. Use of an Electronic Patient Portal Among Disadvantaged Populations. Journal of General Internal Medicine. 2011;26(10):1117–1123. pmid:21647748
- 6. Fox S, Duggan M. Mobile Health 2012. Washington, DC: Pew Research; 2012 Nov 8. http://pewinternet.org/Reports/2012/Mobile-Health.aspx. Accessed February 25, 2016.
- 7. Palen TE, Ross C, Powers J, Xu S. Association of Online Patient Access to Clinicians and Medical Records With Use of Clinical Services. JAMA. 2012;308(19):2012–2019. pmid:23168824
- 8. Lee J, Rho MJ. Perception of Influencing Factors on Acceptance of Mobile Health Monitoring Service: A Comparison between Users and Non-users. Healthcare Informatics Research 2013;19(3):167–76 pmid:24175115
- 9. Goldzweig CL, Orshansky G, Paige NM, Towfigh AA, Haggstrom DA, Miake-Lye I, et al. Electronic Patient Portals: Evidence on Health Outcomes, Satisfaction, Efficiency, and Attitudes: A Systematic Review. Ann Intern Med. 2013;159:677–687. pmid:24247673
- 10. Kohnke A, Cole ML, Bush RG. Incorporating UTAUT Predictors for Understanding Home Care Patients’ and Clinician’s Acceptance of Healthcare Telemedicine Equipment. Journal of Technology Management & Innovation 2014 2014;9(2):13
- 11. de Veer AJE, Peeters JM, Brabers AE, Schellevis FG, Rademakers JJJ, Francke AL. Determinants of the intention to use e-Health by community dwelling older people. BMC Health Services Research 2015;15(1):1–9 pmid:25889884