Who Isn’t Using Patient Portals And Why? Evidence And Implications From A National Sample Of US Adults

The United States has made substantial federal investments in telehealth infrastructure over the past decade, including through the American Recovery and Reinvestment Act of 2009.^1–3 Technologies such as online patient portals, which provide secure internet access to medical records and test results in addition to email communication with providers, can improve health care quality. Evidence thus far shows that access to online portals increases patients’ engagement and adherence^4–7 and may reduce unnecessary utilization and spending.^8,9 While the majority of adults in the United States believe that online access to personal health information is important, disparities in portal access exist. Findings from multiple studies that analyzed different population groups, including nationally representative samples, consistently show that members of racial and ethnic minority groups, older patients, and people of lower socioeconomic status are less likely than others to access an online portal.^10–13 Inequities in access to new and beneficial technologies can exacerbate existing disparities in health.^14,15

One important barrier to patient portal use is finding the time for health care providers to discuss it with patients and encourage them to access it. A 2012 Robert Wood Johnson Foundation report notes that “resistance from physicians to add one more item to talk about in an office visit” limits discussion of portal benefits and patient concerns.¹⁶ Even subtle differences in patient-provider communication or in the usability of health information across patients have the potential to affect disparities in health.^15,17,18

However, even when offered access, most patients do not use portals. One national study found that only about 15–30 percent of patients who were offered access to a portal used it, with lower use among people living in rural and high-poverty areas.¹⁹ Other studies have found that male, nonwhite, and Spanish-speaking patients, as well as those who have the lowest incomes or public insurance, were less likely than their counterparts to ever access a portal.^12,13

Few large-sample studies have explored the reasons for nonuse of patient portals, but those that have done so indicate additional barriers for disadvantaged groups. Technological barriers (such as lack of home internet access)¹⁹ and concerns about privacy and information security²⁰ can impede the use of health information technologies, including online portals.²¹ Yet recent narrowing of the demographic divides in internet access and increasing familiarity with telehealth may lower these barriers.^4,22 To identify appropriate levers that can be used to address inequities in online portal access, policy makers and providers must have a clear understanding of who is and is not accessing portals, as well as the reasons for not accessing them.

In this study we used nationally representative data from 2017 to address two questions. First, what are the characteristics of people who report not being offered access to a portal and of those who report not using a portal in the previous twelve months? Second, among those who did not use a portal, do specific reasons for not doing so vary by demographic or socioeconomic characteristics, type of health insurance, or having a regular provider? By identifying who is not using portals and why, we sought to uncover barriers and reduce disparities.

Study Data And Methods

Data Source

We used data from the 2017 Health Information National Trends Survey 5 (HINTS 5) Cycle 1, administered to 3,285 US adults by the National Cancer Institute in the period January–May 2017 to study people’s access to health-related information. Additional information about the survey design, including its weighting to allow nationally representative estimates, is available on the HINTS website.²³

Study Sample

We restricted our analysis to people who reported making a medical visit during the year before the survey (84.5 percent of the sample) and who reported having health insurance coverage (94.0 percent of the sample). Previous research on nationally representative samples has not specifically addressed whether different types of insurance play a role in patient portal use. Of the 2,682 respondents who met our criteria, we analyzed information about only those 2,325 with complete cases. Many of the incomplete cases (164) were missing only information about race. For analyses examining the reasons for nonuse, of the 1,684 respondents who reported not accessing the patient portal and were asked why not, we included the 1,340 who had complete cases. Once again, many (116) of the incomplete cases resulted from a missing value for race. In sensitivity analyses available in the online appendix (supplementary exhibits 1 and 2),²⁴ we show that our findings were robust to the inclusion of these incomplete cases.

Measures

We identified who was not offered access to a portal using responses to two questions: “Have you ever been offered online access to your medical records by your health care provider?” and “Have you ever been offered online access to your medical records by your insurer?” We counted respondents who answered “no” or “don’t know” as not having been offered access to their online medical records and those who answered “yes” as having been offered access.

We identified portal users based on responses to the question (asked of all respondents), “How many times did you access your online medical record in the last twelve months?” We counted respondents who answered zero as nonusers and those who answered one or more as users.

HINTS 5 asked nonusers whether any of five reasons explained why they had not accessed an online medical record in the past twelve months: no way to access the website (that is, no internet), not having such a record, preferring to speak directly to a provider, having concerns about the privacy or security of the website, and having no need to access the record.

We included as covariates in multivariable models measures of respondents’ sex, race/ethnicity, age, education, employment status, location (urban or rural), insurance type, and having a regular provider.

Statistical Analysis

We first examined the bivariate associations of the covariates with not having been offered access to a portal and not having used a portal. We then estimated two separate multivariable logistic regression models to predict the factors associated with not having been offered access and those associated with not using a portal. We also estimated a third multivariate model for not using a portal among respondents who reported having been offered access to one. In all models, we adjusted for demographic and socioeconomic characteristics, health insurance type, and having a regular provider. To identify patterns in reasons for nonuse among only the nonusers, we estimated five multivariable logistic regression models using the same covariates as in the models above, with one model estimating each reason for nonuse.

Limitations

Our study had several limitations. First, the cross-sectional survey design of HINTS prevented us from making or testing causal claims.

A second important limitation is that the measures of being offered access to a portal and using one were based on the reports of respondents, which could be inaccurate.

A third, and related, limitation is that these measures did not capture the full process by which patients gain access to or use portals. Future research should seek to examine longitudinal data on not only portal usage but also the factors that facilitate (or not) patients’ access, as well as how, when, and why (or why not) they use portals.

Finally, though the HINTS data are nationally representative, the survey does not have large samples of all population subgroups, particularly some racial/ethnic minority groups (for example, Asian Americans). Future research should seek to understand whether disparities exist across all racial and ethnic minority groups.

Study Results

We found that in 2017, 63 percent of insured adults with a health care visit in the previous twelve months reported not using an online patient portal (exhibit 1). Bivariate analyses show that compared to users, nonusers were more likely not to have been offered access to a portal and to be male and age sixty-five or older, have less than a college degree, not be employed, live in a rural location, be on Medicaid, and not have a regular provider.

Ninety-five percent of those who did use a portal said that they were offered access to one, and only 5 percent of those who said they were not offered access reported using a portal. Being offered access may therefore act as a proxy for whether the provider has a portal. However, it may simply be that respondents who accessed a portal are more likely to recall being offered that access. In addition, the fact that 41 percent of those offered access did not use a portal indicates some variance to explain in both outcomes. Given the systematic differences that previous research has found in both offers and access,^10,19 we explored factors associated with not being offered a portal and portal nonuse to fully consider the implications for disparities.

Characteristics Of People Not Offered And Not Using A Portal

Relative to females, males had significantly higher odds of not being offered access to and not using a portal, but among only those who were offered access, males were no more or less likely than females to use a portal (exhibit 2). Members of racial minority groups (specifically, non-Hispanic blacks and non-Hispanics of other races—including Asian Americans, Native Americans, Native Hawaiians, and Pacific Islanders) had significantly greater odds of not being offered a portal. Among only those who were offered a portal, these groups reported rates of using a portal comparable to the rate of non-Hispanic whites.

People with only a high school diploma or less were significantly less likely than those with college degrees to have been offered access to a portal. Among those offered access, those without a college degree and those with a high school diploma or less were significantly more likely not to use a portal, compared to college degree holders. Patients with Medicaid insurance were significantly more likely to report not having been offered access to a portal and not using one, compared to people with other insurance. Patients who lacked a regular provider were significantly more likely to report not having been offered access and not using a portal, including among those offered access. We found no significant variation across age groups or by rurality in being offered access or using a portal.

Reasons For Not Accessing A Portal

Among all who did not use a portal, we found no demographic differences in who was likely to cite a technological barrier, having no internet or having no online medical record, as a reason for not using a portal (exhibit 3). Medicaid and Medicare beneficiaries were significantly more likely than the privately insured to say that they had not used a portal because they preferred to speak directly to a provider.

Patients older than age forty and non-Hispanics of other races were more likely than younger and non-Hispanic white patients, respectively, to report concerns about privacy and security as a reason for not accessing a portal.

Finally, Hispanics and people older than age fifty were less likely than non-Hispanic whites and younger patients, respectively, to say that they had no need to access a portal.

Discussion

Despite significant investments in telehealth to engage patients in their care, about two-thirds of insured US adult patients were not using an online patient portal in 2017. Nonusers were more likely to be male, have less than a college degree, be on Medicaid, and lack a regular provider. These factors, along with race, were also related to whether a patient reported receiving an offer to use a portal.

In considering reasons why people did not use a portal, we found no evidence of disparities in technological barriers—in contrast to previous evidence that information technology use varies with socioeconomic status.^12,25,26 However, respondents reported other reasons for nonuse. For example, some nonusers, particularly those on Medicare or Medicaid, said that they preferred to speak directly to their provider. Other research has found that some patients fear that portals will erode relationships with providers.²⁷ Today the “digital divide” between advantaged and disadvantaged groups appears to be less about access to technological infrastructure (as indicated in our findings) than about differences in knowledge about and skills and comfort in using technology.^11,28–30 Addressing barriers, then, will require more than just increasing access to technology; it will require education as well as technological designs that enable patients to communicate with providers in acceptable ways.^31,32

Provider communication skills may be key to addressing another important reason for not using a portal cited by older patients and some members of racial/ethnic minority groups: concerns about privacy and information security. Privacy concerns can influence health care seeking and sharing of information with providers.^20,33 Providers who can elicit and allay patient concerns about privacy may not only facilitate patients’ use of online portals but also enhance patient trust in and communication with providers.

Provider communication is also important to patients who receive offers to use portals. Previous research has found that groups whose members are less likely to receive initial support for technology can be particularly disadvantaged.^17,27 This is supported by our finding that patients without a regular provider are more likely to be nonusers and to not receive offers to use a portal. Previous research on patients with diabetes has found greater use of online portals (even among some disadvantaged groups) among those with trusting provider relationships.^15,17 Reducing disparities in portal use will require that providers, particularly those serving vulnerable populations, communicate with all patients about portal use and have the capacity to discuss these technologies with patients.

Policy Implications

Our findings have several policy implications. First, socioeconomic and other disparities exist in the use of online patient portals—an important new technology with real potential to improve health and health care. Patients with the lowest education levels, those insured by Medicaid, and those without a regular provider are less likely to report that they were offered access to an online portal or that they used one. In addition, members of racial and ethnic minority groups are less likely to report being offered access to a portal. Only by continuing to track portal use and examine the underlying reasons for and against it can we address disparities across patients. Though the federal meaningful-use objectives require providers to provide access to a certified portal to 80 percent of their patients, the incentives target neither actual portal use nor which patients are using portals. If the objectives are focused on disadvantaged patients specifically, as well as on use (or, better, usefulness), patient portals can be a technology that diminishes health disparities instead of exacerbating them.

A related implication of our findings is the powerful role of providers in facilitating patients’ access to portals. Providers’ communication with patients—a key tenet of patient-centered care—must include discussions of technologies such as online portals that address patients’ concerns, including fears of eroding relationships with providers. In addition, policy makers and payers must recognize, and value accordingly, the amount of time providers need to talk with patients,^16,34 particularly vulnerable and disadvantaged patients, about why and how to be engaged in their care, both online and off.

Finally, practitioners must address issues of confidentiality to reassure patients who are concerned about privacy. Policy makers, health care administrators, and technology developers must also ensure the security and privacy of telehealth systems and infrastructure in the United States.

Conclusion

Using the latest national data, we found that almost two-thirds of insured adults who had had a previous health care visit did not use an online portal in 2017. Those who had only a high school education, did not have a regular provider, and had Medicaid insurance were much less likely to use a portal. Because online patient engagement yields important benefits, it is vital to continue to monitor access to patient portals and related technologies, particularly for disadvantaged groups. Health care providers and plans can increase patients’ use of portals and narrow disparities in that use through direct communication about the benefits of portals, while also addressing patient-specific needs and concerns. Such interventions will require recognition that providers’ communication with patients takes time—an extremely scarce resource in clinical practice today. New communication practices to facilitate patient engagement work best in the context of engaged providers and trusting provider-patient relationships. Careful monitoring of who is and who is not using new technologies, and why, and designing technologies to address patients’ needs, will help ensure that such innovations do not exacerbate disparities but rather lead to improvements for all.

ACKNOWLEDGMENTS

Denise Anthony was partially supported by the National Science Foundation (Grant No. TWC SBE CNS-1408730). The authors thank the editors and reviewers at Health Affairs for helpful advice on the manuscript.

NOTES

• 1 Blumenthal D, Tavenner M. The “meaningful use” regulation for electronic health records. N Engl J Med. 2010;363(6):501–4. Crossref, Medline, Google Scholar
• 2 National Learning Consortium. How to optimize patient portals for patient engagement and meet meaningful use requirements [Internet]. Washington (DC): HealthIT.gov; 2013 May [cited 2018 Oct 26]. (Fact Sheet). Available from: http://www.healthit.gov/sites/default/files/nlc_how_to_optimizepatientportals_for_patientengagement.pdf Google Scholar
• 3 Steinbrook R. Health care and the American Recovery and Reinvestment Act. N Engl J Med. 2009;360(11):1057–60. Crossref, Medline, Google Scholar
• 4 Delbanco T, Walker J, Bell SK, Darer JD, Elmore JG, Farag Net al. Inviting patients to read their doctors’ notes: a quasi-experimental study and a look ahead. Ann Intern Med. 2012;157(7):461–70. Crossref, Medline, Google Scholar
• 5 Kruse CS, Bolton K, Freriks G. The effect of patient portals on quality outcomes and its implications to meaningful use: a systematic review. J Med Internet Res. 2015;17(2):e44. Crossref, Medline, Google Scholar
• 6 Sarkar U, Lyles CR, Parker MM, Allen J, Nguyen R, Moffet HHet al. Use of the refill function through an online patient portal is associated with improved adherence to statins in an integrated health system. Med Care. 2014;52(3):194–201. Crossref, Medline, Google Scholar
• 7 Wright A, Feblowitz J, Samal L, McCoy AB, Sittig DF. The Medicare Electronic Health Record Incentive Program: provider performance on core and menu measures. Health Serv Res. 2014;49(1 Pt 2):325–46. Crossref, Medline, Google Scholar
• 8 Baker L, Rideout J, Gertler P, Raube K. Effect of an internet-based system for doctor-patient communication on health care spending. J Am Med Inform Assoc. 2005;12(5):530–6. Crossref, Medline, Google Scholar
• 9 Palen TE, Ross C, Powers JD, Xu S. Association of online patient access to clinicians and medical records with use of clinical services. JAMA. 2012;308(19):2012–9. Crossref, Medline, Google Scholar
• 10 Peacock S, Reddy A, Leveille SG, Walker J, Payne TH, Oster NVet al. Patient portals and personal health information online: perception, access, and use by US adults. J Am Med Inform Assoc. 2017;24(e1):e173–7. Medline, Google Scholar
• 11 Ancker JS, Hafeez B, Kaushal R. Socioeconomic disparities in adoption of personal health records over time. Am J Manag Care. 2016;22(8):539–40. Medline, Google Scholar
• 12 Gordon NP, Hornbrook MC. Differences in access to and preferences for using patient portals and other eHealth technologies based on race, ethnicity, and age: a database and survey study of seniors in a large health plan. J Med Internet Res. 2016;18(3):e50. Crossref, Medline, Google Scholar
• 13 Wallace LS, Angier H, Huguet N, Gaudino JA, Krist A, Dearing Met al. Patterns of electronic portal use among vulnerable patients in a nationwide practice-based research network: from the OCHIN practice-based research network (PBRN). J Am Board Fam Med. 2016;29(5):592–603. Crossref, Medline, Google Scholar
• 14 Arcaya MC, Figueroa JF. Emerging trends could exacerbate health inequities in the United States. Health Aff (Millwood). 2017;36(6):992–8. Go to the article, Google Scholar
• 15 Lyles C, Schillinger D, Sarkar U. Connecting the dots: health information technology expansion and health disparities. PLoS Med. 2015;12(7):e1001852. Crossref, Medline, Google Scholar
• 16 Aligning Forces for Quality. Lessons learned: the value of personal health records and web portals to engage consumers and improve quality [Internet]. Princeton (NJ): Robert Wood Johnson Foundation; 2012 Jul [cited 2018 Oct 26]. Available from: http://forces4quality.org/af4q/download-document/5596/Resource-value_of_personal_health_records_and_web_portals_12-000--for_rwjf_review.pdf Google Scholar
• 17 Lyles CR, Sarkar U, Ralston JD, Adler N, Schillinger D, Moffet HHet al. Patient-provider communication and trust in relation to use of an online patient portal among diabetes patients: the Diabetes and Aging Study. J Am Med Inform Assoc. 2013;20(6):1128–31. Crossref, Medline, Google Scholar
• 18 Lutfey K, Freese J. Toward some fundamentals of fundamental causality: socioeconomic status and health in the routine clinic visit for diabetes. Am J Sociol. 2005;110(5):1326–72. Crossref, Google Scholar
• 19 Government Accountability Office. Health information technology: HHS should assess the effectiveness of its efforts to enhance patient access to and use of electronic health information [Internet]. Washington (DC): GAO; 2017 Mar [cited 2018 Oct 26]. (Report No. GAO-17-305). Available from: https://www.gao.gov/assets/690/683388.pdf Google Scholar
• 20 Campos-Castillo C, Anthony DL. The double-edged sword of electronic health records: implications for patient disclosure. J Am Med Inform Assoc. 2015;22(e1):e130–40. Crossref, Medline, Google Scholar
• 21 Sakaguchi-Tang DK, Bosold AL, Choi YK, Turner AM. Patient portal use and experience among older adults: systematic review. JMIR Med Inform. 2017;5(4):e38. Crossref, Medline, Google Scholar
• 22 Pew Research Center. Internet/broadband fact sheet [Internet]. Washington (DC): Pew Research Center; 2018 Feb 5 [cited 2018 Oct 26]. Available from: http://www.pewinternet.org/fact-sheet/internet-broadband/ Google Scholar
• 23 Note that HINTS 5 Cycle 2 added a skip pattern between the questions on being offered and accessing a portal so that its results are not directly comparable to those reported here. Westat. Health Information National Trends Survey 5 (HINTS 5): Cycle 1 methodology report [Internet]. Bethesda (MD): National Cancer Institute; 2017 Jul [cited 2018 Oct 26]. Available from: https://hints.cancer.gov/docs/methodologyreports/HINTS5_Cycle_1_Methodology_Rpt.pdf Google Scholar
• 24 To access the appendix, click on the Details tab of the article online.
• 25 Gell NM, Rosenberg DE, Demiris G, LaCroix AZ, Patel KV. Patterns of technology use among older adults with and without disabilities. Gerontologist. 2015;55(3):412–21. Crossref, Medline, Google Scholar
• 26 Arcury TA, Quandt SA, Sandberg JC, Miller DP Jr, Latulipe C, Leng Xet al. Patient portal utilization among ethnically diverse low income older adults: observational study. JMIR Med Inform. 2017;5(4):e47. Crossref, Medline, Google Scholar
• 27 Lyles CR, Allen JY, Poole D, Tieu L, Kanter MH, Garrido T. “I want to keep the personal relationship with my doctor”: understanding barriers to portal use among African Americans and Latinos. J Med Internet Res. 2016;18(10):e263. Crossref, Medline, Google Scholar
• 28 Bailey SC, O’Conor R, Bojarski EA, Mullen R, Patzer RE, Vicencio Det al. Literacy disparities in patient access and health-related use of internet and mobile technologies. Health Expect. 2015;18(6):3079–87. Crossref, Medline, Google Scholar
• 29 Tieu L, Schillinger D, Sarkar U, Hoskote M, Hahn KJ, Ratanawongsa Net al. Online patient websites for electronic health record access among vulnerable populations: portals to nowhere? J Am Med Inform Assoc. 2017;24(e1):e47–54. Medline, Google Scholar
• 30 Sarkar U, Karter AJ, Liu JY, Adler NE, Nguyen R, López Aet al. Social disparities in internet patient portal use in diabetes: evidence that the digital divide extends beyond access. J Am Med Inform Assoc. 2011;18(3):318–21. Crossref, Medline, Google Scholar
• 31 Lazard AJ, Watkins I, Mackert MS, Xie B, Stephens KK, Shalev H. Design simplicity influences patient portal use: the role of aesthetic evaluations for technology acceptance. J Am Med Inform Assoc. 2016;23(e1):e157–61. Crossref, Medline, Google Scholar
• 32 Otte-Trojel T, de Bont A, Rundall TG, van de Klundert J. What do we know about developing patient portals? A systematic literature review. J Am Med Inform Assoc. 2016;23(e1):e162–8. Crossref, Medline, Google Scholar
• 33 Stablein T, Hall JL, Pervis C, Anthony DL. Negotiating stigma in health care: disclosure and the role of electronic health records. Health Sociol Rev. 2015;24(3):227–41. Crossref, Google Scholar
• 34 Kaplan RS, Haas DA, Warsh J. Adding value by talking more. N Engl J Med. 2016;375(20):1918–20. Crossref, Medline, Google Scholar