Disparities in patient portal access by US adults before and during the COVID-19 pandemic

Abstract

Objective

Online patient portals become important during disruptions to in-person health care, like when cases of coronavirus disease 2019 (COVID-19) and other respiratory viruses rise, yet underlying structural inequalities associated with race, socio-economic status, and other socio-demographic characteristics may affect their use. We analyzed a population-based survey to identify disparities within the United States in access to online portals during the early period of COVID-19 in 2020.

Materials and Methods

The National Cancer Institute fielded the 2020 Health and Information National Trends Survey from February to June 2020. We conducted multivariable analysis to identify socio-demographic characteristics of US patients who were offered and accessed online portals, and reasons for nonuse.

Results

Less than half of insured adult patients reported accessing an online portal in the prior 12 months, and this was less common among patients who are male, are Hispanic, have less than a college degree, have Medicaid insurance, have no regular provider, or have no internet. Reasons for nonuse include: wanting to speak directly to a provider, not having an online record, concerns about privacy, and discomfort with technology.

Discussion

Despite the rapid expansion of digital health technologies due to COVID-19, we found persistent socio-demographic disparities in access to patient portals. Ensuring that digital health tools are secure, private, and trustworthy would address some patient concerns that are barriers to portal access.

Conclusion

Expanding the use of online portals requires explicitly addressing fundamental inequities to prevent exacerbating existing disparities, particularly during surges in cases of COVID-19 and other respiratory viruses that tax health care resources.

BACKGROUND AND SIGNIFICANCE

The Medicare and Medicaid Promoting Interoperability (PI) program and its predecessors require US providers to implement certified health information technology that enables patient access to their electronic medical records.¹ Similarly the 21st Century Cures Act created technical standards to facilitate patient access to and exchange of their health information.² Online portals are a common tool through which providers offer patients electronic access to their health information. Although most hospitals report having online portals available for patients,³ only about 40% of US adults reported accessing an online patient portal in 2020⁴ and previous research documents evidence of significant socio-demographic disparities in, and barriers to, portal use.^5–9

Of course, 2020 is also when the public health emergency for the coronavirus disease 2019 (COVID-19) pandemic was declared, and when access to health care services was severely disrupted, at least initially. According to social theories of fundamental causes of health, those with greater and more flexible resources (eg, income, knowledge, and social connections) will be better equipped to respond to changing risks, such as adapting to restricted in-person care during a pandemic by using digital technologies to access health care.¹⁰ For example, during the early months of the pandemic, in-person medical visits declined^11–13 while rates of telehealth visits increased rapidly.^14–16 Technologies like patient portals facilitated access to care for some during this period since they enable patients’ online access to their medical records and provided an avenue for scheduling appointments, requesting prescription refills, and other functions including messaging providers. While HIPAA flexibility enabled providers to use a range of nonpublic facing platforms for telehealth visits, patients who already had telehealth integrated into their patient portal were advantaged because the infrastructure was already in place.¹⁷ Moreover, the skills and resources to access and use a patient portal are comparable to those for telehealth.¹⁸ Yet for patients who do not already own an internet-enabled device, know how to go online, or feel comfortable sharing personal information (typically in English) over the internet, such adaptation to restrictions in accessing in-person care is much more challenging than for those already familiar or comfortable with digital technologies. These issues extend beyond just the recent emergence of COVID-19, as any surge in cases of respiratory viruses (eg, influenza and the respiratory syncytial virus [RSV]) can tax health care systems and increase the use of telehealth.¹⁹ Accordingly, because digital technologies create benefits for those who can access and utilize them effectively, they also can increase disparities and this can become particularly problematic when health care systems are trying to limit in-person contact.²⁰

Extant research suggests this need not always be the case, and the fundamental cause of disease theory can explain why. Much research accords with the tenets of the theory. Though the proportion of people both offered and accessing patient portals increased from 2014 to 2020, socio-demographic disparities in, and barriers to, portal use persisted.^4,7 Studies of telehealth access published during the pandemic find similar evidence of disparities.^21–23 Consistent with the theory, access to resources that are patterned by socio-demographics contributes to disparities. Specifically, systematic structural barriers in access to technologies, such as availability of infrastructure (eg, broadband), as well as lack of provider support, and patient differences in device ownership, digital literacy, and social support, contribute to disparities in access to health information technologies.^6,20,24–28 Critically, the degree to which such resources are available or necessary to access health information technologies can modulate disparities in access.²⁰ When access to infrastructure, including computers and the internet, is either secured or unnecessary, studies show socio-demographic disparities in access to some health technologies, such as telehealth, are reduced.^27,29 Accordingly, understanding the state of disparities in access and use of patient portals in general, and particularly in the early period of the pandemic, has important implications for potential and ongoing disparities in access to care and other outcomes during times in which health care systems are attempting to manage their limited resources, as during the pandemic.

Our study investigates whether socio-demographic disparities in access to online portals persist for insured patients using data from the Health Information National Trends Survey (HINTS), a nationally representative survey of US adults fielded from February through June 2020. We focus on portal access among insured adults who had a healthcare visit in the prior 12 months, that is, a population who may be most expected to have a need, opportunity for, and an interest in accessing a portal. They also may be most able to recall being offered a portal. In addition, we identify whether there are socio-demographic disparities in reasons for nonuse among those who report no use of online portals. Other recent studies finding socio-demographic barriers to portal use have studied earlier time periods (2017–2019) and reported either only bivariable analyses or looked at the entire population rather than focusing on patients with a visit.^8,9 Identifying whether disparities in portal access and reasons for nonuse exist among insured patients help to assess existing policies promoting health information exchange and whether expansions of access to care during the early pandemic were indeed affecting access. Documenting disparities also can help guide the refinement and further development of resources, policies, and practices needed to ensure equitable access to and use of technologies to promote healthcare access and enhance health. Such insights are critical for supporting health care access by patients who may continue to be wary of in-person contact and also the broader population of patients when taxed health care systems are responding to surges in cases of COVID-19 and other respiratory viruses.

MATERIALS AND METHODS

Data source and data availability statement

The data underlying this article were derived from sources in the public domain: Health Information National Trends Survey 5 (HINTS 5): Cycle 4, Available at: https://hints.cancer.gov/data/Default.aspx. HINTS is a population-based survey that investigates the American public’s use and access to health-related information. Additional information about the survey design and access to the publicly available data can be found on the HINTS website.³⁰ The 2020 HINTS survey was administered to US adults by the National Cancer Institute in the period February–June 2020. A total of 3865 surveys were completed, corresponding to a response rate of 37%, that is comparable to the response rates for previous HINTS iterations. Briefly, survey weights are available to approximate benchmarks based on the American Community Survey and the National Health Interview Survey, and to also address nonresponse bias. The derived data generated in this research will be shared on reasonable request to the corresponding author. The following sections report the sample selection and survey measures that were used from the 2020 HINTS data.

Study sample

The analytical sample was restricted to those who reported having made a medical visit in the past 12 months (85.9% of the full sample) and reported having health insurance coverage (93.2% of the full sample). From the 3162 respondents who met both criteria, our analytical sample was further restricted to the 2916 respondents who did not have missing responses on any of the study measures.

Measures

Respondents offered access to online medical records were identified by combining responses from 2 survey 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 health insurer?” Respondents who answered “yes” to either question were counted as having been offered access to online medical records, while respondents who answered “no” or “don’t know” to both questions were counted as not having been offered access.

Those who accessed a portal were identified based on response to the question, “How many times did you access your online medical record in the last 12 months?” Respondents who answered “none” were counted as nonusers, whereas those who answered 1 or more were counted as users. All respondents, regardless of responses to the questions about being offered a portal, were asked about the number of times they accessed an online record.

For nonusers, the survey further asked about possible reasons why they had not accessed an online medical record in the past 12 months. The survey offered a total of 8 possible reasons for nonuse. We included 7 of the 8 in the analysis. We exclude the reason “nonuse because of having multiple online records” because it is unclear whether people were reporting nonuse of all available portals. Furthermore, only 3.2% of total respondents reported having multiple records as a reason for nonuse. The 7 reasons for nonuse of a portal that were analyzed include preference for direct communication with a provider, perceived lack of need for a portal, discomfort with technology, not having an online medical record, concerns for the privacy or security of the website, problem with log-in, and inability to access the online record.

Covariates in the multivariable logistic regression models were respondent’s self-reported (binary) sex, age, race and ethnicity, education level, employment status, location (metro or nonmetro), general health, insurance type, access to a regular provider, the quality rating of care received, internet usage, and whether the survey was completed before or after the official declaration of the pandemic (a variable generated by survey administrators, based on whether the survey was received before or after March 11, 2020). Respondents self-reported race and ethnicity, coded as non-Hispanic Asian, non-Hispanic Black, Hispanic, non-Hispanic White, and non-Hispanic of other race and ethnicity (which included those who were American Indian or Alaska Native and Native Hawaiian or Other Pacific Islander, and those who reported multiple races).

Most measures had less than 1% missing values, except for race and ethnicity, which was missing for slightly above 5% of cases. A separate category for “Missing” on race and ethnicity was created to avoid the loss of data and because past studies have found significant associations between race and ethnicity and portal use.⁵ Sensitivity analysis showed that omitting the “Missing” race and ethnicity category did not meaningfully change the reported results.

Statistical analysis

Bivariate associations were calculated between the covariates and being offered a portal and accessing the portal. The data were further analyzed using multivariable logistic regression for associations of measures with being offered and accessing a portal. Finally, we ran separate multivariable logistic regression models for each of the 7 reasons for nonuse of online medical records. All multivariable models include the previously described covariates.

Results

In 2020, 67% of insured adults who made a medical visit in the prior 12 months reported being offered access to an online portal, while approximately 48% reported to have accessed the online portal (Table 1). Bivariate analysis shows that of those who were offered access to the portal, approximately 67% accessed the online portal.

Table 1.

Weighted characteristics of survey respondents

		Offered access to portal
Characteristics	All (N = 2916)	Yes (n = 1951)	No (n = 965)
Accessed online medical record in the last 12 months***
Yes	48.1 %	69.0 %	6.6 %
No	51.9	31.0	93.4
Sex**
Male	47.3	43.4	55.0
Female	52.7	56.6	45.0
Age (years)**
18–30	16.4	16.4	16.5
31–40	15.1	14.7	15.9
41–50	18.5	20.2	15.2
51–64	27.7	29.0	25.2
65 or older	22.2	19.7	27.2
Race/ethnicity**
Non-Hispanic Asian	64.7	68.3	57.6
Non-Hispanic Black	10.3	10.3	10.4
Hispanic	12.9	10.4	17.9
Non-Hispanic Other Race	4.6	4.7	4.2
Non-Hispanic White	3.4	3.6	3.0
Missing on Race	4.1	2.7	6.8
Education***
College or more	34.6	38.9	26.0
Some college	39.1	39.8	37.7
High school or less	26.3	21.2	36.3
Employment status***
Employed	58.6	61.9	51.8
Not employed	41.4	38.1	48.2
Location
Metro	88.5	88.4	88.8
Nonmetro	11.5	11.6	11.2
General health
Excellent, Very good, Good	85.5	86.4	83.8
Fair, Poor	14.5	13.6	16.2
Insurance type***
Private	57.5	62.8	47.0
Medicaid	16.4	12.2	24.7
Medicare	21.4	20.2	23.7
Other	4.7	4.8	4.6
Has a regular health care provider***
Yes	72.9	78.2	62.3
No	27.1	21.8	37.7
Use Internet***
Yes	89.2	94.3	79.1
No	10.8	5.7	20.9
Pandemic
Before	36.1	37.0	34.4
After	63.9	63.0	65.6
	Mean (SE)	Mean (SE)	Mean (SE)
Quality of care***	3.928 (0.032)	4.017 (0.036)	3.751 (0.063)

^** P < .01.

^*** P < .001.

Characteristics of patients offered and accessing online portals in 2020

In multivariable analysis, females had significantly higher odds than males of reporting both being offered and accessing online portals (Table 2). We find no significant differences by race or ethnicity in being offered a portal, unlike previous studies from earlier periods that showed non-Hispanic Black patients and those of other non-Hispanic race were less likely to be offered a portal than non-Hispanic White patients.^5,8 However, in 2020, we find that Hispanic patients have a 49% lower odds of accessing an online portal than Non-Hispanic White patients. Patients with less than a college degree also were significantly less likely to access an online portal. Respondents who rated themselves to have “Fair” or “Poor” health were significantly more likely to access an online portal relative to those who rated their general health better. Patients with Medicaid insurance were significantly less likely to report being offered a portal and had a 47% lower odds of accessing one than the privately insured. Similarly, patients with no regular healthcare provider had a 55% lower odds of being offered and accessing an online portal, while those reporting high quality of care were more likely to be offered a portal than those rating their care as lower quality. Patients without internet are less likely to be offered a portal and have a 75% lower odds of accessing one compared to those with internet access. Notably, the pandemic variable is not statistically significant, indicating that those who took the survey after the pandemic was declared (between March and June 2020) were no more or less likely to report being offered or accessing a portal. Though this is only a short period of time to assess the potential effects of the pandemic on portal use, it is somewhat surprising that not more people were accessing a portal after the start of the pandemic.

Table 2.

Adjusted odds ratios (95% confidence intervals) for having been offered access to and having accessed online medical records in the past 12 months by respondent characteristics, 2020

	Offered access			Accessed online record
	OR	CI (lower)	CI (upper)	OR	CI (lower)	CI (upper)
Sex (ref: male)
Female	1.68****	1.28	2.21	1.55***	1.19	2.01
Race/ethnicity (ref: non-Hispanic white)
Non-Hispanic Black	1.11	0.72	1.7	0.84	0.51	1.4
Hispanic	0.74	0.44	1.25	0.51**	0.31	0.83
Non-Hispanic Asian	0.92	0.51	1.67	1.18	0.61	2.28
Non-Hispanic Other	1.32	0.37	4.67	0.74	0.27	2.02
Missing	0.57	0.28	1.15	0.57	0.29	1.12
Age (ref: 18–30)
31–40	0.88	0.46	1.67	0.77	0.42	1.43
41–50	1.23	0.59	2.55	0.9	0.5	1.62
51–64	0.96	0.5	1.84	0.8	0.5	1.29
65 or older	0.7	0.33	1.46	0.67	0.33	1.39
Education (ref: college or more)
Some College	0.81	0.57	1.14	0.64**	0.46	0.9
High School or Less	0.62*	0.38	1.00	0.44****	0.31	0.63
Employment status (ref: employed)
Not employed	1.05	0.72	1.53	1.11	0.78	1.58
Location (ref: metro)
Nonmetro	1.01	0.68	1.51	0.77	0.54	1.09
General Health (ref: Excellent, Very good, Good)
Fair, Poor	1.28	0.85	1.92	1.42**	1.02	1.99
Insurance type (ref: private)
Medicaid	0.47***	0.29	0.77	0.53**	0.34	0.84
Medicare	0.88	0.52	1.5	0.65	0.34	1.24
Other	0.84	0.45	1.54	0.66	0.35	1.24
Has a regular healthcare provider (ref: yes)
No	0.45****	0.32	0.65	0.45****	0.31	0.64
Quality of care	1.27***	1.08	1.48	1.06	0.9	1.24
Use internet (ref: yes)
No	0.33****	0.2	0.54	0.25****	0.14	0.46
Pandemic (ref: before pandemic)	1.03	0.77	1.39	1.14	0.82	1.58
After pandemic

OR = odds ratio; CI = confidence interval.

^* P < .10.

^** P < .05.

^*** P < .01.

^**** P < .001.

Reasons for not accessing a portal

Among nonusers, the most prevalent reason insured patients reported for not accessing an online portal was having a preference for direct communication with the provider (77.0%). Nearly, half of nonusers (53.5%) reported they had no need for an online portal, suggesting that many patients do not see value in using a portal. Other reasons included discomfort with technology (29.5%), not having an online medical record (29.0%), concerns for the privacy or security of the website (25.6%), problem with log-in (25.2%), and the inability to access the online record (24.3%).

Compared to non-Hispanic Whites, non-Hispanic Black patients were significantly less likely to cite having no online record and perceived lack of need for one, but significantly more likely to report an inability to access the website as reasons for nonuse (Figure 1). These findings on reasons for nonuse suggest that, compared to nonusing White patients, Black patients who do not use a portal did perceive a need for one but face technology barriers to using it. Patients with less than a college education were significantly more likely to cite being uncomfortable with technology, while those with a high school diploma or less were less likely to report concerns with privacy as a reason for nonuse (Figure 1).

Figure 1.

Adjusted odds ratios for reasons for not accessing an online medical record in the past 12 months by education and race/ethnicity, 2020. Note: Plots with bolded circles indicate statistical significance at P < .05 or less.

Relative to males, females were significantly less likely to cite an inability to access the website and a perceived lack of need for a portal as reasons for not accessing a portal (Figure 2).

Figure 2.

Adjusted odds ratios for reasons for not accessing an online medical record in the past 12 months by sex and age, 2020. Notes: Plots with bolded circles indicate statistical significance at P < .05 or less. The age category of 18–40 was used as the reference for Uncomfortable with Technology reason, as there was an insufficient number of responses in the 18–30 age category.

Compared to younger patients, those 41 years or above were more likely to report discomfort with technology as a reason for nonuse, while those aged 51–64 were also significantly more likely to cite concerns about privacy as a reason for nonuse (Figure 2).

Respondents living in nonmetro versus metro areas and those with Medicaid compared to private insurance were more likely to cite having no online portal as a reason for nonuse (Figure 3). Medicaid recipients also were more likely to report an inability to access a website and concerns about privacy as reasons for not accessing an online portal (Figure 3).

Figure 3.

Adjusted odds ratios for reasons for not accessing an online medical record in the past 12 months by location and insurance type, 2020. Note: Plots with bolded circles indicate statistical significance at P < .05 or less.

Respondents with no internet had a higher odds of citing inability to access the website, preference for direct communication with a provider, concerns about privacy, and discomfort with technology as reasons for nonuse (Figure 4). Finally, those with a higher quality of care were less likely to report concerns about privacy as a reason for nonuse (Figure 4). There were no statistically significant associations between reasons for not accessing an online record and employment status, general health, having a regular provider, or survey completed after the start of the public health emergency (see Supplementary Figure S1).

Figure 4.

Adjusted odds ratios for reasons for not accessing an online medical record in the past 12 months by internet use and quality of care, 2020. Note: Plots with bolded circles indicate statistical significance at P < .05 or less.

DISCUSSION

Given the significant ongoing federal investments and policies that promote health information technologies like online portals for patients, we should not be surprised by reports that the proportion of people offered and accessing online portals has increased since 2014.⁴ However, less than half of insured patients reported accessing a portal anytime in the 12 months through June 2020. This was true even among those interviewed after the start of the pandemic (March–June 2020) who may have been encouraged to use a portal given the limitations on in-person medical visits during this time.¹³ Most importantly, although we find somewhat fewer socio-demographic differences in who was offered a portal compared to previous studies,^5,8 our findings indicate that socio-demographic inequities in who is offered and who accesses a portal persist. Patients who are male, Hispanic, have less than a college degree, have Medicaid insurance, have no regular provider, or have no internet, are less likely than their counterparts to report accessing a portal. These results highlight the challenge of ongoing socio-demographic disparities in access to health technologies even amidst the increasing use of those technologies overall, and even during the early part of the COVID-19 pandemic when the use of digital health technologies like telehealth were rapidly expanding. They also demonstrate how those with fewer social resources, based not only on socio-economic factors like education and Medicaid insurance but also related to racially minoritized status and health care attributes like having a regular provider or high-quality care, are less likely to use interventions that may be beneficial. When it is the case that interventions produce real benefits, such as for use of patient portals,^31,32 fundamental cause of disease theory predicts that disparities in access and use of those interventions can exacerbate disparities in health because those already better off gain the benefits of the intervention.²⁰ Such patterns reproduce and extend inequality by reinforcing rather than disrupting underlying fundamental social inequities.³³

Some reasons for not accessing an online portal are related to patient digital literacy (discomfort using the internet), or with individual preferences and concerns: desire to speak directly with a provider, concerns about privacy, and perceived lack of need for an online record. Improving patient access to portals must distinguish among various types and levels of barriers in order to develop appropriate interventions and policies to address them.^6,34 Lyles et al’s model for Digital Health Equity is based on a socio-ecological framework that recognizes community, practice and policy levels, as well as individual and family factors.³⁴ For example, our finding that older patients and those with less education were more likely to cite discomfort with technology as a reason for nonuse of portals, indicates a need for increased attention to digital literacy, either individually or via supporting others who may assist a patient with accessing the portal. However, interventions also may require channeling additional resources to providers who can offer assistance to patients to sign-up for a portal and to learn how it may be beneficial to them.^6,28,35–37 Providers who use alternative care models that include health coaches and other supports may be better equipped to provide such support to patients.

Providers should also be aware of who is likely to be concerned about privacy. Adults aged 51–64 and those with Medicaid insurance were more likely to report privacy concerns as a reason for nonuse. Privacy concerns can limit health-seeking behavior and information sharing with providers.^38–40 Amid the pandemic and the increase in telehealth, privacy may be an important factor in the willingness of some patients to use such digital tools.^21,26 Establishing positive patient-provider relationships that foster trust is an important factor in mitigating privacy concerns.⁴¹ Of note, those with a higher quality of care were less likely to be concerned about privacy as a reason for not using a portal. Past studies have supported this observation and suggested that privacy concerns may be partly addressed if providers focus on the care quality benefits for patients from engagement through an online portal.^40,42

Medicaid patients and patients in rural areas were more likely to cite not having an online record as the reason for not accessing one, which may indicate resource constraints by the providers who serve those patients rather than (or in addition to) lack of technology resources of the patients. Recent evidence of telehealth use by Medicare beneficiaries during the pandemic similarly finds that those in rural areas have lower levels of telehealth use.⁴³ Current policies such as the Medicare and Medicaid PI program which provides financial incentives to providers to implement and improve patient access to health information (among other health information exchange goals), may be insufficient to support all providers, particularly safety net and rural providers. But it is important to keep in mind that technology barriers faced by patients are not addressed at all by federal incentive programs like PI. As suggested by the fundamental cause of disease theory, unless underlying inequalities and resource constraints faced by both patients and some providers are addressed, disparities in access to health technologies will persist and may even exacerbate other disparities in health or healthcare.²⁰

Significance

Our findings have several policy and practice implications. First, while policies and incentive programs like the PI program can increase the availability and usage of technologies, they may also exacerbate disparities because better-resourced groups will be more likely to adopt and use them.^10,20,24 Federal programs should be carefully monitored for disparities in use and their impact on health. Future federal policy efforts and investments to increase access to and use of health technologies should focus directly (and maybe exclusively) on the most vulnerable patients and the providers who serve them, which would ensure not only overall increases in usage but may also decrease disparities. Public policies that seek to expand basic technology infrastructure like broadband access, are critical for ensuring that technologies like online portals and other digital health tools are accessible to all populations.⁴⁴

Second, our findings have implications for access to digital health resources during the ongoing public health emergency of the COVID-19 pandemic and other instances in which taxed health care systems seek to minimize in-person contact. Early in the pandemic, several policy changes facilitated the expanded use of telehealth as a necessary substitute when in-person care was unavailable or could not be delivered safely.⁴⁵ Initial reports indicated rapid uptake of telehealth,²¹ with some reports suggesting that the use of a broad array of telehealth delivery modes—including via a patient portal—potentially narrows pre-existing inequities in health care access.²⁷ Online portals serve as one means by which some telehealth-specific services like video visits are offered by providers, so having an online portal is sometimes a prerequisite for accessing telehealth visits. Our findings that less than half of insured adult patients accessed an online portal even during the initial period of the pandemic, and worse, that some minoritized and lower-resourced patients were even less likely to access a portal, are consistent with some research showing disparities in telehealth use during the early period of the pandemic.^21,22,28 Such findings have important implications for current debates regarding telehealth payment parity in Medicare and other policies like coverage of audio-only telehealth visits after the end of the public health emergency. To the extent that telehealth becomes an acceptable and high-quality substitute for some in-person visits, it has the potential to expand access to care but only if infrastructure, digital literacy, payment, and other barriers are addressed.

Limitations

There were several limitations to our study. First, it is not possible to infer causal mechanisms from the HINTS survey, as it is cross-sectional data. Second, the HINTS survey asks patients to report online portal use and offers, but patient recall may be inaccurate, particularly for offers of portals. Moreover, the survey does not evaluate whether or how providers attempt to offer online portals, including whether and what support resources may be available to assist patients. Future research should investigate more of the process by which patients are offered online portals. Third, while the response rate is reasonable and survey weights account for nonresponse bias, there may remain additional biases from certain subgroups being underrepresented, such as those who were critically ill. Finally, this analysis does not investigate how patients are using a portal or other aspects of portal usage.

CONCLUSION

As national policies aim to expand the reach of digital health services, it is important to closely monitor the adoption of health technologies, particularly by vulnerable and disadvantaged groups. Although the availability and use of online patient portals have increased steadily among patients in the United States over the past several years, evidence shows that inequalities in access persist. These inequalities persisted even after the start of the pandemic when the possible need or utility of accessing healthcare information and communication was great due to limited access to in-person care.

Furthermore, because only about half of all insured patients accessed an online portal in the prior 12 months through June 2020, including only approximately two-thirds of those offered one, it is important to address the technology and other barriers that hinder portal use. Technology barriers can be partly addressed through policies that expand broadband access and more directly support providers who serve vulnerable patients who may need greater resources for devices and infrastructure. Additionally, resources to facilitate both provider and patient literacy and comfort with technology are also necessary. Finally, ensuring that health data and information systems are secure, private, and trustworthy will alleviate some patient concerns about using digital health tools. The successful adoption and use of online portals, telehealth, and other digital health tools in the healthcare system will require explicitly addressing fundamental inequities in access if we are to ensure such tools are beneficial for everyone.

Funding

A collaborative award to the University of Michigan (Anthony) from the National Science Foundation (NSF) SaTC Frontiers program under award number 1955805. The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of NSF.

AUTHOR CONTRIBUTIONS

All authors contributed equally to the conception of the work; the analysis and interpretation of data; the drafting and revision of the manuscript and the final approval of the version to be published, and are accountable for all aspects of the work.

SUPPLEMENTARY MATERIAL

Supplementary material is available at JAMIA Open online.

CONFLICT OF INTEREST STATEMENT

None declared.