“Getting people access to services is also getting them access to a phone”: Clarifying digital divide dynamics and their consequences in Community Mental Health Care

Abstract

Access to mental healthcare is increasingly technologically-mediated. People with low socioeconomic status (SES) and serious mental illness (SMI) face lower rates of tech ownership and may lack technological skills, called “digital divides.” Yet, little is known about how digital divides may impact mental healthcare access. Therefore, a qualitative study (ethnographic observations and interviews) was conducted with stakeholders working with low-SES SMI patients using community mental health care (CMH) (N=14). Findings showed that consumers struggled to maintain consistent internet—and thus mental healthcare—access despite owning smartphones. Consumers frequently faced care disruptions due to broken, lost, or uncharged phones. Staff and patients created effortful but ad-hoc workarounds to restore access during technological access disruptions. These solutions frequently occurred after healthcare appointments were missed. Digital divide concepts should accommodate the work necessary to maintain technology access even after ownership and its impact on care access—especially among low-SES SMI patients.

Introduction

People with serious mental illness (SMI) are disproportionally represented among those with low socioeconomic status (SES), resulting in a need for additional support to treat their SMI.¹ SMI refers to a mental health condition resulting in serious functional impairment that limits one or more major life activities, such as schizophrenia or bipolar disorder. ² The Substance Abuse and Mental Health Services Administration found that one in four, or 2.5 million, adults with SMI in the US live below the poverty line.¹ People with SMI face challenges related to socioeconomic disparities, such as lower rates of full-time employment, higher rates of food insecurity, and higher rates of homelessness than compared to populations without SMI.^3-5 Those in low-SES neighborhoods are also more likely to have a mental health condition than those in high-SES neighborhoods.⁶ Furthermore, individuals with SMI who have lower socioeconomic statuses (SES) have the lowest mental health service utilization rates.⁷ People with SMI (henceforth, “consumers”) with low SES urgently need more care access to properly treat their mental health condition. Treating SMI can be difficult, but up to 65% of people with SMI can experience recovery over time with access to mental healthcare.⁸

Community mental healthcare (CMH) can effectively treat SMI, facilitating recovery and improved quality of life. Care access is broadly defined as consumers being able “to identify healthcare needs, to seek healthcare services, to reach, to obtain or use healthcare services, and to actually have the need for services fulfilled.”⁹ For those who have low SES, access is primarily enabled through CMH.⁹ CMH is an “inclusive, multidisciplinary, systemic approach to publicly funded mental health services provided for all in need, residing in a given geographical locale, without consideration of ability to pay.”¹⁰ While CMH care provides traditional medical treatments (e.g., therapy, medication), it expands the scope of care to include other supports necessary for low-SES people with SMI, such as employment training, case management, housing support, and coordinating food access. Access to consistent CMH care is associated with fewer deaths by suicide, greater patient acceptance of treatment, and a reduction in hospital admissions.¹¹

In the past 20 years, mental healthcare has become more technologically mediated,¹² often requiring consumers to have digital devices and internet access to easily obtain care. Consumers frequently use technologies like smartphones to “access help, monitor progress, and increase understanding” of mental health needs.¹³ Some of these technologically-mediated means of supporting mental health conditions have proven effective at treating SMI.¹⁴ With the rise in telehealth platforms over the years, many have found that using technologies to access care directly can lower some of the barriers associated with in-person care, such as eliminating the financial and time costs of travel. Furthermore, the COVID-19 pandemic accelerated a shift towards increasingly remote healthcare access.

Many have proposed that patient-facing technologies could bridge care gaps between low SES individuals and mental health treatment. For instance, Schueller et al. highlight that digital technologies for mental health could “provide opportunities to alleviate mental health disparities among marginalized populations by overcoming traditional barriers to care and putting quality mental health services in the palm of one’s hand.”¹⁵ Yet, technology access and utilization patterns continue to differ across socioeconomic strata. Disparities in technology access and skills are often described as “digital divides,”¹⁶ a concept which can sensitize us to some of the issues that arise for low SES people with SMI. Firstly, regarding device ownership, low SES people with SMI are less likely to own a smartphone than the general population (37% ownership in 2016).¹⁷ Secondly, more than one in six people experiencing poverty have no internet access. As young mental health leaders asserted, internet connectivity remains “an unaffordable luxury for families struggling to put food on the table.” ¹⁴ Lastly, issues such as lack of skills or knowledge can cause barriers to technology use for individuals with SMI, with a 2024 study by Spanakis et al. uncovering that over 42% of their sample of people with SMI lacked foundational skills for internet use, such as changing passwords, updating user settings, and interacting with device home screens.¹⁸ While other groups may also face unique challenges when it comes to technology access and use, the nature of living with limited economic resources, joined with the consequences of a major psychiatric condition such as SMI, results in potential care compliance concerns that remain understudied.

To address such digital divides, some healthcare organizations have provided low-SES patients access to devices and support. The Department of Veterans Affairs health system was able to increase use of mental health care via telehealth by providing video-enabled tablets to low-income Veterans who required them to participate in home-based telehealth.^{19, 20} Additionally, there are growing numbers of programs that support low SES community members in acquiring broadband internet. For instance, the nonprofit Link Health establishes health center-based internet clinics in communities throughout the Greater Boston area to assist people in acquiring at-home internet. They support consumers in navigating the eligibility and sign-up processes associated with the Affordable Connectivity Program, a federal benefit that provides wireless internet to low-income households.²¹

Other organizations have expanded efforts to address skills-based digital divides. Some healthcare organizations have incorporated digital navigators into their care models to support technology use for care access. Digital navigators are trained clinical technology specialists who help consumers with their technology needs.²² While these recent programs offer promising ways to confront some of these access issues consumers face, many are only available in certain locales and are not widely accessible.

Despite these efforts, novel programs to facilitate access to technology-mediated care could benefit from better characterization of the relationship between technology access and access to mental health care. Most work in this space investigates technology-specific uses in controlled settings rather than the in-situ, day-to-day use of technology by consumers to access care.²³ Therefore, this research investigates how technologies impact mental healthcare access for low SES people with SMI.

Theoretical Framework: Digital Divides

This research uses van Dijk’s multiple access model of digital divides to guide analyses. In this model, the digital divide is defined as inequities in (1) access to technologies (including access to the internet), (2) technological skills and experience that impact technology use, and (3) outcomes of technology use.²⁴ In the first level of the digital divide, there are gaps in digital access, underscoring access needs concerning device ownership and internet access. Van Dijk underscores that physical access also involves the motivations, attitudes, and expectations surrounding acquiring device access.²⁴ The second level digital divide encompasses the skills needed to “command and use digital media.”²⁴ These skills include technical competencies (such as using software) and information navigation skills (such as finding information on the internet).²⁵ As Schreeder et al. describe, the second-level digital divide also describes differences in use, otherwise known as the usage gap.²⁵ Lastly, the third level digital divide describes the divide present when access to technologies and technological skills do not lead to equal benefits of internet use. The third-level digital divide is a “multidimensional phenomenon” caused by various factors. In medical informatics, factors contributing to the third-level digital divide in health care include the social determinants of health (built environment, social and community context, economic stability, and healthcare access).²⁶

In order to understand the role of the digital divide specifically for low SES people with SMI, this research seeks to address the following research questions (RQs):

• RQ1: Does access to technology among low SES people with SMI influence their mental healthcare access? If so, how?

• RQ2: Do knowledge of and skills related to technology among low SES people with SMI influence their mental healthcare access? If so, how?

• RQ3: Does the ability of low SES people with SMI to benefit from the use of technologies impact mental healthcare access? If so, how?

Methods

This project employed two qualitative methods: ethnographic observations at mental health treatment court (MHTC) systems and interviews with MHTC stakeholders in one Midwestern state. Stakeholders consist of employees and volunteers from CMH, nonprofit organizations, and criminal justice sectors (such as case managers, court staff, shelter staff, and mental health nonprofit volunteers). Mental health treatment courts were chosen as it is a site for enabling mental healthcare access for consumers with SMI, primarily those of low SES. The University of Michigan Institutional Review Board reviewed the study and declared it exempt from ongoing IRB review based on an exemption for minimal-risk research with adults.

Research setting. Data was collected within a Midwestern state’s MHTC system. Mental health treatment courts serve consumers with pervasive SMI and criminal justice involvement, helping to connect consumers to mental health care and preventing incarceration. Notably, MHTC-enrolled consumers are primarily covered under public insurance and experience high rates of unemployment. Given that MHTC-enrolled consumers typically face multiple prior challenges in accessing and staying in mental health care, the study of this setting also offers a valuable opportunity for identifying how technology access may help or hinder access to mental health care among low SES consumers. The consumers primarily receive mental health care through local CMH agencies and receive additional support from local nonprofit agencies.

Court Review Sessions (CRS) are MHTC sessions that happen multiple times a month for people with SMI. The MHTC includes multiple “stakeholders” from CMH, nonprofit organizations, and criminal justice sectors. Most stakeholders interact with consumers between CRSs remotely (via email, phone calls, and text messages). Additionally, stakeholders contribute to regular in-person court team meetings to monitor and assess participants’ progress in engaging with mental healthcare and related social services. During the early COVID-19 pandemic, CRSs were conducted remotely via videoconference, largely returning to in-person meetings in 2022.

Data Collection: Ethnographic Observations. Ethnographic observations were conducted at bimonthly CRSs from July 2022 to September 2023 (15 consecutive months) over the transition from video-conference-based to in-person sessions. Observations averaged 1 hour and 40 minutes in length, with a total of over 38 hours of observation. At CRS sessions, jottings were recorded jottings and later summarized in field notes after the sessions.²⁷ Specific attention was paid to the technology that they use within the courtrooms (tablets, laptops, cellphones), as well as note any mention of technologies (email, videoconferencing, telehealth, remote support groups, phone calls) that factor into each participant’s mental health care.

Data Collection: Interviews. Semi-structured interviews were conducted with MHTC stakeholders (N=14) using an interview guide between January 2022 and November 2023. Stakeholders were asked to complete a demographic survey after the interview. Interviews averaged 56 minutes, and stakeholders were offered a $25 incentive for participation. Using purposive sampling, stakeholders were recruited to represent different care roles in CMH service provision. Interviews were transcribed verbatim or summarized (for one participant). Data collection continued until after the data were saturated.²⁸

Data analysis. Qualitative data analysis began early in the interview process to make any early adjustments to the procedures. Two researchers (A.K.W., a doctoral candidate in information with qualitative methods training, and E.J.L., a master’s student in information with qualitative research experience) independently read three randomly selected interviews and 5 randomly selected field notes to develop inductive codes. A.K.W. developed deductive codes based on digital divide literature and discussed those codes with E.J.L. While coding, analytic memoing was used to synthesize findings, and after coding, memos of individual codes were developed. The interview transcripts and field notes were coded inductively using the process coding approach and A.K.W. wrote analytical memos during analyses.²⁸

Results

Stakeholders characteristics. Stakeholders interviewed (N=14) were majority white (71%) with ages ranging from 30 to 80 years old. All had a secondary education, with the majority (35.7%) having a graduate education, 21.5% having a bachelor’s degree, 21.5% having a professional degree. All stakeholders were current or former employees or volunteers who worked with consumers in this system. Respondents represented those holding roles at non-profit agencies (n=3), MHTCs (n=6), community mental health agencies (n=4), and shelter agencies (n=1). Of note, two stakeholders did not provide demographic information.

RQ1: Does access to technology among low SES individuals with SMI influence their mental healthcare access? If so, how?

Device access: initial acquisition. Consumers with consistent access to smartphones had an easier time accessing mental health care. Consumers regularly used cell phones to coordinate their own care (e.g., online scheduling, messaging providers) or directly access care (e.g., telehealth services). There were supports in place to help consumers access smartphones through public assistance programs. Multiple stakeholders mentioned programs like Assurance Wireless as a mechanism for providing participants access to technology. Some consumers also informally leaned on their family members or friends to help them apply for these programs. Other consumers received support from their CMH caseworker to acquire their first public smartphone. Caseworkers would help consumers directly apply for these phones, or they would provide them with the informational resources to do so on their own. As one stakeholder mentioned, “[CMH] case managers will also [facilitate] applying for people to have smartphones.” [P008] They went on to reiterate, “getting people access to services is also getting them access to a phone.” [P008]

Device access: inconsistent access due to challenges maintaining technology. Despite the existing mechanisms to acquire a smartphone, consumers had limited options if they broke, damaged, or lost those phones, which can negatively impact care access. The public access phones were said to be of relatively low quality, thus leading them to break often. As one stakeholder said, “There’s lots of reasons that our folks’ stuff is more inclined to break. It’s not as probably as good quality as the stuff that I’m purchasing.” [P014] Additionally, publicly provided phones often lacked peripheral items that helped to protect their devices from damage, such as screen protectors or a case. Once broken, consumers could rarely afford repair. A field note highlighted a consumer’s lack of funds to fix a broken phone: “Their phone died recently. They need to patch it and they’ve been borrowing money and they need to borrow some more money to get their phone fixed…” [Field note]

Additionally, the plans were expensive and difficult to maintain for consumers who had to secure their smartphones privately. As a stakeholder mentioned, “Their phone got shut off because they weren’t able to pay for it. It’s challenging to afford consistent access to technology.” [P010] Lastly, if a consumer needed a replacement phone, “it’s kind of the process to get a new one,” [P002] and that it could take weeks to arrive from applying to receiving the device. In such circumstances, one-off access to devices could enable access to care, but these consumers struggled to maintain their technologies overall.

Device access: borrowed devices. Insufficient device access has led consumers and their loved ones to borrow devices in order to access mental health care. This often happened due to the low quality of phones and struggles with keeping track of them. One stakeholder explained, “they gave us a number that might have been a phone that their friend gave them or their parents gave them for a short term.” [P002] Borrowing devices provided either onetime or short-term care access, but was not always a reliable means to access care. A stakeholder recounts a participant describing their technological struggles with a borrowed phone, stating, “I was going to borrow my friends, phone but his phone wasn’t charged.” [P003] While potential privacy issues may be inherent to the shared use of phones in this way, our participants did not note any particular privacy concerns emerging from this type of device sharing.

Internet access. Many consumers lacked sufficient internet access, which could directly impede them from accessing care. Accessing video-enabled telehealth via a smartphone required either data or an internet connection. One major issue consumers faced with public phones was running through available minutes quickly or using up the available data before the end of the month, as these phones “come…with [only] so much data.” [P012] Many consumers either completely lacked access to, or had inconsistent access to, broadband or Wi-Fi. As one stakeholder mentioned, “the pandemic highlighted the fact that individuals, who have low SES often do not have the same level of access to internet.” [P009] Stakeholders also observed that service interruptions often occurred repeatedly to the same people, and stemmed from reasons outside of the consumer’s control. One stakeholder mentioned, “She found the connectivity issues that often pop up happen with the same participants. Usually, it stems from unstable working or living environments.” [P010] Without the funds to acquire a high-quality internet connection, many consumers cannot depend on current data plans or at-home internet to access care.

Impact on care access: stakeholder device workarounds to enable digital connection. Stakeholders mentioned that CMH workers often accommodate the aforementioned digital divides by directly connecting consumers to devices in order to access care. They frequently facilitated ad-hoc and temporary provision of technology or set up shared technology for individual use. One stakeholder mentioned that tablets are available at CMH locations for consumers to use. Another stakeholder discussed how consumers who lived in group homes may have care workers get them directly connected to care through devices that they can borrow. “So this zoom account is like an account, the house made up just to have available so people can access things.” [P009] Workers often provided these solutions in the wake of previous disruptions, enabling consumers to directly access care via the borrowed device.

Impact on care access: requiring additional stakeholder time and labor to compensate for tech inadequacies. The workarounds that consumers use to maintain digital connections (e.g., borrowing phones) often led to challenges for the stakeholders providing care. One stakeholder described how inconsistent consumer phone numbers could be, “It’s not, it’s something that we can always consist of, rely on to have a number, and to assume that it’s gonna be a working order and it’s gonna go to that person.” [P002]. Consumers churning through phones or phone numbers led to CMH relying on out-of-date numbers and constantly updating their systems to reflect the most up-to-date consumer information. Furthermore, consumers’ inability to communicate via technology can result in more work for CMH workers. For example, one stakeholder described the process of reaching out to jails, hospitals, and law enforcement, stating, “there’s a whole kind of decision tree that gets implemented when we can’t find the person from initial phone call to an outreach to then kind of networking with the local agencies they might have interacted with.” [P002] There were multiple ways of reaching consumers, and when consumers could not be reached, they then reached out to other local agencies.

If CMH workers cannot locate a consumer, they may try to physically locate them via an in-person outreach. One stakeholder described the required labor, “if the person has a phone, we certainly try to keep in contact by phone. But if not, then the case manager very much might kind of just do home visits. You know, we do a lot of outreach, so it might be a matter of trying to do an outreach and it could be to an apartment if they have one, if the clients do have a lack of housing, if they’re homeless, if they’re one of the homeless encampments, the case manager very well might go out to the campsite and either hand them a card, or sometimes they, they allow[them to use their work] cell phones.” [P002]

When they are unable to locate the consumers via digital and in-person outreach, CMH stakeholders have to try to locate them via other means. For example, “we might have talked to them on the phone and then we can’t find them. So what we do for those folks is try to like weekly call the local hospitals, the jail, see if their name’s there, check their electronic health record to see if they’ve hit any hospitals.” [P009] There were processes that these stakeholders went through to accommodate consumers who they were unable to reach, but these processes were often labor-intensive. If the original digital outreach was successful, the stakeholders would not need to spend time and resources trying to locate consumers through these other means.

Care impact: public internet and lack of privacy. When consumers did not have access to the internet, their solutions were often ad hoc fixes that facilitated single use rather than ensuring consistent access to technologies. Ad hoc fixes included going to CMH in-person or visiting public libraries to access technology in order to engage with care. When asked how they could conduct a telehealth call without the internet, one stakeholder mentioned, “they can go to the library and use the public [internet] and I think the shelter might have some.” [P011] Additionally, one-time access can be obtained through publicly available Wi-Fi or food establishments. When asked about other locations where participants could access Wi-Fi, one stakeholder stated “There’s probably coffee shops or something like that that may have [internet].” [P004] Thus, the solutions available to either accommodate poor Wi-Fi or provide Wi-Fi for those who do not have it often led consumers to use publically available Wi-Fi.

While these locations often did provide internet, they were often not appropriate places to have a private call with a provider due to privacy concerns. As one stakeholder described this tension saying, “there’s often probably a lack of privacy for phone calls, any conversation that you’re having, you’re having out in the open.” [P014] While stakeholder did not discuss the implications of using semi-public places for accessing care via technology, one can assume that it is not the ideal location for sharing sensitive information about SMI.

Digital divide negatively impacts care compliance. The breakdowns in both device access and internet access frequently had negative impacts on care access. Often, there would be actions to try and repair the technological breakdowns by stakeholders or consumers, either in the moment or after the fact. However, the timing of the breakdowns usually negatively impacted the consumer’s ability to access care, either resulting in a missed appointment or time spent troubleshooting technical issues instead of engaging in care. “The participant mentioned that he wasn’t able to join his therapies because his phone was messed up – he mentioned that his screen was black.” [Field note] Stakeholders saw how lacking technologies threatened patient access to care, “There have been participants who are supposed to show up to see the judge at a certain date and time on zoom, and they didn’t have a laptop so they couldn’t navigate it.”[P003]

RQ2: Do knowledge of and skills related to technology among low SES individuals with S1VII influence their mental healthcare access? If so, how?

Technology skills. When consumers began using a novel technology, they often lacked the skills to fully use new systems or software. One stakeholder described this by saying, “Sometimes if they’ve never had an iPhone or an Android before, they don’t know how to work it, it’s like a foreign object to them” [P008]. Many of these consumers used Zoom for the first time to connect with care and the court, and it was obvious to stakeholders that many consumers lacked the skills to use them. “I think it would take some training, and some help, and some other things that are necessary in order to get the value and what’s needed out of a Zoom.” [P003]

Stakeholders also observed that sometimes providing the technologies was not enough – often, the consumers required direct support to be able to use the technologies that were provided for them through CMH. When asked about consumers’ technological skills, one stakeholder mentioned, “The younger ones are usually more with it. the older ones. Not necessarily, but it seems like they’ve been able to manage it.” [P009] This same stakeholder mentioned how, due to their lack of skills, “They may require some support. So it may be them coming up here and us getting them linked on the iPad or something.” [P009]

However, there were limits to even these one-time supports. For example, stakeholders found that facilitating access was much easier if they were collocated with the consumer. “I don’t know how to say ‘Okay, go to safari, go to, go to this,’ and it’s a lot easier if I just show them how to do it.” [P008] Remotely coaching consumers in connectivity issues was often unsuccessful. Furthermore, some consumers may not have the know-how to use technology enough to learn remotely, and require collocated human supports to set them up with a technology. “Some people can’t sign in on their own. They need supportive case workers or family members cause they don’t know how to use Zoom.” [P011]

Care impact: missed appointments and rescheduling. Insufficient skills in regards to technology at times resulted in an inability to access care. Stakeholders often became aware that consumers lacked the proper skills to access care when it already had a negative impact on care. One stakeholder discussed a missed appointment with a consumer, stating, “[CMH WORKER] would say, ‘you missed your therapy appointment.’ [Consumer’s responded] ‘Well, you know I tried getting on I couldn’t get on.’” [P005] Another field note shows how technologies led to a consumer missing an appointment, “When the [COURT WORKER] asked what happened, they said that they couldn’t get into the portal, they tried, but just couldn’t figure it out, and then eventually had to get things rescheduled.” [Field note] For many of these consumers, that breakdown in their ability to use their technology led to missed appointments or appointments that would be rescheduled when it became apparent that a telehealth video visit was impossible.

Care impact: negative emotions and recovery impediments. Aside from potentially disrupting their care compliance, insufficient technology skills could put more stress on consumers who were already dealing with pervasive SMI, potentially overwhelming them. One stakeholder described having “run into [CONSUMERS] being overwhelmed by it and like, ‘I don’t know what to do with all this.’” [P012] These feelings of defeat may lead consumers to give up on trying to access technologically mediated care altogether. Additionally, some consumers were uncomfortable or uncertain about using these technologies in the first place. Stakeholders reported that many consumers were suspicious of public phones; for example, “individuals who don’t really know how to use them may have some resistance or paranoia associated with them.” [P009]

Overall, the lack of technology skills disadvantaged the consumers, who were often trying their best to keep track of their recovery, leading them to additional stress in accessing care. Those without skills may have fared worse in care “not through any fault of their own.” [P011] This stakeholder expanded on this sentiment, stating, “since COVID, anybody who was more technologically sophisticated had an easier time participating [MHTC]. and the reverse is true. If they don’t have skills, that’s one more thing on their plate that’s a burden, that makes their life harder. Which means some of their energy has to be diverted from thinking about their mental health issue, thinking about their last therapy session and how they can implement the changes.” [P011]

RQ3: Does the ability of low SES people with S1VII to benefit from the use of technologies impact mental healthcare access? If so, how?

Third level digital divide: burden of inequities falling on the most under-resourced The ultimate burden of these digital divides fell on the most marginalized consumers. One stakeholder explained, “if they’re in and out of incarceration and have been in and out of like the jail or prison, their technology skills are very limited because technology advances so fast and they they’ve been in a literal cell where they don’t have access to that and to be able to grow and change what the technology advances.” [P008] Additionally, homeless consumers had additional burdens that took higher priority over maintaining technology. One stakeholder says, “It can be challenging to stay organized in terms of your person and things. It can be easy to lose stuff. You have a lot of things to maintain on you at once. And if we all had to carry our homes with us all the time, we’d probably feel the same way.” [P014] Thus, competing priorities associated with managing an SMI and navigating the stressors that coincide with low SES often pulled people away from their healthcare needs. In addition to technological barriers, those with the least access and skills had the most trouble accessing care.

Discussion

Digital access and digital skills are crucial in enabling healthcare access for low SES consumers. In the face of barriers to technology access and skills, many face care disruptions even though they have engineered workarounds that enable temporary access. Many lack sufficient internet access, leading stakeholders to spend more time sorting out technological issues, which are often only solved once there is a disruption in care. Consumers regularly borrow devices from friends and family or use resources provided publicly or through CMH. Spending time assessing and solving these digital divide issues may pull stakeholders from care provision. The individual work that stakeholders and consumers did often results in one-time solutions that are not always efficient or effective. Sometimes, the workarounds were also insufficient for enabling adequate privacy, which is of utmost importance for consumers with SMI. Missed appointments due to technological issues were observed. Additionally, due to the novelty of devices and inconsistent access to the same devices, they often did not have adequate technological skills to access care. The lack of skills frequently results in missed or rescheduling appointments due to technological disruptions. Findings indicate that the burden of the observed digital divides likely has the most impact on those who are most under-resourced, such as those returning from incarceration or homeless consumers.

Study findings highlight that initial smartphone access was enabled through current structures for many, but broken or lost technologies became a significant barrier to accessing care. For many consumers who could not consistently access functioning technology, there were direct negative consequences related to care, such as missed appointments. The consumers observed in this study often faced issues with broken screens, lost or malfunctioning devices, and inconsistent internet access, with little understanding of how to troubleshoot or fix these issues. Although the lack of access to technologies is not unique to populations with SMI, their access is impeded by a conglomeration of factors necessary to enable access. As Brewer et al. highlight, the “digital disadvantage for the disadvantaged” remains an ongoing issue in health informatics, and despite an increasing proliferation of digital health technologies, populations such as low-SES patients with SMI remain marginalized.²⁹

While there is a considerable amount of work within health informatics that identifies the value of technological maintenance within healthcare systems, little work has been done to understand how the devices that consumers use require direct maintenance. The Public Interest Research Group is one of many groups advocating for the “right to repair” to allow consumers to easily modify or repair technologies to decrease costs and increase access.³⁰ Furthermore, there are calls to manufacturers to increase access to tools and diagnostics that would make it easier to repair or refurbish devices.³⁰ Allowing for increased maintenance and reestablishing access may allow for improved autonomy for the patient and remove one more access challenge. By failing to understand patients’ maintenance needs, the field risks assuming that technology ownership is the main component of access to technology-mediated care when maintenance of said technologies remains incredibly important.

Beyond repairs, maintaining charged, functional smartphones was a challenge for consumers that often resulted in stress and care disruptions. Prior work indicates that the use of mobile technologies has a beneficial impact on chronic illness management,³¹ however, little work has been done to understand how consumers manage their devices. Additionally, while built environments can contribute to digital divides in healthcare, findings suggest greater attention to this dimension of access to technology-mediated health care. Indeed, many consumers lacked environments that were conducive to even charging their devices. Providing devices such as solar chargers may be one solution,³² but there is also a need for more spaces where consumers can charge their devices without purchasing a meal. Expanding services to include charging stations with available internet may also work to meet the needs of these consumers. As the nonprofit Shelterforce stated, the challenges faced by maintaining a working cellphone pose “a threat to their safety and ability to follow up with service providers.”³³ This draws attention to an issue regarding access from early research on eHealth: the quality of technology access for health-related purposes.³⁴

Stakeholders and providers in this study often had to accommodate technological issues, which aligns with the concept of “infrastructuring” from science and technology studies and human-computer interaction fields.^{35, 36} Infrastructuring is the work people do in the face of inefficient or broken infrastructures to meet their needs. One aspect of infrastructuring work is “repairing failures that happen at the individual level, a shared experience by consumers and stakeholders. Stakeholders at times engaged in intermediation work, acting “in a middle space between people and technologies,”³⁷ helping consumers troubleshoot their technologies and cope with the use of subpar technologies. This project also highlights that digital divides lead to individual infrastructuring work for low-SES consumers, creating one-time solutions for systemic issues that ultimately require infrastructural changes. The lack of digital connectivity in this population threatened care compliance, which reaffirms prior work underscoring that digital inclusion continues to be a social determinant of health.³⁸ While this problem of digital exclusion has persisted for decades, the evolving nature of technology requires an evolution in strategies to address these disparities.³⁸

There has been positive progress regarding the public provision of technology through programs like Assurance Wireless. Still, many of the participants in our study often quickly hit their data limits and required increased data on those plans. Others have researched to understand insufficiencies around mental health care within rural populations and in the global south but these projects do not fully explore technological access, management, and skill development and its impact on consumers.¹² Accordingly, this paper contributes empirical results that can help to better characterize infrastructuring work, and thus current infrastructural “exclusions.”³⁹

One potential way to enhance infrastructures to meet consumers’ needs is by building upon and expanding resources at libraries and other public spaces. One example is ensuring that patrons have the ability to reserve private soundproof spaces for one-time telehealth calls or embedding social workers in libraries to “ unmet social service needs where patrons are.” It remains clear that improvements to infrastructure are necessary, and as Gui and Chen advocated, “persuading healthcare consumers to do more burdensome work is not going to improve the fragmented healthcare infrastructure.”³⁵ This work often emerges when there are breakdowns in the infrastructure, which may be increasingly applicable in resource-constrained systems (such as CMH) where those with less power engaging in the system (low SES consumers with SMI) have to accommodate the system’s failings. Together, these infrastructural issues create friction that prevents seamless access to CMH care and reinforces existing disparities for low SES consumers with SMI. As mentioned in the introduction, expanding broadband access and providing devices to consumers is a potential way to mitigate some of the negative impacts of these digital divides and potentially skirt some of them together. Among them, funding digital navigators trained to address digital inclusion with consumers could help address some of the first and second-level divide issues.

Overall, study findings showed that maintaining technology access requires ongoing work. Thus, access is more than just technology ownership and having a broadband internet connection. Instead, it is a process by which access can be alternately had, lost, restored, and disrupted again in multiple cycles. Accordingly, there is a need to develop current concepts of digital divides further, exemplified by van Dijk’s Multiple Access Model,¹⁶ away from static and binary understandings of technology access towards a more dynamic conceptualization. Furthermore, the problem of mental healthcare access using technology highlights several further issues that need to be incorporated into extant theory: quality of access for mental healthcare purposes (e.g., privacy), and concepts of technology access fixes and their respective ad-hoc nature or sustainability.

There are a few study limitations. The interviews and observations for this study were conducted in one state in the US, which may impact the applicability of these findings in other states with newer or less-resourced MHTC systems. However, a large variety of stakeholder roles were involved. The observations were conducted during an unusual period when CRS sessions were remote, hybrid, and in-person, likely contributing to the diverse ways consumers used technology to access care. Furthermore, the stakeholders may have been more attuned to participants’ struggles, and those who did not face access issues may be underrepresented in the findings. Moreover, this research focused on observations and perspectives of stakeholders who served as key informants about consumers with whom they work. Future research will investigate the perspectives of consumers themselves.

Conclusion

Ultimately, access to devices and the internet is crucial to enable care access for low SES people with SMI. Not rectifying the issues posed by the aforementioned digital divide will likely result in increased labor by healthcare consumers and stakeholders. By reducing the time these people spend on enabling technologically-mediated care access, stakeholders can focus more on care provision, and consumers can focus more on their mental health recovery. These issues span beyond the individual and require improved infrastructure to enable improvements at the societal level and take the pressure off those navigating this problematic system. To reiterate the words of our participant, access to technology is access to care, and the consequences of not meeting the technological needs of low SES people with SMI will lead to deepening digital divides. The relationship between SES and SMI is bidirectional, with “with poverty often leading to mental illness and mental illness regularly reinforcing poverty.”⁴⁰ Digital health equity persists as a major informatics issue, and by failing to improve these technological conditions, we run the risk of further disenfranchising a group of people who face numerous barriers to care.