Highlights

  • Evidence points to benefits associated with PHR adoption and uses in chronic conditions

  • Barriers to PHR adoption/use, with a special focus in chronic care, has not been well described and understood

  • Addressing barriers for PHR adoption/use in chronic care should cross the boundary of patient-level barriers

  • Barriers at the provider and healthcare organization levels should be understood and addressed, thoroughly

  • PHRs should fit in the structure of “chronic care models” developed for improving chronic care

Background

Promoting self-care and patient engagement in care management has gradually become key features in efforts to improve health service delivery and care quality in chronic diseases [1, 2]. Electronic personal health records (ePHRs) provide the tools to empower patients and promote self-care [3, 4]. A systematic review found that self-monitoring through ePHR improves health outcomes in chronic conditions [5]. Because of such potentials to enhance quality and patient engagement [6,7,8], the Health Information Technology for Economic and Clinical Health Act (HITECH) and meaningful use phase 2 and 3 have driven the adoption of ePHRs in parallel to Electronic Health Records (EHRs) [9].

Studies have shown that both patients and providers are interested in ePHRs especially as they find them as a means to increase patient empowerment [10,11,12]. Yet, there are barriers to overcome and challenges to embrace when adopting ePHRs. Some of these barriers are related to the implementation of EHRs such as EHR products and capital and human resource issues. For example, from 2,674 general hospitals studied in the United States (US) in 2013, only 5.8 percent of hospitals met measures for stage 2 meaningful-use readiness and several other criteria, including sharing care summaries with other providers and providing patients with online access to their data, as necessary functions for a tethered PHR [13]. Other barriers are more ePHR specific ones such as poorly aligned functionalities with patients’ expectations and self-management practices and concerns about privacy and confidentiality of patient information in ePHRs [14, 15]. Even outside the US healthcare context, similar hurdles have also contributed to a lower adoption rate than what has been expected or hoped for [16]. Such results continue to be reported after the implementation of many health information technologies (HIT) including ePHRs, which highlight a strong need to understand factors and challenges that influence the implementation outcomes [17]. Overcoming these challenges and barriers in implementing and adopting ePHRs can result in increased efficiency and improved quality patient care [18]. Therefore, recognizing and understanding the nature of such barriers is imperative to be well equipped to devise strategies to overcome the barriers and to achieve ePHR’s meaningful use.

There have been a few reviews published on the barriers to ePHR adoption and use. A review of the patient-level barriers categorized them into individual, demographic, capability, health-related, ePHR-related, or attitudinal factors [19]. Another review with similar scope concluded that a lack of awareness of and sufficient training regarding portal use were the two main barriers [18]. In the elderly population, the main barriers were limited technology access and no prior knowledge of the existence of a patient portal, and limited health literacy and motivation to use a patient portal [20]. In rural areas of the US, provider resistance, privacy concerns, and the lack of EHRs, interoperability standards, and funding have emerged as the main barriers [21]. However, these reviews have narrowly focused on patient-level barriers [18, 19], or were limited in terms of age ranges [20], time frame, or geographical location reviewed [21, 22]. To our knowledge, there is a significant gap in the literature on the barriers in the patient, caregiver, and provider levels that may impact ePHR adoption and use in the context of chronic care. To address this gap, we aimed to identify and synthesize evidence on ePHR adoption and use barriers in chronic disease care. More specifically, we were interested to identify the types of barriers and to analyze their extent in this care. The insights gained will inform efforts for effective design, implementation, and use of ePHRs for a patient population at the most need of these tools.

Methods

This review was conducted according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) [23].

Search strategy

We conducted a literature search in OVID versions of MEDLINE, PubMed, Science Direct, Cumulative Index to Nursing and Allied Health Literature (CINAHL), the Cochrane Central Register of Controlled Trials and the IEEE database for English-language, journal or congress proceedings’ full texts published January 1, 2005, till December 31, 2018. We used a Boolean search strategy using keywords and MeSH terms related to two areas of interest i.e., the intervention (e.g., Personal Health Record OR Personal Medical Record OR patient portal OR patient internet portal, etc.) AND the health condition (e.g., Chronic Disease OR Chronic Illness, OR Chronic Condition, etc.). The details of our search strategy are accessible in Additional file 1. We also conducted a manual review of all reference lists of included studies and the pertinent ePHR reviews including [14, 15, 18,19,20,21,22, 24,25,26,27,28,29].

Inclusion and exclusion criteria

We included studies according to the following inclusion criteria: 1) the intervention was an ePHR/patient portal, 2) the targeted users were chronic disease patients, their caregivers and/or their healthcare professionals, 3) the study was an original research article, and 4) the study design was either quantitative, qualitative, or mixed methods.

We excluded ePHR/patient portals that were not aimed at chronic patients, paper-based ePHRs or educational websites, assistive living technologies, or mHealth tools, systematic reviews, proceedings abstracts, commentaries, editorials, and articles describing theoretical background or design reports without having an evaluation nature. The main reasons for exclusions in each phase of this review are accessible in Additional file 2.

Review procedures and data extraction

After removing duplicates, our search identified 3088 unique records, which were screened for eligibility. Figure 1 shows the PRISMA flow diagram of our review. Two reviewers (ET and MCH) were trained on the screening and data extraction tool by ZN, who is an experienced researcher in conducting systematic reviews in the field. The reviewers reviewed a sample of references and compared extraction results to reach an excellent agreement (kappa= 0.77). Then, they screened titles and abstracts of the above-mentioned search result to find relevant studies based on our inclusion/exclusion criteria. In this phase, 143 potentially eligible publications were selected for the full-text review. Further articles were found through the manual review. All articles were independently reviewed in detail by ZN and either ET or MCH. Disagreements were solved by consensus. Endnote version XI was used to manage records.

Fig. 1
figure 1

Flow diagram of study selection (literature search January 1, 2005 till December 31, 2018)

Full size image

We extracted the following main study characteristics in the full review phase: general information (e.g., the authors and publication year), study objectives, study design, patient population, system users, the intervention (e.g., the description of ePHRs and their integration with other systems), and the main study results.

The methodological quality of studies

We used the mixed methods appraisal tool (MMAT) version 2018 to assess the quality of evidence in included studies [30]. This tool can be used to appraise the quality of empirical studies (i.e., primary research based on experiment, observation, or simulation) in three categories of study designs (i.e., qualitative, quantitative, and mixed methods).

Data synthesis

According to a widely used definition, an ePHR is “an electronic application through which individuals can access, manage and share their health information, and that of others for whom they are authorized, in a private, secure, and confidential environment” [3]. We used two well-known models as a theoretical background to analyze and categorize barriers to ePHR adoption/use faced by users. The first was the “Personal Health Records Adoption Model” (PHRAM), developed through integrating several relevant parent models/theories [31]. This model was used to analyze the barriers faced by patients and caregivers in the context of chronic care. We also used the unified theory of acceptance and use of technology (UTAUT) to analyze barriers specifically experienced by care providers [32]. Since conducting a meta-analysis became out of the scope of this study due to the lack of unified quantitative data in included studies, we only provide a narrative description of results based on the PHRAM and UTAUT.

Results

Characteristics of included studies

Our review identified 60 ePHR studies [5, 10, 12, 33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89], among which 24 were qualitative, 22 quantitative, and 14 mixed methods studies. Additional file 4 provides the details of the included studies. These studies were conducted between 2006 and 2018, nine of them in the single year of 2015. Forty-six studies were from the US, followed by five studies in Canada, two studies in the Netherlands, two in the United Kingdom, and the remaining five in Denmark, Sweden, Germany, New Zealand, and Argentina (one study from each country). A majority of studies included older patient populations (compared with younger patients) and diabetics (compared to other chronic patients) in their evaluations. Ten studies had a target population of pediatrics [37,38,39, 43, 44, 59, 71, 72, 78, 85]. The results are provided here according to the personal, environmental/medical practice, technological, and chronic disease factors on the bases of the PHRAM and UTAUT.

Personal factors

In one study involving survivors of pediatric cancers, referring to the issue of age, cancer survivors >18 years old were significantly more likely to use an ePHR compared to those <18 [85]. While a high proportion of patients with age 50 and older had higher frequency and intensity of ePHR use [66], patients over the age of 65 were less likely to intend to use an ePHR [10], and patients aged over 70 were associated with a lack of use [70]. In four studies, more men than women had access to computers or the internet, expressed confidence in using ePHRs, or used it [54, 67, 74, 84], while females were the dominant users in three studies [48, 66, 89].

People with minority race/ethnicity (including African-American, Latino, and Filipino) reported more negative attitudes towards ePHRs, were less likely to use ePHRs and experienced more obstacles compared to Whites and Caucasians [34, 47, 48, 55, 57, 61, 65, 66, 69, 70, 74, 75, 85]. Having a paid job or higher income and living in a higher socioeconomic neighborhood, and being insured were associated with higher use (reported in studies from the US and the Netherlands) [42, 61, 66, 67, 75, 84]; while, having a lower income and being below the poverty level were linked to non-use [34, 48].

Patients with lower educational attainments were less likely to intend to use or use an ePHR [10, 34, 42, 48, 61, 65, 67, 69, 70, 74, 75]. Patients with limited health literacy were less likely to use ePHR or use it ineffectively [5, 34, 41, 46, 54, 65, 69, 74, 76, 77, 87]. The level of knowledge, self-efficacy, and confidence in technology use was associated with ePHR adoption/use [52, 54, 56, 67, 83].

Negative attitudes toward the disease and health care experiences in general, and ePHRs in particular, prevented patients from using ePHRs [10, 40, 44, 47, 59, 71]. Patients were concerned about the reliability of ePHRs to facilitate timely and productive communication with providers [37, 43, 65, 81]. In one study, patients commonly expressed negative attitudes partly because of their experience of confusion and misunderstanding [40].

Fourteen studies highlighted the critical role of computer/technology literacy and skills to effectively use ePHRs [5, 41, 42, 45, 49, 52, 56, 65, 68, 76, 82, 83, 86, 87]. Computer literacy barriers included, but were not limited to, the lack of basic computer skills, inexperience in using search bars or uniform resource locators, difficulty while navigating the portal, and negative experiences with online security breaches/viruses. Three studies noted that computer anxiety negatively affected patients’ behavioral intention to adopt ePHRs [5, 34, 86].

Challenges related to communication preferences were brought up in several studies with a majority pointing out the value of in-person, or telephone contacts between patients and providers [37,38,39,40, 44, 47,48,49, 52, 58, 59, 64, 65, 76, 78,79,80, 82, 86]. The main reasons for such a preference were getting anxious when seeing results online and concerns over technology replacing their providers. The preference for in-person communications was also shared by providers in certain circumstances [64, 80].

Environmental/medical practice factors

Social influence

The impact of the social influence of “important others” (i.e., family members and care providers) on patients was evident [12, 54, 55]. It was shown that living alone and being not currently married were associated with non-adoption and lesser ePHR use [34, 42, 67]. Moreover, studies pointed out the role of providers’ willingness to use portals, their communication about it, and their level of use in patients’ initial portal use [47, 59, 81, 84]. While patients wanted their physicians to get more involved in ePHRs [79], physicians viewed them as more of a patient, receptionist, or nurse tool [68, 79].

Facilitating/impending conditions

Our review identified the existence or otherwise lack of the following organizational and/or technical infrastructures to support or impede ePHR use: being in an organization’s priority list, integration into the EHRs, patients ready access to resources such as computers, the Internet, and ePHRs, adequate technical support, and proper training on ePHR use [5, 10, 12, 34, 42, 46, 50,51,52, 56, 60, 64, 65, 76, 84, 88].

Due to its impacts on physician’s time management and workload, “physician resistance” was mentioned as “the greatest barrier to ePHR implementation” [12]. There were also concerns about the impacts on providers’ available time for care, lack of reimbursement, or professional liability issues [36, 64, 68]. Physicians voiced their concerns about excess time and efforts to handle issues related to the ePHRs due to lack of integration with EHRs [79, 80].

Incentive motivation

Tangible incentives and cost compensations, or otherwise lack thereof, were also an important factor [12, 54, 64, 65]. For example, it was important to be certain about how ePHR-related services would be paid for, who would pay, and under what circumstances [41]. The cost of services was also mentioned as a barrier by patients [76, 83, 88].

Technology factors

This section provides the results related to the perceived usefulness of ePHRs, perception of external control, compatibility, and perceived complexity.

Perceived usefulness

Perceived usefulness featured as a key driving factor for the intention to use ePHRs [10, 49, 59, 65, 79, 80]. For example, non-users mostly expressed concerns about simply not seeing the value of using a portal to manage their health or lack of personalization in using this technology [65].

Perception of external control

Preserving general privacy, confidentiality, and security of health records was one of the most common concerns regarding ePHR use (e.g., confidentially of a stigmatized or sensitive condition, or confidentiality and security of information easily accessible to researchers and industry members, and misuse of information by insurance companies to deny coverage) [10, 45,46,47, 52, 58, 65, 68, 72, 76, 78, 87, 88]. Patients voiced their concerns about caregiver’s access to their information and requested appropriate access limitation [52, 53, 68]. Clinicians’ attitudes towards caregiver involvement in ePHR use were controversial in one study: while 28.3% favored it, 32.1% disagreed because it impaired patients’ privacy [80].

Moreover, patients reported frustration at several instances in which their profile, medication list, lab results or medical history were incorrect or missing in the ePHR but they were unable to correct them [46, 50, 64]

Compatibility

The degree to which an ePHR was perceived as being consistent with the existing values, past experiences, and needs of its potential adopters i.e., chronic patients and their caregivers and providers were mentioned as an important factor for adoption in some studies [36, 40, 44, 64, 90]. When comparing with the traditional chronic care, users asked for much easier navigation through ePHRs, access to additional information (e.g., progress notes, outside test results, personalized medication information, and a structure to track the course of treatment) or a customized ePHR based on their specific chronic illness [36, 39, 41, 44, 46, 50, 60, 62, 78, 80].

Perceived complexity

The difficulty of understanding or navigating an ePHR was one of the most common barriers referred to in the included studies. Use of problematic medical jargon, confusing information display, and unclear presentation of information based on patients’ knowledge (e.g., unclear numeral values and unfamiliar medical terms) were only some of the barriers that prevented effective ePHR use [33, 35, 38,39,40,41, 46, 49, 50, 59, 60, 68, 76, 78, 79, 82].

Characteristics of chronic disease

Attitudes on negotiated collaboration and preferences for self-regulation

On the one hand, the feeling of having more control over the disease was a reason for limited portal use by patients [34, 59, 79]. Providers also doubted whether patients who were proficient at monitoring their disease were the right group to benefit from ePHRs [79]. On the other hand, being an active healthcare consumer and having a worse or higher proportion of co-morbid conditions and taking more prescribed medications were linked to ePHR use [34, 45, 54, 66, 84, 89]. It was also noted that patients’ willingness to take responsibility for their health through ePHR depended on their coping style and perceived competence and autonomy [71]. In a study, patients who “felt too confronted when monitoring the course of their illness” dropped out of an ePHR [80].

The perceived complexity of care

Based on the task-technology-fit model, a study found instances of mismatches between user mental models and the technology, which manifested primarily as vocabulary misunderstandings, as portal functionality that did not perform as the patient expected, and as requests for clarification and help [33]. They did not have a very concrete understanding of how health information management tasks and processes underlying the ePHR worked.

Characteristics of healthcare settings, providers, and chronic illnesses

In one study, patients in rural settings were less likely to use ePHRs compared with those in urban settings [34]. However, if patients received care at multiple sites, they were more likely to use ePHRs. Patients acknowledged the need to consolidate data produced by multiple providers and scattered in different locations through ePHRs [33]. A lack of interoperability between ePHRs and EHRs in provider offices was noted in three studies resulting in excess workload and frustration [46, 50, 80]. In a survey of patients from 29 states across the US, with at least 38 different types of portals, 51% reported having two or more portal accounts creating frustration when it came to patients remembering their names, and managing different portals from different providers [56]. This was a concern in another study, too [65].

Another problem was confusion over who should receive and reply to messages on the provider team i.e., a physician, a nurse, the office staff, or the entire care team; because this would impact the content of patient messages [37, 46, 86]. In another study, patients had unsatisfactory communications with the care team through a portal; for example, they failed to track their health issues in a coherent way [43]. Physicians were concerned about clarity about responsibilities (and potential liabilities) related to responding to patient-added information or commentaries seen by several different clinicians [36]. For the sake of clarity, Table 1 provides a summary of provider-specific barriers.

Table 1 Barriers to the adoption and use of care providers on the basis of the UTAUT
Full size table

The methodological quality of the included studies

Additional file 3 provides the results of the quality check for included studies. There were some quality issues mainly about data collection and interpretation in four studies [5, 87,88,89]. Because the MMAT discourages excluding studies based on methodological quality, we did include all 60 identified studies in our analysis and report.

Discussion

Understanding barriers that prevent realizing the ePHR’s full benefits is a prerequisite to future work aimed at its optimal use. Our comprehensive review identified 60 relevant studies, which reported barriers to ePHR adoption/use associated with the interacting factors of personal, environmental/medical practice, technology, and chronic disease condition. Our findings expand on those of earlier reviews [18, 19] and point out that our knowledge base for this topic is still limited (and one dimensional), with most of the research predominantly focusing on facilitators than barriers and also on barriers at the patient level than those existed beyond the patient level in chronic disease care.

Differences among users of health information technology (HIT) and the implication of their needs and requirements for design and development have recently gained further attention [91,92,93,94]. ePHRs are aiming to empower patients and/or caregivers and engage them in collaborative and productive chronic care with health professionals. Failure to acknowledge the characteristics, needs, and requirements of all these user groups will lead to the development of unpredictable barriers to ePHR adoption leading to its sub-optimal use. In line with the previous literature, our findings highlight the impact of “digital divide” at the patient level (in terms of age, gender, health, and technology literacy, and socioeconomic status) and several attitudinal factors such as coping styles with a chronic condition (e.g., denial of a condition) and preferences for personal communications with care providers (e.g., preference for direct contact) [19, 95, 96]. Our review also points out that the literature has heavily focused on the elderly, probably because they are disproportionately represented among patients with chronic diseases. Thus, it is plausible that the barriers faced by the younger and also middle-aged (<50 years old) chronic patients would be underrecognized and ePHR use in these groups may fall behind. This is particularly important because of the increasing prevalence of chronic diseases such as diabetes in these age groups. These groups increasingly represent users with higher educational levels and technology literacy (with higher needs and expectations), compared with that of the elderly, introducing a niche market for ePHRs and a unique opportunity to tap into their potential. Moreover, while providing the care for young chronic patients (e.g., between 10 and 19 years old), these adolescent users are a different user group when compared to their parent caregivers and this becomes more important especially when these patients transit from pediatric care to receive adulthood care. Thus, their needs and requirements for an effective and useful ePHR should be given much attention helping a smooth and safe care transition. Such an approach will, to a great extent, make sure that using an ePHR becomes an enriching experience for both the adolescents and those involved in their care.

Most of the healthcare systems have important constraints in terms of human resources shortages, inadequate infrastructure, and insufficient finances, which require mindful management to operate and maximize efficiency [97]. ePHRs have the potentials to do so by facilitating self-care and virtual visits. However, in the context of ePHR use, the responsibilities of patients and providers are changed in many ways. For example, they need to make sure that the data available in different locations are accurate, integrated, and updated [3]. This is important particularly because data about chronic care is scattered throughout different EHRs that do not speak together; and then, the task of data integration is informally delegated to patients. ePHRs can be used meaningfully if they are implemented as a component (i.e., a tool for self-management) of comprehensive care models developed for chronic care (such as the Wagner’s Chronic Care Model [98]). If such models are implemented and proper links are made among their components (i.e., self-management, clinical information systems, disease registries, and decision support systems), patients are freed from extra responsibilities and can focus on productive “self-management” through an ongoing collaborative process with their providers via ePHRs. Therefore, as the adoption of ePHRs are very related to the adoption of EHRs, the barriers related to EHRs in the first place and then the interoperability between these two should adequately be addressed [17, 21, 27]. For example, providers should consider ePHRs’ potentials and their fit within the information infrastructure of their practice when they commence investing in EHRs and choosing their vendors. This becomes especially important after changes that the outbreak of the novel coronavirus disease has brought up to the current and future practice in terms of managing virtual visits. Unfortunately, discussion on such issues has been underrepresented in the identified studies, which should be taken into account in future research.

Our review shows that the barriers related to the providers and the organization of chronic care have not fully been studied despite their importance (studied by only 8 studies). The lack of provider interest and even their resistance to adopting ePHRs are important [3, 12, 21, 99]. Provider concerns about the impacts on workload, professional/legal liabilities, relationships with patients, and the lack of reimbursements should be fully addressed [15, 21, 100, 101]. Moreover, their involvement in ePHR use has not been given full attention as it deserves. In a review of 19 ePHRs, only half had enabled user actions taken by physicians [17]. Providers can act as an effective catalyst in this regard by practicing their social influence on patients [102]. Scholars have highlighted that without involving providers in ePHR’s design, implementation, and application and without addressing their barriers, efforts for widespread ePHR adoption/use would be in vain [3, 103, 104]. Therefore, it will be insightful if future studies explore in more depth provider issues and how they can further be engaged with this emerging technology in chronic care.

Functionalities of ePHRs that provide solutions for personalized needs and requirements of chronic patients have important implications for their adoption and use, as also emerged in our review [17, 19, 73, 90, 105]. One review suggested that features such as access to personal health data and general health information, communicating with providers and support groups, and receiving personal decision support were linked to empirical evidence of benefits from ePHR-enabled self-management [19]. Yet, no ePHR in that review described a platform for all those features. Furthermore, the necessity for measures to ensure the privacy and confidentiality in record transactions and communication through ePHRs was a serious concern voiced by clinical directors and health information technology leaders, besides patients [10, 12, 45, 47, 52, 58, 65, 68, 72, 76, 78, 80]. The relevancy of this concern has also been highlighted elsewhere [20, 21, 27, 95, 106]. Reviewing privacy policies of 24 ePHRs showed that such concerns are very relevant and that compliance with privacy standards and regulations were generally low [14]. It has been recommended that institutions should assemble governance groups to develop policies regarding security, privacy, and confidentiality of records to assure ePHR users on preserving their rights [12].

Strengths and weaknesses of our review

To our knowledge, no study to date has analyzed ePHR studies exclusively concerning barriers to its adoption and use in chronic care. Nevertheless, our review has several limitations. First of all, we only included studies published in English. Second, facilitators and barriers to the adoption of technology is a complex concept without an agreed-upon research methodology. It is plausible that many of the discussions about these core concepts have appeared only in non-peer-reviewed or research publications such as white papers, perspectives, editorials, etc. The findings of our systematic review are confined by the content of included articles, and hence may not well reflect a proper balance of what is known on the topic. Such reviews, however, point out the gaps and direct future studies. Third, ePHRs are an evolving technology with new features and functionalities and so is their position in chronic care. Therefore, the barriers identified in this review are possibly not generalizable to all patient populations or different implementation strategies and healthcare systems. For example, a majority of studies are from the US and therefore a Western viewpoint is predominant here. Therefore, it should be born in mind that the barriers faced by users might be different in different healthcare contexts.

Conclusion

If we are to reap the full benefits of ePHRs in chronic care, we ought to understand the unique characteristics of this type of care and the barriers and challenges that ePHR users face in adoption and sustained use, in the first place. This knowledge should be used to make ePHR functionalities that fit in these unique characteristics well. Future research must aim at identifying the barriers experienced especially by younger chronic patients and their requirements and expectations, and also those barriers faced by care providers all beyond the patient level. A deeper understating of these barriers will reveal opportunities that if addressed in the design, development, and implementation can lead to the enhanced use of ePHRs.