A total of 11 review papers analyzed 1 specific type of digital health intervention applied to PHC—(1) 3 (27%) reviewed client-to-provider telemedicine interventions, (2) 1 (9%) reviewed provider-to-provider telemedicine interventions, (3) 3 (27%) reviewed health worker decision support interventions, (4) 1 (9%) reviewed interventions tracking patients’ health status, (5) 2 (18%) reviewed client participation and self-care interventions, and (6) 1 (9%) reviewed interventions providing education and training to health workers (Textbox 3).

A total of 29% (2/7) of the scoping reviews analyzed the literature on the state of client-to-provider telemedicine with a focus on nonurgent PHC practice. One review focused on web-based consultations using mobile devices [30], and the other included all remote health services delivered via real-time communication between the patient and the HCP [31]. Both reviews agreed that this kind of intervention allows for patient follow-up and clinical condition monitoring and aims to provide quick and easy access to meet medical needs.

In addition, a systematic review examined the evidence supporting the use of secure messages associated with other electronic health record (EHR) resources, such as patient portals and personal health records (PHRs; client participation and self-care interventions), focused on diabetes management. Secure messaging was defined as “any electronic communication between a HCP and patient that ensures only those parties can access the communication” [32].

Beheshti et al [31] showed that telehealth has spread in many countries, including papers that analyzed interventions in Europe (half of them, ie, 55.8% from the United Kingdom), United States (34.9%), Asia (2 papers), and Brazil and Zambia (1 paper each). Most studies analyzed in this review focused on investigating the effects of these services on rural and underserved areas.

However, the concentration of studies in Europe and the United States may be associated with the various implementation barriers cited by Hickson et al [30] and Beheshti et al [31]. As obstacles to the implementation of client-to-provider interventions, they cited infrastructure barriers such as lack of security to transmit confidential information, low bandwidth, unsuitable networks, low signal quality, and bad image quality; system barriers such as the lack of integration with electronic medical records, workflow, and other existing systems; legal barriers such as professional regulation when patients and HCPs are in different states or countries; and billing and reimbursement barriers to guarantee the monetary incentives for HCPs to deliver remote health services. These studies also presented patient barriers, such as confidentiality and privacy concerns and the demand for supervisor support, and HCP barriers, such as reluctance toward new approaches to health care delivery, concerns about the absence of patient health literacy, the perception that telehealth technology is unintuitive and inflexible, and the fear of increased workload.

Despite the implementation difficulties, Hickson et al [30] presented studies that showed improvements in health service outcomes. Using remote health services through mobile devices resulted in cost savings for acute care delivery by decreasing in-office spending per member per month and by preventing face-to-face visits in settings that applied web-based consultation reimbursements to a fee-for-service model of patient payment. Beheshti et al [31] presented only 1 study that indicated clinical condition improvements in patients with diabetes with lower blood sugar levels using telemedicine, in addition to other benefits such as better access to health care, less waste of time, and a high degree of satisfaction with health care. In addition, Kuo and Dang [32] reviewed 11 studies, of which 7 suggested a significant improvement in the glycemic control of patients with diabetes (primary outcome) resulting from the use of secure messaging, but the impact on other outcomes such as cholesterol and blood pressure was inconsistent. A total of 10 of the studies included in this review were conducted in North America.

The 2 scoping reviews also specified that there are several communication tools for delivering remote health care but asynchronous technologies (eg, emails and SMS text messages) seemed to be supplementary and did not reduce the frequency of face-to-face visits. The synchronous form is the most dominant approach for delivering telehealth services (eg, telephone calls, videoconferences, and teleconferences). Finally, despite the identified benefits of client-to-provider telemedicine interventions cited by these 2 reviews, such as cost reduction, decreased face-to-face visits, patients’ time savings, and health care access and clinical condition improvements, the 2 scoping reviews pointed out that more research is necessary to replicate and expand findings on access to health care and patients’ clinical conditions to identify impacts on workflow efficiency and cost-effectiveness depending on billing and reimbursement strategies [30,31].

A systematic review examined the impact of electronic consultations on the delivery of care, applying the quadruple aim framework that analyzes the effects on population health, the experience of receiving care (patients), the experience of providing care (HCPs), and per-capita costs [33]. This review defined an electronic consultation (e-consult) as an electronic communication tool that allows PHC HCPs to obtain a specialist consultant’s expert opinion promptly.

Most studies included in the systematic review were from the United States (19/43, 44%) and Canada (12/43, 28%). In terms of implementation outcomes, e-consults had sustained use and spread, high adoption, and little pushback from HCPs. The range of specialties accessed by PHC HCPs through e-consults expanded beyond dermatology, including multispecialty services that comprise endocrinology, hematology, cardiology, gastroenterology, and neurology.

The results presented by this systematic review showed that most PHC HCPs reported being satisfied with e-consults, they intended to use this technology in the future because their questions were answered, and the specialty support was conclusive without impact on the workload. However, there were still some challenges, such as unclear directions from specialists to PHC HCPs, lack of information or pertinent questions delivered to specialists by PHC HCPs, lack of patient follow-up by the specialists, and a potential increase in workload for specialists.

Regarding health service outcomes, Liddy et al [33] pointed out that e-consults are effective in faster access to specialist advice and substantial avoidance of face-to-face visits and showed that the waiting time for a specialist response was less than that for traditional referrals. Related to patients’ clinical condition effects, they indicated a potential for reducing adverse cardiovascular outcomes, but there was a potential for harm due to the lack of patient follow-up.

The authors concluded that there are gaps in the evidence on the implementation of provider-to-provider digital interventions relative to specialists’ workload and patients’ perceptions. Furthermore, the divergent findings related to impacts on cost-effectiveness and clinical conditions indicate the need for more research in this area.

In total, 27% (3/11) of the systematic reviews analyzed health worker decision support interventions [34-36]. All reviews examined aspects related to computerized clinical decision support systems (CDSSs), any software designed to offer patient-specific assessments and recommendations to the HCP based on patient data from EHRs and evidence-based clinical practice guidelines. One review sought to identify and quantify the barriers to and facilitators of the use of CDSSs by PHC HCPs [34]. The second one provided a summary of the effects of CDSS use on the clinical outcomes of adult patients with chronic diseases managed in PHC [35]. The third one focused their analysis on CDSSs accessible via mobile devices by PHC HCPs [36]. Most studies included in these 3 reviews were from North America (United States and Canada), Europe, and Australia, with a few studies from Asia, Africa, and South and Central America.

Meunier et al [34] assessed the implementation of CDSSs using the human, organization, technology, net benefits framework, which comprises the interdependent human, organizational, and technological factors related to health information system adoption in PHC. In the human category, they found that the HCPs’ satisfaction and their perceived usefulness of the CDSS recommendations could facilitate the use and adoption of this digital technology. However, when the CDSS offers information overload and recommendations conflicting with HCPs’ expertise, it might result in the HCPs’ resistance, reluctance, and negative attitude toward CDSS use. Furthermore, providing training in computer skills and software use could be a human facilitator. The main factors related to the organizational category are workflow and teamwork. When the CDSS is integrated into the EHR and the clinical workflow, producing quality reports and reducing workload through the expansion of the skill sets and roles of other health professionals in assisting physicians, it could facilitate the use of this digital technology. Conversely, the workflow disruption caused by CDSS implementation, demanding more teamwork and increasing the workload, might become a barrier. Finally, in the technological category, the ease of use, visual layout, usefulness of system features and functions, full integration with EHR databases (including data on specialty care), and technical support were pointed out as factors that could influence the use of the CDSS. However, the lack of computers and tablets and the slowness and poor performance of the software were identified as barriers to CDSS use. Agarwal et al [36] sought to verify the HCPs’ acceptability of and satisfaction with mobile CDSSs, but they did not find direct evidence of this implementation outcome.

Related to health service outcomes, Agarwal et al [36] reviewed papers that analyzed the HCPs’ adherence to recommended practices, guidelines, or protocols and the time between clinical recommendation presentation and appropriate management only for mobile CDSSs. The results were inconclusive for the first outcome because the certainty of evidence was very low, and no evidence was identified for the second outcome.

El Asmar et al [35] and Agarwal et al [36] searched for CDSS use impact on patients’ clinical conditions or behavior changes. The quality of evidence of the studies included in both reviews was rated as moderate, low, or very low. El Asmar et al [35] focused on clinical outcomes of the management of chronic diseases and presented data on diabetes mellitus, asthma, angina, hyperlipidemia, and hypertension. Most results indicated no statistically significant differences between CDSS intervention and control participants (usual care or baseline), and it is important to highlight that the study that showed a statistically significant impact on blood pressure between the intervention and control groups revealed extensive training for participants and widespread use by HCPs. Agarwal et al [36] reported similar results for clinical outcomes of chronic disease management—mobile CDSS use made little or no difference in clinical conditions for people with high cardiovascular disease risk, poorly controlled diabetes, or hyperlipidemia. The authors still presented inconclusive results for the impact on maternal and neonatal health outcomes because of the uncertainty of the evidence. However, they pointed out that the use of mobile CDSSs by community workers may increase medication adherence in people with high cardiovascular disease risk. However, the use of this technology by physicians made little or no difference in people with poorly controlled diabetes.

A systematic review analyzed the literature on remote health monitoring systems for chronic disease management in PHC settings [18]. This type of technology is classified as an intervention tracking patients’ health status. The included studies focused on interventions that automatically monitor patient indicators (sensors or wearables) continuously or those in which patients must collect and transmit personal electronic health information in real time or in an asynchronous way. Most studies included were from Europe and the Americas, with a few studies from Asia and Africa.

The studies included in this review indicated that there is a need for more research in this area. Most of the articles presented implementation processes as a proof of concept, pilot study, or clinical trial without evidence of the impact on actual health services or patients’ clinical conditions or behavior change outcomes. The authors showed some barriers to the intervention’s large-scale implementation, such as the low involvement of health workers, patients, and other stakeholders in the tool design and implementation. The low adherence by HCPs was explained by the lack of technology integration into the existing workflow, limited integration into the information systems already used, and the divergences between the data presented by the remote monitoring systems and the HCPs’ expertise in chronic disease management.

Peyroteo et al [18] concluded that it is necessary to fill in the implementation gaps mentioned to ensure the correct and large-scale use of the digital tool. This is the only way to assess the impacts of interventions tracking patients’ health status on health services and patients’ clinical conditions or behavior change outcomes.

In total, 18% (2/11) of the systematic reviews examined client participation and self-care interventions in PHC. One aimed to identify the benefits of using electronic PHRs based mainly on studies that assessed the patients’ and HCPs’ perceptions of the implementation of this technology [37]. The other review focused on the PHR features that improve medication adherence in the adult population with chronic diseases [38]. Both reviews defined PHRs as a tool that provides patients with their health care and personal medical information, and most included studies were conducted in North America or Europe.

Regarding implementation outcomes, Tuan Soh et al [37] found evidence indicating users’ favorable opinions on using PHR systems. The positive effects pointed out by these authors were on patient empowerment; improvement in self-care management, communication, and the relationship between patients and HCPs; and improvement in quality of care by supporting shared decision-making based on shared data, responsibility for health information that could be updated anytime by the patient, in addition to supporting interventions tracking patients’ health status and promoting continuity between the different levels of care. However, the authors indicated that more research is needed to understand structural, organizational, and policy barriers to the implementation of PHR systems and the challenges to ensure data accuracy and reliability.

Andrikopoulou et al [38] focused their review on the analysis of one specific patient behavior change outcome: medication adherence. In total, 13 of the 15 studies included reported positive impacts on medication adherence, and only 2 found no difference. This positive tendency was found despite the heterogeneity in participants’ demographics, study location, chronic disease, or medication adherence measurement method. The authors supposed that these results were due to specific PHR features such as reminders, feedback, and alerts on medication management as all studies included at least one of these features. Finally, they highlighted some gaps that demand more research, such as considering the particularities of patients with multimorbidity or polypharmacy when designing PHR resources to guarantee medication adherence.

A systematic review analyzed the provision of education and training to health workers on antibiotic management in PHC to summarize evidence on the effectiveness of digital education compared to traditional education for improving clinical practice [39]. For this reason, there were no reports of implementation outcomes, such as human, organizational, or technological barriers and facilitators. All included studies were from Europe and North America, except for one that was conducted in China.

Regarding health services, the authors presented results of specific knowledge outcomes, clinical practice improvements, and physicians’ attitudes toward the interventions. They found that digital education (mobile phone or PDA device with internet access) may improve knowledge scores measured using questionnaires compared to traditional education. Some forms of digital education (eg, email with feedback and suggestions and web-based learning resources) showed an impact on reducing antibiotic prescribing or dispensing rates in 4 studies included in the review, and another 2 studies indicated no difference between digital and traditional education when using SMS text messaging or web-based blended education. These results seemed to be more relevant when digital education was combined with web-based training on point-of-care diagnostic tests and enhanced communication skills. Finally, one study reported that almost all physicians were satisfied with the intervention and wanted to continue receiving clinical recommendations about antibiotic prescriptions.

In total, 4 studies included in the review by Kyaw et al [39] assessed patients’ clinical conditions measured using reconsultation rate, probability of consultation for a new respiratory tract infection, hospital referral, and postintervention hospital admission rate. In summary, these studies reported little or no difference between digital and traditional education. However, it can be assumed that the real impact of digital education on antibiotic prescription may be on antibiotic resistance.

A total of 28% (5/18) of the review papers searched for all types of digital interventions applied to six specific health issues in PHC settings: (1) preventive care, (2) chronic disease management, (3) behavioral health disorders, (4) the COVID-19 pandemic, (5) multicomponent PHC interventions, and (6) care coordination (Textbox 4).

A scoping review examined the digital interventions used in the US PHC services to enhance and support the delivery of primary, secondary, tertiary, and quaternary preventive care [40]. Most digital interventions identified in this review focused on electronic medical records, with alerts to help health professionals make appropriate clinical decisions (interventions for tracking patients’ health status and health worker decision support), or mobile interventions to improve patients’ self-care (client participation and self-care interventions). In addition to these types, the authors identified the use of client-to-provider telemedicine and targeted client communication interventions to promote preventive care.

Most studies included in this review showed statistically significant improvements in primary prevention health services and implementation outcomes (adolescent and adult vaccination); secondary prevention health service outcomes (screening rates for colorectal cancer, hepatitis C virus, and osteoporosis but not for colonoscopy); and tertiary prevention health service and clinical condition or behavior change outcomes for managing diabetes, hypertension, asthma, obesity, cardiovascular disease, hyperlipidemia, and mental conditions. There were no studies included on digital interventions in quaternary prevention. Finally, the authors concluded that digital interventions enhanced PHC by augmenting the comprehensiveness of the care provided but there are still gaps in the evaluations of mobile interventions [40].

A scoping review summarized the evidence on digital interventions for chronic disease management in PHC in low- and middle-income countries (10 studies from South America, 9 from Asia, and 2 from Africa) [41]. The review identified digital interventions of almost all types: client-to-provider telemedicine, targeted client communication, health worker decision support, client participation and self-care, provision of education and training to health workers, and laboratory and diagnostics imaging management. However, most digital health interventions still used SMS text messaging and smartphone apps for communication, making them easy to use and affordable as they only required cell phones and internet access.

This review indicated consistent positive results for implementation outcomes, such as improvements in accessibility, patient perception of the health service delivery, HCPs’ capabilities, and better cost-effectiveness. However, the clinical conditions and behavior change outcomes were highly mixed, with positive, negative, neutral, or nonsignificant results. In addition, the authors found that communication-related digital interventions (client-to-provider telemedicine and targeted client communication interventions) produced more consistent improvements in clinical conditions or behavioral and implementation outcomes but interventions focused on HCPs’ capacities (interventions for health worker decision support and provision of education and training to health workers) resulted in substantially mixed clinical condition or behavior change outcomes. These authors defend that research with a larger number of participants followed over a longer time may result in more consistent evidence of the impact of digital interventions [41].

A scoping review was conducted to gather and analyze research on how digital health services can enhance the collaborative care model, which addresses behavioral health disorders in PHC settings [42]. The studies included presented digital health interventions of different types: client-to-provider telemedicine, targeted client communication, health worker decision support, tracking of patients’ health status, client participation and self-care, and provision of education and training to health workers.

Regarding implementation outcomes, both patients and HCPs had positive perceptions of these digital interventions, which facilitated the communication between users and freed up HCP time for patients. Despite these positive outcomes, privacy and workload concerns arose when the technology was not fully integrated into the clinical workflow, resulting in low user engagement and high dropout rates. The improvements in clinical conditions or behavior change outcomes presented in the included studies were modest, partly due to methodological factors such as small sample sizes and different control groups [42].

A total of 29% (2/7) of the scoping reviews explored the use of digital interventions in PHC settings during the COVID-19 pandemic. One focused on the Australian PHC response to the pandemic and found interventions on client-to-provider telemedicine, client participation and self-care, and education and training for health workers [43]. The other was a more extensive review including 18 studies conducted in North America, 14 conducted in Europe, 4 conducted in South America, 4 conducted in Asia, and 4 conducted in Oceania; they identified targeted client communication interventions in addition to the digital technologies found in the other review [44]. Both reviews concluded that these digital technologies were implemented to manage COVID-19 as well as other common diseases in PHC settings.

The 2 reviews emphasized that the key facilitators in implementing digital technologies during the COVID-19 pandemic were providing technology training and support for HCPs; offering validated and reliable technologies with a user-centered design that were suitable for the local context; and ensuring user security, acceptability, and satisfaction. However, there were implementation barriers, such as lack of digital facilities in remote and rural areas, safety and privacy concerns, lack of planning for health services and clinical protocols when using digital tools, and the need for new funding arrangements that reimburse for remote services. Although digital technologies during the pandemic had particular importance for chronic condition management, reducing the number of referrals to specialties and hospitalizations and speeding up screening in cases of clinical changes through remote tracking, these tools were not always effective in identifying complex or urgent demands and assisting older adults and children [43,44].

According to Silva et al [44], digital health emerged as the sole substitute for PHC during the COVID-19 pandemic. However, the decrease in the quality of health care could be attributed to the hasty implementation of remote services in not properly equipped facilities and by health professionals who were under immense pressure.

A systematic review explored how digital health has enhanced multicomponent PHC interventions, defined as initiatives comprised of several features aimed at enhancing PHC [45]. The review identified 6 types of digital technologies implemented in this context: client-to-provider telemedicine, provider-to-provider telemedicine, targeted client communication, health worker decision support, tracking of patients’ health status, and client participation and self-care. Most studies included were from the United States and Europe, except for one from Argentina.

In terms of implementation outcomes, this review showed consistent findings for cost savings, decreased health care expenses, and improved perceptions among HCPs regarding the quality of the workplace and services offered. However, the results were mixed for other outcomes, such as hospital admissions, readmissions, emergency department visits, the number and cost of drug prescriptions, and patients’ perception (increased satisfaction with timely appointments and self-management support but not with PHC performance). Regarding clinical conditions or behavior change outcomes, there were consistent improvements in blood pressure and glycated hemoglobin control and reduction in amputation, smoking, and the number of patients with end-stage renal disease, but this was not the case for outcomes related to cardiovascular diseases [45].

A systematic review analyzed how digital interventions have supported the laboratory testing process in PHC considering all its phases, from the decision process on the need to carry out a diagnostic test to the interpretation and communication of the test result, contributing to care coordination [46]. The review identified 4 types of digital interventions: health worker decision support, tracking of patients’ health status, client participation and self-care, and laboratory and diagnostics imaging management. Most studies included in this review were from the United States, with some conducted in Europe and Peru, and most of them focused on health service outcomes, but some implementation outcomes were related to health professionals’ perceptions.

The studies included in this review indicated that digital interventions related to tracking patients’ health status and health worker decision support resulted in reducing the amount of effort spent by the staff on collecting information and promoting best care practices but not necessarily reducing the number of diagnostic tests prescribed in the preanalytical phases. The digital laboratory and diagnostics imaging management interventions seemed to impact health care positively in all phases of the testing process, reducing professional workload and the number of errors when the technology was integrated into ambulatory systems and saving time to order prescriptions, reorganize tasks in the intra-analytical phase, and return the test results. Despite this positive impact on health services, in some studies, health professionals indicated that these systems need to be improved in terms of design and reliability, and technical problems and lack of integration between systems were associated with negative effects on the workflow and increased workload. Finally, client participation and self-care interventions to directly communicate the test results to the patient contributed to better collaboration between patients and HCPs and promoted the self-management of chronic diseases [46].