Over the past few decades, health care has been shifting from a paternalistic to patient-centered model that values patient engagement and shared decision making [1,2]. These values align with the patient-centered care model, where clinicians provide care that is tailored to the distinct needs of the patient. It is based on the development of respectful and dignified therapeutic relationships [3].

To provide patient-centered care, clinicians need to know and incorporate the patients’ context into their communication and care with patients. Patient contextual data (PCD) encompass social determinants of health (SDH) [4] and further comprise patients’ needs, values, goals, and preferences relevant to their care [5]. In the primary care setting, clinicians address most patients’ health care needs through a sustained partnership with patients and within the context of family and community [6]. Therefore, care teams must have access to data about the patients’ perspectives, values, and other contextual considerations to tailor patient-centered conversations and clinical decisions [1,6,7]. PCD can facilitate team-based care by enabling health care team members to build a rapport quickly and to connect with patients on a humanistic level [8].

Evidence suggests that connecting with patients can bolster patient activation. In a cross-sectional study, individuals at the highest level of activation (level 4) received relevant preventative cancer screenings, had 5 of 6 clinical indicators in the normal range, and did not engage in unhealthy behaviors (tobacco smoking and obesity) at statistically significantly higher rates compared with individuals in the lowest level of activation (level 1) [9]. A longitudinal study affirmed the results that indicated that people at the highest level of activation had significantly higher odds of guideline-concordant high-density lipoproteins and serum triglyceride levels, normal Patient Health Questionnaire-9 scores, not smoking, not being obese, having no Emergency Department visits, and having no hospitalizations in the 2-year follow-up period compared with people at the lowest levels of activation [9,10]. These outcomes translated to lower health care costs, with a projected 31% decrease in costs for people who were most activated (level 4) than those who were least activated (levels 1 and 2) [10].

Despite evidence indicating that connecting with patients improves patient outcomes and reduces health care costs [9,10], PCD are often not collected as a routine component of care [11]. A barrier to the integration of PCD is linked to the current limitation of electronic health record (EHR) systems in integrating and facilitating the retrieval of social risks and care preference data [12], even if collected as unstructured data within clinical notes. Clinicians face several limitations in terms of time [5,13] allocated to clinical visits and tools to gather a comprehensive understanding of their patients’ needs, values, preferences, goals, and concerns. System-level barriers lead to missed opportunities to individualize care and act upon PCD that might have a substantial impact on patient outcomes and the experience of care [14].

Previous research has demonstrated that patients reveal more sensitive information via health information technology than during patient visits [15]. Studies show that unvoiced concerns and goals for care disproportionately relate to the patients’ experience of illness [16,17], patients’ expectations of treatment [18], or psychosocial concerns [17,19-21]. These contextual errors (ie, disregard of PCD in care planning) [14] are more costly to the health care system than biomedical errors (ie, guideline-discordant care) [11]. Conversely, when providers incorporate PCD into the care context, patients’ engagement in their self-management and adherence to the agreed-upon care plan increases [22]. Furthermore, when the health care team collects and incorporates PCD during a visit, it facilitates rapport building and aligns patient and provider goals [5].

Research indicates that there are ethnic and racial differences in the adoption of consumer informatics tools [23-26]. In a study of a national sample of US adults, ethnic and racial minorities were less likely to be invited to use a patient portal than ethnic and racial majority populations [27]. Furthermore, individuals who did not use a patient portal were more likely to be unemployed, receive Medicaid insurance, have less than a college-level education, did not have a regular health care provider, were male, and aged 65 years or above [26]. As patient portal usage has been shown to be associated with improved quality measures and is thought to contribute positively to patient safety, digital tools should be assessed for their capability to be adopted by a wide range of the population and to narrow, rather than grow, the gaps in care across groups [28].

Given that disparities exist in the adoption and use of consumer informatics tools [27,29,30], researchers must evaluate ways to reach vulnerable populations when testing new consumer information technologies. Current trends suggest that internet access is no longer the main cause of the digital divide [31]. Instead, some patients lack the knowledge, skills, and confidence in using technology [32]. Providers may be key to reduce differences in consumer technology use by inviting all patients to use the new technology, discussing privacy and security concerns and providing resources tailored to a low health literacy level, on use [26]. Additional strategies to reach the most vulnerable individuals and across racial groups include developers employing patient-centered design strategies such as a simple, clean, and aesthetically appealing interface [33]; incorporating patient education on how to use the technology [34]; and promoting the new technology in various ways [35]. Conceivably, introducing a new consumer informatics technology designed to improve patient activation and communication may not achieve the desired rates of adoption, unless health care team members actively promote and assist in the use of the technology [8,25,35,36].

In this randomized controlled trial, we aim to evaluate the influence of PCD, collected using a consumer informatics tool, for previsit planning and routine clinical visit discussions with the health care team. The goal is to compare postvisit patient-provider communication and changes in patient activation among patient users and nonusers of the PCD tool, accounting for differences between non-Hispanic White and Black participants (hereafter White and Black). We will measure these constructs using 2 validated measures, the Communication Assessment Tool (CAT) [37] and the Patient Activation Measure (PAM) [38]. Furthermore, we will evaluate the impact by race by determining whether PCD could help mitigate any baseline differences in patient activation and postvisit patient-provider communication between White and Black patients.

We hypothesize that inviting patients directly to submit this information may help with several factors, including activation (patient ready to manage their health and care) and communication (helps prepare perspective and helps the clinician identify salient points).

The primary aims of this trial are to (1) assess the effects of using PCD on patient-provider communication (primary outcome) and patient activation (secondary outcome) and examine whether the effects are different for Black and White patients, accounting for age, gender, and other patient factors and (2) evaluate whether baseline measures of patient-provider communication and patient activation modify the effectiveness of PCD in improving either outcomes, accounting for age, gender, and other patient factors.

We have 2 outcomes of interest: (1) patient-provider communication measured using the CAT [37] and (2) patient activation measured using the PAM [38].

This trial will assess the impact of incorporating PCD on patient-provider communication and patient activation of Black and White participants. The trial is registered at ClinicalTrials.gov (NCT03766841). The health network’s ethics review board approved this trial (registered project PRO00031177). The study protocol adheres to the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) 2013 [39] checklist (Multimedia Appendix 1).

Using an experimental study design, we will recruit a prospective cohort of eligible Black and White patients from primary care clinic sites randomized to intervention (invitation to use the PCD tool with facilitated enrollment) or usual care (invitation to use the PCD tool only) and administer questionnaires at baseline and after their primary care visit. The questionnaires assess the perceptions of visit communication and patient activation. The survey results will be adjusted for previsit measures of communication as the CAT [37] is only validated as a postvisit measure of patient-provider communication. Figure 1 presents the randomized controlled trial study design.

An a priori power analysis was performed to estimate the required sample size using G*Power 3 [40,41] based on the study’s primary outcome, the CAT [37]. These were conducted for the more straightforward two-sample t test procedure, as this is known to yield a conservative assessment of power. For a two-sided test at α=.05, a total sample size of 200 results in 80% power for a standardized effect size of 0.4 and 94% power for an effect size of 0.5. Increasing the sample size to 250 raises the power to 88% and 98%, respectively. A sample size of 250 provides 79% power to detect a standardized effect size of 0.35. To account for up to 20% potential dropout over time, we aim to enroll 300 participants (targeting 150 Black and 150 White participants). Once participants provide consent, REDCap (Research Electronic Data Capture) [40] will randomize them into 1 of 2 experimental arms: (1) PCD tool (ie, facilitated enrollment for PCD tool intervention) or (2) usual care (ie, email invitation only for PCD tool). We will use stratified random sampling to ensure equal representation of Black and White participants in each arm. Stratified randomization prevents an imbalance of racial representation between arms.

There is increasing attention on the role of health information technology and digital health tools to enable patients to collaborate with providers by sharing and acting upon PCD [12,63-70]. Adoption of consumer-facing informatics tools may overcome the barriers of current EHRs by directly engaging patients to share PCD. Moreover, with the advent of application programming interfaces and the increasing level of interoperability of EHR systems, these consumer applications can integrate PCD information into current EHR systems to make the data available for use by clinicians and health care teams [1,71]. In particular, consumer informatics tools that gather and share PCD are hypothesized to improve communication [21] and health outcomes [72,73]. Effectively, these tools would support the EHR in providing a more holistic understanding of the patient.

In an earlier study [8], digitally engaged patients reported that the completion of their profiles in a consumer informatics tool (ie PatientWisdom, Inc.) promoted reflection of their health goals, challenges, and priorities. The reflection led to actions toward goal attainment and targeted conversations with their health care team about issues important to them [8]. Research further supports that individuals who have robust digital engagement using consumer informatics tools have higher participation in treatment follow-up and self-care [74,75]. When care goals were aligned, racial and ethnic minority populations experienced improvements in patient-provider communication and decision quality outcomes similar to racial and ethnic majority populations [25,76]. Therefore, it is critical to understand whether interventions that elicit and share patients’ social risks and care preferences with the health care team serve as a mechanism to improve individualized care and lessen the gap in health outcomes.

The clinical trial will provide crucial empirical evidence on the effects of a consumer informatics tool that elicits and aggregates PCD for use in the clinical exchange of patient-provider communication and patient activation across populations. The study will occur within the most racially segregated metropolitan area in the United States, where racial disparities in health and health care represent a significant public health concern [77-80]. The study sample may not reflect the population or the complex contextual issues associated with the area.

We acknowledge the following limitations and significant threats to the study and present contingency strategies. This research study will occur within 1 academic medical center, which may limit the generalizability of the results. To mitigate this limitation, we will recruit participants from various academic and community primary care clinics with different staff, providers, milieu, and the composition of patients who receive care. The participants in this study will be Black and White and limited to individuals who can speak English. This inclusion criterion excludes other diverse populations. This limitation is because of the population of patients who are served at the academic medical center. To reach equal racial representation of participants, we will oversample Black primary care patients. Recruitment difficulties for participation may occur. We employ several recommended strategies to recruit Black populations into this trial but lack other strategies such as community involvement and informational sessions [81]. Furthermore, the study may be threatened by volunteer bias, where the participants’ characteristics or outcomes differ from those of nonparticipants [41]. An efficacious strategy to improve participant recruitment and retention is to compensate individuals for their time [41]. Participants will receive a modest financial incentive for participation in this research project. The incentive is US $25 for each survey completed. The study team has also allocated substantial time and resources for personalized telephonic or in-person contact during recruitment, retention, and follow-up procedures. Researchers have successfully used these strategies to recruit and retain historically disenfranchised populations in clinical trials [41].

We also considered the alternative design of an efficiency trial with its advantages of high internal validity [53]. Although there are methodological advantages to this design, the real-life variability of clinical practice precludes the strict adherence to a study protocol mandated in an efficiency trial. Therefore, we chose a pragmatic clinical trial with somewhat diminished internal validity but a high degree of external validity of the results, which is valued in implementation research [53].

Health care stakeholders, clinicians, and patients increasingly call for the evaluation of clinically relevant interventions that are tested in heterogeneous clinical settings with the inclusion of diverse study participants [56,82]. In this clinical trial, we will test an intervention (PCD tool) that is deployed across an academic health network. We focus on understanding the differences by use, adjusting for problems with bias and self-selection of users for the PCD tool. The study design intends to overcome self-selection bias by creating randomization to treatment using various facilitation processes to improve the usage of the tool beyond its baseline. In addition to the ITT analysis, typical in pragmatic trials [55], we will conduct a TOT analysis [57,58] to model estimates of whether a participant used the PCD tool in the follow-up period and then identify the causal impact of using the PCD tool on outcomes (patient activation and patient-provider communication). In this study, the TOT analysis will adjust for participants’ nonadherence to the group assignment, a common occurrence in pragmatic trials [56].

When patients’ preferences and life circumstances drive health care decisions, their quality of involvement in their care improves [9,83-85]. PCD are essential information that, when known and incorporated, may promote the development of a person-centered plan for care [14]. Therefore, interventions that test these relationships must be explored to understand how to optimize individuals’ involvement in self-care. Researchers must also investigate whether the outcomes differ between Black and White patients who experience different social, political, and economic injustices that affect health [86].