Cancer clinical trials (CCTs) are essential for advancing cancer treatment and ensuring quality of care for patients [1]. However, for cancer research to benefit all patients, it is critical to include participants who represent the diversity of the American population. Unfortunately, participation in cancer treatment trials remains low, with an overall accrual rate of only 7.1% [2]. Moreover, only approximately 5% of trial participants are Black or Hispanic, although these groups represent 15% and 13%, respectively, of people with cancer [3]. The ongoing underrepresentation of minoritized populations in CCTs not only raises concerns about equitable access to care but also limits the generalizability of the findings across diverse patient groups.

Various barriers at the system, institutional, clinician and research team, and patient levels contribute to low CCT enrollment [4-11]. One significant barrier is that a substantial portion of eligible patients are not approached about the possibility of receiving treatment through clinical trials [12]. This issue is especially pronounced for patients from demographic groups who are typically underrepresented in trials (eg, uninsured individuals, members of racial and ethnic minoritized groups, and people living in rural areas) [12-19] and who experience higher cancer mortality rates [20]. Even when the option of clinical trials is discussed with eligible patients, the communication is frequently unclear or inequitable, often reflecting racial disparities in how comprehensively information about CCTs is conveyed [21].

Research has shown that educating patients before their first oncologist visit can improve their knowledge, attitudes, and readiness for treatment decision-making, as well as increase their willingness to ask about and consider cancer treatment trials [22]. Therefore, it is important to normalize receiving treatment through clinical trials and encourage patient inquiries about trials before starting treatment, particularly for underrepresented groups. To effectively inquire about clinical trials, patients need to understand that quality care can be provided through CCTs, that asking about CCTs is encouraged and supported, that their participation in treatment decisions is encouraged [23,24], and that they are capable of fulfilling this role [8-10]. Patient education can start with the patient’s trusted primary care provider (PCP).

PCPs are a potential gateway facilitating access to CCTs because they interact with patients at the time of cancer diagnosis, provide ongoing care, and are trusted sources of information [25,26]. PCPs also report believing that they play an important role in patient care across the cancer continuum, including being actively involved during treatment [27]. Moreover, studies have shown that a trusted physician’s recommendation is a primary factor influencing patients’ decisions to enroll in clinical trials when offered [5,10,28-34] and that this influence may even be more acute for minoritized populations [35]. Therefore, improving PCPs’ attitudes, beliefs, and behaviors around CCTs may significantly influence patients’ attitudes and openness toward participation [36-39].

Although PCPs have expressed a desire for more of their patients with cancer to participate in CCTs, they often feel inadequately prepared to discuss trials due to their own limited understanding of CCTs as a high-quality treatment option [40,41]. In addition, they need guidance on effectively communicating with patients about potential trial participation. This gap in knowledge and skills presents a significant opportunity to educate PCPs about CCTs and their role in preparing patients for the oncology referral. These new skills can potentially improve trial accrual and patient access, especially for underrepresented groups.

Previous research, including our own, has demonstrated that conducting education with PCPs who serve underrepresented populations can contribute to breaking down key barriers to trial participation [42-44]. This can be achieved by improving PCPs’ capacity to both (1) educate recently diagnosed patients about the possibility of trial participation at the time of treatment referral to cancer care and (2) provide decision-making support to patients if advice is sought about trial participation after the patient has already met with the oncologist.

The purpose of this study was to pilot-test an innovative web-based training intervention for PCPs, including both practicing clinicians and trainees, designed to help them prepare patients for discussions about cancer treatment and CCTs. There are several types of CCTs, including prevention, screening, behavioral, and supportive care trials. However, for the purpose of this training, we focused specifically on treatment trials (eg, those involving chemotherapy or immunotherapy). We aimed to evaluate (1) PCPs’ general impressions of the training intervention and how the intervention impacted (2) their knowledge about CCTs, (3) their perceptions of their roles in caring for patients with cancer, (4) their approach toward and communication with patients, and (5) their willingness and behaviors in enacting practice-level changes.

A 1-hour web-based training intervention for PCPs was developed and evaluated in a single-arm pilot study with assessments conducted before the intervention, immediately after the intervention, and at 3-month follow-up. We used a mixed methods approach, using quantitative and qualitative data collection and analysis to evaluate the training intervention [45]. We used the Kirkpatrick model to guide our evaluation [46]. The Kirkpatrick model proposes 4 sequential levels of training evaluation that become progressively more distal from the original training. In this study, we focused on levels 1 (reaction), 2 (learning), and 3 (behavior).

After conducting separate analyses, we integrated the results to allow the qualitative data to enhance and provide context for the quantitative data [47,48]. This approach integrates the data for triangulation, which strengthens the validity of the findings and allows for a more comprehensive understanding of the data [49-51]. This study followed the SQUIRE 2.0 (Standards for Quality Improvement Reporting Excellence) guidelines for assessing quality and reporting results [52].

We collaborated with instructional designers (DD and JA) from the University of Florida’s E-Learning, Technology, and Communications department to develop a 1-hour web-based training intervention for PCPs titled “Preparing Patients for Cancer Treatment Decisions: The Critical Role of Primary Care Providers in Facilitating Equitable Access to Care and Clinical Trials.” The content of the intervention was based on the authors’ collective research on the topic [40-43,53] as well as current literature on the role of PCPs in cancer treatment and referral, barriers to CCTs, patient activation, and best practices in medical education.

The asynchronous training intervention was hosted by a PCP (AN-P) and a radiation oncologist (RMV) who introduced content through engaging conversations and videos. Using a model of cognitive dissonance, the course introduced new information about CCTs, with the intent of helping providers realize inaccuracies in their knowledge, attitudes, and behaviors around CCTs and the referral to cancer treatment [54,55]. A clinical trials content expert (MM) additionally provided facts about CCTs, cancer care disparities, barriers to CCT participation, and the role of PCPs in cancer care. The 4 intervention modules covered content to enhance PCPs’ knowledge and communication skills surrounding cancer treatment discussions, including clinical trials and strengthening preparation related to the oncology referral process. The communication skills taught in the curriculum centered on the 5 E’s communication model (ie, explore, educate, encourage, engage in planning, and emphasize partnership), a framework designed to support patients in becoming more active participants in cancer treatment decision-making. Informed by principles of patient activation, treatment decision support, shared decision-making, and patient-centered communication, the model provides a simple mnemonic that clinicians can use to educate patients; prepare them for oncology referrals; and encourage inquiry about treatment options, including cancer treatment trials [56-59]. The learning objectives for each of the 4 modules are displayed in Table 1.

Throughout the modules, video vignettes featuring actors as patients and physicians further illustrate the 5E's communication model. The modules also include case studies where participants can explore patients’ disease histories, backgrounds, and cancer diagnoses. Participants complete each module at their own pace.

To gain further insights into the survey results, we conducted follow-up interviews with PCPs who completed the 3-month follow-up survey.

After conducting individual quantitative and qualitative analyses, we jointly interpreted the findings from both phases to enhance reporting. We report quantitative data first, which are then elaborated with the qualitative findings. When appropriate, we created joint displays to illustrate this data integration and to organize the results [48,68].

The study procedures were approved by the University of Florida Institutional Review Board (202200894). Before the PCPs participated in the surveys and interviews, they were provided a statement of participant rights and responsibilities and given the option to withdraw from the study at any time. All survey participants checked a box agreeing to participate before taking the survey. All interview participants verbally confirmed their willingness to participate, had the opportunity to ask questions before the interview, and were informed that interview data would be transcribed for analysis but only available to members of the research team. All data were deidentified before reporting in this study. Participants who completed the intervention and the pre- and postintervention surveys were compensated US $150. Participants who completed the 3-month follow-up survey were compensated an additional US $45.

A total of 29 PCPs completed the study and preintervention and postintervention surveys; the sample included 15 (52%) women and 14 (48%) men. Of the 29 PCPs, 28 (97%) completed the 3-month follow-up survey (Figure 1). Of these 28 participants, 8 (29%) took part in the interviews. Of the 29 participants, 26 (90%) had MD degrees, 2 (7%) were NPs, and 1 (3%) had a DO degree. More than half were physicians in training, with 17 (59%) describing themselves as residents or fellows in PGY2 (n=7, 41%), PGY3 (n=9, 53%), or PGY4 (n=1, 6%). Participants described their practice areas as internal medicine (16/29, 55%), family practice (11/29, 38%), or other (2/29, 7%). Most reported that their practice or training setting was an academic-affiliated hospital (17/29, 59%) or community-based and owned by an academic medical center (7/29, 24%). Complete demographic and professional characteristics of both survey and interview participants are shown in Table 3.

The mixed methods findings were used to (1) characterize PCPs’ general impressions of the training intervention; in addition, we used these evaluation findings to capture how the training intervention impacted their (2) knowledge, (3) perceptions of their roles in caring for patients with cancer, (4) communication with their patients, and (5) practice-level change willingness and behaviors. The quantitative results are reported first and then elaborated upon with the qualitative findings within each of the 5 areas of impact.

Both survey and interview findings illustrated PCPs’ positive impressions of, and high satisfaction with, the course. Specifically, PCPs expressed satisfaction with the course content, design and functionality, course elements, and course length. Their general impressions are presented in a joint display (Table 4) to depict the mixed methods results comprehensively.

PCPs’ positive feedback about the course aligns with their significant improvement in CCT knowledge after participating in the course. The survey results (Table 5) indicated that PCPs’ knowledge of CCTs was low before the training, with a mean score of 55% (SD 17.9%) across all 7 CCT knowledge items. Immediately after the training, PCPs’ knowledge of CCTs improved significantly to a mean score of 81% (SD 12.1%; P<.001). This improvement from before the training was sustained at the 3-month follow-up, with a mean score of 73% (SD 18.4%; P=.005).

This mixed methods pilot study provides a comprehensive evaluation of an innovative web-based training intervention aimed at enhancing PCPs’ knowledge, attitudes, communication, and practices regarding referrals to cancer care and CCTs. The Kirkpatrick evaluation model [46] provided a helpful framework for these evaluation findings. It is widely used in medical education to assess the outcomes of training and learning programs.

Level 1 of the Kirkpatrick evaluation model focuses on user reaction to the training intervention. Participants had high levels of satisfaction with the training intervention and reported high scores and positive comments about various aspects. Although satisfaction is important for participant engagement, it is not necessarily sufficient to produce desired changes.

Kirkpatrick level 2 focuses on learning as the next important outcome. Before the intervention, PCPs demonstrated relatively low levels of CCT knowledge, with a mean score of 55% (SD 17.9%) across 7 knowledge items, consistent with our earlier assessments of PCPs’ understanding of clinical trial concepts [40]. We found that PCPs’ knowledge about CCTs improved significantly after the intervention and was sustained at the 3-month follow-up, replicating what our team found in a prior study of PCPs in the New York City area [40]. This improvement is particularly important given PCPs’ previously self-identified limitations in cancer-specific knowledge [33,41]. Our evaluation also found an improvement in PCPs’ self-reported attitudes and beliefs about their role in working with patients with cancer, which was sustained at the 3-month follow-up, as well as willingness to communicate with their patients about CCTs, although this willingness to communicate was not sustained at the 3-month follow-up. However, willingness may be less important than actual behaviors.

With a foundation of change in knowledge, attitudes, and behavioral intentions, Kirkpatrick level 3 examines the impact of the educational intervention on behavior. According to our survey data, PCPs reported using the communication skills taught in the course with a higher percentage of patients about cancer treatment and CCTs at the 3-month follow-up than they did before the training within the context of making an initial referral for cancer care. Of the 7 communication items on the PRCP, only 1 (14%)—emphasize partnership as patients go through cancer care—did not show an improvement in scores. There may have been a ceiling effect on this item because the baseline score was already high. With respect to communication skills with patients who had already met with an oncologist, only 2 (29%) of the 7 items on the PRCP saw significant increases, and these 2 items focused specifically on the discussion of CCTs as a potential treatment option. It may be that the other items are not as relevant for a patient who has already had their initial meeting with the oncologist.

Another type of behavior we assessed was participants’ willingness to change their referral behavior after the training and the reports of these behavior changes at the 3-month follow-up. As these are slightly different concepts, we did not compare them; however, it was not unexpected to see lower scores on actual behavior change than intention.

Overall, the qualitative data underscored the positive impact that the training intervention had on the participating PCPs in terms of their evaluation of the intervention, their learning, and their behavior change. The qualitative data highlighted ways in which the intervention facilitated a more informed and patient-centered approach to care by increasing PCPs’ self-efficacy and confidence.

In addition to these positive outcomes, PCPs expressed some barriers to changing their CCT communication behaviors. They voiced concerns about their perceived lack of cancer expertise despite the course addressing the boundary of what PCPs should be expected to discuss. In our prior work, PCPs expressed a similar reluctance to discuss cancer treatments with patients [40], further highlighting a need to expand education regarding the role of PCPs in encouraging patients to engage in decision-making about their cancer treatment, including the consideration of CCTs. Importantly, the training does not suggest that PCPs determine patient eligibility for specific trials but rather prepares them to initiate supportive, patient-centered conversations that help patients feel empowered to ask questions and engage in screening discussions with their oncology team. PCPs can also support patients by reviewing laboratory or imaging results, helping to translate complex findings into accessible language, and explaining how these may relate to cancer treatment decisions or clinical trial eligibility without making determinations about eligibility themselves.

PCPs also relayed broader systemic challenges in their work settings that hindered making behavioral changes. The interview data helped to explain the lower scores on the willingness to change referral practices, as PCPs discussed the systematic barriers hindering collaborative efforts among health care professionals; these barriers in turn hindered their ability to integrate new knowledge and practices into their clinical settings.

Our findings underscore the need for ongoing education to address gaps in the PCPs’ knowledge and skills needed to effectively communicate with patients about the potential of participating in CCTs as a high-quality treatment option. The evaluation of this educational intervention suggests that sustained change in knowledge, attitudes, and behavior can be achieved through an engaging, self-directed web-based intervention. As this training intervention is easily scalable, further dissemination may be able to positively impact quality of care and participation in clinical trials.

As this was a pilot study, the sample size was small, although not too small to have enough power to detect significant results. Notably, the majority of the participants (n=17, 59%) were trainees working in academic settings, which may influence the generalizability of findings to the broader population of practicing PCPs. Furthermore, their receptiveness to the training and self-reported outcomes may not fully reflect the attitudes and behaviors of more experienced PCPs. Future research using a larger and more diverse sample of practitioners, including both trainees and practicing PCPs, would allow us to look for differences in evaluation metrics among different types of PCPs (residents, practicing physicians, physician assistants, and NPs). This study is also limited by its reliance on self-reported measures. Future research could incorporate patient interviews to provide additional insights into the effectiveness of the intervention from the patients’ perspectives. Using larger, more diverse samples to validate the intervention’s effectiveness across different practice settings and demographics is also recommended. Furthermore, including more follow-up surveys (eg, at 6 mo and 12 mo) could evaluate the sustained impact of educational interventions on PCPs’ knowledge retention and clinical practices over time. While this study identified several systemic barriers that may hinder CCT recruitment, addressing these challenges was beyond the scope of this pilot intervention. Future research should explore multilevel strategies to overcome these barriers and support sustained changes in clinical practice. This study, while focused on oncology referrals, is intended to be proof of concept for an intervention that can be adapted for other disease areas requiring specialist referrals, such as cardiology, rheumatology, and gastroenterology.

This pilot study found that a self-guided, 1-hour web-based training intervention for PCPs improved their knowledge about CCTs as well as their attitudes and beliefs regarding their role in discussing treatment options with patients. The intervention also enhanced PCPs’ ability to communicate with patients about CCTs and prepare them for subsequent steps in the referral process, including discussions with specialists. Further testing of the training intervention in a larger sample can lead to future dissemination, with the potential to improve CCT accrual, especially among underrepresented groups. Furthermore, this study underscores the effectiveness of targeted educational interventions in equipping PCPs with the knowledge and confidence to communicate with patients about the potential of trial participation as a high-quality treatment option.