All study procedures were approved by the Seattle Children’s Hospital Institutional Review Board (IRB approval number STUDY00004962; approval date June 11, 2024). The study was conducted under IRB-exempt status in which survey information obtained is recorded in a manner that the identity of the survey respondents cannot be ascertained. Survey participation was voluntary and anonymous, and no compensation was offered for completing the survey. A one-page REDCap (Research Electronic Data Capture) Study Information Sheet containing purpose of the study, investigator information, study description, confidentiality statement, data storage details, and contact information was provided on the main screen of the online questionnaire. Survey respondents were informed via the Study Information Sheet that their responses would remain anonymous and research records would be stored exclusively within secure institutional intranet databases. The survey was administered through Seattle Children’s Hospital REDCap, a secure web-based survey platform that uses password-protected access and role-based permissions to prevent unauthorized data access.

We disseminated an online survey called the “Oncology Provider Experiences National Survey (OPENS): Mobile Health” (hereafter referred to as “OPENS”) requesting responses on provider attitudes and experiences toward the use of mHealth apps in oncology. The target population for this study consisted of health care professionals currently practicing in oncology settings within the United States. Participation was open to all who self-identified as an oncology provider. Although we did not include an eligibility screener, eligibility was controlled with targeted recruitment through oncology provider list servers (listservs). Our survey was distributed nationally across the United States to members of the American Psychosocial Oncology Society (APOS) and provider listservs at the home institutions of several of the coauthors (eg, Seattle Children’s Hospital Cancer and Blood Disorders Center, Fred Hutchinson Cancer Center, MD Anderson Cancer Center, St Jude Children’s Research Hospital). We distributed the survey to several organizations that are pediatric-focused: Children’s Oncology Group (COG) Behavioral Science Committee, Society of Pediatric Psychology Hematology/Oncology/Bone Marrow Transplant Special Interest Group, and Association of Pediatric Oncology Social Workers (APOSW). Permission to distribute our online survey was obtained by research committees or administrators, as appropriate for each respective listserv. Each listserv was sent a single request with a designated survey end date and with no follow-up reminders. We staggered survey distribution pending submission and review of materials by each listserv’s respective research committee and administrators. Prospective participants were informed via the survey dissemination email and the REDCap Study Information Sheet that the survey was intended for oncology providers only. The survey took approximately 10 minutes to complete. The survey remained open from June 2024 to December 2024. This study was reported in accordance with CHERRIES (Checklist for Reporting Results of Internet E-Surveys). See Multimedia Appendix 1 for the complete checklist.

To the best of our knowledge, no validated measure currently exists to assess oncology provider attitudes toward mHealth apps. We developed the 30-item OPENS to describe mHealth-related usage, attitudes, perceived benefit, facilitators, and barriers. OPENS was designed as a descriptive survey to assess provider perspectives and usage patterns rather than as a psychometric scale intended to measure a single latent construct. These survey items directly address our study aims to inform the clinical integration of mHealth technologies in oncology care. Internally developed surveys are a standard approach in digital health research when established measures are unavailable [51-54]. Survey items were collaboratively developed by a multidisciplinary research team (including experts in psychology, psychiatry, social work, oncology, and digital health) following a targeted literature review. We established content validity using a modified Delphi approach [55-58]. This process involved asynchronous feedback in which collaborators independently reviewed survey items for clarity and relevance to both the study aims and the oncology clinical context. Facilitated group consensus meetings were held with external oncology experts to refine survey items. Eight items assessing mHealth attitudes and beliefs were adapted from the “Questionnaire to Assess Public Health Staff Viewpoints about Mobile Phone Use in the Delivery of Their Services” [59]. These items were updated for oncology relevance. Survey items were finalized asynchronously among the research team members.

Before national dissemination, the survey underwent pilot-testing with a small group of oncology providers to assess technical functionality within REDCap, item comprehension, and user burden. All 33 items were distributed across 4 pages and administered in a forced-response format to ensure complete data capture. Questions included single-choice items, multiple-choice items, “5-point Likert scale” items (eg, strongly disagree to strongly agree), and “select all that apply” items (including “none of the above” and “other” options). Respondents were able to review and modify their responses prior to survey submission, with no time limits or automatic expiration imposed for completion. Consistent with the scoring method of the validated instrument from which eight survey items were adapted [48], Likert-scale responses were summarized using means (SDs) to characterize the strength and variability of providers’ beliefs and attitudes on mHealth apps. The brief close-ended survey was designed to reduce participant burden and maximize response rates for uncompensated national listserv recruitment methods, targeting a completion time of approximately 10 minutes. To ensure that our study met IRB-exempt criteria, we did not include open-ended survey questions in which human subjects’ identifying information may have inadvertently been obtained. See Multimedia Appendix 2 for all survey items.

We summarized provider demographics and other survey items using descriptive statistics and data visualizations created in Canva. Exploratory statistical analyses were performed using IBM SPSS Statistics (version 29). Study results are presented as frequency counts, percentages, and means (SDs). We also conducted exploratory analyses on survey responses broadly categorized by provider type (medical providers vs psychosocial providers) who may differ in their attitudes toward and the use of mHealth tools for disease management and behavioral/emotional support based on their training backgrounds and areas of expertise. For the purposes of this study, medical providers were defined as those whose primary training and clinical focus address disease treatment and the management of physiological symptoms. This group included those who identified their role as oncologist, nurse or advanced practice provider, physician, or health care administrator. Psychosocial providers were defined as clinicians whose primary training and clinical focus address the psychological, social, and emotional aspects of the cancer experience. This group included those who identified their role as psychologist, social worker, mental health therapist, or occupational therapist. This categorization also allowed for a balanced distribution of providers for comparison (90/188, 47.9%, medical vs 89/188, 47.3%, psychosocial). We performed chi-square analyses to examine differences between medical providers and psychosocial providers on the use of mHealth apps in clinical practice, mHealth apps endorsed as potentially helpful or beneficial, and the perceived facilitators of and barriers to mHealth app adoption. We also performed chi-square analyses to examine differences between younger providers (age≤45 years) vs older providers (age≥46 years) and pediatric care providers (children, teens, or young adults) vs adult care providers (adults, older adults) on the use of mHealth apps in clinical practice. Given that the study used categorical data, traditional assumptions of normality were not applicable. As all chi-square analyses were exploratory and intended for hypothesis generation rather than formal hypothesis testing, we did not control for multiple comparisons.

A total of 188 providers responded to the survey (Table 1). Participants ranged in age from under 24 to over 65 years. Over one-third (70/188, 37.2%) were between the ages of 36 and 45 years. Most providers identified as female (150/188, 79.8%), White (161/188, 85.6%), and not Hispanic or Latino (174/188, 92.6%).

Of the 188 respondents, 47.9% (90/188) were medical providers, 47.3% (89/188) were psychosocial providers, and 4.8% (9/188) had administrative staff or research roles. Psychologists represented the largest group of survey respondents (66/188, 35.1%), followed by oncologists (54/188, 28.7%) and nurses or advanced practice providers (32/188, 17%), as shown in Table 2. Most providers practiced in academic or teaching hospitals (155/188, 82.4%) and urban settings (130/188, 69.1%). Over half of the sample provided care for children (120/188, 63.8%), teens (124/188, 66%), and young adults (143/188, 76.1%).

Nearly all providers (178/188, 94.7%) reported recommending or using at least one mHealth app with their patients (Figure 1). Most providers recommended or used telehealth and virtual consultation apps (139/188, 73.9%). Other types of mHealth apps that were commonly recommended to or used with patients included communication with the medical team (136/188, 72.3%) and mental health and wellness (98/188, 52.1%) apps. Only a small proportion of providers reported recommending or using cancer and disease management (21/188, 11.2%) or health literacy (15/188, 8%) mHealth apps with their patients. Both medical and psychosocial providers predominantly used mHealth apps for medical team communication or telehealth (Figure 2 and Figure 3, respectively). Psychosocial providers (25/89, 28.1%) were significantly more likely than medical providers (6/90, 6.7%) to recommend or use pain management apps (χ²₁=14.34, P<.001, φ=0.28). Psychosocial providers (70/89, 78.7%) were significantly more likely than medical providers (23/90, 25.6%) to recommend or use mental health apps (χ²₁=50.54, P<.001, φ=0.53). Psychosocial providers (48/89, 53.9%) were significantly more likely than medical providers (16/90, 17.8%) to recommend or use sleep health apps with their patients (χ²₁=25.47, P<.001, φ=0.38). All other chi-square analyses comparing medical and psychosocial provider groups were nonsignificant.

Providers stratified by age (≤45 and ≥46 years) recommended or used similar types of mHealth apps with patients (Multimedia Appendix 3). Older providers (16/88, 18.2%) were significantly more likely than younger providers (5/100, 5%) to recommend or use disease management apps with their patients (χ²₁=8.20, P=.005, φ=0.21). All other chi-square analyses comparing older and younger providers were nonsignificant. Providers working with pediatric and adult patients also recommended or used similar types of mHealth apps (Multimedia Appendix 4). Adult care providers (8/32, 25%) were significantly more likely than pediatric care providers (6/101, 5.9%) to recommend or use disease management apps (χ²₂=9.22, P=.01, Cramer V=0.22). Pediatric care providers (83/101, 82.2%) were more likely than adult care providers (15/32, 46.9%) to recommend or use medical team communication apps (χ²₂=15.66, P<.001, Cramer V=0.29). All other chi-square analyses comparing adult and pediatric care providers were nonsignificant.

All providers endorsed mHealth apps to be potentially useful or beneficial for patients (Figure 1). Most providers perceived mental health and wellness apps to be potentially useful (153/188, 81.4%). Other types of mHealth apps that were perceived as potentially useful were telehealth and virtual consultation (151/188, 80.3%), communication with the medical team (150/188, 79.8%), exercise and physical activity (136/188, 72.3%), social support (134/188, 71.3%), health monitoring and medical symptom tracking (133/188, 70.7%), medication management (131/188, 69.7%), and sleep health (130/188, 69.1%). Although providers predominantly used mHealth apps for patient communication (eg, medical team communication, telehealth), they endorsed interest in prescribing mHealth apps for a broad range of holistic care functions.

Interest in a wide array of psychosocial support functions was fairly consistent across medical and psychosocial providers (Figure 2 and Figure 3, respectively). Medical providers (73/90, 81.1%) were significantly more likely than psychosocial providers (52/89, 58.4%) to perceive medication management apps as potentially beneficial for their patients (χ²₁=10.93, P<.001, φ=0.25). Medical providers (31/90, 34.4%) were significantly more likely than psychosocial providers (17/89, 19.1%) to perceive energy therapy apps as potentially beneficial (χ²₁=5.37, P=.02, φ=0.17). In contrast, psychosocial providers (69/89, 77.5%) were significantly more likely than medical providers (54/90, 60%) to endorse sleep health apps as potentially helpful for their patients (χ²₁=6.40, P=.01, φ=0.19). All other chi-square analyses comparing medical and psychosocial provider groups were nonsignificant.

Providers had positive attitudes and beliefs toward specific facets of mHealth apps (Table 3). Providers had the highest mean scores for specific items pertaining to health equity, indicating that they “agree” or “strongly agree” that mHealth apps reduce travel costs for services (mean 4.18/5, SD 0.68) and expands the range of health services (mean 4.18/5, SD 0.69). Providers had the highest mean scores for specific items pertaining to confidence and a growth mindset, indicating that they “agree” or “strongly agree” that they are confident in their ability to navigate mobile apps (mean 4.14/5, SD 0.83) and can adapt to health technology changes over time (mean 4.18/5, SD 0.70). Providers had the lowest mean scores for specific items pertaining to the patient-provider relationship and training, indicating that they “neither agree nor disagree” that mHealth improves provider-patient interactions (mean 3.21/5, SD 0.95) or are trained and informed about mHealth regulatory and ethical issues in health care (mean 3.24/5, SD 1.07).

The top five perceived facilitators of mHealth adoption in clinical settings were “alignment with patients’ expressed interest and needs” (147/188, 78.2%), “improved health care accessibility” (135/188, 71.8%), “cost-effectiveness for patients and health care systems” (127/188, 67.6%), “seamless integration with existing health care systems and/or electronic medical records” (113/188, 60.1%), and “improved job performance and efficiency” (109/188, 58%); see Multimedia Appendix 5. Medical and psychosocial providers reported similar top facilitators overall (Figure 4). However, medical providers (47/90, 52.2%) were significantly more likely than psychosocial providers (23/89, 25.8%) to endorse patient health literacy as a perceived facilitator (χ²₁=13.08, P<.001, φ=0.27).

The top five perceived barriers to mHealth adoption in clinical settings were “disparities in access to technology” (132/188, 70.2%), “digital literacy” (ie, technical knowledge to access, understand, and use digital resources and tools; 117/188, 62.2%), “data privacy and safety concerns” (114/188, 60.6%), “lack of knowledge regarding which digital tools to recommend or where to find relevant information” (110/188, 58.5%), and “being overwhelmed by too many digital health technology options” (105/188, 55.9%); see Multimedia Appendix 6. Medical and psychosocial providers reported similar top barriers overall (Figure 5). However, medical providers (52/90, 57.8%) were more likely than psychosocial providers (38/89, 42.7%) to endorse cost as a perceived barrier (χ²₁=4.07, P=.04, φ=0.15).

This study provided an update to and extension of prior research regarding the integration of mHealth apps in oncology care. Regarding mHealth app usage and its perceived benefit (RQ1), oncology providers commonly use mHealth apps, with an emphasis on patient-provider communication tools. Although current usage is primarily logistical, there is clear interest in whole person health–supporting apps that providers consider potentially beneficial but that currently see lower rates of adoption. With regard to provider attitudes, facilitators, and barriers related to mHealth adoption (RQ2), providers consistently endorsed a growth mindset, reporting high confidence in their ability to learn and adopt digital tools. Providers recognized the potential for mHealth to improve health equity by expanding access to care. The top perceived facilitators of mHealth adoption were alignment with patient needs, improved health care accessibility, and cost-effectiveness. The top perceived barriers to mHealth adoption were disparities in access to technology, digital health literacy, and data privacy.

With regard to exploratory differences by clinical role, provider age, and patient age (RQ3), we found several group differences. Reflecting their respective training backgrounds, medical providers were more likely to endorse the potential benefit of medication management apps, whereas psychosocial providers were more likely to use and/or perceive potential benefit in pain management, mental health, and sleep health apps. The use of mHealth apps was generally consistent across oncology provider age groups (≤45 and ≥46 years) and patient population (pediatric care providers and adult care providers). However, older providers (age≥46 years) and those providing care to adult patients used more disease management tools, whereas pediatric providers used more medical team communication apps.

Our study expands on the background literature on oncology providers’ attitudes toward mHealth apps by soliciting perspectives of those in various clinical roles, not exclusively oncologists; expanding mHealth app functions beyond disease management; and identifying perceived barriers to and facilitators of implementation in oncology care settings. Our findings are consistent with previous studies that have reported that oncologists primarily use mHealth tools for telehealth and medical team communication and rarely use mHealth tools for disease management and follow-up care [40]. Of note, we also found that medical and psychosocial providers alike are interested in using mHealth apps focusing on a wide range of psychosocial care functions to their patients and perceived them to have clinical utility. This was consistent with the literature on patients with cancer and those who have survived cancer, showing high rates of acceptance of mHealth apps but low rates of adoption [60-64]. Specific patient facilitators and barriers beyond demographic and disease characteristics have not yet been examined. Studies examining patient preferences have predominantly focused on disease management apps [60,62-66]. Consistent with the background literature, we found that a primary barrier identified by providers is disparities in access to technology [21,40,67]. However, research has shown that mHealth technologies are universally accessible to individuals regardless of income level, rurality, or marginalized status [3,4,68]. In the United States, 79%-94% of individuals from low-income households (<US $30,000) have access to smartphones, computers, and tablets [3,4,69]. Digitally deployed health interventions have also been proposed for populations/persons experiencing homelessness as viable alternatives to in-person services to overcome traditional access care barriers [68,70,71]. In the United States, 91% of individuals own a smartphone, 95% use the internet, and 90% have a high-speed internet subscription at home [1,3,4,69]. Universal use of smartphones holds across the age continuum, with 95% of teens/young adults, 95%-97% of adults aged>45 years [2-4,72], and 76%-79% of older adults aged>65 years owning a smartphone [4,69,72]. Digital technologies (smartphones, computers, tablets) have become significantly more affordable in recent years, and their rapid adoption has been further accelerated by the COVID-19 pandemic.

Consistent with the background literature, a primary barrier to mHealth adoption, as identified by providers, was limited digital health literacy [61]. Previous studies have shown that health literacy and digital health literacy are significant mediators of the efficacy of digital health interventions [73] and are not fixed but malleable patient characteristics [74]. Thus, providers may be more likely to recommend and use mHealth apps with intuitive user interfaces, accessibility for those with impairments, provision of health information at lower reading levels, and multimedia options (eg, text-to-speech and animations) [61,75,76]. Recommendations for mHealth may also be paired with apps that identify a user’s health literacy level and provide tailored definitions of medical terminologies to their reading level (eg, MediReader [77]).

Previous research has found that younger oncologists (<45 years old) and younger cancer patients and survivors (<45 years old) are more likely to use internet-enabled mHealth tools for disease management and cancer care [40]. Our results showed consistent use of mHealth apps in younger as well as in older providers (>45 years old) and among providers of patients across the age continuum. For younger pediatric patients, caregivers may primarily interact with mHealth apps, while teens and young adults may be more independent app users [78]. Digital inclusion for older adults is recognized as an important social determinant of health and a crucial component of a healthy aging society [79-88]. These findings highlight the importance of developing, testing, and implementing mHealth apps that are tailored to the diverse needs of patients with cancer and those who have survived cancer, of all ages.

The Technology Acceptance Model (TAM) proposes that perceived usefulness (ie, the belief that using a particular technology enhances job performance) and perceived ease of use (ie, the belief that using a particular technology is effort free) are the primary drivers of user adoption of new technologies [89,90]. Interpreted through the TAM, our findings suggest high readiness for mHealth adoption across the oncology workforce. Specifically, providers’ endorsement of mHealth apps for reducing costs and expanding care access aligns with high perceived usefulness. Providers’ reported growth mindset and confidence in their ability to learn and adapt to mHealth technologies reflect high self-efficacy, which aligns with high perceived ease of use. Although TAM pathways were not formally tested in our study, the alignment between our findings and these TAM constructs suggest that oncology providers have the positive attitudes and behavioral intentions necessary to facilitate mHealth adoption.

mHealth apps that address holistic patient care have the potential to improve patient outcomes across the cancer care continuum and patient age continuum. Despite the narrow use of mHealth apps in current practice, the perceived utility of mHealth across medical and psychosocial domains suggests important opportunities for broader integration. Guided by the RE-AIM (Reach, Effectiveness, Adoption, Implementation, and Maintenance) framework [91-93], future research should focus on expanding reach by partnering with medical and psychosocial providers on codesign strategies of mHealth tools in the innovation design-to-implementation pipeline. Future research should also enhance effectiveness via rigorous clinical testing and addressing the complex ethical and regulatory landscape of mHealth apps. Although oncology providers are interested in using mHealth tools, the lack of standardized clinical guidelines for recommending non-FDA-approved apps remains an important barrier. Endorsing or recommending mHealth clinical tools that lack rigorous scientific testing poses risks to patient safety and data confidentiality. Future research and policy work should focus on developing clinical guidelines to identify evidence-based mHealth apps and pathways for determining their clinical efficacy/effectiveness. In addition, future research should support adoption and implementation by examining providers’ nuanced perspectives on the perceived barrier of technology access and investigate the specific sources of these concerns (eg, financial burden of cancer, lack of reliable cellular or internet service in rural areas). Efforts to mitigate barriers, such as concerns about data privacy and security, will also be critical to ensuring adoption of mHealth solutions. Successful integration of mHealth apps into oncology care will require accessible, user-centered design (eg, intuitive interfaces, appropriate reading levels, multimedia supports) and shared decision-making informed by ongoing examination of concordant and discordant beliefs and attitudes toward mHealth apps among providers and patients. Long-term maintenance will depend on sustained institutional and policy support (reimbursement pathways, standardized evaluation criteria, regulatory guidance on data privacy and security), engagement of local providers or stakeholder champions to promote uptake through education and shared learning in navigating digital health systems, and preservation of alternative care options (eg, in-person services) as technologies continues to evolve. Future research using a mixed methods approach is also warranted to provide richer context to these exploratory findings.

This study has several limitations. First, this is a convenience sample of oncology providers, which introduces the potential for selection and response bias due to voluntary participation. It is possible that providers with a pre-existing interest in mHealth were more motivated to participate in the study, potentially leading to nonresponse bias where favorable attitudes are overrepresented. In addition, social desirability bias may have influenced participants to overreport favorable attitudes to align with current digital health trends. To mitigate potential bias, we distributed the survey through multiple national professional listservs representing diverse professional roles. In addition, survey questions assessed diverse app functions, barriers, and facilitators, which resulted in variability in study findings. This suggests that the survey captured an array of provider perspectives. Second, most respondents identified as White and female, practiced in urban settings, and did not see older adults in their clinical practice. This limits the generalizability and transferability of our findings to underrepresented providers because of a lack of diversity in our sample. Future studies should implement recruitment strategies to target provider demographic diversity, rural health settings, and those in older adult and geriatric care to ensure findings are representative of all clinical contexts. Third, this is a small sample relative to the number of oncology providers who practice nationally, and we relied on our institutional and national professional organization listservs to distribute the survey. Sample size was determined by convenience sampling within the study recruitment time frame. As the primary aim of this study was descriptive, we did not establish a predetermined target sample size. Fourth, as there are no well-validated measures of provider attitudes toward mHealth interventions with established cutoff scores, we used an internally developed survey. This influenced the study by limiting our ability to categorize and compare scores with existing studies. To mitigate the lack of a pre-existing validated scale, we used a modified Delphi process involving a multidisciplinary team and external oncology experts to ensure content validity. Fifth, to comply with IRB-approved protocols and IRB-exempt status, we did not collect identifiers from survey respondents, so it is not possible to determine response rates from specific organizations or listservs or analyze group differences based on affiliations. Future research could include such identifiers in order to assess representativeness and potential cohort effects. Sixth, due to the exploratory nature of our analyses, we did not control for multiple comparisons or include demographic covariates. Although this approach increases the risk of type 1 errors, we mitigated the risk of overinterpretation by explicitly framing these analyses as hypothesis generating, and our findings should be viewed as preliminary. Finally, this was a cross-sectional study who findings provide a snapshot in time of provider attitudes in a post-COVID-19 pandemic landscape that has normalized digital health care. Because attitudes and perspectives may continue to shift over time and technology evolves rapidly, future longitudinal research is warranted.

The findings of this study reveal a significant “readiness gap” in oncology care. Although current use remains largely focused on telehealth services and patient-provider communication, providers expressed interest in integrating a wider array of mHealth apps encompassing health information, disease and symptom monitoring, mental health resources, and psychosocial support. Our findings suggest that the oncology workforce is prepared to move beyond logistical communication tools toward an integrated holistic model of oncology care that encompasses broad disease management and psychosocial support functions. Because these positive attitudes are consistent across clinical roles and patient populations, mHealth interventions have broad appeal and should be implemented as interdisciplinary standards of care. Health care systems can leverage existing provider buy-in to create a more equitable and accessible oncology care model.