We designed a cross-sectional web-based survey to examine the preferences of adults in Israel for adopting HMO-mHealth apps and receiving health care services versus HMO-mHealth as a substitute for face-to-face interaction. Data for this study were collected from December 2020 to February 2021. Respondents were not required to provide any identifying information, and any participant could stop the survey at any point.

The questionnaire was designed on the Qualtrics platform and distributed by research assistants via social media (WhatsApp groups, Facebook, etc). Our target population was Israeli adults aged ≥18 years, who were sampled using convenience sampling. Responders were allowed to fill in the questionnaire electronically only once. To estimate the required sample size, we relied on data from the Central Bureau of Statistics in Israel. The population of the State of Israel in January 2024 was estimated to be around 9.855 million [33]. Since the proportion of mHealth use in the population is unknown and is assumed to be a maximum of 50%, with a significance level of 5%, a sample size of 384 residents was calculated for this study. In practice, there was a high response rate, and 6321 Israeli adults completed the research questionnaire. It was decided to use all the data to obtain greater validity.

All services available in the main HMO-mHealth apps of the 4 Israeli HMOs were mapped. A total of 29 health care services, which were identical across all 4 HMO-mHealth apps in Israel, were selected for this study and included in the HMO-mHealth/face-to-face interaction preference questionnaire. These services included administrative matters such as setting a visit or changing a visit date as well as medical services such as receiving a medical diagnosis or treatment. For each of the 29 items, participants were asked whether they preferred mHealth or face-to-face interaction to receive the health service. Each item was tested as a separate outcome. In addition, the following information was self-reported by the participants: gender, age (years), marital status (married/single/divorced/widowed), number of children, birthplace (Israel/other), religion (Jewish/non-Jewish), religiosity (secular/religious), education (uneducated/elementary-middle school/high school/vocational training/BA/MA/PhD), residence (Center, North, or South Jerusalem), subjective health condition (from 1: worst to 5: very good), and previous diagnosis of chronic illness by a licensed physician (yes/no).

The main characteristics of the study population were described using percentages for categorical variables and means with standard deviations for continuous variables. Chi-square tests or independent-sample 2-sided t tests were employed to evaluate the association of categorical or continuous predictors, respectively, with each of the 29 health services included in this study. Variables that were significant (P<.05) in the univariable tests for at least one of the health care services were included in the final multivariable logistic models. The association between significant personal characteristics with the preference of mHealth or face-to-face interaction was assessed separately for each health care service by using multivariable logistic regression analysis. In each model, a particular health service was evaluated as the dependent variable (0=face-to-face; 1=mHealth). The predictor variables were entered simultaneously into each model and included sociodemographic characteristics (gender, age, marital status, birthplace, number of children, education, religion, religiosity, residence), subjective health condition, and previous diagnosis of chronic illness. To mitigate the potential for spurious statistical significance arising from multiple tests, Benjamini-Hochberg adjusted P values were computed using WinPepi version 11.65 [34]. P values less than .05 were considered statistically significant.

This study was approved by the ethics committee of Ariel University (AU-HEA-AZ-20201217). All participants were informed of the aims of the study in an introductory section, and they gave their informed consent to participate in the study. No conditioning questions were asked to conduct the survey, as health care services are provided to all residents of Israel by 4 health care providers to which the residents are registered.

Adoption of mHealth by patients is a complex and multidimensional process with its own advantages, disadvantages, and barriers. This study was the first to examine the extent to which adults in Israel prefer to receive health care services through HMO-mHealth as a substitute for face-to-face interaction and to identify factors associated with these preferences. Approximately 46.9% (2969/6321) of our study participants preferred face-to-face interaction for the initial treatment or diagnosis rather than through HMO-mHealth apps (represented by the items, namely, personalized medical diagnosis, physical examination by a physician, and receiving medical diagnosis results). However, 80.9% (5115/6321) to 88.2% (5578/6321) of our study participants were interested in receiving administrative services through HMO-mHealth apps (demonstrated in items such as changing visit details, setting a family physician visit, and prescription renewal request). Nevertheless, 11.7% (743/6321) to 19.1% (1206/6321) of the participants were still interested in receiving administrative and information services through face-to-face meetings. Our axiom is that the respondents have adequate digital literacy since the questionnaire was distributed and filled out using a digital tool. However, good digital literacy is not necessarily a guarantee of the ability to use digital health services [1], and the validity of information in mHealth apps that focus on emotional content may influence the adoption of HMO-mHealth services, even among those with high digital literacy [6,35]. The high usage rate of HMO-mHealth in this study can be explained by the Unified Theory of Acceptance and Use of Technology (UTAUT), which states that effort expectancy (defined as the degree of ease of use) affects the acceptance and use of HMO-mHealth [36]. Seven main variables were found to be associated with HMO-mHealth adoption, including gender, age, education, marital status, religious affiliation, and subjective health condition.

Our results show that women are more likely to adopt HMO-mHealth apps for medical information or administrative services (eg, prescription renewal request, setting a family physician visit, account details inquiry). However, they prefer face-to-face interactions with physicians for receiving medical services with treatment or diagnostic significance (eg, personal medical diagnosis, medical examination, receiving the results of medical diagnoses, information on treatment options). This result is interesting because one might assume that gender has a similar effect on the adoption of all HMO-mHealth services and that there are no differences between the different types of services. Previous studies suggest that there are generally no clear trends in gender differences in the willingness to adopt mHealth technologies [3,14,25,27,36-38]; gender is mentioned as a moderating variable in mHealth studies dealing with UTAUT, and it plays a moderating role with threat appraisal and coping appraisal factors in mHealth adoption behavior [39,40]. However, traditional face-to-face health care is frequently preferred over digital health care [1], and women, in particular, tend to be more skeptical about relying solely on mHealth for treatment and place higher value on in-person doctor visits [41]. This inclination is likely influenced by greater health anxiety and risk aversion among women [42-45].

Consistent with previous studies [11,14,27,37], our study also shows that age, education, religious affiliation, and ethnicity were barriers to HMO-mHealth adoption. On the one hand, people with a higher level of education and younger individuals had a higher tendency to adopt HMO-mHealth services compared to older individuals and those with a lower level of education who preferred face-to-face meetings. On the other hand, societal minorities (non-Jews) and religious individuals tended to prefer face-to-face meetings over the adoption of HMO-mHealth services. It is possible that internet use is perceived as more accessible among young people, and those with higher education are perceived as having better digital literacy [4,46]. However, our study shows that people with a higher level of education and younger individuals who have a poor subjective health condition generally prefer face-to-face meetings. Based on this finding, it can be assumed that despite the effectiveness of HMO-mHealth services, young and educated people still find it necessary to have human contact with a health provider for diagnosis and treatment.

This study also shows that ethnic minorities (non-Jewish) and religious people would prefer face-to-face meetings over HMO-mHealth for all health care provider services examined in this study. There is evidence of the effectiveness of internet-based interventions that improve the health of ethnic minorities [47]. These ethnic minority groups reported greater interest in using mHealth technologies than the nonminority population [48]. A recent study found that there are significant ethnic differences in the use of telemedicine between different ethnic groups [27]. However, the association of religion with the adoption of health technologies has, to our knowledge, not yet been sufficiently investigated. The adoption of technologies by religious people is considered at higher-than-average risk, and previous studies have shown that religious leaders can influence the opinions and behaviors of religious people when adopting medical technologies [49]. We link these 2 facts because we believe that barriers to HMO-mHealth adoption can be associated with ethnicity and religion. Both cultural and community characteristics, language barriers, emotional state, and health and digital literacy can be assumed to play an important role in HMO-mHealth adoption [46,50]. A study examining the influence of cultural aspects on the adoption of mHealth in 3 different countries, based on UTAUT, found that cultural differences have a decisive effect on the adoption behavior model and concluded that mHealth services must be adapted to the culture of the population for which the services are developed [51].

Greater involvement of minorities in the development of HMO-mHealth will lead to the adoption of these tools by minorities [52], and the involvement of leaders from these communities in the development of technological medical products will enable broader adoption of HMO-mHealth in these populations. Another interesting finding of this study is that married people would prefer to adopt HMO-mHealth over face-to-face meetings. A systematic literature review [53] and narrative synthesis showed that the effects of marital status on mHealth uptake are inconsistent across studies. Some studies suggest that marital status has no significant effect on the use of mHealth services, while others propose possible differences in the intention to use depending on marital status [53]. In our opinion, addressing family health needs and marital responsibilities can increase interest in technologies that improve health monitoring and save time. Since there is a gender dynamic in most marital relationships regarding domestic responsibilities and health care decisions, this is likely to have an impact on the adoption of HMO-mHealth.

This study raises an interesting question: for whom are HMO-mHealth services really intended for? In this study, it was found that patients with chronic diseases do not prefer to use HMO-mHealth services. Conversely, people who describe their subjective health as good are more likely to use HMO-mHealth services. It is generally assumed that patients with chronic diseases will adopt more mHealth services [54], as mHealth is seen as part of medical practice and a factor that can support patients with chronic diseases [37]. There is evidence in the literature that subjective health condition is a factor that influences the use of digital means to access health care services [55]. However, studies show that the presence of a chronic disease does not predict demand for medical services [54,56-58]. It can be assumed that the support provided by the family of a patient with chronic diseases will reduce the need for HMO-mHealth services [58].

Our study has several limitations. First, due to the cross-sectional nature of our study, it is not possible to establish causation or demonstrate cause-and-effect relationships. Second, the data were self-reported. The researchers were not able to assess whether the questionnaire was completed during work or leisure time or whether health limitations influenced the completion of the questionnaire. Third, the services in the HMO-mHealth apps are similar, but the interaction between the patient and the health care provider in each app may be different. It is possible that a better response from the health care provider for certain services in the HMO-mHealth apps could influence the information collected in the questionnaire. Fourth, health literacy was not examined in this study. It is likely that the level of health literacy may also influence patients’ interactions with HMO-mHealth services.

HMO-mHealth proves to be a robust and efficient tool for health care service delivery when compared to face-to-face health care interaction. However, barriers that affect vulnerable populations in HMO-mHealth adoption still exist. HMO-mHealth services will not be able to completely replace face-to-face interactions with health care providers and will be a complementary tool for face-to-face meetings with therapists. The utilization of health applications within health services constitutes an increasingly substantial component of the communication interface between the health care system and its patients, accompanied by a rise in the array of services offered through HMO-mHealth. On the one hand, this study’s outcomes furnish insights for policy makers engaged in the development of HMO-mHealth services, facilitating the formulation of culturally sensitive HMO-mHealth services. On the other hand, there exists a compelling necessity for policy makers to institute a comprehensive training regimen aimed at equipping patients with the requisite skills for utilizing HMO-mHealth. This is particularly imperative for aiding demographic segments such as the older adults, chronically ill patients, and societal minorities, who are in greater need of these services. Health care providers need to develop intervention plans for accessibility, adaptation, and implementation of HMO-mHealth in this population. Such adaptations are essential for mitigating health care inequalities in Western societies. To provide an optimal response to these populations, future studies should focus on identifying the barriers that affect the utilization of HMO-mHealth in these groups. Future studies should also use longitudinal designs to better establish cause-and-effect relationships between variables; assess health literacy, which is a potential factor in the adoption of HMO-mHealth; and include comparisons with mHealth adoption in other countries. These future studies could significantly enhance our understanding of the cultural and systemic variations.