This cross-sectional study was conducted via a web-based questionnaire survey administered across Japan by Cross Marketing Inc, an internet research firm. At the beginning of the questionnaire, participants were clearly informed that their personal data would be strictly protected and anonymized. Furthermore, they were explicitly instructed to respond to questions about their attitudes and intentions toward data sharing on the premise that any shared data would exclude personally identifiable data and privacy would be guaranteed. This explanation was included to ensure participants understood that the protection of personal data was a fundamental prerequisite for data sharing scenarios described in the questionnaire.

The survey was developed in accordance with the CHERRIES (Checklist for Reporting Results of Internet E-Surveys) checklist (Figure 1, Checklist 1). A pilot test was conducted between October 28 and November 6, 2023, to assess the questionnaire’s clarity, readability, interface usability, and time burden. The final survey was distributed from November 11 to 18, 2023.

A minimum sample size of 385 was calculated using a normal approximation of the binomial distribution, considering Japan’s population of 126 million, a 95% confidence level, and a 5% margin of error [44]. Based on previous studies, 1000 questionnaires were collected to improve the validity of the study [36,38,40].

Participants were recruited from Cross Marketing Inc.’s panel and were aged 18 years or older. To ensure demographic representativeness, stratified random sampling was used based on age and gender. Participants were equally allocated into seven age categories (18‐27, 28‐37, 38‐47, 48‐57, 58‐67, 68‐77, and 78 years or older) and by gender, resulting in equal numbers in each age category and gender group. Within each stratum, participants were randomly selected.

Exclusion criteria included discrepancies between registered and self-reported gender or age, extremely short response times, failure to follow instructions, or provision of inconsistent or implausible responses. Out of 23,434 distributed questionnaires, 2203 participants provided informed consent, and after applying exclusion criteria, 1000 valid responses were analyzed. Participant demographics are presented in the Results section (Table 1).

The questionnaire was developed with reference to previous studies on willingness to share health-related data, including biobank and electronic medical record data, and attitudes toward data repositories. Five factors were investigated as predictors of participants’ attitudes toward sharing digital health data for social benefit: (1) individual sociodemographic characteristics, (2) types of health data shared, (3) motivation for sharing data, (4) concerns factors of data sharing, and (5) reasonable access and control over the data (Figure 2) [36-38,45,45]. A pilot study involving 5 academic staff and researchers helped refine the instrument. The selection and formulation of questionnaire items were based on previous studies identifying factors influencing data sharing behavior.

The questionnaire included the following domains, all using uniform response options (“Yes,” “No,” “Don’t know,” and “Other” with free-text input). Questions were randomized, and multiple responses were allowed.

Full wording of all items and response options is available in Multimedia Appendix 1 questionnaire.

Participants indicated their willingness to share various types of digital health data, including physical and mental clinical diagnoses, family health history, genetic and blood test data, dietary and nutritional data, alcohol consumption, sleep status, social communication (frequency and content), physical activity, stress, and geolocation data.

Participants selected motivations such as helping future patients, supporting research, receiving study results, financial incentives, and proposing research topics.

Participants selected concerns from a list including privacy risks, data misuse, unauthorized access, and ambiguity in consent terms. The wording related to access restrictions was revised for clarity. Participants chose from options indicating who may access their data (eg, medical professionals, researchers, and family members) and preferences on data control.

Participants reported preferences on controlling access to their data, transparency in data usage, trust in data controllers, and openness to sharing with public or academic institutions.

Data on age, gender, educational attainment, annual household income, marital status, and ownership of digital devices (smartphones, PCs, tablets, smartwatches, and smart home assistants or artificial intelligence [AI] appliances) were collected using predefined categories. Participants also reported previous participation in health research (epidemiological or clinical trials).

eHealth literacy was assessed using the Norman eHealth Literacy Scale (eHEALS), an 8-item, 5-point Likert scale measuring knowledge, comfort, and perceived skills in finding, evaluating, and applying eHealth data [45,46].

Self-rated health status was assessed on a 5-point Likert scale, and responses were grouped into “good” (somewhat good, very good) or “not-so-good” for analysis. Logistic regression analyzed the relationship between health status and willingness to share data.

Participants’ concern about their own and family’s health was measured on a 6-point scale, dichotomized into “interested” or “not interested,” with logistic regression used to assess associations with willingness to share digital health data.

Descriptive statistics (frequencies, means, and SDs) summarized participant characteristics. We used the intention to share digital health data as the outcome variable. Logistic regression analysis was performed to examine its associations with a range of predictor variables, including motivations for sharing data (eg, helping patients and supporting research), concerns about data sharing (eg, privacy risks and misuse), attitudes toward data access (eg, preferences for control and trust in institutions), and sociodemographic and health-related factors (eg, age, gender, education, income, marital status, digital device ownership, and previous participation in health research). Analyses were conducted using SPSS version 28 (IBM Corp), with significance set at P<.05.

This study fully complied with the Declaration of Helsinki and Japan’s Personal Data Protection Law [47]. Ethical approval was granted by the Waseda University Ethics Committee (application 2023‐250), confirming that all protocols and consent procedures were reviewed and sanctioned before data collection.. No personally identifiable information was collected in accordance with both institutional and legal privacy requirements. All persons described are not identifiable and cannot be identified through the details of the story because this research is an anonymous Internet survey.

Informed consent was obtained electronically before participation. Participants received a small monetary incentive (points equivalent to approximately 10 Japanese yen [US$1=147.1 JPY]) via the survey provider’s point system. Low-quality or inconsistent responses were screened and excluded by the research firm.

Table 1 shows participant characteristics. The men-to-women participant ratio is 1:1. The mean age of the participants and the SD is 52.8 (SD 19.8) years. Approximately half of the participants had a university or postgraduate degree in education. In total, 93% of the respondents had a smartphone and 83% had a personal computer. About 80% of the respondents were interested in their health (very interested and somewhat interested).

Approximately 70% of respondents were willing to share digital health data. Respondents who were willing to share digital health data were defined as having a positive willingness to share, and subsequent analyses were conducted.

The results of the logistic analysis of the association between positive attitudes toward sharing digital health data and sociodemographic characteristics are shown in Table 2. Compared with men, women were less willing to share data (odds Ratio [OR] 0.722, 95% CI 0.539‐0.967; P<.05).

Those who had more experience participating in research (OR 3.308, 95% CI 1.391‐7.868; P<.01), and had awareness of research data repositories (OR 2.711, 95% CI 1.364‐5.388; P<.01) were more willing to share digital health data. Moreover, owning a smartphone (OR 1.841, 95% CI 1.078‐3.145; P<.05), personal computer (OR 1.658, 95% CI 1.175‐2.341; P<.05), tablet (OR 1.688, 95% CI 1.212‐2.349; P<.01), or smart home assistance appliances (OR 3.384, 95% CI 1.077‐5.276; P<.05) was positively associated with higher willingness to share digital health data. Conversely, there was no association between annual income or education and the willingness to share digital health data. Logistic analysis of e-health literacy and digital health data sharing suggested that the higher the e-health literacy, the more positive the willingness to share digital health data.

Willingness to share digital health data was significantly associated with a high level of health concern and willingness to share digital health data (OR 2.1871, 95% CI 1.5750‐3.03710; P<.001) but not associated with health status (P=.73).

The percentage of respondents who indicated a willingness to share shareable data (multiple responses) is shown in Multimedia Appendix 2 Figure S2. The results show that 44%, 33%, 33%, 32%, and 31% of participants were highly motivated to share data regarding physical clinical diagnoses, sleep patterns, blood samples, stress, and food consumption, respectively. Moreover, the content and frequency of social communication tended to be low (11% and 14%, respectively).

Multimedia Appendix 3 Figure S3 shows the percentage of respondents who indicated each item as a motivation for sharing digital health data. Helping future patients was the most common motivation for sharing digital health data (59%), followed by receiving one’s results (48%), supporting researchers (39%), and financial gain (31%).

Table 3 shows the results of the logistic analysis of the association between motives and positive willingness to share data. Motives found to be associated with positive motivation to share digital health data were helping future patients (OR 2.586, 95% CI 1.885‐3.548; P<.001), receiving one’s results (OR 2.266, 95% CI 1.624‐3.561; P<.001), and receiving financial benefits (OR 1.806, 95% CI 1.263‐2.582; P=.001).

The percentage of respondents who selected concerns about their digital health data is shown in Multimedia Appendix 4 FIgure S4 . Of the total respondents, 52% cited unknown use of data for the benefit of others as a concern, approximately 39% agreed to terms and conditions they did not fully understand, and approximately 34% said that data sharing would make them vulnerable to cyber attacks.

Table 4 shows the results of the logistic analysis of the relationship between positive intentions toward digital health data and concerns about sharing. The concern found to be associated with a positive willingness to share data was agreeing to contract terms that were not fully understood (OR 0.711, 95% CI 0.525‐0.968; P=.03). The results showed that those who felt concerned tended to have lower positive intentions than those who had no concerns about their data being used for unethical projects (OR 0.510, 95% CI 0.361‐0.722; P=.001).

Multimedia Appendix 5 FIgure S5 shows the percentage of participants’ responses regarding their preferred access to digital health data sharing. Receiving information about projects using their data was the most common response (43%), followed by receiving information about who is using their data (35%) and deciding who has access to what parts of their data (31%). However, 13% stated they would not need to be contacted after providing their data and having it anonymized.

Table 5 shows the results of the logistic analysis between positive intentions and each of the access restriction items regarding data sharing. The access restrictions found to be associated with positive intentions were receiving information about projects using their data (OR 2.313, 95% CI 1.700‐3.147; P=.001), giving public or academic institutions access to their data (OR 2.800, 95% CI 1.9248‐4.0734; P=.001), and not receiving information after providing their data (OR 1.929, 95% CI 1.292‐3.029; P=.004).

Internationally, secondary use and third-party provision of personal health-related data have become very important for public health, including surveillance of infectious diseases and the development of preventive methods and treatments [48]. Therefore, it is significant to investigate the factors related to data sharing, such as citizens’ attitudes toward, barriers to, and motivations for data sharing. However, research on citizens’ personal health-related data sharing is limited [49].

Our study identifies factors influencing citizens’ willingness to share a wide range of personal digital health data about health in Japan, including data in medical records, biobank samples, and digitized social communication data. This study showed that approximately 70% of Japanese participants were willing to share digital health data about their health. A survey of public attitudes toward digital research registries among Danish and Brazilian citizens showed that 91% of Brazilian and 70% of Danish citizens had a positive attitude toward digital research registries; our study shows a similar willingness as the results from the Danish participants [36]. Our research focuses on whether people are willing to share data, which directly evaluates behavioral intent. This approach is essential as it provides actionable data for organizations or governments to design effective policies. For instance, it can guide the creation of incentives to promote data sharing or enhance privacy protection measures.

While the Danish and Brazilian study asks about how “comfortable” people feel with data sharing, focusing on emotional aspects, our research directly addresses the question of “willingness to share.” This specificity enables deeper insights into both emotional responses and actual behaviors, bridging the gap between attitudes and actions. Our research design, which evaluates the intent to share data, is not limited by cultural or emotional factors. This makes it versatile and applicable across different global contexts. Conducting similar studies in other countries allows us to uncover universal trends and contrasts in data sharing behavior. Our study suggests that helping future patients is an important motivator for sharing health-related digital health data. Furthermore, participants were willing to share their data in the interest of research.

The participants’ high level of interest in their health suggested that gaining knowledge about their health is a motivating factor. Moreover, social benefits, such as benefits to future patients and benefits to oneself, were simultaneously motivating for sharing digital health data. Researcher benefits were not found to be associated with motivation for data sharing. A study of attitudes toward sharing health data among United States and United Kingdom citizens also reported a lower willingness to share data for commercial purposes compared to that for public benefit [43].

Based on our analysis, willingness to share digital health data regarding physical illness diagnoses, sleep patterns, food intake, blood samples, and stress was relatively high, while participants had a lower willingness to share social communication data. This may occur because social communication can be considered personal data that citizens may be hesitant to share due to privacy concerns [50]. However, social communication provides data for research into mental health and infectious disease control; thus, it holds important implications in public health research. Therefore, there is a need to increase public understanding of the importance and significance of digital health data in this area.

Our results suggest that concerns about providing data to unethical projects and agreeing to contract terms that were not fully understood were barriers to the willingness to share digital health data. A study on public attitudes toward the sharing of personal medical records and genomic data in Japan similarly reported concerns about the unauthorized use of data [35]. Although the indicators differ from this study, a link between literacy in the knowledge of genomics, AI, and digital devices, and the sharing of medical records and genomic data has also been noted [35]. Furthermore, trust is important for data sharing [51,52]. Privacy concerns deter data sharing, particularly attitudes toward sharing health data for health policy creation [53].

Aggregating and analyzing multiple types of health data can have significant benefits. We can gain comprehensive insights into health patterns and conditions by combining diverse sources, such as genetic data, clinical records, and lifestyle habits. This approach supports personalized health care, allowing tailored treatments and interventions to address individual needs effectively. Furthermore, the correlation of health data enables predictive analysis, helping forecast potential health issues and promoting preventative measures [54]. Citizens’ attitudes and receptivity to digital health data sharing differ depending on the type of data; however, concerns persist regarding personal data protection and transparency in data sharing [55].

While one measure to remove barriers of concern is restricting access to digital health data, our results showed that receiving information about projects that use data, getting information about data users, and granting data access to public or academic institutions are linked to a willingness to share data. These results may suggest that people avoid providing data to unethical projects and wish to exercise their right to self-determination regarding data. Other studies have also identified the importance of controlling the sharing of digital health data [36,37].

Our analysis suggests that the sociodemographic factors associated with willingness to share digital health data are being a man; owning smartphones, PCs, and smart appliances; having experience in participating in health research; and having high digital health literacy. It is necessary to consider measures for sharing digital health data that consider individual characteristics. The association between willingness to share digital health data and educational background reported by Grande et al [56] was not found in our study.

This study identifies factors that make Japanese citizens willing and able to share their data for research. These factors may be applicable not only in Japan but also internationally across borders. An increase in data sharing can contribute to research using big data, for example, the development of diagnostic, preventive, and therapeutic methods for rare diseases for which data are scarce, as well as international surveillance for infectious diseases.

The results of our study suggest that there are differences in the types of data that people are willing to share and that they may be less willing to share data that is related to personal data or not related to their health. Therefore, it is possible to fully explain the significance of sharing through communication with people, motivating them to share, and identifying the possible contributions to their health. Concerns about unethical data use projects and consenting before understanding the full terms of the agreement, identified as barriers to data sharing, can be addressed by disclosing information about the research project when requesting digital health data.

Concerns about unethical data usage and misunderstood consent indicate a demand for stronger ethical governance. Data governance frameworks must incorporate feedback mechanisms, such as providing people with updates about the projects that use their data, thereby enhancing perceived control and accountability.

Receiving this data may contribute to increased confidence in data sharing. Furthermore, ensuring that people have the right to control their data is also key to facilitating data sharing.

The study highlights demographic variations in willingness to share data, suggesting that public engagement strategies must be tailored to specific groups. For instance, women, people with lower eHealth literacy, and those without previous research participation may require more targeted education and reassurance regarding privacy and data control mechanisms.

The willingness of the Japanese public to share health data opens opportunities for international data pooling, particularly in the context of rare diseases and emerging infectious diseases, where domestic data may be insufficient for robust research. These findings support the establishment of cross-border data sharing frameworks that are ethical, secure, and culturally sensitive.

The association between device ownership and willingness to share digital health data suggests that disparities in access to digital technologies may translate into inequities in participation in health research. Policymakers should be aware of the digital divide when designing inclusive health data collection frameworks. These findings underscore the importance of policy measures that enhance public trust in digital health data sharing. Efforts to improve data literacy, transparency in data usage, and the ethical oversight of research projects are vital to increasing citizens’ willingness to share their data. Government and academic institutions should consider developing communication strategies that convey how data will be used, who will access it, and the potential social benefits of data sharing.

Two methodological limitations exist in this study. First, the web-based survey could have created the possibility of sampling bias as well as bias in education levels and ownership of digital devices.

Another limitation relates to our sampling design. Although we used random selection within strata defined by age and gender, this constitutes stratified random sampling rather than simple random sampling. In our analysis, we treated the sample as if drawn via simple random sampling, which may lead to inaccurate estimation of standard errors and affect inferential validity, including the risk of type I error. However, this design was intentionally chosen to ensure adequate representation across demographic subgroups, particularly to explore associations between data sharing intentions and age or gender. Given the total sample size of 1000, we believe the precision of estimates remains acceptable for our exploratory purposes, although results should be interpreted with appropriate caution.

Third, this study was a survey that asked questions about willingness to share digital health data based on future assumptions. However, the study did not explore the perspectives of data providers regarding implementing actual digital health data sharing.

While the quantitative data provides a broad understanding of attitudes, future research should include qualitative approaches, such as interviews or focus groups, to explore the underlying motivations, fears, and expectations surrounding data sharing in greater depth. Therefore, it is important to survey providers during actual data sharing in a future study. Conducting ongoing research into the topic may also be necessary.

This study identified people’s receptivity and willingness to share personal digital health data about their health and its influencing factors. Approximately 70% of the respondents viewed sharing digital health data about their health positively. The majority of those with a positive view were motivated to share their data by receiving data about themselves as well as helping future patients. Unwillingness to share digital health data about personal health is a barrier to promoting digital sharing and research for societal benefit. Concerns about data being used in unethical research projects and agreeing to contractual terms and conditions that are not understood were suggested to be associated with reduced willingness to share data. It is also important to improve the understanding of the information of users’ institutions, disclosure of information on the projects used, terms of contracts, and access control of data sharing. Furthermore, associations between willingness to share digital health data for public associations with personal characteristics were found regarding gender, interest in health issues, ownership of digital devices, history of participation in research, awareness of digital health databases, and digital health literacy. Therefore, a digital sharing system that considers individual characteristics and provides information regarding the usage of the data could facilitate data sharing.