This cross-sectional study was based on data obtained from a Dutch community-based sample (N=1013) as part of an annually recurring survey study (for additional details, see the study by Kop et al [24]). Research assistants were instructed to apply quota sampling within their personal network by approaching a minimum number of participants from each sex and age group to ensure equal representation (eg, to ensure the number of men aged between 31 and 40 years in the total sample is equal to the number of women aged between 31 and 40 years in the total sample). Inclusion criteria were age above 18 years and being fluent in Dutch. Although there were no geographical restrictions, most participants lived in the south of the Netherlands, as a result of the sampling procedure. Data for this study were collected between February 2022 and April 2022. Data collection took place online using Qualtrics.

Ethical approval was obtained from the Tilburg School of Social and Behavioral Sciences ethical review board (protocol RP55). All participants gave informed consent before taking part in this study, and their data were deidentified prior to data analysis. Participants were not compensated for their participation.

Data collection was conducted using a web-based survey after participants provided informed consent for their participation. All data were obtained using validated questionnaires or purpose-designed questions (see Multimedia Appendix 1 for an overview of the used questionnaires).

Descriptive statistics were obtained for all demographic and psychological variables by calculating the mean and SD for continuous variables and the number of occurrences and percentages for all categorical variables. Pearson and nonparametric Spearman correlation analyses were used to obtain an overview of the bivariate associations between the psychological measures and willingness to share health technology data. For illustrative purposes, odds ratios (ORs) with 95% CIs were also calculated using tertile-based dichotomized data of the psychological measures and the willingness to share health data (a value higher than 1 indicates a higher chance of willingness to share data based on the upper tertile of the psychological measure).

Multiple linear regression analyses were performed to assess which variables were independently associated with the willingness to share health technology data. Assumptions for regression analysis, including linearity, normality of residuals, homoscedasticity, and the presence of influential cases, were evaluated using the plot function in R (version 4.2.2; R Foundation for Statistical Computing) (findings are shown in Table S1 and Figure S1 in Multimedia Appendix 2). Data are presented in terms of the overall explained variance of the model (R²) and the standardized regression weights per variable (β). Analyses first focused on the results of the full sample (N=1013). Subsequent sensitivity analyses were conducted for participants who had used or currently used health technology apps (N=244) and those not using such apps (N=769).

Total scores were coded as missing values if participants answered <60% of all items of that specific scale. If at least 60% of the items of a questionnaire were completed, then missing items of that scale were imputed by the mean of the answered questions of that scale. The percentage of missing values among all the psychological item scores was <0.1%, and there appeared to be no patterns in missingness. For the multiple regression models, listwise deletion was used to deal with missing values on the total scores of each predictor. The potential of biased parameters estimated as a result of mean imputation was limited, as it was applied only within variables, the percentage of missing values was extremely low, and listwise deletion was used [32].

Exploratory network analyses were performed to further investigate the associations between the item scores of all the psychological scales. Network analyses were visualized with the “qgraph” package (version 1.9.3) in R [33]. Each node represents an individual item of the questionnaire, while edges represent the associations between 2 nodes (ie, questionnaire items).

Table 1 displays the participant characteristics. The mean age was 48.6 (SD 16.6) years, and 51.5% (522/1013) of the participants were women. Levels of education were relatively high (52.3% higher education). A current or history of a medical disorder was reported by 36.9% (374/1013) of the participants, and average anxiety and depressive symptom levels were relatively low.

The variable willingness to share health technology was approximately normally distributed, with scores ranging from 8 to 40 (mean 23.8, SD 6.0, median 24.0) (Figure S2 in Multimedia Appendix 2). Furthermore, 24.1% (244/1013) of the participants were currently using digital health apps or other forms of health technology (ie, wearable devices or smartphone apps aimed at improving physical activity, sleep, and medication adherence, or at reducing stress, alcohol use, and smoking). Participants who used technological devices were younger, were more likely to have a high education level, and had higher psychological flexibility levels than participants who did not use such devices, whereas no differences were found on the other variables listed in Table 1.

As shown in Table 2, high levels of optimism (r=0.07; P=.03) and psychological flexibility (r=0.13; P<.001) were significantly correlated with higher levels of willingness to share health technology data. In addition, higher levels of SI were associated with a lower willingness to share health technology data (r=−0.08; P=.01). Other bivariate correlations between psychological factors and willingness to share data were nonsignificant. When repeating these analyses with dichotomized scores, it was found that participants scoring in the highest tertile of psychological flexibility were 63% more likely to score high (upper tertile) on willingness to share health technology data (OR 1.63, 95% CI 1.25-2.12). Participants scoring in the lowest tertile of SI were also more likely to be willing to share their data (OR 1.53, 95% CI 1.17-2.00), whereas none of the other dichotomized psychological measures revealed significant ORs with a high level of willingness to share data.

Of the background factors, male participants (Cohen d=−0.14; P=.03) and technology users (Cohen d=−0.25; P<.001) scored significantly higher on willingness to share technology data (Table S2 in Multimedia Appendix 2). When adjusting for age, sex, and education level in separate linear regression models, optimism (β=.093; P=.004), psychological flexibility (β=.127; P<.001), and SI (β=−.096; P=.002) remained significantly associated with willingness to share health technology data.

To evaluate which psychological factor was associated with willingness to share health technology data, independent of other psychological factors, a multiple linear regression analysis was used entering all psychological factors simultaneously while also adjusting for age, sex, and education level (adjusted R²=0.03; P<.001). Higher psychological flexibility was the psychological variable that was independently associated with more willingness to share data (β=.097; P=.008) while adjusting for covariates. Other correlates of willingness to participate in health technology data sharing were younger age (β=−.086; P=.01), male sex (β=.073; P=.02), and lower education level (β=−.074; P=.02) (Table 2).

When repeating the analyses after stratifying the sample by the use of health technology, a similar pattern of results was found for both groups. Among health technology users (N=244), higher levels of psychological flexibility were independently associated with more willingness to share data (β=.204; P=.006). Other associations with the willingness to share data among technology users were nonsignificant (adjusted R² for the full model=0.04; P=.03). In the subset of participants who did not use health technology (N=769), lower SI was associated with a higher willingness to share data (β=−.092; P=.03), while male sex (β=.076; P=.04) and lower education level (β=−.089; P=.02) were also independently associated with willingness to share data (Table S3 in Multimedia Appendix 2).

Figure 1 shows the results of the network analysis based on the correlational structure of all questionnaire items. Every edge represents a correlation larger than 0.1. Each underlying construct is represented by a different color. The network visualizes the extent to which items cluster together, indicating the measurement of the same underlying construct. The network visualizes that the items on data sharing do not cluster together with most of the psychological predictors. Within the health technology data-sharing items, it appeared that items 7 (“I am worried about the confidentiality of the private information being exchanged by the health technology”) and 8 (“Health technology will violate my privacy”) were distinct from the other items (eg, “I would allow my personal data to be shared with my physician if it would improve my treatment”), based on their relative spatial position. The item-level comparisons show that items 7 and 8 cluster more toward items on optimism, while the other data-sharing items cluster more toward psychological flexibility. A regularized network analysis can be found in Figure S3 in Multimedia Appendix 2.

This study shows that higher levels of optimism and psychological flexibility and lower levels of SI are associated with higher levels of willingness to share health technology data. Multivariable analyses further indicated that psychological flexibility was the strongest and independent psychological factor associated with the willingness to share health technology data. Younger age, male sex, and lower education level were also independently associated with a higher willingness to share health technology data. Item-level network analysis revealed that optimism was associated with less concerns about confidentiality and privacy, whereas psychological flexibility was associated with a higher willingness to share health technology data if it could benefit the user or others.

The current findings are consistent with previous research indicating that positive psychological traits, such as optimism and trust, are associated with a higher willingness to share health data [17,19]. The association of psychological flexibility with willingness to share health technology data has not yet been described in the scientific literature. People high on psychological flexibility are able to adapt to situational demands and balance competing desires, needs, and life domains [20]. Furthermore, they are more open toward new forms of behavior or shifting behavioral repertoires to achieve goals related to their personal values, which could play a central role in the decision to share personal health technology data with others. When balancing between giving up a part of their privacy compared and helping themselves or others by sharing their data, their prosocial values can take precedence over their protective intuitions, resulting into a higher willingness to share data. By shifting behaviors toward sharing data, people accept potential privacy concerns and remain true to personal values such as helping others, pursuing optimal health outcomes, or other values related to the benefits of data sharing [34].

The findings of this study need to be interpreted in the context of several limitations. The questionnaire to evaluate willingness to share health technology data has not been validated in prior settings, although the psychometric properties of the questionnaire were good. There was a lack of diversity in the sample with regard to education level and ethnic background. The levels of generalized anxiety and depressive symptoms were also low in this community-based sample, potentially resulting in a floor effect for their relationship with willingness to share health data. The cross-sectional design precludes causal inferences and a more in-depth qualitative research approach, such as conducting individual interviews or focus groups, might provide additional information about the underlying values and arguments in the decision to share health technology data. However, even with additional knowledge on these motivations, there is a gap between data-sharing intention and action [35]. This observation indicates the need for future studies measuring actual privacy-related behaviors instead of evaluating intentions to engage in such behaviors. A strength of this study is the large sample covering a wide range of ages and an equal number of women and men, in which the relative importance of multiple psychological constructs could be studied. However, the set of psychological constructs was limited in the already existing dataset, and the models indicated very small explained variance, which pleads for an extension of other relevant psychological factors, such as technological self-efficacy, performance expectancy, trust, or other personality traits [14,17,36-39].

This study documents a significant association of optimism, psychological flexibility, and low SI with a higher likelihood of individuals being willing to share health technology data. Of these variables, psychological flexibility appeared to be the strongest factor, particularly among individuals who use health technology. In addition, younger age, male sex, and low education levels were associated with a higher willingness to share health technology data, particularly among people who currently do not use health technology. Examining the underlying network structure of a large set of psychological variables revealed that psychological flexibility is mainly associated with the benefits of sharing health data, while people low on optimism show more concerns about confidentiality and privacy. Still, the magnitude of effect sizes was low, and additional work is needed on the underlying psychological mechanisms and reasons behind the disclosure and sharing of health-related data in general and health technology data in particular. Future studies that investigate psychological factors with reasons for sharing health data could identify which specific reasons are important for whom on an individual level instead of a population level. Additional studies with a broader set of psychological variables should be conducted to identify those variables that contribute the most to the willingness to share health data before specific clinical implications can be formulated. To increase the explained variance of the regression models, a broader set of psychological measures should be considered in studying their relationship with the willingness to share health technology data. Findings from these studies would provide important building blocks for the design of future health technology and promote individually tailored functionalities. This would enable customization of the data-sharing settings of personal health technology data sharing in accordance with individual preferences, rather than a binary decision to share all data or no data at all. The possibility to retract or modify prior consent to health technology data sharing with minimal effort could also increase people’s initial participation in data sharing [40]. This personalized approach might result in less privacy concerns, higher user satisfaction with the technology, higher satisfaction with health care, and potentially also better health outcomes.