Policy approaches for increasing participation with personal health information and data sharing (HIDS)
Daniel J. Caron PhD a, b | Vincent Nicolini, PhD a | Alexandre Prud’homme c
ABSTRACT
Governments, public healthcare providers and researchers are urged by health experts to make evidence-based policy decisions and to support data-driven public health interventions, medical research and healthcare systems management. Facilitating personal health information and data sharing (HIDS) can generate multiple positive externalities, benefitting individuals, healthcare organizations and society at large. Nevertheless, sharing health information and data also comes with certain risks such as privacy breaches and misuses of confidential health information and files. This can have both negative social and individual impacts. In this paper, based on a survey of 2,016 persons in Quebec, we show that providing more information on the individual and collective benefits of HIDS is key to achieving participation from citizens. Survey results showed that 143 respondents were willing to share their health information data unconditionally while 1,706 respondents changed their opinion from resistant to cooperative, after being better informed on the benefits of health data sharing, plus another 127 after being informed of the conditions of a potential data management regime. Our survey shows the importance of building trust to foster individuals’ willingness to share their health information and data and contributes to the evolving debate revolving around the social acceptability of HIDS. Public policies and regulations designed to reassure citizens about privacy protection and thereby facilitate sharing personal health information are important, but likely to be less effective if the public is unaware of the safeguards provided by these measures and uninformed about the benefits of HIDS. Our results demonstrate that trust is the first ingredient that will allow an effective HIDS regime, and that trust-building is a long-term communication and knowledge translation task that requires ongoing and transparent conversation between citizens and institutions.
Author Affiliations
- École nationale d’administration publique (ENAP), Montreal, Quebec.
- School of Public Policy and Administration, Carleton University, Ottawa.
- Université de Montréal, Montreal, Quebec.
Correspondence
[email protected]
Cite Article
Caron, Daniel J et al (2023). Policy approaches for increasing participation with personal health information and data sharing (HIDS). Canadian Health Policy, MAR 2023. https://doi.org/10.54194/GRLE9589 www.canadianhealthpolicy.com
Key Words
Canada, health, information, data, informatics
Disclosure
This research received funding from the Health Department of the Government of Quebec (MSSS).
Article
Submitted: 9 FEB 2023. Status: Peer reviewed. Open Access sponsor: n/a
Journal
Publisher: Canadian Health Policy Institute Inc. www.chpi.ca
ISSN 2562-9492 | DOI https://doi.org/10.54194/XYCP5241
(to access tables and figures download the PDF)
BACKGROUND
Governments, public healthcare providers and researchers are urged by health experts to make evidence-based policy decisions and to support data-driven public health interventions, medical research and healthcare systems management. Facilitating personal health information and data sharing (HIDS) can generate multiple positive externalities, benefitting individuals, healthcare organizations and society at large. According to research, the two main benefits of sharing health data are improved health outcomes (Green, 2019; Edelstein et al., 2018; Hutson et al., 2019; Malboeuf, 2018) and greater efficiency in the provision of care and its management (Green, 2019; Hutson et al., 2019; Malboeuf, 2018).
Nevertheless, sharing health information and data also comes with certain risks such as privacy breaches and misuses of confidential health information and files. This can have both negative social and individual impacts. For instance, non-consensual disclosure of an individual’s actual or previous health conditions could make it difficult to obtain private insurance. Furthermore, the growing frequency of data infringements (Webster 2020) and the pervasiveness of data collection by tech companies and data brokers through social media, search engines and websites has expanded the collection and usage of HIDS way beyond healthcare providers (Gostin, Halabi and Wilson 2018). As noted by Ghafur and co-authors (2020), recent trends in the commodification of data “have increased public awareness of the risks and consequences of sharing data with commercial and tech industries.” Consequently, the commercial usage of health information and data may lead to perceived or real imbalances between the pursuit of private profits and public benefits. As a case in point, various researchers have shown that the willingness to share health information is lower when the recipient has a commercial purpose (Bell, Ohno-Machado and Adela Grando 2014; Ghafur et al. 2020; Jones et al. 2022). This could explain why privacy apprehensions are more noticeable when health information data are shared with certain healthcare professionals like community pharmacists who privately own their pharmacies (Rolf von den Baumen et al. 2020).
Research has shown that individual attitudes towards HIDS are generally related to the above-mentioned risks and benefits (Kalkman et al. 2019; Hutchings et al. 2020) and accordingly to the social, political and technological environment within which data is collected, shared, utilized and exploited, and to the rhetoric that justifies such actions (Ostherr et al. 2017). The social and political sensitivity of HIDS was exemplified through recent experiences with the implementation of electronic medical records and health data sharing infrastructures by governments and healthcare providers in Australia (Duckett 2018), in the UK (Carter, Laurie and Dixon Woods 2015; Sterckx et al. 2016), and in France (Bévière-Boyer 2021). These international projects have disclosed the lack of communication towards citizens and indeterminacies of some key aspects, which produced uncertainties for individuals and contributed to undermine their acceptance. With the COVID-19 pandemic amplifying demand for health data and putting at the forefront of the debate the need for health data sharing (Boudreau and Caron 2022), governments and healthcare providers will be more than ever confronted with the necessity of generating public support for HIDS.
METHODOLOGY
This paper describes the results of a study which aimed to answer two questions: (1) what are the benefits of HIDS with reference to how information is provided and (2) how do conditions regulating HIDS influence the willingness of respective individuals to share their health information data?
A literature review was conducted on HIDS allowing us to identify two distinctive sets of elements influencing the levels of acceptance of individuals to share their health information, namely the main benefits of HIDS and the conditions under which the data are being managed by organizations. Based on the results of the literature review we hypothesized that the more citizens are informed about the benefits of HIDS, the more willing they will be to share their personal health information data, if it contributes to materialize those benefits. Building on studies examining the relationships between information and the formation of individual opinion (Tversky and Kahneman 1974; Kahneman 2011; Sheela and Mannering 2019), a survey strategy was developed in line with a sequential approach. This was intended to measure if and how new evidence-based information on the benefits of HIDS positively influenced individuals’ propensity to share their health information. The survey data collection strategy also aimed to allow for statistical analysis revolving around several variables permitting us to identify correlations between socio-demographic variables and indicators of support for health data sharing, and cooperation with stakeholders, health professionals, university researchers and public institutions, and under what conditions if any.
We used a sequential decision process to measure the impact on the willingness of individuals to share their health information data when they are being provided with more information. Our analysis was based on an extensive survey (carried out by SOM, a research firm specializing in data collection, analysis and visualization) reaching respondents via phone and the Internet within Quebec. The target population was adults of 18 years and older. The initial sample was randomly generated from all valid telephone numbers within the province. 1,008 respondents were selected from land line, and cellular phone numbers. For land line phone numbers, a random selection of an adult in the household was applied by using an age-based randomization computerized selection procedure. For cell phone numbers, respondents were automatically selected if they were 18 years of age or older. Another 1,008 respondents were selected for the Internet survey component, using a probability panel made up of Internet users randomly recruited via telephone by the survey firm. In addition, an external panel was also used to complete some profiles. The sampling consisted of a total of 2,016 respondents and it was representative of the adult population of the province by region, age, sex, education, and the size of households. The telephone and Internet surveys were carried out between October 29, 2020, and November 16, 2020. The length of the interviews was approximately 11.6 minutes for the telephone part and 6.9 minutes for the web component.
The first set of questions was developed to determine the knowledge and values of respondents with reference to sharing health information, regardless of their desire or willingness to share. We measured participants’ level of trust in the provincial Ministry of Health). Confidence in the MSSS was an indicator constructed from three questions:
- In your opinion, are Quebecers’ health information currently well protected against data theft?
- In general, do you believe that the Ministère de la Santé et des Services sociaux (MSSS) manages your health information well?
- In general, do you believe that the Ministère de la Santé et des Services sociaux (MSSS) is telling the truth about sharing your health information?
A three-point score was structured (Yes = 1 point). We agreed that a score of 3 out of 3 is a high level of trust in HHS. Then, 2 out of 3 is considered as ‘Medium’ and 0-1 translates into a low level of trust.
A second variable, named General Opinion on Sharing Health Information, was derived from a two-question construct. Note that this indicator differs from individuals’ intention to share their own health information.
- Do you think that in the future, health information should be shared to help the government better manage health system challenges?
- In your opinion, in the future, should health information be shared to improve the performance of Quebec’s health care system?
Individuals who indicated ‘Yes’ to the two previous questions were classified as ‘In Favour of Sharing Health Information’. Those respondents who did not indicate a positive answer to either question was classified as ‘Unfavorable to Sharing Health Information’, while those who indicated one positive response out of two were classified as ‘Mitigated’.
The second set of questions looked at participants’ preferences for sharing their health information. Those questions constituted the three decisional questions at the crux of our experiment. According to answers provided, the surveyed respondents were redirected towards two avenues with a third option. First, in the case where respondents said they agreed with sharing their health information with all the proposed stakeholders (health network stakeholders, researchers in universities, and private sector), the respondents were directed to the second series of questions on their preferences regarding the sharing conditions (e.g., security, anonymity). Second, when respondents were undecided or disagreed with at least one of the stakeholders proposed for sharing health information data, they were directed to subsequent questions based on the first responder who received a negative response (e.g., healthcare network stakeholders, university researchers, etc.). The following questions put forward scientific statements drawn from scientific literature. Based on this, they were asked new questions about whether they would change their minds about sharing their health information considering this new information. Third, if they changed their perspective due to the scientific information provided on HIDS impacts, they were directed to a series of questions to find out their preferences for sharing. If reading the scientific statements did not change their perspective and they were still undecided or disagreed with sharing their health information, they were then directed to another set of questions about other factors that could potentially cause them to reconsider their perspective on sharing their health information. After pre-testing we removed references to the respondent’s willingness to share data with private sector entities from the questionnaire because it introduced confusion. The study was focused on the provincial health institutional environment, and on sharing health information data with government departments, inside the public healthcare system, and with researchers affiliated to public institutions.
In the third set of questions, all participants were required to answer a question about their risk aversion using a proposed fictitious scenario. All the participants also answered questions aimed at determining the trust they placed in the Quebec’s Provincial Health Department. Socio-demographic questions were also asked. An open-ended question designed to allow participants to express themselves on sharing health information ended the questionnaire.
Ordered logistic regressions were performed to determine the factors related with trust in the MSSS and general opinion in sharing health data. Sociodemographic characteristics and perceived health status indicators were used as independent variables. The initial willingness to share health information models (for each stakeholder) were estimated by a logistic regression (0 = Refuse to share; 1 = Agree to share). Trust in the MSSS, knowledge about sharing and general opinion were used as explanatory variables. We included sociodemographic characteristics and perceived health status as control variables. The second decision model included only individuals who initially refused or agreed but under certain conditions. For this model, the dependant variable reflected the change of intention to share (0 = Would not change initial opinion; 1 = Agree to share if at least one condition is respected). Once more, we carried out a logistic regression including trust in the MSSS, knowledge about sharing and general opinion as explanatory variables, and sociodemographic characteristics and perceived health status as control variables. The third decision model included those individuals who did not change their opinion after the second attempt. Once again, the dependant variable calculated the change of intention to share (0 = Would not change initial opinion; 1 = Agree to share if at least one condition is respected). As for the other regression models, we performed a logistic regression including trust in the MSSS, knowledge about sharing and general opinion as explanatory variables, and sociodemographic characteristics and perceived health status as control variables.
RESULTS
Literature Review
Benefits of HIDS
Previous research highlights several benefits connected with sharing health data both from a collective and an individual point of view (Summarized in TABLE 1). At the collective level, sharing health data offers an opportunity to increase the efficiency of the healthcare system. For example, better sharing of health data reduces the cost of healthcare by, among other things, avoiding duplication of services and reducing the time and resources required for healthcare without compromising on quality (Simon et al. 2009; American Hospital Association 2019; Van Panhuis et al. 2014). In addition, based on health data, the development of artificial intelligence and robotics also promises to positively impact health systems efficiency (Gruson 2019).
Likewise, the sharing of health data reinforces scientific research and innovation. Scientific research on health is fuelled by the data to which it has access. Lack of access to quality health data reduces the likelihood of medical progress and innovation (Pisani and AbouZahrb 2010; Kim et al. 2017). Some authors even advocate for opening up health data to bring forth scientific collaborations (Huston et al. 2019). Similarly, diverse players in the health network produces their own data. Most of the time, such data are kept in closed databases. These locked databases reflect the organizational silos that characterizes the health system and many public organizations. The pooling of the latter offers immense potential for healthcare research by making it possible to build a less fragmented portrait of research revolving around health (King et al. 2016). This more transversal vision and the new relations that can emerge from it open the door to a better understanding of diseases and possible treatments (Shah et al. 2019a). Moreover, since healthcare research is often publicly funded, taxpayers should be able to derive maximum benefit from it (Pisani and AbouZahrb 2010; Shah et al. 2019b).
Furthermore, the sharing of health data can help public health decision-making. Access to health data enables decision-makers to make more informed decisions about public health issues (Huston et al. 2019; Van Panhuis et al. 2014). For instance, Nutley and Reynolds (2013) underscore the importance of data-informed decision-making within their work. Analogously, health data also contribute to the monitoring of public health issues (Edelstein et al. 2018). Such a public health surveillance ensures more coordinated risk management and more effective public health interventions when needed (Edelstein et al. 2018).
The benefits of sharing health data are also characterized at the individual level although it is to be noted that these two perspectives are not so distinct and can intersect. At multiple instances, there can be such a close link between individual and collective benefits that improving the individual health status will necessarily generate benefits for the common good through a decrease in public health expenditures.
Several researchers report positive outcomes concomitant with healthcare quality (Ceccato and Price 2019; Dimitropoulos et al. 2011; Kostkova et al. 2016). These include better decision-making related to the patient’s condition because decisions are better informed by up-to-date and comprehensive data (Maiorana et al. 2012; American Hospital Association 2019). The development of artificial intelligence based on health data also contributes to improved decision-making (Becuwe and Thébault 2020). At the same time, access to timely and quality data allows for earlier detection of disease and targeted follow-up for people with chronic diseases (Simon et al. 2009). Better coordination among health care providers is also a benefit for sharing health data. It allows more integrated care management and support for the patient (American Hospital Association 2019). For example, if all health professionals treating the same patient have access to all their health data, it is conceivable to ensure more rigorous follow-up. It also helps to avoid errors or duplication of care. The patient experience is therefore enhanced and rendered safer. This benefit is particularly relevant for patients who are suffering from chronic diseases and who are receiving care from manifold providers (O’Donnell et al. 2011).
In addition, the sharing of health data offers new opportunities to inform and communicate with the patient. Sharing health data may improve the relationship between health professionals and patients by promoting communication among them (Maiorana et al. 2012). As such, a telephone survey study conducted in the United States among 1,542 respondents showed that 67% of them felt that the use of a shared electronic record would improve their communication with their physician (O’Donnell et al. 2011). From the patient’s perspective, having better access to their medical record and being better informed empowers individuals to become active stakeholders in maintaining their own health (American Hospital Association 2019). This leads to greater patient empowerment.
One of the greatest benefits of sharing health data is the ability to save more lives. Sharing health data leads the healthcare system to greater performance. This increased efficiency positively influences the number of lives saved (Van Panhuis et al. 2014). By allowing the right people to have access to the right information at the right time can indeed foster health data sharing to reduce mortality rates (Platt et al. 2018). Here, the benefit can be envisioned both collectively and individually. Besides, a faster flow of information between health care providers within the health care system reduces the time required to make a diagnosis. This decrease in diagnostic delays also has a major consequence on the quality of care received and on the possibility of saving lives (Courbier et al. 2019).
Factors Impacting the Acceptance to Disclose HIDS
Studies have drawn attention to many factors impacting the acceptance of health information and data sharing by individuals, with the most frequently identified being perceived benefits, apparent concerns or risks, trust, individual control over sharing (Cumyn et al. 2019) and transparency (Esmaeilzadeh and Sambavasian 2017; Moon 2017; Kim et al. 2017; Kalkman 2019). We can regroup these factors under four categories: (1) trust, (2) security and privacy, (3) sharing dispositions, and (4) socio-demographic.
Trust is one of the most influential factors regarding the willingness of individuals to share their health data (Sterckx et al. 2016). Trust refers to a dynamic multidimensional relationship between two parties characterized by an expectation or willingness to confer authority and acceptability to another person to perform a given set of tasks (Platt et al. 2018). Platt and co-authors (2018) distinguished three main reasons why trust is a factor of interest for the sharing of health data. Firstly, it improves relationships and information transfer across the healthcare network. For instance, an increased level of trust promotes effective work organization between providers and patients. Secondly, trust between the patient and the healthcare professional is paramount because of the significant asymmetry of information that exists between the two parties. Thirdly, despite security precautions for the technology operated and the framework used to govern the information shared, there is still a risk for privacy and confidentiality. Lack of trust is an important cause of failure for governmental health data sharing initiatives, as was the case with care.data in the UK (Carter, Laurie and Dixon-Woods 2015).
The factors influencing trust are neither subjective nor objective because they stem from a learning process (Luhmann, 2017). Consequently, they are multiple and include, for example, technological, sociological, or psychological aspects (Maiorana et al. 2012). From a technological point of view, trust particularly revolves around security and respect for personal data. For that reason, it is necessary to ensure that shared data are used conscientiously and are not made available to other parties or the public without prior consent (Edelstein et al. 2018). From a sociological vantage point, it plays out more at the level of trust in the government, the healthcare system, and the professionals who work within it (Platt et al. 2018). Psychologically, it is more about the trust that healthcare professionals are acting in the best interest of patients and the patients’ awareness that limited sharing of health data has negative effects on health care delivery. It can also be contemplated as an affective force that shapes the acceptance of health data sharing (Lupton 2019). Hence, trust is a significant factor with multiple dimensions that influences the individual’s acceptance for HIDS.
In terms of collection and handling of HIDS, the scientific literature reviewed articulates two other conditions: security and privacy. On the one hand, security of health data refers to the protection against unauthorized access and maintenance of integrity and availability (Abouelmehdi et al. 2018). It focuses on protecting health data against malicious attacks and health data theft for profit-making purposes (Abouelmehdi et al. 2018). On the other hand, the protection of privacy refers to the ability to protect sensitive information that could lead to personal identification. Although security and the protection of privacy may be grasped as two clear-cut elements, they are intrinsically linked, particularly in the collective thinking. There is no doubt that security and privacy are key factors that play a role in the willingness to share health data (Kalkman et al. 2019; Ghafur et al. 2020). Cyber-attacks events like data breaches have a high visibility and they subsequently affect the perceptions of individuals. According to a UK study of 2018, people were 53% to be more apprehensive with data security than three years before and they were generally more aware of negative outcomes than positive ones (Healthwatch England 2018). Nonetheless, the relationship between privacy and security concerns and the willingness to share health information data can be mitigated by other factors. Contemporary studies have shown that individuals, while concerned about privacy and security, tend to confer a higher value to potential benefits (Dimitropoulos et al. 2011) or quality of care (Walker et al. 2017) when deciding about sharing or not their health information data.
The scientific literature reports that sharing dispositions are a factor affecting the acceptance of HIDS. For instance, on the one hand, patients are fearful about the idea of sharing their data when they do not know what it will be used for. Additionally, they are scared that their data may be shared outside the health care system, for example, with pharmaceutical companies or with researchers (Pisani et al. 2016; Bell et al. 2014). As a result, the patient’s level of access and control over the shared health data influences the decision to share (McCormick et al. 2019). The less control the patient has over the sharing process, the less inclined he or she will do so (Moon 2017).
On the other hand, the demand for explicit consent or permission is associated with a greater willingness to share (Moon 2017). Adding a time limit to the health data sharing also positively influences acceptance of health data sharing (Moon 2017). In addition, limiting the information available to what is essential (e.g., the need-to-know principle) also contributes to stimulate the willingness to share. In this sense, acceptance is more significant if the data are shared with stakeholders who need them to reach a specific objective within a defined context (e.g., patient management or for scientific research purposes). Motivation for the usage of health information data has been identified as one of the most important concerns of individuals when assessing where they should share their data (Stockdale, Cassell and Ford 2018).
Moreover, the type of data shared also influences the willingness to share. Some types of health information data are considered more sensitive by patients (e.g., data on mental health, drug and alcohol use, sexually transmitted diseases, and HIV) (Moon 2017). Hence, the importance of ensuring a form of sovereignty for the patient over sharing his or her data. Choosing which data will be shared is connected with a greater desire to share (Moon 2017). Likewise, ensuring the anonymity of the health data also positively motivates the willingness to share health information data (Bell et al. 2014). Many patients are worried that the shared data could lead to their identification, which could cause them prejudice in obtaining certain services. In the scientific literature examined, several methods are proposed to reduce the risks allied with patient’s identification (Abouelmehdi et al. 2018).
Therefore, transparency is a crucial factor that can influence trust (Mabillard and Caron 2022). The lack of transparency manifests itself particularly through the lack of information given to the citizen about how his or her health data would be made accessible, would be used and eventually shared (Moon 2017). As a general rule, knowledge about privacy measures and information on data governance (processes and responsibilities) have a positive impact on an individual’s acceptance of HIDS (Kalkman et al. 2019).
Certain sociodemographic characteristics stand out in the scientific literature appraised as they may influence the willingness to share health data. Firstly, people with chronic diseases tend to disagree or express more reservations regarding sharing their health data (Woldaregay et al. 2020; Moon 2017; Bell et al. 2014). Secondly, these individuals are generally greater users of the health care network. Because they produce more health data than a healthy individual, it is, consequently, possible to infer that, in their opinion, taking into consideration the amount of data in their name and the frequency with which they use the healthcare system, the risks associated with health data sharing may appear difficult to control and, thus, have more direct consequences. In addition, people with chronic diseases are generally more likely to be older adults. The digital divide can also play a role in the willingness to share (Moon 2017).
Although the scientific community does not seem to agree on the influence of ethnic origins in the decision to share health data, it is still put forward in some research as an influencing factor. Indeed, some studies have shown correlations between ethnicity and willingness to share health information data (Grande et al. 2013; Dimitropoulos et al. 2011), while others suggest no significant correlation between the two (Kim et al. 2017). Finally, the research of Kim and co-authors (2017) shows that a person’s degree of technical ability is not a guarantee of his or her willingness to agree to share his or her health data.
In summary, the scientific literature identifies numerous collective and individual benefits to HIDS. While these benefits are proven, multiple factors have at the same time an impact on the acceptance of HIDS, which leads to the question of how the former interact with the latter. If both are usually recognized as determinants of an individual’s acceptance for HIDS, for instance through the form of perceived benefits and perceived risks, it is yet to be examined more closely how a clear portrayal of the benefits can have an impact on an individual’s dispositions to accept or reject HIDS. Accordingly, our research aimed to gauge more closely the extent to which information provision plays a role in sharing health data, that is how the knowledge of the benefits of HIDS interacts with the factors persuading individuals in their decision to share or not their health information data. As well, our explorative study sought to assess the nature of the other elements that can contribute to or preclude that acceptance.
Analysis
TABLE 2 portrays the descriptive statistics of the independent variables of our model. FIGURE 1 summarizes the structure of the survey around the three decisional questions. It also depicts an outline of how the sharing of health information data progresses along with the participants of the survey. TABLES 3, 4, 5, and 6 show the analytical results. The data analysis shows that trust is one of those elements playing an essential role on individuals’ willingness to share their health information data. As shown in TABLE 3, nearly half of the respondents (45.9%) mistrust the Quebec’s Provincial Health Department, while 63.5% have a favorable opinion on HIDS in general. When we looked more closely at the determinants of those variables as detailed in TABLE 4, we saw that trust towards the MSSS was clearly interconnected with a favorable general opinion on HIDS and a very good or excellent perceived health status. For individuals having a favorable opinion on sharing health information, the odds of having a high trust in the MSSS (i.e., high or likely medium versus low) were 2.508 times higher than for individuals who reported an unfavorable general opinion on sharing health information, holding constant all other variables. Furthermore, and as shown within TABLE 5 and TABLE 6, the level of trust towards the MSSS and the general opinion on HIDS were the strongest factors correlated with a change of opinion and a willingness to share HIDS with all stakeholders. During the first iteration, only a small number (7.1%, see FIGURE 1) of participants agreed to share their health data without conditions. But these participants were more likely to have a high trust in the Quebec Ministry of Health and a favorable opinion about HIDS. Those who changed opinion in the second iteration were in a similar way more likely to have declared a high level of trust in the MSSS. While we cannot say that those who were reluctant to share their health information data in each of the iterations had a lower trust towards the MSSS, we can, nonetheless, conclude that trust plays an important role in individuals’ decision to share their HIDS with most stakeholders.
Knowledge of the benefits contributed to change the opinion of 95.3% (1,706 out of 1,790) of the respondents. Furthermore, when informing the respondents of hypothetical conditions that could be applied and reiterating some potential benefits, of those who were reluctant to change their opinion after the second decision plus those who disagreed to share at the first and second iteration, 76.0% changed their mind and agreed to share their health information data. After the third iteration, we ended up with around 2% of participants (40 out of 2,016) who would not change their opinion whatever be the benefits and/or conditions. The important role of information on the benefits holds true for all stakeholders. These findings are in line with previous and more recent studies which assert that individuals tend to be highly responsive to the benefits of health data sharing, even when knowing about the risks (Dimitropoulos et al. 2011; Walker et al. 2017; Bearth and Siegrist 2021; Summers et al. 2022). Nevertheless, we must not minimize the fact that an inclusion of the private sector in the second and third question could have generated different answers, possibly more negative ones (Braunack-Mayer et al. 2021; Tosoni et al. 2022). When participants were questioned about their knowledge of the current data management regime of the government and healthcare system, our study results demonstrate that it is not well understood. For example, answers determine that most participants (77%) think that any healthcare professionals that they consult can have access to their health information data and believe (75%) that their healthcare professional needs to file their electronic medical record system diagnosis. More than 30% think that their health information can be shared between government departments without their consent. These perceptions do not reflect the current rules and practices. This lack of knowledge is in agreement with what is found elsewhere in the literature (Atkin et al. 2021), namely that individuals do not have a “spontaneous understanding about the range of ways patient data is used in health” (Understanding Patient Data 2018). This could impact perceptions vis-à-vis the Quebec’s Health Department capacity to effectively manage health information data sharing. As a matter of fact, it follows that many individuals forge their decisions on whether to share their health data or not based on inaccurate information. While this does not seem to have a direct incidence on the refusal to share health information data, is it still meaningful.
Socio-demographic factors were often not statistically significant, but when they were, these factors did not seem to play a significant role in determining individuals’ willingness to share their health information data. For example, during the first iteration, people with a higher revenue were more likely to agree to share their HIDS with stakeholders from the healthcare system, but the same was not found for sharing with all stakeholders. The level of education, which can be related to a higher revenue, was not linked to a higher willingness to share HIDS. Here and there, age, sex or language appear statistically revelatory, but we could not find any pattern that could point to a constant relevance of socio-demographic factors. This differs from findings in other recent studies (Helou et al. 2021; Joseph et al. 2022). These differences suggest that further studies may be necessary to properly grasp the role and influence of this variable. In overview, we had 143 respondents willing to share their health information data unconditionally while 1,706 respondents have changed their opinion after being better informed on the benefits of health data sharing, plus another 127 when they were being informed of the conditions of a potential data management regime. Subsequently, our hypothesis that with more information provided, individuals tend to be more willing to share their health information data – is corroborated by the data. The study has also shown other significant elements that have an influence on the willingness of individuals to share their health information. As illustrated in FIGURE 2, other conditions are also playing a role in building an acceptable health data sharing regime. Amongst those requirements, security and privacy are ranking very high.
IMPLICATIONS
The results confirm our assumption that people are willing to modify their initial position on health data sharing if they are better informed about the personal and social benefits of such data sharing. Several implications for public policy can be derived from our research outcomes. First, trust has been found as the fundamental pillar to HIDS acceptability and viability. Regarded as a learning process (Luhmann, 2017), trust building must become an ongoing concern and task for public health administrations.
Second, if trust building is a long-term undertaking, inferring from our results gives information a key role in building trust. However, one of the main corollaries to these outcomes are that, although desirable, one-time arrangements such as consent or annual reporting from institutions to citizens may not be a sufficient condition. One of the reasons for that is the weak level of knowledge that citizens have of health data management regimes. Those regimes are usually quite complex (Caron 2021) and unfamiliar to citizens or not known well enough to rely on sporadic communications (annual reporting) or one-off contract (consent) to expect to build trust at a sufficient level to piece together HIDS acceptability and viability. Policymakers must become more attentive to not only the role that providing information can play but also on the type, frequency and channel used to share such information. For instance, this could be implemented through constant communications about benefits as well as on the management regime of data sharing that trust building can be undertaken over time.
Third, we found that clarity of assertions about the benefits and the conditions under which HIDS could operate induced a change of opinion for a very high proportion of individuals. Simultaneously and as already mentioned, our study showed that a high number of individuals were not familiar with the workings of the data sharing regime of Quebec’s government and public healthcare system and they carried false assumptions on some of its aspects. Based on our analysis, we understand that governments, public healthcare providers and researchers seeking to establish a HIDS infrastructure must provide clear information to individuals on both the benefits of HIDS and on how the data will be collected, used, and protected. Clarity and simplicity are paramount to trust building and acceptability.
Fourth, to be effective, information must not only be shared as a hypothetical outcome communicated at the beginning of a research project based on health information data or at the moment of the implementation of a HIDS infrastructure. Ongoing communication of clear and proven benefits in the medium and long term, as well as during the course of the project would contribute to maintain individuals’ acceptance of HIDS and build trust. Practically, this can be achieved through evaluation committees including patients, citizens, administrators, and researchers, and appraisals of the main secondary usages of health data and information and how they helped to provide collective and individual benefits such as higher efficiency or better treatment for specific diseases. But above all requirements, it must be communicated in a very plain language, on a continuing basis and with a high degree of transparency.
Fifth, and accordingly, whereas the level of trust towards the MSSS reported by respondents was not particularly high, with almost half of them having a low level of trust, it was nevertheless not a factor that prevented them from changing their opinions during the second and third iteration. Nonetheless, one must keep in mind that the way the research was conducted was precisely to help remediate to the lack of trust by providing information unknown to most respondents in a plain language. Notwithstanding this, given the importance of such a factor in the scientific literature reviewed and the fact that trust was a high determinant of acceptance to share with all stakeholders during the first iteration, our results suggest to pay careful attention to how trust is impacted by certain decisions or some circumstances. For instance, when health information data is shared in the healthcare system with other governmental entities or with researchers, individuals should be able to know with precision the expected benefits and the rules that govern such a sharing, if not they should have control over how the data is used (Rivas Velarde et al. 2021).
Sixth, our research results suggest that more information should be readily available about measures taken to secure the health data and ensure its anonymity. Concerning the latter, HIDS regime could include an explanation of anonymization procedures with a clear statement of their limitations. In that sense, clear regulations and better communication will appear as a key driver for motivating citizens to accept the sharing of their health information data. This can be captured under the broader term “transparency” of the regime.
CONCLUSION
This study contributes to the evolving debate revolving around the social acceptability of HIDS. Our research raises questions about dependence on experts to decide what the design of an acceptable HIDS regime should be. By experts we mean the users of data, that is health professionals and researchers. Instead, it suggests putting the “owner” of the data, precisely the patient, at the heart of the equation. Even to make consent a viable solution, it appears elemental to engage citizens within a continuous dialogue to co-construct both the policies around HIDS and its operational framework. Doing so would altogether favor trust building as well as progressively gaining social acceptability for HIDS. Our study makes evident that we cannot simply assume that citizens will be satisfied to hear that sharing health information data will be beneficial to them and society and that it will be done properly. This needs to be fostered and managed jointly with them.
REFERENCES
Abelson, J., Miller, F.A., Giacomini, M. (2009). What does it mean to trust a health system?: A qualitative study of Canadian health care values, Health Policy, 91, 63-70, https://doi.org/10.1016/j.healthpol.2008.11.006.
Abouelmehdi, K., Abderrahim B.-H. & Hayat K. (2018). Big healthcare data: preserving security and privacy. Journal of Big Data, 5, 1–18. https://doi.org/10.1186/s40537-017-0110-7
American Hospital Association (2019). Sharing Data, Saving Lives: The Hospital Agenda for Interoperability. Retrieved from https://www.aha.org/system/files/2019-01/Report01_18_19-SharingData-Saving-Lives_FINAL.pdf. Accessed August 25, 2022.
Atkin, C., Crosby, B., Dunn, K. & al. (2021). Perceptions of Anonymised Data Use and Awareness of the NHS Data Opt-Out Amongst Patients, Carers and Healthcare Staff. Research Involvement and Engagement, 7, Article 40. https://doi.org/10.1186/s40900-021-00281-2.
Bearth A. & Siegrist M. (2020). Psychological Factors that Determine People’s Willingness-to-Share Genetic Data for Research. Clinical Genetics 2020, 97, 483-491. https://doi.org/10.1111/cge.13686.
Becuwe, A. & Thébaut, C. (2020). Introduction du numéro spécial : les impacts des nouvelles technologies sur les systèmes de santé. Marché Et Organisations, 38, 9-13. https://doi.org/ 10.3917/maorg.038.0009
Bell, E. A., Ohno-Machado, L. & Adela Grando, M. (2014). Sharing My Health Data: A Survey of Data Sharing Preferences of Healthy Individuals. AMIA Annual Symposium Proceedings. AMIA Symposium: 1699–1708. PMC4419941
Bévière-Boyer, B. (2021). La gestion des données de santé par le Heath Data Hub : le recours à la société Microsoft, entre risques et précautions. Droit, Santé et Société, 3, 42-48.
Boudreau, C. et Caron, D. J. (2022). Covid-19 et surveillance automatisée: une analyse comparée entre le Canada et la Corée du Sud. Administration publique du Canada, 65, 261-277.
Carter P., Laurie G. T. & Dixon-Woods M. (2015). The Social Licence for Research: Why care.data Ran into Trouble, Journal of Medical Ethics 2015, 41, 404-409. http://dx.doi.org/10.1136/medethics-2014-102374.
Caron, D.J. (2021), in collaboration with Nicolini, V., Bernard, S. and Lamontagne, R. Partage des données en santé au Québec : modélisation et cartographie des trajectoires de partage des données. Rapport de recherche. Chaire de recherche en exploitation des ressources informationnelles. Gatineau (Québec) : École nationale d’administration publique (ÉNAP).
Ceccato, N. & Courteney P. (2019). When Personal Health Data is No Longer “Personal”. Healthcare Management Forum, 32, 326–328. https://doi.org/10.1177/0840470419865851.
Courbier, S., Dimond, R. & Bros-Facer, V. (2019). Share and Protect our Health Data: An Evidence-Based Approach to Rare Disease Patients’ Perspectives on Data Sharing and Data Protection – Quantitative Survey and Recommendations. Orphanet Journal of Rare Diseases, 14, 1-15. https://doi.org/10.1186/s13023-019-1123-4.
Cumyn A, Barton A, Dault R, Cloutier A-M, Jalbert R, Ethier J-F (2019). Informed consent within a learning health system: a scoping review. Learn Health Syst. https://doi.org/10.1002/lrh2.10206
Dimitropoulos, L., Patel, V., Scheffler, S. A. & Posnack., S. (2011). Public Attitudes toward Health Information Exchange: Perceived Benefits and Concerns. The American Journal of Managed Care, 17, SP111–SP116. PMID: 22216769.
Duckett, S. (2018). Case Study: What Can We Learn from Australia’s My Health Record Experience?. Philips. Retrieved from https://www.philips.com/a-w/about/news/archive/future-healthindex/articles/20181107-case-study-what-can-we-learn-from-australias-my-health-recordexperience.html. Accessed August 25, 2022.
Edelstein, M., Lee, L. M., Herten-Crabb, A., Heymann, D. L. & Harper, D. R. (2018). Strengthening Global Public Health Surveillance through Data and Benefit Sharing. Emerging Infectious Diseases, 24, 1324–1330.
Esmaeilzadeh P, Sambasivan M. (2017). Patients’ Support for Health Information Exchange: A Literature Review and Classification of Key Factors. BMC Medical Informatics and Decision Making, 17, Article 33. https/doi.org/10.1186/s12911-017-0436-2.
Ghafur, S., Van Dael, J., Leis, M., Darzi, A. & Sheikh, A. (2020). Public Perceptions on Data Sharing: Key Insights from the UK and the USA. The Lancet Digital Health, 2, 444-446. https://doi.org/10.1016/S2589-7500(20)30161-8.
Gostin L. O., Halabi S. F., Wilson K. (2018). Health Data and Privacy in the Digital Era. JAMA, 320, 233-234. https/doi.org/10.1001/jama.2018.8374.
Grande D., Mitra, N., Shah, A., Wan, F. & Asch, D. A. (2013). Public Preferences about Secondary Uses of Electronic Health Information. JAMA International Medicine, 173, 1798–1806. https://doi.org/10.1001/jamainternmed.2013.9166.
Gruson, D. (2019). « Le numérique et l’intelligence artificielle en santé : surveillance généralisée ou avancée majeure ? » Les Tribunes De La Santé, 60, 23–29. https://doi.org/10.3917/seve1.060.0023.
Healthwatch England (2018). How do people feel about their data being shared with the NHS. https://www.healthwatch.co.uk/report/2018-05-17/how-do-people-feel-about-their-data-being-shared-nhs. Accessed 26 August, 2022.
Helou, S., Abou-Khalil, V., El Helou, E. & Kiyono, K (2021). Factors Related to Personal Health Data Sharing: Data Usefulness, Sensitivity and Anonymity. Studies in Health Technology Information, 281, 1051-1055. https://doi.org/10.3233/SHTI210345.
Hutchings, E., Loomes, M., Butow, P. & Boyle F. M. (2020). A Systematic Literature Review of Health Consumer Attitudes towards Secondary Use and Sharing of Health Administrative and Clinical Trial Data: A Focus on Privacy, Trust, and Transparency. Systematic Reviews 9, Article 235. https://doi.org/10.1186/s13643-020-01481-9.
Huston P., Edge V. L. & Bernier, E. (2019). Tirer profit des données ouvertes en santé publique. Relevé des maladies transmissibles au Canada, 45, 277–282.
Jones L. A., Nelder J. R., Fryer J. M., et al. (2022). Public Opinion on Sharing Data from Health Services for Clinical and Research Purposes without Explicit Consent: An Anonymous Online Survey in the UK. BMJ Open,12, e057579. https://doi.org/10.1136/ bmjopen-2021-057579.
Joseph, C. L. M., Tang, A., Chesla, D. W., Epstein, M. M., Pawloski, P. A., Stevens, A. B., Waring, S. C., Ahmedani, B. K., Johnson, C. C. & Peltz-Rauchman, C. D. (2022). Demographic Differences in Willingness to Share Electronic Health Records in the All of Us Research Program, Journal of the American Medical Informatics Association, Volume 29, 1271–1278. https://doi.org/10.1093/jamia/ocac055.
Kalkman S., van Delden J., Banerjee A., Tyl B., Mostert M. & van Thiel G. (2019). Patients’ and Public Views and Attitudes towards the Sharing of Health Data for Research: A Narrative Review of the Empirical Evidence. Journal of Medical Ethics, 48, 3-13. https://doi.org/10.1136/medethics-2019-105651.
Kim, K. K., Sankar, P., Wilson, M. D. & Haynes, S. C. (2017). Factors Affecting Willingness to Share Electronic Health Data among California Consumers. BMC Medical Ethics, 18, 1–10. https://doi.org/10.1186/s12910-017-0185-x.
King, R. J., Garrett, N., Kriseman, J., Crum, M., Rafalski, E. M., Sweat, D., Frazier, R., Schearer, S. & Cutts, T. (2016). A Community Health Record: Improving Health Through Multisector Collaboration. Information Sharing, and Technology, 13, 1-9. https://doi.org/ 10.5888/pcd13.160101.
Kostkova, P., Brewer, H. de Lusignan, S. Fottrell, E. Goldacre, B., Hart, G., Koczan, P., Knight, P., Marsolier, C., McKendry, R. A., Ross, E., Sasse, A., Sullivan, R., Chaytor, S., Stevenson, O., Velho, R. & Tooke, J. (2016). Who Owns the Data? Open Data for Healthcare. Frontiers in public health, 4, Article 7. https://doi.org/ 10.3389/fpubh.2016.00007.
Lane, J. & Shur, C. (2020). Balancing Access to Health Data and Privacy: A Review of the Issues and Approaches for the Future. Health Services Research, 45, 1456–1467. https://doi.org/10.1111/j.1475-6773.2010.01141.x.
Luhmann, N. (2017). Trust and Power. Wiley.
Lupton, D. (2019). ‘I’d like to think you could trust the government, but I don’t really think we can’: Australian Women’s Attitudes to and Experiences of My Health Record. Digital Health. January 2019. https://doi.org/10.1177/2055207619847017.
Mabillard, V. et Caron, D.J. (2022). Plus de transparence, plus de confiance ? Regard critique sur un principe clé de gouvernance et ses attentes. Administration publique du Canada, 65, 482-496. https://doi.org/10.1111/capa.12487.
McCormick N., Hamilton C. B., Koehn C. L., English K., Stordy A., Li L. C. (2019). Canadians’ Views on the Use of Routinely Collected Data in Health Research: A Patient-Oriented Cross-Sectional Survey. CMAJ Open, 7, E203-E209. https://doi.org/10.9778/cmajo.20180105.
Moon, L. (2017). Factors Influencing Health Data Sharing Preferences of Consumers: A Critical Review. Health Policy and Technology, 6, 169–187. https://doi.org/10.1016/j.hlpt.2017.01.001.
Nutley, T. & Reynolds, H. W. (2013). Improving the Use of Health Data for Health System sSrengthening. Global Health Action, 6, 1-10. https://doi.org/10.3402/gha.v6i0.20001.
O’Donnell, H. C., Patel, V. Kern, L. M., Barrón, Y., Teixeira, P., Dhopeshwarkar R. & Kaushal, R. (2011). Healthcare Consumers’ Attitudes towards Physician and Personal Use of Health Information Exchange. Journal of General Internal Medicine, 26, 1019–1026.
Ostherr K., Borodina S., Bracken R. C., Lotterman C., Storer E. & Williams B. (2017). Trust and Privacy in the Context of User-Generated Health Data. Big Data & Society, 4. https://doi.org/10.1177/2053951717704673.
Pisani, E., Aaby, P., Breugelmans, G., Carr, D., Groves, T., Helinski, M. & al. (2016). Beyond Open Data: Realising the Health Benefits of Sharing Data. BMJ (Clinical Research Ed.), 355, 1-5. https://doi.org/10.1136/bmj.i5295.
Pisani, E. & AbouZahr, C. (2010). Sharing Health Data: Good Intentions Are Not Enough. Bulletin of the World Health Organization, 88, 462–466. https://doi.org/ 10.2471/BLT.09.074393.
Platt, J. E., Jacobson, P. D. & Kardia, S. L. R. (2018). Public Trust in Health Information Sharing: a measure of system trust. Health Services Research 53, 824–845. https://doi.org/10.1111/1475-6773.12654.
Rivas Velarde, M.C., Tsantoulis, P., Burton-Jeangros, C. & al. (2021). Citizens’ Views on Sharing their Health Data: The Role of Competence, Reliability and Pursuing the Common Good. BMC Medical Ethics 22, Article 62 (2021). https://doi.org/10.1186/s12910-021-00633-3.
Rolf von den Baumen, T., Lake, J., Everall, A. C., Dainty, K., Rosenberg-Yunger, Z., & Guilcher, S. J. T. (2020). “Clearly they are in the circle of care, but . . .”: A Qualitative Study Exploring Perceptions of Personal Health Information Sharing with Community Pharmacists in an Integrated Care Model. Canadian Pharmacists Journal / Revue des pharmaciens du Canada, 153, 378–398. https://doi.org/10.1177/1715163520956686.
Shah, N., Coathup, V., Teare, H., Forgie, I., Giordano, G. N., Hansen, T. H., Groeneveld, L., Hudson, M., Pearson, E., Ruetten, H. & Kaye, J. (2019). Sharing Data for Future Research—Engaging Participants’ Views about Data Governance beyond the Original Project: a DIRECT Study. Genetics in Medecine, 21, 1131-1138. https:// 10.1038/s41436-018-0299-7.
Shah, T., Wilson, L., Booth, N., Butters, O., McDonald, J., Common, K., Martin, M., Minion, J., Burton, P. & Murtagh, M. (2019). Information-Sharing in Health and Social Care: Lessons from a Socio-Technical Initiative. Public Money & Management, 39, 359-363. https://doi.org/10.1080/09540962.2019.1583891
Simon, S. R., Stewart Evans, J., Benjamin, A., Delano, D. & Bates, D. W. (2009). Patients’ Attitudes toward Electronic Health Information Exchange: Qualitative Study. Journal of Medical Internet Research, 11, e30. https://doi.org 10.2196/jmir.1164.
Sterckx S., Rakic V., Cockbain J., Borry P. (2016). “You hoped we would sleep walk into accepting the collection of our data”: controversies surrounding the UK care.data scheme and their wider relevance for biomedical research. Medical Health Care Philosophy, 19, 177-90. https://doi.org/10.1007/s11019-015-9661-6.
Stockdale J., Cassell J. & Ford E. (2018). “Giving something back”: A systematic review and ethical enquiry into public views on the use of patient data for research in the United Kingdom and the Republic of Ireland. Wellcome Open Research, 3, Article 6. https://doi.org/10.12688/wellcomeopenres.13531.2.
Summers C., Griffiths F., Cave J., Panesar A. (2022). Understanding the Security and Privacy Concerns About the Use of Identifiable Health Data in the Context of the COVID-19 Pandemic: Survey Study of Public Attitudes Toward COVID-19 and Data-Sharing. JMIR Formative Research, 6, e29337. https://doi.org/10.2196/29337.
Tosoni, S., Voruganti, I., Lajkosz, K. et al. (2022). Patient Consent Preferences on Sharing Personal Health Information during the COVID-19 Pandemic: “the more informed we are, the more likely we are to help”. BMC Medical Ethics, 23, Article 53. https://doi.org/10.1186/s12910-022-00790-z.
Understanding Patient Data (2018). Public Attitudes to Patient Data Use. A Summary of Existing Research. https://understandingpatientdata.org.uk/sites/default/files/2019-05/Public%20attitudes%20key%20themes%200.pdf. Accessed on August 25, 2022.
Van Panhuis, W. G., Proma, P., Emerson, C., Grefenstette, J., Wilder, R., Herbst, A. J., Heymann D. & Burke, D. S. (2014). A Systematic Review of Barriers to Data Sharing in Public Health. BMC Public Health, 14, 1–9. https://doi.org/10.1186/1471-2458-14-1144.
Webster, Paul. (2020). Canadian Digital Health Data Breaches: Time for Reform. The Lancet: Digital Health, 2, e113-e114. https://doi.org/10.1016/S2589-7500(20)30030-3.
Woldaregay, A. Z., Henriksen, A., Issom, D.-Z., Pfuhl, G. Sato, K., Richard, A., Lovis, C., Årsand, E., Rochat, J. & Hartvigsen, G. (2020). User Expectations and Willingness to Share Self-Collected Health Data. Studies in Health Technology and Informatics, 270, 894–898. https://doi.org/10.3233/SHTI200290.
