General consent: a mixed-method and patient-inclusive approach to explore and optimise information strategies

DOI: https://doi.org/https://doi.org/10.57187/4812

Introduction

General consent for research is a form of informed consent by which individuals grant permission for the secondary use of their health-related personal data and biological samples in yet-to-be-defined research projects. General consent covers all data and samples collected during hospital visits and stays (both before and after accepting and signing the consent form), but precludes the collection of additional samples for research for which additional and specific informed consent is necessary. Even though researchers are legally required to provide participants with information and consent documents that are clear and understandable, these documents are often lengthy and complex, hindering patients’ understanding, informed choice and recall of their decision [1–6]. General consent has the particularity that future studies are not yet defined, and therefore the information provided at the time of the consent request can only be generic.

General consent has been implemented at Lausanne University Hospital (CHUV) since November 2012 [7] and in the other Swiss university hospitals since 2014, as a consequence of the entry into force of the Federal Act on Research involving Human Beings (Human Research Act / SR 810.30) [8]. At CHUV, the current general consent form includes two questions: the first is about secondary use of health-related personal data and biological samples, the second is about participation in the genomic biobank of the hospital (supplementary document 3 in the appendix). Regardless of its evolution over time, the aim of general consent at CHUV has remained the same, namely to facilitate research, while providing patients with sufficient clear information to allow them to make an informed decision. However, the modes of general consent administration and the forms used to document patients’ general consent have undergone several modifications.

There is evidence that medical and demographic factors influence the way patients respond to general consent [9], including some ambiguous findings. In an early analysis conducted in 2017, CHUV reported that the consent rate was declining progressively with age. Moreover, women, patients born outside Switzerland and patients admitted through the emergency ward were less likely to consent [7]. More recently, Zurich University Hospital also observed that women are less likely to grant general consent, whereas age, a higher frequency of hospital visits and multiple diagnoses were associated with an increased decision rate [10]. Similar results were reported in Geneva University Hospital: people more prone to grant general consent were older, had more comorbidities, were of higher socioeconomic status and were more often of Swiss nationality [11]. We observed that the number of patients informed about general consent and response rates also fluctuated depending on the modes of administration. However, overall, evidence regarding the impact of the mode of general consent administration on patients’ decision-making is limited [12]. Our study aims to fill this knowledge gap and illustrate the influence of the clinical context.

The first objective of the study was to identify the factors impacting the response and the rate of general consent, in particular the influence of general consent administration modes implemented at CHUV between December 2012 and April 2022. The second objective was to evaluate patients’ understanding, preferences, needs and expectations concerning the content, format and administration of general consent. Overall, the results of this study provide a solid base on which pragmatic recommendations can be drawn on how to optimise general consent content and how the information is administered.

Materials and methods

This was a mixed-methods study, including quantitative and qualitative analysis.

Data sources

The evaluation of factors impacting the response and consent rate to general consent relies on administrative data and general consent-associated data from 178,223 adult patients informed about general consent, collected at CHUV from December 2012 to April 2022. Data were extracted from three existing institutional administrative databases by the CHUV IT division team. During this period, general consent was gradually implemented in adult hospital departments, culminating in 2019 with systematic general consent information provided to adult inpatients and outpatients, except for those visiting the Psychiatric Department. A single research database was created after an iterative process of data extractions and data management, a check of data consistency and removal of duplicates. Random spot checks were undertaken to ensure data quality after this process. Before the statistical analysis, it became clear that certain variables had a very large number of categories, some of which had very few observations. As a result, the categories of certain variables were grouped. Subsequently, during logistic regression modelling, several observations had to be excluded and variables dropped in order for model assumptions to be respected and to ensure that the models could converge. In particular, all variable categories with fewer than 50 observations were dropped as the odds ratios for these categories were very unstable.

The database created for this study will be kept on CHUV’s secure server for 20 years after the publication of the study, then destroyed.

The assessment of patients’ understanding, preferences, needs and expectations regarding general consent is based on data collected via three channels: 2445 anonymous questionnaires, 32 semi-structured interviews and 4737 patient communications:

Questionnaire: A 36-item 6-page questionnaire was developed by the study team. It was sub-divided into 3 sections: demographic information; self-assessed comprehension; and needs and expectations relating to the information (supplementary document 1 in the appendix). Question types were multiple-choice, scale and open-ended. During a 3-month period (1 March 2022 – 3 June 2022), the paper questionnaire was added to the general consent documents that are systematically mailed to adult patients who visit CHUV. An electronic form was also available via a QR code. The French version of the questionnaire was mailed and various languages were available upon request (German, Italian, English, Spanish and Portuguese). Only the English version was requested. The questionnaire explored respondents’ understanding as well as their needs and expectations regarding the content, format and administration of general consent. Data from the paper questionnaires were manually entered into an electronic REDCap® database. The questionnaire was anonymous.

Semi-structured interviews: Given that individuals not returning a general consent decision (non-responders) were very unlikely to answer the questionnaire, a random sample of individuals who did not fill in the general consent form after the reminder was selected based on their age, sex and whether they received general consent information before or after the hospital visit/stay. These individuals were invited to participate, on a voluntary basis, in a semi-structured phone interview during which researchers probed participants about their recall of the information received, their understanding, and their needs and expectations about this information. In total, 32 interviews were conducted in July and August 2022 and in May and June 2023, with 15 women (6 aged 18–42, 9 aged 43–67) and 17 men (6 aged 18–42, 10 aged 43–67 and 1 aged 68–90). Discussions were conducted according to the interview guide (supplementary document 2 in the appendix), recorded, transcribed verbatim and anonymised. All recordings were destroyed and the transcripts were archived.

Patient communications (hotline phone calls, e-mails and patient letters) received at the Research Consent Unit are systematically summarised in a dedicated database.We extracted the notes taken during the period from 15 May 2020 to 7 March 2023, and analysed them.

Data analysis

Quantitative analyses

Quantitative analyses were conducted on the retrospective dataset (December 2012 – April 2022) to identify factors impacting response and decision about general consent using logistic mixed-effect regression models.

The modalities of general consent administration were as follows:

• Modality 1: general consent oral presentation by staff dedicated to general consent (Dec 2012 – Oct 2016),
• Modality 2: general consent oral presentation by medical doctors, nurses or administrative staff within the wards (Since Sep 2014),
• Modality 3: Information by mail before visit/hospitalisation, with or without reminder letter (Since Oct 2016),
• Modality 4: Information by mail after visit/hospitalisation, with or without reminder letter (Since Oct 2016).

General consent modality, version, status for information reiteration (once or more than once), age, sex, marital status, religious affiliation, country of citizenship, type of visit (outpatient, inpatient or emergency), reason for care (illness or accident) were introduced as fixed effects, and medical units as a random effect. After relevant groupings, small categories (fewer than 50 patients) were excluded (corresponding to 315 participants). Forest plots of odds ratios and 95% confidence intervals are presented. It was assumed that the effect of each variable is the same across all categories. We assumed that there are no interactions between the different variables in the model. No correction for multiplicity of tests was applied. Due to the large number of variables and for readability reasons, only those relevant to the discussion are displayed in figure 1. All analysed variables are available in figure S1 in the appendix. Missing values were considered to occur completely at random.

Figure 1Logistic regression of response rate among informed patients (A) and consent rate among responders (B). The number of values for each category is shown in brackets. References are visible on the figure with an odds ratio value of 1 and no confidential interval. Response rate was modelled among informed patients and consent rate was modelled among patients who responded to general consent. Modality 1: General consent oral presentation by staff dedicated to general consent; Modality 2: General consent oral presentation by medical doctors, nurses or administrative staff within wards; Modality 3: Information by mail before visit/hospitalisation, with or without reminder letter; Modality 4: Information by mail after visit/hospitalisation, with or without reminder letter.

Descriptive statistics were performed to present the results of the questionnaire, applied to answers to queries and Likert scale data; mainly counts and percentages. All quantitative statistical analyses were performed with R 4.1.3 and 4.5.1. No weighting for underrepresentation of certain groups of patients was applied.

Qualitative analyses

Qualitative data collected via the questionnaire, semi-structured interviews and patient communications were analysed using the thematic content analysis method of Braun V and Clarke V [13], chosen for its flexibility and suitability in capturing complex, subjective experiences. This method entails an iterative process in which two researchers independently familiarise themselves with the data prior to coding them to identify interesting recurring content. To enhance the credibility and trustworthiness of the findings, discrepancies between coders were resolved through discussion and consensus. Data coding and analysis were performed using MAXQDA (Analytics Pro 2022) to help organise data and facilitate interdisciplinary exchange between researchers. In the Results section, the number of occurrences per theme are indicated as well as the total number of occurrences for one question. The total number of occurrences excludes answers with no relevance to the question. This clarification is provided to ensure methodological transparency and prevent any misinterpretation of the results.

Ethics approval and consent to participate

This study was conducted in compliance with the Helsinki Declaration. According to its aims, the study is a quality project and so falls outside the scope of the Swiss research legislation (the Human Research Act), which applies to research concerning human diseases, and the structure and function of the human body (see article 2). As such, Swiss research legislation and institutional guidelines waive the requirement for approval by an Ethics Committee.

Furthermore, to process this monitoring study, the dataset constituted with mainly administrative data of patients was anonymised to comply with the applicable data protection legislation. Given the type of study and the anonymisation of data, patient consent was not required. Data concerning patients’ understanding and opinions about general consent documents and modalities were collected on a voluntary basis through anonymised questionnaires and interviews.

The institutional Legal Department formally confirmed that this study is exempt from Ethics Committee authorisation in compliance with Swiss legislation and institutional guidelines, as well as from patient consent.

Nonetheless, before starting, we informed the cantonal ethics committee for research involving humans (CER-VD) about this study.

Results

Patient characteristics

Sample for the evaluation of factors impacting the response and consent rate to general consent

To be able to determine the factors influencing the response and consent rate, a database was created as described in “Materials and methods”. Several observations were excluded, as shown in the patient selection chart (figure 2). This results from complexities associated with how data were recorded and anomalies in the information process. In total 36,712 (17%) observations were excluded from the initial extraction file (n = 214,935) for various reasons, resulting in the final database containing 178,223 observations. The descriptive analysis of the database is available in table S1 in the appendix.

Figure 2Patient selection chart used to create a single research database containing 178,223 observations for the quantitative analysis. In total, 36,712 (17%) observations were excluded resulting from data cleaning from the initial extraction file (n = 214,935); exclusion was linked to various reasons resulting in a database. A summary of the reasons for exclusion of variables is provided in the flowchart.

Questionnaire sample

The questionnaire was mailed to 13,798 adult patients of CHUV between 1 March and 3 June 2022. The research team accepted returned questionnaires (paper 92% and electronic 8%) until 3 August 2022. During this period, 2616 questionnaires were returned (response rate: 19%). In comparison, the response rate to general consent forms over the same period (i.e. individuals who provided a response to the general consent request) was 46%. Among the returned questionnaires, 6.5% (171) had to be excluded (incomplete electronic records [n = 86]; only the administrative part of the questionnaire completed [n = 75]; fewer than 5 items of the questionnaire filled in [n = 9]; questionnaire completed by a minor [n = 1]), which resulted in the inclusion in the analysis of 2445 questionnaires of different levels of completeness (providing a consolidated response rate of 18%). Among questionnaire respondents, 36% identified as male, 48% as female and 16% did not specify sex. The average age was 51 and most questionnaire respondents were identified as patients (80%). The descriptive table of questionnaire respondents as well as questionnaire results are extensively reported in tables S2 and S5 in the appendix.

Based on the responses to question 8 of the questionnaire (How did you fill in/do you think to fill in general consent form), we observed that patients who stated in the questionnaire that they responded to or intended to respond to the general consent request were overrepresented. This overrepresentation was particularly pronounced for individuals who stated that they would agree to the secondary use of their health-related personal data and biological material: 70% of the questionnaire respondents indicated that they intended to or had consented while the observed actual consent rate among those informed about general consent in the same period is 34%. In contrast, questionnaire respondents who stated they did not intend to respond to general consent (2% of questionnaire respondents) or would refuse (5% of questionnaire respondents) were drastically underrepresented, when compared to the population who received the questionnaire accompanying general consent documents (54% non-respondents and 11% refusers) (table S6). Due to this lack of representation of non-respondents, a random sample of individuals who did not fill in the general consent form was selected and interviewed, as described in the “Materials and methods” section. Although individuals who declined general consent were underrepresented among questionnaire respondents, engaging this group was not feasible, as it would have required contacting those who had refused general consent.

Factors influencing response and consent rate

The quantitative analyses conducted on factors influencing patients’ response revealed that most variables related to general consent administration, to hospital visit/stay and to patient demographics impact the odds for responding to general consent and consenting to secondary use of health-related personal data and biological samples (figure S1, tables S3 and S4 in the appendix).

The way general consent information is administered is associated with various response and consent rates. In-person oral presentation by staff of the unit/ward (modality 2) is associated with the highest odds for responding and consenting to the secondary use of health-related personal data and biological samples (compared to modality 1: oral presentation by dedicated staff): adjusted odds ratio (aOR) = 4.44, p <0.001, for responding and aOR = 1.58, p <0.001) for consenting. In contrast, written information sent by mail before (modality 3) or after (modality 4) the visit/hospitalisation is associated with the lowest odds for responding and consenting: modality 3 vs modality 1, aOR = 0.12, p <0.001 for responding and aOR = 0.92, p = 0.018 for consenting; and modality 4 vs modality 1, aOR = 0.07, p <0.001 for responding and aOR = 0.89, p <0.001 for consenting (figure 2A illustrates the logistic regression of response rate among informed patients, and figure 2B illustrates consent rates among responders).

Inpatients are more likely to respond to general consent than outpatients (aOR = 1.10, p = 0.001) and emergency patients (aOR = 0.87, p <0.01); however, outpatients are the most likely to grant general consent. Emergency patients are less prone to respond and grant general consent. This is related to the fact that patients whose type of visit/stay is categorised as “illness” are more prone to respond to and grant general consent than those categorised as “accident” (aOR = 1.23, p <0.001 and aOR = 1.24, p <0.01, respectively) since these patients often enter via the emergency room.

Compared to patients informed once (with the first mailing +/- the reminder letter), those informed multiple times are less prone to respond (aOR = 0.41, p <0.02) and grant general consent (aOR = 0.83, p <0.001). Patients were informed multiple times because either they did not respond before April 2019 and were informed again when the new general consent management application was implemented or they did not reply to the first version of general consent and were informed again when the version changed. The same trend in response and consent rate is observed when analysing the effect of the reminder letter. During the period from January 2018 to April 2020, 39% of patients made their decision after receiving the information by a single mail. In May 2020, a reminder letter was introduced and sent 6 weeks after the initial information letter if the patient had not responded. During the following period, 55% of patients made their decision when receiving general consent information by mail, which suggests that despite a lower overall consent rate in responders during the latter period (76% vs 81%), the reminder mailing was followed by a 16% increase in overall response rate among informed patients.

Patients’ self-assessed comprehension of general consent information

Using a series of 12 statements (questions 9–10 and 14–23; questions 11–13 concerning the genomic biobank are not discussed in this article), questionnaire respondents were asked to indicate the extent to which they understood the information provided on a Likert scale of 1 “I didn’t understand” to 5 “I understood”. Most questionnaire respondents (60%, n = 1454) indicated that they understood all these statements, rating a score of 5 for each statement. Similarly, when asked on a scale of 1 to 10 whether they easily understood the general consent documents (Q25), 89% (n = 2060) of respondents gave a rating greater than 8, with 59.7% (n = 1378) giving the maximal score (table S5 in the appendix). An analysis to compare the understanding of acceptors’, refusers’ and non-responders’ groups was not statistically possible, since the questionnaire return rate of the non-responders and refusers was too low.

While most questionnaire respondents self-assessed complete understanding of general consent information, we identified some topics with varying levels of understanding (figure 3). The greatest variability in understanding is related to (i) the use of health-related personal data and biological material collected in the past and the future, (ii) the use of samples and data by public and private entities, and (iii) the use of samples and data in Switzerland and abroad. The possibility that, in cases of non-response, health-related personal data and biological samples may exceptionally be used for research under Article 34 of the Swiss Human Research Act – outlining how permission to use data and samples from non-respondents can be granted – remains a subject for which patients’ and questionnaire respondents’ understanding is limited.

Figure 3Self-assessed comprehension of general consent information. Using a series of 12 statements (questions 9–10 and 14–23; questions 11–13 concerning the genomic biobank are not illustrated in this figure), questionnaire respondents were asked to indicate the extent to which they understood the information provided on a Likert scale of 1 “I didn’t understand” to 5 “I understood”. Ratings of people who did not rate 5 to all comprehension questions are illustrated (n = 991 patients). “Missing” means no response.

Examination of the 4737 patient communications data revealed several recurring questions from patients: 350 people asked why they had received this consent form, 320 needed help understanding the information and completing the form and 220 asked whether they had to complete the general consent form considering they felt too old, too sick or misunderstood the scope of the consent. Individual interviews confirmed that the reason why a patient receives general consent needs to be better explained as well as the existence of a provision whereby data may still be used in specific circumstances in case of non-response (i.e. that under Article 34 of the Swiss Human Research Act, permission to use the data may still be granted in some cases).

Main sources of concern for patients

When questionnaire respondents were asked whether they were concerned about some aspects of general consent information (Q31), the majority responded that they were not (91.4%, n = 2149) (table S5). Among respondents who indicated having concerns (8.6%, n = 203), 3 themes recurred in the comments. The first concern was the use of health-related personal data and biological material in the private sector (59/175 occurrences). The second concern was the fact that biological samples and health-related data could be shared with institutions in a foreign country (32/175 occurrences). The third concern was data protection (how data are stored, the level of protection, the risk of hacking, …) with 25/175 occurrences.

Preferences about general consent information modalities and timing

When asked about the best way to receive information about general consent (Q29), 73.3% of the questionnaire respondents (n = 1677) stated that the printed materials, the website and support options offered by the hospital are sufficient to make an informed decision, while 26.7% (n = 611) of respondents would appreciate an oral presentation. When asked who should provide oral information, the preference goes to a medical doctor (38.2%), followed by a general consent team member (26.6%), a nurse (25.9%) and, lastly, a member of the admission office (14.3%) (table S5).

Concerning the best moment to receive general consent information (Q33), opinions vary. We observe a slight preference for the option “during the stay or visit” compared to “after the stay or visit” and “before the stay or visit” (33.3% of questionnaire respondents vs 29.8% and 24.5%, respectively) (table S5). When examining the reasons why respondents prefer to receive the information during their stay or visit, the main motivations are that it is the best moment to ask/clarify certain points with hospital staff (98/227 occurrences) and that patients have time while waiting at the hospital (77/227 occurrences).

Evaluation of the layout and presentation of general consent information

Questionnaire respondents were asked to rate the presentation and layout of general consent information on a scale of 1 to 10 (Q24). The majority rated the presentation and layout positively, with 88% using scores of 8 or more (n = 2038).

Respondents were also probed on whether they felt any information was missing from the documentation (Q26). The overwhelming majority, 93.1% (n = 2137), considered that there was no information missing from the documentation. Among respondents indicating that information was missing, comments were made on the type of research to be conducted with their health-related personal data and biological material, as well as on the need to show examples of research being done with general consent (31/110 occurrences). Some respondents mentioned that they would like to have more information on research being done in the private sector and abroad (15/110 occurrences). Details on how health-related data and biological samples are protected would also be appreciated (14/110 occurrences).

During individual interviews, several suggestions for improvement were raised, such as simplifying the information to make it more understandable and accessible, illustrating the importance of general consent by referring to concrete examples of research projects carried out within the framework of general consent and clarifying the consequences of not responding to general consent.

Most questionnaire respondents (94.4%, n = 2241) stated that general consent documents were sufficient to make a decision (Q28). Of those who needed more information (n = 132), 45.4% (n = 60) indicated that they wanted to discuss the topic with a relative and 20.5% (n = 27) wanted to discuss it with their doctor or nurse (table S5).

Willingness to participate in research and overall satisfaction with information received

When asked to rate the importance of contributing to research on a scale from 1 “Not at all” to 10 “It’s essential” (Q34), an overwhelming majority of questionnaire respondents (89.1%, n = 2068) answered towards the positive end of the scale (greater than 8).

When asked why it is important to contribute to research, the two most prevalent reasons were civic duties (567/647 occurrences) and a personal involvement in medical research (student, physician, researcher in the family, …) (47/647 occurrences).

88% (n = 2044) of the respondents were satisfied with the information received on general consent, rating satisfaction at more than 8 on a scale from 1 to 10 (Q35). Two questions sought suggestions on enhancing the clarity and comprehensibility of the documents (Q27) and improving how patients are informed about general consent (Q36). Among 406 suggestions, 130 emphasised the need to simplify the documents and make them more concise. While 86 respondents would like to have oral information during their visit/stay at the hospital and have human contact, 34 respondents were interested in a more electronic process. In line with observations reported above, 19 respondents emphasised the importance of providing examples of research conducted within the framework of the general consent.

Discussion

Factors predicting response and consent

Our quantitative analyses spanning over a decade of general consent practice at a university tertiary hospital reveal that information modalities, as well as the context of the patient’s visit, influence response and consent rates.

The quantitative analyses highlight that oral information provided by hospital unit staff during the patient’s stay/visit is associated with the highest likelihood of responding to and granting general consent. The questionnaire data also show a slight preference for receiving general consent information during the hospital stay/visit as it provides an opportunity to ask questions, and because patients often have waiting time in hospital that could be used wisely to be informed about general consent. These results are aligned with previous studies demonstrating that patients’ preferences regarding the use of their samples and health-related data in research is influenced by relational considerations [4].

Patients whose visit/stay is classified as emergency and patients whose visit/stay is related to an accident are less likely to grant general consent than patients presenting at the hospital for a scheduled appointment and patients whose visit is related to an illness. The unforeseen situation, concomitant stress and possibly a temporary incapacity to discern create an unfavourable context for making decisions regarding general consent. This may be a barrier to a representative engagement of patients in research conducted in critical care or emergency units. Repeating general consent information in a more favourable clinical context may increase the chances of obtaining a response to general consent request.

A decision to participate in research initiatives depends on the patients' trust towards the hospital and their staff [9, 14–16]. Over time, the trust relationship of a patient with the hospital may change, as well as the context of their care or the reason for the visit (transition from an acute episode to a chronic illness, repeated hospital stays, scheduled versus emergency stay, etc), their interest in research and general consent might change accordingly.

We also observed that the probability of responding and granting consent is influenced by the patient's sociodemographic profile (sex, age, religion, nationality and marital status), which will not be discussed in more detail as it has been described in several previous studies [7, 9–11]. It does, however, question the representativeness of people included in research projects conducted using data covered by general consent. As previously reported, the extent to which patients understand information and consent documents influences the response and consent rates [17]. Offering multiple ways to inform patients about general consent might be a solution to better reach the diverse patient profiles and increase the variety of patients responding to general consent. Several studies reported that users’ comprehension improved when an e-consent medium was used [18]. The potential of digital solutions that provide flexible and tailored information for patients, such as dynamic consent platforms [19], needs to be explored with the help of patients and researchers. However, digitisation of the consent process should be thought of as a complement to improve patient autonomy and understanding, not as a substitute for human interaction [20].

Patients’ preferences about general consent documents

Most questionnaire respondents stated that the printed material, website and support options offered (hotline and e-mail) are sufficient to help them make their decision. However, the analysis of the questionnaires, interviews and patient communications reveals that the information brochure should be more concise, easier to understand and more attractive. Examples of research made possible by general consent would also be greatly appreciated. Additionally, individual interviews showed that the reason why patients receive general consent and the consequences of not responding need to be better explained.

Furthermore, the questionnaire revealed that the questions concerning the use of health-related personal data and biological material by the private sector and/or abroad were also those that raised most concerns. As previously reported, the main concerns when deciding whether to grant general consent are lack of trust, data privacy and an objection to profit-making through the use of health-related data [21–24]. Emphasising the patient-centred benefit of this third-party use of health-related personal data and biological material, as well as better communication about the means used to protect privacy may help patients feel more comfortable about sharing their data with researchers from the private sector or abroad [25]. Furthermore, digital dynamic consent platforms offering transparency and more consent choices may offer patients greater confidence and peace of mind in their choices [24, 26].

We should keep in mind that, even though a random sample of individuals that did not fill in the general consent form was selected for interviews, most respondents in the three qualitative models were favourable towards general consent, meaning that the interpretation of the data discussed must be taken with caution.

Limitations of the study

The large sample size gives the results their strength, but it could also lead to some effects appearing statistically significant without representing an operationally relevant difference. In addition, data are based on administrative data extracted from hospital databases. Although efforts have been made to ensure that these data are collected accurately (through iterative quality checks), some inaccuracies may persist due to data entry errors, despite careful data consistency and data cleaning checks when deriving the analysis dataset.

The extended duration of data collection – over 10 years – cannot exclude a time effect. Intercurrent events, such as the SARS-CoV-2 pandemic, could have influenced perceptions of the healthcare system and biomedical research, and therefore the reception of information on general consent by patients [24] .

Questionnaire respondents displayed a recruitment bias: patients who responded to general consent were overrepresented, particularly those who granted general consent. In contrast, non-respondents to general consent and general consent refusers were severely underrepresented, something that cannot be compensated for by statistical methods. Furthermore, the questionnaire was distributed only via mailings to patients receiving general consent via modality 1, which may influence some responses. Therefore, the data collected via the questionnaire must be interpreted with caution as respondents’ profile reveals a specific population that is generally favourable towards research and general consent.

Conclusion

The results presented in this study are based on a substantial volume of information. These data allowed us to confirm the validity of several practices, draw lessons from more than ten years of experience in the field of general consent and propose four practical recommendations for general consent practice:

• Promote general consent information before or during the hospital visit or stay, whether in written or oral form.
• If a patient has not replied to the general consent request, repeat the information in a more favourable clinical context.
• Diversify information channels and explore digital options to offer more-personalised and adapted information.
• Simplify, clarify and illustrate general consent documentation with concrete examples of research carried out using general consent.

These recommendations should help increase the number of patients properly informed and therefore patients who complete the general consent form. Indeed, if modalities of general consent administration are adapted according to patients’ needs and preferences, and if the content and format of the information allow for better understanding of general consent, we assume that the response rate will increase and, more importantly, so will the rate of truly informed and autonomous decisions.

This study focused on patients’ perspectives on general consent information within a broader institutional approach to patient and public involvement and engagement. The relatively high response rate to the questionnaire suggests that many patients value the idea of improving understanding of the consent request and legitimise its informed and well-understood nature. It is essential to involve patients in the development of general consent and in the establishment of guidelines relating to its administration.

Availability of data and materials

The datasets generated and analysed during the current study were deposited under restricted access in the institutional repository of CHUV due to the sensitive nature of the data. Metadata are publicly available on Zenodo and accessible via the DOI provided in the dataset reference (https://doi.org/10.5281/zenodo.15683552). Access to the data may be granted upon reasonable request to the Principal Investigator and is subject to a data transfer and use agreement.

Acknowledgments

We would like to thank the following: Professor Mauro Oddo, Director of innovation and research at CHUV, for allocating institutional funding dedicated to Swiss personalized health network to the project; Léa Stiefel, research fellow, for individual interviews and qualitative analysis; Francesca Bosisio, Ordinary Professor at Haute École d'Ingénierie et de Gestion du Canton de Vaud (HEIG‑VD), for methodological support in developing the questionnaire and qualitative analyses; the Datawarehouse team of CHUV, particularly Laïla Nadich, for extracting and combining the data from IT systems; Éric Eboulet, statistician at CHUV Clinical Research Center, for support in creating the study database; Vassili Soumas, datamanager at CRC, for support in creating the questionnaire database; Pierre-Antoine Dubois, graphic designer at the communications service of CHUV, for the graphic design of the questionnaire and the layout of the document; Dominique Sprumont, President of the cantonal research ethics committee, for fruitful discussions; Maria Ramos and Amédée Kibalabala, research collaborators at the Research Consent Unit, for their support in recruiting participants for individual interviews; and finally all the patients who took time to answer the questionnaire and participate in interviews.

Author contributions: BS, MF and CA contributed to the conception and the supervision of the study. BS, MF, CA, NK and JZ were involved in designing the methodology. NK, JZ and MJC collected and processed the data. NK, JZ, MJC and MM analysed and interpreted the data. JZ and CA reviewed the literature. NK, CA, MM and BS wrote the first draft of the manuscript. All the authors were involved in the revision of the manuscript.

References

1. D'Abramo F. Biobank research, informed consent and society. Towards a new alliance? J Epidemiol Community Health. 2015;69(11):1125-8. Epub 20150210. doi: 10.1136/jech-2014-205215. PubMed PMID: 25669218. 

2. Anderson EE, Newman SB, Matthews AK. Improving informed consent: Stakeholder views. AJOB Empir Bioeth. 2017;8(3):178-88. Epub 20170802. doi: 10.1080/23294515.2017.1362488. PubMed PMID: 28949896; PubMed Central PMCID: PMCPMC5749407. 

3. Ballard LM, Horton RH, Dheensa S, Fenwick A, Lucassen AM. Exploring broad consent in the context of the 100,000 Genomes Project: a mixed methods study. Eur J Hum Genet. 2020;28(6):732-41. Epub 20200109. doi: 10.1038/s41431-019-0570-7. PubMed PMID: 31919452; PubMed Central PMCID: PMCPMC7253456. 

4. Barazzetti G, Bosisio F, Koutaissoff D, Spencer B. Broad consent in practice: lessons learned from a hospital-based biobank for prospective research on genomic and medical data. Eur J Hum Genet. 2020;28(7):915-24. Epub 20200221. doi: 10.1038/s41431-020-0585-0. PubMed PMID: 32086443; PubMed Central PMCID: PMCPMC7316733. 

5. Bosisio F, Barazzetti G, Koutaissoff D, Spencer B. Patients' decision to contribute to a biobank in the light of the patient-recruiter relationship-a qualitative study of broad consent in a hospital setting. J Community Genet. 2021;12(1):15-25. Epub 20200810. doi: 10.1007/s12687-020-00479-z. PubMed PMID: 32779150; PubMed Central PMCID: PMCPMC7846645. 

6. Cheah PY, Jatupornpimol N, Hanboonkunupakarn B, Khirikoekkong N, Jittamala P, Pukrittayakamee S, et al. Challenges arising when seeking broad consent for health research data sharing: a qualitative study of perspectives in Thailand. BMC Med Ethics. 2018;19(1):86. Epub 20181107. doi: 10.1186/s12910-018-0326-x. PubMed PMID: 30404642; PubMed Central PMCID: PMCPMC6223036. 

7. Bochud M, Currat C, Chapatte L, Roth C, Mooser V. High participation rate among 25 721 patients with broad age range in a hospital-based research project involving whole-genome sequencing - the Lausanne Institutional Biobank. Swiss Med Wkly. 2017;147:w14528. Epub 20171020. doi: 10.4414/smw.2017.14528. PubMed PMID: 29063527. 

8. Humanforschungsgesetz. Die Bundesversammlung der Schweizerischen Eidgenossenschaft. Bundesgesetz über die Forschung am Menschen. https://www.admin.ch/opc/de/classified-compilation/20061313/index.html#2014 

9. Sanderson SC, Brothers KB, Mercaldo ND, Clayton EW, Antommaria AHM, Aufox SA, et al. Public Attitudes toward Consent and Data Sharing in Biobank Research: A Large Multi-site Experimental Survey in the US. Am J Hum Genet. 2017;100(3):414-27. Epub 20170209. doi: 10.1016/j.ajhg.2017.01.021. PubMed PMID: 28190457; PubMed Central PMCID: PMCPMC5339111. 

10. Griessbach A, Bauer A, Jorger Lebet F, Grossmann R. The concept of General Consent in Switzerland and the implementation at the University Hospital Zurich, a cross-sectional study. Swiss Med Wkly. 2022;152:w30159. Epub 20220414. doi: 10.4414/smw.2022.w30159. PubMed PMID: 35429239. doi: https://doi.org/10.4414/SMW.2022.w30159 

11.       Bosmani C, Carboni S, Samer C, Lovis C, Perneger T, Huttner A, et al. REPRESENT: REPresentativeness of RESearch data obtained through the 'General Informed ConsENT'. BMC Med Ethics. 2023;24(1):10. Epub 20230213. doi: 10.1186/s12910-022-00877-7. PubMed PMID: 36782161; PubMed Central PMCID: PMCPMC9926654. 

12. Gerber DE, Rasco DW, Skinner CS, Dowell JE, Yan J, Sayne JR, et al. Consent timing and experience: modifiable factors that may influence interest in clinical research. J Oncol Pract. 2012;8(2):91-6. Epub 20111206. doi: 10.1200/JOP.2011.000335. PubMed PMID: 23077435; PubMed Central PMCID: PMCPMC3457835. 

13. Braun VC. Thematic analysis. APA handbook of research methods in psychology, Vol 2: Research designs: Quantitative, qualitative, neuropsychological, and biological. psychology® Ahi, editor: American Psychological Association; 2012 2012. 14 p. doi: https://doi.org/10.1037/13620-004 

14. Nielsen MEJ, Kongsholm NCH. Blanket Consent and Trust in the Biobanking Context. J Bioeth Inq. 2022;19(4):613-23. Epub 20220906. doi: 10.1007/s11673-022-10208-5. PubMed PMID: 36068428. 

15. Hong SJ, Drake B, Goodman M, Kaphingst KA. Race, Trust in Doctors, Privacy Concerns, and Consent Preferences for Biobanks. Health Commun. 2020;35(10):1219-28. Epub 20190605. doi: 10.1080/10410236.2019.1623644. PubMed PMID: 31167570; PubMed Central PMCID: PMCPMC6893100. 

16. Grande D, Mitra N, Shah A, Wan F, Asch DA. Public preferences about secondary uses of electronic health information. JAMA Intern Med. 2013;173(19):1798-806. doi: 10.1001/jamainternmed.2013.9166. PubMed PMID: 23958803; PubMed Central PMCID: PMCPMC4083587. 

17. de Man Y, Wieland-Jorna Y, Torensma B, de Wit K, Francke AL, Oosterveld-Vlug MG, et al. Opt-In and Opt-Out Consent Procedures for the Reuse of Routinely Recorded Health Data in Scientific Research and Their Consequences for Consent Rate and Consent Bias: Systematic Review. J Med Internet Res. 2023;25:e42131. Epub 20230228. doi: 10.2196/42131. PubMed PMID: 36853745; PubMed Central PMCID: PMCPMC10015347. 

18. Kassam I, Ilkina D, Kemp J, Roble H, Carter-Langford A, Shen N. Patient Perspectives and Preferences for Consent in the Digital Health Context: State-of-the-art Literature Review. J Med Internet Res. 2023;25:e42507. Epub 20230210. doi: 10.2196/42507. PubMed PMID: 36763409; PubMed Central PMCID: PMCPMC9960046. 

19. Budin-Ljosne I, Teare HJ, Kaye J, Beck S, Bentzen HB, Caenazzo L, et al. Dynamic Consent: a potential solution to some of the challenges of modern biomedical research. BMC Med Ethics. 2017;18(1):4. Epub 20170125. doi: 10.1186/s12910-016-0162-9. PubMed PMID: 28122615; PubMed Central PMCID: PMCPMC5264333. 

20. Goldschmitt M, Gleim P, Mandelartz S, Kellmeyer P, Rigotti T. Digitalizing informed consent in healthcare: a scoping review. BMC Health Serv Res. 2025;25(1):893. Epub 20250702. doi: 10.1186/s12913-025-12964-7. PubMed PMID: 40604763; PubMed Central PMCID: PMCPMC12225439. 

21. Tosoni S, Voruganti I, Lajkosz K, Habal F, Murphy P, Wong RKS, et al. The use of personal health information outside the circle of care: consent preferences of patients from an academic health care institution. BMC Med Ethics. 2021;22(1):29. Epub 20210324. doi: 10.1186/s12910-021-00598-3. PubMed PMID: 33761938; PubMed Central PMCID: PMCPMC7992944. 

22. Ghafur S, Van Dael J, Leis M, Darzi A, Sheikh A. Public perceptions on data sharing: key insights from the UK and the USA. Lancet Digit Health. 2020;2(9):e444-e6. Epub 20200724. doi: 10.1016/S2589-7500(20)30161-8. PubMed PMID: 32838250; PubMed Central PMCID: PMCPMC7380931. 

23. Richter G, Borzikowsky C, Hoyer BF, Laudes M, Krawczak M. Secondary research use of personal medical data: patient attitudes towards data donation. BMC Med Ethics. 2021;22(1):164. Epub 20211215. doi: 10.1186/s12910-021-00728-x. PubMed PMID: 34911502; PubMed Central PMCID: PMCPMC8672332. 

24. Tosoni S, Voruganti I, Lajkosz K, Mustafa S, Phillips A, Kim SJ, et al. 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 Med Ethics. 2022;23(1):53. Epub 20220520. doi: 10.1186/s12910-022-00790-z. PubMed PMID: 35596210; PubMed Central PMCID: PMCPMC9122733. 

25. Trinidad MG, Platt J, Kardia SLR. The public's comfort with sharing health data with third-party commercial companies. Humanit Soc Sci Commun. 2020;7(1). Epub 20201111. doi: 10.1057/s41599-020-00641-5. PubMed PMID: 34337435; PubMed Central PMCID: PMCPMC8320359. 

26. Lee AR, Koo D, Kim IK, Lee E, Yoo S, Lee HY. Opportunities and challenges of a dynamic consent-based application: personalized options for personal health data sharing and utilization. BMC Med Ethics. 2024;25(1):92. Epub 20240831. doi: 10.1186/s12910-024-01091-3. PubMed PMID: 39217356; PubMed Central PMCID: PMCPMC11365279. 

Appendix

The appendix is available in the pdf version of the article at https://doi.org/10.57187/4812.