The research team developed 2 online questionnaires and tested them among colleagues from both academic and nonacademic backgrounds, as well as among individuals from their social circles. Technical details were worked out in collaboration with the digital agency that technically manages the website and also programmed both questionnaires. Visits to Embryotox drug fact sheets were counted via an anonymous session-tracking procedure by a commercial provider. To identify potential multiple entries, datasets of questionnaires were analyzed regarding entry times and similarity of entries. IP addresses or cookies were not used for this purpose. In accordance with the General Data Protection Regulation in Germany, questionnaires were designed in such a way that no conclusions could be drawn about individuals. Banner ads leading to the questionnaires were based directly on the drug fact sheets. The banner leading to the first questionnaire (2022) invited users to participate with the words “We ask for your feedback.” On the banner leading to the second questionnaire (2023), users were asked, “How do you use embryotox.de? Questionnaire about your experiences.” Participation was voluntary and not a prerequisite for using the drug fact sheets. According to their answers, participants were automatically guided through the questionnaire. Clicking back was not possible. Progress within the questionnaire was visible via a bar. Only one question at a time displayed on the screen, and participants could move on to the next question only after answering the previous question. Only the question about the educational qualifications of lay users was optional and could be skipped. The selection of mutually exclusive answer options and the selection of more than one answer (if not explicitly requested) were blocked. The answers were only saved if the users completed the questionnaire to the end and then selected the button “Send data.”

Online questionnaire 1 was placed on all drug fact sheets for 7 months (May-December 2022). Over the same period, the total number of visits to each fact sheet was counted. Users were asked to complete the questionnaire only once. Questionnaire 1 included questions on user characteristics, in particular demographic information (age and gender) and, for HCP, occupational information such as profession (for physicians also specialization) and working area (outpatient care, hospital, pharmacy, and other). Patients and other non-HCPs were asked about their level of education and how they got to know about Embryotox. HCPs were asked whether they were looking for information for a specific patient and whether they had found the essential information they needed to make a well-founded treatment decision. All users were asked to specify the frequency and duration of Embryotox use and the clinical focus of their search. Furthermore, all users were asked to evaluate the comprehensibility of the fact sheet they had accessed and whether its information resulted in changes in risk perception or preferred medication. Possible answers comprised either multiple-choice options or ratings on a Likert scale ranging from 0 (not at all) to 10 (fully). For a detailed flow chart of questionnaire 1, refer to Multimedia Appendix 2.

In addition, weekday, daytime, and the accessed fact sheet were recorded.

Questionnaire 2 was placed for 7 months (February-September 2023) on fact sheets for prescription-only drugs that typically require shared decision-making (refer to Table S3 in Multimedia Appendix 1). The questionnaire was aimed at the main user groups identified in questionnaire 1, primarily at physicians and patients as the main decision-makers. For the detailed flowchart, refer to Multimedia Appendix 3. The questionnaire was developed based on the findings from semistructured interviews on the functions of website use. For all user groups, answers comprised multiple-choice options or ratings on a Likert scale ranging from 0 to 10 or from “strongly agree” via “moderately agree” and “moderately disagree” to “strongly disagree.” If appropriate, “question not applicable” was also an answer option.

Questions and answer options were tailored to the respective user groups. Physicians answered up to 15 questions focusing on (1) their medical specialization, (2) typical use of fact sheets (eg, with or without patient, during or outside of consultation; maximum 3 out of 7 possible answers), (3) fact sheet use in the process of risk-benefit assessment and decision-making (eg, risk assessment and decision with or without patient, and satisfaction with fact sheet use), (4) further effects and functions of fact sheet use (eg, perceived effects on patient adherence and confidence in decision-making), and (5) potentially useful supplementary materials to the information provided on the website (maximum 2 out of 7 possible answers). Patients answered up to 16 questions about (1) fact sheet use for risk-benefit assessment and decision-making, (2) potential benefit or harm from the use of fact sheets, (3) effects and functions of fact sheet use, (4) potentially missing information, and (5) involvement of other persons in the decision-making process (multiple answers possible). Pharmacists and midwives answered up to 6 questions about the clinical situation, their typical fact sheet use (maximum 3 out of 5 possible answers) and functions of fact sheet use.

Of the 14,562 completed questionnaire 1, 10,860 were filled by patients and 1676 by physicians, the latter mostly obstetricians and gynecologists (27.2%, 456/1676) or general practitioners (22.7%, 381/1676). Details of participating user groups are presented in Table 1.

Overall, a median of 0.58% (IQR 0.42%‐0.77%) of visits to a drug fact sheet resulted in the completion of a questionnaire. The top 10 most frequently reviewed fact sheets are presented in Figures 1 and 2 for physicians and patients, respectively.

For user characteristics of HCP, refer to Table 2. The majority of participating patients were aged between 31 and 40 years (69.3%, 7525/10,860), most of them had a high level of education (university degree 58.2%, 6319/10,860; qualification for higher education 20.7%, 2247/10,860; school-leaving certificate 10th or 11th grade 17.4%, 1887/10,860; school-leaving certificate 9th grade 2.3%, 247/10,860; no educational qualification or not specified: 1.5%, 160/10,860). Figure 3 shows how patients became aware of Embryotox.

Drug fact sheets were accessed to a considerable extent outside standard working hours, that is, between 8 PM and 7:59 AM (physicians: 30.4%, 509/1676; patients: 42.9%, 4655/10,860; pharmacists: 28.4%, 156/550; midwives: 29.4%, 107/364). Most of the participating HCP stated that they had used the same fact sheet more than once (physicians: 73.3%, 1229/1676; pharmacists: 69.1%, 380/550; midwives: 77.5%, 282/364), as did the majority of patients (71%, 7708/10,860). At the time of fact sheet access, most of the participating patients were pregnant (46.8%, 5085/10,860) or were breastfeeding (42.8%, 4650/10,860). A total of 6% (649/10,860) of patients and 9.4% (131/1388) of physicians seeking information for a specific patient accessed Embryotox when pregnancy was still being planned. Among psychiatrists, the proportion was even higher (25.2%, 27/107). Details about fact sheet use by different user groups are presented in Table S4 in Multimedia Appendix 1.

For all user groups combined (N=14,562), the mean of comprehensibility ratings was 8.88 (Likert scale from 0 to 10). For all HCPs, the mean of comprehensibility ratings was higher than 9 (physicians: 9.39, pharmacists: 9.25, midwives: 9.14, and other HCP: 9.01). For patients, the mean was 8.80, and for other lay persons (partners, relatives, and friends), it was 8.83. Patients’ assessment of comprehensibility depended on their level of education. The mean comprehensibility ratings was 7.29 for patients without a school-leaving certificate, while it was 8.92 for patients with an academic degree.

HCPs were asked to indicate if they had found the information required to make a well-founded treatment decision. This was answered in the affirmative (“Yes, definitely” and “Somewhat”) by 94.1% (1577/1676) of physicians, 92.7% (510/550) of pharmacists, and 92.6% (337/364) of midwives, with the majority of users fully agreeing (“Yes, definitely”: physicians 72.1%, 1208/1676; pharmacists 65.3%, 359/550; and midwives 70.3%, 256/364).

A total of 66.6% (9694/14,562) of all users stated that they had changed their risk perception of a particular drug during pregnancy or breastfeeding after reading a fact sheet. Risk assessment remained unchanged more often in HCPs than in patients (physicians: 46.9%, 786/1676; pharmacists: 42.9%, 236/550; midwives: 51.9%, 189/364; and patients: 29.9%, 3252/10,860). Compared to their perception before reading the fact sheet, 16.2% (2358/14,562) of all users estimated the risk to be higher and 50.4% (7336/14,562) to be lower. The change of risk perception was also linked to the safety profile of the drug: After reading fact sheets with a red label on Embryotox, that is, fact sheets for drugs with known teratogenicity and/or severe fetotoxicity, 43.1% (81/188) of users estimated the risk to be higher (no change in risk assessment: 31.4%, 59/188). After reading fact sheets with a green label, that is, for drugs of choice for specified treatment indications, 58% (3642/6279) rated the risk to be lower (no change in risk assessment: 33%, 2070/6279). As a result of fact sheet information, 22.3% of users to whom the question applied (2252/10,083) considered changing drug treatment (“definitely yes”: 8.5%, 862/10,083; “possibly yes”: 13.8%, 1390/10,083). For physicians, the proportion was 27.9% (355/1274; with “definitely yes” 11.4%, 145/1274 and “possibly yes” 16.5%, 210/1274).

A total of 41 users were interviewed from February to November 2022, including 14 physicians and nine pharmacists, midwives, and patients each. Physicians worked in various specialized areas (gynecology, pediatrics, psychiatry, dentistry, and general medicine) and settings (hospitals and outpatient care). Ten physicians were female and 4 male, with a mean age of 41.8 (range 26-69) years. Most of the pharmacists worked in public pharmacies (8/9). Six were female and 3 were male, with a mean age of 37.8 (range 28-56) years. The majority of midwives worked freelance (8/9). All were female, with a mean age of 44.2 (range 34-56) years. Of the patients, 6 were pregnant, 2 were breastfeeding, and 1 patient was planning a pregnancy. Mean age was 34.6 (range 24-42) years.

All 41 interviewees stated to trust the information on Embryotox drug fact sheets. Interviewees from all user groups considered the association of Embryotox with a trustworthy institution to be a reason for trust, with the Charité – Universitätsmedizin Berlin and the Federal Ministry of Health mentioned most frequently. Users explicitly stated that they placed trust in the scientific basis of the website and perceived that drug fact sheets were developed by experts in the field. A physician summarized:

And a pharmacist stated:

In addition, the differentiated description of complex issues was named as a reason for trust, as a patient explained:

Interviewees from all user groups trusted the Embryotox Center as the content provider of the website, which they explained by citing positive experiences with other Embryotox activities such as telephone counseling, lectures, and research projects. A midwife explained:

Furthermore, the absence of commercial interests and the long-standing experience of the Embryotox Center were given as reasons. Patients and pharmacists also emphasized that the recommendation of the website by other users, especially by physicians, built trust in the presented information. One patient said:

Physicians said to trust the website because they had corroborated the information elsewhere, and also because of its clear structure and layout. One of them stated:

Further details on users’ assessment of trust can be found in Textbox 1.

Functions and purpose of typical fact sheet use considering different stakeholder groups are summarized in Textbox 2.

Of 1093 questionnaires, 793 were completed in full and 300 ended after 2 or 3 questions, because the decision-making process was still ongoing (214 patients) or fact sheets had not been used in the process of decision-making (23 physicians and 63 patients). The fully completed questionnaires (n=793) provided the data basis for further analyses; contributing user groups are shown in Table 3.

To characterize their typical use of fact sheets, HCP could select a maximum of 3 out of 7 possible answers: 56.3% (215/382) of physicians used fact sheets typically before or after patient consultations, 76.3% (309/382) exclusively or additionally during the consultation. A total of 30.1% (93/309) of the latter routinely used fact sheets online together with their patients. A total of 46.9% (179/382) of physicians encouraged patients to inform themselves on Embryotox (gynecologists: 59.4%, 57/96) and 9.2% (35/382) printed out fact sheets for them (psychiatrists: 20.9%, 14/67). For further details, refer to Table S5 in Multimedia Appendix 1. Pharmacists and midwives used fact sheets to pass on selected information to patients (79.7%, 63/79 and 66.2%, 43/65, respectively), read the fact sheets together with patients (20.3%, 16/79 and 30.8%, 20/65), and encouraged patients to inform themselves on Embryotox (24.1%, 19/79 and 50.8%, 33/65). Fact sheets were also used to check the medication prescribed by physicians (pharmacists: 57%, 45/79; midwives: 52.3%, 34/65) and to consult with the prescribing physician if a medication was considered problematic (pharmacists: 58.2%, 46/79; midwives: 36.9%, 24/65).

Figure 4 shows the extent to which physicians and patients considered themselves involved in making the final decision about drug therapy during pregnancy and breastfeeding after fact sheets have been used to obtain information for the decision-making process.

The majority of physicians and patients agreed to benefit from fact sheet use. Details are given in Figures 5 and 6.

Patients also indicated that Embryotox drug fact sheets had helped them to convince their physician that a specific medication was compatible with pregnancy or breastfeeding (52.1%, 139/267) or that alternatives had to be considered (33%, 88/267). A total of 98.6% of pharmacists (71/72) agreed “moderately” or “strongly” to use Embryotox drug fact sheets for clarification in cases of disagreement between patients and physicians regarding drug therapy (question not applicable; n.a.: 7/79); the proportion for midwives was similar (95.2%, 59/62; n.a.: 3/65). Both user groups also indicated to use Embryotox drug fact sheets for informing and, if necessary, reassuring worried patients (pharmacists: 98.7%, 74/75; n.a.: 4/79; midwives: 100%, 62/62; n.a.: 3/65).

A total of 53.4% (204/382) of physicians wished for more drug fact sheets on the website (gynecologists: 77.1%, 74/96). Furthermore, 42.9% (164/382) of physicians, particularly internists (55.6%, 20/36) and psychiatrists (47.8%, 32/67), considered “decision aids” to be useful, that is, structured guides through the decision-making process. Internet links to other reliable sources were requested by 19.6% (75/382), references to scientific literature by 14.1% (54/382), increased use of visualizations by 6.5% (25/382), and further explanations on given content by 4.7% (18/382). A total of 16.8% (64/382) did not consider any additions to be necessary.