GPs’ decision-making behaviour in the prescription of aspirin for primary prevention in elderly individuals: a qualitative study using semistructured interviews in France

Abstract

Background

Low-dose aspirin has long been recommended for primary prevention. This indication has been reconsidered owing to an unfavourable benefit–risk balance, no proven reduction in cardiovascular risk and an increase in the risk of haemorrhage. The recommendations concerning this indication are still contradictory and are a source of confusion for general practitioners (GPs), especially in the management of patients over 70 years of age who are at greater risk of complications. We explored the determinants of the decision-making behaviour of general practitioners regarding aspirin for primary prevention in patients aged 70 years and older.

Methods

Grounded theory was the basis of this qualitative study, which involved individual semistructured interviews with GPs in the Auvergia region. The interviews were recorded and analysed by two researchers.

Results

Twelve GPs were interviewed, and 4 themes were observed. The distinction between primary and secondary prevention was unclear, particularly for patients with diabetes or asymptomatic atheromatous lesions. Cardiovascular risk was assessed using a standardized approach rather than a risk scoring tool. GPs considered physiological age rather than chronological age when assessing the benefits and risks associated with regular low-dose aspirin use. In the end, the opinion of the specialist and that of the patient were paramount in decision-making.

Conclusion

Our study provides a better understanding of the determinants of GPs’ decision-making in the prescription of aspirin for primary prevention in elderly patients. This occurs in a context of uncertainty and is a source of ambivalence; thus, a clear recommendation would be beneficial.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12875-025-03063-x.

Background

Aspirin is recommended for the secondary prevention of atherosclerotic cardiovascular disease (ASCVD) [1]. Low doses of aspirin have long been used for the prevention of cardiovascular disease. However, 3 trials published in 2018 revealed a small reduction in the incidence of cardiovascular events and cardiovascular mortality in patients at high cardiovascular risk (CVR) receiving primary prevention, and this small benefit was offset by an increased risk of haemorrhage [2–4]. Moreover, the ASPREE trial revealed that the risks outweighed the benefits of aspirin as a risk reduction measure for cardiovascular disease in patients over 70 years of age [5]. In addition, according to the European Society of Cardiology, aspirin may be considered for primary prevention in patients with diabetes mellitus with high or very high CVD risk in the absence of clear contraindications; however, the US Preventive Service Task Force does not recommend aspirin in adults 60 years or older [6, 7]. Currently, in France, the only recommendation concerning the use of aspirin as a risk reduction measure dates back to 2012 [8].

Given the contradictory recommendations, it is not always easy for general practitioners (GPs) to adopt the most appropriate clinical practice. Aspirin is still frequently prescribed for primary prevention, especially for elderly patients and patients with diabetes, with rates varying between 10% and 45% depending on the study [9–13]. This relatively high prescription rate is likely characteristic of the frequent prescription of low-dose aspirin for primary prevention by primary care practitioners GPs. Factors associated with this prescribing behaviour have not been explained by any published studies.

The aim of our study was to explore the determinants of the decision-making behaviour of general practitioners regarding aspirin for primary prevention in patients aged 70 years and older using qualitative methods.

Methods

Design and setting

This qualitative study was inspired by grounded theory with the construction of an explanatory model and was based on semistructured interviews conducted between November 2022 and March 2023 with GPs practising in the Allier and Puy-de-Dôme departments. In France, patients must choose a referring physician and declare him or her to the French National Health Insurance Body. Reimbursement of their medical expenses depends on this process. The referring physician, usually a general practitioner (GP), provides initial access to care and coordinates medical follow-up, including access to other specialties. GPs are generally responsible for primary cardiovascular prevention.

This study was reported in accordance with COREQ recommendations (Additional File 1) [14].

Recruitment

GPs were recruited by the investigators to obtain a theoretical purposive sample. GPs practising exclusively in general practice and who were not opposed to voice recording were selected. They were recruited using a snowball effect. To obtain a varied panel, the sample was selected on the basis of demographic variables—age, sex, type and location of practice—and strategic variables—seniority of internship and length of time in practice. GPs were recruited on an ongoing basis until data saturation was achieved.

Data collection

The investigator conducted the semistructured interviews. The investigator (ML) was a locum GP. The interviews were conducted in the context of GP practices.

A thematic interview guide was created beforehand on the basis of studies retrieved during a literature search [7, 8, 13, 15]. The interview consisted of open-ended questions, which allowed the GPs to express themselves freely while covering various themes (see Table 1). Three GPs who did not participate in the study underwent mock interviews to ensure that they understood the questions. After the third interview, given the frequent confusion between primary prevention and secondary prevention, questions 1 and 2 were reversed on the interview grid to ensure that the answers corresponded with primary prevention. Each interview opened with a conceptual clarification question to ensure a shared understanding of primary versus secondary prevention before the participants explored prescribing practices. The participants completed a sociodemographic questionnaire at the start of the interview, and the participants were categorized accordingly.

Table 1.

Interview guide

The aim of the first part of this interview is to discuss your practices for prescribing low-dose aspirin for the primary prevention of cardiovascular disease. The questions are open-ended so that you can answer freely and tell me your opinion and practice.

- What does primary prevention in the field of cardiovascular disease mean to you?

- Can you tell me about the last time you prescribed or combined antiplatelet therapy with aspirin for primary prevention?

Please share your specific practice for prescribing aspirin for primary prevention among elderly people (≥ 70 years of age). The second part of this interview focuses on this specific population.

- What do you think about prescribing aspirin for primary prevention in elderly individuals (≥ 70 years of age)?

- What circumstances or clinical context clinical context prompts the introduction or renewal of aspirin for primary prevention in these patients?

- Do you feel that your attitude is supported by clear scientific data?

- What factors might limit your prescription? What difficulties or obstacles have you encountered regarding the prescription of low-dose aspirin in the context of primary prevention?The interviews were recorded using a digital recorder dedicated to the study. The interviews were then transcribed verbatim by the investigator using Microsoft Word, and all the data were anonymised. The study data were sent by mail to each of the participants to obtain their approval of the content reported. None of the participants made any changes or added any new data.

Data analysis

Analyses were conducted without coding software, using an approach inspired by grounded theory [16–18]. To obtain theoretical saturation, the interview contents were analysed iteratively during the research period. The open analysis benefited from triangulation by comparing the results of two researchers (ML and PV). In the event of disagreement, a third investigator (EC) was solicited to reach a consensus. The use of coding strategies, memo writing, and constant comparative analysis enabled the identification and construction of categories and concepts, leading to the development of a substantive theory represented as a schematic conceptual framework. This was a three-stage process. An initial longitudinal analysis of each interview was conducted, with the creation of descriptive codes. The open codes were then grouped into themes. By comparison and integration, the themes relevant to the research question were subsequently grouped into coherent units representing the different dimensions of the topic studied [19]. The integrative analysis allowed us to develop an explanatory theoretical model.

Ethical and regulatory aspects

As this study fell outside the scope of the Jardé Law, since it was not considered research involving human people and did not involve patients, it did not require ethical approval under French research law [20]. Data collection was declared to the Commission Nationale de l’Informatique et des Libertés (CNIL) by the Université Clermont Auvergne.

Each GP was verbally informed about the research process and objectives, and written information and a consent form were provided before the start of each interview. Written consent was obtained from the participants. To guarantee their anonymity, each participant was given an alphanumeric code, and all information that could identify the participants was removed from the interviews. Only the investigator has access to the correspondence between the participant and the assigned code. The sociodemographic data were processed independently so that no comparisons could be made.

Results

Twelve interviews were conducted between December 2022 and March 2023. Data saturation was achieved after 10 interviews; however, 2 additional interviews were conducted to confirm data saturation. The average duration of the interviews was 27 min. The characteristics of the participants are summarised in Table 2. Four themes emerged from the analysis of the interviews.

Table 2.

Characteristics of the participants

Participants
Gender
Female	n = 5
Male	n = 7
Age
< 30 years	n = 1
30–39 years	n = 3
40–49 years	n = 4
50–59 years	n = 2
> 60 years	n = 2
Number of years in practice
< 10 years	n = 5
> 10 years	n = 7
Type of practice
Solo practice	n = 1
Group practice	n = 9
Multiprofessional health centre	n = 2
Location of practice
Rural (< 2000 habitants)	n = 4
Semirural (2000–20000 habitants)	n = 4
Urban (> 20000 habitants)	n = 4

Vague notions of primary and secondary prevention

Unclear lines of primary prevention

The participants were familiar with the concept of primary prevention, i.e., the absence of cardiovascular events. For some, the difference between primary prevention and secondary prevention was less well known and was a source of confusion. Among all the participants, only one had a clear definition of primary prevention.

The participants were concerned about cardiovascular risk factors, such as diabetes. They considered patients with diabetes to have a very high risk of stroke. They saw them as suitable for secondary prevention, and prescribing aspirin seemed legitimate.

Asymptomatic atheroma: a source of confusion

Similarly, in one patient, an ultrasound during a systemic examination revealed an atheromatous lesion of the supra-aortic trunks, which was quite puzzling; notably, cardiologists usually perform such examinations. They reported that this was a frequent occurrence in elderly patients with no cardiovascular complications. They considered these patients to have a high CVR but felt that the boundary between primary and secondary prevention was very blurred. Quotes with demographic identifiers are presented in Additional File 2.

In fact, in certain clinical situations, I confuse primary and secondary prevention (GP5).

For high-risk diabetic patients… I think of it as secondary prevention, given that the risk is greater in diabetic patients (GP7).

When there are already visible atheromatous lesions, there is indeed a high risk, so I feel that primary my patient is beyond primary prevention (GP1).

If they have had further tests that show blocked carotid arteries, it makes me wonder (GP8).

Cardiovascular risk assessment

The participants linked the indications for prescribing aspirin for primary prevention to CVR. The risk factors they most frequently reported included diabetes, hypertension, hypercholesterolaemia, smoking and obesity. The combination of several of these factors led GPs to question the appropriateness of prescribing aspirin. The GPs explained that they were making an overall estimate, accounting for metabolic factors, lifestyle and heredity, without calculating a score. Only 2 interviewees said that they used a score to assess the level of risk. All agreed that diabetes was a major risk factor because of frequent cardiovascular complications.

I sometimes have to prescribe antiaggregants for people who have not yet had a problem but who have risk factors (GP9).

In primary prevention, it is obviously diabetes first, then the other risk factors such as obesity and smoking. If these are present we may consider prescribing. Once again, I will adapt to the patient’s condition (GP7).

Thus, I use the SCORE 2 risk calculator. I try to calculate it and then I adapt the treatment according to it (GP10).

Ambivalence towards the elderly

Distinction between chronological age and physiological age

For the participants, patients who were aged 70 years were not always considered to be of advanced age. They stated that they preferred to take physiological age into account rather than chronological age. Advanced age did not seem to be a barrier to prescribing aspirin. They considered those aged 85 years to be elderly and at higher risk of cardiovascular disease and complications.

In cases where preventive treatment was needed, participants said they were taking other factors into account, such as the patient’s life expectancy, autonomy and, above all, cognitive status. The loss of autonomy seemed to be a contraindication to the prescription of aspirin for primary prevention.

I admit that for people aged 70–80, I do not truly ask myself the question (…) If a person who is still active or plays a sport starts treatment at 75, then it’s fine (…) For people aged 85, I start to wonder whether there are too many risks of side effects (GP 2).

The patient’s life expectancy is also of interest; is there any point in treating them with primary prevention if they have several serious pathologies that reduce their life expectancy? (GP7).

Of course, if there are cognitive problems, I’m not going to bother him with that, I’m going to be a little less intensive than with a 75-year-old who’s still in good shape and, above all, still in his right mind (GP4).

The benefit‒risk balance is difficult to assess in elderly patients

Given that the risk of cardiovascular disease increases with age, the GPs believed that prescribing aspirin was more justified. However, they were more hesitant because of the less favourable risk–benefit balance for elderly patients, who are more susceptible to adverse effects. Among these patients, they identified the risk of intracranial haemorrhage associated with falls, the risk of digestive haemorrhage and the rarer risk of allergy. They found decision-making difficult and explained that their attitude was based on a case-by-case assessment of the context, always considering the best interests of the patient.

Is it truly in the patient’s best interest to increase their life expectancy by potentially preventing a heart attack or reducing the risk of cerebral haemorrhage? (GP7).

I think that the indication is there, but the real particularity is that the population is much more at risk, and as such, we need to move away from a simple standardized recommendation to a case-by-case analysis depending on the patient’s frailty and calculate the benefit‒risk balance according to the benefit of using this treatment (GP7).

To prescribe or not to prescribe?

Aspirin, a reassuringly low dose

The participants reported that adverse effects of aspirin were minimal, particularly the risk of haemorrhage, which they felt was not very common. The fact that aspirin is prescribed at low doses for this indication tended to reassure them. Among those who said they were aware of the risk of bleeding, some felt that the risk‒benefit ratio indicated that prescribing aspirin was favourable. Other GPs questioned the risk of trivialising prescriptions in relation to low doses of aspirin, which could be considered harmless. They were unanimous in their view that the undesirable side effects of aspirin, particularly haemorrhage, were far less serious than those of oral anticoagulants.

It is not something that worries me in terms of bleeding risk, especially for people aged 75 in fact (GP4).

Yes, there is perhaps a tendency to trivialise things a little, especially as these are small doses of antiaggregants (GP3).

Need for specialist advice

Even though they felt that the risk of haemorrhage was relatively low, few participants said that they initiated treatment on their own. Many relied on the advice of a cardiologist or angiologist. The cardiologist’s opinion helped them to reinforce their position in cases of doubt. GPs who said that they initiated treatment on their own reported that the retrospective approval of another specialist reassured them. Several participants said that they concentrated on reducing risk factors and delegated the prescription of aspirin to the cardiologist. The convergence of views with the cardiologist seemed to them to improve the quality of the information provided and patients’ understanding, thereby enhancing compliance with treatment.

I often rely on the cardiologist. Often, the cardiologist adds it even though I have not prescribed it yet (GP2).

Because of diabetes and cholesterol, I’m going to take an active approach by sending you straight to the cardiologist, who will tell you that you need to start taking anti-aggregants (GP4).

If the cardiologist has recommended it, then there are two people insisting on it, and that makes it simpler (GP6).

GPs explained their uncertainty about prescribing and the need for additional advice when professional recommendations were absent. Even the participants who were aware of studies exploring the prescription of aspirin for primary prevention had no clear conclusion other than to avoid prescribing it to elderly patients. They all felt that they needed clear guidelines to guide their practice.

It is true that we do not have any established recommendations, and it would be interesting to have clear guidelines specifying what needs to be done (GP1).

In fact, it seemed quite clear to me, but now I’m a bit confused by the study I have read, which shows that there is a greater risk of haemorrhage, whereas up until then it seemed quite consistent […] so I’m wondering about prevention, what do we do? […] I think we are opting for less treatment (GP8).

It is finally easier to stop prescribing

In several clinical contexts, participants considered the discontinuation of aspirin to be sensible, especially when the benefit‒risk balance in elderly patients was not favourable. First, they cited the risk of cerebral haemorrhage in the event of a fall and the risk of digestive haemorrhage. For them, digestive haemorrhage occurred less frequently; however, if it did occur, it was not severe. The combination of aspirin and an oral anticoagulant seemed dangerous to them, particularly for elderly patients, which led them to consider discontinuing aspirin. GPs also thought that loss of autonomy and poor medication compliance were good reasons for the discontinuation of treatment.

If the person is already being treated with an anticoagulant, such as Xarelto or Eliquis, I do not recommend combining the two, as it increases the risk of bleeding (GP10).

We also learned that he was not taking it. He was happy to take one less medication (GP8).

On the other hand, in certain clinical situations, GPs reported that it was difficult to stop prescribing aspirin to elderly patients. First, if the treatment had been prescribed by a colleague who was a specialist, they feared that the patient would be resistant to stopping the treatment. Another patient, despite questioning the appropriateness of the treatment, continued to take aspirin for fear of an ischaemic event. The patient’s previous ischaemic event was heavily weighted in their decision to continue treatment, which they did not necessarily consider useful. Some GPs felt constrained by the patient himself, who did not wish to discontinue treatment, even after the unfavourable benefit–risk balance was explained.

I have two or three patients who saw a cardiologist several years ago who told them to take a little to thin their blood […] but for no obvious reason. I have tried to stop them from taking it, but they do not want to (GP3).

It was the specialist who prescribed it. Did the patient meet any other criteria that I’m not necessarily aware of, particularly studies? And then the patient also sees it that way, he asked a specialist for advice and the specialist told him to take it… so as a GP I find it hard to recommend that they stop. Maybe I do not have all the information I need to recommend that they stop (GP6).

In a person in good general condition, you always think: if in 6 months they have a stroke, have I made a mistake by stopping them? (GP2).

Discussion

Main results

Our study explored the decision-making behaviour of GPs in the prescription of aspirin for primary prevention among elderly people. The participants did not truly question its use for this purpose. They reported that they occasionally found it difficult to distinguish between primary and secondary prevention, especially in diabetic patients or those with asymptomatic peripheral atherosclerosis. Before considering prescribing aspirin, the GPs reported analysing patients as a whole. They considered cardiovascular risk factors, lifestyle and heredity and often estimated patients’ “physiological” age. They felt that it was important to estimate the benefit–risk balance; the older the patient was, the more valuable it was. They reported that the decision to prescribe low-dose aspirin was rather difficult. It was not based on the results of clinical trials or a French clinical recommendation, which they would have found useful. Even though the low dosage of aspirin is reassuring, as there are few, if any, adverse effects, they preferred to delegate its prescription to cardiologists or angiologists. They were more comfortable deciding whether to stop treatment, especially if the person has cognitive problems, no autonomy and a risk of falling. On the basis of these data, an explanatory theoretical model was constructed (see Fig. 1).

Fig. 1.

Explanatory theoretical model of the attitude of GPs regarding the prescription of aspirin for primary prevention in elderly patients

Comparison of data from the literature

Although recent studies revealed that the benefit‒risk balance of aspirin use for primary prevention in patients aged over 70 years was not favourable [21–23], aspirin remains frequently prescribed and is often inappropriate [10, 24]. For the GPs in the study, this habit of prescription was based on the presence of cardiovascular risk factors, even though they feel that it is necessary to act on modifiable factors (e.g., smoking, hypercholesterolemia, hypertension, obesity) in a patient-centred approach. They did not systematically calculate CVR scores, as recommended in the European guidelines [6]. In the study by Bonner et al. [25], Australian GPs were also influenced by their subjective perception of risk and rarely relied on absolute risk thresholds. Similarly, in the study by Jansen et al. [26], GPs’ decisions to prescribe antihypertensive medication were based more on the management of individual risk factors than on an approach based on absolute risk.

Another source of concern is type 2 diabetes, which is considered a major risk factor for cardiovascular disease, making GPs more likely to consider prescribing aspirin; however, this is still debated [10]. One of the explanations for this prescribing habit by French GPs is that until 2017, aspirin prescription to diabetic patients was an indicator that was remunerated by the French National Health Insurance Body as part of its public health objectives. Finally, the discovery of asymptomatic peripheral atherosclerosis, especially in the carotid artery, raised questions for GPs and led them to consider antiplatelet therapy, which is not recommended on the basis of recent guidelines of the European Society of Cardiology [27].

However, advanced age is a concern for GPs, who more frequently assess the benefit–risk balance before prescribing aspirin. In the LESS study [28], Mantelli et al. reported that compared with analgesics, which they tend to retain, Swiss GPs more readily deprescribe cardiovascular prevention drugs when they are not indicated. GPs reported that their decision was influenced by the ‘risk’ and “benefit” of the drugs, ‘quality of life’ and ‘life expectancy’, and that they gave priority to the patient’s wishes and priorities when deprescribing. The uncertainties involved in assessing the balance between benefits and risks among elderly people may explain why GPs tend to delegate primary prescribing to their specialist colleagues, who are considered in a better position to make a decision.

Another important finding raised in our study is that GPs associated low-dose aspirin use with a low risk of undesirable effects. Paradoxically, they say they take the risk of cerebral haemorrhage as well as digestive haemorrhage into account as potential contraindications. According to the literature, this is a major risk to be accounted for. A meta-analysis from the review by Marquis-Gravel et al. [29] confirmed the increased risk of major haemorrhage (risk ratio, 1.47; 95% CI, 1.31–1.65; P < 0.0001) with aspirin as primary prevention. The ASPREE trial [5], included in this meta-analysis, which specifically studied community-dwelling elderly patients, showed higher rates of major haemorrhage and an increase in all-cause mortality. Inconsistency in the reported risk assessments in our study highlights the difficulty of predicting this risk of haemorrhage. The need for a tool to assess the risk of bleeding was also emphasised by Dasa et al. in their review in 2020 [30]. The frailty of the elderly makes GPs see aspirin as a potentially inappropriate drug and makes them consider its deprescription as a necessity, as specified in the American Geriatrics Society Beers Criteria^® [31].

The inappropriate prescription of aspirin can be an example of overmedication, which is defined as a healthcare service that is provided under circumstances in which its potential for harm exceeds the possible benefit [32]. In a study by Warkentin L. in Germany, GPs reported that, on average, 30.2% of all GP services can be attributed to medical overuse [33]. Like the GPs in our study, 90% of them wished to have more guidelines with clear recommendations of what not to do. Even with clear recommendations, healthcare professionals may find it difficult to apply these guidelines to older people [34, 35], who are often multimedicated, preventing them from tailoring medicines to individual patients and ensuring that the decisions they make are safe and/or defensible. Such difficulties in adapting their decisions on the basis of limited evidence, comorbidities, polypharmacy, frailty or life expectancy were also reported in a qualitative study conducted by Jansen [36]. Decision-making related to prescriptions reported by GPs is consistent with elements of the overmedication models, such as the lack of clear responsibility for deprescribing. Although GPs are often recognised as being well positioned to take on the role of medicine manager, they may find it difficult to do so because of a lack of clear responsibility caused by multiple prescribers and transitions between primary and secondary care. In our study, this was reflected in GPs’ frequent referrals to specialists, illustrating how unclear responsibility for deprescribing can hinder appropriate medication management. Similarly, in our study, GPs reported obstacles such as fear of negative consequences, patient ambivalence, or resistance to change, which have also been highlighted in studies on potentially inappropriate prescribing [37].

Finally, the decision to prescribe or withdraw a prescription is based primarily on patient–clinician interactions [38]. Previous studies suggest that such interactions can be improved through training in communication, shared decision-making and access to information that can be easily shared with patients [39]. In the context of uncertainty surrounding the use of aspirin in primary prevention in elderly patients, several authors have also highlighted shared decision-making as a possible strategy to support appropriate deprescribing in the context of evidence-based medicine [39, 40]. As emphasised in the review by Marquis-Gravel et al., informed shared decision-making is considered a suitable approach to guide clinical practice when the net clinical benefit of aspirin remains uncertain [29]. Our findings suggest how this approach could be observed in real-world encounters.

Strengths and limitations

One limitation is that we interviewed GPs in a single region between November 2022 and March 2023. The results may not reflect the experiences and views of GPs in other regions in France.

Another limitation is that the interviewer (ML) was a locum GP and novice qualitative researcher who conducted his first interviews. His presence and interactions may have influenced the interviewees’ responses, and his lack of experience may have reduced interactivity at the beginning of the study. During the interviews, his self-questioning and positioning (e.g., the fact that participants were unfamiliar with scientific data, the impression some participants had of being judged by the interviewer) were noted and discussed with a researcher with experience in medical practice and research (PV). Similarly, during data analysis, to limit interpretations, the initial analyses were discussed with two supervisors (PV, EC).

The interviews were transcribed verbatim by the interviewer who conducted them, ensuring accuracy and completeness. The interview transcripts were read and analysed independently by two researchers to support consistent interpretation and enhance credibility. Participants were also invited to review the results, which further supported the credibility of the findings.

Prospects for care and research

Overall, the GPs in the study did not question the usefulness of low-dose aspirin as primary prevention in elderly patients (≥ 70 years). However, they based their decisions on an assessment of the benefit–risk balance, which they considered particularly important in this population. In most cases, thy deferred to the cardiologist’s opinion. This study highlights the need for clear professional recommendations to guide their practice.

On the basis of the experience of the GPs in our study and in line with prior work indicating that any net benefit may be limited to very high-risk patients with low bleeding risk [41], evaluating aspirin for primary prevention in this subgroup would be relevant. In this context of uncertainty regarding the use of aspirin in primary prevention, practitioners should prioritize a shared decision-making approach in which patients’ values and preferences are considered [42, 43]. Such an approach could be supported by improved communication training, the development of decision aids, and access to information that can be easily shared with patients.

Conclusion

Our study offers a deeper understanding of the determinants of GPs’ decision-making behaviour to prescribe aspirin for primary prevention in elderly patients. It appears that GPs are not very familiar with trials investigating the indications for low-dose aspirin use. Although it is difficult for GPs to distinguish primary prevention from secondary prevention, risk factors should be weighed against adverse effects, particularly the risk of haemorrhage, which increases with age. GPs tend to rely on cardiologists to decide whether to prescribe or discontinue aspirin.

Abbreviations

CVR

Cardiovascular Risk

GP

General Practitioner

Authors’ contributions

PV, ML and EC, conceptualised and designed the study, analysed and interpreted the interview data, and were responsible for drafting and writing the original manuscript. ML and EC conducted the literature review. BC was involved in the interpretation of the data. SB, MB and BC critically reviewed and edited the manuscript. All the authors have read and approved the final version of the manuscript.

Funding

There was no funding for the study.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

All methods were conducted in accordance with the relevant guidelines and regulations. Informed consent was obtained from all the subjects and/or their legal guardian(s). Given that this study was outside the scope of the Jardé Law, since it was not considered research involving a human person and did not involve patients, it did not require ethical approval under French research law. Informed consent was obtained from all participants. Data collection was declared to the CNIL (Commission Nationale de l’Informatique et des Libertés) by the University of Clermont Auvergne.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.