The Antiplatelet Trialists Collaboration provided convincing evidence of the effectiveness of aspirin in the secondary prevention of cardiovascular and cerebrovascular disease.1 Despite this evidence, many patients remain untreated.24
Many districts have embarked on initiatives to increase uptake in eligible patients in primary care, the best known of which is probably the Framework for Appropriate Care Throughout Sheffield (FACTS) project.
In November 1996, using a model developed after visiting the FACTS project, the Portsmouth Aspirin Project was launched.
This was initially targeted at GPs, with a pack containing a brief summary of the evidence, guidance on patient identification, and an audit template. PGEA accreditation was awarded for participation in the project, together with funding for audit support, and on completion of a report.
However, only 8 of the 79 practices in the district completed the audit cycle. In those practices, uptake of aspirin for secondary prevention rose from 55% to 75%, with a further 18% apparently having contraindications. This improvement was encouraging, but participating practices covered less than 10% of the district population.
A questionnaire was sent to practices that had expressed initial interest, but who had not taken part, to ascertain the reasons for this. The two main reasons were lack of time, and that eligible patients were already being prescribed aspirin appropriately. This seemed unlikely, given experience in this and other districts, but was difficult to disprove.
Community pharmacists seemed the obvious next step in the campaign. Not only would a substantial number of patients in the at-risk groups be taking other cardioactive drugs, and thus present to pharmacies on a regular basis, but so too would those purchasing aspirin.
Community pharmacies therefore offered an opportunity to audit uptake in a broader group, which would also include patients from practices that had not taken part in the original study.
Pharmacists are key primary care professionals, but their involvement in audit locally had to date been minimal. Promoting their role in the implementation of evidence-based practice, and in partnership with general practice and the health authority, was another key potential benefit.
Following positive discussions with both the Local Medical Committee (LMC) and the Local Pharmaceutical Committee, the project was promoted at an evening meeting attended by more than 50 community pharmacists. Presentations were given by a local cardiologist, two of the authors (MS and KH), and the lead GP in audit for the district.
The aims of the project were set out:
|Increase awareness of the effectiveness of low-dose aspirin in secondary prevention of coronary heart disease among pharmacists, and as a result, among patients and GPs|
|Audit the uptake of low-dose aspirin in groups most at risk|
|Increase the uptake of low-dose aspirin in those groups most at risk who attended pharmacies|
|Increase the involvement of community pharmacists in clinical care|
|Increase cooperative working between community pharmacists and GPs|
Discussion at the evening meeting and with local representative committees refined the original options for participation. The LMC did not support the most radical option, in which pharmacists were to encourage immediate purchase of aspirin in eligible patients not taking it. The protocol consisted of four stages:
- All patients presenting prescriptions for cardioactive drugs, defined as those in Chapter 2 of the British National Formulary, were eligible for inclusion in the project. If the patient was already purchasing aspirin, or being prescribed aspirin or warfarin, this was noted on the data collection form. For these patients, no further action was taken other than to ask the patient the reason for aspirin purchase, if appropriate.
- If there was no evidence that the patient was already taking aspirin, he/she was given a questionnaire to complete (Figure 1) while waiting for the prescription to be dispensed. The purpose of the questionnaire was to identify 'at-risk' patients, and to exclude contraindications to aspirin.
- When the patient had completed the questionnaire, he/she returned it to the pharmacist, who quickly assessed it for the presence of indications for, and contraindications to, aspirin.
- If the patient appeared to be eligible for secondary prevention, and without contraindications, the pharmacist completed the top half of a tear-off sheet (Figure 2), which informed the patient that they may benefit from taking aspirin, stressing that the GP was best placed to make this decision. The bottom half of the sheet was sent to the patient's GP, together with a copy of the completed questionnaire. At the base of this note was a request that once the GP had made a decision about the patient's suitability for aspirin, he/she should indicate that decision and return the form to one of the authors (MS) at the health authority.
|Figure 1: Questionnaire given to patients presenting a prescription for cardioactive drugs where there was no evidence that the patient was already taking aspirin|
|Figure 2: Letter and tear-off sheet to be completed by the pharmacist for patients who appeared to be eligible for secondary prevention and had no contraindications|
The protocol and forms were successfully piloted by a local pharmacy, and GPs who received a note responded positively. Eye-catching posters were also developed for display in participating pharmacies, and a patient information leaflet was designed and printed.
A small financial incentive was offered to pharmacies participating in the pilot, and for completion of the data collection forms. The project was intended to run for 2 months, to avoid patients on quarterly prescriptions being questioned twice.
Thirty-six pharmacists expressed interest in the project, either directly following the meeting, or in response to a subsequent article in the district prescribing newsletter.
Twenty-five community pharmacies subsequently accepted visits by one or more members of the team (the authors MS, KH, DLM). The purpose of these visits was to distribute posters, leaflets and forms, and to enable further discussion of the project. Although all 25 of these premises displayed the poster and leaflets, only 18 participated fully by sending in completed data collection forms. The results are shown in Table 1.
|Community pharmacies submitting data||18|
|Patients entered on data collection form||1712|
|Being prescribed warfarin/aspirin||1153|
|Patients who should have been offered questionnaire||450|
Completed questionnaires were thus not received for 43 patients who should have been offered one. Likely reasons for this were that busy pharmacists were not offering them appropriately, or that some patients may well have taken them home. Table 2 shows the results for the next stages of the protocol.
|Questionnaires suggesting indication and no contraindication||92|
|Letters sent to GPs||79|
|Completed letters returned to health authority by GP||62|
|Aspirin not indicated||14|
|Aspirin already being purchased||1|
|No decision yet made (not yet seen; under hospital etc)||9|
|Aspirin now being prescribed||28|
|Aspirin now being purchased||3|
Thus an additional 31 at-risk patients began taking aspirin as a result of the study. On the basis of applying the average risk reductions noted in the Antiplatelet Trialists' paper, at least one of these patients can be expected to benefit through the direct avoidance of either a myocardial infarction or a stroke, which could have been fatal.
Although these 31 patients represent only 1.8% of the total number of patients entering the study, it should be noted that almost 74% of patients were already taking either aspirin or warfarin. Given that patients entered the study purely on the basis of taking a cardioactive drug, many of which may have been prescribed for hypertension, this initial uptake seems high.
It is not possible to extrapolate these figures to the proportion of at-risk (i.e. secondary prevention) patients in the district now receiving aspirin. The denominator in this study was patients on cardioactive drugs, some of whom would not have establiWhed cardiovascular or cerebrovascular disease and thus require secondary prevention.
The role of aspirin in primary prevention remains controversial, and it is not clear whether some of these patients were indeed taking aspirin on that basis.
The study cost around £3000. Most of the money was spent on printing costs and payments to community pharmacists. This does not take into account the time spent by the team at the health authority, nor fully that of the participating pharmacists.
On the face of it, these results appear somewhat disappointing. A substantial investment of time and effort in arranging the project has led to only an additional 31 patients taking aspirin, only one of whom is likely to benefit. However, even if this were the only benefit, the cost of that prevented adverse event is not high in comparison with some other health service interventions.
Though less easily measured, the more substantial benefits were probably reaped from achieving the other aims of the project, namely raising public and professional awareness of the benefits of aspirin, and increasing the involvement of community pharmacists in clinical care, and with partner agencies. Requests for leaflets continued long after the project had finished, and several pharmacists indicated that their participation had boosted morale.
|Word forms carefully, and pilot them first. The protocol and forms were adjusted several times before the pilot, and again after the pilot study.|
|Do not expect academic rigour from a study carried out in a busy primary care setting. There were significant numbers of missed or missing forms, and this is understandable in the context of a pharmacist dispensing many prescriptions in the course of a day, only a small proportion of which related to the project. This is also likely to be a feature of projects that require data collection at the time of consultation. Most audits extract the data as a separate process subsequent to the actual consultation.|
|Including all cardioactive drugs may be too ambitious. Restricting eligible drugs to nitrates, lipid lowering agents and calcium antagonists would have missed some 'at-risk' patients, but would have increased the specificity of the data collection.|
|Be aware of professional boundaries. Although relations between GPs and community pharmacists in the district are reasonably good, professional sensitivities remain. The questionnaire and letters neededMto avoid any suggestion that previous care had been in any way suboptimal.|
|Community pharmacists are an invaluable and enthusiastic professional resource. The proportion of community pharmacies participating in the project was higher than the proportion of practices in the area.|
|Would we do this again? Probably not. We are now convinced that the best ways to promote effective care and audit practice involve the use of dedicated additional staff who are able to promote good practice, and collect data, without imposing additional workload on frontline staff. This is the model that we have used, and continue to use, for recent projects.|
|Community pharmacists are key members of the primary care team. A significant proportion are willing to be involved in interface audit.|
|Despite convincing evidence of effectiveness, low cost and infrequent adverse effects, some patients who require aspirin for secondary prevention of vascular disease are not taking it. Continued efforts are required to overcome this unmet need.|
- Antiplatelet Trialists' Collaboration. Collaborative overview of randomised trials of antiplatelet therapy. Br Med J 1994; 308: 81-106.
- McCallum A, Whincup P, Morris R et al. Aspirin use in middle-aged men with cardiovascular disease: are opportunities being missed? Br J Gen Pract 1997; 47: 417-21.
- Bradley F, Morgan S, Smith H et al. Preventive care for patients following myocardial infarction. Fam Pract 1997; 14: 220-6.
- McCartney P, Macdowall W, Thorogood M. A randomised controlled trial of feedback to general practitioners of their prophylactic aspirin prescribing. Br Med J 1997; 315: 35-