Pharmaceutical Journal Vol 263 No 7071 p788-791
November 13, 1999 Original Papers

Community pharmacist reporting of suspected ADRs: (2) Attitudes of community pharmacists and general practitioners in Wales

By Jane Houghton, BSc, MRPharmS, Fiona Woods, BSc, MRPharmS, Sarah Davis, PhD, Rosalind Coulson, MSc, FRPharmS, and Philip A. Routledge, MD, FRCP

Aim To determine the level of awareness of the involvement of community pharmacists in adverse drug reaction reporting, and to elicit the views of general practitioners and community pharmacists on this new and potentially important role for community pharmacists.
Design A questionnaire survey designed to elicit information on the level of awareness of the ADR reporting scheme, how widely accepted it had become, and general opinions on ADR reporting.
Subjects and Setting A randomly selected sample of GPs and community pharmacists in Wales.
Results Questionnaires were returned by 167 (67%) of 250 GPs and 142 (57%) of 250 community pharmacists. Although 135 (95%) of the community pharmacists were aware they had the authority to report suspected ADRs, only 56 (34%) of the GP respondents were aware of this. 110 GPs (66%), but only 52 community pharmacists (37%) recalled that they had seen an ADR in the past year. 29 of those community pharmacists (54%) stated that they had discussed the ADR with the patient's GP. 86% of GPs considered that community pharmacists had a responsibility to report suspected ADRs to the CSM/MCA, with 96% indicating that it was also the GP's responsibility. 94% of community pharmacists stated that they considered it to be the pharmacist's responsibility to report, with 97% indicating that it was also the GP's responsibility.
Conclusions The low reporting rate (38 yellow cards) from the 1,598 community pharmacists in Wales during the demonstration period indicates that further efforts are needed to increase community pharmacists' awareness about ADR reporting. However, the questionnaire indicates a high degree of support among both GPs and community pharmacists for a role for community pharmacists in reporting suspected ADRs via the yellow card scheme.

An adverse drug reaction (ADR) is any undesirable effect of a drug beyond its expected therapeutic effects occurring during clinical use. ADRs are a significant cause of morbidity and mortality. A survey of a Yorkshire general practice showed that 1 in 40 consultations were for an ADR.¹. A further survey of a practice population showed that 41 per cent of patients either "certainly" or "probably" had a reaction to a drug prescribed for the first time. The gastrointestinal and central nervous systems were most frequently involved and in this survey 90 per cent of the reactions occurred by the fourth day of treatment.². ADRs are often the cause of hospital admissions. It has been suggested that approximately 5 per cent of all admissions are due to suspected ADRs.³
Clinical trials cannot detect all suspected ADRs because the trials are often of short duration and the patient populations are carefully selected for the study. In addition, the conditions under which clinical trials are undertaken do not reflect the conditions under which the drug will subsequently be prescribed. It is only once a drug is marketed and used in much larger numbers of patients that rarer ADRs, and those resulting from interactions or long-term toxicity, may become apparent. Postmarketing surveillance or pharmacovigilance is therefore essential for detecting less common ADRs.⁴ In the United Kingdom, a system for the spontaneous reporting of suspected ADRs to the Committee on Safety of Medicines (CSM)/ Medicines Control Agency (MCA) (the yellow card scheme) has now been operating for more than 30 years. Reports of suspected ADRs are primarily submitted voluntarily by general practitioners (GPs), hospital doctors, dentists, coroners and more recently by hospital pharmacists. Reports are also received via the pharmaceutical industry, which has a statutory obligation to report suspected serious ADRs.
During its 30-year history, the yellow card scheme has been critically important in monitoring drug safety in normal clinical practice, increasing knowledge about known ADRs, and acting as an early warning system for the identification of previously unrecognised reactions.⁵ Nevertheless, the scheme is kept under continual review in order to develop ways of increasing reporting in areas where there has been little reporting to date, eg, suspected ADRs to over-the-counter medicines and herbal medicinal products.
In 1992 and 1993, a pilot scheme run by the CSM's regional monitoring centre (Northern region) concluded that reporting of suspected ADRs by hospital pharmacists would enhance the yellow card scheme. In 1997, the CSM recommended that the yellow card scheme should be extended to include hospital pharmacists nationwide.
In April, 1997, the CSM/MCA introduced a demonstration scheme to evaluate the reporting of suspected ADRs by community pharmacists in the four regions covered by the CSM regional monitoring centres: Mersey, Northern, West Midlands and Wales. The results of this scheme would be reviewed before any further extension would be considered.⁶
Community pharmacists were requested to report suspected ADRs in accordance with the UK ADR reporting guidelines, ie, serious reactions to established drugs and all reactions to newly introduced "black triangle" drugs. They were requested also to focus their reporting on those areas where there was limited reporting by doctors and in which community pharmacists have special expertise, ie, over-the-counter (P and GSL) and unlicensed medicinal (herbal) products. Community pharmacists are ideally placed to assist in this task, since they are often the health care professional in the first line of contact for the use of these products. They were asked to discuss the particular case with the relevant GP prior to sending a yellow card report. While it was not recommended that a report be made against the advice of the GP directly responsible for the patient concerned, the pharmacist could exercise his or her professional judgment in sending in such a report to the CSM/MCA.⁶
In February, 1997, the CSM/MCA informed all doctors and pharmacists of its intention to extend the ADR reporting scheme to pharmacists (community pharmacists working in the four regions of the CSM's monitoring centres and all hospital pharmacists) with an article in Current Problems.⁷
In April, 1997, two copies of an information pack, "Pharmacovigilance, the yellow card scheme: information pack for pharmacists", were sent by the CSM/MCA to every community pharmacy in Wales. Training sessions were held in conjunction with the Welsh Centre for Postgraduate Pharmaceutical Education (WCPPE), and a comprehensive ADR teaching pack, which was produced by the CSM's regional monitoring centre for Wales (CSM Wales), was made available to all tutors. Since fewer than 10 per cent of all community pharmacists in Wales attended these training sessions, CSM Wales sent each community pharmacist a news-sheet informing them of the demonstration scheme and included a self-addressed card for them to obtain their own personal copy of the information pack. Further newsletters were sent at six-monthly intervals. The newsletters reported on the progress of suspected ADR reporting by pharmacists in Wales and included encouragement to send further reports to the regional monitoring centre.
CSM Wales undertook a questionnaire survey of a sample of GPs and community pharmacists in Wales towards the end of the two-year demonstration period. The questionnaire was designed to elicit information on the level of awareness of the scheme, how widely accepted it had become and opinions on the subject of ADR reporting.

Method

Structured questionnaires were designed and piloted on a small number of GPs and community pharmacists prior to the main study. The questionnaires were designed to be straightforward and to take no longer than five minutes to complete. They were a combination of tick boxes and sections for free text.
The names and practice addresses of GPs registered in the UK and practising in Wales were supplied by Health Solutions Wales. This amounted to a total of 2,253 doctors. The WCPPE provided a list of the 1,598 pharmacists working in community practice in Wales. A total of 30 community pharmacists, who had reported suspected ADRs to CSM Wales via the demonstration scheme were excluded from the list of community pharmacists; the 28 general practitioners for whose patients these community pharmacists had submitted reports were excluded from the list of general practitioners.
Two hundred and fifty GPs and the same number of community pharmacists were randomly chosen from the remaining 2,225 doctors and 1,568 pharmacists on the lists. In February, 1999, a personalised letter, the questionnaire and a Freepost reply envelope addressed to CSM Wales were sent out to the randomly selected sample of doctors. The sample of community pharmacists were also sent a personalised letter, a slightly different questionnaire and a Freepost reply envelope.
After four weeks, a reminder letter and questionnaire were sent to all non-responders in the two groups. Finally, a third personalised letter from the directors of CSM Wales was sent to each non-responder together with a third copy of the questionnaire and a Freepost envelope. All data were manually collated and anonymised prior to analysis and the results are expressed as a percentage of the total responses for each individual question.

Results

Questionnaires were returned to CSM Wales from 167 GPs (a 67 per cent reply rate). Of the respondents, 118 (70 per cent) replied to the initial questionnaire, 26 (16 per cent) to the second and 23 (14 per cent) to the third. Questionnaires were received from 142 community pharmacists (a 57 per cent reply rate). Of the respondents, 91 (64 per cent) replied to the initial questionnaire, 28 (20 per cent) to the second and 23 (16 per cent) to the third.

Fifty-six of the general practitioner respondents (34 per cent) and 135 of the community pharmacists (95 per cent) were aware that community pharmacists had the authority to report suspected ADRs. Eighty-eight of the community pharmacist respondents (62 per cent) stated that they had read a copy of the information pack which gave guidance on the demonstration scheme. Only 39 per cent of community pharmacists had read a copy of the CSM Wales newsletter which gave details of the progress of the scheme. Between April and October, 1997, lectures and workshops on identifying and reporting suspected ADRs were held at a number of venues throughout Wales in association with the WCPPE. Thirty-five of the community pharmacists who responded to the questionnaire (25 per cent) stated that they had attended one of these lectures. Both GPs and community pharmacists were asked whether they had seen an ADR in the past year. One hundred and ten GPs (66 per cent) and 52 community pharmacists (37 per cent) recalled that they had seen an ADR in the past year. Table 1 indicates the percentage of respondents who had seen differing numbers of adverse reactions. Both GPs and community pharmacists were asked how the suspected ADRs had been brought to their attention. The results are shown in Table 2.

Table 1: How many suspected adverse drug reactions have you seen in the past year? Number of reactions seen Community pharmacists General practitioners 0 60% 30% 1 1% 5% 2-4 13% 20% 5-10 8% 12% 11-15 1% 2% 16-20 1% 2% >20 1% 8% *Other   4% Non-respondents 15% 17% *"Other" refers to those GPs who had seen reactions but could not say how many Data are expressed as a percentage of the total number of questionnaires returned by community pharmacists and general practitioners.

Table 2: How were the suspected ADRs brought to your attention?
	Community pharmacists		General practitioners
	Yes	No	Yes	No
By the patient	51 (98%)	0 (0%)	101 (92%)	1 (1%)
By noticing changes in the patient's medication	20 (38%)	8 (15%)
By discussing the patient's medication	23 (44%)	4 (8%)	57 (52%)	8 (7%)
By the doctor	3 (6%)	14 (27%)
By a hospital doctor			11 (10%)	31 (28%)
By the pharmacist			7 (6%)	31 (28%)
Fifty-two community pharmacists and 110 general practitioners recalled that they had seen an ADR in the past year. The Table shows the number of respondents who replied to each question

Of the 52 community pharmacists who had seen an ADR in the past year, 29 (54 per cent) stated that they had discussed it with the patient's GP. Information was also sought as to whether the reactions which the community pharmacists had seen were reported to CSM Wales. Only 48 community pharmacists responded to this question, with three (6 per cent) indicating that they had reported the suspected ADR. These three community pharmacists reported in the period between the selection of the random sample and the distribution of the questionnaire. The most common reasons for non-reporting of suspected ADRs by the community pharmacists who had seen a suspected ADR but not reported it were that the reaction was already known, or they had doubt regarding causality (Table 3).

Table 3: Reasons for non-reporting by community pharmacists who indicated that they had seen a suspected ADR but had not reported it to CSM Wales
Reason	Yes	No
Recognised reaction	29 (64%)	0 (0%)
Uncertain that the reaction was caused by the drug	11 (24%)	4 (9%)
Already reported by doctor	9 (20%)	6 (13%)
GP did not want pharmacists to report the reaction	3 (7%)	5 (11%)
Unable to contact the doctor	2 (4%)	6 (13%)
Unaware community pharmacists could report	1 (2%)	6 (13%)
The Table shows the numbers of community pharmacists who responded to each question; the data were also expressed as a percentage of community pharmacists who had seen a suspected ADR but had not reported it to CSM Wales (45 community pharmacists)

Only six GPs (4 per cent) who responded stated that they had been contacted by a community pharmacist regarding a suspected ADR occurring in one of their patients. This low figure is to be expected since all GPs whose patient had been the subject of a report had been excluded from the survey. Three GPs (2 per cent) stated that they had a patient who had been the subject of a community pharmacist yellow card. However, an extensive search of the Adverse Drug Reactions On-line Information Tracking (ADROIT) database, into which yellow card reports are entered, was unable to locate these reports.
Both GPs and community pharmacists were questioned to determine whose responsibility they thought it was to report suspected ADRs. It is of note that 143 GPs (86 per cent) considered that community pharmacists had a responsibility to report suspected ADRs to the CSM/MCA, with 96 per cent indicating that it was also the GPs' responsibility. Ninety-four per cent of community pharmacists stated that they considered it to be the pharmacist's responsibility to report to the CSM/MCA, with 97 per cent indicating that it was also the GP's responsibility (Table 4).

Discussion

It is clear that this randomly selected sample of community pharmacists in Wales were almost all aware of their authority to report suspected ADRs to the CSM/MCA. The vast majority also believed that it was their responsibility to report. However, only 52 of respondents (37 per cent) remembered having seen a suspected ADR in the previous year. Only 32 yellow cards were received from a total of 1,598 community pharmacists working in Wales during the demonstration period. This questionnaire identifies several possible reasons for this low number.
Only 62 per cent of the community pharmacist respondents stated that they had read the original information pack from the CSM/MCA. This pack had been sent to each pharmacy premises in Wales and not to each pharmacist. Virtually all of the respondents stated that they were aware of the existence of the demonstration scheme. However, less than half could remember reading either of the newsletters which had been sent to all community pharmacists from CSM Wales informing them of the progress of the pharmacist reporting scheme.
A total of nine training sessions were arranged in association with WCPPE and held at several different venues throughout Wales during the study. Twenty-five per cent of community pharmacist respondents stated that they had been able to attend one or more of these sessions, compared with the actual overall attendance of less than 10 per cent of the total number of community pharmacists in Wales. This may be explained by responder bias, with those who attended the sessions being possibly more likely to have completed the questionnaire than those who did not attend.
Many of the community pharmacist respondents stated that they would like to receive additional education on identifying suspected ADRs. Other surveys have also identified this need.⁸ A card giving details of the availability of a distance learning pack on ADR reporting was included in the original information pack, but uptake of this resource was low.
In this study, 66 per cent of GPs and 37 per cent of community pharmacists said they had seen one or more ADRs in the past year. However, only 31 per cent of GPs stated that they had reported a suspected ADR to CSM Wales. A survey carried out on reporting by Welsh GPs in 1985 showed that 61 per cent claimed to have reported a reaction.⁹ This is a similar result to that found in another survey in which 58 per cent of 348 GP respondents claimed to have reported a reaction to the CSM at some time, 38 per cent saying they had done so in the previous year.¹⁰ Another investigator reported that 77 per cent of GP responders stated that they had reported one or more suspected ADRs to the CSM previously.¹¹ Thus GPs' recollections of their own reporting habits consistently give relatively high reporting rates. However, when the actual reporting rate of GPs in Wales was examined only 63 per cent of general practitioners had submitted at least one yellow card over a four-year period between 1992 and 1995.¹² There is thus a difference between perceived and actual reporting rates. Since the perceived reporting rates quoted were examined by questionnaire, this difference could again be explained by responder bias.
The commonest stated reason for non-reporting was that the community pharmacists felt the ADR was already well recognised. We did not explore whether such suspected ADRs were also serious reactions to established drugs or ADRs in association with newly introduced (black triangle) drugs. The CSM/MCA requests reporting in both these circumstances. It is thus unclear how many appropriate ADRs were not reported because they were thought to be already well-recognised. The second most frequently stated reason for not reporting was uncertainty about whether the reaction was caused by the drug. This should not be a barrier to reporting suspected ADRs, but it has been cited in the past as a common reason for non-reporting by doctors.¹³ Both these reasons suggest uncertainty among community pharmacists as to the reporting guidelines issued by the CSM/MCA and further promotion of the guidelines may be needed.
Another reason for non-reporting may have been confusion concerning whether reports could be sent by the community pharmacist without the knowledge and agreement of the GP. Clarification of this issue was often requested at the meetings attended by the CSM Wales staff. Several respondents to the questionnaire cited difficulties in contacting the GP as a reason why they did not report the suspected ADRs they had seen. When contact was made, most of those who commented on the interaction found the discussion with the GP to be positive and helpful. In another study it was indicated that involvement in a reporting scheme had not adversely affected the relationship with their GP.⁸
Nearly 90 per cent of the GP respondents felt that community pharmacists should have a responsibility to report suspected ADRs. Nevertheless, the vast majority still believed that they also had a responsibility to report. Indeed 914 yellow cards were received from GPs in Wales over the same period, representing or corresponding to the highest annual reporting rate since the establishment of CSM Wales in 1983. Although only a third of the GPs were aware that community pharmacists in Wales had been given the authority to report suspected ADRs since April, 1997, those who commented specifically on this initiative were positive and felt it was a good idea that was long overdue. Other investigators have reported that many GPs believed that community pharmacists should be involved in ADR reporting.¹⁴ Only one of the 167 GPs who responded to our questionnaire stated that pharmacists should not be responsible for reporting suspected ADRs to the MCA/CSM.
The intention of the questionnaire was to obtain the general views of GPs and community pharmacists on the subject of the demonstration scheme rather than the views of community pharmacists who had reported, or GPs whose patients had been the subject of a report made via the scheme. For this reason, those community pharmacists who did report during the study and the corresponding GPs for the patients involved were excluded before the random selection of community pharmacists and GPs who were sent the questionnaire. These individuals were also sent questionnaires, but results were not added to the data from the random sample. It is not surprising that these 30 community pharmacists were, without exception, enthusiastic about the scheme. This enthusiasm was shared by the GPs whose patients they had reported on, and who responded to this survey.
It is unfortunate that only 57 per cent of the community pharmacists and 67 per cent of the GPs responded to the questionnaire, despite three attempts to obtain a reply. Comments from some of the respondents indicated that pressure of work might have been one factor in their poor ADR reporting rate and it may also have had an impact on response rate to the questionnaire. This low reporting rate should be considered when generalising the findings of the questionnaire to the region as well as to the UK as a whole. Nevertheless, Wales is around the same size as many English health regions and the reporting rate of suspected ADRs by GPs is very similar. There are therefore unlikely to be substantial differences between the attitudes of community pharmacists and GPs in Wales and those in the UK as a whole.

Conclusion

The study highlights the importance of the role of community pharmacists in ADR reporting. In addition it clearly demonstrates the support of both professions for the role of community pharmacists in pharmacovigilance. In order to ensure that this role is performed effectively, continuing joint professional education should be provided. This will help to improve the under-reporting of suspected ADRs that has been shown in previous studies over many years. This initiative comes as a challenge and great opportunity for every community pharmacist to increase their involvement in patient care.

ACKNOWLEDGMENTS We thank Professor Martin Kendall and Dr Peter Arlett for reviewing the manuscript, Dr David Temple and his colleagues at the Welsh Centre for Postgraduate Pharmaceutical Education for organising the workshops in pharmacovigilance, Mrs Mandy Tipping and Miss Siān Houghton for their help in collating the questionnaires, and all the general practitioners and community pharmacists who completed the questionnaire.

References

Jane Houghton is senior adverse drug reaction pharmacist,CSM Wales, and Fiona Woods is principal pharmacist at the Welsh drug information centre, University Hospital of Wales, Cardiff. Sarah Davis is ADROIT signal generation co-ordinator and Rosalind Coulson is specialist in pharmacovigilance information management at the Medicines Control Agency, London. Philip Routledge is director of CSM Wales and professor of clinical pharmacology at the University of Wales College of Medicine, Heath Park, Cardiff CF4 4XN. Correspondence to Professor Routledge