• C h r i s t i n e M . B o n d a n d M a r g a r e t C . Wa t s o n o n b e h a l f o f t h e m e m b e r s o f t h e G r a m p i a n Evidence Based Community Pharmacy Guidelines Group
Pharm World Sci 2003; 25(4): 177–181. © 2003 Kluwer Academic Publishers. Printed in the Netherlands. C.M. Bond, M.C. Watson (correspondence): Department of General Practice and Primary Care, University of Aberdeen, Westburn Road, Aberdeen, AB25 2AY, UK (e-mail:
[email protected]) Key words Community pharmacy Evidence-based practice Guidelines United Kingdom Vulvovaginal candidiasis Abstract Objective: The purpose of this study was to develop evidencebased guidelines for over-the-counter (OTC) treatment of vulvovaginal candidiasis with non-prescription anti-fungal medicines purchased from community pharmacies. Method: A multidisciplinary guideline development group was recruited from the locality where the guidelines were to be tested. A Nominal Group Technique (NGT) was used to achieve formal consensus within the group regarding the issues that the guidelines would address. Guideline recommendations were developed from the results of two systematic literature reviews that assessed which symptoms were predictive of vulvovaginal candidiasis (using data from epidemiological studies) and estimated the relative effectiveness of oral and intra-vaginal anti-fungals using data from randomised controlled trials. Main outcome measures: Evidence-based guideline recommendations. The guideline statements were linked to the evidence using a standard hierarchy. Results: The guideline development group met four times. The use of NGT was an effective way of achieving consensus on guideline content. Two systematic reviews carried out as part of the guideline development process identified evidence for the guidelines on the efficacy of OTC treatments and symptoms suggestive of vulvovaginal candidiasis. The guideline recommendations were presented as a booklet and a laminated algorithm. In summary, the guidelines highlighted symptoms suggestive of vulvovaginal candidiasis, and symptoms associated with other vaginal conditions that should be referred to the GP. The guidelines stated that oral treatment and intra-vaginal treatment are equally effective, and that selection of an anti-fungal should be based upon safety, cost and patient preference. Many of the recommendations were influenced by OTC licence restrictions of each antifungal product. Contra-indications to, and special precautions with, antifungals were also listed. In addition, the guidelines stated that the male sexual partner does not require treatment unless symptomatic. Conclusion: There is sufficient evidence available to develop evidence-based guidelines for the treatment of vulvovaginal candidiasis in the community pharmacy setting. The NGT is a useful component in the guideline development process. Accepted December 2002 * This article is accompanied by a Commentary on pages 129–134 of this issue.
Introduction Community pharmacists have an established and increasingly recognised role in the provision of advice to customers presenting with symptoms of minor illness1,2. This role is supported by the increased availability from community pharmacies of a range of reclassified medicines, previously only supplied on a doctor’s prescription.
Approximately 75% of pre-menopausal women3 will experience vulvovaginal candidiasis (VVC) at some time. This infection is treated with topical and oral antifungal agents. Since 1992, several topical antifungals and one oral antifungal have been reclassified from prescription only medicines (POM) to over-the-counter (OTC) status in the UK Whilst it is generally recognised that VVC can be treated safely and effectively in this way, there are nonetheless some concerns that not all antifungal sales are appropriate4,5. Clinical guidelines are now widely used to influence clinicians and support evidence based practice. Although most guidelines developed to date have focussed on medical practice, a limited number have been produced for OTC treatment of symptoms presented in community pharmacies6–9. These are mostly consensus guidelines based on best practice, expert input and published research and have been welcomed by the pharmacy profession10. Few guidelines for OTC medicines have been based upon systematic reviews. Different professional hierarchies amongst guideline development groups may inhibit contribution from some members, particularly during the early meetings. The Nominal Group Technique (NGT)12 was developed to facilitate small group discussion which considers the beliefs and opinions of all participants. In this study, the feasibility of using the NGT as part of the guideline development process was explored. The treatment of VVC with OTC antifungals is an ideal topic for guideline development because of the widespread sales of these products as well as the concerns about their safe and appropriate use. The purpose of this study was to develop evidencebased guidelines for OTC treatment of vulvovaginal candidiasis with non-prescription anti-fungal medicines purchased from community pharmacies. The guidelines were subsequently used in a randomised controlled trial to test the effectiveness and efficiency of various guideline dissemination strategies. The guidelines were disseminated to pharmacies participating in the trial. The pharmacies were randomised to four groups: control; educational outreach; continuing education; or, educational outreach and continuing education. The effects of the guidelines and these strategies were tested using simulated patients. The results of the RCT are reported elsewhere13.
Research article
The development of evidence-based guidelines for over-the-counter treatment of vulvovaginal candidiasis*
Method Guideline development group The members of the guideline development group were recruited from the same locality where the study was conducted (Grampian, Scotland). The group comprised 13 members including 3 community pharmacists, 3 general practitioners (GPs), consultants of
177
Table 1 Guideline development meetings Meeting
Purpose
1 2 3 4
Results of Stage 1 of Nominal Group Technique (NGT) to identify and rank issues Discussion of NGT issues and literature review and OTC license Further discussion of literature reviews’ resultsa Agreement of guideline content
a
The systematic review of the relative effectiveness of oral and intra-vaginal anti-fungals was published as a Cochrane review15,16.
microbiology, genito-urinary medicine and gynaecology, 1 patient representative, 1 health economist and the researcher/facilitator (MCW). Four guideline development meetings were held between August and December 1999 (Table 1) with each meeting being chaired by a principle grantholder. At the end of each meeting the researcher also formally recorded details of the meeting (e.g. numbers and identity of attendees, and duration. The purpose of the first meeting was to conduct a NGT12,14 to address the question: “What are the main issues that need to be addressed by guidelines for community pharmacists for overthe-counter treatment of vulvovaginal candidiasis (thrush)?” The question was mailed to group members who were asked to list their responses and return them prior to the first meeting. During the meeting, the responses were discussed and ranked. The first round responses were anonymised. During the meeting however, responses that were added to this list were not anonymous as they were suggested by group members in the presence of the group. On completion of the process the participants were asked to complete a brief questionnaire about their satisfaction with the NGT process and their previous experience of this technique. The NGT results were disseminated to the participants prior to the second meeting.
178
Systematic literature reviews The intention of the development process was to derive evidence-based recommendations for OTC treatment of VVC. A key component of the development process, therefore, was the identification of evidence. Two systematic literature reviews were undertaken to generate the evidence: (1) a Cochrane review of the relative effectiveness of oral versus intravaginal anti-fungals; and (2) a review of epidemiological studies to identify symptoms that are predictive of VVC. When a customer presents in a community pharmacy either to purchase an anti-fungal preparation or describing vaginal symptoms suggestive of VVC, pharmacy staff are reliant solely upon the description of symptoms to make a diagnosis. Therefore, the purpose of the second literature review was to identify symptoms that were more likely to be associated with this fungal infection and thereby increase the likelihood of making a correct “diagnosis” of VVC. Epidemiological studies that evaluated the association between symptoms and a diagnosis of vulvovaginal
candidiasis were included. Additionally, studies that assessed the association between vaginal symptoms and other infections were included to identify symptoms that were predictive of infections other than VVC. Effectiveness review The systematic review of the relative effectiveness of antifungals was conducted according to Cochrane methodology and has been published elsewhere15,16. Randomised controlled trials conducted worldwide and published in any language were included. The primary outcome measure was clinical cure. Mycological cure, patient preference and safety were secondary outcome measures. The drugs that were included in the review were oral and intra-vaginal triazole and imidazole anti-fungals. Ketoconazole was excluded from the review because it is only licensed for the treatment of acute vulvovaginal candidiasis in a few countries and because of its toxicity profile. Nystatin and other anti-fungals were excluded primarily because they do not have OTC licences in the UK. The results of the literature reviews were presented and discussed during the second and third development meetings. Format and content of guideline recommendations The purpose of the fourth and final meeting was to finalise the detailed content of the guidelines and to discuss the results of the relative effectiveness systematic review in the context of the guideline recommendations. The guidelines were completed in April 2000 and approval was obtained from each member of the guideline development group prior to their dissemination, as part of the factorial RCT.
Results First meeting Ten of the thirteen members of the guideline development group participated during the first meeting. Only one participant had previous experience of the NGT. The first stage of NGT generated numerous issues many of which were combined following clarification amongst group members. After the discussion and first round ranking 21 items were selected (Table 2). All participants rated the technique as useful and most felt that their opinions had been represented. One participant was dissatisfied with the NGT process because the final list did not include the points that they felt were important.
Table 2 List of 21 issues selected from first round of the NGT Response 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Diagnostic criteria When not to treat Appropriate treatment Key information from patient When to refer List of symptoms indicating other condition Written materials Concurrent medication Counselling points Contraindications Current licensing Action if treatment unsuccessful Consequences of misdiagnosis Indicators of unsuccessful treatment When does pharmacist need to be involved Safety of treatment Privacy Treatment of sexual partner Continuity of questioning Ensure patient consults GP if treatment unsuccessful Cost to patient
The final list of topics for inclusion in the guidelines clarified the exact questions to be addressed by the systematic reviews (see Method) and emphasised the need to review the OTC licensing conditions of the antifungal products.
Table 3 Major licence restrictions for OTC antifungals In the UK, OTC antifungals are not licensed for the treatment of: • Vaginal candidiasis in pregnancy • Women who have had more than two episodes of candidiasis in 6 months • Women aged under 16 years or over 60 years • Women who have not had vaginal candidiasis diagnosed by a doctor (in the past)
licensing conditions for different products. The main restrictions are summarised in Table 3. Second meeting The second meeting was attended by 10 group members. The results of the literature review and the conflicting OTC licensing requirements were considered in the context of the issues identified by the NGT. There was little consensus and a general sense of lack of progress. Some of these concerns have been used to inform the discussion of this paper. Third meeting This meeting was attended by eight group members. The concerns identified at the second meeting were further discussed and agreement reached on the main guideline recommendations. These were translated into a flow chart by the main researcher (MCW) which was circulated prior to the next meeting.
Systematic reviews Symptoms review The epidemiological review identified several symptoms that were more suggestive of VVC. Pruritus (itch) was the most useful symptom in predicting a diagnosis of vulvovaginal candidiasis17–20. Dysuria (burning or stinging pain on passing urine)18 and dyspareunia (painful sexual intercourse) were also possible symptoms of vulvovaginal candidiasis. The latter is unlikely to be reported to community pharmacy staff17. The presence of unpleasant or offensive vaginal odour is more likely to be associated with other types of vaginal infection e.g. bacterial vaginosis21. This review is currently being updated.
Fourth meeting This was the final meeting and was attended by nine group members. The final draft guideline materials (flow chart and booklet) were presented and agreed after final refinements. It was agreed that all group members would have a further opportunity to comment once the changes had been incorporated. There was a general feeling of satisfaction that an evidence based guideline had been developed and that this was an important development for pharmacy. The guideline details will be reported elsewhere.
Discussion
This study has shown that it is possible to devise evidence based guidelines for non-prescription medicines informed by systematic literature reviews. The results of the NGT gave the guideline development process a framework around which the guideline recommendations could be constructed and encouraged equal contribution from all group members. The vast majority of the group members felt that the NGT had been a useful way to clarify the guideline content. However, there was some concern that items believed to be important by individual members were not retained in the final list. The NGT is a democratic way of ensuring that every suggestion is treated equally and is subjected to group decision, thus avoiding OTC licensing conditions Minor inconsistencies were identified in the OTC dominance of the results by specific individuals. Effectiveness review The systematic review of relative effectiveness concluded that there was no difference between oral and intra-vaginal antifungals. The full review has now been published15. An additional finding of this review was that the male sexual partners of women with VVC do not routinely require treatment for this infection22–25. Antifungal treatment of male sexual partners should only be recommended if the man is symptomatic or if the woman is experiencing chronic infections.
179
180
The OTC licences of antifungals influenced much of the information presented in the guidelines. The lack of consistency in some of the information is a result of the regulatory process and the safety data included in the application for reclassification. Two of the licence requirements for all OTC antifungals are having a previous diagnosis of vaginal candidiasis and not having more than two episodes in the previous 12 months. These statements were not evidence linked but their inclusion in the guidelines was necessary to ensure that supplies of OTC antifungals conformed to the legal requirements. A previous diagnosis of vaginal candidiasis might assume that the symptoms experienced with a “diagnosed” episode of this infection will be consistent and recognisable with subsequent episodes. However, women treated for this infection in general practice are often treated clinically (i.e. on the basis of their symptoms) with no swab being taken26. Furthermore, women’s ability to predict Candida infection from vaginal symptoms is also variable, with 41% of women in one study making an incorrect diagnosis27. Anti-fungals should be given systemically or intravaginally to achieve cure. Anti-fungal preparations that are licensed for external use provide only symptomatic relief and are unlikely to cure the infection. This information is provided in the OTC licence of anti-fungals licensed for external use only. However, from the results of the RCT that followed guideline development, it was apparent that topical antifungals are often sold on their own as a treatment rather than for symptomatic relief13. In order to apply the guideline recommendations, community pharmacy staff need specific information about the person requiring treatment. This requires the member of staff firstly to know what information is necessary to make a diagnosis of vulvovaginal candidiasis, secondly, to know when it is appropriate to sell an antifungal for the treatment of this infection, and thirdly, to know which product is appropriate for a specific individual. In the community pharmacy setting, reported and elicited symptoms are the only information that pharmacists and medicine counter assistants have available to inform their decision regarding whether to supply an anti-fungal. The results of the systematic review of symptoms showed that vaginal candidiasis is more likely to be associated with some symptoms (i.e. pruritus) than others, but these symptoms are not solely predictive of this type of infection. Therefore, the presence of pruritus may be suggestive of vaginal candidiasis but it does not guarantee that this is the cause of the symptom. Ideally all treatment should be based on mycological tests to ensure only confirmed episodes of infection are treated. In the GP setting this is possible but not universally applied. In the community pharmacy setting mycological confirmation could only be achieved using a self-administered diagnostic test of which none are currently available. One problem associated with all paper-based guidelines is that they quickly become obsolete if new evidence becomes available. In the original version of the systematic review the results suggested that fluconazole was marginally more effective than clotrimazole. However, the addition of the more recent evidence changed the conclusions of the review i.e. that there was no difference in efficacy between the oral and intra-vaginal treatments.
Finally, the guidelines recommended that the selection of an antifungal should be based on a range of factors, which would include customer preference, and not solely upon effectiveness. Thus cost and delivery route will be taken into account in decision making. Fluconazole is the only oral anti-fungal available for OTC supply in British community pharmacies. At the time of guideline development fluconazole cost £12.50 for a single dose, making it prohibitively expensive for many women. The reclassification process widens the public’s direct access to more medicines, exploits community pharmacists’ expertise and saves unnecessary GP consultations. Mechanisms to improve the quality of care from community pharmacies, such as evidence based guidelines, are essential to ensure this alternative access to medicines results in care comparable to that received elsewhere in the NHS. Concerns regarding OTC costs may compromise decision making, and schemes for NHS supply through community pharmacists (http://www.rpsgb.org.uk/pdfs/minailrepa.pdf) need to be implemented nationally. Guidelines should be developed for therapeutic groups where there are sufficient data to derive evidence based recommendations for their use.
Conclusion There is sufficient evidence available to develop evidence-based guidelines for the treatment of vulvovaginal candidiasis in the community pharmacy setting. The NGT is a useful component in the guideline development process.
Acknowledgements The study and the Health Services Research Unit were funded by the Chief Scientist Office, Scottish Executive Health Department. The grant-holders for this study were Professor Christine Bond, Professor Jeremy Grimshaw, Jill Mollison and Dr Arthur Winfield. The authors thank the members of the guideline development group: Mr G Largue, Mr J Strachan, Ms Sam Melrose, Dr Blair Smith, Dr Tom Reid, Dr Aileen Downie, Dr Alison Glenesk, Ms Anne Ludbrook, Professor Phil Hannaford, Mrs Anne Mutton, Dr Gillian Flett and Miss Vivian Pirie. The views expressed here are those of the authors and may not represent the funding body.
References 1 Department of Health. Pharmacy in the future – Implementing the NHS plan. A programme for pharmacy in the National Health Service. London: Department of Health, 2000. 2. Scottish Executive. The right medicine. A strategy for pharmaceutical care in Scotland. Edinburgh: HMSO, 2002. 3 Sobel J, Faro S, Force RW, Foxman B, Ledger WJ, Nyirjesy P et al. Vulvovaginal candidiasis: epidemiologic, diagnostic and therapeutic considerations. Am J Obstet Gynecol 1998; 178: 203–11. 4 Nyirjesy P, Weitz M, Grody M, Lorber B. Over-the-counter and alternative medicines in the treatment of chronic vaginal symptoms. Obstet Gynecol Surv 1997; 90: 50–3. 5 Sihvo S, Ahonen R, Mikander H, Hemminki E. Self-medication with vaginal antifungal drugs: physicians' experiences and women's utilization patterns. Fam Pract 2000; 17: 145–9. 6 Matheson C, Bond CM, Flett G, Kennedy E, Duthie I. Over the counter emergency contraception: a feasible option. Fam Pract 1998; 15: 38–43. 7 Porteous T, Bond CM, Duthie I, Matheson C. Guidelines for
8 9 10 11 12 13
14 15
16
17
the treatment of self limiting upper respiratory tract ailments. Pharm J 1998; 260: 134–9. Porteous T, Bond CM, Duthie I, Matheson C. Guidelines for the treatment of hayfever and other allergic conditions of the upper respiratory tract. Pharm J 1997; 259: 62–5. Bond CM. Lower gastrointestinal symptoms. Pharm J 1995; 253: 656–8. Bond CM, Grimshaw J, Taylor RJ, Winfield A. An evaluation of clinical guidelines for community pharmacy. J Soc Admin Pharm 1998; 15: 33–9. Bond CM, Grimshaw J. Multidisciplinary guideline development: a case study from community pharmacy. Health Bull 1995; 53: 26–33. Gallagher M, Hares T, Spencer J, Bradshaw C, Webb I. The nominal group technique: a research tool for general practice? Fam Pract 1993; 10: 76–81. Watson MC, Bond CM, Grimshaw JM, Mollison J, Ludbrook A, Walker A. Educational strategies to promote evidencebased community pharmacy practice: a cluster randomised controlled trial (RCT). Fam Pract 2002; 19: 529–36. Leape LL, Freshour MA, Yntema D, Hsiao W. Small-group judgement methods for determining resource-based relative values. Med Care 1992; 30: ns28–ns39. Watson MC, Grimshaw JM, Bond CM, Mollison J, Ludbrook A. Oral versus intra-vaginal imidazole and triazole anti-fungal agents for the treatment of uncomplicated vulvovaginal candidiasis (thrush): a systematic review. BJOG 2002; 109: 85–95. Watson MC, Grimsha JM, Bond CM, Mollison J, Ludbrook A. Oral versus intra-vaginal imidazole and triazole anti-fungal treatment of acute, uncomplicated vulvovaginal candidiasis (thrush) (Cochrane Review). The Cochrane Library 2, 2001. Geiger AM, Foxman B, Sobel JD. Chronic vulvovaginal candidiasis: characteristics of women with Candida albicans, C glabrata and no candida. Genitour Med 1995; 71: 304–7.
18 Eckert LO, Hawes SE, Stevens C, Koutsky LA, Eschenbach DA, Holmes KK. Vulvovaginal candidiasis: clinical manifestations, risk factors, management algorithm. Obstet Gynecol 1998; 92: 757–65. 19 McCormack WM, Starko KM, Zinner S. Symptoms associated with vaginal colonization with yeast. Am J Obstet Gynecol 1988; 158: 31–3. 20 Mårdh P, Tchoudomirova K, Elshibly S, Hellberg D. Symptoms and signs in single and mixed genital infections. Int J Gynaecol Obstet 1998; 63: 145–52. 21 Eschenbach DA, Hillier SL, Critchlow C, Stevens C, DeRouen T, Holmes K. Diagnosis and clinical manifestations of bacterial vaginosis. Am J Obstet Gynecol 1988; 158 :819–28. 22 Bisschop MPJM, Merkus J, Scheygrond H, Van Cutsem J. Cotreatment of the male partner in vaginal candidosis: a double-blind randomized control study. BJOG 1986; 93: 79–81. 23 Centers for Disease Control and Prevention. 1998 Guidelines for treatment of sexually transmitted disease. MMWR Morb Mortal Weekly Report 47(RR-1), 6–10, 1998. 24 Clinical Effectiveness Group (Association of Genitourinary Medicine and the Medical Society for the Study of Venereal Diseases). National guideline on the management of vulvovaginal candidiasis. Sex Transm Infect 1999; 75: S19–S20. 25 Fong IW. The value of treating the sexual partners of women with recurrent vaginal candidiasis with ketoconazole. Genitour Med 1992; 68: 174–6. 26 O'Dowd T, Parker S, Kelly A. Women's experience of general practitioner management of their vaginal symptoms. BJGP 1996; 46: 415–8. 27 Chaponis R, Bresnick P, Weiss R, Edwards L. Candida vaginitis: Signs and symptoms aid women's self-recognition. J Clin Res Drug Dev 1993; 7: 17–23.
181