- Email: [email protected]
A Feasibility Study Comparing Pharmacist and Physician Recommendations for Sildenafil Treatment
1463
ORIGINAL RESEARCH—ED PHARMACOTHERAPY A Feasibility Study Comparing Pharmacist and Physician Recommendations for Sildenafil Treatment jsm_2212
1463..1471
Tara Symonds, PhD,* John D. Dean, MB, BS, FRCGP,† Alison Carr, PhD,‡ Martin Carlsson, MS,§ Aditya Marfatia, BSPharm, MS,*¶ and Gabriel Schnetzler, MD*¶ *Pfizer Inc, New York, NY, USA (at time of research); †St. Peter’s Andrology Centre, London, UK; ‡Hamell Communications, London, UK; §Pfizer Inc, New York, NY, USA; ¶Pfizer International Operations, Paris, France (at time of research) DOI: 10.1111/j.1743-6109.2011.02212.x
ABSTRACT
Introduction. In Europe , pharmacists may be an important first point of contact for men with erectile dysfunction (ED) asking for advice and treatment. Aim. To determine if European community pharmacists could appropriately recommend suitability for supply of sildenafil 50 mg for the treatment of ED. Methods. For this cross-sectional, observational study, the current Summary of Product Characteristics was adapted to create a study drug information sheet for use in a pharmacy setting in which, for certain patients, supply is not suitable and referral to a physician is recommended. After training and with use of a guidance questionnaire, pharmacists assessed the suitability of supply of sildenafil 50 mg for men presenting to their pharmacy. Men with self-reported ED who were not currently using a phosphodiesterase type 5 inhibitor were recruited. Within 7 days of the pharmacist-patient interaction, a physician with experience in the management of ED telephoned the subject to assess suitability. If there was discordance between the pharmacist and physician recommendations, the case was independently reassessed by a physician specialist in sexual medicine. Main Outcome Measures. The primary end point was the concordance rate (with 95% confidence intervals) between pharmacist and physician recommendations. Rates were weighted by country sample sizes. Results. Concordance (95% confidence interval) was 0.70 (0.66–0.74) between pharmacist and physician recommendation, indicating agreement in 70% of cases, and was 0.90 (0.86–0.94) between pharmacist and physician specialist in sexual medicine. Furthermore, if the cases in which the pharmacist did not put subjects at risk (i.e., gave an acceptable recommendation) are assessed, the success rate is 83.5% (79.6–87.4%) and 92.8% (90.1–95.5%), respectively. Conclusion. Pharmacists were accurate in providing suitable treatment recommendation, generally not recommending sildenafil for men without ED and recommending physician assessment when there was any question about cardiovascular health, other comorbidity, or co-medication. Symonds T, Dean JD, Carr A, Carlsson M, Marfatia A, and Schnetzler G. A feasibility study comparing pharmacist and physician recommendations for sildenafil treatment. J Sex Med 2011;8:1463–1471. Key Words. Erectile Dysfunction; Nonprescription Drugs; Pharmacist; Sildenafil; Treatment Guidance; Phosphodiesterase Type 5 Inhibitors
Introduction
I
n Europe, the public recognizes community pharmacies as a “vital, integral part of the health services” that are conveniently accessible and “. . . where sound, objective advice on health © 2011 International Society for Sexual Medicine
issues can be obtained, from a knowledgeable health professional, in an informal environment in which they feel relaxed, without the need to make an appointment.” [1] A corresponding role for pharmacists as providers of patient-focused and medicine-focused care is now common [2]. Thus, J Sex Med 2011;8:1463–1471
1464 community pharmacists are an important first point of contact with a healthcare professional for people seeking advice and guidance on treatment. These services are increasingly used and studied in various countries in Europe, either in the context of improving patient adherence to prescription treatment, additional medication review for potential drug–drug interactions, or supply of nonprescription medicines [3–7]. Benefits of specific education, training programs, and ongoing support on pharmacists’ confidence in determining patient suitability for medicine supply and in providing patient counseling have been demonstrated for emergency hormonal contraception [8], simvastatin [9], and nicotine replacement therapy [10]. Community pharmacists may also be an important first point of contact with a healthcare professional for men with erectile dysfunction (ED). Recently, an observational study compared men with vs. without a prescription for a phosphodiesterase type 5 inhibitor (PDE5i: sildenafil, tadalafil, or vardenafil) who asked for ED treatment at one of more than 500 pharmacies in Spain [11]. There were no statistically significant differences between the two groups of men in ED severity, the proportion without ED, or the duration of symptoms before first consultation with a healthcare professional (mean of 26 months in each group). In the nonprescription group, the pharmacy visit was the first discussion about ED for 60%, 50% of those who had previously discussed ED had done so with a pharmacist in the first instance, and 85% asked for a PDE5i. Some recommendations exist for ED management in community pharmacies [12], and a small patient group direction initiative is ongoing in the United Kingdom [13]. Aim
The aim of this study was to determine if community pharmacists across several European Union (EU) countries, given support by tailored training documents and a Pharmacy Guidance Questionnaire (PGQ [14], available upon request), could make an appropriate recommendation for supply of sildenafil 50 mg to treat ED. Methods
A study drug information sheet was adapted from the current Summary of Product Characteristics [15], with additional restrictions, for use in the pharmacy setting. The PGQ was based on the study drug information sheet. The PGQ was initially developed J Sex Med 2011;8:1463–1471
Symonds et al. and studied in the United Kingdom. However, to identify whether this assessment at a community pharmacy is also practical in other EU countries, the study was expanded to Germany, Spain, and the Czech Republic, which were chosen to represent different languages, cultures, and daily practice.
Participants Eligible pharmacists were in full-time practice in community pharmacies in cities in the United Kingdom (n = 18), Germany (n = 11), Spain (n = 15), and the Czech Republic (n = 9). Eligible primary care physicians had experience in the management of ED and performed telephone consultations in the United Kingdom (n = 5), Germany (n = 3), Spain (n = 3), and the Czech Republic (n = 2). Eligible men were 18 years or older and had self-reported ED or erection problems likely to constitute ED, defined as a positive response to the question “Do you have difficulty getting or keeping an erection that is satisfactory for sexual intercourse?” Those currently using a PDE5i were excluded. The study was advertised using local media advertisements (Germany, the United Kingdom, and the Czech Republic) or in the pharmacy (Spain and the United Kingdom), depending on legal and regulatory limitations. Interested subjects rang a call center, where they received additional information about the study and were screened for eligibility. Study Design This was a cross-sectional, observational study; no medication was dispensed. Subjects visited a participating pharmacy and completed the patientscreening questionnaire of the PGQ and, to confirm ED status, the Sexual Health Inventory for Men (SHIM), a validated five-item questionnaire developed from the International Index of Erectile Function as a diagnostic tool for ED [16]. Using the guidance from the PGQ and the study drug information sheet, pharmacists assessed suitability for sildenafil 50 mg use—suitability was not communicated to the men. Within 7 days of the pharmacist assessment, a study physician telephoned the subject to perform an independent clinical assessment (using the clinical questions typical to his or her practice) and judge suitability. If there was discordance between the recommendations of the pharmacist and physician, a physician who was a specialist in sexual medicine reassessed the case using all available information from the PGQ, the SHIM, and the physician’s and
1465
Feasibility of In-Pharmacy PDE5i Prescription the pharmacist’s notes. When a vague diagnosis was provided by the pharmacist, such as “heart problems,” the worst-case scenario was assumed, so that the estimate of risk was the highest possible. When a physician diagnosed “heart problems” and identified sildenafil as being suitable for the patient, the physician assessment was considered a reliable “gold standard.” The protocol was reviewed and approved by independent ethics committees where required. All patients provided written informed consent.
Study Materials Educational materials were developed, which covered the topics of ED as a medical condition, product information on sildenafil 50 mg (including instructions on proper use [dosing interval, dietary guidance predosing, need for sexual stimulation]), and additional information and practical tips for supplying the medicine in a pharmacy setting. Tailored training materials were provided, which consisted of a series of 10 fictional case studies that incorporated responses to the patient-screening questionnaire and were to be assessed for suitability using the pharmacy guidance notes of the PGQ. The study drug information sheet indicated that pharmacy supply would not be appropriate and referral to a physician would be recommended for certain patients. These included those who do not have ED; are physically unfit; are taking any nitrates or nitric oxide donors, potent cytochrome P3A4 inhibitors, a-blockers, or other ED medicine; have heart, liver, or kidney problems; have a history of certain rare eye disease (nonarteritic anterior ischemic optic neuropathy or retinitis pigmentosa); or had a recent heart attack or stroke. The PGQ was developed in focus groups with independent community pharmacists (n = 22) in various geographic locations in the United Kingdom to improve comprehensibility, acceptability, ease of use, and ability to discriminate between men who were suitable and unsuitable for supply of sildenafil 50 mg in the pharmacy setting. Wording and content were refined, and it was then translated using standard translation methodology. Additional brief questionnaires were completed by the patient and pharmacist to evaluate the patient-screening questionnaire of the PGQ and the pharmacist guidance portion of the PGQ, respectively, for ease of completion and acceptability. Statistical Analysis Sample size determination was based on a pooled projected concordance rate of ⱖ0.66 for the four
countries. Using a two-tailed a-level of 0.05, and assuming that the true concordance rates were ⱖ0.64 in Germany, Spain, and the Czech Republic (based on concordance data from the initial phase of this study performed in the United Kingdom) [17], enrollment of 66 subjects per country was required to demonstrate that the lower bound of the confidence interval (CI) for the pooled concordance rate was ⱖ0.60, which would indicate good concordance [18]. Missing data were not imputed. Data were analyzed using SPSS for Windows version 14.0 (SPSS Inc., Chicago, IL, USA). Main Outcome Measures
The primary end point was concordance between pharmacist and physician assessments of suitability for supply of sildenafil 50 mg in the pharmacy setting according to the study drug information sheet. Concordance rates were estimated with 95% CIs that were calculated using the large sample normal approximation to the binomial distribution. If the positive and negative agreement rates between the pharmacist and physician were balanced, concordance was to be assessed with the kappa statistic and would be considered substantial if there was ⱖ61% agreement [18]. Homogeneity of the concordance rates across the countries was assessed with a chi-square test; if heterogeneity was present, the overall concordance rate and CI would be based on weighted country proportions, with individual country sample sizes as the weights. There were several secondary end points: • Acceptable recommendation: Number of men not put at risk, which included the false-negative recommendations (i.e., men suitable for treatment considered unsuitable by the pharmacist). • False-positive rate: Number of men considered suitable for treatment but who were actually unsuitable/total number considered suitable • False-negative rate: Number of men considered unsuitable for treatment but who were actually suitable/total number considered unsuitable • Sensitivity of the pharmacist assessment: Number of men agreed by the sexual medicine specialist and the pharmacist to be suitable for treatment/total number considered suitable by the sexual medicine specialist • Specificity of the pharmacist assessment: Number of men agreed by the sexual medicine specialist and the pharmacist to be unsuitable for treatment/total number considered unsuitable by the sexual medicine specialist J Sex Med 2011;8:1463–1471
1466 • Positive predictive value: Number of true positives/(number of true positives + number of false positives) • Negative predictive value: Number of true negatives/(number of true negatives + number of false negatives) • Pharmacist inter-rater variability: Based on the 10 case studies • Feedback on the PGQ: Patient and pharmacist
Symonds et al.
Results
The study was designed to recruit subjects regardless of their suitability for supply of sildenafil 50 mg, to test the pharmacists’ ability to make correct recommendations. Those judged unsuitable were subjects with heart problems (n = 43), inability to walk briskly for 15 minutes without becoming breathless or experiencing chest pains (n = 39), or liver/kidney problems (n = 24). More than half of the men (n = 189 [55%]) were currently taking medicine for some other medical condition.
Patients Of the 2,182 men who responded to the advertisements, 1,147 met inclusion criteria and were eligible for entry; 515 of these decided not to participate when the study procedures were explained. Of the 632 who agreed to participate, 187 were withdrawn without undergoing any assessments because of logistic difficulties (loss of contact or inability to attend an appointment in a participating pharmacy), 35 attended only the pharmacist assessment and were withdrawn, and 64 were withdrawn as protocol violators because they either completed the physician assessment before the pharmacist assessment or were current users of a PDE5i. A total of 346 men completed the study and contributed data to the analysis: 96 from the United Kingdom, 99 from Germany, 59 from Spain, and 92 from the Czech Republic. The mean age was 52 years (standard deviation [SD] 13.6; range 18–85 years), with 61 (18%) aged ⱕ 39 years, 174 (51%) aged 40 to 59 years, and 106 (31%) aged ⱖ 60 years (age data were missing for five men). The majority of the men were white (94%). Many men (n = 143/ 335 [43%]) had a previous physician diagnosis of ED but were not currently taking a PDE5i (11 men did not respond to the question). SHIM scores indicated that 334 men (97%) had ED (Table 1), including 77 (22%) with mild ED, 109 (32%) with mild-to-moderate ED, 76 (22%) with moderate ED, and 72 (21%) with severe ED.
Appropriateness of the Pharmacist Recommendation Concordance Rates Heterogeneity was present among the countries (Table 2). Therefore, concordance rates were based on weighted country proportions, with individual country sample sizes as the weights. Concordance between the pharmacist and physician recommendations was 0.70 (95% CI, 0.66–0.74), indicating agreement between the healthcare professionals in 70% (243/346 [weighted data]) of cases. After the physician specialist in sexual medicine resolved cases for which the recommendation of the pharmacist and the physician did not agree, the adjusted concordance was 0.90 (95% CI, 0.86–0.94), indicating agreement in 90% (311/346 [weighted data]) of cases. Of the cases that the pharmacist found suitable but the physician found unsuitable, 67% (35/52) were reclassified by the sexual medicine specialist as suitable (true positive; Figure 1). Reasons for reclassification were that the physician’s cited contraindication or “unsuitable” recommendation was not supported by the study drug information sheet or the responses on the PGQ (n = 29) or because the physician’s diagnosis of no ED (n = 5) or of low desire as the etiology of the sexual dysfunction (n = 1) was contradicted by the ED diagnosis provided by the SHIM. Of the 17 cases for which the pharmacist incorrectly recommended suitability, the responses on the PGQ did not support the
Table 1 Erectile dysfunction diagnosed by score on the Sexual Health Inventory for Men vs. by patient self-report and by physician assessment ED diagnosis by SHIM score*, n (%)†
Patient self-reported ED (PGQ question 1: Do you have difficulty getting or keeping an erection that is satisfactory for sexual intercourse?) Physician diagnosis of ED
ED No ED ED No ED
ED (score ⱕ 21)
No ED (score ⱖ 22)
Total
328 6 326 8
8 (67%) 4 (33%) 12 (100%) 0
336 10 338 8
(98%) (2%) (98%) (2%)
*A score ⱕ 21 is defined as ED [16,19]. † Shading indicates positive and negative concordance. ED = erectile dysfunction; PGQ = Pharmacy Guidance Questionnaire; SHIM = Sexual Health Inventory for Men.
J Sex Med 2011;8:1463–1471
(97%) (3%) (98%) (2%)
1467
Feasibility of In-Pharmacy PDE5i Prescription Table 2
Concordance rates Rate of concordance in pharmacist recommendation of suitability for PDE5 inhibitor treatment
United Kingdom (n = 96) Germany (n = 92) Spain (n = 59) Czech Republic (n = 99) Overall (N = 346) weighted data (95% CI)
Rate of concordance in pharmacist acceptable recommendation*
vs. physicians
vs. sexual medicine specialist
vs. physicians
vs. sexual medicine specialist
0.71† 0.73† 0.59† 0.77† 0.70 (0.66–0.74)
0.83† 0.91† 0.88† 0.97† 0.90 (0.86–0.94)
0.85‡ 0.83‡ 0.65‡ 0.96‡ 0.84 (0.80–0.87)
0.92† 0.91† 0.88† 0.99† 0.93 (0.90–0.96)
*Number of men not put at risk, which included the false-negative recommendations (i.e., men who were suitable for treatment per the physician and the sexual medicine specialist, respectively but were considered unsuitable by the pharmacist). †P < 0.05 between countries. ‡ P < 0.0001 between countries. CI = confidence interval; PDE5 = phosphodiesterase type 5.
pharmacist’s recommendation (n = 11) or the contraindication cited by the physician was not reported to the pharmacist on the PGQ (n = 6). Thirteen of the pharmacist false-positive classifications were considered critical: cardiovascular symptoms with moderate exercise or cardiovascular problems (n = 10); contraindicated medications (n = 2, tamsulosin and alprostadil); and painful erections (n = 1). Of the cases that the pharmacist found unsuitable but the physician found suitable, 81% (39/48) were reclassified by the sexual medicine specialist as unsuitable (true negative; Figure 1). Reasons for the reclassification were that the physician’s recommendation was not supported by the study drug information sheet (n = 30), was at odds with his or her clinical opinion (e.g., a recommendation
of “suitable” but a comment that treatment requires prior medical assessment [n = 4]), or the contraindication identified by the pharmacist was neither included in the study drug information sheet nor cited by the physician but considered legitimate (diabetes and hypertension [n = 1], psychological state and lack of clarity about concurrent medications [n = 3], and vascular abnormalities [n = 1]). Of the nine cases for which the pharmacist incorrectly recommended unsuitability, the pharmacist recommendation was not supported by the responses on the PGQ and/or the SHIM (n = 5) or by the study drug information sheet (high alcohol intake that the pharmacist thought warranted liver function testing before sildenafil treatment [n = 3] and heart problems reported on the PGQ [n = 1]).
Figure 1 Sexual medicine specialist reassessment of cases with discordance between the pharmacist recommendation and the physician recommendation.
J Sex Med 2011;8:1463–1471
1468
Acceptable Recommendation Concordance between the pharmacist and physician recommendations, excluding the false positives, indicated that pharmacists made an acceptable recommendation in 83.5% (95% CI, 79.6–87.4%) of cases. After reclassifications by the sexual medicine specialist, pharmacists made acceptable recommendations in 92.8% (95% CI, 90.1–95.5%) of cases. False-Positive and False-Negative Rates The false-positive rate was 0.09 for pharmacists and 0.14 for physicians (9% and 14%, respectively). The false-negative rate was 0.12 for pharmacists and 0.34 for physicians (12% and 34%). Sensitivity, Specificity, and Predictive Value The pharmacist assessment showed sensitivity of 0.96 (96% of men suitable for treatment identified) and specificity of 0.74 (74% of men unsuitable for treatment identified). Pharmacists were proficient at correctly identifying men who were suitable (positive predictive value, 0.90) and unsuitable (negative predictive value, 0.88) for treatment. Of the men who did not meet the SHIM criteria for ED, treatment was recommended by pharmacists for half (6/12) vs. by physicians for 92% (11/12). Pharmacist Inter-Rater Variability The case study recommendations had an overall concordance rate across the pharmacists of 89% (95%CI, 86–92%) and an estimated overall kappa value of 0.79 (95% CI, 0.73–0.84), indicating substantial inter-rater agreement. The false-positive rate was 11%, and the false-negative rate was 10%, indicating a balanced sample. Feedback on the Pharmacist Guidance Questionnaire The patient-screening questions were rated easy or very easy to answer by 92% (315/340) of men, and 73% (245/338) felt very or quite comfortable answering the questions in their local pharmacy. Pharmacists rated the notes of the PGQ as very or quite easy to use in 96% (n = 330/344) of cases and the men’s responses as very or quite easy to interpret in 97% (n = 329/340) of cases. The median time to assess suitability for treatment was 14.5 minutes (interquartile range 89.93; range 0.17–90). Discussion
This study demonstrated that community pharmacists are cautious and responsible in recommendJ Sex Med 2011;8:1463–1471
Symonds et al. ing sildenafil 50 mg for treatment of ED. The concordance rate was 0.70 (95% CI, 0.66–0.74) between pharmacist and physician recommendation and was 0.90 (95% CI, 0.86–0.94) between pharmacist recommendation and assessment by a sexual medicine specialist. Furthermore, if the cases in which the pharmacist did not put subjects at risk (i.e., gave an acceptable recommendation) are assessed, the success rate is over 80% (0.84 [95% CI, 0.80–0.87]) between pharmacist and physician recommendation and is over 90% (0.93 [95% CI, 0.90–0.96]) between pharmacist and sexual medicine specialist. In each country, concordance rates between pharmacists and the sexual medicine specialist were at least 83% and rates of acceptable recommendation were at least 88%. Thus, pharmacists were accurate in providing appropriate treatment recommendation when using guidance from the PGQ and the study drug information sheet, which included additional restrictions compared with the current Summary of Product Characteristics. To enable appropriate patient counseling, the educational materials ensured that pharmacists were alerted to the proper use of sildenafil (i.e., not taking more than one tablet a day, avoiding heavy meals, need for sexual stimulation). In 17 cases, the pharmacist incorrectly recommended suitability, including 13 cases which were considered critical. Unfortunately, it cannot be established why the pharmacists recommended treatment suitability in these cases, against the guidance of both the PGQ and the study drug information sheet—in situations for which precautions and contraindications already exist in the current Summary of Product Characteristics. Further training materials may be necessary or cardiovascular risks and contraindicated medications could be highlighted in greater detail in the PGQ. However, it is also noteworthy that physicians recommended suitability in 30 cases in which precautions or contraindications existed that were identified by pharmacists using the PGQ. This outcome supports the results of a survey among more than 1,650 delegates of a European Association of Urology congress—only 12% of respondents reported that they assess contraindications in men asking for a PDE5i prescription [20]. Most patients and pharmacists rated the PGQ positively. The pharmacist–patient interaction, which included responding to the seven questions of the PGQ, lasted only a few minutes. Almost all pharmacists (96%) rated the PGQ notes as easy to follow and 92% of men rated the patient-
1469
Feasibility of In-Pharmacy PDE5i Prescription screening questions as easy to answer. Also, despite the potential embarrassment of talking about ED, the majority of men (73%) felt comfortable answering the questions in a local pharmacy. A potential limitation of this study is its inability to assess the clinical outcomes of actual use based on the pharmacist recommendation because, in the European Union, supply and subsequent patient follow-up are currently only possible for prescription-only medicines under medical supervision. This is in contrast to the United States, where product supply and patient behavior can be studied before a potential switch to nonprescription status [21,22]. Another possible limitation is that the study was conducted in only four EU member states. However, the concordance rates between pharmacist and sexual medicine specialist in the blinded review being consistently >80% suggest that differences in culture and daily practice across the countries have little impact on the pharmacist’s ability to adequately assess PDE5i suitability. Thus, it is reasonable to assume that the study outcomes could be representative of all EU countries, all of which apply the same professional pharmacy principles [1]. Another potential limitation is that ED is not conventionally managed by telephone consultation, but by face-to-face consultation with a physician, whereas pharmacist assessment is likely to be similar to the method here. Finally, because the results of the study, including the diagnosis of ED, are based on patient-reported clinical data, the accuracy of the assessments and the strength of the results are dependent upon the openness and honesty of the patients, whether the healthcare professional eliciting the data is a pharmacist or a physician. For example, a diagnosis of ED is based on the selfreported symptom of inability to attain and maintain penile erection sufficient to permit satisfactory sexual intercourse [23]. In the current study, inclusion was based on a simple question about ED symptoms, which was identical to the first question on the PGQ and was asked by an anonymous call center operator. This question screened out approximately half of the men responding to the study advertisement, such that 98% of the men included in the study population had ED using a gold-standard questionnaire (SHIM) [19]. This result supports the validity of pharmacist use of the initial question of the PGQ to diagnose ED and to screen out those without ED in men requesting sildenafil 50 mg. ED often occurs concomitantly with other diseases (e.g., diabetes and cardiovascular disease) or
with conditions that are associated with an increased risk for a cardiovascular event (e.g., smoking, obesity, hypercholesterolemia, and hypertension) [24,25]. Based on a study conducted in the United Kingdom, most men receiving a prescription for a PDE5i already have a medical diagnosis of an underlying condition such as hypertension or hypercholesterolemia, and only a few men receive a new such diagnosis at the time of PDE5i prescription [26]. Furthermore, an observational study conducted in Spanish community pharmacies of men asking for ED advice or treatment showed that there were no statistically significant differences in the severity of ED or the comorbidity profile between those men who approached the pharmacist with or without a prescription for a PDE5i [11]. Thus, community pharmacists can play an important role in identifying patients with ED, providing education and counseling on potential cardiovascular risk factors and recommending further medical care [12]. A reminder for further medical care to check for potential underlying conditions was also highlighted in a more recent version of the patientscreening questionnaire, used in subsequent research [27]. Conclusions
In this feasibility study in several EU member states, which included men across a broad age range (18–85 years), pharmacists provided appropriate recommendation for supply of sildenafil 50 mg in the pharmacy setting according to the study drug information sheet in 90% of cases, closely following the guidance within the PGQ (based on the study drug information sheet), not recommending for men without ED, and generally recommending physician assessment when appropriate. This suggests that pharmacists are capable of assessing patient suitability for ED treatment with sildenafil 50 mg. Acknowledgment
The authors would like to thank Ms. Anwen Russell of TBWA Health for her invaluable support. Editorial support was provided by Deborah M. CampoliRichards, BSPHA, RPh, of Complete Healthcare Communications, Inc., and was funded by Pfizer Inc. Corresponding Author: Tara Symonds, PhD, Pfizer Ltd Global Research and Development, Outcomes Research (IPC160) Ramsgate Road, Sandwich, Kent J Sex Med 2011;8:1463–1471
1470
Symonds et al.
CT13 9NJ, UK. Tel: +44 (1304) 641553; Fax: +44 (1304) 658823; E-mail: tara.symonds@pfizer.com Conflict of Interest: John D. Dean is President and CEO of the International Society for Sexual Medicine (Incorporated in California); a consultant/advisor and meeting participant/lecturer for Bayer Schering Pharma, Boehringer-Ingelheim, Johnson & Johnson, Lilly, Pfizer, Plethora, Procter & Gamble, and ProStrakan; and a scientific study/trial investigator for Bayer Schering Pharma, Boehringer-Ingelheim, Lilly, Pfizer, Plethora, Procter & Gamble. Alison Carr is an employee of Hamell Communications, an independent healthcare research company that was a paid consultant to Pfizer in designing and overseeing the running of the study and analyzing the data. At the time of this research, Martin Carlsson, Aditya Marfatia, Gabriel Schnetzler, and Tara Symonds were employees of Pfizer.
4
5
6
7
8
9
10
Statement of Authorship
Category 1 (a) Conception and Design Tara Symonds; John D. Dean; Alison Carr; Martin Carlsson; Aditya Marfatia; Gabriel Schnetzler (b) Acquisition of Data Tara Symonds; Martin Carlsson; Gabriel Schnetzler (c) Analysis and Interpretation of Data Tara Symonds; John D. Dean; Alison Carr; Martin Carlsson; Aditya Marfatia; Gabriel Schnetzler
11
12
13 14
Category 2 (a) Drafting the Article Tara Symonds; John D. Dean; Alison Carr; Martin Carlsson; Aditya Marfatia; Gabriel Schnetzler (b) Revising It for Intellectual Content Tara Symonds; John D. Dean; Alison Carr; Martin Carlsson; Aditya Marfatia; Gabriel Schnetzler
15 16
Category 3 (a) Final Approval of the Completed Article Tara Symonds; John D. Dean; Alison Carr; Martin Carlsson; Aditya Marfatia; Gabriel Schnetzler References 1 Pharmaceutical Group of the European Union (PGEU). The commitment of community pharmacists in Europe. Available at: http://www.pgeu.eu/tabid/596/Default.aspx (accessed July 20, 2009). 2 Anderson C, Bates I, Beck D, Brock TP, Futter B, Mercer H, Rouse M, Whitmarsh S, Wuliji T, Yonemura A. The WHO UNESCO FIP pharmacy education taskforce. Hum Resour Health 2009;7:45. 3 Hughes CM, Hawwa AF, Scullin C, Anderson C, Bernsten CB, Bjornsdottir I, Cordina MA, da Costa FA, De Wulf I, Eichenberger P, Foulon V, Henman MC, Hersberger KE, Schaefer MA, Sondergaard B, Tully MP, Westerlund T,
J Sex Med 2011;8:1463–1471
17 18 19
20
21
22
23
McElnay JC. Provision of pharmaceutical care by community pharmacists: A comparison across Europe. Pharm World Sci 2010;32:472–87. Berenguer B, La Casa C, de la Matta MJ, Martin-Calero MJ. Pharmaceutical care: Past, present and future. Curr Pharm Des 2004;10:3931–46. Eickhoff C, Schulz M. Pharmaceutical care in community pharmacies: Practice and research in Germany. Ann Pharmacother 2006;40:729–35. Herborg H, Sorensen EW, Frokjaer B. Pharmaceutical care in community pharmacies: Practice and research in Denmark. Ann Pharmacother 2007;41:681–9. Gastelurrutia MA, Faus MJ, Fernandez-Llimos F. Providing patient care in community pharmacies in Spain. Ann Pharmacother 2005;39:2105–10. Bacon L, Savage I, Cook S, Taylor B. Training and supporting pharmacists to supply progestogen-only emergency contraception. J Fam Plann Reprod Health Care 2003;29:17–22. Hansford D, Cunningham S, John D, McCaig D, Stewart D. Community pharmacists’ views, attitudes and early experiences of over-the-counter simvastatin. Pharm World Sci 2007;29: 380–5. Sinclair HK, Bond CM, Stead LF. Community pharmacy personnel interventions for smoking cessation (Review). Cochrane Libr 2009;1–14. Martin Morales A, Ibáñez J, Machuca M, Pol-Yanguas E, Schnetzler G, Pascual Renedo V. The EPIFARM study: An observational study in 574 community pharmacies in Spain characterizing patient profiles of men asking for erectile dysfunction medication. J Sex Med 2010;7:3153–60. Vic JM. Specific aspects of erectile dysfunction and its treatment for community pharmacists. Int J Impot Res 2004;16: S50–2. Mayer S. UK pilot allows pharmacists to supply sildenafil without prescription. BMJ 2007;334:387. Dean J, Carr A, Carlsson M, Marfatia A, Schnetzler G, Symonds T. Correlation between community pharmacist and physician recommendations for treatment with a phosphodiesterase type 5 inhibitor. Poster presented at the Joint Congress of the European and International Societies for Sexual Medicine, Brussels, Belgium, December 7–11, 2008. J Sex Med 2009;6(suppl 2):135. VIAGRA Summary of Product Characteristics. European Medicines Agency, London, UK, 2008. Rosen RC, Cappelleri JC, Smith MD, Lipsky J, Peña BM. Development and evaluation of an abridged, 5-item version of the International Index of Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction. Int J Impot Res 1999;11: 319–26. Pfizer Inc. Data on file. Pfizer Inc, New York, NY. 2009. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74. Cappelleri JC, Rosen RC. The Sexual Health Inventory for Men (SHIM): A 5-year review of research and clinical experience. Int J Impot Res 2005;17:307–19. Schnetzler G, Marfatia A, Zou KH, Muniain J, Symonds T. Current erectile dysfunction management: A survey among delegates of the 2008 annual EAU congress in Milan. Eur Urol 2009;(suppl 8):212(#368). Schwartz SM, Bansal VP, Hale C, Rossi M, Engle JP. Compliance, behavior change, and weight loss with orlistat in an over-the-counter setting. Obesity (Silver Spring) 2008;16: 623–9. Brass EP. Consumer behavior in the setting of over-thecounter statin availability: Lessons from the consumer use study of OTC Mevacor. Am J Cardiol 2004;94:22F–9F. NIH Consensus Development Panel on Impotence. NIH Consensus Conference. Impotence. JAMA 1993;270:83–90.
Feasibility of In-Pharmacy PDE5i Prescription 24 Thompson IM, Tangen CM, Goodman PJ, Probstfield JL, Moinpour CM, Coltman CA. Erectile dysfunction and subsequent cardiovascular disease. JAMA 2005;294:2996–3002. 25 Hodges LD, Kirby M, Solanki J, O’Donnell J, Brodie DA. The temporal relationship between erectile dysfunction and cardiovascular disease. Int J Clin Pract 2007;61:2019–25. 26 Kirby M, Marfatia A, Zou KH, Symonds T, Schnetzler G. Diagnosis of underlying disease in men with erectile
1471 dysfunction receiving phosphodiesterase type 5 inhibitors in the United Kingdom. Eur Urol 2009;(suppl 8):172(#208). 27 Symonds T, Dean J, Coyne KS, Margolis MK, Hackett G, Edwards D, Marfatia A, Schnetzler G. The ability of men to assess their suitability to take 50 mg sildenafil: An assessment of the comprehension of patient information materials. J Sex Med 2010;7:2217–25.
J Sex Med 2011;8:1463–1471
ORIGINAL RESEARCH—ED PHARMACOTHERAPY A Feasibility Study Comparing Pharmacist and Physician Recommendations for Sildenafil Treatment jsm_2212
1463..1471
Tara Symonds, PhD,* John D. Dean, MB, BS, FRCGP,† Alison Carr, PhD,‡ Martin Carlsson, MS,§ Aditya Marfatia, BSPharm, MS,*¶ and Gabriel Schnetzler, MD*¶ *Pfizer Inc, New York, NY, USA (at time of research); †St. Peter’s Andrology Centre, London, UK; ‡Hamell Communications, London, UK; §Pfizer Inc, New York, NY, USA; ¶Pfizer International Operations, Paris, France (at time of research) DOI: 10.1111/j.1743-6109.2011.02212.x
ABSTRACT
Introduction. In Europe , pharmacists may be an important first point of contact for men with erectile dysfunction (ED) asking for advice and treatment. Aim. To determine if European community pharmacists could appropriately recommend suitability for supply of sildenafil 50 mg for the treatment of ED. Methods. For this cross-sectional, observational study, the current Summary of Product Characteristics was adapted to create a study drug information sheet for use in a pharmacy setting in which, for certain patients, supply is not suitable and referral to a physician is recommended. After training and with use of a guidance questionnaire, pharmacists assessed the suitability of supply of sildenafil 50 mg for men presenting to their pharmacy. Men with self-reported ED who were not currently using a phosphodiesterase type 5 inhibitor were recruited. Within 7 days of the pharmacist-patient interaction, a physician with experience in the management of ED telephoned the subject to assess suitability. If there was discordance between the pharmacist and physician recommendations, the case was independently reassessed by a physician specialist in sexual medicine. Main Outcome Measures. The primary end point was the concordance rate (with 95% confidence intervals) between pharmacist and physician recommendations. Rates were weighted by country sample sizes. Results. Concordance (95% confidence interval) was 0.70 (0.66–0.74) between pharmacist and physician recommendation, indicating agreement in 70% of cases, and was 0.90 (0.86–0.94) between pharmacist and physician specialist in sexual medicine. Furthermore, if the cases in which the pharmacist did not put subjects at risk (i.e., gave an acceptable recommendation) are assessed, the success rate is 83.5% (79.6–87.4%) and 92.8% (90.1–95.5%), respectively. Conclusion. Pharmacists were accurate in providing suitable treatment recommendation, generally not recommending sildenafil for men without ED and recommending physician assessment when there was any question about cardiovascular health, other comorbidity, or co-medication. Symonds T, Dean JD, Carr A, Carlsson M, Marfatia A, and Schnetzler G. A feasibility study comparing pharmacist and physician recommendations for sildenafil treatment. J Sex Med 2011;8:1463–1471. Key Words. Erectile Dysfunction; Nonprescription Drugs; Pharmacist; Sildenafil; Treatment Guidance; Phosphodiesterase Type 5 Inhibitors
Introduction
I
n Europe, the public recognizes community pharmacies as a “vital, integral part of the health services” that are conveniently accessible and “. . . where sound, objective advice on health © 2011 International Society for Sexual Medicine
issues can be obtained, from a knowledgeable health professional, in an informal environment in which they feel relaxed, without the need to make an appointment.” [1] A corresponding role for pharmacists as providers of patient-focused and medicine-focused care is now common [2]. Thus, J Sex Med 2011;8:1463–1471
1464 community pharmacists are an important first point of contact with a healthcare professional for people seeking advice and guidance on treatment. These services are increasingly used and studied in various countries in Europe, either in the context of improving patient adherence to prescription treatment, additional medication review for potential drug–drug interactions, or supply of nonprescription medicines [3–7]. Benefits of specific education, training programs, and ongoing support on pharmacists’ confidence in determining patient suitability for medicine supply and in providing patient counseling have been demonstrated for emergency hormonal contraception [8], simvastatin [9], and nicotine replacement therapy [10]. Community pharmacists may also be an important first point of contact with a healthcare professional for men with erectile dysfunction (ED). Recently, an observational study compared men with vs. without a prescription for a phosphodiesterase type 5 inhibitor (PDE5i: sildenafil, tadalafil, or vardenafil) who asked for ED treatment at one of more than 500 pharmacies in Spain [11]. There were no statistically significant differences between the two groups of men in ED severity, the proportion without ED, or the duration of symptoms before first consultation with a healthcare professional (mean of 26 months in each group). In the nonprescription group, the pharmacy visit was the first discussion about ED for 60%, 50% of those who had previously discussed ED had done so with a pharmacist in the first instance, and 85% asked for a PDE5i. Some recommendations exist for ED management in community pharmacies [12], and a small patient group direction initiative is ongoing in the United Kingdom [13]. Aim
The aim of this study was to determine if community pharmacists across several European Union (EU) countries, given support by tailored training documents and a Pharmacy Guidance Questionnaire (PGQ [14], available upon request), could make an appropriate recommendation for supply of sildenafil 50 mg to treat ED. Methods
A study drug information sheet was adapted from the current Summary of Product Characteristics [15], with additional restrictions, for use in the pharmacy setting. The PGQ was based on the study drug information sheet. The PGQ was initially developed J Sex Med 2011;8:1463–1471
Symonds et al. and studied in the United Kingdom. However, to identify whether this assessment at a community pharmacy is also practical in other EU countries, the study was expanded to Germany, Spain, and the Czech Republic, which were chosen to represent different languages, cultures, and daily practice.
Participants Eligible pharmacists were in full-time practice in community pharmacies in cities in the United Kingdom (n = 18), Germany (n = 11), Spain (n = 15), and the Czech Republic (n = 9). Eligible primary care physicians had experience in the management of ED and performed telephone consultations in the United Kingdom (n = 5), Germany (n = 3), Spain (n = 3), and the Czech Republic (n = 2). Eligible men were 18 years or older and had self-reported ED or erection problems likely to constitute ED, defined as a positive response to the question “Do you have difficulty getting or keeping an erection that is satisfactory for sexual intercourse?” Those currently using a PDE5i were excluded. The study was advertised using local media advertisements (Germany, the United Kingdom, and the Czech Republic) or in the pharmacy (Spain and the United Kingdom), depending on legal and regulatory limitations. Interested subjects rang a call center, where they received additional information about the study and were screened for eligibility. Study Design This was a cross-sectional, observational study; no medication was dispensed. Subjects visited a participating pharmacy and completed the patientscreening questionnaire of the PGQ and, to confirm ED status, the Sexual Health Inventory for Men (SHIM), a validated five-item questionnaire developed from the International Index of Erectile Function as a diagnostic tool for ED [16]. Using the guidance from the PGQ and the study drug information sheet, pharmacists assessed suitability for sildenafil 50 mg use—suitability was not communicated to the men. Within 7 days of the pharmacist assessment, a study physician telephoned the subject to perform an independent clinical assessment (using the clinical questions typical to his or her practice) and judge suitability. If there was discordance between the recommendations of the pharmacist and physician, a physician who was a specialist in sexual medicine reassessed the case using all available information from the PGQ, the SHIM, and the physician’s and
1465
Feasibility of In-Pharmacy PDE5i Prescription the pharmacist’s notes. When a vague diagnosis was provided by the pharmacist, such as “heart problems,” the worst-case scenario was assumed, so that the estimate of risk was the highest possible. When a physician diagnosed “heart problems” and identified sildenafil as being suitable for the patient, the physician assessment was considered a reliable “gold standard.” The protocol was reviewed and approved by independent ethics committees where required. All patients provided written informed consent.
Study Materials Educational materials were developed, which covered the topics of ED as a medical condition, product information on sildenafil 50 mg (including instructions on proper use [dosing interval, dietary guidance predosing, need for sexual stimulation]), and additional information and practical tips for supplying the medicine in a pharmacy setting. Tailored training materials were provided, which consisted of a series of 10 fictional case studies that incorporated responses to the patient-screening questionnaire and were to be assessed for suitability using the pharmacy guidance notes of the PGQ. The study drug information sheet indicated that pharmacy supply would not be appropriate and referral to a physician would be recommended for certain patients. These included those who do not have ED; are physically unfit; are taking any nitrates or nitric oxide donors, potent cytochrome P3A4 inhibitors, a-blockers, or other ED medicine; have heart, liver, or kidney problems; have a history of certain rare eye disease (nonarteritic anterior ischemic optic neuropathy or retinitis pigmentosa); or had a recent heart attack or stroke. The PGQ was developed in focus groups with independent community pharmacists (n = 22) in various geographic locations in the United Kingdom to improve comprehensibility, acceptability, ease of use, and ability to discriminate between men who were suitable and unsuitable for supply of sildenafil 50 mg in the pharmacy setting. Wording and content were refined, and it was then translated using standard translation methodology. Additional brief questionnaires were completed by the patient and pharmacist to evaluate the patient-screening questionnaire of the PGQ and the pharmacist guidance portion of the PGQ, respectively, for ease of completion and acceptability. Statistical Analysis Sample size determination was based on a pooled projected concordance rate of ⱖ0.66 for the four
countries. Using a two-tailed a-level of 0.05, and assuming that the true concordance rates were ⱖ0.64 in Germany, Spain, and the Czech Republic (based on concordance data from the initial phase of this study performed in the United Kingdom) [17], enrollment of 66 subjects per country was required to demonstrate that the lower bound of the confidence interval (CI) for the pooled concordance rate was ⱖ0.60, which would indicate good concordance [18]. Missing data were not imputed. Data were analyzed using SPSS for Windows version 14.0 (SPSS Inc., Chicago, IL, USA). Main Outcome Measures
The primary end point was concordance between pharmacist and physician assessments of suitability for supply of sildenafil 50 mg in the pharmacy setting according to the study drug information sheet. Concordance rates were estimated with 95% CIs that were calculated using the large sample normal approximation to the binomial distribution. If the positive and negative agreement rates between the pharmacist and physician were balanced, concordance was to be assessed with the kappa statistic and would be considered substantial if there was ⱖ61% agreement [18]. Homogeneity of the concordance rates across the countries was assessed with a chi-square test; if heterogeneity was present, the overall concordance rate and CI would be based on weighted country proportions, with individual country sample sizes as the weights. There were several secondary end points: • Acceptable recommendation: Number of men not put at risk, which included the false-negative recommendations (i.e., men suitable for treatment considered unsuitable by the pharmacist). • False-positive rate: Number of men considered suitable for treatment but who were actually unsuitable/total number considered suitable • False-negative rate: Number of men considered unsuitable for treatment but who were actually suitable/total number considered unsuitable • Sensitivity of the pharmacist assessment: Number of men agreed by the sexual medicine specialist and the pharmacist to be suitable for treatment/total number considered suitable by the sexual medicine specialist • Specificity of the pharmacist assessment: Number of men agreed by the sexual medicine specialist and the pharmacist to be unsuitable for treatment/total number considered unsuitable by the sexual medicine specialist J Sex Med 2011;8:1463–1471
1466 • Positive predictive value: Number of true positives/(number of true positives + number of false positives) • Negative predictive value: Number of true negatives/(number of true negatives + number of false negatives) • Pharmacist inter-rater variability: Based on the 10 case studies • Feedback on the PGQ: Patient and pharmacist
Symonds et al.
Results
The study was designed to recruit subjects regardless of their suitability for supply of sildenafil 50 mg, to test the pharmacists’ ability to make correct recommendations. Those judged unsuitable were subjects with heart problems (n = 43), inability to walk briskly for 15 minutes without becoming breathless or experiencing chest pains (n = 39), or liver/kidney problems (n = 24). More than half of the men (n = 189 [55%]) were currently taking medicine for some other medical condition.
Patients Of the 2,182 men who responded to the advertisements, 1,147 met inclusion criteria and were eligible for entry; 515 of these decided not to participate when the study procedures were explained. Of the 632 who agreed to participate, 187 were withdrawn without undergoing any assessments because of logistic difficulties (loss of contact or inability to attend an appointment in a participating pharmacy), 35 attended only the pharmacist assessment and were withdrawn, and 64 were withdrawn as protocol violators because they either completed the physician assessment before the pharmacist assessment or were current users of a PDE5i. A total of 346 men completed the study and contributed data to the analysis: 96 from the United Kingdom, 99 from Germany, 59 from Spain, and 92 from the Czech Republic. The mean age was 52 years (standard deviation [SD] 13.6; range 18–85 years), with 61 (18%) aged ⱕ 39 years, 174 (51%) aged 40 to 59 years, and 106 (31%) aged ⱖ 60 years (age data were missing for five men). The majority of the men were white (94%). Many men (n = 143/ 335 [43%]) had a previous physician diagnosis of ED but were not currently taking a PDE5i (11 men did not respond to the question). SHIM scores indicated that 334 men (97%) had ED (Table 1), including 77 (22%) with mild ED, 109 (32%) with mild-to-moderate ED, 76 (22%) with moderate ED, and 72 (21%) with severe ED.
Appropriateness of the Pharmacist Recommendation Concordance Rates Heterogeneity was present among the countries (Table 2). Therefore, concordance rates were based on weighted country proportions, with individual country sample sizes as the weights. Concordance between the pharmacist and physician recommendations was 0.70 (95% CI, 0.66–0.74), indicating agreement between the healthcare professionals in 70% (243/346 [weighted data]) of cases. After the physician specialist in sexual medicine resolved cases for which the recommendation of the pharmacist and the physician did not agree, the adjusted concordance was 0.90 (95% CI, 0.86–0.94), indicating agreement in 90% (311/346 [weighted data]) of cases. Of the cases that the pharmacist found suitable but the physician found unsuitable, 67% (35/52) were reclassified by the sexual medicine specialist as suitable (true positive; Figure 1). Reasons for reclassification were that the physician’s cited contraindication or “unsuitable” recommendation was not supported by the study drug information sheet or the responses on the PGQ (n = 29) or because the physician’s diagnosis of no ED (n = 5) or of low desire as the etiology of the sexual dysfunction (n = 1) was contradicted by the ED diagnosis provided by the SHIM. Of the 17 cases for which the pharmacist incorrectly recommended suitability, the responses on the PGQ did not support the
Table 1 Erectile dysfunction diagnosed by score on the Sexual Health Inventory for Men vs. by patient self-report and by physician assessment ED diagnosis by SHIM score*, n (%)†
Patient self-reported ED (PGQ question 1: Do you have difficulty getting or keeping an erection that is satisfactory for sexual intercourse?) Physician diagnosis of ED
ED No ED ED No ED
ED (score ⱕ 21)
No ED (score ⱖ 22)
Total
328 6 326 8
8 (67%) 4 (33%) 12 (100%) 0
336 10 338 8
(98%) (2%) (98%) (2%)
*A score ⱕ 21 is defined as ED [16,19]. † Shading indicates positive and negative concordance. ED = erectile dysfunction; PGQ = Pharmacy Guidance Questionnaire; SHIM = Sexual Health Inventory for Men.
J Sex Med 2011;8:1463–1471
(97%) (3%) (98%) (2%)
1467
Feasibility of In-Pharmacy PDE5i Prescription Table 2
Concordance rates Rate of concordance in pharmacist recommendation of suitability for PDE5 inhibitor treatment
United Kingdom (n = 96) Germany (n = 92) Spain (n = 59) Czech Republic (n = 99) Overall (N = 346) weighted data (95% CI)
Rate of concordance in pharmacist acceptable recommendation*
vs. physicians
vs. sexual medicine specialist
vs. physicians
vs. sexual medicine specialist
0.71† 0.73† 0.59† 0.77† 0.70 (0.66–0.74)
0.83† 0.91† 0.88† 0.97† 0.90 (0.86–0.94)
0.85‡ 0.83‡ 0.65‡ 0.96‡ 0.84 (0.80–0.87)
0.92† 0.91† 0.88† 0.99† 0.93 (0.90–0.96)
*Number of men not put at risk, which included the false-negative recommendations (i.e., men who were suitable for treatment per the physician and the sexual medicine specialist, respectively but were considered unsuitable by the pharmacist). †P < 0.05 between countries. ‡ P < 0.0001 between countries. CI = confidence interval; PDE5 = phosphodiesterase type 5.
pharmacist’s recommendation (n = 11) or the contraindication cited by the physician was not reported to the pharmacist on the PGQ (n = 6). Thirteen of the pharmacist false-positive classifications were considered critical: cardiovascular symptoms with moderate exercise or cardiovascular problems (n = 10); contraindicated medications (n = 2, tamsulosin and alprostadil); and painful erections (n = 1). Of the cases that the pharmacist found unsuitable but the physician found suitable, 81% (39/48) were reclassified by the sexual medicine specialist as unsuitable (true negative; Figure 1). Reasons for the reclassification were that the physician’s recommendation was not supported by the study drug information sheet (n = 30), was at odds with his or her clinical opinion (e.g., a recommendation
of “suitable” but a comment that treatment requires prior medical assessment [n = 4]), or the contraindication identified by the pharmacist was neither included in the study drug information sheet nor cited by the physician but considered legitimate (diabetes and hypertension [n = 1], psychological state and lack of clarity about concurrent medications [n = 3], and vascular abnormalities [n = 1]). Of the nine cases for which the pharmacist incorrectly recommended unsuitability, the pharmacist recommendation was not supported by the responses on the PGQ and/or the SHIM (n = 5) or by the study drug information sheet (high alcohol intake that the pharmacist thought warranted liver function testing before sildenafil treatment [n = 3] and heart problems reported on the PGQ [n = 1]).
Figure 1 Sexual medicine specialist reassessment of cases with discordance between the pharmacist recommendation and the physician recommendation.
J Sex Med 2011;8:1463–1471
1468
Acceptable Recommendation Concordance between the pharmacist and physician recommendations, excluding the false positives, indicated that pharmacists made an acceptable recommendation in 83.5% (95% CI, 79.6–87.4%) of cases. After reclassifications by the sexual medicine specialist, pharmacists made acceptable recommendations in 92.8% (95% CI, 90.1–95.5%) of cases. False-Positive and False-Negative Rates The false-positive rate was 0.09 for pharmacists and 0.14 for physicians (9% and 14%, respectively). The false-negative rate was 0.12 for pharmacists and 0.34 for physicians (12% and 34%). Sensitivity, Specificity, and Predictive Value The pharmacist assessment showed sensitivity of 0.96 (96% of men suitable for treatment identified) and specificity of 0.74 (74% of men unsuitable for treatment identified). Pharmacists were proficient at correctly identifying men who were suitable (positive predictive value, 0.90) and unsuitable (negative predictive value, 0.88) for treatment. Of the men who did not meet the SHIM criteria for ED, treatment was recommended by pharmacists for half (6/12) vs. by physicians for 92% (11/12). Pharmacist Inter-Rater Variability The case study recommendations had an overall concordance rate across the pharmacists of 89% (95%CI, 86–92%) and an estimated overall kappa value of 0.79 (95% CI, 0.73–0.84), indicating substantial inter-rater agreement. The false-positive rate was 11%, and the false-negative rate was 10%, indicating a balanced sample. Feedback on the Pharmacist Guidance Questionnaire The patient-screening questions were rated easy or very easy to answer by 92% (315/340) of men, and 73% (245/338) felt very or quite comfortable answering the questions in their local pharmacy. Pharmacists rated the notes of the PGQ as very or quite easy to use in 96% (n = 330/344) of cases and the men’s responses as very or quite easy to interpret in 97% (n = 329/340) of cases. The median time to assess suitability for treatment was 14.5 minutes (interquartile range 89.93; range 0.17–90). Discussion
This study demonstrated that community pharmacists are cautious and responsible in recommendJ Sex Med 2011;8:1463–1471
Symonds et al. ing sildenafil 50 mg for treatment of ED. The concordance rate was 0.70 (95% CI, 0.66–0.74) between pharmacist and physician recommendation and was 0.90 (95% CI, 0.86–0.94) between pharmacist recommendation and assessment by a sexual medicine specialist. Furthermore, if the cases in which the pharmacist did not put subjects at risk (i.e., gave an acceptable recommendation) are assessed, the success rate is over 80% (0.84 [95% CI, 0.80–0.87]) between pharmacist and physician recommendation and is over 90% (0.93 [95% CI, 0.90–0.96]) between pharmacist and sexual medicine specialist. In each country, concordance rates between pharmacists and the sexual medicine specialist were at least 83% and rates of acceptable recommendation were at least 88%. Thus, pharmacists were accurate in providing appropriate treatment recommendation when using guidance from the PGQ and the study drug information sheet, which included additional restrictions compared with the current Summary of Product Characteristics. To enable appropriate patient counseling, the educational materials ensured that pharmacists were alerted to the proper use of sildenafil (i.e., not taking more than one tablet a day, avoiding heavy meals, need for sexual stimulation). In 17 cases, the pharmacist incorrectly recommended suitability, including 13 cases which were considered critical. Unfortunately, it cannot be established why the pharmacists recommended treatment suitability in these cases, against the guidance of both the PGQ and the study drug information sheet—in situations for which precautions and contraindications already exist in the current Summary of Product Characteristics. Further training materials may be necessary or cardiovascular risks and contraindicated medications could be highlighted in greater detail in the PGQ. However, it is also noteworthy that physicians recommended suitability in 30 cases in which precautions or contraindications existed that were identified by pharmacists using the PGQ. This outcome supports the results of a survey among more than 1,650 delegates of a European Association of Urology congress—only 12% of respondents reported that they assess contraindications in men asking for a PDE5i prescription [20]. Most patients and pharmacists rated the PGQ positively. The pharmacist–patient interaction, which included responding to the seven questions of the PGQ, lasted only a few minutes. Almost all pharmacists (96%) rated the PGQ notes as easy to follow and 92% of men rated the patient-
1469
Feasibility of In-Pharmacy PDE5i Prescription screening questions as easy to answer. Also, despite the potential embarrassment of talking about ED, the majority of men (73%) felt comfortable answering the questions in a local pharmacy. A potential limitation of this study is its inability to assess the clinical outcomes of actual use based on the pharmacist recommendation because, in the European Union, supply and subsequent patient follow-up are currently only possible for prescription-only medicines under medical supervision. This is in contrast to the United States, where product supply and patient behavior can be studied before a potential switch to nonprescription status [21,22]. Another possible limitation is that the study was conducted in only four EU member states. However, the concordance rates between pharmacist and sexual medicine specialist in the blinded review being consistently >80% suggest that differences in culture and daily practice across the countries have little impact on the pharmacist’s ability to adequately assess PDE5i suitability. Thus, it is reasonable to assume that the study outcomes could be representative of all EU countries, all of which apply the same professional pharmacy principles [1]. Another potential limitation is that ED is not conventionally managed by telephone consultation, but by face-to-face consultation with a physician, whereas pharmacist assessment is likely to be similar to the method here. Finally, because the results of the study, including the diagnosis of ED, are based on patient-reported clinical data, the accuracy of the assessments and the strength of the results are dependent upon the openness and honesty of the patients, whether the healthcare professional eliciting the data is a pharmacist or a physician. For example, a diagnosis of ED is based on the selfreported symptom of inability to attain and maintain penile erection sufficient to permit satisfactory sexual intercourse [23]. In the current study, inclusion was based on a simple question about ED symptoms, which was identical to the first question on the PGQ and was asked by an anonymous call center operator. This question screened out approximately half of the men responding to the study advertisement, such that 98% of the men included in the study population had ED using a gold-standard questionnaire (SHIM) [19]. This result supports the validity of pharmacist use of the initial question of the PGQ to diagnose ED and to screen out those without ED in men requesting sildenafil 50 mg. ED often occurs concomitantly with other diseases (e.g., diabetes and cardiovascular disease) or
with conditions that are associated with an increased risk for a cardiovascular event (e.g., smoking, obesity, hypercholesterolemia, and hypertension) [24,25]. Based on a study conducted in the United Kingdom, most men receiving a prescription for a PDE5i already have a medical diagnosis of an underlying condition such as hypertension or hypercholesterolemia, and only a few men receive a new such diagnosis at the time of PDE5i prescription [26]. Furthermore, an observational study conducted in Spanish community pharmacies of men asking for ED advice or treatment showed that there were no statistically significant differences in the severity of ED or the comorbidity profile between those men who approached the pharmacist with or without a prescription for a PDE5i [11]. Thus, community pharmacists can play an important role in identifying patients with ED, providing education and counseling on potential cardiovascular risk factors and recommending further medical care [12]. A reminder for further medical care to check for potential underlying conditions was also highlighted in a more recent version of the patientscreening questionnaire, used in subsequent research [27]. Conclusions
In this feasibility study in several EU member states, which included men across a broad age range (18–85 years), pharmacists provided appropriate recommendation for supply of sildenafil 50 mg in the pharmacy setting according to the study drug information sheet in 90% of cases, closely following the guidance within the PGQ (based on the study drug information sheet), not recommending for men without ED, and generally recommending physician assessment when appropriate. This suggests that pharmacists are capable of assessing patient suitability for ED treatment with sildenafil 50 mg. Acknowledgment
The authors would like to thank Ms. Anwen Russell of TBWA Health for her invaluable support. Editorial support was provided by Deborah M. CampoliRichards, BSPHA, RPh, of Complete Healthcare Communications, Inc., and was funded by Pfizer Inc. Corresponding Author: Tara Symonds, PhD, Pfizer Ltd Global Research and Development, Outcomes Research (IPC160) Ramsgate Road, Sandwich, Kent J Sex Med 2011;8:1463–1471
1470
Symonds et al.
CT13 9NJ, UK. Tel: +44 (1304) 641553; Fax: +44 (1304) 658823; E-mail: tara.symonds@pfizer.com Conflict of Interest: John D. Dean is President and CEO of the International Society for Sexual Medicine (Incorporated in California); a consultant/advisor and meeting participant/lecturer for Bayer Schering Pharma, Boehringer-Ingelheim, Johnson & Johnson, Lilly, Pfizer, Plethora, Procter & Gamble, and ProStrakan; and a scientific study/trial investigator for Bayer Schering Pharma, Boehringer-Ingelheim, Lilly, Pfizer, Plethora, Procter & Gamble. Alison Carr is an employee of Hamell Communications, an independent healthcare research company that was a paid consultant to Pfizer in designing and overseeing the running of the study and analyzing the data. At the time of this research, Martin Carlsson, Aditya Marfatia, Gabriel Schnetzler, and Tara Symonds were employees of Pfizer.
4
5
6
7
8
9
10
Statement of Authorship
Category 1 (a) Conception and Design Tara Symonds; John D. Dean; Alison Carr; Martin Carlsson; Aditya Marfatia; Gabriel Schnetzler (b) Acquisition of Data Tara Symonds; Martin Carlsson; Gabriel Schnetzler (c) Analysis and Interpretation of Data Tara Symonds; John D. Dean; Alison Carr; Martin Carlsson; Aditya Marfatia; Gabriel Schnetzler
11
12
13 14
Category 2 (a) Drafting the Article Tara Symonds; John D. Dean; Alison Carr; Martin Carlsson; Aditya Marfatia; Gabriel Schnetzler (b) Revising It for Intellectual Content Tara Symonds; John D. Dean; Alison Carr; Martin Carlsson; Aditya Marfatia; Gabriel Schnetzler
15 16
Category 3 (a) Final Approval of the Completed Article Tara Symonds; John D. Dean; Alison Carr; Martin Carlsson; Aditya Marfatia; Gabriel Schnetzler References 1 Pharmaceutical Group of the European Union (PGEU). The commitment of community pharmacists in Europe. Available at: http://www.pgeu.eu/tabid/596/Default.aspx (accessed July 20, 2009). 2 Anderson C, Bates I, Beck D, Brock TP, Futter B, Mercer H, Rouse M, Whitmarsh S, Wuliji T, Yonemura A. The WHO UNESCO FIP pharmacy education taskforce. Hum Resour Health 2009;7:45. 3 Hughes CM, Hawwa AF, Scullin C, Anderson C, Bernsten CB, Bjornsdottir I, Cordina MA, da Costa FA, De Wulf I, Eichenberger P, Foulon V, Henman MC, Hersberger KE, Schaefer MA, Sondergaard B, Tully MP, Westerlund T,
J Sex Med 2011;8:1463–1471
17 18 19
20
21
22
23
McElnay JC. Provision of pharmaceutical care by community pharmacists: A comparison across Europe. Pharm World Sci 2010;32:472–87. Berenguer B, La Casa C, de la Matta MJ, Martin-Calero MJ. Pharmaceutical care: Past, present and future. Curr Pharm Des 2004;10:3931–46. Eickhoff C, Schulz M. Pharmaceutical care in community pharmacies: Practice and research in Germany. Ann Pharmacother 2006;40:729–35. Herborg H, Sorensen EW, Frokjaer B. Pharmaceutical care in community pharmacies: Practice and research in Denmark. Ann Pharmacother 2007;41:681–9. Gastelurrutia MA, Faus MJ, Fernandez-Llimos F. Providing patient care in community pharmacies in Spain. Ann Pharmacother 2005;39:2105–10. Bacon L, Savage I, Cook S, Taylor B. Training and supporting pharmacists to supply progestogen-only emergency contraception. J Fam Plann Reprod Health Care 2003;29:17–22. Hansford D, Cunningham S, John D, McCaig D, Stewart D. Community pharmacists’ views, attitudes and early experiences of over-the-counter simvastatin. Pharm World Sci 2007;29: 380–5. Sinclair HK, Bond CM, Stead LF. Community pharmacy personnel interventions for smoking cessation (Review). Cochrane Libr 2009;1–14. Martin Morales A, Ibáñez J, Machuca M, Pol-Yanguas E, Schnetzler G, Pascual Renedo V. The EPIFARM study: An observational study in 574 community pharmacies in Spain characterizing patient profiles of men asking for erectile dysfunction medication. J Sex Med 2010;7:3153–60. Vic JM. Specific aspects of erectile dysfunction and its treatment for community pharmacists. Int J Impot Res 2004;16: S50–2. Mayer S. UK pilot allows pharmacists to supply sildenafil without prescription. BMJ 2007;334:387. Dean J, Carr A, Carlsson M, Marfatia A, Schnetzler G, Symonds T. Correlation between community pharmacist and physician recommendations for treatment with a phosphodiesterase type 5 inhibitor. Poster presented at the Joint Congress of the European and International Societies for Sexual Medicine, Brussels, Belgium, December 7–11, 2008. J Sex Med 2009;6(suppl 2):135. VIAGRA Summary of Product Characteristics. European Medicines Agency, London, UK, 2008. Rosen RC, Cappelleri JC, Smith MD, Lipsky J, Peña BM. Development and evaluation of an abridged, 5-item version of the International Index of Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction. Int J Impot Res 1999;11: 319–26. Pfizer Inc. Data on file. Pfizer Inc, New York, NY. 2009. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74. Cappelleri JC, Rosen RC. The Sexual Health Inventory for Men (SHIM): A 5-year review of research and clinical experience. Int J Impot Res 2005;17:307–19. Schnetzler G, Marfatia A, Zou KH, Muniain J, Symonds T. Current erectile dysfunction management: A survey among delegates of the 2008 annual EAU congress in Milan. Eur Urol 2009;(suppl 8):212(#368). Schwartz SM, Bansal VP, Hale C, Rossi M, Engle JP. Compliance, behavior change, and weight loss with orlistat in an over-the-counter setting. Obesity (Silver Spring) 2008;16: 623–9. Brass EP. Consumer behavior in the setting of over-thecounter statin availability: Lessons from the consumer use study of OTC Mevacor. Am J Cardiol 2004;94:22F–9F. NIH Consensus Development Panel on Impotence. NIH Consensus Conference. Impotence. JAMA 1993;270:83–90.
Feasibility of In-Pharmacy PDE5i Prescription 24 Thompson IM, Tangen CM, Goodman PJ, Probstfield JL, Moinpour CM, Coltman CA. Erectile dysfunction and subsequent cardiovascular disease. JAMA 2005;294:2996–3002. 25 Hodges LD, Kirby M, Solanki J, O’Donnell J, Brodie DA. The temporal relationship between erectile dysfunction and cardiovascular disease. Int J Clin Pract 2007;61:2019–25. 26 Kirby M, Marfatia A, Zou KH, Symonds T, Schnetzler G. Diagnosis of underlying disease in men with erectile
1471 dysfunction receiving phosphodiesterase type 5 inhibitors in the United Kingdom. Eur Urol 2009;(suppl 8):172(#208). 27 Symonds T, Dean J, Coyne KS, Margolis MK, Hackett G, Edwards D, Marfatia A, Schnetzler G. The ability of men to assess their suitability to take 50 mg sildenafil: An assessment of the comprehension of patient information materials. J Sex Med 2010;7:2217–25.
J Sex Med 2011;8:1463–1471