Influence of pharmacists' opinions on their dispensing
medicines without requirement of a doctor's prescription
Francisco Caamaño / Manuel Tomé-Otero / Bahi Takkouche / Juan Jesús Gestal-Otero
Department of Preventive Medicine and Public Health. University of Santiago de Compostela.
Santiago de Compostela. A Coruña. Spain.
Objective: To assess the influence of pharmacists' opinions on their dispensing medicines with a «medical prescription only» label without requiring a doctor's prescription.
Objetivo: Evaluar la influencia de las opiniones de los farmacéuticos que no solicitan receta médica para dispensar fármacos que la requieren para su venta.
Correspondencia: Francisco Caamaño Isorna. Departamento de Medicina Preventiva y Salud Pública. Facultad de Medicina.
San Francisco. s/n. 15705 Santiago de Compostela. A Coruña. Spain.
Correo electrónico: email@example.com
Recibido: 11 de mayo de 2004. Aceptado: 19 de octubre de 2004.
The professional activity of the community pharmacist has changed dramatically in the last decades. With the increase of ready-to-use drugs, the activity in a pharmacist's laboratory experienced a sharp decrease1. Today, the main health-related activity of a pharmacist is to assure quality of dispensing2,3.
Several studies associated quality of dispensing with factors such as pharmacist's age, educational background and social and demographic factors of the practice4-6. The pharmacists' opinions about their activity have been proposed as potential determinants of the quality of dispensing7,8. However, these factors have been analyzed individually and so far, no comprehensive theoretical model has been proposed to explain their effect.
According to previous studies2,7-12, we propose a model to investigate the opinions of the pharmacist that are associated with dispensing. Following this model, dispensing by the pharmacists is determined by their opinions about the prescription practice of the physicians, about their own competence to prescribe and about the pharmacists' responsibility in the control of consumption of medicines. The model also considers that the dispensing practice is subject to changes that are due to the workload of the pharmacist and to the socio-economic and socio-cultural characteristics of the customers.
The aim of this study is to assess the validity of the model proposed and to measure the effect on the quality of dispensing (measured as the requirement of a medical prescription to dispense) exerted by each opinion.
Design, population, sample
We carried out a cross-sectional study of the population of community pharmacists in Northwest Spain (n = 875, who work in 490 pharmacies). We used multistage cluster sampling (pharmacies are clusters and pharmacist are the population). Pharmacies were used as sampling units in the first stage (n = 150), and pharmacists were sampled at random within pharmacies. All pharmacists present during the interviewer's visit were selected (n = 166). The schedule of the visit to the pharmacy was selected at random during the opening hours.
We collected data collection by means of a personal interview (February and March, 2002) with a closed questionnaire, a method that can maximize participation13.
The interviews were carried out by a trained qualified interviewer (pharmacist). The questionnaire included four blocks of questions: a) socio-demographic variables and variables of the formation of the pharmacist; b) practice in relation to requirement of prescription to dispense different drugs; c) level of agreement with 24 items about prescription practice of the doctors, pharmacist's qualification to prescribe, responsibility of the pharmacist about dispensed drugs, clients' qualification for self-medication, and pharmacists' perception of their work, and d) services offered in the pharmacy and characteristics of its socio-cultural and socio-economic surroundings.
Pharmacists were asked about whether they had specialty training. In Spain, pharmacists may work either with or without specialty degrees. The three-year special training is officially regulated and takes place in pharmacies of the National Health Service. During the three year special training, the pharmacists work in a hospital pharmacy and they study clinical pharmacology, pharmacoepidemiology and clinical epidemiology. So, specialty is a dichotomous variable: yes or no.
Pharmacists were asked about their work status. Two situations were considered: a pharmacist who is the owner or responsible for the pharmacy and a pharmacist who is under contract. So, work status is a dichotomous variable: owner or responsible (yes or no).
The models also included two variables that measure the work environment of the pharmacist: number of pharmacists in the pharmacy: number of pharmacist who work in the pharmacy (this variable is used as a proxy of the customers per day of the pharmacy) and the socio-economic level of the population attended. The socio-economic level of the population was valued through the perceptions of the pharmacists in charge, by means of a 5-point Likert-like response scale ranging from «very low» to «very high».
The opinions were measured by asking the pharmacists about their level of agreement with 16 items (from 0 = completely in disagreement, to 10 = completely in agreement). The items were grouped in 5 topics: prescription practice of the doctors (3 items); pharmacist's qualification to prescribe (3 items); responsibility of the pharmacist about dispensed drugs (5 items); clients' qualification for self-medication (2 items), and pharmacists' perception of their work (3 items).
Requirement. We gathered information on the requirement of a prescription to dispense 5 drugs (no requirement = 0; requirement = 1), which in Spain have to be dispensed with a prescription (drugs with «medical prescription only» label). These drugs included an oral contraceptive (Diane 35®), an angiotensin-converting enzyme (ACE) inhibitor (Capoten®), an antibiotic (Clamoxyl®), a benzodiazepine (Lexatin®), and a nonsteroidal anti-inflammatory drug (Voltaren®). Prescription requirement was considered for 5 drugs and two types of client: well-known customers (habitual customers) and unknown customers. So, the variable ranges from 0 (the pharmacist does not demand any prescription for any of the 5 drugs, either for well-known or unknown customers) to 10 (the pharmacist demands a prescription for the 5 drugs for all customers).
We measured the effect that each of the opinions exerts on the variable requirement through linear regression. We adjusted the multiple linear regression models for the variables: age of the pharmacist, work status, specialty, number of pharmacists in the pharmacy and perception of socio-economic level of the population. There variable were associated with the quality of dispensing in previous studies6.
Other candidates for potential confounders included sex, number of years of work, former work experience, additional degree of studies, and urban/rural environment. These covariates were introduced in the final model if their inclusion changed the coefficient of the main independent variable by more than 10%. SPSS package was used to analyze the data.
Of the 150 pharmacies sampled, 4 were excluded (due to vacation closing). Out of the 146 remaining pharmacies, 123 participated (84.2%). Of the 166 pharmacists, 164 participated in the study (98.8%). Table 1 shows the characteristics of the participants. The mean age of the pharmacists was 40.2 years and their average work experience was 12.3 years. Our sample showed a proportion of 72% of women and a proportion of pharmacists with a specialist degree obtained though the National Health Service of 3.1%.
Table 2 shows the proportion of pharmacists that demanded a prescription for dispensing the five drugs considered, according to the type of client. Table 3 shows the variables in the multivariate model for the dependent variable requirement. Age of the pharmacist, work status, specialty, number of pharmacists in the pharmacy, and perception of socio-economic level of the population, were associated to the quality of dispensing.
Table 4 presents the influence of the pharmacist's opinions on the variable requirement, the agreement with item (mean), the regression coefficients (β), the p-value and the proportion of multivariate explained variance (r2). Pharmacists who have a heavier workload and who underestimate the physicians' qualification to prescribe but overestimate their own qualification to prescribe required medical prescriptions less often. Those pharmacists who stress the importance of their duty in rationalizing the consumption of drugs demanded medical prescriptions more often.
According to our model, the dispensing practice of the pharmacists is associated with their opinions on the prescription practice of the physicians, on the qualification of the pharmacist to prescribe, and on the perception of the pharmacist's responsibility toward the rational use of drugs.
Prescription practice of the doctors. The association between pharmacists' perception of the quality of the medical prescription and prescription requirement (table 4, items 1 and 2) reflects the doubts that pharmacists raise on the adequacy of the diagnosis and prescription by the physician. The pharmacist who is not convinced by the adequacy of the physician's diagnosis will be more flexible towards the mandatory character of the prescription demand. These results are consistent with those found in Nepal in a study of excessive prescription and quality of dispensing9. The similarity of the results of these studies that were carried out in very different cultural environment and with different methodologies suggests that the relation between those variables is causal14.
Furthermore, the high degree of agreement of the pharmacists with the third affirmation «doctors do not stop enough to give explanations to their patients», (mean, 6.32; 95%CI, 5.98-6.68) and the negative correlation with the variable requirement show that the lower confidence in the clinical practice is associated with a higher independence in the behavior of the pharmacist.
Pharmacists' qualification to prescribe. We analyzed this determinant of the quality of dispensing by means of 3 items. The first and very explicit affirmation, «in minor pathologies, the pharmacist is as prepared as a doctor to prescribe» is negatively correlated with a lower demand of a medical prescription (p = 0.048). The items 5 and 6, in which the level of knowledge of the patient is assessed, are also negatively associated with the variable requirement, although this association is not statistically significant.
These results show that the opinions of the pharmacists about their qualification to prescribe translate into dispensing practice6. This practice is facilitated by their large autonomy and the scarce control that exists over dispensing in Spain15. To improve the quality of dispensing, it is necessary to implement a stricter administrative control over dispensing. Pharmacist's prescription could be accepted only for over-the-counter drugs and continuation of treatments initially prescribed by a physician16.
Responsibility on dispensed drugs. The higher degree of agreement with affirmation 7 and 8 «never dispense a drug with a «medical prescription only» label without a medical prescription» and «the pharmacist must ask the client what the drug is for if he/she does not have a prescription» is significantly related to a higher demand of the medical prescription (p = 0.015 and p = 0.006, respectively). Those pharmacists who perceive as important their role in the rationalization of the consumption of drugs dispense in a stricter manner and demand every legal contro15,17. The higher degree of agreement with the remaining items (9, 10 and 11) is correlated positively but not statistically significantly with the variable requirement.
According to the proposed model, the interaction of the pharmacists with their environment will modify their dispensing practice. Our study measured two aspects: pharmacists' perception of their work and opinion on the qualification of the customer for self-medication.
The pharmacist's perception of the customers' qualification for self-medication is not significantly associated with prescription requirement. However, both items 1 and 2 show a negative correlation with the prescription requirement: those pharmacists who consider that «a drug prospectus can be understood by patients» and that «the population makes a rational use of medicines» tend to require the medical prescription less often. In addition, if we take into account that the model is adjusted for the pharmacists' perception of the socio-economical level of the population (p = 0.016), and that this level and the qualification for self-medication are highly correlated17, we cannot rule out the possibility that the lack of statistical significance is due to over-adjustment.
In relation to pharmacists' perception of their work, previous studies associated a large number of customers with less pharmaceutical advice10,11. Other studies demonstrated the existence of a relation between large dispensing loads and the number of dispensing mistakes12. Our results confirm those results, as item 14 «the pharmacist puts up with an excessive work-load» shows a negative correlation with the variable requirement, which is statistically significant (p = 0.045). This could be explained by the fact that those pharmacies with excessive workload function in a more commercial way, reducing the time a pharmacist spends with each customer and exerting less control over prescriptions6.
Previous studies found that a low degree of pharmaceutical advice is related to a high cultural degree of the population attended11,18; a similar relation exists between medical prescriptions and the level of autonomy of the patient19-21.
Four possible limitations to this study have to be considered. First, since the current analysis is based on cross-sectional data, the validity of the conclusions could be limited by the difficulty in differentiating between cause and effect. However, in our study, those factors that are associated with quality dispensing are variables that are unlikely to change during the period of time in which the dependent variable is measured. A cross-sectional analysis provides then results that are close to those of a longitudinal one14. Second, the questionnaire has not been validated. Third, the measure of the prescription requirement of the pharmacist could be biased by the pharmacist's tendency to provide answers that are legally acceptable. The indicators are then probably too optimistic as far as it concerns the evaluation of the quality of dispensing. Finally, the quality of dispensing is assessed only by the prescription requirement. There are probably other indicators the measure of which was not feasible in our study22. In conclusion, our results suggest that, in order to increase the quality of dispensing, it is necessary: a) to stress the importance of the duty of the pharmacists in controlling the consumption of medicines; b) to optimize the pharmacies workload, and c) to improve the perception of the prescription practice of the physicians among pharmacists. It is also necessary to restrict drug prescription by the pharmacist to over-the-counter drugs and to continuation of treatments initially prescribed by a physician.
Acknowledgements This work was funded by the Spanish Ministry of Health (Grant FIS 01/1688). We thank María Xesús Cebro for her comments during the drafting of this report.
1. Rawlin MD. Extending the role of the community pharmacist. BMJ. 1991;302:427-8.
2. Rupp MT. Value of community pharmacists' interventions to correct prescribing errors. Ann Pharmacother. 1992;26:1580-4.
3. Beney J, Bero LA, Bond C. Expanding the roles of outpatient pharmacists: effects on health services utilisation, costs, and patient outcomes. Cochrane Database Syst Rev. 2000; 3:CD000336.
4. Odedina FT, Hepler CD, Segal R, Miller D. The pharmacists' implementation of pharmaceutical care (PIPC) model. Pharm Res. 1997;14:135-44.
5. Yesalis CE 3rd, Norwood GJ, Lipson DP, Helling DK, Fisher WP, Burmeister LF. Capitation payment for pharmacy services: impact on generic substitution. Med Care. 1980;18:816-28.
6. Caamaño F, Tomé-Otero M, Figueiras A, Takkouche B. Factors related with prescription requirement to dispensing in Spain. Pharmacoepidemiol Drug Saf. 2004;13:405-9.
7. Pendergast JF, Kimberlin CL, Berrardo DH, McKenzie LC. Role orientation and community pharmacists' participation in a project to improve patient care. Soc Sci Med. 1995;40:557-65.
8. Cancrinus-Matthijsse AM, Lindenberg SM, Bakker A, Groenewegen G. The quality of the professional paractice of community pharmacists: what can still be improved in Europe? Pharm World Sci. 1996;18:217-28.
9. Holloway KA, Gautam BR. Consequences of over-prescribing on the dispensing process in rural Nepal. Trop Med Int Health. 2001;6:151-4.
10. Rupp MT, DeYoung M, Schondelmeyer SW. Prescribing problems and the pharmacist interventions in community practice. Med Care. 1992;30:926-40.
11. Raisch DW. Patient counseling in community pharmacy and its relationship with prescription payment methods and practice settings. Ann Pharmacother. 1993;27:1173-9.
12. Guernsey BG, Ingrim NB, Hokanson JA, Doutre WH, Bryant SG, Blair CW, et al. Pharmacists' dispensing accuracy in a high-volume outpatient pharmacy service: focus on risk management. Drug Intell Clin Pharm. 1983;17:742-6.
13. Caamaño F, Ruano A, Figueiras A, Gestal-Otero JJ. Data collection methods for analyzing the quality of the dispensing in pharmacies. Pharm World Sci. 2002;24:217-23.
14. Rothman KJ, Greenland S. Types of epidemiology study. En: Rothman KJ, Greenland S, editors. Modern epidemiology. Philadelphia: Lippincott and Raven; 1998. p. 75-6.
15. Guerra L. Informe sobre resistencia microbiana: ¿qué hacer? Med Clin (Barc). 1995;106:267-79.
16. Dowell J, Cruikshank J, Bain J, Staines. Repeat dispensing by community pharmacists: advantages for patients and practitioners. B J Gen Pract. 1998;48:1858-60.
17. Figueiras A, Caamaño F, Gestal-Otero JJ. Sociodemographic factors related to self-medication in Spain. Eur J Epidemiol. 2000;16:19-26.
18. Ruegg A. Contribution of the pharmacist to safety in self medication. Soz Praventivmed. 1986;31:160-4.
19. Chewning B, Sleath B. Medication decision-making and management: a client-centered model. Soc Sci Med. 1996;42:389-98.
20. Soctt A, Shiell A, King M. Is general practitioner decision making associated with patient socio-economic status? Soc Sci Med. 1996;42:35-46.
21. Virji A, Britten N. A study of the relationship between patients' attitudes and doctors' prescribing. Fam Pract. 1991;8:314-9.
22. Mullins CD, Baldwin R, Perfetto EM. What are outcomes? J Am Pharm Assoc (Wash). 1996;36:39-49.