- Email: [email protected]
Improving private pharmacy practice
Journal of Clinical Epidemiology 55 (2002) 1148–1155
Improving private pharmacy practice: A multi-intervention experiment in Hanoi, Vietnam Nguyen T.K. Chuca,b,*, Mattias Larssonb, Nguyen T. Doc, Vinod K. Diwanb,d, Goran B. Tomsonb,e, Torkel Falkenbergb a Hanoi Medical University, Hanoi, Vietnam Division of International Health (IHCAR), Department of Public Health Sciences, Karolinska Institutet, SE171 76 Stockholm, Sweden c Hanoi College of Pharmacy, Hanoi, Vietnam d Nordic School of Public Health, Gothenburg, Sweden e Medical Management Centre, Karolinska Institutet, Sweden
b
Received 3 October 2001; received in revised form 19 May 2002; accepted 21 May 2002
Abstract The objective of this study was to assess the effects of a multicomponent intervention on private pharmacy practice. From 641 private pharmacies in Hanoi, 68 pharmacies were randomly selected and matched into 34 pairs. Each pair consisted of a control and an intervention pharmacy. Three interventions were applied sequentially: Regulatory enforcement, Education, and Peer influence. Four tracer conditions were selected: uncomplicated acute respiratory infection (ARI), sexually transmitted disease (STD), requesting the prescription-only drugs prednisolone, and a short course of cefalexin. Practice was assessed through the Simulated Client Method (SCM). The intervention pharmacies improved significantly compared to the control pharmacies (P .05) in all tracer conditions. For ARI, antibiotic dispensing decreased (P .02) and questions regarding breathing increased (P .01). For STD, advice to go to the doctor and dispensing the correct syndromic treatment increased (P .01). Dispensing of prednisolone and cefalexin decreased (P .01) and prescription requests increased (P .01). Our conclusion is that it is possible to improve private pharmacy practice with a multicomponent intervention. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Private pharmacy practice; Randomized controlled trial; Multi-intervention; Simulated Client Method (SCM); Acute Respiratory Infection (ARI); Sexually Transmitted Disease (STD)
1. Introduction In many low income countries, pharmaceuticals make up 50% or more of health care costs [1]. With health sector reforms, private pharmacies are becoming increasingly the first and only contact with health delivery services [2–4]. “Prescribing” by pharmacy staff sometimes can result in irrational use of drugs, waste of scarce resources, and the sale of harmful drugs, possibly due to the “business-profession” dilemma in private pharmacy practice [5–7]. In 1986, an economic reform, known as Doi Moi, was launched in Vietnam introducing market economy with privatization in all sectors of society including drug provision. Pharmacists with 5 years or more of working experience were allowed to open private pharmacies [8], and the number of private pharmacies increased from none in 1986
* Corresponding author. Fax: 0046-83-11-590. E-mail address: [email protected] (N.T.K. Chuc).
to more than 6,000 in 1996 [9]. This, combined with the increased domestic production of drugs and drug imports, considerably improved the availability of drugs. However, the regulations governing the pharmaceutical sector and information to providers, prescribers, and users could not keep up with the rapid increased supply of drugs [10]. Currently, most patients buy drugs including antibiotics from private pharmacies without prescription or adequate user instructions [8,11,12]. Standards to improve the quality of pharmacy services have been set out in the Good Pharmacy Practice (GPP) document. In line with the greater demand for pharmacy services, the pharmacists’ new role in the health system includes being a communicator, a quality drug supplier, a trainer, and supervisor and a health promoter [13]. Professional practice in private pharmacies is influenced by many factors including staff education, customer demand, physician practice, regulations, and economic incentives [2,4,5]. Regulation is one important factor influencing pharmacy practice [2,14]. Even with the existing regulations, effective
0895-4356/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S0895-4356(02)00 4 5 8 - 4
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155
1149
Fig. 1. Sequence and time intervals between the interventions.
enforcement mechanisms are often absent in low and middle income countries [15,16]. There are few studies published showing the effects of regulatory enforcement at pharmacies in low income countries [16]. Good knowledge is a prerequisite for good practice. However, knowledge does not always guarantee good practice [12,17]. Educational interventions that are effective have been shown in some studies [7,18,19]. With growing competition, economic incentives are important, and customer demand has an increasing influence on practice [4,8]. The social and professional culture within a profession may have a major impact on the practice [5]. Peer influence, building on the concept of group responsibility for self-education and monitoring, as well as multicomponent interventions, have shown to be effective in improving quality of service in the public sector of high income countries [14,19]. However, there is no research published combining different types of interventions in the private sector of low income countries. The aim of this study was to evaluate the effectiveness of a multi-intervention package on pharmacy practice using tracer conditions: uncomplicated acute respiratory infection (ARI), sexually transmitted disease (STD), requesting the prescription-only drugs prednisolone, and a short course of cefalexin. This study is part of the Towards Good Pharmacy Practice in Thailand and Vietnam project. The members of the team jointly developed the design and the instruments used in this study.
whether or not the pharmacist was the license holder; 3. whether or not the pharmacy was close to a hospital; and 4. located in a different ward or more than 150 meters from all other pharmacies selected. The sample size was chosen to detect a difference in degree of 20% between the intervention and control group using the t-test with a significance level of 5% and a power of 90%. This sample size was assumed to be applicable also for the Wilcoxon signed rank test used in the final analysis. The pharmacies in each pair were randomly assigned as intervention or control. 2.2. Tracer conditions Four tracer conditions of public health importance were selected: (1) case management of an uncomplicated upper respiratory tract infection (ARI) in a child under 5 years of age with a mild cough. (2) Case management of a sexually transmitted disease (STD), presented as an urethral discharge in an adult man. (3) Request to purchase a couple of capsules of an antibiotic (cefalexin) without a prescription.
2. Method 2.1. Study design, sampling, and sample size A clustered randomized controlled trial with a time series design and a multicomponent intervention was carried out (Fig. 1). There were 789 private pharmacies registered in the urban area of Hanoi; 148 located inside a hospital or mainly wholesalers were excluded. From the remaining 641 pharmacies 34 pairs were randomly selected (Fig. 2) according to the following matching criteria: 1. turnover: high, medium, or low according to district inspectors; 2.
Fig. 2. Sample size and sampling.
1150
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155
(4) Request to purchase a couple of tables of a steroid (prednisolone) without a prescription.
quests for a short course of antibiotics; in the second, STD case management and requests for oral steroids.
2.3. Simulated Client Method (SCM)
3.3. Peer influence (PI)
The Simulated Client Method (SCM) [20] was used to assess private pharmacy practice. Four rounds of SCM were conducted, a baseline and after (1 month) each of the three interventions. Twenty simulated clients were selected, five per tracer condition, and trained to act in a reproducible way according to the tracer condition scenario (detailed information is available on request from the first author). In the first three SCM rounds the simulated clients were research assistants from the Hanoi College of Pharmacy and in the last SCM round from the Hanoi Medical University. For every SCM round, each simulated client encountered all the selected pharmacies, presented the scenario, bought the drugs recommended by the pharmacy staff, and filled in the record sheets at a maximum of 15 min after leaving the pharmacy. The drugs purchased and the record sheets were submitted to the research team at the College of Pharmacy. The simulated clients knew that they were measuring the practice of drug sellers but did not know which pharmacies belonged to the intervention and control group.
The intervention pharmacies were geographically divided into five groups. The research group appointed one person in each group to be the leader based on the experience gained during the educational intervention. The five group leaders and representatives of the Hanoi Pharmacy Association attended a 1-day seminar where the research group informed about the peer influence strategy and reviewed case management of the four tracer conditions. Each group then organized a meeting where the group leader informed about the peer influence strategy and instructed the participants to make case management records of pharmacy customers with complaints related to the tracer conditions. Three meetings were held during 3 months where the participants presented and discussed their records. A summary meeting with all participating intervention pharmacy staff was held to discuss the intervention strategy. The intervention pharmacies did know that the interventions were taking place, but did not know about the study design or the SCM. Ethical approval was obtained from the Ministry of Health, Vietnam, and from the Karolinska Institutet. The study was done in collaboration with the Hanoi Pharmacy Association and the Hanoi Health Bureau.
3. Interventions 3.1. Regulatory enforcement (RE) The first intervention, done in collaboration with the Hanoi Health Bureau, focused on the regulation of prescriptiononly drugs, particularly the drugs related to tracer conditions in the study (antibiotics, with emphasis on cefalexin and perfloxacin, and corticosteroids including prednisolone). Four inspectors from the Hanoi Health Bureau were trained to cover these areas of inspection. In pairs they visited the intervention pharmacies twice with an interval of 1 month. During the first visit, in addition to normal inspection procedures, they disseminated and explained the regulations regarding prescription-only drugs [21]. Cefalexin and prednisolone were used as examples of prescription only drugs that should be dispensed only if a prescription was provided (detailed information is available on request from the first author). In the second visit, these messages were repeated. 3.2. Education (ED) Pharmacy treatment guidelines for the four tracer conditions were developed based on the national treatment guidelines by the research team in collaboration with Hanoi Health Bureau, the Pharmacy Association, experienced clinicians, and pharmacologist (detailed information is available on request from the first author). Two senior research team members and two clinicians conducted the intervention. Each intervention pharmacy received two 45-min face-to-face educational sessions. Written and verbal information regarding the importance of GPP [13], and adequate questions, advice, and treatments in relation to the four tracer conditions was provided. In the first session, the topics were ARI case management and re-
3.4. Data analysis The study unit was the pharmacy. The results were coded into computers using Microsoft Access software. SPSS 10.0 and SAS 8.0 software were used for the data analysis. Analysis of effects of the whole intervention package was done using the method of summary statistics for the selected tracers and variables [22]. This method is based on finding one or a few statistics that summarize the information of interest in a longitudinal series. The individual effect is then separated from the cohort effect, and the analysis incorporates the within-pharmacy dependence. In the study, the main question was about the difference in trend between the intervention and control pharmacies over the study period. Here, the regression coefficient, that is, the slope, from a linear regression on each pharmacy was selected as a suitable summary statistic. The tracer variable measured with SCM, for example, antibiotics dispensed, was used as the dependent variable, and the four rounds of SCM were used as the independent variable. The slopes of the selected variables were compared in pairs using the Wilcoxon signed-rank test. A significant difference (P .05, two tailed) was interpreted as a significant effect of the intervention package.
4. Results During the study, four pharmacies were excluded because of irregular opening hours. As a consequence, the other phar-
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155
macy in each pair was also excluded as the results were analyzed in pairs. In total, eight pharmacies were excluded. Thirty-four intervention pharmacies received regulatory enforcement, 30 received educational face-to-face sessions, and 29 took part in the peer influence intervention; one refused for the reason that it was “too busy.” In the cross-sectional analysis 60 pharmacies with 4,445 encounters were used, in the longitudinal analysis 58 pharmacies with 4,356 encounters. The estimated total cost of the interventions was approximately 5,700 USD (less than 200 USD per pharmacy). 4.1. ARI case management After the interventions, dispensing of antibiotics decreased in the intervention pharmacies (from 45 to 30%) while it increased in the control pharmacies (from 39 to 42%) with a significant difference in trends (P .05). Questions regarding breathing increased (11 to 30%) in the intervention pharmacies and decreased in the control pharmacies (10 to 7%) with a significant difference in trends (P .01) (Table 1). Dispensing of acceptable drugs, defined as paracetamol, cough syrups, and traditional medicines not containing antihistamines and opiates, showed no significant difference in trends between the two groups. 4.2. STD case management Advice regarding consultation with a physician increased in the intervention pharmacies (from 22 to 37%) compared to the control pharmacies (from 20 to 18%) and showed a significant difference in trends (P .05). Correct syndromic treatment (details information is available on request from the first author) increased in both the intervention (from 3 to 30%) and the control pharmacies (from 4 to 19%), however, to a greater extent in the intervention pharmacies with a significant difference in trends compared to the control pharmacies (P .05). Advice to use condom during intercourse and asking about sexual activities showed no significant differences in trends between the intervention and control pharmacies (Table 1). 4.3. Cefalexin request Dispensing of cefalexin decreased (from 95 to 56%) in the intervention pharmacies compared to control pharmacies (from 94 to 89%) with a significant difference in trends (P .01). However, dispensing of nonprescription antibiotics, such as ampicillin and penicillin, increased. In total, antibiotics were dispensed in 69 and 95% of the encounters in the intervention and control pharmacies, respectively. Prescription requests increased in the intervention pharmacies (from 0 to 21%) compared to control pharmacies (from 0 to 2%) with a significant difference in trends (P .01) (Table 1). 4.4. Prednisolone request Steroids dispensed decreased both in intervention (from 78 to 17%) and control pharmacies (from 73 to 58%) with a significant difference in trends (P .01) Prescription re-
1151
quests increased (from 1 to 18%) in the intervention compared to the control pharmacies (from 1 to 4%) with a significant difference in trends (P .01) (Table 1). Dispensing of no drugs and acceptable drugs, defined as nonsteroid anti-inflammatory drugs (NSAID), acetyl salicylic acid (ASA), paracetamol, and traditional medicines increased more in the intervention group (from 20 to 83%) compared to the control group (from 26 to 44%) with a significant difference in trends (P .01) (Table 1). 4.5. Drug costs The cost trends for all four conditions in both the control and intervention groups showed an increase. As the increase was similar in the intervention and control pharmacies there were no significant difference in trends. However, in the STD condition, the cost in the intervention pharmacies increased more than in the control pharmacies with a significant difference (P .01) in trends.
5. Discussion This study is, to our knowledge, the first reported multiintervention experiment in the private pharmacy sector in a low income country. Significant improvements in practice are shown subsequent to the interventions including regulatory enforcement, face-to-face education, and peer influence. Intentionally, the interventions were applied in a sequence; hence the study design does not permit separation of the effects of the different interventions. Thus, the effect of the whole intervention package was assessed. The basic contents of the messages were constant throughout the different interventions, but increased in quantity and complexity. In the regulatory enforcement intervention, the regulation regarding prescription-only drugs was disseminated and explained. This was further emphasized in the educational as well as in the peer influence intervention. In the educational intervention, pharmacy treatment guidelines, and a Question, Advice, and Treatment strategy were introduced in relation to the four conditions. The information was repeated during the peer influence intervention. Major findings in this study are significant improvements of quality of pharmacy practice, including ability to ask questions, give advice, and appropriate treatment of the four tracer conditions. For uncomplicated childhood ARI, questions regarding breathing increased and dispensing of antibiotics decreased. By posing this question, pharmacy staff can determine whether the patient needs a medical examination and antibiotic treatment or symptom-relieving drugs. Considering the common utilization of private pharmacies, the early detection of pneumonia in children and prompt correct treatment may reduce mortality and morbidity. Moreover, considering the documented frequent use of antibiotics in too short courses for ARI, improved rational use of antibiotics is of major importance for the containment of antibiotic resistance in respiratory pathogens [23].
1152
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155
Fig. 3. Mean (%) and standard deviation (SD) of encounters for ARI, STD, Cefalexin and Prednisolone tracer variables, Summary statistics and test of difference between the intervention (I) and control (C) groups across the study time and Trend-line graphs with the means (%) plotted out.
In response to a case history of urethral discharge, indicating STDs as chlamydia and gonorrhoea, advice to see a physician and dispensing of the correct syndromic treatment increased. STD treatment is relatively expensive; thus, there
are economic incentives for the pharmacies to follow the treatment guidelines. This improvement is similar to results from a study in Ghana, where correct syndromic drug provision for urethral discharge was improved by an educational
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155
intervention [24]. In Zambia, public health center personnel improved STD examination, treatment, and patient information after interactive health training [25]. Patients with symptoms of STD are commonly hesitant to consult a physician, as the disease is stigmatizing and the patients might feel embarrassed. Visiting a pharmacy is more anonymous, often quicker, and less expensive [26]. Untreated or incorrectly treated STDs increase the risk for transmission during sexual intercourse and potentially also transmission of HIV [27]. It is of mutual interest for the public health system and private pharmacies to dispense correct drugs to clients who are not willing to see a physician [17]. The educational intervention made clear to the pharmacy staff that the appropriate drugs for this condition were prescription-only drugs, but they were not actively discouraged from dispensing these without a prescription. This strategy was agreed upon with the Hanoi Health Bureau in the interests of public health. In the educational and peer influence interventions, the pharmacy staff was taught to ask about sexual activity and symptoms of chlamydia and gonorrhoea. Although emphasized during the educational intervention and peer influence, very few drug sellers advised the clients to use a condom, after the interventions. Similar results were also shown in the Ghana study where the intervention led to improvements in treatment of urethral discharge but not in advice to use a condom [24]. This may be an important focus for future preventive interventions in relation to health information on these diseases to reduce the spread of STDs including HIV, and to promote a more open attitude. There was an increase of correct syndromic treatment also in the control pharmacies. As the Hanoi Health Bureau issued no official information to pharmacies regarding STDs during the intervention period it might be assumed that information could have spread from intervention to control pharmacies. Prescription requests increased and dispensing of prescription-only drugs, cefalexin and prednisolone, decreased. The reasons for the major improvements might be both that the messages were repeated in the sequential interventions and the simplicity of the messages. Other studies have shown the effectiveness of simple, frequently repeated messages [14]. To purchase small amounts of cefalexin is relatively cheap and prednisolone is even cheaper. Thus, over-the-counter drugs can substitute for those drugs without any loss of profit. As shown in the results, prednisolone was substituted by nonsteroidal anti-inflammatory drugs and traditional medicines. Sales of these increased threefold in the intervention pharmacies. Cefalexin was substituted with over-the-counter drugs, antibiotics, traditional medicines, or paracetamol. Similarly, it has been shown in one study that when the dispensing of antibiotics decreased for treatment of diarrhoea, the dispensing of antidiarrhoeals increased [7]. Our results indicate improved compliance with the regulations and may contribute to reduce irrational steroid and antibiotic use, and thus decrease the risks of side effects and antibiotic resistance. Generally, the costs of treating the four conditions increased for both intervention and control pharmacies. As
1153
this increase matched the inflation during the 2-year period the study took place the latter is the likely reason. For the ARI and drug request conditions the type of drugs dispensed changed although there were no significant changes in cost. For STD, the cost for correct treatment was expensive, equal to one-third of a civil servant’s minimum monthly salary. The cost trend of STD case management at intervention pharmacies increased significantly, probably due to the increase in correct syndromic treatment at these pharmacies. Thus, pharmacy staff both learned from the intervention and changed their practice and maintained or increased their profit [5]. This intervention study was a randomized controlled trial. As the pharmacies were selected in the same areas throughout Hanoi it may be assumed that confounding factors in the environment were similar for the intervention and control pharmacies. The pharmacies were paired to ensure a similar composition for the intervention and control groups. It is thus assumed that the differences found between the intervention and control groups were due to the interventions. The pharmacies were very heterogeneous, as shown by the high standard deviation in most variables. The Simulated Client Method (SCM) has frequently been used to assess practice in both low and high income countries [20]. Here, the simulated clients present complaints to pharmacy staff and purchase the drugs recommended. The pharmacy staff does not know that they are being assessed; thus, it is assumed that the clients witness common practice. To avoid attention and recognition from pharmacy staff, five clients per condition visited all the pharmacies, once per each SCM monitoring period, which is to say about once every 7 months. The clients worked simultaneously according to a prearranged rotation schedule. To prevent the client’s individual behavior from affecting the pharmacy staff, all clients were trained to act in a reproducible way, and they were not informed about which pharmacies belonged to the intervention or control groups. After the third SCM round, it was found that some did not perform their tasks accurately, for example, one client asked another to take over his work. To avoid stigmatizing those clients, all clients were replaced in the fourth and last SCM round by new clients who received the same training. As the simulated clients did not know which pharmacies belonged to the intervention or control group, any inaccurate performance is likely to have effected both groups similarly, decrease the group difference and thus the measured effect of interventions. In the last round of SCM, questions regarding sexual activity increased in both the intervention and control pharmacies compared to the two previous SCM rounds. This effect might partly be due to the fact that the clients in the last round of SCM were younger than those used in the preceding SCMs. In Vietnam, questions regarding sexual activity and advice on condom use are not considered polite, particularly if the clients are older than the pharmacy staff. The ethics of performing an act using benevolent deception [28] to gather information from pharmacy staff could
1154
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155
be raised. In support of the Simulated Client Method, it should be pointed out that the potential harm of low quality of pharmacy practice including frequent irrational provision of drugs by the fairly unregulated private pharmacies was considered to be an increasing problem by the national health authorities. Hence, the experiment was supported, planned, and conducted together with the Hanoi Pharmacy Association and the Hanoi Health Bureau. Moreover, one should consider the major difficulties involved in obtaining the necessary information on actual practice with other methods in these contexts and the obvious advantages of evidence obtained from real practice.
[2]
[3] [4] [5]
[6] [7]
6. Conclusion This study shows that it is possible to improve practice in private pharmacies with a combination of improved enforcement of regulation, education, and personal involvement through peer networks. It is important that the interventions are designed so that profit can be maintained or increased. Issues that are considered sensitive, such as condom use, are however, more difficult to handle. Our study shows that if pharmacy staff gets appropriate support to fulfill their public health role more rational provision of drugs will follow. The impact of this for the containment of antibiotic resistance, prevention of the spread of STDs including HIV and avoiding waste of scarce public resources cannot be underestimated considering the high utilization of the private pharmacy services.
[8]
[9] [10]
[11]
[12]
[13] [14]
Acknowledgments We would like to thank the studied pharmacy staff for their participation, the Health Strategy and Policy Institute, the College of Hanoi Pharmacy School, the Hanoi Medical University, the Hanoi Health Bureau and, the Hanoi Pharmacy Association. The project was financially supported by the European Union DG XII, INCO-DC, ERB3514PL950674. The WHO Essential Drugs and other Medicines program provided funds for drug purchases. Max Petzold has been most valuable in support with parts of the data analysis. Pharm. Binh NT and Dr. Hiep HT for their major role in the organizing and monitoring of the SCM process. We also thank our colleagues in the project “Good Pharmacy Practice in Vietnam and Thailand,” including the national coordinators at the Health System Research Institute in Thailand, Drs. Yupadee Javrongrit and Sauwakon Ratanawijitrasin, and at the London School of Hygiene and Tropical Medicine, Prof. Gill Walt and Dr. John Chalker, for being partners in design and method development, as well as all other colleagues who made the project implementation possible.
[15]
[16]
[17]
[18]
[19]
[20]
[21] [22] [23]
References [1] Quick J, Laing R, Ross-Degnan D. Intervention research to promote clinically effective and economically efficient use of pharmaceuti-
[24]
cals: the International Network for Rational Use of Drugs. J Clin Epidemiol 1991;44(Suppl 2):57S–65S. Goel P, Ross-Degnan D, Berman P, Soumerai S. Retail pharmacists in developing countries, a behaviour and intervention framework. Soc Sci Med 1996;42:1155–61. Tomson G, Sterky G. Self-prescribing by way of pharmacies in three Asian developing countries. Lancet 1986;13:620–2. Kamat VR, Nichter M. Pharmacies, self-medication and pharmaceutical marketing in Bombay, India. Soc Sci Med 1998;47:779–94. Cederlof C, Tomson G. Private pharmacies and the health sector reform in developing countries—professional and commercial highlights. J Soc Adm Pharm 1995;3:101–11. Thamlikiktul V. Antibiotic dispensing by drug store personnel in Bangkok. Thailand. J Antimicrob Chemother 1988;21:125–31. Ross-Degnan D, Soumerai S, Goel P, Bates J, Makhulo J, Dondi M, Sutoto Daryono A, Ferraz-Tubor L, Hogan R. The impact of face-toface educational outreach on diarrhoeal treatment in pharmacies. Health Policy Plan 1996;11:308–18. Chuc NT, Tomson G. “Doi moi” and private pharmacies: a case study on dispensing and financing issues in Hanoi, Vietnam. Eur J Clin Pharmacol 1999;55:325–32. Ministry of Health. Health statistic year book. Hanoi, Vietnam: MoH; 1996. p. 49. Falkenberg T, Nguyen TB, Larsson M, Nguyen TD, Tomson G. Pharmaceutical sector in transition—a cross sectional study in Vietnam. Southeast Asian J Trop Med Public Health 2000;31:590–7. Duong Van D, Binns CW, Van Le T. Availability of antibiotics as over-the-counter drugs in pharmacies: a threat to public health in Vietnam.Trop Med Int Health 1997;2:1133–9. Chuc NTK, Larsson M, Falkenberg T, Do NT, Binh NT, Tomson G. Management of childhood acute respiratory infections at private pharmacies in Vietnam. Ann Pharmacother 2001;35:1283–8. World Health Assembly Resolution WHA 47.12, 1994 (FIP). Standards for quality of pharmacy services. Tokyo: The Tokyo Declaration; 1993. Brugha R, Zwi A. Improving the quality of private sector delivery of public health services: challenges and strategies. Health Policy Plan 1998;13:107–20. Kumaranayake L, Mujinja P, Hongoro C, Mpembeni R. How do countries regulate the health sector? Evidence from Tanzania and Zimbabwe. Health Policy Plan 2000;15:357–67. Stenson B, Syhakhang L, Lundborg CS, Eriksson B, Tomson G. Private pharmacy practice and regulation—a randomized trial in Lao P.D.R. Int J Technol Assess Health Care 2001;17:579–89. Chalker J, Chuc NT, Falkenberg T, Do NT, Tomson G. STD management by private pharmacies in Hanoi: practice and knowledge of drug sellers. Sex Transm Infect 2000;76:299–302. Bexell A, Lwando E, von Hofsten B, Tembo S, Eriksson B, Diwan VK. Improving drug use through continuing education: a randomized controlled trial in Zambia. J Clin Epidemiol 1996;49:355–7. Lundborg CS, Wahlstrom R, Oke T, Tomson G, Diwan KV. Influencing prescribing for urinary tract infection and asthma in primary care in Sweden: a randomized controlled trial of an interactive educational intervention. J Clin Epidemiol 1999;52:801–12. Madden JM, Quick JD, Ross-Dengan D, Kafle KK. Undercover careseekers: simulated clients in the study of health care provider behavior in developing countries. Soc Sci Med 1997;45:1465–82. Ministry of Health. Prescription and dispensing drug regulation No 488. Hanoi, Vietnam: Ministry of Health; 1995. Diggle P. Analysis of longitudinal data. Oxford: Oxford University Press; 1995. Larsson M, Kronvall G, Chuc NT, Karlsson I, Lager F, Hanh HD, Tomson G, Falkenberg T. Antibiotic medication and bacterial resistance to antibiotic: a survey of children in a Vietnamese community. Trop Med Int Health 2000;5:711–21. Adu-Sarkodie Y, Stener MJ, Attafuah J, Tweedy K. Syndromic management of urethral discharge in Ghanaian pharmacies. Sex Transm Infect 2000;76:439–42.
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155 [25] Faxelid E, Ahlberg BM, Freudenthal S, Ndulo J, Krantz I. Quality of STD care in Zambia. Impact of training in STD management. Int J Qual Health Care 1997;9:361–6. [26] Benjarattanaporn P, Lindan CP, Mills S, Barclay J, Bennett A, Mugrditchian D, Mandel JS, Pongswatanakulsiri P, Warnnissorn T. Men with sexually transmitted diseases in Bangkok: where do they go for treatment and why? AIDS 1997;11(Suppl 1):S87–95.
1155
[27] Gilson L, Mkanje R, Grosskurth H, Mosha F, Picard J, Gavyole A, Todd J, Mayaud P, Swai R, Fransen L, Mabey D, Mills A, Hayes R. Cost-effectiveness of improved treatment services for sexually transmitted diseases in preventing HIV-1 infection in Mwanza Region, Tanzania. Lancet 1997;350:1805–9. [28] Bok S. The ethics of giving placebo. Sci Am 1974;231:17–23.
Improving private pharmacy practice: A multi-intervention experiment in Hanoi, Vietnam Nguyen T.K. Chuca,b,*, Mattias Larssonb, Nguyen T. Doc, Vinod K. Diwanb,d, Goran B. Tomsonb,e, Torkel Falkenbergb a Hanoi Medical University, Hanoi, Vietnam Division of International Health (IHCAR), Department of Public Health Sciences, Karolinska Institutet, SE171 76 Stockholm, Sweden c Hanoi College of Pharmacy, Hanoi, Vietnam d Nordic School of Public Health, Gothenburg, Sweden e Medical Management Centre, Karolinska Institutet, Sweden
b
Received 3 October 2001; received in revised form 19 May 2002; accepted 21 May 2002
Abstract The objective of this study was to assess the effects of a multicomponent intervention on private pharmacy practice. From 641 private pharmacies in Hanoi, 68 pharmacies were randomly selected and matched into 34 pairs. Each pair consisted of a control and an intervention pharmacy. Three interventions were applied sequentially: Regulatory enforcement, Education, and Peer influence. Four tracer conditions were selected: uncomplicated acute respiratory infection (ARI), sexually transmitted disease (STD), requesting the prescription-only drugs prednisolone, and a short course of cefalexin. Practice was assessed through the Simulated Client Method (SCM). The intervention pharmacies improved significantly compared to the control pharmacies (P .05) in all tracer conditions. For ARI, antibiotic dispensing decreased (P .02) and questions regarding breathing increased (P .01). For STD, advice to go to the doctor and dispensing the correct syndromic treatment increased (P .01). Dispensing of prednisolone and cefalexin decreased (P .01) and prescription requests increased (P .01). Our conclusion is that it is possible to improve private pharmacy practice with a multicomponent intervention. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Private pharmacy practice; Randomized controlled trial; Multi-intervention; Simulated Client Method (SCM); Acute Respiratory Infection (ARI); Sexually Transmitted Disease (STD)
1. Introduction In many low income countries, pharmaceuticals make up 50% or more of health care costs [1]. With health sector reforms, private pharmacies are becoming increasingly the first and only contact with health delivery services [2–4]. “Prescribing” by pharmacy staff sometimes can result in irrational use of drugs, waste of scarce resources, and the sale of harmful drugs, possibly due to the “business-profession” dilemma in private pharmacy practice [5–7]. In 1986, an economic reform, known as Doi Moi, was launched in Vietnam introducing market economy with privatization in all sectors of society including drug provision. Pharmacists with 5 years or more of working experience were allowed to open private pharmacies [8], and the number of private pharmacies increased from none in 1986
* Corresponding author. Fax: 0046-83-11-590. E-mail address: [email protected] (N.T.K. Chuc).
to more than 6,000 in 1996 [9]. This, combined with the increased domestic production of drugs and drug imports, considerably improved the availability of drugs. However, the regulations governing the pharmaceutical sector and information to providers, prescribers, and users could not keep up with the rapid increased supply of drugs [10]. Currently, most patients buy drugs including antibiotics from private pharmacies without prescription or adequate user instructions [8,11,12]. Standards to improve the quality of pharmacy services have been set out in the Good Pharmacy Practice (GPP) document. In line with the greater demand for pharmacy services, the pharmacists’ new role in the health system includes being a communicator, a quality drug supplier, a trainer, and supervisor and a health promoter [13]. Professional practice in private pharmacies is influenced by many factors including staff education, customer demand, physician practice, regulations, and economic incentives [2,4,5]. Regulation is one important factor influencing pharmacy practice [2,14]. Even with the existing regulations, effective
0895-4356/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S0895-4356(02)00 4 5 8 - 4
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155
1149
Fig. 1. Sequence and time intervals between the interventions.
enforcement mechanisms are often absent in low and middle income countries [15,16]. There are few studies published showing the effects of regulatory enforcement at pharmacies in low income countries [16]. Good knowledge is a prerequisite for good practice. However, knowledge does not always guarantee good practice [12,17]. Educational interventions that are effective have been shown in some studies [7,18,19]. With growing competition, economic incentives are important, and customer demand has an increasing influence on practice [4,8]. The social and professional culture within a profession may have a major impact on the practice [5]. Peer influence, building on the concept of group responsibility for self-education and monitoring, as well as multicomponent interventions, have shown to be effective in improving quality of service in the public sector of high income countries [14,19]. However, there is no research published combining different types of interventions in the private sector of low income countries. The aim of this study was to evaluate the effectiveness of a multi-intervention package on pharmacy practice using tracer conditions: uncomplicated acute respiratory infection (ARI), sexually transmitted disease (STD), requesting the prescription-only drugs prednisolone, and a short course of cefalexin. This study is part of the Towards Good Pharmacy Practice in Thailand and Vietnam project. The members of the team jointly developed the design and the instruments used in this study.
whether or not the pharmacist was the license holder; 3. whether or not the pharmacy was close to a hospital; and 4. located in a different ward or more than 150 meters from all other pharmacies selected. The sample size was chosen to detect a difference in degree of 20% between the intervention and control group using the t-test with a significance level of 5% and a power of 90%. This sample size was assumed to be applicable also for the Wilcoxon signed rank test used in the final analysis. The pharmacies in each pair were randomly assigned as intervention or control. 2.2. Tracer conditions Four tracer conditions of public health importance were selected: (1) case management of an uncomplicated upper respiratory tract infection (ARI) in a child under 5 years of age with a mild cough. (2) Case management of a sexually transmitted disease (STD), presented as an urethral discharge in an adult man. (3) Request to purchase a couple of capsules of an antibiotic (cefalexin) without a prescription.
2. Method 2.1. Study design, sampling, and sample size A clustered randomized controlled trial with a time series design and a multicomponent intervention was carried out (Fig. 1). There were 789 private pharmacies registered in the urban area of Hanoi; 148 located inside a hospital or mainly wholesalers were excluded. From the remaining 641 pharmacies 34 pairs were randomly selected (Fig. 2) according to the following matching criteria: 1. turnover: high, medium, or low according to district inspectors; 2.
Fig. 2. Sample size and sampling.
1150
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155
(4) Request to purchase a couple of tables of a steroid (prednisolone) without a prescription.
quests for a short course of antibiotics; in the second, STD case management and requests for oral steroids.
2.3. Simulated Client Method (SCM)
3.3. Peer influence (PI)
The Simulated Client Method (SCM) [20] was used to assess private pharmacy practice. Four rounds of SCM were conducted, a baseline and after (1 month) each of the three interventions. Twenty simulated clients were selected, five per tracer condition, and trained to act in a reproducible way according to the tracer condition scenario (detailed information is available on request from the first author). In the first three SCM rounds the simulated clients were research assistants from the Hanoi College of Pharmacy and in the last SCM round from the Hanoi Medical University. For every SCM round, each simulated client encountered all the selected pharmacies, presented the scenario, bought the drugs recommended by the pharmacy staff, and filled in the record sheets at a maximum of 15 min after leaving the pharmacy. The drugs purchased and the record sheets were submitted to the research team at the College of Pharmacy. The simulated clients knew that they were measuring the practice of drug sellers but did not know which pharmacies belonged to the intervention and control group.
The intervention pharmacies were geographically divided into five groups. The research group appointed one person in each group to be the leader based on the experience gained during the educational intervention. The five group leaders and representatives of the Hanoi Pharmacy Association attended a 1-day seminar where the research group informed about the peer influence strategy and reviewed case management of the four tracer conditions. Each group then organized a meeting where the group leader informed about the peer influence strategy and instructed the participants to make case management records of pharmacy customers with complaints related to the tracer conditions. Three meetings were held during 3 months where the participants presented and discussed their records. A summary meeting with all participating intervention pharmacy staff was held to discuss the intervention strategy. The intervention pharmacies did know that the interventions were taking place, but did not know about the study design or the SCM. Ethical approval was obtained from the Ministry of Health, Vietnam, and from the Karolinska Institutet. The study was done in collaboration with the Hanoi Pharmacy Association and the Hanoi Health Bureau.
3. Interventions 3.1. Regulatory enforcement (RE) The first intervention, done in collaboration with the Hanoi Health Bureau, focused on the regulation of prescriptiononly drugs, particularly the drugs related to tracer conditions in the study (antibiotics, with emphasis on cefalexin and perfloxacin, and corticosteroids including prednisolone). Four inspectors from the Hanoi Health Bureau were trained to cover these areas of inspection. In pairs they visited the intervention pharmacies twice with an interval of 1 month. During the first visit, in addition to normal inspection procedures, they disseminated and explained the regulations regarding prescription-only drugs [21]. Cefalexin and prednisolone were used as examples of prescription only drugs that should be dispensed only if a prescription was provided (detailed information is available on request from the first author). In the second visit, these messages were repeated. 3.2. Education (ED) Pharmacy treatment guidelines for the four tracer conditions were developed based on the national treatment guidelines by the research team in collaboration with Hanoi Health Bureau, the Pharmacy Association, experienced clinicians, and pharmacologist (detailed information is available on request from the first author). Two senior research team members and two clinicians conducted the intervention. Each intervention pharmacy received two 45-min face-to-face educational sessions. Written and verbal information regarding the importance of GPP [13], and adequate questions, advice, and treatments in relation to the four tracer conditions was provided. In the first session, the topics were ARI case management and re-
3.4. Data analysis The study unit was the pharmacy. The results were coded into computers using Microsoft Access software. SPSS 10.0 and SAS 8.0 software were used for the data analysis. Analysis of effects of the whole intervention package was done using the method of summary statistics for the selected tracers and variables [22]. This method is based on finding one or a few statistics that summarize the information of interest in a longitudinal series. The individual effect is then separated from the cohort effect, and the analysis incorporates the within-pharmacy dependence. In the study, the main question was about the difference in trend between the intervention and control pharmacies over the study period. Here, the regression coefficient, that is, the slope, from a linear regression on each pharmacy was selected as a suitable summary statistic. The tracer variable measured with SCM, for example, antibiotics dispensed, was used as the dependent variable, and the four rounds of SCM were used as the independent variable. The slopes of the selected variables were compared in pairs using the Wilcoxon signed-rank test. A significant difference (P .05, two tailed) was interpreted as a significant effect of the intervention package.
4. Results During the study, four pharmacies were excluded because of irregular opening hours. As a consequence, the other phar-
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155
macy in each pair was also excluded as the results were analyzed in pairs. In total, eight pharmacies were excluded. Thirty-four intervention pharmacies received regulatory enforcement, 30 received educational face-to-face sessions, and 29 took part in the peer influence intervention; one refused for the reason that it was “too busy.” In the cross-sectional analysis 60 pharmacies with 4,445 encounters were used, in the longitudinal analysis 58 pharmacies with 4,356 encounters. The estimated total cost of the interventions was approximately 5,700 USD (less than 200 USD per pharmacy). 4.1. ARI case management After the interventions, dispensing of antibiotics decreased in the intervention pharmacies (from 45 to 30%) while it increased in the control pharmacies (from 39 to 42%) with a significant difference in trends (P .05). Questions regarding breathing increased (11 to 30%) in the intervention pharmacies and decreased in the control pharmacies (10 to 7%) with a significant difference in trends (P .01) (Table 1). Dispensing of acceptable drugs, defined as paracetamol, cough syrups, and traditional medicines not containing antihistamines and opiates, showed no significant difference in trends between the two groups. 4.2. STD case management Advice regarding consultation with a physician increased in the intervention pharmacies (from 22 to 37%) compared to the control pharmacies (from 20 to 18%) and showed a significant difference in trends (P .05). Correct syndromic treatment (details information is available on request from the first author) increased in both the intervention (from 3 to 30%) and the control pharmacies (from 4 to 19%), however, to a greater extent in the intervention pharmacies with a significant difference in trends compared to the control pharmacies (P .05). Advice to use condom during intercourse and asking about sexual activities showed no significant differences in trends between the intervention and control pharmacies (Table 1). 4.3. Cefalexin request Dispensing of cefalexin decreased (from 95 to 56%) in the intervention pharmacies compared to control pharmacies (from 94 to 89%) with a significant difference in trends (P .01). However, dispensing of nonprescription antibiotics, such as ampicillin and penicillin, increased. In total, antibiotics were dispensed in 69 and 95% of the encounters in the intervention and control pharmacies, respectively. Prescription requests increased in the intervention pharmacies (from 0 to 21%) compared to control pharmacies (from 0 to 2%) with a significant difference in trends (P .01) (Table 1). 4.4. Prednisolone request Steroids dispensed decreased both in intervention (from 78 to 17%) and control pharmacies (from 73 to 58%) with a significant difference in trends (P .01) Prescription re-
1151
quests increased (from 1 to 18%) in the intervention compared to the control pharmacies (from 1 to 4%) with a significant difference in trends (P .01) (Table 1). Dispensing of no drugs and acceptable drugs, defined as nonsteroid anti-inflammatory drugs (NSAID), acetyl salicylic acid (ASA), paracetamol, and traditional medicines increased more in the intervention group (from 20 to 83%) compared to the control group (from 26 to 44%) with a significant difference in trends (P .01) (Table 1). 4.5. Drug costs The cost trends for all four conditions in both the control and intervention groups showed an increase. As the increase was similar in the intervention and control pharmacies there were no significant difference in trends. However, in the STD condition, the cost in the intervention pharmacies increased more than in the control pharmacies with a significant difference (P .01) in trends.
5. Discussion This study is, to our knowledge, the first reported multiintervention experiment in the private pharmacy sector in a low income country. Significant improvements in practice are shown subsequent to the interventions including regulatory enforcement, face-to-face education, and peer influence. Intentionally, the interventions were applied in a sequence; hence the study design does not permit separation of the effects of the different interventions. Thus, the effect of the whole intervention package was assessed. The basic contents of the messages were constant throughout the different interventions, but increased in quantity and complexity. In the regulatory enforcement intervention, the regulation regarding prescription-only drugs was disseminated and explained. This was further emphasized in the educational as well as in the peer influence intervention. In the educational intervention, pharmacy treatment guidelines, and a Question, Advice, and Treatment strategy were introduced in relation to the four conditions. The information was repeated during the peer influence intervention. Major findings in this study are significant improvements of quality of pharmacy practice, including ability to ask questions, give advice, and appropriate treatment of the four tracer conditions. For uncomplicated childhood ARI, questions regarding breathing increased and dispensing of antibiotics decreased. By posing this question, pharmacy staff can determine whether the patient needs a medical examination and antibiotic treatment or symptom-relieving drugs. Considering the common utilization of private pharmacies, the early detection of pneumonia in children and prompt correct treatment may reduce mortality and morbidity. Moreover, considering the documented frequent use of antibiotics in too short courses for ARI, improved rational use of antibiotics is of major importance for the containment of antibiotic resistance in respiratory pathogens [23].
1152
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155
Fig. 3. Mean (%) and standard deviation (SD) of encounters for ARI, STD, Cefalexin and Prednisolone tracer variables, Summary statistics and test of difference between the intervention (I) and control (C) groups across the study time and Trend-line graphs with the means (%) plotted out.
In response to a case history of urethral discharge, indicating STDs as chlamydia and gonorrhoea, advice to see a physician and dispensing of the correct syndromic treatment increased. STD treatment is relatively expensive; thus, there
are economic incentives for the pharmacies to follow the treatment guidelines. This improvement is similar to results from a study in Ghana, where correct syndromic drug provision for urethral discharge was improved by an educational
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155
intervention [24]. In Zambia, public health center personnel improved STD examination, treatment, and patient information after interactive health training [25]. Patients with symptoms of STD are commonly hesitant to consult a physician, as the disease is stigmatizing and the patients might feel embarrassed. Visiting a pharmacy is more anonymous, often quicker, and less expensive [26]. Untreated or incorrectly treated STDs increase the risk for transmission during sexual intercourse and potentially also transmission of HIV [27]. It is of mutual interest for the public health system and private pharmacies to dispense correct drugs to clients who are not willing to see a physician [17]. The educational intervention made clear to the pharmacy staff that the appropriate drugs for this condition were prescription-only drugs, but they were not actively discouraged from dispensing these without a prescription. This strategy was agreed upon with the Hanoi Health Bureau in the interests of public health. In the educational and peer influence interventions, the pharmacy staff was taught to ask about sexual activity and symptoms of chlamydia and gonorrhoea. Although emphasized during the educational intervention and peer influence, very few drug sellers advised the clients to use a condom, after the interventions. Similar results were also shown in the Ghana study where the intervention led to improvements in treatment of urethral discharge but not in advice to use a condom [24]. This may be an important focus for future preventive interventions in relation to health information on these diseases to reduce the spread of STDs including HIV, and to promote a more open attitude. There was an increase of correct syndromic treatment also in the control pharmacies. As the Hanoi Health Bureau issued no official information to pharmacies regarding STDs during the intervention period it might be assumed that information could have spread from intervention to control pharmacies. Prescription requests increased and dispensing of prescription-only drugs, cefalexin and prednisolone, decreased. The reasons for the major improvements might be both that the messages were repeated in the sequential interventions and the simplicity of the messages. Other studies have shown the effectiveness of simple, frequently repeated messages [14]. To purchase small amounts of cefalexin is relatively cheap and prednisolone is even cheaper. Thus, over-the-counter drugs can substitute for those drugs without any loss of profit. As shown in the results, prednisolone was substituted by nonsteroidal anti-inflammatory drugs and traditional medicines. Sales of these increased threefold in the intervention pharmacies. Cefalexin was substituted with over-the-counter drugs, antibiotics, traditional medicines, or paracetamol. Similarly, it has been shown in one study that when the dispensing of antibiotics decreased for treatment of diarrhoea, the dispensing of antidiarrhoeals increased [7]. Our results indicate improved compliance with the regulations and may contribute to reduce irrational steroid and antibiotic use, and thus decrease the risks of side effects and antibiotic resistance. Generally, the costs of treating the four conditions increased for both intervention and control pharmacies. As
1153
this increase matched the inflation during the 2-year period the study took place the latter is the likely reason. For the ARI and drug request conditions the type of drugs dispensed changed although there were no significant changes in cost. For STD, the cost for correct treatment was expensive, equal to one-third of a civil servant’s minimum monthly salary. The cost trend of STD case management at intervention pharmacies increased significantly, probably due to the increase in correct syndromic treatment at these pharmacies. Thus, pharmacy staff both learned from the intervention and changed their practice and maintained or increased their profit [5]. This intervention study was a randomized controlled trial. As the pharmacies were selected in the same areas throughout Hanoi it may be assumed that confounding factors in the environment were similar for the intervention and control pharmacies. The pharmacies were paired to ensure a similar composition for the intervention and control groups. It is thus assumed that the differences found between the intervention and control groups were due to the interventions. The pharmacies were very heterogeneous, as shown by the high standard deviation in most variables. The Simulated Client Method (SCM) has frequently been used to assess practice in both low and high income countries [20]. Here, the simulated clients present complaints to pharmacy staff and purchase the drugs recommended. The pharmacy staff does not know that they are being assessed; thus, it is assumed that the clients witness common practice. To avoid attention and recognition from pharmacy staff, five clients per condition visited all the pharmacies, once per each SCM monitoring period, which is to say about once every 7 months. The clients worked simultaneously according to a prearranged rotation schedule. To prevent the client’s individual behavior from affecting the pharmacy staff, all clients were trained to act in a reproducible way, and they were not informed about which pharmacies belonged to the intervention or control groups. After the third SCM round, it was found that some did not perform their tasks accurately, for example, one client asked another to take over his work. To avoid stigmatizing those clients, all clients were replaced in the fourth and last SCM round by new clients who received the same training. As the simulated clients did not know which pharmacies belonged to the intervention or control group, any inaccurate performance is likely to have effected both groups similarly, decrease the group difference and thus the measured effect of interventions. In the last round of SCM, questions regarding sexual activity increased in both the intervention and control pharmacies compared to the two previous SCM rounds. This effect might partly be due to the fact that the clients in the last round of SCM were younger than those used in the preceding SCMs. In Vietnam, questions regarding sexual activity and advice on condom use are not considered polite, particularly if the clients are older than the pharmacy staff. The ethics of performing an act using benevolent deception [28] to gather information from pharmacy staff could
1154
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155
be raised. In support of the Simulated Client Method, it should be pointed out that the potential harm of low quality of pharmacy practice including frequent irrational provision of drugs by the fairly unregulated private pharmacies was considered to be an increasing problem by the national health authorities. Hence, the experiment was supported, planned, and conducted together with the Hanoi Pharmacy Association and the Hanoi Health Bureau. Moreover, one should consider the major difficulties involved in obtaining the necessary information on actual practice with other methods in these contexts and the obvious advantages of evidence obtained from real practice.
[2]
[3] [4] [5]
[6] [7]
6. Conclusion This study shows that it is possible to improve practice in private pharmacies with a combination of improved enforcement of regulation, education, and personal involvement through peer networks. It is important that the interventions are designed so that profit can be maintained or increased. Issues that are considered sensitive, such as condom use, are however, more difficult to handle. Our study shows that if pharmacy staff gets appropriate support to fulfill their public health role more rational provision of drugs will follow. The impact of this for the containment of antibiotic resistance, prevention of the spread of STDs including HIV and avoiding waste of scarce public resources cannot be underestimated considering the high utilization of the private pharmacy services.
[8]
[9] [10]
[11]
[12]
[13] [14]
Acknowledgments We would like to thank the studied pharmacy staff for their participation, the Health Strategy and Policy Institute, the College of Hanoi Pharmacy School, the Hanoi Medical University, the Hanoi Health Bureau and, the Hanoi Pharmacy Association. The project was financially supported by the European Union DG XII, INCO-DC, ERB3514PL950674. The WHO Essential Drugs and other Medicines program provided funds for drug purchases. Max Petzold has been most valuable in support with parts of the data analysis. Pharm. Binh NT and Dr. Hiep HT for their major role in the organizing and monitoring of the SCM process. We also thank our colleagues in the project “Good Pharmacy Practice in Vietnam and Thailand,” including the national coordinators at the Health System Research Institute in Thailand, Drs. Yupadee Javrongrit and Sauwakon Ratanawijitrasin, and at the London School of Hygiene and Tropical Medicine, Prof. Gill Walt and Dr. John Chalker, for being partners in design and method development, as well as all other colleagues who made the project implementation possible.
[15]
[16]
[17]
[18]
[19]
[20]
[21] [22] [23]
References [1] Quick J, Laing R, Ross-Degnan D. Intervention research to promote clinically effective and economically efficient use of pharmaceuti-
[24]
cals: the International Network for Rational Use of Drugs. J Clin Epidemiol 1991;44(Suppl 2):57S–65S. Goel P, Ross-Degnan D, Berman P, Soumerai S. Retail pharmacists in developing countries, a behaviour and intervention framework. Soc Sci Med 1996;42:1155–61. Tomson G, Sterky G. Self-prescribing by way of pharmacies in three Asian developing countries. Lancet 1986;13:620–2. Kamat VR, Nichter M. Pharmacies, self-medication and pharmaceutical marketing in Bombay, India. Soc Sci Med 1998;47:779–94. Cederlof C, Tomson G. Private pharmacies and the health sector reform in developing countries—professional and commercial highlights. J Soc Adm Pharm 1995;3:101–11. Thamlikiktul V. Antibiotic dispensing by drug store personnel in Bangkok. Thailand. J Antimicrob Chemother 1988;21:125–31. Ross-Degnan D, Soumerai S, Goel P, Bates J, Makhulo J, Dondi M, Sutoto Daryono A, Ferraz-Tubor L, Hogan R. The impact of face-toface educational outreach on diarrhoeal treatment in pharmacies. Health Policy Plan 1996;11:308–18. Chuc NT, Tomson G. “Doi moi” and private pharmacies: a case study on dispensing and financing issues in Hanoi, Vietnam. Eur J Clin Pharmacol 1999;55:325–32. Ministry of Health. Health statistic year book. Hanoi, Vietnam: MoH; 1996. p. 49. Falkenberg T, Nguyen TB, Larsson M, Nguyen TD, Tomson G. Pharmaceutical sector in transition—a cross sectional study in Vietnam. Southeast Asian J Trop Med Public Health 2000;31:590–7. Duong Van D, Binns CW, Van Le T. Availability of antibiotics as over-the-counter drugs in pharmacies: a threat to public health in Vietnam.Trop Med Int Health 1997;2:1133–9. Chuc NTK, Larsson M, Falkenberg T, Do NT, Binh NT, Tomson G. Management of childhood acute respiratory infections at private pharmacies in Vietnam. Ann Pharmacother 2001;35:1283–8. World Health Assembly Resolution WHA 47.12, 1994 (FIP). Standards for quality of pharmacy services. Tokyo: The Tokyo Declaration; 1993. Brugha R, Zwi A. Improving the quality of private sector delivery of public health services: challenges and strategies. Health Policy Plan 1998;13:107–20. Kumaranayake L, Mujinja P, Hongoro C, Mpembeni R. How do countries regulate the health sector? Evidence from Tanzania and Zimbabwe. Health Policy Plan 2000;15:357–67. Stenson B, Syhakhang L, Lundborg CS, Eriksson B, Tomson G. Private pharmacy practice and regulation—a randomized trial in Lao P.D.R. Int J Technol Assess Health Care 2001;17:579–89. Chalker J, Chuc NT, Falkenberg T, Do NT, Tomson G. STD management by private pharmacies in Hanoi: practice and knowledge of drug sellers. Sex Transm Infect 2000;76:299–302. Bexell A, Lwando E, von Hofsten B, Tembo S, Eriksson B, Diwan VK. Improving drug use through continuing education: a randomized controlled trial in Zambia. J Clin Epidemiol 1996;49:355–7. Lundborg CS, Wahlstrom R, Oke T, Tomson G, Diwan KV. Influencing prescribing for urinary tract infection and asthma in primary care in Sweden: a randomized controlled trial of an interactive educational intervention. J Clin Epidemiol 1999;52:801–12. Madden JM, Quick JD, Ross-Dengan D, Kafle KK. Undercover careseekers: simulated clients in the study of health care provider behavior in developing countries. Soc Sci Med 1997;45:1465–82. Ministry of Health. Prescription and dispensing drug regulation No 488. Hanoi, Vietnam: Ministry of Health; 1995. Diggle P. Analysis of longitudinal data. Oxford: Oxford University Press; 1995. Larsson M, Kronvall G, Chuc NT, Karlsson I, Lager F, Hanh HD, Tomson G, Falkenberg T. Antibiotic medication and bacterial resistance to antibiotic: a survey of children in a Vietnamese community. Trop Med Int Health 2000;5:711–21. Adu-Sarkodie Y, Stener MJ, Attafuah J, Tweedy K. Syndromic management of urethral discharge in Ghanaian pharmacies. Sex Transm Infect 2000;76:439–42.
N.T.K. Chuc et al. / Journal of Clinical Epidemiology 55 (2002) 1148–1155 [25] Faxelid E, Ahlberg BM, Freudenthal S, Ndulo J, Krantz I. Quality of STD care in Zambia. Impact of training in STD management. Int J Qual Health Care 1997;9:361–6. [26] Benjarattanaporn P, Lindan CP, Mills S, Barclay J, Bennett A, Mugrditchian D, Mandel JS, Pongswatanakulsiri P, Warnnissorn T. Men with sexually transmitted diseases in Bangkok: where do they go for treatment and why? AIDS 1997;11(Suppl 1):S87–95.
1155
[27] Gilson L, Mkanje R, Grosskurth H, Mosha F, Picard J, Gavyole A, Todd J, Mayaud P, Swai R, Fransen L, Mabey D, Mills A, Hayes R. Cost-effectiveness of improved treatment services for sexually transmitted diseases in preventing HIV-1 infection in Mwanza Region, Tanzania. Lancet 1997;350:1805–9. [28] Bok S. The ethics of giving placebo. Sci Am 1974;231:17–23.