Strengthening pharmacy practice in vietnam: findings of a training intervention study

Authors


Corresponding Author Pham Duc Minh, PATH, 2nd floor, Hanoi Towers, 49 Hai Ba Trung Street, Hoan Kiem District, Hanoi, Vietnam. E-mail: minhp1974@gmail.com

Abstract

Objective

To assess the effectiveness of a training and supportive supervision intervention in strengthening the capacity of pharmacy staff in Vietnam to deliver client-oriented, accurate healthcare information and appropriate services for childhood diarrhoea and emergency contraceptive pills (ECP).

Methods

Pre- and post-intervention study using a cross-sectional design. Pharmacy staff participated in 3 days of training on customer relations, good pharmacy practice, childhood diarrhoea and ECP over a period of 1 month, consisting of lectures, discussion, question-and-answer sessions and role-playing. We compared baseline and 6-month post-intervention surveys to ascertain changes in knowledge, attitudes and practice of pharmacists, using univariate statistics to find significant differences.

Results

More than 1200 pharmacists received training and supportive supervision. After interventions, pharmacy staff knowledge was significantly improved on most of the measured indicators. Knowledge of dehydration symptoms for diarrhoea increased from 19% to 88%, and for side effects of ECP increased from 27% to 77%. While assessment of actual practice revealed that this knowledge was not always used, significant improvement was observed. Before interventions, 12% gave information on dehydration symptoms but 45% did so afterwards. The proportion giving information on side effects of ECP increased from 13% to 54%.

Conclusions

Providing a programme of training and supportive supervision is an effective way to improve knowledge and practice of pharmacists at private pharmacies in Vietnam. These improvements have the potential to lead to better community health care.

Abstract

Objectif

Evaluer l'efficacité d'une intervention de formation et de supervision d'appui pour le renforcement de la capacité du personnel de pharmacie au Vietnam à fournir des informations précises sur les soins de santé, axées sur le client et des services appropriés pour la diarrhée infantile et les pilules contraceptives d'urgence (PCU).

Méthodes

Etude pré-et post-intervention utilisant un concept transversal. Le personnel de pharmacie a participé à three jours de formation sur les relations avec les clients, la bonne pratique pharmaceutique, la diarrhée infantile et les PCU, sur une période d'un mois, comprenant des cours, des discussions, des séances de questions-réponses, des jeux de rôle. Nous avons comparé les données de base et à six mois après l'intervention afin de vérifier les changements dans les connaissances, les attitudes et les pratiques des pharmaciens, à l'aide des statistiques univariées pour trouver des différences significatives.

Résultats

Plus de 1200 pharmaciens ont reçu la formation et l'encadrement. Après les interventions, les connaissances du personnel de pharmacie ont été significativement améliorées dans la plupart des indicateurs mesurés. La connaissance des symptômes de déshydratation en cas de diarrhée est passée de 19 à 88% et celle sur les effets secondaires des PCU est passée de 27 à 77%. Bien que l’évaluation de la pratique actuelle ait montré que cette connaissance n’était pas toujours utilisée, une amélioration significative a été observée. Avant les interventions, 12% donnaient des informations sur les symptômes de la déshydratation, mais 45% l'ont fait par la suite. La proportion donnant des informations sur les effets secondaires des PCU est passée de 13% à 54%.

Conclusions

Offrir un programme de formation et de supervision d'appui est un moyen efficace d'améliorer la connaissance et la pratique des pharmaciens dans les pharmacies privées au Vietnam. Ces améliorations ont la possibilité de mener à de meilleurs soins de santé communautaires.

Abstract

Objetivo

Evaluar la efectividad de una intervención de entrenamiento y supervisión de apoyo para fortalecer las capacidades del personal farmacéutico en Vietnam, con el fin de entregar información sanitaria precisa, orientada al cliente y servicios apropiados para la diarrea infantil y las píldoras anticonceptivas de emergencia (PAE).

Métodos

Estudio pre- y post-intervención, utilizando un diseño croseccional. El personal farmacéutico participó en un entrenamiento de three días, en un periodo de un mes, sobre las relaciones con el cliente, las buenas practicas farmacéuticas, la diarrea infantil y las PAE, consistente en clases, discusiones y sesiones de preguntas y respuestas, y juego de roles. Comparamos los resultados antes del estudio y six meses después de haber realizado la intervención para determinar cambios en el conocimiento, actitudes y prácticas de los farmacéuticos, utilizando una estadística univariada para encontrar diferencias significativas.

Resultados

Más de 1200 farmacéuticos reciben entrenamiento y supervisión de apoyo. Tras las intervenciones, los conocimientos del personal farmacéutico habían mejorado significativamente para la mayoría de los indicadores utilizados. El conocimiento de los síntomas de deshidratación por diarrea aumentó del 19 al 88% y el de los efectos secundarios del PAE aumentó del 27 al 77%. Mientras que la evaluación de las prácticas actuales reveló que dicho conocimiento no se utilizaba siempre, se observó una mejoría significativa. Antes de las intervenciones, un 12% daba información sobre los síntomas de deshidratación pero un 45% lo hizo después. La proporción que daba información sobre los efectos secundarios del PAE aumentó del 13% al 54%.

Conclusiones

Proveer un programa de entrenamiento y supervisión de apoyo es una forma efectiva de mejorar los conocimientos y las prácticas de los farmacéuticos en las farmacias privadas de Vietnam. Estas mejoras tienen el potencial de resultar en unos mejores cuidados sanitarios comunitarios.

Introduction

Pharmacies in low- and middle-income countries provide quick access to medicines and health advice – they are ubiquitous, have convenient hours and locations, require no appointment and offer anonymity (Goel et al. 1996). Consumers in these countries often make pharmacies their first, and sometimes only, healthcare visit for common ailments (Kamat & Nichter 1998; Smith 2009a; Hoa et al. 2011); however, staff at these shops do not always recommend appropriate or adequate medicines or treatment regimens, causing concerns about public health issues such as antibiotic resistance, incorrect treatment of sexually transmitted infections (STIs) and complications from diarrhoea in children. Studies on the quality of pharmacies in low-resource settings also have raised concerns about their limited involvement in preventive services and a high degree of discrepancy between pharmacy staff stated intentions and practice (Smith 2009a).

Reviews of pharmacy intervention studies in low- and middle-income countries in Asia, Central and South America, and sub-Saharan Africa indicate that most researchers and public health officials have used educational and/or training methods. A review of 18 articles in low and middle-income countries worldwide (Smith 2009a,b) noted that all focused on educational methods, while three also included regulatory (coercive) components. Another review analysed 10 intervention studies in sub-Saharan Africa (Wafula & Goodman 2010) and found that all used educational interventions. Both reviews concluded that the training provided in these studies had resulted in some improvements in practice but that more rigorous studies are needed before recommendations on the most effective types of interventions can be made.

The estimated number of drug outlets in Vietnam was more than 40 000 in 2011, excluding drug stores within commune health stations, with private pharmacies comprising nearly 12 000 of the total (Drug Administration of Vietnam 2012). Self-treatment is common, and pharmacies are the most frequently used healthcare facilities, accounting for approximately two-thirds of all health service contacts (World Bank et al. 2001). Several observational studies of pharmacies in Vietnam have been reported. A study in Hanoi observed more than 1800 customers at pharmacies and found that 95% decided themselves which treatments to purchase and 17% of these were antibiotics (Chuc & Tomson 1999). Other studies on the quality of pharmacy practice report that improvements are needed, especially in primary healthcare areas such as supplying antibiotics, managing STIs, treating diarrhoea in children, achieving tuberculosis care and control and managing childhood acute respiratory infection (van Duong et al. 1997; Chalker et al. 2000; Chuc et al. 2001; Lonnroth et al. 2003).

Among the few intervention studies reported from Vietnam, a randomised trial in Hanoi (Chalker et al. 2002; Chuc et al. 2002) showed that knowledge and reported practice of pharmacists who received training and support improved significantly for syndromic treatment of STIs and acute respiratory infections, appropriate referrals and rational dispersal of drugs compared with a group that had no interventions. The interventions also included regulatory enforcement and a peer-influence component.

In 2008, the international non-governmental organisation PATH began a project with the ultimate goal of improving health outcomes among community members in five provinces in Vietnam through a public–private partnership among provincial health authorities, pharmacies and public and private health facilities in the intervention area. Objectives of the project included strengthening the capacity of pharmacy staff to deliver client-oriented primary healthcare information, services and referrals; and increasing community awareness of and demand for pharmacy-based primary healthcare services for two health issues of community concern, childhood diarrhoea and use of emergency contraceptive pills (ECP). While under-five mortality in general and deaths from childhood diarrhoea have fallen in Vietnam, 10% of under-five deaths still were attributed to diarrhoea in 2010 (World Health Organization 2012). According to a 2009 report, only 41% of children with diarrhoea received increased fluids for the condition and 26% received an oral rehydration solution packet (UNICEF & World Health Organization 2009).

In regard to use of emergency contraception, data on abortions in recent years in Vietnam show that the rate is very high. About 1.4 million are performed annually and the rate among married women aged 25–29 years is 0.40 (Nguyen 2012). Surveys do not measure the rate for unmarried women, but data from national surveys suggest that young people in Vietnam are engaging more frequently in premarital sexual relationships, with nearly 14% doing so in 2008 compared with 11% in 2003, although their knowledge on reproductive health and pregnancy remained limited and unchanged (General Statistics Office of Vietnam 2009). The surveys also have suggested that ECP is among the ‘best sellers’ of over-the-counter drugs at private pharmacies. According to a report on incidence and trends in legal abortion worldwide from the Guttmacher Institute, ‘Where the abortion rate is high, it likely reflects that levels of contraceptive use are not sufficient to meet the fertility desires and family planning needs of women and couples’ (Sedgh et al. 2007).

This article reports on strengthening the knowledge and practice of pharmacists in delivering client-oriented, accurate healthcare information and appropriate services for childhood diarrhoea and ECP through a programme of training and supportive supervision.

Methods

Study design and setting

This was a pre- and post-intervention study using a cross-sectional design. The interventions were training and supportive supervision of pharmacists, carried out in selected urban areas in five provinces in Vietnam representing a diversity of geographical settings: Thai Nguyen, Thua Thien Hue, Da Nang, Khanh Hoa and Vinh Long. Urban areas were chosen because the majority of private pharmacies are found in these locations. Effects of interventions were measured by comparing knowledge, attitudes and practice of pharmacists assessed in two types of surveys before and after the interventions. Two coordinators – one from the medical professional section and one from the pharmaceutical management unit of the provincial health department – in each province served as liaisons between the PATH team (two authors of this article and two other staff) and all provincial partners.

Licensed private pharmacies in each province were mapped, and owners of all identified pharmacies were invited to participate in 1-day technical update and orientation workshops. Those who subsequently decided to participate in the training programme signed informed consent documents and were registered for the study. Others invited to the orientations were key stakeholders from the provincial health departments, district health departments, medical schools, pharmacy schools, public and private clinical health services, the Women's Union and the Youth Union.

Intervention content and implementation

We worked with local partners to develop training tools, drawing on materials developed in Vietnam and elsewhere on similar issues, and we incorporated information from the baseline surveys into training plans and materials. In discussions with local partners, we identified teams of provincial-level trainers with experience leading workshops and conducted 5-day training-of-trainers sessions. Pharmacy staff then participated in 3 days of training on customer relations, good pharmacy practice, childhood diarrhoea and ECP over a period of 1 month, with methodology including lectures, discussion, question-and-answer sessions and role-playing. At the end of training, each pharmacy was provided reference materials, such as job aids and information, education and communication (IEC) materials on childhood diarrhoea and ECP, as well as leaflets on diarrhoea and ECP for client distribution.

To further strengthen the capacity of pharmacy staff, PATH and partners developed a system of supportive supervision to provide on-the-job guidance, staff orientations and tools to help ensure the quality of pharmacy staff work on an ongoing basis. The team of supervisors consisted of 11–13 pharmacists and physicians in each province who visited all participating pharmacies quarterly.

Sample selection and size

A simple random sampling technique was applied to identify subjects for pharmacy staff surveys, in which the sampling unit was an individual staff member. The minimum sample size was calculated using WHO's recommended formula (Lwanga & Lemeshow 1991) to detect the difference in staff knowledge before and after the interventions at a level of significance (α) of 5% and power of test (1- β) of 80%. The calculated sample size was 270; a 4% non-response rate for the baseline survey and 2% for the post-intervention survey were taken into account, giving a final sample size of 281 and 275 for the two surveys, respectively. A similar method was applied to calculate the sample size for the mystery client survey, in which the sampling unit was one pharmacy; the required sample size was 220 for each survey (Figure 1).

Figure 1.

Sampling process for pharmacy staff and mystery client surveys. *During the time from registration until training started, a number of new pharmacies opened and asked to participate in the study. After consultations with the Provincial Health Departments, these pharmacies gave written consent to participate and were included.

Pharmacy staff questionnaires

For baseline and post-intervention surveys of pharmacy staff, we developed questionnaires to gather demographic data as well as information on general attitudes and practices, and on knowledge, attitudes and practices of pharmacy staff for diarrhoea in children and emergency contraception issues. The questions for childhood diarrhoea were developed by the PATH team in consultation with clinical experts. The questionnaire and scenario for ECP were adapted from another PATH project (PATH 2006) that had been used in Vietnam and several other countries (Cambodia, Nicaragua and Kenya). Both questionnaires were piloted in two provinces in Vietnam (Vinh Long and Thai Nguyen), and revisions were made to adapt them for local terminology and context.

Mystery client assessment of pharmacy staff

In order to further assess pharmacy staff knowledge and practice before and after interventions, we trained a group of eight men and women in each province, selected from the Youth and Women's Unions, as simulated clients (Chuc et al. 2001; Garcia et al. 2003) who went to project pharmacies to seek health care. These simulated or ‘mystery clients’ were aged 20–30 years and had no medical or pharmaceutical background, but expressed interest and enthusiasm for the work. They participated in a 2-day training programme where scenarios involving clients visiting pharmacies for issues of childhood diarrhoea and need for emergency contraception were presented and explained. Mystery clients practiced observation, memorisation skills and role-playing and participated in a ‘pilot survey’ on at least one of the two scenarios at a real pharmacy that was not included in the formal surveys.

For the baseline and post-intervention surveys, mystery clients visited selected pharmacies in pairs, with one performing the diarrhoea scenario and the other the need for ECP. Both men and women could enact the ECP scenario, with men asking for medicine for a girlfriend or partner, and women asking for themselves. Each pair of mystery clients was assigned three to four pharmacies to visit per day for 3 days; thus, four pairs of clients visited a total of 44 pharmacies in each province. Mystery clients were instructed to buy medicines recommended by the pharmacy staff, observe staff attitudes and actions and remember recommendations and guidance. No later than one hour after mystery clients visit a pharmacy, an interviewer debriefed them and recorded their observations.

In each province, one supervisor was recruited from the provincial health department to attend the training and to supervise the performance of the mystery clients. Supervisors were responsible for seeing that all mystery clients visited their assigned pharmacies each day and for screening all questionnaires completed by these clients.

Timeline of interventions and surveys

After pharmacies registered for the project, we conducted baseline surveys to ascertain the knowledge level and service quality among pharmacy personnel. The baseline survey was conducted in June 2008 and the post-intervention survey in January 2010. Training for pharmacy staff on the selected topics concluded 6 months before the post-intervention survey (June 2009) (Table 1).

Table 1. Timing and sequence of intervention and surveys
ActivityMapping pharmacies; orientation workshops; registrationBaseline surveyDeveloping and printing training and IEC materialsTraining of trainers; training pharmacistsSupportive supervisionPost-interven-tion survey
TimelineJanuary–May 08June 08July 08–February 09April–June 09September 09January 10

Data entry and analysis

We computerised collected data using Microsoft Access 2003 and transferred data to Stata, version 11.0 (College Station, TX, USA), for analysis. We used univariate statistics (descriptions, frequencies, cross-tabulations, chi-squared test) to describe the characteristics of pharmacy staff as well as to look for significant differences in the knowledge level, attitudes and practice of pharmacy staff between the baseline and post-intervention evaluations. Statistical tests used an alpha of 0.05.

Because at least 90% of the pharmacy staff respondents were either intermediate or assistant pharmacists, we did not perform subanalyses of results for participants with different levels of education.

Ethical issues

PATH's Research Determination Committee approved the survey protocols for non-research determination. We asked all participants in the project to take part voluntarily and gave all the right to withdraw at any time without threat or disadvantage. PATH conducted the project in collaboration with officials from the five provincial health departments and the medical schools of each province. In Vietnam, for this type of study where the major interventions are training and supportive supervision or quality assurance on public health issues for healthcare professionals, there is no requirement for ethical approval at the national level. In each of the five provinces, the project protocol was assessed for both technical and ethical aspects and received approval from the Provincial People's Committee for implementation.

Results

Profiles of pharmacists

A total of 734 pharmacies in the five provinces participated in the project, with 396 owners and an additional 890 staff trained, for a total of 1286 people working in pharmacies who received training. Table 2 presents characteristics of pharmacy staff who were interviewed.

Table 2. Characteristics of pharmacists surveyed
ParticipantBaselinePost-intervention
GenderN (%)N (%)
 Female225 (80)223 (81)
 Male56 (20)52 (19)
Professional degree
 Pharmacist (post-graduate)9 (3.2)1 (0.3)
 Pharmacist (university)14 (5)3 (1.1)
 Intermediate pharmacist155 (55.2)154 (56)
 Assistant pharmacist99 (35.2)114 (41.5)
 Others4 (1.4)3 (1.1)
Age and experienceYearsYears
Average age41.640.8
Average experience in pharmacy sector14.513.7

In the baseline and post-intervention surveys, grade 1-specialised pharmacists (post-graduate degree) and pharmacists (degree from a 4-year medical and pharmacy university programme) comprised fewer than 10% of the interviewees. 80% of respondents were female, and the average length of time staff had worked in pharmacy was 14 years.

Pharmacy staff surveys on knowledge of childhood diarrhoea and ECP

We used structured questionnaires to gather information from 281 pharmacy staff at baseline and 275 at the post-intervention survey across the five provinces. For the baseline survey, numbers of respondents in each province were as follows: Da Nang: 53; Thua Thien Hue: 53; Thai Nguyen: 59; Vinh Long: 61; Khanh Hoa: 55. In the post-intervention survey, the number was 55 in each province.

Data showed a statistically significant improvement in knowledge on childhood diarrhoea in all five provinces. As can be seen in Table 3, the proportion of respondents who knew dehydration symptoms after the interventions was more than four times that in the baseline evaluations, increasing from 19% to 88%. Improvements were observed in the proportions of post-intervention interviewees with good knowledge on danger signs of diarrhoea and diarrhoea prevention, reaching 94% and 93%, respectively. Only half as many respondents post-intervention recommended antibiotics, a positive trend.

Table 3. Knowledge of pharmacy staff on childhood diarrhoea and ECP at baseline and post-intervention evaluations
IndexAverage in five provincesOR (95% CI)
Baseline% (= 281)Post-intervention% (n = 275)
  1. OR, Odds Ratio; CI, Confidence Interval.

  2. a

    ORS: oral rehydration solution.

  3. b

    Result is not statistically significant at P < 0.05.

Childhood diarrhoea
 Knew ≥ 2 symptoms indicating diarrhoea22432.7 (1.8–4.0)
 Knew ≥ 3 dehydration symptoms198832.4 (19.7–53.7)
 Knew ≥ 3 danger signs of diarrhoea499415.9 (9.1–29.1)
 Knew ≥ 3 means to prevent diarrhoea579310 (5.8–17.8)
 Would recommend ORSa97992.6 (0.6–15.6)b
 Would recommend antibiotics1572.3 (1.3–4.2)
Emergency contraception
 Knew ≥ 3 side effects of ECP27779.2 (6.2–13.9)
 Knew when to use ECP41858.4 (5.5–12.9)
 Knew time to use ECP after unprotected sex61721.7 (1.2–2.4)

Table 3 also presents results of the two surveys regarding ECP, showing that pharmacy staff knowledge improved significantly after the interventions. After training, 77% of staff could name at least three side effects of ECP compared with 27% at baseline. The number of staff knowing when to use ECP after training more than doubled.

Mystery client surveys of pharmacy staff for attitudes and practice for childhood diarrhoea and ECP

Mystery client surveys allowed comparison of what pharmacy staff said they were doing in direct questionnaires with what they practiced when interacting with customers. For these measurements, mystery clients assessed 220 staff at baseline and 220 after interventions (44 per province). While 220 assessments were conducted, 3 questionnaires at baseline and 4 in the post-intervention survey had missing or inconsistent information and were deemed invalid; these were excluded from the analyses. Table 4 shows changes in knowledge, attitude and skills among pharmacy staff after programme interventions. Overall, 73% of mystery clients at the post-intervention evaluation reported that pharmacy staff had a positive attitude (defined as ‘friendly, helpful, welcoming, or attentive’) when asked about childhood diarrhoea, an improvement from 59% at baseline. In regard to knowledge levels of pharmacy staff on childhood diarrhoea, more than 75% of mystery clients were asked by staff at the post-intervention evaluation about symptoms indicating diarrhoea in the child for whom they sought help, compared with 47% at baseline. In comparison with baseline information, the practice of providing advice to prevent diarrhoea in children improved significantly, from 1% at baseline to 18% post-intervention, although this is still very low. Notably, while staff self-reporting of recommending ORS was much higher than that found by mystery client survey, it did increase substantially after intervention (Tables 3 and 4). Recommendations for antibiotic use were low in both staff survey and survey by mystery client at baseline and decreased after training as well.

Table 4. Mystery client evaluation of pharmacy staff attitudes and practice for childhood diarrhoea and ECP at baseline and post-intervention evaluations
IndexAverage in five provincesOR (95% CI)
Baseline% (n = 217)Post-intervention% (n = 216)
  1. OR, Odds Ratio; CI, Confidence Interval.

  2. a

    ORS: oral rehydration solution.

  3. b

    Result is not statistically significant at P < 0.05.

Childhood diarrhoea
 Pharmacist had positive attitude59731.8 (1.2–2.8)
 Client understood information fromPharmacist61671.3 (0.9–1.9)b
 Pharmacist asked about other symptoms47773.8 (2.4–5.9)
 Pharmacist gave information on dehydration12455.8 (3.5–9.9)
 Pharmacist gave information on danger signs23605.2 (3.3–8.1)
 Pharmacist gave information on prevention11815.2 (4.7–78.0)
 Pharmacist offered ORSa428810.1 (6.0–17.2)
 Pharmacist offered antibiotic1543.9 (1.8–9.7)
Emergency contraception
 Positive attitude48662.0 (1.4–3.0)
 Client understood information from pharmacy staff57631.3 (0.9–1.9)b
 Pharmacist gave information on side effects of ECP13548.3 (4.9–14.2)
 Pharmacist warned ECP cannot protect against STIs11822.4 (5.6–193.2)
 Pharmacist recommended regular contraception methods instead of frequent use of ECP84610.4 (5.7–19.7)

Mystery client surveys on ECP across the five provinces are also reported in Table 4 and showed a trend towards improved attitude and skill in providing information. The proportion of staff with a positive attitude towards these clients rose to 66% from a baseline of 48%. Among other significant improvements were giving information on side effects of ECP and advising use of regular contraception methods instead of frequent use of ECP.

Discussion and Conclusions

Because of the ubiquity of pharmacies in developing countries and the number of people who obtain medicines and advice there, these enterprises have a substantial impact on public health. According to a review by authors from WHO and the World Bank, ‘The private sector exerts a significant and critical influence on child health outcomes in developing countries, including the health of poor children … Pharmacies, drug sellers, private suppliers, and food producers … have an impact on the health of children … governments and programmes that fail to integrate these actors in child health policy and programmes will be seriously constrained’ (Bustreo et al. 2003). The article calls out training of pharmacists as a good strategy for contributing to child health outcomes. Other publications from World Health Organization (1994) and the International Pharmaceutical Federation (1998) emphasise the role of pharmacists in the healthcare system. It follows that strengthening the knowledge and practice of pharmacists could improve public health, although measuring the latter in large populations is difficult.

The study reported here is one of a few pre- and post-intervention studies that have compared knowledge and practice reported by staff as well as actual practice and attitude of pharmacy staff measured by client assessment in a developing country. Another study in Vietnam (Chalker et al. 2002) trained pharmacists in 22 control and 22 matched intervention pharmacies and showed significant improvement in the intervention group as measured by direct questionnaire, but had no assessment for actual practice. We saw considerable discrepancies between pharmacists' answers to direct questions and their practice assessed by mystery clients in our study, so it seems important not to rely only on reported practice. Our results add to evidence (Ross-Degnan et al. 1996; Chalker et al. 2002; Chuc et al. 2002; Garcia et al. 2003; Qidwai et al. 2006) that providing training to staff at private pharmacies is an effective way to improve pharmacy practice, which has the potential, ultimately, to improve community health outcomes.

In our study, pharmacy staff improved their knowledge of how best to care for clients with concerns about childhood diarrhoea or ECP after participating in training and a programme of supportive supervision. Staff were also significantly more likely to put this knowledge into practice when approached by customers after training, although at a lower rate than indicated by direct questioning.

Questions arose in regard to some of our findings; for example, in the assessment of pharmacy staff knowledge on childhood diarrhoea, knowledge improved substantially for identifying danger signs and dehydration symptoms after training, but not so impressively for identifying ‘symptoms indicating diarrhea’. The reason for this discrepancy is unclear, but in the mystery client assessment of practice for childhood diarrhoea, pharmacists were significantly more likely to ask about symptoms and give information on dehydration and danger signs after training, so the training clearly had a positive impact.

Our study had some limitations. First, there was no control group of pharmacies that received minimal training such as provision of printed information, or training on another topic (Garcia et al. 2003). Further, we did not arrange to assess the same individual staff via direct questionnaire and mystery client survey: while a pharmacist who answered a questionnaire also may have been the subject of a mystery client visit, these coincidences, if any, were not recorded purposively. However, the large number of observations for both surveys provides confidence that the interventions strengthened pharmacy practice in the survey populations. The questions in our surveys were all open-ended – there were no questions with response options involving confirming a statement and no prompting by interviewers – minimising the possibility of acquiescence response bias (Saris et al. 2010). Another limitation of the study was that we chose to work with urban pharmacies, as the majority of pharmacies in Vietnam are located in urban areas. The results thus may not be generalisable to rural and mountainous areas of the country.

The fact that Vietnam has more than 40 000 drug outlets makes the task of improving practices at these facilities daunting. While our study showed that interventions at the private pharmacy level can produce change, clearly the efforts must expand to improve pharmacy curricula in colleges, thereby building correct practice into the profession from the beginning.

Acknowledgements

The study reported in this article was part of the project ‘Enhancing the Role of Pharmacies as Community Health Care Providers’ funded by the Atlantic Philanthropies. We would like to thank all pharmacists/pharmacy staff who participated in this study. Special thanks go to leaders and officials of the five provincial health departments for their cooperation and support; PATH staff, provincial trainers, supervisors and data collectors who involved in trainings of pharmacy staff and data collections in the field. We also thank Molly Derrick and Kate Bagshaw for proofreading the article.

Ancillary