Background
The burden of malaria persists in many parts of Africa despite the availability of many interventions that are focused on preventive and therapeutic strategies [
1]. Artemisinin-based Combination Therapy (ACT) has been adopted as the most effective treatment option against malaria in many countries following the widespread malaria parasite resistance to more affordable anti-malarial drugs such as chloroquine and sulphadoxine-pyrimethamine (SP) [
2]. Furthermore, quinine remains the most widely used anti-malarial drug in the treatment of severe and complicated malaria in many malaria-endemic regions [
3,
4]. Kenya adopted the new malaria policy in 2004, which recommended the use of artemether 20 mg-lumefantrine 120 mg (AL) as the first-line drug for treatment of uncomplicated malaria [
4]. Apart from AL, other artemisinin-based combinations and monotherapy drugs are widely available and easily accessed in private outlets in Kenya [
5,
6] and other malaria-endemic countries [
7].
The implementation of a policy is a continuous process and involves many activities including, but not limited to, in-service training (to update the personnel in the field with new knowledge) and adoption of new practices (which comes along with changes in treatment policy). In Kenya, there is a greater effort in training of health service providers in the public sector on the use of ACT relative to the private sector despite the vital role they play in malaria treatment [
7,
8]. Data on evaluations of knowledge and practices before or during implementation of the ACT, and quinine as a second-line anti-malarial in the private sector in relation to the other sectors in malaria-endemic region of Kenya, is scanty [
8‐
11]. In addition, the multiple-dose regimens for most ACT and quinine are rather more complicated in comparison to a single dose needed for SP. This situation has further been complicated by the fact that previous findings in Kenya demonstrate that only 11% of health workers dispensing AL in public facilities were without formal clinical qualification [
8].
Studies on health-worker practices have reported mixed findings. For example, a previous study to determine the predictors of quality of health-worker practices reported failure of health workers to prescribe AL to all deserving cases due to insufficient supply of AL [
11]. This observation further raised fears of stockouts and patients’ preferences for SP over AL because of its simple dosage [
11]. In the same study, AL was provided free to consumers, however, the health workers had to assess and prioritize cases that they deemed deserved to receive AL as it was considered expensive by the government. Other practices included prescription of available drugs such as amodiaquine since they were continuously supplied to the health facilities despite the policy change to ACT [
11].
The successful policy outcome in the appropriate use of ACT and quinine, therefore, is dependent on provision of suitable knowledge to the health providers in all drug outlets and to consumers. Currently, literature on providers’ knowledge and practices on treatment policy and dosing regimens on the use of AL following the policy change in Kenya is dwindling. Despite the use of quinine for decades, no study has been carried out to evaluate providers’ knowledge on its use for malaria treatment in Kenya. The current study evaluated providers’ knowledge and practices of treatment policy and dosing regimens with AL and quinine in the public, private and not-for-profit drug outlets.
Discussion
This survey was performed six years after the government of Kenya had adopted the policy on ACT for treatment of malaria. This policy recommended the use of AL as a first-line treatment for uncomplicated malaria, while quinine was still preferred for treatment of complicated and severe malaria. In 2010, the policy was universally changed to include the use of ACT for only laboratory-confirmed malaria cases [
15,
16]. The current study therefore compared knowledge and practices in providers in public, private, and not-for-profit outlets on treatment policy and dosing regimen on recommended anti-malarials in malaria-prone areas of western Kenya.
The results revealed that there were more in-service trained personnel in the past two years dispensing drugs in public outlets (72.2%). This figure surpassed the target 60% set up by the Ministry of Health (MoH) in 2006 [
17] and was much higher than the figure (46%) previously reported in another study conducted one year after the policy change in Kenya [
8]. However, this proportion trained is below a recent proposed target of 100% as per the Division of Malaria Control (DOMC) Monitoring and Evaluation report for 2013 [
18]. The expansion of the in-service training coverage in public facilities is encouraging despite the fact that the private sector is still lagging behind in training its drug providers. Further findings demonstrated that a considerable bulk (45%) of private outlet staff dispensing drugs had no clinical qualification. This confirms previous findings in Kenya and Democratic Republic of Congo [
19,
20] in which it was shown that a lower proportion of drug providers at the private outlets were trained. The private-sector qualification has been a staggering issue for a very long time in developing countries despite the significant role they play in medication management and provision of relevant information to patients [
21]. The findings in the private sector were contrary to that in the not-for-profit outlets in that a higher proportion of drug providers (54.5%) were trained in the not-for-profit outlets. This disparity in the proportion of those trained in private
vs not-for-profit outlets may be attributed to the fact that the mission hospitals (which are the major outlets in the not-for-profit category), receive government support and tend to effect policies enacted by the government. The 54.5% training received in this sector is nevertheless inadequate and may have been contributed by the recent mushrooming of non-governmental interventions in Nyanza Province (e.g. The Millenium Villages Projects and Ogra Foundation), most of which targets integrated management of HIV and AIDS and malaria. The qualification of providers in this sub-sector has not yet been evaluated. Meanwhile, for better results, both the private and all players in the not-for-profit sector need to be involved in the implementation of ACT policy. In addition, the providers should have regular and adequate training in drug dispensation to ensure correct administration of anti-malarials.
An interesting observation was that about half (49%) of providers in private outlets could not mention AL as first-line treatment for uncomplicated malaria, while approximately 52% mentioned quinine incorrectly for treatment of severe malaria. This observation provides additional evidence of inadequate knowledge and skills on the treatment policy. The reason behind low knowledge of quinine could be attributed to it being less frequently stocked by the private sector due to reasons, such as: quinine is mostly prescribed as a second-line drug and therefore most customers would only ask for it after the failure of other drugs, or it is either less publicized or not publicized at all in media, and there is decreasing awareness about it, as compared to first-line AL. Private providers tend to stock drugs that are frequently sought by consumers as it was demonstrated in a previous study that the type of drug acquired for use in a particular region is influenced by awareness on the type of anti-malarial in the market [
22].
Questions were further raised on the dosing regimen with AL and quinine for particular weights (children weighing 9 kg and adults weighing 45 kg) for uniformity and easy recording. Disparity was recorded in knowledge of dosing regimen with the two anti-malarials in public, private and not-for-profit outlets. For example, a generally higher knowledge of dosing regimen for both children and adults was observed in public and not-for-profit outlets. This was in part influenced by in-service training by non-governmental bodies reported in this study, an indication of need for training. Some studies related to prescription of AL reported that incorrect weight-specific prescriptions of AL were sporadic and the packaging would have influenced dosing regimen [
23]. One previous study carried out in Tanzania concurs with the current findings in that most dispensers in private pharmacies could not state the dosing schedules of AL without referring to the package leaflets [
24]. A higher knowledge of treatment regimen with quinine for severe malaria in public outlets reported in the current study could have potentially been influenced by the concurrent in-service training and health briefs, contrary to previous observations in Democratic Republic of Congo, in which it was shown that training of the providers did not improve the knowledge of the dosing schedule with quinine among the Village Health Volunteers (VHV) and pharmacy owners [
25]. However, the reasons for the differences on awareness level on the most recent ban on anti-malarials being observed in the outlets still remains unclear.
In the private and not-for-profit outlets, the providers knew more than three symptoms of malaria, a good indication for subsequent prescription, even though the malaria treatment guidelines still confine AL and quinine prescription to registered pharmacies and should only be provided to confirmed cases of malaria. It is important to mention that although the guidelines insist on laboratory tests before selling the policy-recommended drugs to customers, the practice of providers in private outlets selling anti-malarials without prescription has been observed in various outlets in the study region and this practice was significantly influenced by in-service training.
The providers at the drug outlets are charged with the responsibility of advising their customers on all matters pertaining to adherence to treatment schedule and proper usage of drugs. These practices are to be enhanced by the providers since the policy indicates that there should be inclusion of a written prescription, sale of full doses of drugs, and even advice to take the first dose immediately. In addition, the providers should educate the patients on the need for adherence, the possible side effects and the need to take more fluids during the treatment. It is encouraged that the providers verbally describe the treatment course and the importance of adherence.
The discrepancy in the practices by the providers in private outlets is disquieting. The sale of partial packs of AL shows lack of commitment to change even in the face of ACT, despite the fact that the type and duration of treatment in private outlets is determined by the clients’ ability to pay [
26,
27]. The low prescription of AL in private and not-for-profit outlets could be attributed to lack of or insufficient in-service training. This followed the observation that training influenced the prescription practice with better prescription seen in the public sector (67.9%). In a previous study exploring reasons for health workers not prescribing AL despite the drug being in stock at the public health facilities, it was observed that most of the health workers were only responding to general health system weaknesses leading to non-adherence to the treatment guidelines [
11]. A low level of prescriptions of the nationally recommended drug was reported in yet another study where only 26% of children who needed treatment with AL received a prescription for this drug according to national guidelines [
23]. The deviation from the guidelines observed in the current study shows inadequate skills and knowledge in preparation for implementation of the new policy in the private sector.
It is important to note that the difference in the outcome observed in the current study is attributed to in-service training and health briefs, which in essence should take precedence in readiness for the on-going implementation process. There is critical need to find an intervention which would address this disparity, owing to the importance of this sector in health provision and given that absence of training in AL policy influence drug-prescribing practices in Kenya [
28]. One way of increasing coverage can be through training the trainers from the same sector who will further train their counterparts. This has been shown to improve malaria treatment in private drug outlets in Bungoma District in western Kenya prior to policy change [
29]. Another approach may be to train the drug dispensers directly in organized workshops, give information, sensitization and education on the new policy and health briefs on quinine regimen, although this method might be a more expensive endeavour for the government. A previous study reported improvement on health-worker performance in care and treatment of patients after conducting educative seminars and training of health workers [
30], confirming a hypothesis of increased training and awareness prior to implementation of a change in drug policy.
Focus on public sector to implement policy change was reported in a study previously carried out in Kenya a year after the implementation process of the new malaria policy [
8]. In that study health workers were trained in a cascade manner leading to an increased number of trained providers [
8]. The tendency to focus on public sector to implement changes in new policy guidelines downplays the call by the World Health Organization to include private sector in malaria treatment due to their significant role in pharmaceutical management and provision of relevant information to patients, thus enhancing the improvement of rational drug use [
21]. Formulation of good policies may not be a guarantee for proper interpretation of knowledge and practice. It must be accompanied by sensitization for the achievement of the outcome. A recent study carried out in neighbouring Tanzania reported positive results in achieving policy change through sensitization of targeted communities prior to the implementation process [
31].
The current study had limitations in that it was only a one-time point survey and thus it could not evaluate the impact of other factors on provider’s knowledge and practices on dispensation of antimalarial drugs. Furthermore, the current findings are based on a sample of 288 outlets. As such, it would be critical to carry out an extensive longitudinal survey in a wider coverage of the outlets in the endemic areas to exhaustively delineate the impact of other factors on provider’s knowledge and practices on dispensation of anti-malarial drugs.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
CAW designed, carried out the survey studies in the drug outlets and participated in the drafting of the manuscript. JHO, ROO, BOA and CO participated in the drafting of the manuscript. CAW and CO performed the statistical analysis. All authors read and approved the final manuscript.