Assessment of the knowledge, attitude and practices of prescribers regarding malaria diagnosis: a cross sectional study among Ghanaian prescribers

Introduction Malaria has proven to be the most fatal parasitic disease known to man. Among the pillars to malaria control are early and accurate diagnosis. In 2010, the World Health Organization launched its test, treat and track initiative which seeks to ensure that all suspected cases of malaria are tested. However, after several years of implementation, the use of malaria tests in diagnosing malaria has not been optimum. This study was conducted to assess the level of knowledge of prescribers on malaria Rapid Diagnostic Test and microscopy and to determine factors influencing prescribers' decision to request and use malaria tests in practice. Methods A cross sectional study was carried out among 100 prescribers of various categories working in 4 hospitals in Ghana in March 2019. A pre-tested self-administered questionnaire was used to collect information on knowledge, malaria diagnostic practices and challenges faced by prescribers regarding parasitological testing for malaria in their health facilities. Results Overall, 73% of respondents had good knowledge on malaria diagnostics. Routine use of malaria tests in diagnosing malaria was reported as 84%. Only 9% reported complete reliance on test results. Most participants (90%) reported awareness of the test-based case management of malaria. Conclusion This study demonstrated that even though there was a high level of awareness of the test-before treatment policy among prescribers, significant numbers did not routinely request a malaria test for all suspected cases of malaria. Factors cited as barriers by prescribers were both health worker and health-system related that are all potentially modifiable.


Introduction
Malaria continues to be the most dangerous parasitic disease afflicting man. According to the World Health Organization (WHO) in 2017, there was an estimated 219 cases of malaria worldwide with about 435000 deaths occurring as a result of malaria [1]. In Ghana, malaria is hyper-endemic and continuous to impose a major health and economic burden on the people. Malaria accounted for 38.1% of all out patient illness and 27.3% of all admissions in 2015 according to the National Malaria Control Programme (NMCP) annual report [2].
There is a regional variation in malaria parasite prevalence among children aged 6-59 months. The Eastern and Central regions have the highest (31%) and the second highest (30%) malaria prevalence, followed by the Volta (28%) and Northern (25%) regions. Malaria prevalence is lowest (5%) in the Greater Accra region [3]. Meanwhile, some researchers have discovered over-diagnosis and overtreatment of malaria as one of the main reasons for the higher cases [4-6].
Following evidence of over-diagnosis of malaria, the WHO launched the T3: Test. Treat. Track initiative. This global initiative was developed to urge all malaria endemic countries, donors and the global malaria community to scale up diagnostic testing, treatment and surveillance for malaria. Every suspected malaria case is therefore to be tested and every confirmed case should be treated with a quality antimalarial medicine [7]. In its annual report in 2010, the national malaria control programme (NMCP) of Ghana stated that presumptive diagnosis of malaria was very high and most febrile illness were wrongly captured as malaria. Ghana therefore adopted a test-based management of malaria and is expected that all prescribers confirm all suspected cases of malaria before commencing treatment for malaria [8].
Considering the fact that, the cost of using artemisinin-based therapies to treat malaria is very expensive, over-diagnosis of malaria coupled with its attendant over-treatment increases the economic burden of the disease. There is also a concern for possible resistance of the Plasmodium parasites to the artemisinin-based combination therapies. The NMCP reports that Ghana has recorded a rise in the proportion of OPD malaria cases tested by microscopy or RDT since 2012 from a low figure of 38.9% in 2012 to 73.6% in 2015. However, there is still more to be done so as to achieve the NMCP target of reducing malaria morbidity and mortality by 75% and to provide parasitological diagnosis to all suspected malaria cases and provide prompt and effective treatment to 100% of confirmed malaria cases by 2020 [2]. This objective cannot be achieved if prescribers continue to treat malaria without laboratory evidence. To achieve this objective, it is important that all prescribers involved in caring for patients with fever possess adequate knowledge of the various malaria diagnostic tests available in their facilities. Previous studies in Ghana have shown the reluctance of some clinicians to test before treating every suspected case of malaria. There is however paucity of information in all literature reviewed on the knowledge level of prescribers in Ghana on microscopy and mRDT as well as their attitudes and practices towards the use of these malaria diagnostic methods. This study was therefore conducted to determine factors that influence adherence of prescribers to the test-before-treatment policy of treating malaria in Ghana.

Methods
Study design: this was a cross sectional study conducted in March 2019. The study made use of a semi-structured self-administered questionnaire.
Population: all prescribers working in four major hospitals in two different regions in Ghana were eligible for recruitment into the study.
These hospitals are located in the capital cities of two regions (Central and Western) which all lie along the coast of Ghana. Malaria is hyper endemic in these areas. The selected hospitals have large numbers of prescribers and were therefore suitable for this cross-sectional study.
The selected hospitals in the central region were the Cape Coast Teaching Hospital which is a tertiary level health facility with an estimated 60 prescribers and the University of Cape Coast Hospital which is a secondary level hospital with an estimated 15 prescribers.
In the western region of Ghana, the selected hospitals were Efia Nkwanta Hospital which is the Regional hospital with an estimated number of 30 prescribers and the Ghana Ports and Harbors hospital with an estimated number of 5 prescribers.
Sampling: total population sampling technique was used. In this type of purposive sampling, the entire population of prescribers at all study sites were eligible for recruitment into the study. Prescribers working outside the selected hospitals and those not willing to participate were excluded.

Study
Instrument: a semi-structured self-administered questionnaire developed from existing literature [9-11] was used.
Steps were taken to validate the questionnaire in the population.
These included establishing face validity by asking experts in malaria diagnostics to go through it. The questionnaire was then pre-tested using 10 prescribers at the Cape Coast Metropolitan hospital.
Cronbach Alpha of 0.78 was found after test for the internal consistency of the questionnaire was carried out. The questionnaire was arranged into the following sections: (a) socio-demographic information; (b) knowledge on malaria rapid diagnostic tests and microscopy; (c) malaria diagnostic practices of prescribers; (d) training issues and challenges and (e) attitudes of prescribers towards malaria tests as well as recommendations of prescribers that will improve malaria diagnosis. The socio-demographic section was used to collect data such as age, sex, profession and years of practice of respondents. Prescribers' knowledge on malaria diagnostic tests was assessed using six questions. A correct answer scored 1 and a wrong answer scored 0. An "I don't know" response attracted a score of 0.
A respondent was deemed to have adequate knowledge if he scored more than 50% (at least 4 out of 6).
Ethical consideration: ethical clearance was obtained from the Institutional Review Board of University of Cape Coast (UCCIRB). The main areas of concern in ethical involvement with participants included the issues of privacy, anonymity and confidentiality. These issues were addressed by training all those involved in the study to ensure confidentiality. Also, no names, staff numbers or any form of personal identification were used. A coding system was used to identify participants. The study was funded by the investigators. None of the investigators have declared any conflict of interest with regards to this study. Permission was sought from the management of the various hospitals before data collection began. Informed consent was obtained from each respondent before the administration of questionnaires.
Data analysis: data collected was entered into SPSS version 20.
Descriptive statistics was used to summarize the data. Pearson's chisquare was used to examine the relationship between prescribers' demographic characteristics and their knowledge level on malaria diagnosis. Level of significance for all tests of association was put at p<0.5.

Results
Out of a total of 110 questionnaires were distributed, 100 prescribers responded giving a response rate of 90.9%. Most of the participants were house officers (57.0%), followed by general practitioners (31.0%). Participants were aged between 24 years and 63 years with a mean age of 29.7± 6.2 years. Majority of respondents were in the age group of 26-30 years. Male prescribers were in the majority (63.0%). Clinical experience of participants ranged from <1 year to over 10 years. The socio-demographic characteristics of respondents are summarized in Table 1.
Knowledge on malaria diagnosis: the basic knowledge of participants on malaria diagnosis was assessed using 6 questions all carrying equal scores (

Discussion
The primary aim of this study was to determine the knowledge level of prescribers on malaria diagnosis. This challenge has been cited as a major barrier to the compliance to national guidelines by clinicians in many previous studies conducted in other malaria endemic low-income settings [20,21]. Unavailability mRDT and microscopy services will in no doubt encourage prescribers to resort to presumptive treatment of malaria. It is widely documented that a very important factor in improving malaria diagnosis is the availability of malaria diagnostic tests in health facilities [22]. In Ghana RDTs are provided by the government and some NGOs for free to Public health facilities. The government of Ghana and its partners through the NMCP must therefore address the shortage of the RDT kits and scale-up mRDTs availability to private facilities as well. Among respondents found to have inadequate knowledge on malaria diagnosis were newly trained doctors who are expected to be up to date with current management strategies. This finding means that training institutions for doctors, physician assistants and nurse practitioners must give more attention to training on malaria given the public health significance of malaria in our setting. Studies have shown that young clinicians are more likely to comply with current case management guidelines compared with older colleagues [23]. Thus, the best strategy to achieve a behaviour change with regards to malaria tests is to inculcate the policy in clinicians whilst still undergoing training at school. There was a general consensus among the respondents in this study that the WHO policy on test-based management of malaria is good. This is an endorsement suggests that when steps are taken to address the barriers identified in this study the aim of the NMCP of provision of parasitological diagnosis to all suspected malaria cases and prompt effective treatment to 100% of confirmed malaria cases by 2020 can be achieved.
Limitations of study: the study made use of a questionnaire which was used to capture self-reported experiences, hence primarily relied on provision of the right information by respondents. There is a possibility of reporting bias as respondents may provide socially desirable answers. This was however minimized by assuring a high degree of confidentiality. Also, the study did not verify information provided by participants as case notes were not examined and also

Competing interests
The authors declare no competing interests.

Authors' contributions
James Kojo Prah: conception, design, acquisition and interpretation of data and drafting the manuscript. Richard Pinkrah and Elias Ewudzi-Acquah: acquisition, analysis and interpretation of data. Atta Yeboah-Sarpong: reviewing of several drafts of the manuscript. Table 1: frequency distribution of socio-demographic characteristics of the respondents