Ethics statement

Ethical approval for this study with approval number AD 13/479/4121 A was obtained from the Oyo State Ministry of Health, Ibadan prior to the conduct of the study. Permission and approval to conduct the study was also obtained from the Head of all the selected PHCs. In addition, the purpose of the study was explained to the participants and their written consents were obtained before the questionnaires were administered. Confidentiality and anonymity were ensured and the execution of the research was conducted according to the guidelines of World Medical Association (WMA) declaration of Helsinki regarding ethical guidelines and principles for conduct of research involving human subjects [12].

This study utilized a cross-sectional study design. Oyo State, with Ibadan as its state capital, is located in Southwest, Nigeria. Oyo state is situated in the southwest geo-political zone of Nigeria with a population of approximately 6,190,000 spreads over an estimated land mass of about 28,000.00 square kilometres. Oyo State consists of 33 local government areas (LGAs) that function as administrative units out of which 5 of the 33 local government areas make up the state capital [13]. Ibadan metropolis was the centre of administration of the old Western Region Nigeria since the days of the British colonial rule. The principal inhabitants of the state are the Yorubas. The city hosts quite a number of small, medium and large-scale industries involved in the production of food and beverages, clothing and textiles, chemicals and pharmaceuticals, confectionaries, leather-works and furniture, plastics, blocks etc [13]. The study was carried out in 261 selected government-owned primary health facilities in Oyo State.

The sample size calculation was done using the Kish formula.

[14]

Considering design effect of 1.5

Considering a non-response rate of 10%

Primary healthcare workers directly involved in malaria case management in the selected government-owned primary health facilities in Oyo State were enrolled as participants in this study. This study employed a multistage sampling technique.

Stage 1: From the three senatorial districts in Oyo state, 8 LGAs were selected from each senatorial district using simple random sampling by balloting. Stage 2: Out of the selected LGAs in each senatorial district, 6 wards were selected using systematic sampling technique. Stage 3: From the 6 selected wards, using proportional allocation, a total of 261 Primary Healthcare Centres (PHCs) was selected from the six wards. Stage 4: One healthcare worker providing malaria services (preferably the head or any health worker that was delegated by the head of malaria services) from each of the 261 selected PHCs was invited to participate in the study. All heads of primary health care workers who were not present in the facilities while data collection was ongoing were traced and enrolled. Visits were repeated at least thrice to ensure that all eligible frontline health workers were enrolled in the study.

A pretested standardized electronic structured questionnaire was used to collect information from the participants. The questionnaire consisted of 5 sections which included the socio-demographic characteristics of respondents, the knowledge of healthcare workers on malaria diagnosis and COVID-19, the practices of healthcare workers regarding malaria diagnosis, the practices of healthcare workers regarding COVID-19 differential diagnosis and challenges that healthcare workers face regarding malaria diagnosis. Data were coded, cleaned and analysed using SPSS version 25 software and MS-Excel. The dependent variables for this study were knowledge and practices of healthcare workers regarding malaria diagnosis and COVID-19 diagnosis while the independent variables were sociodemographic and socioeconomic characteristics. Means and SD were used to summarise the continuous variables while frequencies, proportions and charts were used to summarise the categorical variables.

Each knowledge and practice questions were scored and categorized into good and poor knowledge on malaria diagnosis, good and poor malaria-related diagnosis practices and good and poor ability to differentiate between malaria and COVID-19. Primary healthcare providers that scored 80% and above of the total possible scores at all levels (Knowledge and Practice) were dichotomized as good while those that scored below 80% were categorised as poor.

Chi-square tests were used at the bivariate analysis level to test the association between categorical outcomes. Binary logistics regression with respective odd ratios and confidence intervals were used at multivariate analysis to explore the relationship between the sociodemographic, socioeconomic characteristics and categorical outcome variables. The level of significance was set at p-value < 0.05.

Access to laboratory diagnosis

The gold standard for malaria diagnosis is the malaria microscopy. Access to malaria microscopy especially in rural areas is limited, however due to donor funding, the use of the Rapid Diagnostic Test kits (RDT) has become a little popular in malaria diagnosis in Nigeria. On the other hand, suspected COVID-19 infection requires the use of a reverse transcriptase polymerase chain reaction (RT-PCR) in addition to clinical and radiological signs for diagnosis [17] in addition, rapid antigen test kits are largely unavailable for use in health facilities [18]. The RT-PCR for the diagnosis of COVID-19 is not available at the primary healthcare or district levels in Nigeria. A patient presenting with symptoms suggestive of COVID-19 at this level, will leave the healthcare worker with her clinical acumen to suspect COVID-19 while she battles with determining whether it is malaria due to unavailability of malaria RDTs. The COVID-19 pandemic has no doubt also led to the disruption of supply chain for malaria RDTs as companies now focus more on the production of rapid tests for COVID-19 [15]. According to the most current guidelines for diagnosis and treatment of malaria, parasitological confirmation is required in all suspected cases of malaria. Based on the national malaria policy, all forms of treatment based on clinical diagnosis is disallowed [19, 20]. The challenge remains availability of test kits and skills to perform the tests accurately resulting in a high rate of clinical diagnosis for malaria and unreliable test results respectively. This highlights the need to improve supply of these test kits to PHCs.

Wrong diagnosis and wrong treatment

Most of our respondents were CHEWS and they constitute majority of healthcare service providers in primary healthcare centres in Nigeria [21] Our study showed that 75.9% and 95.8% of respondents had participated in trainings on COVID-19 and malaria diagnosis and treatment respectively resulting in good knowledge on both diseases. However, the good knowledge did not translate to practice, as 85.6% of respondents had poor malaria diagnosis practices and could not properly differentiate between malaria and COVID-19. This raises concerns as COVID-19 and malaria have many symptoms in common including fever, difficulty in breathing, fatigue and headache, this makes differential diagnosis difficult [22] especially in resource poor settings with limited access to laboratory diagnosis. Healthcare providers in primary health facilities being the first responders in most cases should be able to identify, correctly diagnose and refer to where COVID-19 cases can be managed if necessary.

While it is necessary for frontline health care providers to have access to and be able to accurately carry out malaria RDT, patients with negative test results must be carefully managed [23] and should further raise the suspicion for COVID-19. Despite the high knowledge of malaria and COVID-19 noted in our study, 86.2% and 69.7% of respondents had poor diagnosis practices for malaria and COVID-19 respectively. Several factors can affect the poor diagnosis practices noted in our study including the level of education of service providers found at this level of care and their ability to be able to clinically suspect malaria and or COVID 19. Further, being able to accurately conduct mRDT, interpret test results [24] and adhere to treatment guidelines remains a challenge [21]. The sensitivity and specificity of malaria rapid test kits may contribute to poor practice. It has been noted that at the population level, varying degree of transmission with varying immunity levels affects parasite density and subsequent exhibition of symptoms triggering the need for a parasite-based diagnosis. Lower test sensitivity has been observed in areas of low transmission and low parasite density [25].

Multiple organ failure in adults and respiratory distress in children are commonly seen in COVID-19 cases and remain major signs of severe malaria [11]. Laboratory diagnosis for COVID-19 remains largely unavailable in primary healthcare centres in Nigeria, thus identifying and differentiating between cases presenting with such symptoms remains a huge challenge. The symptom overlap may lead to delay in treatment which may inadvertently affect the treatment outcome [26]. Hence, there is a need to improve malaria diagnosis practices and COVID-19 differential diagnosis practices among primary healthcare workers to improve patient health outcomes.

Training of healthcare providers and service integration

The primary healthcare remains the first line of reach to the communities [27] and several healthcare providers have been trained to manage malaria. To properly handle suspected COVID-19 cases, the training on Infection, Prevention and Control (IPC) was rapidly scaled up to most healthcare providers [28], while online COVID-19 trainings were made available to providers [29]. As revealed by this study, over 96% respondents have had recent training on COVID-19 related IPC while 75.9% had online courses on COVID-19, and 95.8% trained on malaria diagnosis and treatment. These trainings were done in isolation of each other and did not converge to link the two diseases to create a better understanding. Linking these trainings will improve case identification, diagnosis and management for both illnesses. Provision of malaria supplies in laboratories where COVID-19 are carried out and vice versa will reduce missed opportunities for both diseases. Furthermore restructuring service provision, as was done for tuberculosis and HIV to provide a more collaborative approach will aid easy management of cases, this may include making available malaria RDTs where COVID-19 tests are done, further making diagnosis and case identification easy [30]. The need for training, retraining and on-the-job mentoring of primary healthcare workers on the proper use of mRDT for malaria diagnosis to minimise clinical diagnosis and unreliable test results is vital to improve management of suspected cases at this level.

Limitations

Our study focused on the healthcare providers at the primary healthcare level thus limiting its applicability in terms of scope. The calculated sample size was 262, however, 261 health facilities were selected for the study as the last facility was non-functional as at the time of the study. The addition of secondary and tertiary healthcare facilities (with specialized COVID-19 diagnostic and malaria testing services) would have been desirable and is recommended for further studies.