Correlation Analysis of Malaria Cases and Meteorological Factors in Lagos State, Nigeria.

Background Malaria transmission affects malaria infection rates. There are several potential drivers of malaria transmission. A suitable meteorological factor such as rainfall, temperature, and relative humidity encourages the breeding of the vector. This improves the survival of the parasite in the host. The female Plasmodium falciparum plays a crucial role in the variability of malaria prevalence. Lagos State is a coastal malaria-endemic area in Nigeria. This study presents a correlation analysis of malaria cases and meteorological factors between the periods of January 2015 to April 2018 in Lagos state. Methods The study was a secondary data analysis of meteorological variables and records of malaria cases reported by health facilities in Lagos state. We accessed weather variables through free access “weather underground.com” a meteorological data sharing service system (MDSSS). The MDSSS provides real-time online weather information from four meteorological monitoring stations in Lagos state. We accessed the malaria cases through the district health information system 2 databases. It is used to report cases of malaria by all the private and public health facilities in the state. We performed the correlation analyses to show the relationship between temperature, humidity, rainfall, and malaria cases at a 5% level of significance. We analysed data using the statistical package for social sciences version 25. Malaria cases peaked between the period of July to November 2016 and the period of April to May 2017 and declined between March to May 2017. The temperature, relative humidity, and rainfall showed a positive correlation with malaria cases. The temperature is most correlated with the occurrence of malaria cases (r = 0.65, p< 0.02). This correlation analysis provides an approach for studying the impact of meteorological variability on the prevalence of malaria cases. This can help to forecast the malaria epidemic while preparing for the elimination of malaria in Lagos state.


Background
Malaria has remained a notable public health challenge. According to the 2019 malaria report, it affects about 228 million people worldwide (95% confidence interval [CI]: 206-258 million(1).
It is the leading cause of morbidity and mortality among over 90% of the populace in Africa(1,2). Nigeria has the largest burden of malaria in West Africa because of poor prevention, and control measures (3). Occurring in Nigeria is over half of these malaria cases in West Africa(1,2). Endemic in Nigeria, malaria poses severity in vulnerable groups like pregnant women and under-five (3,4). Most Nigerians live in areas of mesoendemic transmission conversely, some live in areas of hyper-holoendemic transmission (3,5). The climatic factors relate to the seasonality and transmission of malaria. Invariable Nigeria accounts for several varying climatic conditions across states. In the last decade, there have been changes in malaria endemicity with variations in patterns of parasite risk across the various states (3,6). Studies show variation in the prevalence of malaria across the country. The prevalence of malaria in Lagos state is 5.4%, and Ogun State is 56.6% (3).
There is little information on the direct potential effects of weather factors on malaria incidence in Lagos state(4). The predisposing climatic factors of malaria are high temperature, high humidity, and persistent rainfall. An accurate weather prediction is an essential tool for planning malaria prevention and control, and for estimation of disease burden (6). There has been a huge financial commitment to eliminate the disease (7). Poor control practices, ineffective disease surveillance, and fluctuating environmental factors have made malaria to remain a major public health challenge. The entire population is at the risk of infection with malaria, but the vulnerability levels vary by age, place of residence, and atmospheric conditions(2,8) Studies have shown the seasonal variation of other tropical diseases like dengue fever and a host of others (9). There has been an association between annual variation in malaria incidences and climatic factors like rainfall and temperature (10). In this study, we conducted a correlation between temperature, humidity, rainfall, and the confirmed cases of malaria. The findings from this study will aid malaria programmers and Lagos state Government to implement effective control measures.

Study Area
Lagos state is in the Southwestern part of Nigeria, at the bay of Benin. The map of Nigeria showing Lagos state (Figure 1). It lies on longitude 3.406448 and latitude 6.465422. Ogun state borders it on the Northeast(10,11). Lagos state shares a border to the west with the Benin Republic(11).
Lagos state has two major climatic seasons: dry season within the periods of November to March and wet season within the periods of April to October. The mean temperature range is between 17 and 36 0 C. The mean annual rainfall is about 1300-1800 mm, even up to 25, mm in the coastal area. Malaria transmission is endemic in the state. Plasmodium falciparum is the main malaria species(2,12). Lagos state is one of the most populated states in the country(3,11,13) It has a population of 9,013,534, according to the 2006 National census(11). Lagos state has 20 LGA and 37 Local Council Development Areas as administrative units in the state(11). There are 256 functional public primary health centers, 26 public secondary facilities, 3 public tertiary facilities, and 2886 private facilities (14).

Study Design
The study was a secondary data analysis of meteorological variables and malaria cases. The data include the period between January 2015 and April 2018. We accessed data on the malaria cases from the District Health Information system-2 (DHIS-2). The database is open-source software used for reporting routinely collected facility data (15). The DHIS-2 platform displays data elements into the state, local government areas (LGA), and wards (15). The rollback malaria managers in the 20 LGA of the states report all malaria cases through the malaria surveillance system. They combine malaria reports into indicators and display them through tables on the Dhis-2 platform. We collected meteorological data online from the free access, "weather underground.com". The database is a meteorological data sharing service system (MDSSS). The MDSSS forecasts and provides free, real-time online weather information from four meteorological monitoring stations in the state (16). These are Murtala Muhammed International Airport, Ikeja, Lagos yacht club, Lagos Island, Lekki, and Apapa (16). The distributions of the location represent Lagos state.

Data Abstraction
We extracted meteorological variables, monthly rainfall (mm), relative humidity (%), and temperature (°C). Malaria data included records of confirmed malaria cases attended to at the primary, secondary and tertiary centers compiled at the LGA level, and imputed to the Dhis-2 database monthly. Rollback malaria managers validate the malaria reports by routine data quality assurance across all the facilities in the state. Health workers confirm malaria cases either by rapid diagnostic tests or microscopically. We analyzed data using the Statistical Package for Social Sciences (SPSS version 25, Inc., Chicago, IL, USA).

Data Analysis
The monthly malaria cases were the dependent variable and meteorological variables (monthly temperature, relative humidity, and the mean rainfall) the independent variables. We estimated a monthly mean value. We represented the frequency distribution of the mean values of temperature, humidity, rainfall, and malaria cases on the line graph. We performed four rounds of analyses to show the relationship between mean temperature (mt), mean humidity (mh), mean rainfall (mr), and malaria cases at a 5% level of significance as shown in the equations below.

Results
The  Figure 4).

Spearman Correlation
The mean temperature mean humidity and mean rainfall positively correlate with malaria cases over the study period. The mean humidity most correlated positively to the malaria incidence (r1 = 0.65, p<0.02) over the study period (Table 1).

Trend Series
The trend series by deseasonalization of data illustrates a positive trend with an increase in the cases of malaria and vice versa. The higher the size of the quarterly variable values, the more the expected cases of malaria. The data show that the highest reported cases of malaria occurred in the first quarter (January to March) and the lowest cases in the second quarter (April to June).
Similar trends occurred in the first and third quarters (Table 3).

Discussion
The meteorological factor affects the seasonal and temporal patterns of infectious agents borne by a vector (17)(18)(19). Malaria thrives in tropics where the weather is hot and wet and makes inhabitants of that area prone to the disease. Meteorological factors affect the number and distribution of malaria according to the Intergovernmental Panel on Climate Change Fourth Assessment Report (19). Several other factors (such as changes in land use, population density, and human behavior) influence the distribution of disease vectors and the extent of infection (20).
The malaria incidence fluctuates over the months depending on the most favorable climatic conditions. This creates a pattern of infection transmission (12). The female Anopheles mosquito, the malaria parasite carrier, depends on a favorable climatic condition to transmit the malaria parasite to humans (21). Climate influences the three major aspects of the female Anopheles mosquito life cycle. It determines the seasonality and pattern of malaria infection. Thus, for Anopheles mosquitoes to thrive, for about 9-12 days there must be adequate rainfall for the breeding of the vector (20,21). This study showed that increased humidity and rainfall poses a risk to malaria infection Other studies has shown also that increased precipitation predisposes to malaria infection (22,23). It has been demonstrated that rainfall and flooding resulted in about 30% more risk of an individual having a positive result of a malaria diagnostic test in areas affected by flooding (20,24). Research has shown that common water bodies, in the wet months and early dry months; aid the development of the parasite consequently affects its transmission in urban areas.
Rainfall creates mosquito-breeding sites (20). Climatic variables like rainfall have also affected temperature, and humidity as rainfall increases accordingly humidity increases (23,25).
The existence of a malaria parasite within the adult mosquito requires temperatures of ≥ 20 0 C for Plasmodium falciparum ≥ 15 0 C for Plasmodium vivax (26). Warmer temperature shortens the extrinsic cycle of the parasite that occurs in the female Anopheles mosquito. This increases the chances of the parasite surviving in the mosquito. This study observed that malaria cases increase with temperature increase.
In contrast, another study describes a lower temperature may favor the survival of the parasite in the mosquito which can translate to an increase in incidence (27). shown malaria vector and predictability of malaria infection to be socioeconomic development, drug resistance, and immunity of the human host (29). Other studies have shown that higher malaria infection distribution does not only result from weather change but there are other key drivers of malaria infection such as economic situations, climate change, land conversion deforestation host movement, and demography (30).
As the World Health Organization (WHO) framework for malaria elimination stresses, most countries have diverse transmission intensity (31). Factors such as ecology, immunity, vector behavior, social factors, and health system characteristics influence both the diversity of transmission and the effectiveness of tools, intervention packages, and strategies in each locality (32,33). The WHO Framework goes further to encourage strategic planning and interventions appropriate for the diverse settings or strata within a country(2). This implies that the nature of malaria transmission in these strata will change as temperature, rainfall, humidity, and human response change(2). Countries not only need to adapt malaria activities to existing strata, but also be alert to changes in transmission and thus changes needed in strategies. Climate, vector habitat, and transmission changes affect vector control activities. The receptivity of an area (to vector control interventions) is not static but is affected by determinants such as environmental and climate factors. Case detection will become even more crucial as transmission drops and the success of elimination programs depends on identifying, tracking, and responding to remaining cases.
The data used is secondary data. The data is facility-based and there is the possibility of underreporting of malaria cases and self-treatment of malaria in the community. The accessibility of the facility in the community can affect facility data. Some confounding factors might have influenced the results of the study. These include the impact of malaria control activities and other related interventions. Preventive practices such as distribution and use of LLIN and larviciding may be responsible for the malaria pattern seen in this study. The accuracy of the Malaria surveillance system is a problem common to other low-income malaria-endemic regions.

Conclusion
This study describes a correlation of meteorological factors: temperature, humidity and rainfall, and malaria dynamics over a period. The measurement of the fluctuation of the climatic variables can predict Malaria incidence. Public health interventions can target months of likely high incidence. The State should direct resource allocations to the months with the highest malaria risk.
In planning for eliminating malaria in Lagos State, prevention measures, indoor house spraying, larviciding seasonal chemoprophylaxis, community distribution of LLIN should target the first and third quarter months.

List of Abbreviations
DHIS 2 District health information system-2.
MDSS Meteorological data sharing service system. mh mean relative humidity.
mr mean rainfall mt mean temperature.
WHO World Health Organization.

Ethics approval and consent to participate
As the study was based on secondary data, no human subjects were involved and hence were exempted from human subject ethical requirements.

Consent for publication
Available

Availability of data and materials
All data supporting the conclusions of this study are available upon reasonable request.