Asymptomatic Malaria And Glycemic Control Among Type 2 Diabetes Mellitus Patients In A Rural Tertiary Health Facility In Ido- Ekiti, Southwestern Nigeria- A Cross Sectional Study

Azeez Oyemomi Ibrahim (  ibrahimazeez1434@gmail.com ) federal teaching hospital, ido-Ekiti, Ekiti State Nigeria Tosin Anthony Agbesanwa Ekiti State University College of Medicine Segun Mathew Agboola Afe Babalola University College of Medicine and Health Sciences Olabode Muftau Shabi Federal Teaching hospital, ido-Ekiti Adewumi Oluwaserimi Ajetunmobi Federal Teaching hospital, Ido-Ekiti Waheed Olalekan Ismail Obafemi Awolowo University Teaching Hospital, Ile-Ife

Ironically, report by the International Diabetes Federation (IDF) states that, the incidence of T2DM is increasing globally and by the year 2030, about 80% of T2DM patients will be living in low predisposition. [4], [5] More worrisome than the absolute number of people with T2DM in Nigeria, is and middle income countries due to western lifestyles adaptation and genetic about 70-80% of the 4 million people who remain undiagnosed or untreated. [6] This relatively large number poses a signi cant host and potential reservoirs of Plasmodium falciparum which has a negative impact on glycemic control. [7], [8] The implications of the co-occurrence of T2DM with asymptomatic malaria in SSA will further deplete the poorly appropriated health budget of such region. The overall effect is that majority of patients present to health care facility with complications of advanced disease and its attendant high morbidity and mortality. [9] A study by Udoh et al in Nigeria showed that the prevalence of ASM among T2DM was 7.2% and 16.8% using light microscopy and Polymerace Chain Reaction (PCR) respectively. [8] The study further revealed that ASM was signi cantly associated with poor glycemic control (p < 0.04). [8] The study also suggested that in SSA, the probability of co-occurrence of T2DM and ASM in the same individual is high, making the need to study the association between ASM and glycemic control among T2DM long overdue. [8] To the best of our knowledge, there is currently paucity of data on the co-occurrence of ASM and T2DM among Nigeria patients, and where it exists, the study was done in the urban health care facility, and exclude the rural areas where malaria most often threatens the lives of people. [8] Furthermore, the impact of such co-occurrence on diabetes management outcomes such as glycemic control remains unknown.
In view of the predicted increased incidence of T2DM in SSA and the recent report of its co-occurrence with ASM. [7], [8] The present study therefore set out to determine the level of glycaemic control, and the prevalence of ASM parasitaemia among T2DM patients; and to investigate the in uence of sociodemographic characteristics and ASM parasitaemia on glycemic control among T2DM patients attending a rural tertiary health facility in Ido-Ekiti, Ekiti State, Southwestern Nigeria. Findings from this study would guide development of strategies to reduce the increasing T2DM trend and improve its case management.
It would also contribute to the global elimination of ASM through case detection and treatment, in addition to integrated vector control efforts in transmission-prone areas. [10] Research questions This research answered the following questions; 1. What is the level of glycemic control among T2DM respondents attending FTH, Ido-Ekiti?, 2. What is the prevalence of ASM among the study respondents?
3. What is the relationship between sociodemographic characteristics of the respondents and their glycemic control?
4. What is the relationship between ASM parasitaemia and glycemic control among the study respondents?
Two null hypothesis were tested at 0.05 level of signi cance. Ado Ekiti, its state capital. [11] The hospital offers secondary and tertiary care including specialist diabetic clinic to the people of its catchment area and its neigbouring states. The hospital was an accredited centre for postgraduate residency training in family medicine and other core clinical specialties. The people are mainly farmers and traders in the informal sector with a relatively small portion of the working population and retirees in the formal sector. In the study area, malaria transmission is perennial during the raining ( April -October) and dry ( November -March) seasons, with Plasmodium falciparum being the major causative agent. [11] T2DM is also a health challenge among adult in Ekiti State and case of poor glycemic control among patients attending FTH had been reported. [12] Study population The study population comprised T2DM patients who have been receiving treatment at FTH, Ido-Ekiti.

Inclusion criteria
Consented T2DM patients on follow up for at least six months, who were asymptomatic for malaria, and were ≥ 95% treatment adherence in the previous three months (i.e, not missing more than six doses during the period for twice-daily or thrice-daily regimen).

Exclusion criteria
Respondents who were critically ill or have a major psychiatric illness and could not follow the study protocol. Other exclusion criteria were obese patient with Body Mass Index ( BMI) ≥ 30Kg/m2 and hypertension ( Mean Blood Pressure of ≥ 130/80 mmHg for two measurements) and other diabetic related complications.
Sample size determination: Using the formula: n = Z 2 Pq/d 2 . [13] Where n is the minimum sample size, p is the 7.2%,being the prevalence of ASM among T2DM patients on oral hypoglycemic agents for at least three months with a con dence Interval of 95% at 5% margin of error. [7] Attrition of 10% was added to allow for unexpected data losses and drop-out, a total of 150 eligible respondents were enrolled.

Sampling technique
A systematic random sampling technique was used to recruit the 150 respondents who ful lled the inclusion criteria. The medical record of T2DM patients on follow up at Family Medicine clinic of FTH, Ido-Ekiti, showed that in the year 2018, an average of 10 T2DM patients were seen daily. This translated to 50 patients per week (Monday-Friday) and 800 (sample frame) patients over a period of four (16 weeks) months that the study was conducted. Using the formula, K = N/n, where k is the sample interval, N the sample frame (800) and n represented the minimum sample size 150. Therefore k was approximately 5. The rst respondent on each clinic day was chosen by simple random sampling and thereafter, every 5th respondent was chosen by systematic random sampling until the sample size of 150 was attained. A sticker was placed on each of the selected folders to avoid its re-sampling at the subsequent clinics.

Recruitment procedure
Respondents for the study were chosen using the inclusion criteria. Individual respondents were told about the methods and the objectives of the study and after which an informed consent was applied.
This was either by appending signature or thumb printing. The data was collected using a pretested semistructured interviewer administered questionnaire which was developed by the researchers. The face and content validity of the questionnaire was established by two Consultant Family Physicians working at FTH, Ido-Ekiti. Construct validity and reliability coe cient (α = 0.80) was established by using the Cronbach's alpha. The reliability coe cient 0.80 was within the acceptable range. The pretesting was conducted on 15 T2DM patients, also on follow up, and who were picked at random at the Family Medicine clinic of the Ekiti State University Teaching Hospital ( EKSUTH), and lasted for two days. The pretesting was to show the fairness of the questionnaire and its applicability. Modi cations of the questionnaire were done using the feedback from the pretest and validity assessement. With the assistance of the interpreter, translation of the questionnaire from English to local language was done for those who did not understand English Language while back translation to English language was later done to achieve response consistency. The questionnaire took about 10 minutes to complete.

Data Collection Instruments
The two instruments for data collection were the interviewer administered questionnaire and data collection form. The questionnaire sought information about the respondents' socio-demographic characteristics including the duration of their diabetes. The data was collected during the raining season over a period of 6 months (April-October 2019).
Clinical parameters of the respondents: 1. Blood pressure: With the aid of a reliable Accosson brand of Mercury Sphygmomanometer, blood pressure of the respondents was recorded twice after about 5 minutes apart and the Mean B.P of ≤ 130/80 mmHg was considered as controlled. The recording was done by a Senior Resident doctor.
2. Body Mass Index: All respondents were directed to be in light clothing and barefooted for their weight and height measurement using an electronic weighing and a portable height board respectively. The weight was measured to the nearest 0.1 kg and height to the nearest 0.1 cm. Normal BMI was de ned as 18.5-24.9 kg/m2, underweight as < 18.5 kg/m2, overweight as 25-29.9 kg/m2 and obese as ≥ 30.0 kg/m2.

Glycosylated Hemoglobin ( HbAIc):
After an overnight fast, 4mls of blood samples were collected through the venous route into uoride and ethylenediaminetetra-acetic acid (EDTA) immunochemically on DCA 2000; HbAIc auto analyzer using kits supplied by Boehringer Mannheim auto analyser. [14] Accuracy of the test result was ensured using commercially prepared standards and control samples. Glycemic level was grouped into good control if HbAIc is < 7.0% (i.e < 53 mmol/mol) and uncontrolled if HbAIc is > 7.0% (i.e > 53 mmol/mol). [15] 4. Microscopy: Two blood lms, one thin and one thick was made from the blood sample collected for malaria parasitaemia. The thick and thin smears were prepared on clean, dry microscope glass slides and was allowed to dry. The thin smear was xed in methanol and both smears were stained with 2% Giemsa BDH Laboratory supplies; Poole BH 15 ITD England. [16] The slide was then viewed under a microscope using oil immersion at 100x magni cation. Staining of the slide and parasite counting was done by a laboratory scientist of the hospital. The lm was earlier viewed for the presence of parasite. A positive result means the presence of one parasite per 100 power eld while a negative result means absence of parasite at 200 high power elds. The parasite density for the positive lm was determined by counting the number of parasite in relation to the standard number of leucocyte per microlitre. If fewer than 10 parasites were counted, then, more of leucocytes were counted up to 500 and the result was recorded as the number of parasite per 500 leucocytes.
Parasite density was then calculated using the formula; number of parasites over the number of leucocytes multiply by 8000 to give the number of parasites per microlitres of blood. A senior scientist was recruited to examine the slide for quality control, who also ensured that the authorized standard operating procedure for thin and thick blood lms microscopy was strictly adhered to. [16] De nition of terms: 1. T2DM Patient: Patient receiving documented treatment with anti-diabetes medication or having a glucose concentration of ≥ 7.0 mmol/l. [17] 2. Asymptomatic malaria (ASM): This was de ned as being free of malaria symptoms ( including body temperature ≤ 37.2 0 C) with presence of malaria parasitaemia as at the time of sample collection. [18] Data entry and analysis Data collected was checked, cleaned and entered into EPI Info Version 7.0 and exported to SPPSS version 22.0 for analysis. Computations, tabulations of proportions, percentage and other summary statistics were done. Chi-square analysis was used to determine the signi cance of association. Mean value was compared with students t-test. Chi-square (x2) and Fisher's exact test was used to compare the proportion of categorical variables. A p-value of < 0.05 was considered signi cant. Binary logistic regression was done to identify the variables that were predictors of glycemic control and such variables were exported to the multivariate logistic regression model. The strength of the association between independent and dependent variables was measured using odds ratio and 95% con dence Interval (CI) with signi cant level ( p-value < 0.05).
Ethical clearance, consideration and consent: Ethical approval for the study was received from the Ethics and Research Committee of FTH, Ido-Ekiti. The methods and objectives of the study were carefully explained to the participants individually, after which written informed consent was obtained either by appending signature or thumbprint for consent to participate in the study. For those participants who could not read or write, the questionnaire was translated from English language to their local language by an independent interpreter who served as their legal guardian while back translation to English language was later done to maintain response consistency. The respondents were free to refuse or disengage participation at any time without losing any bene t of care or favour to those that participated. Con dentiality and privacy were ensured throughout the study. The study was at no cost to the respondents Results: A total of one hundred and fty respondents were studied. The mean age of the respondents was 64.  (Table 3.0).
Twenty one (14.0%) of the respondents was positive for malaria parasitaemia. Therefore, the prevalence of ASM parasitaemia was 14%. Furthermore, the mean parasite density of the respondents was 1123.4±433.8 and ranged from 480-1123.4 (Table 4.0). ). There was also a statistically signi cant association between glycemic control and ASM parasitaemia (p<0.001) and parasite density (p<0.001), (Table 5).
Using multivariate binary logistic regression analysis, old age (OR, 6.334, 95% CI:      The mean age of 64.8 ± 11.1 years found in this study was closed to 60.67 ± 13.85 obtained from a study by Gabriel et al, [19] in the same study area and 60.96 ± 10 years recorded in another study in Abeokuta, Nigeria. [20] This nding was not surprising because most of the chronic medical conditions develop during the middle age period. In this study, the prevalence of respondents with poor glycemic control was 30.0%. This was similar to nding by Ajayi et al in the same study area, where he reported a prevalence of 32.5% and 29.3% of the study population using American Diabetes Association (ADA) and IDF-Europe criteria respectively. [12] However, our nding was less than report of a study by Ali et al conducted in the United States, where he found that the prevalence of poor glycemic control was 12.9%. [21] The difference with our nding could be due to difference in socio-economic characteristic of the respondents, diabetes management guideline and level of care. Furthermore, our nding was better than 47.0% reported by Danquah et al [7] and 48.7% reported by Yigazu and Colleaque, [22] based on HbAIc measurements. The difference between our study and the other studies could be due to variations in clinical characteristics of the respondents. For example, majority of the respondents in the other studies had obesity, dyslipidaemia and hypertension, whereas in our study, these co-founding factors have been excluded. [7], [22] Previous studies have established the relationship between obesity, dyslipidaemia, hypertension and poor glycemic control. [7], [22] In this study, 21 of the 150 afebrile T2DM respondents were found to harbor ASM parasitaemia. This nding is consistent with previous study conducted by Udoh et al, [8] and Danquah et al, [7] where they found that T2DM patients harbour Plasmodium falciparum asymptomatically. Therefore, these collective ndings suggest that adult T2DM patients are potential reservoirs of malaria. Although, previous studies have not been able to de ne the actual reason for the presence of P. falciparum among T2DM patients, however, the risk increases with increasing glucose concentration and is a sign of biologic plausibility. [7], Page 15/20 [8] Probable reason for such risk could be due to impaired defence against the liver leading to a decrease in T-cell mediated immunity. [1] Other suggested that P. falciparum grows faster in intensity with increased glucose availability as demonstrated invitro. [23] The implication of this nding is that the increasing trend in the incidence of T2DM in developing countries of SSA may therefore further worsen the already increased number of person at risk of malaria infection. Thus, the burden of co-occurrence of T2DM and malaria in developing countries warrants further investigation into the signi cance of this nding. Our nding of 14.0% prevalence of ASM is greater than 7.2% reported in a study by Udoh et al, [8] and 0.9% from a study by Danquah et al, [7] using microscopy to detect ASM parasitaemia. The difference from our study could be due to difference in the sample size, study design and the geographical location between the study areas.
In the current study, the mean parasite density of the respondents was 1123.4 ± 433.8/ul. This nding is less than 1580/ul reported from a study by Udoh et al, [8] but higher than 880/ul reported from a study by Danquah et al. [7] The difference with our nding could be due to the difference in the screening tools. This is because, in this study, we made use of microscopy to screen for ASM, whereas, studies by Udoh et al, [8] and Danquah and colleaque, [7] made use of Polymerase Chain Reaction ( PCR). However, studies have established that PCR is more sensitive in identifying P. falciparum reservoirs than microscopy. [7], [8] Therefore, using only microscopy to screen for ASM in this study, we may therefore have underestimated the burden of ASM among this set of patients.
In this study, using multivariate binary logistic regression analysis, ASM was found to be signi cantly associated with poor glycemic control. This nding is consistent with previous studies that have reported a statistically signi cant association between ASM parasitaemia and poor glycemic control among T2DM. [8], [24] The implication of this nding to the global malaria elimination target by 2030 implies that in malaria endemic settings like this study area, if T2DM patients have poor glycemic control despite adherence to medication among other factors, screening for malaria parasitaemia is necessary to completely rule out ASM. [8], [25] Report s from previous studies that investigated the in uence of socio-demographic characteristic of the respondents on glycemic control have been inconsistent. [20], [26], [27] In this study, old age, female gender and lack of formal education were associated with poor glycemic control. This is consistent with previous studies, [26] but inconsistent with other studies. [20], [27] while other studies found no relationship. [28][29][30] The reason for the poor glycemic control among the older patients in this study could be due to the fact that beta cell function begins to decline as age increases. This leads to decrease in insulin secretion and poor glucose utilization and therefore poor glycemic control. [27], [30] Also, female gender was linked to poor glycemic control in this study as compared to other studies. This could be due to the differences in the study design, gender-based roles in the study area, and differences in the health -seeking behavior among the genders. [20], [30] In view of this, interventional approaches should go beyond gender differences but patient -centre approach. [20] In this study, lack of formal education was a predictor of poor glycemic control. Previous studies have linked informal education to lack of knowledge of the disease, leading to poor understanding of the treatment regimen and decrease in treatment outcomes. 31,32 Therefore, continue health education and communication strategies by the stakeholders are needed as a measure to control the increasing trend of poor glycemic control among T2DM in the study area and Nigeria. Limitation: The study was limited by its cross sectional design and its sampling coverage. Therefore, nding from it cannot address the causal relationships between poor glycemic control and the factors found to be associated with it. Secondly, microscopy instead of PCR, was used to screen for malaria parasitaemia in this study. We may therefore have underestimated the burden of ASM among this set of patients.

Strength of the study:
It is noteworthy that some of the studies that have been conducted on glycemic control did not use HbAIc, which is an index of control. Hence, this study was conducted in Family Medicine clinic of a tertiary health facility in a rural area of Ekiti State using HbAIc as an index of control.
Bene ts of the study: 1. The respondents with ASM parasitaemia were treated with safe antimalaria drug to prevent further complications associated with the co-occurrence of T2DM and malaria infection.
2. Regular follow up visit and detailed health education were given to the respondents with poor glycemic control in order to improve their glycemic control level.
Implications for policy makers: The study revealed the co-existence of ASM and T2DM patients. Therefore, policy makers should canvass for more funding in the national health budget to cater for the integration of free ASM screening in the management of T2DM patients. Recommendations: 1. Large community based studies are required to clarify fully the relationship between ASM parasitaemia and glycemic control among T2DM patients.
2. Further research is needed to identify other factors negating good glycemic control among the study population. Conclusion: This study found that in Ido-Ekiti, Southwestern Nigeria, T2DM patients are potential reservoirs of ASM parasitaemia and that ASM parasitaemia, old age, female gender and illiteracy were negative predictors of glycemic control. Therefore, respondents with these negative predictors should be targeted with focus care and qualitative health education to improve their glycemic control. The authors also recommend integration of ASM screening into the management of T2DM patients with poor glycemic control despite treatment adherence. Ethical approval for the study was received from the Ethics and Research Committee of FTH, Ido-Ekiti. Informed consent was obtained from each respondents and participation was fully based on the willingness of the respondents. For those participants who could not read or write, the questionnaire was translated from English language to their local language by an independent interpreter who served as their legal guardian while back translation to English language was later done to maintain response consistency. The respondents were free to refuse or disengage participation at any time without losing any bene t of care or favour to those that participated. Con dentiality and privacy were ensured throughout the study. The study was at no cost to the respondents.

Consent for publication: Not Applicable
Availability of data and materials: The datasets for this study would be made available from the corresponding author on a reasonable request Declaration of con icts of interest: The authors declare that they have no con icts of interest.
Funding: The researcher received no speci c grant from any funding agency in the public, commercial or not-for-pro t sectors.
Authors' contribution: AOI-Conceptualization of the study, designed the study protocol, data acquisition and analysis and drafted the initial manuscript