PREVALENCE AND ASSOCIATED RISK FACTORS OF TYPHOID FEVER IN CHILDREN ATTENDING “DEO GRATIAS” HOSPITAL IN DOUALA, LITTORAL REGION

Typhoid fever is a communicable disease transmitted by the bacteria Salmonella typhi, related to serotype paratyphi A, B and C. The disease is of a significant health concern in most developing countries especially Cameroon. Objectives: The study aimed at determining the prevalence and associated risk factors of typhoid fever in children ( 0-18 years) attending the “Deo Gratias” hospital in Douala Method: A hospital base cross sectional study from August to September 2018 was carried out in patients’ age 0-18 years suffering from typhoid fever at the Deo gratias Catholic hospital. Widal slide agglutination test was the diagnostic test used. Positive tested patients were administered questionnaires to evaluate the level of knowledge, attitude and practice toward the disease, as well as their self-management abilities. Data obtained from respondents was analysed by descriptive statistics. One-way ANOVA and means comparison using Tukey’s test (α = 0.05) was performed to check whether the population of respondents differed significantly across risk factor practices. Results were finally presented on bar charts, tables and pie chart. Results: Typhoid fever was more prevalent in females (52.3%) than in males (47.7%), with a high proportion in the ages 5-9 years (38.6%). A significant difference was observed in population of respondents across risk factor practices. Conclusion: Water quality have a great impact on the burden of typhoid fever among children. The identification of risk factors associated to the disease is of great importance in the development of rational control strategies of the disease.


Introduction
Typhoid fever is an infection having as causative agent Salmonella typhi related to the serotype paratyphi A, B and C (Geoffrey et al., 2010). This bacteria is a significant cause of morbiditity and mortality especially in developing countries and exhibits multiple antibiotic resistance (Akinyemi et al., 2005). Studies by and Mike, (2008) shows that this disease is associated to low socio-economic status and poor hygiene, having humans as the only natural host of the infection since the bacteria grows best at 37 ℃ which corresponds to the human body temperature. Transmission of the disease is through faecal oral route from contaminated food or water (WHO, 2018). Major symptoms of the disease includes; malaise, fever, vomiting constipation, splenomegaly and hepatomegaly (Nsutebu et al., 2003). The disease can result to major complications such as internal haemorrhage and perforation (Evanson, 2008). In the absence of effective treatment, this disease has a fatality rate of about 10 to 30 % (Buckle et al., 2010). Typhoid fever is a threat to many tropical countries showing a worldwide estimate of about 212 million cases with 129,000 deaths yearly with children and young adults being the vulnerable groups (Steele et al., 2016).
Reports from the Cameroons' Public Health ministry shows a frequent diagnosis of typhoid fever in children in health facilities in Cameroon and has resulted in a public scare (Nsutebu et al., 2003). It is thus considered an endemic disease in Cameroon. One major challenge in the treatment of this disease in Cameroon is the high costs of its drugs. Control strategies to the disease is a possible way out to reduce the disease spread. However, absence of information associated to the risk factors of typhoid fever especially in children in Cameroon has made it not really possible to bring about effective control strategies to manage the disease. From the findings of this study, the knowledge of associated risk factors of typhoid fever will help to bring about rationale control strategies of the disease thus mitigating its spread.

Study design
A hospital based cross sectional study was conducted from August to September 2018 with the goal of determining the prevalence and associated risk factors of typhoid fever in children attending "Deo Gratias" hospital in Douala, Littoral region of Cameroon. The age 0-18years was considered as children. Patients who were tested positive for typhoid fever were administered structured questionnaires. For patients less than 12 years of age their parents or guardian were required to fill the questionnaire. Questions were based on demographics of patients and typhoid fever associated risk factors. Questions on risk factors were related to water sources and treatment practices.

Study area
The study site was the "Deo Gratias" hospital in Douala, Littoral region of Cameroon.
Cameroon is a country located in the central part of Africa. The country is comprised of ten regions. The Littoral region of Cameroon is the largest in size and the most populated of the all the ten regions that make up Cameroon with a population of about 2, 768 436 inhabitants (INS, 2017). Douala is the capital of the littoral region and also the economic capital of Cameroon. It is the most populated town in Cameroon (BUCREP, 2010). Water sanitation in Douala is poor which greatly contributes to water borne diseases such as typhoid and cholera (Ndjama et al., 2008)

Study participants and collection of samples
Participants of the study were patients of age between 0 to 18 years who tested positive for typhoid fever. Testing of typhoid disease was done with the use of blood samples. Blood specimens were collected into vacutainer tubes containing no preservative/additive (red cap tubes) and tests were performed using the Widal slide agglutination method. Structured questionnaire were also administered to positive tested patients to evaluate the level of knowledge, attitude and practice towards the prevention and control of the disease, as well as their self-management abilities.

Laboratory analysis
The widal test was used as the presumptive serological diagnostic test for typhoid fever. The Positive results were indicated by the appearance of a visible agglutination within a minute, formed due to the reaction occurring between antibodies present in the infected person's blood (serum) and the antigens specific for S. typhi and S. paratyphi.
Results were recorded as 1/40, 1/60, 1/80, 1/160 etc. depending on the concentration of the agglutination observed. Negative results were indicated by the absence of agglutination between the patient's antibodies in serum and specific Salmonella antigens. Negative results were noted as "non-reactive" (NR), indicating the absence of a reaction (agglutination).
A semi quantitative test was further performed on the patients' serum that showed visible agglutination in order to determine the specific salmonella antigen responsible for the agglutination.

Result Analysis
The data obtained from questionnaires by respondents was analysed by descriptive statistics.
The data was entered in a spread sheet, Microsoft Excel and normality determined. One-way ANOVA and means comparison using Tukey's test (α = 0.05) was performed to check whether the population of respondents differed significantly with respect to risk factors tested. Results were finally presented on bar charts, tables and pie chart.

Research Ethics
Prior to the sample collection, verbal and written details of the study was provided in both English and French. Written informed consent was obtained from all the participants or their guardians which was approved by the hospital management.

Age and Gender
All patients who were confirmed positive for typhoid took part in the study during that period.

Clinical presentation of participants
Common symptoms shown by patients who participated in the study included fever, fatigue, headache and anorexia. Amongst the symptoms, most of the patients presented with fever (77.2 %) having temperatures ≥ 37.5ºC (figure 2). Fatigue was also common in the patients. Some already taken medications to reduce fever. Laboratory analysis showed that all the typhoid fever cases detected were due to S. paratyphi A.

Source of Drinking Water
Sources of drinking water identified by patients included pipe borne, river, stream and wells.  A one way ANOVA (Table 1) carried out on the data obtained from drinking water sources showed a significant difference (P ˂0.05) between the mean population of respondents' on the sources. A majority of the respondents used pipe borne as the major source of drinking water (23.98 ±3.20). There was no significant difference in the population of respondents whose source of drinking water was wells, river, streams and other sources.

Household Water Treatments Method used
Household water treatment methods outlined in the questionnaire included; boiling of water, filtering of water using purchased water filters and use of cotton wool as local household filters.
A significant difference was recorded among participants on use of treatment methods.
Majority of the respondents did not use any treatment method on water before drinking. Others used either boiling or filtering of the water as their water treatment technique.

Number of Members in Households
Patients who participated in the study lived in household with size ranging from one to eight members. A higher proportion of patients was obtained in households of size between of 3 to 5 members (66.7 %) Table 3. More than 8 0

Socioeconomic status index
Socioeconomic status index was considered based on monthly income of parents or guardians in households. The status was categorized as follows; high socio economic status index (>150,000 CFA), medium socioeconomic status index (between 100,000 CFA to 150,000 CFA) and low socioeconomic status index (˂ 100,000 CFA). A greater proportion of the patients (50 %) came from families with medium socioeconomic status index .

DISCUSSION
Based on our knowledge, this is the first study done on the association of risk factors to typhoid fever in this part of Cameroon. The gender distribution of typhoid disease in this study was 47.7% for males and 52.3% for females, suggesting that typhoid fever was more prevalent in females than in males among the age group in that locality. Similar research done by Butler et al. (1991) and Khan et al. (1999) in Bangladesh and South Africa showed that typhoid fever correlated with gender and case fatality is higher in females compared to males. A greater proportion of positive cases was detected among children with age range 5 to 9 (38.8 %) while a lesser proportion of patients was found in the age group below 5years (18.2 %). One reason for the high prevalence observed in the age group 5 to 9 is the underdeveloped immune system in growing children, this makes them more vulnerable to this enteric pathogen. A low prevalence noted in children less than 5 years of age may probably be due to their controlled diet and drinking water at these tender ages by their parents.
As regards socioeconomic status index, high income category had a lesser prevalence of typhoid (15.9 %) relative to lower income category (34.1 %) and middle income category (50 %). Similar studies done by Vollaard, (2004) show that the prevalence of typhoid infection was higher among lower income category households. Low income category household have high tendency of purchasing and eating cooked food from street vendors which predisposes them to typhoid infection. Street vendors have limited facilities for storing food and cleaning of dishes.
This poor hygiene practice is a vehicle for disease transmission. Furthermore low income category household practice poor household hygiene due to lack of means of available portable water connected to their houses. Ram et al. (2007) also identified socioeconomic status as a significant risk factor associated in the occurrence of typhoid fever.
Patients who took part in the study lived in household of varied sizes. Research indicates that household contact is a major risk factor associated to the spread of typhoid infection. Vollaard, (2004) found that the prevalence of typhoid was higher in households containing more than 6 members. Crowding was seen to be a risk factor associated with typhoid fever among households.
Most epidemiological studies have related the risk factors to typhoid fever of being waterborne or foodborne (Swaddiwudhipong et al., 2001). Findings obtained from the data showed a With respect to household water treatment methods, a significant difference (P ˂0.05) was observed among the population of respondents. Some respondents did not use any household treatment method for water. This could contribute greatly to the prevalence of typhoid fever.
Studies carried out by Ram, (2007) in Bangladesh demonstrated that drinking of unboiled water at home was a major risk factor in the occurrence of typhoid fever. Boiling of water in clean containers before drinking could reduce the risk of typhoid fever. This is due to the fact that the Salmonella typhi bacteria grows best at a temperature of 37 ℃ thus very high temperatures kills the bacteria. Boiling, the use of ceramic filters, bleach addition and solar disinfection has been household water treatment interventions introduced by the WHO (Farooqui et al., 2009)

LIMITATIONS
One possible limitation of this study was the limited number of participants which could greatly affect the statistical power of the study. Responses provided in the questionnaire for age group below 12 years was provided by parents and guardians which could introduce recall bias as regards the study.

CONCLUSION
The results from the study have a lot of significance to health experts. Firstly, it highlights improvement of sanitation and hygiene as the most effective way to prevent the spread of the disease especially in children. Nonetheless, our findings also highlight the need for more sensitisation of the public concerning the mechanism of transmission and effective control or preventive methods of the disease.

USED ABBREVIATIONS
WHO; World Health Organisation, UNICEF; United Nations International Children Emergency Fund, ANOVA; Analysis of variance

CONFLICT OF INTEREST
The authors declare that there is no conflict of interest regarding the publication of this article and there have been no significant financial support from anywhere that might have influenced its outcome.

ACKNOWLEDGEMENTS
The authors are thankful for the support provided by the administration of the "Deo Gracias" Catholic hosptital for making it possible for the research to be carried. Many thanks go to the patients or parents and guardians of patients who assisted responding to the questionnaires.