Public health issues associated with Samonella enterica in rural world remain alarming Widal test commonly used may not bring out the right diagnostic and cultural technics which is an alternative is expensive. In the 19^th century, it emerged as an infectious disease and was known to affect only humans. Transmission is basically bay fecal route and water [1] . Typhoid fever (TF) continues to be a substantial public health problem in developing countries likely due to rapid population growth, increased urbanization, and limited safe water, infrastructure, and health systems [2] . Salmonella typhi are restricted exclusively to human hosts and are associated with systemic infection, and prolonged fever and may result in an asymptomatic carrier state. Following resolution of infection, a small number of persons, called carriers, continue to carry the bacteria. Around 2% - 5% of those who contract typhoid fever become chronic carriers, as the bacteria persists in the biliary tract after symptoms have resolved [3] Sero-epidemiological surveys have also revealed that some persons in the population have non-immunizing antibodies present in them irrespective of previous vaccination status, and will be seropositive, which is a limitation of employing sero-diagnostic rapid diagnostic tests (RDTs) in surveillance [4] . Globally in 2017, there were an estimated 14.3 million cases of typhoid and paratyphoid fever in 2017, resulting in 135,900 deaths, 15.8% of which were in Sub-Saharan Africa [5] . Serological diagnostic methods still suffer from limitations of sensitivity and specificity emphasizing the need for a quick and reliable serologic test for acute infection of typhoid fever as an alternative to the old Widal test and as a complement to blood culture [6] . Diagnostic methods for typhoid fever currently used in health facilities in Cameroon include the Widal test, which is cheaper, easy to perform and does not require complicated expertise [7] . A disturbing issue with the Widal test in rural set is that the technicians are not available and when available they are poorly trained and at times lack the concentration to appreciate the reaction of the test worst of all nothing is done to confirm if the Widal test its positive or negative which then lead to an inappropriate antimicrobial treatment of patients without the typhoid fever due to misdiagnosis. Culture of the causative organism is considered an effective diagnostic procedure in suspected typhoid fever [8] but they are expensive techniques and some bacterial culture facilities are often unavailable and also it requires laboratory equipment and technical training. Widal test alone for the detection of typhoid fever without confirmation may lead to unnecessary development of antimicrobial resistance to antibiotics which is a big problem worldwide today. The aim of this study was to compare the Widal test against stool culture in typhoid-suspected cases and to evaluate the agreement between test methods.

With advancing technology and research, the world continues to quest for better ways of fighting infectious disease and antimicrobial resistance, correct, rapid and accurate diagnosis methods are needed to achieve this goal. In many parts of the tropics like Cameroon in particular, febrile presenting diseases such as typhoid fever are among the diseases which are commonly diagnosed [10] . Therefore, a study was carried out to evaluate the diagnostic accuracy of the commonly performed Widal test and stool culture in some local health centers in the Adamawa region of Cameroon.

The actual prevalence of typhoid fever was found to vary from one area to another in the different health districts. With Widal test the prevalence in Vina division was 43.16% while stool culture gave a prevalence of 66.24% statistically there was a significance difference (p < 0.05) between the Widal test and stool culture in this health establishment on the contrary to other works published by Abdulhai [11] in Kano Nigeria. Meanwhile, in other health districts, the prevalence of Widal test and stool culture were equally high but statistically, there was no significance difference (Mayo Banyo Widal test = 44.62%), stool culture = 58.46%), Djerem Widal test = 31.88%, stool culture = 46.38%). The prevalence of Widal test and stool culture were higher than those obtained by Wam [8] who reported a prevalence of 39.3% in patients using the stool culture method and 57.1% using the Widal test method in Akum health district. High prevalence values obtained in this exercise could be due to the quality of the reagents used and continuous training in laboratory techniques offered to health workers to improve the quality of work in the quest for better ways to diagnose infectious diseases in Cameroon and Adamawa region in particular.

Another contributing factor to the high prevalence of bacteria in the stool culture could be due to overcrowding with poor access to clean water and sanitation as reported by Tangwa et al. [12] . Faecal matter may gain access to the water through runoff water or through sewage from sewers and pit latrines. This could have been a major contributing factor to the high prevalence considering that a number of patients come from rural and riverine areas where facilities for sewage disposal are not controlled The consequences of false positive results include miss-use of antibiotics, likely danger of increasing the antibiotic resistance, increased cost of treatment due increased hospital stay for inpatients and missing of fatal disease of febrile illnesses.

This study revealed that Salmonella infection cuts across all age groups. The age groups most susceptible in the three study areas range between 0 to 20 and 20 to 40 years old. These results differ from the study carried out by Ramyil [13] colleagues which showed that those between 24 - 29 years were more susceptible to typhoid fever’ The age range of 1 to 40 years is considered tender and active age. In this light, both children and adults can get typhoid fever through ingestion of contaminated food and water. People living in overcrowded areas with poor hygienic conditions and lack of access to safe drinking water are prone to infection by typhoid fever. In this study age was not significantly associated with Salmonella infection, on the contrary, a study by Lunguya [14] reported that older age and longer duration of fever were indication of typhoid fever.

The sensitivity and specificity of Widal test indicate the probability of a true typhoid patient to be positive by Widal test. The sensitivity, specificity, PPV and NPV of Widal test in this study were not the same in all the study areas, these values are seen in Table 5 . Sensitivity was 51.61% in the Vina division, 60.53% in the Mayo Banyo division and 53.18% in the Djerem division at 95%CL respectively. Similar results were observed in a study conducted in Vietnam (64%) [15] .

The Widal slide agglutination true-negative results against stool culture were 73.42% in Vina division, 77.75% in the Mayo Banyo division and 83.73% at the Djerem division at 95% CI. This has a major impact on provision of effective patient management and reduces trust of patients in health services. As it designates the ability of an individual who does not have a disease to be tested negative. Other studies conducted in Nigeria on adults in hospitals showed that Widal test had low specificity [16] .

This variation in negative predictive values may be due to geographical variations. The Widal agglutination varies with geographical location based on the indemnity of enteric fever, prevalence of no typhoid salmonella infection, and other infections which cross-react with salmonella antigen [17] . Low sensitivity may be associated equally with the sample collection time stool is not sensitive in early infection [18] .

The positive predictive values in this study varied from 79.21% in the health districts in Vina to 79.31% in the health districts of the Mayo Banyo and finally to 73.91% in the health district of Djerem division. The positive predictive value is the ratio of participants truly diagnosed as positive to those who show positive test results. This characteristic can predict how likely it is for a participant to be truly positive in case of positive test results. It is the most important measure of a clinical diagnostic method since it represents the proportion of patients with positive test results that are correctly diagnosed [19] The negative predictive values in the area of study were all average and ranged from 43.31% in the Vina division to 58.33% in the Mayo Banyo division and finally to 67.39% in the Djerem division. This indicates that a negative Widal test result has an average predictive value for the absence of the disease. Though the PPV and NPV are low it does not concord with results recorded by Wam [18] with PPV of 28.13% and NPV of 6.44% at the Akum health district in Cameroon.

The assessment of agreement between two different diagnostic methods can also indicate how the test methods are close to each other. Statistically, the kappa values varied from 0.201 in the Vina division to 0.37 in the Mayo Banyo division and finally to 0.37 in the Djerem division. Theses kappa values demonstrated a fair agreement between Widal test and the stool culture methods These results concord with those obtained by Gemechu in Ethiopia [18] . Most people prefer Widal test to stool culture under the presumption that at early infection stool culture is positive in only 30 to 40% of the patients and that stool culture may bring about false positive results as a result of contaminations.

In this study, the prevalence of salmonella in stool culture results was relatively higher than the prevalence in the Widal test. Hence, the Widal test as the only laboratory test for the diagnosis of typhoid fever will produce a wrong diagnosis. The Widal test had a fair agreement with the stool culture. This means the Widal test should not be used alone but in combination with stool culture. Since the study area was vast other studies should be carried out to determine which of the serotypes are found in each zone

This research was financed with money contributed by the authors most especially from the Ph.D. student who is the first author and second corresponding author. No contribution came in from any other individuals or organizations.

Conceptualization: Ngueguim Lucie; Methodology: Ngueguim Lucie, Tangwa Viban and Ngu Ngwa Victor; Resources: Ngueguim Lucie and Ngueguem Romain; Writing original draft: Ngueguim Lucie and Tangwa Viban; Validation; and supervision: Mamadou Abdoulmoumini and Ngu Ngwa Victor.

*1st corresponding author.

^#2nd corresponding author.