In this study, people who had never heard about dengue fever were found to be more at risk of getting dengue infection, with an adjusted OR of 2.8 (95 % CI: 1.24–6.53). This could be explained by a lack of knowledge regarding susceptibility to the disease and the appropriate protective measures. This lack of knowledge is directly exposing participants to a greater risk of contracting the disease compared with others who are well aware about dengue disease dynamics and take preventive measures [
17]. Population knowledge is a vital social determinant of dengue disease epidemiology because community participation is believed to be a key element for current recommended dengue control and prevention strategies worldwide [
18]. Household density, which refers to the number of people per room, was found to be a statistically significant predictor of dengue sero-prevalence amongst the study population, with an odds ratio of 2.1 (95 % CI: 1.06–4.09). This indicates that residents of a household with a density of greater than 3 people per room are more likely to get dengue infection compared with those who live in less crowded accommodation. This finding supports existing beliefs regarding dengue in different parts of the world-that human host density is considered one of the important social determinants of dengue infections. Furthermore, small household size is being confounded by overall human density in other studies [
19,
20]. This could be attributed to increased exposure of susceptible people to the dengue vector in overcrowded households. The responsible vector, the Aedes mosquito, feeds on multiple humans per day, facilitating transmission of the virus in an efficient manner [
11]. In addition, household density itself is always associated with poverty and low socioeconomic status, particularly in developing countries, and this is another social predictor for dengue infections [
19]. Age of participants in the current study was not statistically associated with dengue fever seropositivity (OR 1.17 (95 % CI: 0.55–2.5)). This is in contrast to a published study that found a higher prevalence of dengue seropositivity with increasing age, which may be due to the increased likelihood of being exposed to a mosquito bite and becoming dengue seropositive over a life span [
19]. In this study, sex did not show a significant association with dengue infection (OR 1.55 (95 % CI: 0.74–3.25)). A similar study, conducted in the southern part of Sudan, had concluded females were at higher risk of developing dengue, which was explained by the fact that in such conservative communities, females usually spend the majority of their time at home and the dengue vector breeds inside homes [
19,
21]. Any association between dengue susceptibility and sex is debatable, as there are contradictions between the available studies to date [
22,
23].
Kassala is an urban area located on the highway connecting Khartoum, the capital of Sudan, with the main sea harbor of the country, Port-Sudan city on the Red Sea coast with large population movements and social connections between Kassala and Red Sea state. Families from Port-Sudan usually spend the hot humid months of summer in Kassala taking advantage of more pleasant weather during that season. Port-Sudan has been known as an endemic area for dengue since the 1980s, with repeated dengue outbreaks in recent years particularly since 2005 onward [
5,
6,
24‐
27]. It was assumed that dengue disease was imported from Port-Sudan to Kassala; however, in this study, travel of a participant or one of his household members to Red Sea state in the last 6 months was not statistically significantly associated with dengue prevalence in Kassala. Movement of dengue viremic patients is believed to facilitate introduction of new dengue serotypes to other areas and sustain the circle of transmission and propagation [
19,
28]. Having a guest from Red Sea, storing water at home, the absence of window screens, the presence of Aedes aegypti pupa and adult mosquitos in the house or a history of yellow fever vaccination were all found not to be statistically associated with dengue infection in this study.
Although this is the first attempt to study factors associated with dengue sero-conversion in Kassala, the current study has its limitations, including the fact that it was carried out among community participants regardless of their symptomatic state; this makes it difficult to differentiate between primary and secondary dengue infections. Sometimes IgG levels wane below the IgG ELISA cut-off and this leads to missing cases (both primary and secondary) several years following exposure. Monotypic and polytypic plaque reduction neutralization tests (PRNT) responses could be assessed in the future for more accurate estimation of overall prevalence. Moreover, there is a possibility of cross reaction with other favivirus antibodies that may be circulating in the study area, particularly Yellow fever and Zika viruses that have been documented in other parts of the country, although there are no specific studies conducted in Kassala to date.