Determinants of Dengue Hemorrhagic Fever Incidence: Community Knowledge, Attitudes, and Practices from a Health Promotion Perspective

Abstract

Background: Gunungsitoli City remains a dengue-endemic area in North Sumatra Province. In 2022, Ilir Village reported the highest number of dengue hemorrhagic fever (DHF) cases within the Gunungsitoli District Primary Health Center catchment area, making it a priority for dengue prevention and control. Despite ongoing control efforts, evidence on the association between community knowledge, attitudes, and practices (KAP) and DHF occurrence in this setting remains limited, highlighting the need for context-specific evidence to inform prevention strategies. Methods: A surveillance-based unmatched case-control study was conducted among 99 participants, including 33 laboratory-confirmed DHF cases identified from routine dengue surveillance records at the Gunungsitoli Primary Health Center and 66 controls recruited from the same neighborhoods using a 1:2 case-control ratio. Controls originated from the same source population as the cases and had no documented history of DHF during the previous two years. Primary data on knowledge, attitudes, and preventive practices (KAP) were collected through structured interviews and household observations, while secondary data were obtained from routine dengue surveillance records. Associations between KAP and DHF were analyzed using chi-square tests and multivariable logistic regression with the Backward Likelihood Ratio (Backward LR) method. The results are presented as odds ratios (ORs), adjusted odds ratios (AORs), 95% confidence intervals (CIs), and p-values. Results: In the multivariable logistic regression analysis, only age and preventive practices remained independently associated with DHF. Respondents aged <15 years had significantly higher odds of DHF than those aged ?15 years (AOR = 3.480; 95% CI: 1.122–10.793; p = 0.031). Respondents with poor preventive practices had 4.756 times higher odds of DHF than those with good preventive practices (AOR = 4.756; 95% CI: 1.854–12.204; p = 0.001), representing the strongest independent factor associated with DHF. Although age, sex, knowledge, attitudes, and preventive practices were significantly associated with DHF in the bivariate analysis, only age and preventive practices remained significant after adjustment. Conclusions: Poor preventive practices and younger age (<15 years) remained independently associated with DHF. Dengue prevention programs should prioritize sustained household vector control through regular elimination of mosquito breeding sites, supported by community participation and targeted health promotion for households with children in endemic areas.