Dengue is a serious mosquito borne disease common in tropical and sub-tropical countries including Malaysia. There is at present a lack of specific treatment and an effective tetravalent vaccine against dengue. The control of dengue depends solely on the suppression of the two most important vectors namely, Aedes aegypti and Ae albopictus. Despite intensive and extensive control efforts by health agencies, the disease continues to spread. This paper updates various innovations on control of dengue vectors. Gene-based sterile insect technique using the RIDL technology for both Aedes aegypti & Ae albopictus control has now been actively researched and field trials are pursued to evaluate the effectiveness of the technology. The release of Wolbachia-infected Ae aegypti is another dengue control innovation. The infected mosquito cannot support development of dengue virus and has shorter life span. Other innovations include: auto-dissemination of insect control agents using ovitrap, autocidal adult and larva trap, outdoor residual spraying, insecticidal paint and biocontrol agent. In other innovation, outbreak prediction capability is enhanced by developing model based on environmental data and analysis utilising neural network.

Early detection of a dengue outbreak is an important first step towards implementing effective dengue interventions resulting in reduced mortality and morbidity. A dengue mathematical model would be useful for the prediction of an outbreak and evaluation of control measures. However, such a model must be carefully parameterized and validated with epidemiological, ecological and entomological data. A field study was conducted to collect and analyse various parameters to model dengue transmission and outbreak. Dengueprone areas in Kuala Lumpur, Pahang, Kedah and Johor were chosen for this study. Ovitraps were placed outdoor and used to determine the effects of meteorological parameters on vector breeding. Vector population in each area was monitored weekly for 87 weeks. Weather stations, consisting of a temperature and relative humidity data logger and an automated rain gauge, were installed at key locations in each study site. Correlation and Autoregressive Distributed Lag (ADL) model were used to study the relationship among the variables. Previous week rainfall plays a significant role in increasing the mosquito population, followed by maximum humidity and temperature. The secondary data of rainfall, temperature and humidity provided by the meteorological department showed an insignificant relationship with the mosquito population compared to the primary data recorded by the researchers. A well fit model was obtained for each locality to be used as a predictive model to foretell possible outbreak.