Abstract

Various entomological indicators and sampling techniques are used to monitor and evaluate the impact of many vector control interventions. A number of methods have been used in sampling mosquitoes for the purpose of estimating the entomological inoculation rate (EIR) and each is subject to some bias or shortcomings. It was the aim of this paper to critically evaluate the most common mosquito sampling techniques in relation to their reliability in the estimation of EIR. The techniques include man-landing, light trap, light trap/bednet combination and odour-baited traps. Although man-landing technique is the most reliable, it however, expose the catcher to mosquito-borne infections. On the other hand, light traps have been found to capture mosquitoes with higher sporozoite rates as compared to those from human bait catch thus leading to an overestimation of EIR. From an epidemiological point of view, the use of light-trap-bed net combination is an approach that is more meaningful than using light trap alone because, a light trap functions more efficiently when placed near the normal flight paths of mosquitoes such as inside huts or under the eaves. Unfortunately, it has been shown that estimates of EIR are influenced by trap position, hence affecting the number caught and the sporozoite rates. A variety of bednets have been used to sample mosquitoes attracted to man. Studies have shown that bednet traps normally catch fewer mosquitoes than do human baits outside them. Although the collections by indoor resting technique give a good estimate of the mean house density in a given area, they may not necessarily give a good estimate of EIR. Thus the development of improved sampling systems based on an improved understanding of host-oriented behaviour is needed. Moreover, there is need to standardise all the sampling techniques in use to enable us make valid comparisons between various studies done by different people and in different areas. In this article, the inherent limitations of conventional mosquito sampling techniques when used in estimating the EIR are discussed.