Coverage and beliefs about temephos application for control of dengue vectors and impact of a community-based prevention intervention: secondary analysis from the Camino Verde trial in Mexico

Abstract

Background

Temephos in domestic water containers remains a mainstay of Latin American government programmes for control of Aedes aegypti and associated illnesses, including dengue. There is little published evidence about coverage of routine temephos programmes. A cluster randomised controlled trial of community mobilisation in Mexico and Nicaragua reduced vector indices, dengue infection, and clinical dengue cases. Secondary analysis from the Mexican arm of the trial examined temephos coverage and beliefs, and the impact of the trial on these outcomes.

Methods

The trial impact survey in December 2012, in 10,491 households in 45 intervention and 45 control clusters, asked about visits from the temephos programme, retention of applied temephos, and views about temephos and mosquito control. Fieldworkers noted if temephos was present in water containers.

Results

Some 42.4% of rural and 20.7% of urban households reported no temephos programme visits within the last 12 months. Overall, 42.0% reported they had temephos placed in their water containers less than 3 months previously. Fieldworkers observed temephos in at least one container in 21.1% of households. Recent temephos application and observed temephos were both significantly more common in urban households, when other household variables were taken into account; in rural areas, smaller households were more likely to have temephos present.

Most households (74.4%) did not think bathing with water containing temephos carried any health risk. Half (51%) believed drinking or cooking with such water could be harmful and 17.6% were unsure.

Significantly fewer households in intervention sites (16.5%) than in control sites (26.0%) (Risk Difference − 0.095, 95% confidence interval − 0.182 to −0.009) had temephos observed in their water; more households in intervention clusters (41.8%) than in control clusters (31.6%) removed the applied temephos quickly. Although fewer households in intervention sites (82.7%) compared with control sites (86.7%) (RD -0.04, 95% CI -0.067 to −0.013) agreed temephos and fumigation was the best way to avoid mosquitoes, the proportion believing this remained very high.

Conclusion

Coverage with the government temephos programme was low, especially in rural areas. Despite an intervention encouraging non-chemical mosquito control, most households continued to believe that chemicals are the best control method.

Trial registration

ISRCTN:27581154.