Highly heterogeneous residual malaria risk in western Thailand

Abstract

Over the past decades, the malaria burden in Thailand has substantially declined. Most infections now originate from the national border regions. In these areas, the prevalence of asymptomatic infections is still substantial and poses a challenge for the national malaria elimination program. To determine epidemiological parameters as well as risk factors for malaria infection in western Thailand, we carried out a cohort study in Kanchanaburi and Ratchaburi provinces on the Thailand-Myanmar border. Blood samples from 999 local participants were examined for malaria infection every 4 weeks between May 2013 and Jun 2014. Prevalence of Plasmodium falciparum and Plasmodium vivax was determined by quantitative PCR (qPCR) and showed a seasonal variation with values fluctuating from 1.7% to 4.2% for P. vivax and 0% to 1.3% for P. falciparum. Ninety percent of infections were asymptomatic. The annual molecular force of blood-stage infection (_molFOB) was estimated by microsatellite genotyping to be 0.24 new infections per person-year for P. vivax and 0.02 new infections per person-year for P. falciparum. The distribution of infections was heterogenous, that is, the vast majority of infections (>80%) were found in a small number of individuals (<8% of the study population) who tested positive at multiple timepoints. Significant risk factors were detected for P. vivax infections, including previous clinical malaria, occupation in agriculture and travel to Myanmar. In contrast, indoor residual spraying was associated with a protection from infection. These findings provide a recent landscape of malaria epidemiology and emphasize the importance of novel strategies to target asymptomatic and imported infections.

Keywords: Epidemiology; Force of infection; Incidence; Malaria; Plasmodium falciparum; Plasmodium vivax; Prevalence; Thailand.

Graphical abstract

Fig. 1

Map of the two study sites in Bong Ti, Kanchanaburi and Suan Phueng, Ratchaburi on the Thailand-Myanmar border. Study households are shown (●) as well as households in which infections with Plasmodium falciparum or Plasmodium vivax were detected over the course of the study (★).

Fig. 2

Prevalence and molecular force of blood-stage infection (_molFOB) of Plasmodium vivax (A) and Plasmodium falciparum (B) at each study site by month of study visit. Error bars (prevalence) represent the upper limit of the 95% confidence interval; shaded area (_molFOB) represents S.E.M. Note that _molFOB could not be determined for visits one and two as it depends on the observations from two previous study visits.

Fig. 3

Distributions of the molecular force of blood-stage infection (_molFOB). The proportion of study participants at each level of _molFOB is shown for Plasmodium vivax (A) and Plasmodium falciparum (B). Red: normalized distributions of observed _molFOB in the study participants. Grey: Poisson distributions computed based on the mean _molFOB. The Poisson distributions are overlaid on top of the observed distributions.

Fig. 4

The molecular force of blood-stage infection (_molFOB) as a function of age in the study population. (Aa) The average Plasmodium vivax_molFOB in each age group. (Ab) The proportion of individuals with P. vivax_molFOB > 0 in each age group. (Ba) The average Plasmodium falciparum_molFOB in each age group. (Bb) The proportion of individuals with P. falciparum_molFOB > 0 in each age group. Error bars indicate the S.E.M. (Aa, Ba) or the upper limit of the 95% confidence interval (Ab, Bb).