Benefit of targeted sampling for lymphatic filariasis surveillance in Samoa depends on antigen prevalence

Background

In Samoa, lymphatic filariasis (LF) remains endemic. Targeted sampling strategies based on locations of known infections could be more efficient than random sampling for locating infected individuals and hotspots, providing valuable information to develop more efficient and cost-effective interventions. However, the level of benefit may depend on the prevalence of the chosen indicator in the area being surveyed. This study aims to assess the efficiency of targeted versus random sampling for identifying LF antigen (Ag)- and microfilaria (Mf)-positive individuals in Samoa for varying background Ag prevalence levels.

Methodology

In 2023, six primary sampling units (PSUs) were surveyed using random and targeted sampling strategies. PSUs were selected based on Ag prevalence in 2019, including two low (3-5%), medium (6-7%) and high Ag prevalence (13-17%). The randomly selected group included residents aged ≥5 years in 15 houses per PSU. The targeted group included residents aged ≥5 years in up to eight households within 200 metres of a household where Ag-positive resident(s) were identified in 2019. Finger prick blood samples were tested for Ag and Ag-positive samples were examined for microfilaria (Mf).

Principal Findings

The targeted sampling strategy (n=400 people) identified more positives (57 Ag-positive, 23 Mf- positive) than the random sampling strategy (n=494, 39 Ag-positive, 16 Mf-positive), with an overall targeted:random sampled case ratio of 1.8 (95% CI 1.3-2.5) for Ag and 1.8 (95% CI 1.1-3.1) for Mf. Gain in efficiency was greatest in medium prevalence PSUs for both Ag-positives (ratio=2.4, 95% CI 1.3-5.2) and Mf-positives (ratio=2.6, 95% CI 0.9-12.8).

Conclusions

In Samoa, a targeted sampling strategy was more efficient for locating Ag-positive and Mf-positive individuals compared to random sampling, with the highest efficiency gain in medium Ag prevalence settings. The findings have design implications for LF surveillance in Samoa and other Pacific Island countries.

Author Summary

Effective surveillance activities are essential for achieving elimination targets for lymphatic filariasis (LF). Cost-effective surveillance strategies are needed to locate infections for targeted treatment of individuals or high-risk communities. The aim of this study was to investigate the efficiency of targeted sampling compared to random sampling for locating antigen (Ag)-positive and microfilaria (Mf)-positive individuals in Samoa and explore how differences in efficiency gains depend on the Ag prevalence in the village. In 2023, six villages with varying Ag prevalence in 2019 were surveyed using both random and targeted sampling strategies. For the random group, we selected 15 houses in each village and invited all household residents aged five years and over to participate. The targeted group consisted of residents in these same villages who lived within 200 metres of an Ag-positive participant from the 2019 survey. Finger prick blood samples were collected and tested for Ag and Mf.

Overall, significantly more cases were identified per person tested by using the targeted strategy. This gain was particularly evident in the medium prevalence villages. These results show that targeted surveys of households neighbouring the residence of an Ag-positive person may be more efficient than random sampling, particularly in medium Ag prevalence villages. The findings have specific design implications for future LF surveys in Samoa and other Pacific Island countries.