Lot quality assurance sampling techniques in health surveys in developing countries: advantages and current constraints

Abstract

Traditional survey methods, which are generally costly and time-consuming, usually provide information at the regional or national level only. The utilization of lot quality assurance sampling (LQAS) methodology, developed in industry for quality control, makes it possible to use small sample sizes when conducting surveys in small geographical or population-based areas (lots). This article describes the practical use of LQAS for conducting health surveys to monitor health programmes in developing countries. Following a brief description of the method, the article explains how to build a sample frame and conduct the sampling to apply LQAS under field conditions. A detailed description of the procedure for selecting a sampling unit to monitor the health programme and a sample size is given. The sampling schemes utilizing LQAS applicable to health surveys, such as simple- and double-sampling schemes, are discussed. The interpretation of the survey results and the planning of subsequent rounds of LQAS surveys are also discussed. When describing the applicability of LQAS in health surveys in developing countries, the article considers current limitations for its use by health planners in charge of health programmes, and suggests ways to overcome these limitations through future research. It is hoped that with increasing attention being given to industrial sampling plans in general, and LQAS in particular, their utilization to monitor health programmes will provide health planners in developing countries with powerful techniques to help them achieve their health programme targets.

PIP: Costly, time-consuming, traditional survey methods usually provide information only at national or regional levels. Information from the health center and community levels is, however, also of interest particularly in managing and directing supervisory activities. An industrial method is described with practical applications for conducting health surveys to monitor health programs in developing countries. This lot quality assurance sampling (LQAS) methodology was developed in industry for quality control, and allows the use of small sample sizes when surveying small geographical or population-based areas. The paper describes the method, explains how to build a sample frame, and how to conduct the sampling necessary for field application of LQAS. Sampling unit selection for health program monitoring is described in detail. Simple- and double-sampling schemes are discussed, as well as interpretation of survey results and the planning of subsequent rounds. Constraints limiting use by health planners are explored with suggestions provided on modes of overcoming obstacles through future research.