Menu ENN Search
Language: English Français

SLEAC and SQUEAC

This question was posted the Coverage assessment forum area and has 1 replies. You can also reply via email – be sure to leave the subject unchanged.

» Post a reply

Anonymous 2352

Normal user

5 Dec 2013, 06:00

Between the SLEAC and SQUEAC, which one should be done first and can SLEAC be done in a SMALL Area, say a district with a population of around 30,000 and geographically the area is not too big.

Mark Myatt

Consultant Epideomiologist

Frequent user

5 Dec 2013, 10:50

One approach used in national coverage surveys is to do SLEAC everywhere and use the SLEAC result to decide where to do SQUEACs. This is covered in the SQUEAC / SLEAC Technical Reference on pages 114 - 115 and 182 - 189 and in this Field Exchange article which covers a lot of the same ground. If you have a SLEAC result before you do a SQUEAC then you can incorporate the SLEAC result into you prior for the SQUEAC stage III survey. You can also use the barriers data to inform SQUEAC stage I and stage II activities and the SLEAC class (or even the SLEAC data as a point estimate) to inform the prior mode.

Both SQUEAC and SLEAC can be done in small areas. For SLEAC there is a sample size issue. If the population is small and prevalence is low then you will have difficulty getting a sample size of n = 40 SAM cases. You can, in these small populations, use a smaller sample sizes without increasing error. See page 118 of the SQUEAC / SLEAC Technical Reference.

Doing SLEAC after SQUEAC is not the most efficient way of working. A better approach would be to use ongoing SQUEAC activities (e.g. look at patterns of admissions over time, look at routine program monitoring statistics) and the posterior or the previous SQUEAC stage III survey to create a new prior for a new SQUEAC stage III survey. For example, you might see improved cure rates, lower defaulting, and increased admission and so would establish a prior mode above the posterior mode of the preceding SQUEAC. Since the prior would be based on a previous SQUEAC and new data you could (reasonably) have a moderately strong prior. This means that your sample size for the SQUEAC stage III likelihood survey could be quite small. The survey would return an estimate and credible interval which is more informative than the classification returned by a SLEAC survey and it would very probably use a smaller sample size. That is more information for less work!

This last point means that you would be better to do SLEAC then SQUEAC rather than SQUEAC then SLEAC.

I hope this makes sense.

Back to top

» Post a reply