You could choose between two strategies: using the “regular” SQUEAC method, which builds the Prior using 20-80% (or 10-90%) OR “building up” the Prior from the last single estimator (46.9%). In your case, you will use 46.9% from which you will add and subtract the “weight” given to the new boosters and barriers; your qualitative work will focus on issues pointed out in the former survey (for which I understand the recommendations have not been implemented). Using this second strategy, for which you already have a coverage estimation and C.I, should allow you having a smaller sample size for Stage 3 (likelihood). However, it looks like the last SQUEAC was done more than one year ago and thus it might be inappropriate using the previous coverage estimation since, during this time, several other and new boosters/barriers might have affected coverage. The second method (building up from the previous coverage estimation) is more appropriate when the second (or third of fourth) investigation is carried out every 3-6 months (clinical audit strategy) and specific actions have been implemented following the boosters/barriers identified. In this case, it is much easy to assume that the new coverage estimate is related to the actions undertaken. Whichever strategy you will use, it will be important that in stage 2 you focus on issues identified in the first survey in order to understand the current situation (i.e. the impact of the non implementation of certain recommendations). Briefly, in your case, I would do a “new” SQUEAC investigation but information (boosters/barriers) from the previous one will help me to give the “weight” of the new boosters/barriers.

Hi,
One important purpose of SQUEAC assessment is to identify new barriers to coverage of CMAM program and take necessary action to address those identified barriers to increase coverage and quality of CMAM program.
From your statements, it looks that there are not community mobilization staffs/CHWs and barriers identified from previous SQUEAC have not been addressed, recommendations have not been followed. I am afraid that coverage may be even less than previous one as there are not community mobilization staffs/CHWs since Jan 2016 and I can assume poor community mobilization, case finding, referral and follow up in this situation. As we know, community mobilization is one of the important component and backbone of CMAM. I can not expect good coverage of CMAM without strong community mobilization component. So, I think it will be better to use existing resources to strengthened community mobilization activities, address the already identified barriers and follow the recommendations made by previous SQUEAC before implementing another SQUEAC.

It is not very fruitful to conduct SQUEAC if we have not addressed already identified barriers.

Lio and Sanjay both make very good points.

I will adress a technical point ... what are theses p-values? The only p-value that is routinely associated with SQUEAC stage III surveys is the test for a conflict between the prior and the likelihood. It does not say much about the coverage estimate except that it is likely to be credible since your prior and likelihood say pretty much the same thing. This can be viewed as a SQUEAC quality measure but interpretation is not straightforward as almost any weak prior will yield a non-significant p-value.

To add the previous replies, the CMN has developed some brief guidelines on conducting a follow up SQUEAC which are available here:
http://www.coverage-monitoring.org/wp-content/uploads/2015/09/CMN_Follow-up-SQUEAC-Companion.pdf

As Sanjay noted however, if minimal activities have been implemented since the initial coverage assessment, then there is no strong reason why coverage will have changed. If it is a requirement of the donor to have a coverage estimate, then as Lio suggests, you can form a stronger prior based on the previous coverage estimate and a reduced qualitative survey. A stronger prior (i.e. a narrow curve) will mean that your sample size will be smaller and you will therefore have to find a smaller sample during the wide area survey.