Menu ENN Search
Language: English Français

Low OTP Coverage

This question was posted the Coverage assessment forum area and has 4 replies. You can also reply via email – be sure to leave the subject unchanged.

» Post a reply

Anonymous 2352

Normal user

28 Oct 2013, 10:10

I have just concluded a SQUEAC Coverage Assessment in one of the Districts in Marsabit County, Kenya and found an OTP Coverage of 20.2%. This is quite low and even below the 50% threshold. My greatest barrier was low community level activities such as Active Case Finding, Mass Screening among others... What I wonder is whether there is any coverage assessment in the past which have found a lower value than this and if there was, what would have been the greatest barrier?

Lio

CMAM advisor

Technical expert

29 Oct 2013, 05:00

Hi Samuel,

Yes, I understand that you must be worried about such a low figure. MCN just published a very interesting document called “Access for all” (you can download it from the website http://www.coverage-monitoring.org/) where coverage figures are reported for several years. I am sure they can give you more information. In my experience, I must have had one of the lowest coverage figure of 16.1% (but it the reality if the Prior wouldn’t have been over over estimated, the true coverage would have been <10% which basically means that we could not really say that the organization was supporting a CMAM programme). But yes, most of low coverage figures are related to poor development of the community component but it is important to identify which aspect: level of activity of the CHW in terms of identification/Referral? Awareness of the programme? Appreciation of the programme? Recognition of malnutrition signs? Etc. What you say looks related to the level of activity of the CHW. You certainly have a good starting point for action but at the same time you know as well that it is often difficult to maintain a good performance of the CHW when they work on a volunteer basis. I believe you also have to look at the other aspects of the community work and focus on elements such recognition of malnutrition, awareness and appreciation of the programme, etc. ; overall if people know the programme and they are happy with the outcomes, they will spontaneously come at the HF, but for this, they should be able to recognize the signs of malnutrition. As you can see, all elements of the community components are related and a good CMAM programme should address every component. So, in order to make a long story short: yes……………….. there are example of cases of low coverage, and even lower of what you found, but the important is to clearly identify the reason/s and focus on these reasons.

Lio

CMAM advisor

Technical expert

29 Oct 2013, 05:05

Sorry for my previously typing mistake: of course I meant CMN and not MCN!!!

Anonymous 2352

Normal user

29 Oct 2013, 08:13

Thank's Lio for the above information. In fact, the prior estimate which we obtained through averaging FOUR prior i.e. Histogram, Weighted Barriers and Boosters, Concept Map and the Unweighted Barriers and Boosters gave us an average prior of 24%. In the wide area survey, we only had 3 SAM cases in the program out of the 18 that we got. The same trend was also observed in the same place when I also did a MAM Coverage which was basically 29%. However, the happier side of it is that we were able to get very insightful barriers to the program which we have also developed very workable action points with an aim of improving both the OTP and SFP Coverage.

Mark Myatt

Consultant Epideomiologist

Frequent user

29 Oct 2013, 10:19

Coverage of about 20% is typical of a program that has decent clinical services but little or no outreach / sensitisation / mobilisation / extension activities. When CTC (now called CMAM) was being developed we used to say that 20% coverage was what you got when you did nothing very wrong but had ignored the community aspects of CTC programming.

As far as I know, the lowest coverage found by a direct survey was about 9% (by contrast the highest was about 90%). I did hear of a SQUEAC being done in a program that had distributed no RUTF for 2 to 3 months. Here the coverage was (effectively) zero. These "failing" programs ignore the community aspect of programming but also have other problems such as RUTF stock-outs, bad behaviours by staff (e.g. verbal, physical, sexual abuse of mothers and children; demands of money for treatment; sale of drugs and RUTF; &c.), very low spatial coverage (e.g. clinic catchment areas with 30 km radius' ... that is about 3000 sq. km. per clinic), and very poor patient flow (e.g. patients having to wait several hours for a three minute consultation) often caused by habitual absenteeism by clinic staff. I have also seen coverage sabotaged by traditional healers and religious leaders but this is indicative of inadequate community activities ... these people are key to mobilisation and case-finding.

It is easy to be disappointed by low coverage. It is important to remember that 20% is 5 to 10 times higher than was typically achieved by the TFC model of care delivery. It is also important to remember that you have (as Lio says above) diagnosed the problem and have a good idea of what needs to be done to improve coverage. The time for disappointment is when you come back a year later and find coverage stuck at 20%.

Your reference to a "50% threshold" is. I think, to SPHERE minimum standards. These standards were developed for emergency contexts and neglect time in the sense that they do not specify how long we should allow before the coverage standard is met or exceeded. This is not a simple question. The answer will vary by context. A simple example of contrasting contexts is emergency vs. development settings. In an emergency setting we would want a very short attack phase, measured in days or weeks, and resources will usually be available to achieve this (usually by vertical intervention). In a development setting we often find ourselves working in poorly functioning health systems operating with severely constrained resources. In such settings we accept, or are forced to accept, a longer attack phase measured in months or years. From this perspective, 20% at the first SQUEAC is "not so bad".

WRT your analysis ... you have a prior mode of 24% and a likelihood with mode of 17% (i.e. 3/18 = 0.1667). Your 20% posterior mode suggests a balanced (in terms of strength) prior and likelihood (20% is close to the means of 24% and 17%). Your 24% vs. 17% suggests a realistic prior and no prior-likelihood conflict. I think Lio's example involved a strong and non-realistic prior which may have biased the posterior estimate towards overestimating coverage.

I hope this helps.

Back to top

» Post a reply