Menu ENN Search
Language: English Fran├žais

Beneficiary calculation for multiple years

This question was posted the Prevention and treatment of moderate acute malnutrition forum area and has 3 replies.

» Post a reply

Sanne Bakker

Nutrition Officer

Normal user

19 Aug 2014, 10:00

Dear Nutrition colleagues, I am looking for guidance regarding the following: We are implementing a targeted supplementary feeding programme for the treatment of MAM which includes a nutrition BCC component for programme beneficiaries, caretakers at the community at large. How can you calculate the expected case load for the coming 5 years? Does the correction factor become K = Prevalence +incidence factor above=1 + {1 x(60months/7.5months)} = 9? How can we take into account the expected decrease in MAM prevalence? 3 extra challenges for this case load estimation is that i) we would need the annual caseload and that ii) we might follow a phase-wise implementation and therefore the under 5 population changes per year and iii) we would like to factor in the population growth rate. Thanks in advance for any advice!

Mark Myatt

Epidemiologist at Brixton Health

Frequent user

19 Aug 2014, 11:01

I think you could tackle this on a phase-by-phase basis in a spreadsheet. That way you can specify phase specific changes in population, coverage, and prevalence using a system of simple equations which can then be summed over years and over the program duration. The process will be a lot like working out an expected program budget or a business plan. This should give you what you want. Once you get a working model you should check the maths and the assumptions and then do a sensitivity analysis so you get some idea of the uncertainty of the results (e.g. what would happen if my coverage changed at a different rate). I usually take a sampling-base approach to sensitivity analysis this but there are other methods. I hope this is of some use.

Zola

responsable nutrition/ Care International

Normal user

21 Jun 2019, 12:44

Hello Mark I work in an NGO and I am faced with the same problem as that expressed by my predecessor on the calculation of caseload of children from 6 to 59 months for a project of a duration of 5 taking into account the reduction of cases of acute malnutrition, the rate of growth of the population. Could you enlighten me more? I did not understand your answer, or rather it seems complicated to me. An example?

Mark Myatt

Epidemiologist at Brixton Health

Frequent user

24 Jun 2019, 10:26

Zola,

We start with the base calculation:

    case-load = NPCK

where N is the population, P is prevalence, C is coverage, and K is a fudge factor. We can implement this in a simple spreadsheet. 

I have put an example together here.

We know that population will change over time by growth (i.e. deaths - births) and migration (exits - entries) in excess of normal growth. This needs to be accounted for. This is done in columns, B, C, and D of the example spreadsheet.

Prevalence will change. You can make informed guesses based on past results of (e.g.) SMART surveys and the example expected trend. This is in column F of the example spreadsheet.

Coverage may start low and increase over time. I have done this in G of the example spreadsheet.

Estimated case-load is calculated in column H of the example spreadsheet.

This is a simple model but it allows you to do what-if type simulations. For example, you may want to see what might happen if the expected decrease in prevalence did not occur.

It is usually a good idea to create (at least) best case, expected case, and worst case scenarios. 

I hope this is of some use.

If you have any problem posting a response, please contact the moderator at post@en-net.org.

Back to top

» Post a reply