Against Malaria Foundation - March 2012 Update

Against Malaria Foundation (AMF) has received significant funding as a result of GiveWell's recommendation (details). This page provides periodic updates on its activities and plans.

Table of Contents

Progress in Ntcheu district of Malawi and plans for future distributions

Summary

AMF has provided regular, public updates on the large, ongoing net distribution in the Ntcheu district of Malawi. Expected data collection has occurred and the distribution has proceeded close to schedule. AMF's distribution partner, Concern Universal, has been transparent about problems it has encountered, and seems to have a robust process to catch problems (such as attempts to steal nets) when they arise.

Distribution in Ntcheu district of Malawi

As discussed in our 2011 review of AMF, in late 2011, AMF sent about 250,000 nets to its distribution partner, Concern Universal, to distribute to households in the district of Ntcheu, Malawi. As of March 8, Concern Universal reported that it had distributed 242,745 nets to households in the district; for its final distribution location, Concern Universal requires another 16,574 nets. The final distribution will take place in late March/early April.1

This distribution was not a result of GiveWell's top recommendation (the distribution was planned before we made AMF our #1 charity); however, AMF has told us that future distributions are likely to resemble this one, and we therefore report in detail below on the progress of this distribution.

Progress to date:

  • Concern Universal completed a household-level pre-distribution survey for the district. AMF shared the full results for this survey. To conduct the survey, Concern Universal sent surveyors door to door to each household in the villages. They filled out surveys with the following information: head of household name, number of people in household (under 5 and over 5), number of sleeping spaces, number of usable LLINs present in household, and the number of nets required (calculated by subtracting number of usable LLINs present from number of sleeping spaces).2
  • Concern Universal provided a schedule for the distribution. To date, distributions have been conducted mostly on schedule.3
  • Concern Universal has reported weekly on how many nets have been distributed in each distribution location (referred to in the reports as "health centres"). The number of nets distributed has corresponded closely with the number of nets thought to be needed in the area (based on the findings of the pre-distribution survey).4

  • Concern Universal has also reported weekly on problems encountered during the distribution, including:
    1. Theft: Concern Universal has reported instances in which Health Surveillance Assistants (HSAs) assisting with the distribution process stole or attempted to steal nets. Of the six reports from distributions in December 2011 to February 2012, three mention incidents of HSA corruption (describing six distinct incidents). In two of the cases (one confirmed and one probable), HSAs had registered fake names for nets, so that they would have the opportunity to steal them when no one came to collect them. Concern Universal also confirmed an instance in which two entire villages were fabricated.5

      Concern Universal reports that theft "has made the project team not to rely so much on the HSAs when distributing but instead rely on the village members who are the beneficiaries."6 Concern Universal mentioned that they had never found instances of theft from villagers helping with the distributions.7


      Concern Universal appears to have a relatively robust process for both preventing and catching cases of fraud. The updates note that there are Concern Universal staff members present at the distributions with HSAs, and also make clear that Concern Universal is aware of potentially problematic situations where nets might be stolen.8

      In at least one case, Concern Universal noted when a number of villagers did not collect their nets, and investigated whether this was a case of HSAs registering fake names.9

    2. Double registration: There was an instance in which villagers double-registered for nets. However, nets are distributed at the same time of day at different points in the same area, making it difficult for villagers to double-register and collect in multiple places.10
    3. Incomplete registration: In some cases villagers hadn't registered for nets, so that when they went to collect them, there were none available for them.11 Concern Universal informed as that this problem had been dealt with and villagers had received their nets.12
    4. Logistical problems: The updates mention a few other minor problems that arose during the distribution (need for additional transport, issues with heavy rain, etc); these do not appear to have caused any serious problems for the distribution.13

Room for more funding and plans for future distributions

We provide a full update on AMF's future plans in our blog post.

Expectations for the next update

We expect an overview report on how the Ntcheu distribution went as well as photos and video within a few months of the end of the distribution.

Post-distribution surveys to check the condition and usage of the nets are scheduled 6, 18, 30 and 42 months after the distribution ends.

Sources

  • 1

    Rob Mather, Email to GiveWell, March 8, 2012.

  • 2

    Concern Universal, "Pre-Distribution Survey Data"

  • 3

    Against Malaria Foundation, "Distribution Plan."

  • 4

    Data from Concern Universal, "Distribution Updates" compiled in GiveWell, "Nets Tables."

  • 5
    • Case of ghost names added to the list: "For the Tsangano Health Centre catchment area we recorded a significant variance of 996 nets (10% fewer were distributed than we anticipated). We believe that fraud by a few of the Health Surveillance Assistants (HSAs- the 451 government extension staff who conducted the initial registration) in this area may account for the bulk of this variance. One HSA confessed that he added ghost names to the initial register which were not removed during verification. He thought that he would be the one responsible for distributing the nets and earmarked 90 nets for himself as a result of these ghost names which he apparently planned to sell. He was unhappy to find out that CU staff would be present for all distributions hence meaning that he was unable to take these nets." Concern Universal, “Distribution Update (December 16-31, 2011)," Pg 2.
    • Case of HSA theft: "The greatest challenge and lesson learnt during the week was when one HSA from Nsiyaludzu health centre collected 7 nets without the knowledge of the distribution officer. This made five beneficiaries from Pembereka village not to receive the nets. After investigations the HSA was identified, called, questioned and he gave back all the 7 missing nets and two days later the five beneficiaries who did not receive their nets were called and have received their nets. This has made the distribution team to take additional care with the HSAs to ensure that such an incident doesn’t happen again. The distributing team has also learnt to collect the exact number of the nets per village requirement from the cluster net requirement when distributing so that they can easily identify the village that has caused the net shortage if there is any." Concern Universal, "Distribution Update (January 16-21, 2012)," Pg 2.
    • Case of attempted HSA theft: "There was an attempt by one HSA to steal 10 nets from Kasinje health centre. The nets were hidden in one of the officers’ office and were recovered by one of the distribution team members who happens to be a supervisor in the government health system. The matter was reported to the health centre authorities. This has made the project team not to rely so much on the HSAs when distributing but instead rely on the village members who are the beneficiaries. A few HSAs are frustrated that our model has not provided them with an opportunity to siphon out nets for their personal benefits hence their subsequent attempts to steal the nets." Concern Universal, "Distribution Update (January 23-28, 2012)," Pg 3.
    • Case of fabricated villages: "The final situation on ghost villages is that there were two that we have been able to confirm: Kasadzu 1 village under Kandeu Health Centre has been confirmed as a ghost village (on the records provided by the Health Surveillance Assistants it had 44 households and required 79 nets). In fact in Kandeu there is a village called Kasadzu which had 111 households and required 202 nets- all these people received their nets and Kaasadzu 1 never existed. Then the other ghost village was Kadzombe village under Kampanje Health centre which had 23 households and required 48 nets according to the Health Surveillance Assistant’s records." Robin Todd, email to GiveWell, March 15, 2012.
    • Case of suspected registration fraud: "There was a high return of 120 nets from one village of Sanjani under Dzunje health centre as the concerned beneficiaries were not known by the villagers. We suspect that this was a case of registration fraud." Concern Universal, "Distribution Update (January 16-21, 2012)," Pgs 1-2.
    • Case of suspected registration fraud: "Nsipe had a large number of nets not collected by the beneficiaries mainly due to double entry‐ this is because there were villages in the register named Chizindeni and Gwedeza Chizindeni which in reality is one village but it was registered as two separate villages which left us to believe that this had been done deliberately by the relevant HSA in order to access the excess nets to sell at the local trading centre. There was also a data issue at Ntonda health centre in the villages of Dambule 1 and 2. It was observed that most beneficiaries from Dambule 1 also existed in Dambule 2 data base hence they had already accessed their nets from Dambule 1. This situation had not been dealt with during data cleaning as it was only when both villages were asked to collect nets from the same distribution point that it became clear that there had been duplicate entries. The team distributes from 10 distribution centres at the same time everyday meaning that it is practically impossible for villagers to collect two nets from different sites by double registering." "Distribution Update (January 9-14, 2012)," Pg 2.

  • 6

    Concern Universal, “Distribution Update(January 23-28, 2012)," Pg 3.

  • 7

    "We have involved the village leadership (headman etc. and members of the Village Health Committee) closely at the distribution stage rather than just using the HS to distribute the nets with our team. This is because we have come across several cases of attempted theft by HSAs but none to date from the village leadership." Robin Todd, quoted in Rob Mather, email to GiveWell, March 8, 2012.

  • 8

    For example: "One HSA confessed that he added ghost names to the initial register which were not removed during verification. He thought that he would be the one responsible for distributing the nets and earmarked 90 nets for himself as a result of these ghost names which he apparently planned to sell. He was unhappy to find out that CU staff would be present for all distributions hence meaning that he was unable to take these nets." Concern Universal, “Distribution Update (December 16-31, 2011)," Pg 2.

    "Heavy rains have also been another challenge in that it is making the distributions to take long to finish per distribution point as the beneficiaries have to fetch shelter for some time till the rains stop. This is mostly occurring in places where the distributions are taking place on open grounds where shifting the nets to the shelters has potential to expose the nets to theft. Bridge wash away has also been another challenge in that we had to reallocate one distribution point meaning beneficiaries had to cross the river to access the nets. These challenges are delaying the distribution process hence the distribution staff have been forced at times to work under darkness so that they finish the day’s distribution. However we have been receiving good support from both health centre personnel and the local lead" Concern Universal, “Distribution Update (January 23-27, 2012)," Pg 3.

  • 9

    "At Kasinje health centre, 95.1% of the sleeping spaces are under mosquito nets now since 21,810 nets were received by the beneficiaries of the 22,932 sleeping spaces requiring the nets. There are five villages (Gambatula, Mchokera, Mitongwe, Donyo and Saguga) under the health centre that have not been issued with the nets. This is due to communication breakdown for the Gambatula, Mchokera and Mitongwe villages, as the beneficiaries anticipated that the distribution will be one day ahead as for Donyo and Saguga. We are investigating the existence of these villages as no beneficiaries have so far claimed that they haven’t received their nets from such villages. We believe that they were fabricated villages. We have planned to arrange a day during the mop up week to verify this and serve the three villages with the nets they are supposed to receive. "Distribution Update (January 23-28, 2012)," Pg 2.

    "There was a high return of 120 nets from one village of Sanjani under Dzunje health centre as the concerned beneficiaries were not known by the villagers. We suspect that this was a case of registration fraud." Concern Universal, "Distribution Update (January 16-21, 2012)," Pgs 1-2.

  • 10

    "There was also a data issue at Ntonda health centre in the villages of Dambule 1 and 2. It was observed that most beneficiaries from Dambule 1 also existed in Dambule 2 data base hence they had already accessed their nets from Dambule 1. This situation had not been dealt with during data cleaning as it was only when both villages were asked to collect nets from the same distribution point that it became clear that there had been duplicate entries. The team distributes from 10 distribution centres at the same time everyday meaning that it is practically impossible for villagers to collect two nets from different sites by double registering." Concern Universal, "Distribution Update (January 9-14, 2012)," Pg 2.

  • 11

    "The major challenge during the week was that were a good number of villagers claiming that their names had been skipped. It turned out upon investigations that many of these people had not bothered to participate in the registration process as they did not believe that nets would actually be distributed as apparently there had in the past been registrations which were not then followed up by the actual delivery of intended items." Concern Universal, Concern Universal, "Distribution Update (January 30-Februry 4, 2012)."

  • 12

    Robin Todd, quoted in Rob Mather, email to GiveWell, March 8, 2012.

  • 13
    • "In order to deliver 250,000 LLINs before the end of February 2012 CU will need to add additional manpower and vehicles to the distribution fleet. It is particularly important to have a reliable 3-tonne truck to ferry net stocks to Health Centres. These issues have all been
      addressed and extra resources are now available for the distribution...There is a need for distribution teams to leave Ntcheu town earlier in the morning in order to ensure that all nets are distributed as planned during that day. There is also a need to further stockpile diesel for distribution given the supply problems in Malawi at present." Concern Universal, "Distribution Update (December 16-31, 2011)," Pg 2.
    • "In the three Health Centres of Doviko, Matanda and Katsekera there was a variance of between 99 and 169 nets (between 2% and 6% of the total anticipated). We believe that the bulk of these nets remained undelivered as registered people did not attend to collect their nets. This will be addressed during our distribution mop-ups in February 2012 when all communities with remaining nets to collect will be revisited at the end of the main distribution." Concern Universal, "Distribution Update (December 16-21, 2011)," Pg 2.
    • "There is need to just send exact number of nets to a distribution point, as the extra nets sent can spark controversy in the villagers thinking that the extra nets were also meant for them. The sending of the extra nets has been mainly due to the project deeming sending incomplete bales as compared to removing the nets from the bales as it will create risk especially in the warehouse." Concern Universal, "Distribution Update (January 9-14, 2012)," Pg 2.
    • "Since the rains have started in earnest some of the bridges to some sites have been washed away making such villages inaccessible hence we have to go through other alternative roads which can be time consuming and thereby affecting our distribution timeliness" Concern Universal, "Distribution Update (January 16-20, 2012)," Pg 2.