EVALUATION OF COMMUNITY LLIN DISTRIBUTION, NIGERIA

List of Abbreviations

ANC / Ante-Natal Care
BCC / Behavioral Change Communication
CD / Continuous Distribution
CDD / Community Drug Distributor
CI / Confidence Interval
FCT / Federal Capital Territory
HF / Health Facility
HH / Households
ITN / Insecticide Treated Nets
JHU / Johns Hopkins University
LGA / Local Government Area
LLIN / Long-lasting Insecticidal Nets
PHC / Primary Healthcare Center
PMI / President’s Malaria Initiative
UNICEF / United Nations Children’s Fund
USAID / United States Agency for International Development
WDC / Ward Development Committee
WHO / World Health Organization

Table of Contents

List of Abbreviations

Table of Contents

Acknowledgements

Executive Summary

Background

Study Site

Community Distribution

Distribution design

Implementation

Evaluation Methods

Evaluation objectives

Evaluation design

Sampling and sample size

Data collection

Data analysis

Ethical clearance

Results

The sample

Result of community distribution

Access to new nets

Net ownership

Equity aspects

Behavior Change Communication

Net use

Discussion

Conclusion and Recommendations

Acknowledgements

This report is made possible by the generous support of the American people through the United States Agency for International Development (USAID) and the President’s Malaria Initiative (PMI) under the terms of USAID/JHU Cooperative Agreements No.GHS‐A‐00‐09‐00014‐00 and AID-OAA-A-14-00057. The contents do not necessarily reflect the views of PMI or the United States Government.

The implementation of the CDD distribution was carried out as part of the MAPS program while the evaluation surveys were implemented through Malaria Consortium.

Executive Summary

This report presents the outcome of an evaluation of a community-based, push-pull system of LLIN distribution in Nasarawa State, Nigeria.Households in selected areas of the State were encouraged to contact their local Community Drug Distributor (CDD) if they felt they need new or additional nets to protect their family from malaria. The CDD would then verify the need and if satisfied, issue a net coupon against which the household could obtain a new LLIN from the nearest distribution point at a health facility.

The community based distributions started in June 2013, 31 months after the last mass campaign and continued for 10 months until the evaluation survey in April 2014. Data from the endline survey was compared to the findings from a baseline survey undertaken in October 2012, 11 months after the campaign. Both surveys were standard population representative household surveys with a cluster sampling design. Targeted sample size for each survey was 1020 and sample achieved was 1015 and 1003 for baseline and endline respectively. The evaluation design was a before-after comparison with an embedded “per protocol” comparison between households aware and not aware of the CDD scheme at the endline survey.

The major findings can be summarized as follows:

  • ITN ownership indicators decreased from baseline to endline, but at endline were twice as high among households aware of the CDD distributions than those not aware, and the decline from baseline to endline was small for the CDD households and statistically not significant. In detail the indicators were
  • Ownership of any ITN: baseline 63%; endline CDD 55%, non CDD 28%
  • Owning 1 ITN/ 2 people: baseline 25%; endline CDD 17%, non CDD 6%
  • Population access to ITN: baseline 45%; endline CDD 35%, non CDD 16%
  • Program effectiveness was overall not very good with only 18% of households that were aware of the CDD distribution getting a new LLIN through the system. The biggest loss was by people not requesting nets because either nets or net coupons were out of stock (55% of non-registration) or people did not understand the scheme (42%). Less than 2% said that they already had enough nets. Considering only the steps from requesting to getting new LLIN the effectiveness was significantly better with 52% and again losses were mainly due to logistical reasons with nets or coupons out of stock. Willingness of people to redeem the coupon once given was very high with 93% effectiveness from getting the coupon to going to the distribution point.
  • Equity of distribution showed almost perfect equity at baseline and was slightly pro-rich at endline but less so in the CDD group compared to the non CDD group.Concentration indices were 0.05 and 0.12 for any ITN and population access, respectively, for the CDD group, and 0.13 and 0.15 for the non-CDD group.
  • The proportion of respondents who had been exposed to any messages around ITN use in the last six months was highest at baseline with 68%, but at endline was also significantly higher for households aware of the CDD program (49% vs. 27%).
  • The BCC linked to the CDD distributions contributed to a better net utilization with 49% of those with access to an ITN using it at endline in the CDD group, compared to only 31% in the non CDD group. However, ITN use was generally much lower in the dry season (endline) compared to the rainy season (baseline), when 71% of those with access had used the ITN.

The conclusions from the findings are that

  • A community-based push-pull distribution system can work well to reach households missed by the campaign, those who did not get enough nets, or those that lost the ones they had, provided the scheme is well advertised and the logistics of supply function smoothly
  • These distributions do not oversupply households but help to supply just enough for all members to use ITN
  • There was very little overlap with other supply channels showing that these distributions complement each other in reaching all households
  • Community-based distribution in this setting is reasonably equitable and equity can be further improved if the poorest wealth quintile is encouraged to request new nets
  • BCC linked to the community-based distribution is an important contributor to enhance messages on ITN use resulting in better utilization of nets by those with access

The following major recommendations can be made:

  • Community-based distributions – like all other continuous channels – need to be started at least within one year following a successful mass LLIN campaign which implies that preparations should always start even before the campaign takes place
  • Informing the population about the distribution scheme and carefully explaining how it works is critical and must be emphasized during roll-out as this will a the most important step to achieve high program effectiveness
  • Special attention should be paid to involving the poorest households to ensure that they feel encouraged to participate in the scheme
  • Logistics systems need to function smoothly and all efforts should be made to avoid stock-outs of LLIN and net coupons
  • BCC messages supporting use of nets should always be incorporated in such distributions

Background

Following the dramatic scale up of LLIN distributions in Nigeria through mass distribution campaigns, it is increasingly realized by the Roll Back Malaria Partnership that repeated campaigns are not the best solution to sustain the achieved gains and that a comprehensive distribution strategy which includes continuous distribution mechanisms isneeded. The purpose of such continuous LLIN distribution is to

  • Avoid decline of population coverage (and hence protection) to levels below the threshold of the mass effect between repeated campaigns caused by the loss of nets through wear and tear
  • To provide replacement nets as well as supply new families when and where this is needed and to do this in an equitable and sustainable manner

Two principle approaches are possible in the implementation of continuous distribution. The most common way is to use routine health services to give LLIN to persons at particular risk such as pregnant women through ANC or children when they come for immunization. The eligibility to receive an LLIN is defined exclusively by the attendance at the specific services and therefore these distributions are called routine distributions in the strict sense. They can also been described as a push system.

The second approach consists of distribution mechanisms which are initiated by those who have a perceived demand for nets and then gain access to a full-price or subsidized system. This can also be described as a pull or demand-driven system and includes among others the classical social marketing.

A combination of the two would be a push-pull system where a certain amount of nets is pushed to a certain point in the supply chain (hub) from where the net users can pull down nets according to their demand and based on defined eligibility criteria.

In December 2010, nets were distributed in a mass campaign in Nasarawa state, where each household/family received two nets. NetWorks Project designed and supported the implementation of a community-based long lasting insecticidal nets (LLIN) distribution strategy, to help improve and sustain net coverage beyond the mass distributions in all LGAs in Nasarawa state.

This community-led LLIN distribution strategy has two separate phases:

1)The push of LLIN to the storage/ health facility

2)The pull of LLIN by households, which requirestwo active steps by the household (supported by adequate BCC)

  • Visit to a Community Drug Distributor (CDD) to declare the need for an additional or replacement net
  • Visit to a LLIN storage hub (health facility) to redeem a LLIN coupon issued to the household upon assessment of LLIN need by CDD. LLIN coupon is redeemed for a new LLIN

Study Site

Nasarawa State was created in 1996 through a split from Plateau State and has 13 Local Government Areas (LGA). It is located in the North Central zone of Nigeria bordering the Abuja Federal Capital Territory (FCT) to the West, Tabara and Plateau States to the East, Kaduna State to the North and Kogi and Benue States to the South (Figure 1). With a surface area of 27,117 km² and an estimated population in 2010 of 2,097,132 based on the 2006 census results and a reported 3.0% growth rate [4] the mean population density is relatively low with 77 persons/km². Due to the multitude of ethnic groups in the state Nasawara has been called a “mini-Nigeria” in this respect. The major ethnic groups include Eggon, Tiv, Alago, Hausa, Fulani, Mada, Rindre, Gwandara, Koro, Gbagyi, Ebira, Agatu, Bassa, Aho, Ake, Mama, Arum and Kanuri. While English and Hausa are widely spoken in the state, all the ethnic groups indicated above also have their own languages.

Figure 1: Location of Nasarawa State and its 13 LGAs within Nigeria

Nasarawa State lies within the Guinea Savannah eco-geographical zone. Accordingly, overall rainfall is moderate to high varying between 1300-1550 mm per annum. The rainy season lasts from April to November with the peak of rains between July and October.

Community Distribution

Distribution design

Following meetings with the State authorities for Health discussions were held with key partners and stakeholders (UNICEF, WHO) and departments of the State Ministry of Health (Onchocerciasis Control and HMIS). This was followed by field visits to two LGAs (Kokona and Akwanga) including discussions with the LGA health teams, visit of a health facility and discussions with a number of CDD. A one-day workshop then presented the options the team had developed to the stakeholders and preferences and challenges were discussed. Building on the consensus reached with representatives at all levels (State, LGA, Wards, communities and the CDDs), the community distribution channel was designed with the aid of the NetCALC tool.

The CDD distribution system:

  • Builds on existing systems and mechanisms and supports or enhances these as much as possible (integration).
  • Makes adjustments where needed to take into account the specific characteristics of the nets (e.g. their bulkiness) as well as aspects of use (need for a net culture and motivation to use)
  • Complements ANC-based distributions as recommended by WHO and NMCP since this will guarantee that pregnant women receive an LLIN as early in their pregnancy as possible without having to wait for the next distribution period
  • Is community-driven, requiring a pro-active role of households to initiate the process of obtaining new or additional LLIN (push-pull system)

The NetCALC tool estimated the output that can be expected of various distribution channels (routine services including schools or community or market based) and compared these with the calculated need to sustain the defined target. Using data on access to services from the 2008 DHS such projections were made for Nasarawa and results are shown in Figure 2. These suggest that even if ANC and EPI distributions are used together with school-based distributions in primary and secondary schools, the target of 80% LLIN household ownership could not quite be sustained. However, if ANC distributions to protect pregnant women throughout their pregnancy were combined with a community-based distribution system (e.g. through CDD), approximately 50% of all households would need to receive a net through the community system each year, and the target of 80% LLIN ownership could be sustained. In such a scenario for 2012/13 about 58,000 LLIN annually would need to be channeled through ANC and 220,000 through the CDD system.

Figure 2: Output from NetCALC for Nasarawa: light green line expected coverage from distributions; dark green diamond ITN ownership at survey; blue line need of LLIN to sustain target of 80%; red lines different continuous distribution scenarios

The Nasawara state community-based distribution was designed as a State-wide implementation, comprising 13 LGAs and 147 wards, with each ward providing a good number of active CDDs based on the population size. At the time of designing community-based distribution, it was estimated that each CDD covered between 250-2,000 people during their routine drug-distribution activities. In the first year of implementation, lacking consumption data to base LLIN allocations to the wards, the design team used the NetCALC tool to estimate that a CDD covering 2,000 people would distribute approximately 200 LLIN in the first six months, and 130 LLIN in the second six months.

Figure 3: Basic design of the community distribution system. HH= household, DC=development committee, HF=health facility

Implementation

Microplanning

As micro-level planning for the community-based distribution continued beyond the design stage, it was decided that a scaled approach would be used. The distribution began in four wards of each of the 13 LGAs, with an initial cadre of 260 CDDs and 52 health facilities (13 general hospitals and 39 primary health care centers) serving as local net storage and distribution hubs. During implementation, the number of wards reached by the CDDs gradually increased.

Stakeholders and RBM Partners conducted a micro-planning meeting that clearly delineated roles and responsibilities. The micro-plan also produced implementation timelines, developed LLIN Coupon, health facility register and summary sheets at various levels, quantification of LLIN need by LGA and a supervision checklist.

Training

The trainings were partitioned into two parts. First, State-level training of trainers workshop for State and LGA (malaria and onchocerciasis officers) and, secondly, LGA level cascade trainings for in-charge of PHCs, CDDs and WDCs. A state-level training of trainers was held over a two-day period with two representatives per LGA (Roll Back Malaria focal persons and the Neglected Tropical Diseases focal person for each LGA). These 26 trainees in turn trained the five CDDs in each of the four selected wards of their LGA (260 CDDs) and the in-charge/head of the 52 health facilities that would serve as the hubs for storage and issuing of nets.

The cascade trainings brought together different caliber of personnel on a one-day basis on different days, to train the CDDs and the health facility head involved in the distribution. The training gave an overview of the LLIN distribution process through CDDs and how to complete the data collection tools, such as the LLIN gap assessment form, LLIN Coupon, health facility register and summary forms. Trainees also reviewed what messages to share with clients to health facilities to raise awareness about using LLINs and how to obtain one through the CDD system. The cascade training approach also created a mechanism for relaying important updates and feedback from the State Malaria Control Program through the LGA Roll Back Malaria focal persons to the health facility staff and CDDs.

Mobilization and BCC

The pull mechanism of this strategy depends on the demand for nets by potential beneficiaries. Demand generation activities have been conducted primarily through interpersonal communication directly with community members by the CDDs, the Ward Development Committees, and community-based organizations and volunteers involved with MAPS activities in Nasarawa.

Flow of activities:

  1. LLIN are allocated monthly/quarterly to LGA and hubs based on expected need, prioritization and LLIN availability, after the initial phase consumption vs. expected need will also be considered
  2. Nets and other supplies are transported to hub
  3. In parallel communities (and local leaders) are sensitized and reminded to select CDD where there are no currently active ones
  4. CDD are called for a one day training on the processes and given materials (LLIN coupon booklets, assessment form booklet, registers and forms)
  5. Communities are informed that during a given period, they can approach the CDD and express their need for more nets
  6. CDD will visit these households and assess the need based on household members and existing, viable nets and give LLIN coupon(s). This can be combined initially with a prioritization (based on clear criteria and an assessment aid) e.g. based on size of gap in household in order to ensure settlements and families missed during campaign are served first without exhausting the net supply. Once a steady state is reached (second year) this would no longer be necessary
  7. Households take the LLIN coupon to the health facility (storage hub) or ward development committee to receive their net(s)
  8. Supervisors (PHCs, LGA) meet with CDD monthly/quarterly to discuss issues, collect summary forms and issue supplies if needed
  9. Summary reports from CDD and hub (distribution data) are sent monthly/quarterly to LGA and State.

Distribution