DRAFT A framework of strategic options For the integrated delivery Insecticide-treated nets and immunization
London School of Hygiene and Tropical Medicine, and
Jenny Hill, Child and Reproductive Health Group
Liverpool School of Tropical Medicine and Hygiene.
This document will be technically edited and published in English by May 2006 Acknowledgements
This framework, commissioned by WHO, was written for the Global Malaria Programme by Jayne Webster, TARGETS Consortium, London School of Hygiene and Tropical Medicine, UK, and Jenny Hill, Child and Reproductive Health Group, Liverpool School of Tropical Medicine, UK. The authors wish to thank all individuals who provided documents and contributed information from their own experiences. The views expressed in the paper do not necessarily reflect the policies of WHO and statements regarding countries, their borders and populations do not imply the expression of any opinion on the part of WHO. This document should not be quoted or disseminated without the agreement of WHO. Queries should be addressed to Dr M. K. Cham, Global Malaria Programme ([email protected]).
accelerated child survival and development
Bacille Calmette-Guérin (vaccine against tuberculosis)
Community Based Malaria Prevention and Control Programme
Canadian International Development Agency
Global Alliance for Vaccines and Immunization
Global Fund to fight Tuberculosis, AIDS and Malaria
Deutsche Gesellschaft fur Technische Zusammenarbeit
Haemophilus influenza type b vaccine
International Federation of the Red Cross and Red Crescent Societies
Integrated Management of Childhood Illness
intermittent presumptive treatment for pregnant women
London School of Hygiene and Tropical Medicine
tetanus toxoid vaccine (TT1 = first dose, TT2 = second dose etc)
Table of contents
SUMMARY 7 1. INTRODUCTION 1.1 Goals and targets 1.2 Immunization delivery: schedules and rationale 2. MODELS OF INTEGRATED DELIVERY OF ITNS AND IMMUNIZATION 2.1 Integration of ITNs with routine immunization services 2.2 Integration of ITNs with expanded routine child health services
2.2.1 Child Health Week / Child Health Days
2.3 Integration of ITNs with vaccination campaigns 3. OUTPUTS AND OUTCOMES OF INTEGRATED DELIVERY MODELS 3.1 Routine systems 3.2 Expanded routine 3.3 Campaigns 4. CONSIDERATIONS IN SELECTING STRATEGIES FOR INTEGRATION 27 4.1 Target groups 4.2 Current coverage of ITNs and delivery systems for ITNs 4.3 Current coverage of EPI 5. ISSUES IN PLANNING, IMPLEMENTATION, MONITORING AND EVALUATION 30 5.1 Opportunities and challenges of specific models 5.2 Cross-strategy issues in planning, implementation, monitoring and evaluation 6. PARTNERSHIPS 6 .1 Key partnerships spearheading integrated campaigns 6 .2 Partnership Coordination for integrated campaigns 6 .3 Partnership lessons REFERENCES 47 Summary
The World Health Organization (WHO) and the United Nations Children’s Fund (UNICEF) have recently published a joint statement on their commitment to developing more integrated programming approaches between malaria control and immunization. National malaria control programmes and their partners are showing increasing interest in joining WHO and UNICEF in their proposed integrated approach to insecticide-treated net (ITN) and immunization programming. Reviews of published and grey literature were undertaken, as were three country visits, in order to analyse and collate past, current and planned integration experiences. Three types of integrated delivery of ITNs and immunization are defined. These are: routine, enhanced routine, and campaign. They are further categorized by the service through which they are delivered, for example antenatal clinics (ANC) and measles campaigns. These categories of service reach varying portions of the malaria target groups of pregnant women and children under five years. No single category of delivery system integrated with immunization reaches the full complement of these targets. Combinations of integrated delivery models are therefore necessary to reach all target groups. Scale and duration of experiences has varied between and within the different categories as has the emphasis on assessment of, and dissemination of, outputs and outcomes. Experiences of national level implementation are few. Although there is a general focus on targeting of subsidies on ITNs to pregnant women and children under five years, blanket subsidies are being delivered in some countries. Insufficient funding for nationwide delivery through routine systems has resulted in geographical targeting of ITNs even within malaria endemic areas. Selection of areas for such geographical targeting has been based mainly on health and poverty indicators. To date more ITNs have been delivered through routine integrated systems than through campaigns. There have been no rigorous comparisons of the impact of different integrated delivery systems. Available data are from evaluations of single systems, using different indicators and operational definitions of indicators. However, available data show that global targets have been reached on a small scale (district level) by several of the models of integrated delivery, including ANC/Mother and Child Health clinic (MCH), intervention packages, and measles campaigns. Within national scale delivery, coverage outcomes vary between districts, some meeting the Abuja targets and others not. No single model of integrated delivery has yet reached 60% coverage of pregnant women and/or children under five years at the national level. Evidence suggests that integrated measles campaigns achieve a greater equity of coverage of ITNs across socioeconomic groups than do other integrated systems of delivery employed, or assessed, to date. Standard packages of indicators, instruments, and methods for monitoring and evaluation of the different models are needed which embrace all opportunities of collection of rigorous data at the national and sub-national level. Integrating the delivery of ITNs and immunization through a single system will not achieve the Abuja targets of 60% of pregnant women and children under five years sleeping under ITNs. Combining delivery models offers the best possibility for achieving this goal at the national level and with a minimum level of disparity across districts and socioeconomic groups. Increased mobilization of resources, effective strategic planning
and partnership coordination are required. Whilst focusing on the integrated delivery of ITNs with immunization, it is important to remember that this group of models is not exclusive. To date, the vast majority of children covered by nets in Africa got these nets through the commercial sector, and these nets are reaching the malaria target groups. The advent of new technologies will enable these nets to achieve the same level of effectiveness as the long-lasting insecticidal nets (LLINs) that are currently in great demand and relatively short supply.
1. Introduction The World Health Organization (WHO) and the United Nations Children’s Fund (UNICEF) have recently published a joint statement on their commitment to developing more integrated programming approaches between malaria control and immunization (1). The statement emphasizes that although collaboration has been limited to date, there is scope for exploring, developing and expanding across health programme synergies within the context of going to scale with national insecticide-treated net (ITN) programmes. In addition, the 11th and 12th Meetings of the Task Force on Immunization in Africa requested WHO to facilitate integration of immunization activities with other interventions such as ITNs, and deworming. The Task Force further recommended the development of technical guidelines for integrated programming of maternal and child health services, including monitoring and evaluation. Potential areas for collaboration identified were: delivery; social mobilization and education; geographical reconnaissance, surveying and monitoring; and training and supervision. National malaria control programmes are showing increasing interest in joining WHO and UNICEF in their proposed integrated approach to ITN and immunization programming. A review of published and grey literature was undertaken together with visits to three countries, namely Ghana, Kenya and Zambia, in order to gather information on country experiences relating to the integrated delivery of ITNs and immunization. This document synthesises the findings of the review in the form of a conceptual framework of strategic options for the integrated delivery of ITNs and immunization; the specific findings of the review are presented elsewhere. The major issues raised by countries with experience of such integrated delivery are presented together with potential solutions based on the available evidence. The document aims to provide a resource to national programme managers and other partners supporting the national scale-up of ITNs through integrated programming with immunization. 1.1 Goals and targets Malaria and Expanded Programme on Immunization (EPI) programmes have defined global goals and targets for ITNs and for immunization respectively, but integration of delivery systems for both interventions will require use of combined targets to guide programme strategy. 1.1.1 Malaria target for ITNs The goal of the Roll Back Malaria (RBM) initiative is to halve the burden of malaria by 2010. ITNs have been shown to reduce the number of child deaths by about one-fifth, saving around six lives for every 1000 children under five protected per year in sub- Saharan Africa (SSA). Targets to reduce child mortality established at the Abuja Malaria Summit in April 2000 therefore include a target for ITNs (2). Coverage of ITNs across many countries is however currently unacceptably low and the scale-up of coverage is vigorously being promoted by RBM. An RBM consensus document on ITN scale-up (3) suggests that the way forward for achieving and sustaining high level coverage with ITNs is a two-pronged approach: sustained and targeted subsidies for the most vulnerable, to promote equity, balanced with efforts to create an enabling environment to support expansion of the private sector, to promote sustainability. 1.1.2 Immunization targets The EPI was launched by WHO in 1974, at which time less than 5% of the world’s children were immunized against the six target diseases selected for inclusion in EPI. In
1984 the Universal Child Immunization (UCI) initiative was launched to accelerate EPI with target coverage of 80% for all six antigens. UCI was declared by UNICEF in 1990. However, the reality was that the 80% coverage achieved reflected high coverage in some relatively heavily populated countries (4) , masking very low coverage in other countries. The aggregated data hid the reality that 107 countries had not achieved universal coverage (5), and the fact that there were significant intra-country disparities in coverage. After 1990, there was a call to reach target populations that had not been immunized and the coverage goal was raised to 90% (for all six antigens). In acknowledgement of the distributional concerns over coverage, the Global Alliance for Vaccines and Immunization (GAVI) set a goal that 80% of developing countries should have routine immunization coverage (with three doses of diphtheria-tetanus-pertussis vaccine [DTP]) in at least 80% of all districts by 2005 (6). By 2010 or sooner, the goal of routine EPI immunization is 90% national coverage (with three doses of DTP [DTP3] in children one year of age), with at least 80% coverage in all districts1. In May 2005, governments at the World Health Assembly welcomed the launch of a new “Global Immunization Vision and Strategy” (GIVS) developed by WHO and UNICEF. The GIVS offers a unified vision of immunization and a set of strategies to meet the immunization challenges of the coming decade. GIVS has three main aims: to immunize more people against more diseases; to introduce a range of newly available vaccines and technologies; and to provide a number of critical health interventions with immunization.2
By 2005, 60% of children under-five and pregnant women are sleeping under ITNs
By 2010 or sooner, 90% national coverage (with three doses of DTP in children one
year of age), with at least 80% coverage in all districts
1.2 Immunization delivery: schedules and rationale Scaling-up the delivery of ITNs to national level is relatively new in comparison with that of delivery of childhood vaccination through EPI. EPI has developed a routine schedule for all children, which is complemented by intermittent campaigns. The schedules and rationale are presented below. The methods and frequency for delivering ITNs are less well defined and there is still much debate about the best way to achieve and sustain national level coverage among target groups. 1.2.1 Routine delivery Children up to about nine months of age are targeted for routine EPI, although children above this age who have not completed their course will be immunized. The recommended schedule of immunizations is presented in Table 1 below. The six primary antigens are Bacille Calmette-Guérin (BCG), DTP, measles and oral polio vaccine (OPV). With GAVI support many countries are now introducing new vaccines, often in combinations containing Hepatitis B (HepB) and Haemophilus influenza type b (Hib). Measles vaccine and yellow fever (where applicable), are the last of the routine childhood vaccines and are given at nine months of age.
1 UN General Assembly Special Session on Children 2002
Table 1. Immunization schedule for children Age Vaccines
BCG – Bacille Calmette-Guérin; OPV – Oral polio vaccine; HepB – Hepatitis B; DTP – Diptheria-tetanus-pertussis Hib – Haemophilus influenza type b Immunization schedules for protection against tetanus begin in the newborn period (Table 1), with reinforcing doses of tetanus toxoid (TT) when older. In addition, every woman of childbearing age (15–49 years) should receive five doses of TT (Table 2) in order to protect the newborn from neonatal tetanus. A minimum of two doses at least four weeks apart are required for protection. Table 2. Tetanus toxoid immunization schedule Dose When given Period of protection
At first contact with woman of childbearing
age, or as early as possible in pregnancy
1.2.2 Campaigns Eradication of polio by the year 2000 was adopted as an immunization goal in 1988. Intensive efforts continue to interrupt the transmission of poliovirus in the six remaining endemic countries (Afghanistan, Egypt, India, Nigeria, Niger and Pakistan). The three strategies of the Global Polio Eradication Initiative include: attaining high routine coverage with at least three doses of oral polio vaccine; conducting national immunization days (NIDs); and mop-up vaccination in high-risk areas as identified through surveillance in areas where polio is reduced to focal transmission. The goal of polio NIDs is to achieve 90% coverage of children aged 0 to 59 months in polio-endemic countries. Polio NIDs aim to interrupt poliovirus transmission through giving oral polio vaccine to all children in a large geographic area over a short period of time. Therefore special efforts are needed to reach children who are missed by routine immunization services. During NIDs doses of OPV are given to children within a defined age group, which is usually 0 to 59 months of age, regardless of their vaccination history. These are considered to be additional doses and do not replace those received through routine EPI. The aim is to provide these vaccinations during each round of NIDs in as short a period of time as possible, and preferably within two days. NIDs are conducted in two rounds and the second round should take place four to six weeks after the first. NIDs are expected to be conducted annually for three years after which time polio should be reduced to focal transmission and NIDs replaced by mopping-up campaigns and sub-national immunization days (SNIDs) as appropriate.
A global goal to reduce measles deaths by half by the end of 2005 (compared with 1999 levels) was adopted by the UN General Assembly Special Session on Children (2002). The effectiveness of measles vaccine is low in the presence of maternal antibodies which are present up to about six months of age. By the age of nine months, effectiveness of the vaccine is around 85%. The lower age limit for measles vaccination is therefore nine months. If a significant proportion of measles cases are known to have occurred in children between six and nine months of age then children within this age group may be vaccinated. However, they will then need to be revaccinated at nine months of age as below this age the vaccination is only about 50% effective. During measles campaigns, known as Supplementary Immunization Activities (SIAs), all children in the target age group are vaccinated regardless of their history of immunization. Traditionally the target age groups have been 9 to 59 months. However, in many countries significant levels of morbidity and mortality in older children mean that children aged up to 15 years may also be included in the target group. There are recommended intervals between follow-up SIAs dependent upon percentage coverage. Classification of delivery of measles vaccination is presented in the box below.
Classification of measles vaccination delivery: ¾ Catch-up
Efforts to vaccinate all children aged 9 months to 15 years in order to reduce the numbers of susceptible persons in the population (those never vaccinated and those in whom the primary vaccination failed). ¾ Follow-up
Periodic mass immunization campaigns conducted every three to four years following catch-up campaigns to reduce numbers of susceptible persons born since the last supplementary immunization activities e.g. if two years since last campaign — target 9 to 33 months. ¾ Keep-up Maintaining high coverage through routine activities during inter-campaign periods
The 42nd World Health Assembly (1989) called for global neonatal tetanus (NT) elimination. Elimination is defined as less than one case of NT per 1000 live births in every district. Areas burdened by maternal and neonatal tetanus (MNT), called high-risk districts (HRDs), typically have very limited health infrastructure. Supplementary immunization activities (SIAs) delivering TT to women of child-bearing age in HRDs have proven to be a cost-effective way to reach and protect these populations. This high-risk approach (HRA) continues to be recommended. Tetanus toxoid SIAs have been implemented in thirty-four countries during the period between 1999 and 2004, and in about eighteen countries in earlier years. To prevent an epidemic of yellow fever, at least 80% of the population must have immunity to the virus. This can only be achieved through incorporation of yellow fever into childhood immunization programmes and implementation of mass catch-up campaigns to vaccinate the entire population older than nine months of age. Protective immunity occurs within one week for 95% of people vaccinated.
Measles Supplementary Immunization Activities (SIAs) target:
Supplement routine measles vaccinations, target the susceptible and ensure
coverage of >90% of children aged 9 months to 59 months, or in some cases 9
National Immunization Days (NIDs) target:
90% coverage in children aged 0 to 59 months in polio-endemic countries
2. Models of integrated delivery of ITNs and immunization Routine and campaign systems of delivery for immunizations are well defined; however the same does not apply to delivery of ITNs. There are a multitude of systems to deliver nets, insecticides and ITNs to households. Many are historical while others have developed over the last two decades. Within a focus on integrated delivery of ITNs with immunization we include public sector delivery and mixed (public-private) models, but pure private sector delivery is excluded. Voucher schemes are an example of a public- private model where the voucher is delivered through the public sector and the ITN through the private sector. Whilst excluding pure private sector delivery from this analysis, we acknowledge that the private sector has been to date the most significant delivery system for mosquito nets (7). The impact of integrated public sector delivery systems on commercial ITN markets should be considered when evaluating strategies for delivering ITNs, as the private sector plays an important role in the sustainable supply of ITNs to non-target groups. Models for integrated delivery of ITNs and immunization may be broadly typified as routine, expanded routine, or campaign. We may then further define each of these types of delivery system by the category of service through which the end user receives the ITN (Figure 1). Routine integrated delivery of ITNs together with immunization may be achieved through services delivering tetanus toxoid to pregnant women (that is, through ANC), through services delivering childhood vaccinations (EPI), and through programmes delivering packages of the two together with other child survival interventions, for example through MCH. So far there are experiences of integrating the delivery of ITNs through all three types of integrated system (routine, expanded routine and campaign) but not all categories of service have been used. Child Health Weeks are being undertaken in a growing number of countries and involve concentrated promotion and delivery of child health interventions, such as de-worming alongside immunization, usually during a one-week period. These are not considered to be campaigns but rather “enhanced routine” delivery, and have been used to deliver both insecticide (re)treatment and ITNs. Immunization campaigns to which the delivery of ITNs may be integrated include measles SIAs, polio NIDs, yellow fever and tetanus toxoid. Experiences of ITN integration with immunization campaigns to date involve only the first two — for measles and polio. This section provides a brief introduction to the experiences of delivering ITNs through ANC, MCH, packages of interventions, expanded routine, and campaigns. It is followed by a section that gives examples of outputs and outcomes through these systems (section 3). Issues relating to delivery may differ within each of the three types of system, and sections 4 and 5 provide a summary of issues to consider when planning and implementing the integrated delivery of ITNs through these models. Questions being asked by countries and their partners currently implementing or considering implementing integrated delivery of ITNs and immunization are addressed. Section 6
describes the partnerships operating at global, regional and country levels, including the factors contributing to successful partnerships. 2.1 Integration of ITNs with routine immunization services Routine immunization is delivered through ANC clinics, and through Mother and Child Health Clinics/ Child Health Clinics (MCH/CHC), each of these offering an opportunity to reach different ITN target groups. There is more experience to date with targeting the delivery of subsidized ITNs to pregnant women through ANC than to children under five through MCH/CHC. Although the main target groups attending ANC and MCH are pregnant women and children under five respectively, it is also possible to take the contact opportunity with multigravidae at ANC to reach children under five, and at MCH to reach pregnant women. A combination of both ANC and MCH/CHC provides the best opportunity for reaching the majority of the ITN target groups. 2.1.1 Antenatal clinics ANC offers a good opportunity for targeting pregnant women with the delivery of ITNs and has been the focus for targeting subsidized ITNs to pregnant women over the last few years. The delivery of ITNs through ANC has involved two ways of providing the ITN subsidy: 1) giving a subsidized ITN (i.e., direct product) or 2) giving a discount voucher which can be exchanged for an ITN at a commercial or other pre-identified outlet. The level of subsidy for both direct product and vouchers ranges from 100% to 40%. Figure 1. Typology and categorization of models for integrated delivery of ITNs and immunization
Routine services Intervention packages Child Health Weeks/Days Enhanced Enhanced outreach Campaigns
Key Pink = some documented experiences Blue = no documented experiences Direct product Delivery of ITNs to pregnant women through ANC is either currently under way or planned in many countries of sub-Saharan Africa (SSA). In some countries, such as Cameroon and Ghana, the ITNs are distributed by the National Malaria Control Programme (NMCP) or its equivalent to districts and from there to peripheral health facilities. In other countries, partnerships with a social marketing organization are involved. Population Services International (PSI) is currently supporting the Ministry of Health (MoH) in ten countries to deliver ITNs to pregnant women and children under five
through ANC and MCH. The scale of these distributions ranges from one district (e.g. Angola) to national level (e.g. Malawi). The cost of the products to the end user varies depending on where distribution is undertaken by MoH and where social marketing organizations are involved. In Cameroon, ITNs are delivered free to pregnant women through ANC, whereas in Ghana the cost to the pregnant woman is approximately US$ 2.20. The cost to the end user of ITNs delivered through ANC and MCH with the support of PSI varies from US$ 0.40 in Malawi to US$ 2.80 in Angola. The biggest constraint on these programmes has been the supply of ITNs, which has been predominantly related to funding availability. Voucher systems Voucher systems for the delivery of ITNs to pregnant women, of varying scale and duration, have been implemented in Ghana, Mali, Senegal, the United Republic of Tanzania, Uganda and Zambia and are planned in Ethiopia and Nigeria. The design of these voucher schemes varies but the basic mode of operation is that discount vouchers are given to pregnant women upon attendance at ANC. The voucher entitles them to a discount on an ITN available from retail or other designated outlets. Voucher schemes have been piloted in Ghana, Senegal, the United Republic of Tanzania, Uganda and Zambia. Scaling-up is now underway in both Ghana and the United Republic of Tanzania, in seven out of ten regions in Ghana and nationwide in the United Republic of Tanzania. In Zambia, the vouchers continue to be delivered in ANC “along the line of rail”, which is the commercial hub of the country covering several districts. In six out of nine provinces of Zambia, ITNs are delivered as a direct product through ANC at a subsidized price. Where data are available the level of subsidy on ITNs delivered through ANC using vouchers has ranged between 20%–70%, and pregnant women are paying from US$ 0.60 to over US$ 6.00 for their ITN. Where vouchers are used to deliver ITNs the woman has a degree of choice (variable depending upon local availability), and therefore the amount she pays depends upon the type of ITN she selects. 2.1.2 Maternal and child health clinics Delivery of ITNs to children along with EPI presents an opportunity for reaching children less than one year. Where EPI is delivered through MCH, children aged 0 to 59 months may be targeted, together with their mothers if they are pregnant. As with ANC, ITNs may be delivered along with EPI as a direct product or via a voucher. There are substantially fewer experiences of delivery of ITNs through EPI than there are through ANC. The major limitation has been levels of funding in relation to the size of the target group — that of pregnant women is smaller than that of children under five (approximately 5% versus approximately 15% of the total population). Delivery of ITNs through ANC may be considered to have been a pathfinder for the integrated delivery of ITNs with EPI. The product delivered through EPI varies in some models, such as the NATNETS programme in Tanzania where ITNs are delivered to pregnant women through ANC whilst insecticide for (re)treatment of nets is delivered through EPI at DTP3 and measles immunization contacts. 2.1.3 Packages of health interventions The UNICEF Accelerated Child Survival and Development (ACSD) programme includes a package of interventions: EPI+, IMCI+ and ANC+. The interventions included in each category are for EPI+: EPI, vitamin A, and ITNs; for IMCI+: management of malaria,
pneumonia and diarrhoea; for ANC+: ITNs, intermittent preventive treatment in pregnancy (IPTp), TT, and iron and folate (IFA). This package was initially implemented during 2002 in selected districts of four countries (Benin, Ghana, Mali and Senegal) which were termed “high impact” (HIP) districts. The programme has since expanded to further districts in each of these four countries and to a further seven countries (Burkina Faso, Cameroon, Chad, Gambia, Guinea Bissau, Guinea Conakry, and Niger). These new districts are termed ‘expansion districts’ and include an initially reduced range of interventions including EPI+ and ANC. The total population targeted in the initial districts of the four high impact countries was 4,321,670, with a further 9,123,588 in the expansion districts. The total target population including the seven expansion countries is 20,951,423. Strategies for delivering ITNs through ACSD vary between countries and certain aspects of the programme also vary within countries. ACSD is a district-based strategy, however, in Ghana, as the programme has scaled up there has been some shifting of responsibility to the regional level. In Ghana, delivery points for ITNs include ANC, Child Welfare Clinics, outreach, and sales agents. The original plan was for ITNs to be sold through health facilities, but upon realizing that targeting children through EPI would only reach those aged less that one year, the strategy was adapted to include sales within the community through sales agents. The sales agents sell ITNs to all age groups. However, there is a quota for numbers of ITNs to be sold to the target groups, pregnant women and children under five (95%), and to the general population (5%).The policy within both regions implementing ACSD in Ghana is that health facility staff should not be involved in selling ITNs, but rather that the sales agents should sell ITNs within the health facilities and within the community. In Mali the ACSD programme targets pregnant women and children aged 0–11 months who are given an ITN on completion of their vaccination schedule. Senegal uses the same target group for ITN delivery, but ITNs are delivered using a voucher system. The level of subsidy on the ITNs varies in the three countries and, consequently, so does the amount that the end user needs to pay. In Ghana, ITNs are sold to pregnant women and children under five for 5000 cedis (approximately US$ 0.60), and to the general population at 20,000 cedis (approximately US$ 2.40). In Mali, ITNs are delivered free of charge to the target group, and in Senegal the voucher is worth a discount of around US$1.50 from the cost of an ITN. 2.2 Integration of ITNs with expanded routine child health services 2.2.1 Child Health Week / Child Health Days Although static and outreach facilities have potential for reaching children under1 year with EPI, and therefore with other child survival interventions, the potential is much reduced for children aged 12 months to 5 years. Child Health Week (CHW) is an intensified delivery and promotion of child health interventions through routine services. Where outreach services are generally provided on a monthly schedule thereby reaching each outreach area at intermittent intervals, all areas are covered during CHW. CHW is used as a one-stop opportunity to provide intensified delivery of a minimum package of services together with health education on preventive care to children aged 0 to 59 months. The goal of CHW is both to increase coverage with child survival interventions and to encourage increased use of routine services for these interventions by creating awareness and demand.
Child Health Weeks and Child Health Days are not campaigns; they are “expanded routine”. They generally involve variable packages of child survival interventions such as EPI vaccines, vitamin A supplementation, growth monitoring, and ITN (re)treatment (delivery of ITNs has been less frequent-see below). Other services have included education of caregivers on home management of fevers, promotion of use of iodated salts, awareness creation on HIV/AIDS and promotion of male and female condoms, family planning services, distribution of iron tablets, distribution of de-worming tablets and birth registration. A major difference between CHW and campaigns is that during campaigns children of the target age are immunized regardless of their immunization status, whereas during CHW immunizations are conducted according to health cards. Scheduling of CHWs/CHDs varies; in Ghana CHW is conducted once a year, whereas in Uganda and Zambia it is twice yearly. In Kenya although there is currently no CHW, CHDs are spearheaded by UNICEF at the sub-national level and implemented intermittently. To date CHWs have been used more for the delivery of (re)treatment than of ITNs, although ITNs have been delivered in some districts in both Ghana and Zambia. In Ghana in particular the way in which CHW has been implemented, and to an extent the interventions that have been delivered, have varied considerably between districts. Services have predominantly been delivered free of charge through CHW.
2.2.2 Enhanced outreach Outreach is used to cover areas where access to fixed facilities is low. Enhanced outreach involves broadening the package of interventions available through outreach. In Ethiopia, access to fixed health facilities is low. The MoH is planning to establish a cadre of trained health workers to provide expanded coverage with child survival interventions through the Health Extension Package (HEP). The HEP includes outreach services such as immunization, Vitamin A supplementation, oral rehydration treatment (ORT), family planning and focused Antenatal Care. UNICEF is supporting the distribution of LLINs through this Expanded Outreach Strategy (EOS). 2.3 Integration of ITNs with vaccination campaigns 2.3.1 Measles SIAs Integration of ITN delivery with measles campaigns has been carried out in four countries to date. The first was in Lawra district in the Upper West region of Ghana in 2002; the second in five districts of Zambia in 2003; the third was a nationwide integrated child health campaign in Togo in 2004; and the fourth in Lindi region of Tanzania in 20053. The campaigns have been conducted over a period of three to seven days. In Ghana and the United Republic of Tanzania, measles campaigns fixed and mobile sites were used as delivery points for ITNs. In Ghana, ITNs were also taken to the homes with eligible children who were known not to have attended the campaign. A similar system was used in four out of the five Zambian districts, whilst in the fifth, Kalulushi District, a discount voucher was given with immunization, which could be exchanged for an ITN at nearby retail outlets. During the national distribution linked to
3 At the last minute, the Red Cross also distributed ITNs in three additional districts (Tanga Urban, Pangani, Rufigi) during the same campaign.
the measles SIA in Togo, delivery points included 480 clinics and 149 mobile teams. Both polio and measles vaccinations were given alongside LLINs and mebendazole. All ITNs delivered in the integrated measles campaigns have been free of charge to the target group.
2.3.2 Polio national immunization days There has been less focus on the delivery of ITN through NIDs than through measles campaigns. With only six countries continuing to have endemic transmission of indigenous wild poliovirus, the goal of eradication is fast approaching, and NIDs are therefore a time-limited intervention. The number of countries in which NIDs are conducted, and their frequency within countries, will decrease. Integrated delivery of ITNs with NIDs was carried out in Central Region, Ghana, in October 2004. The campaign was conducted in all districts of the region. Caretakers of children under five were given a coupon which could be exchanged for an ITN with a top-up fee of 20,000 cedis (approximately US$ 2.40). Health staff sell the ITNs at fixed sites and some are now taking ITNs with them on outreach visits so that coupon sales can be made. There are also plans for distributing ITNs during the second round of polio NIDs in Niger in December 2005. 2.3.3 Tetanus toxoid SIAs There is a current push to eliminate neonatal tetanus through the use of TT SIAs. The aim of the SIA is to vaccinate at least 90% of women of childbearing age with three properly spaced doses of TT in high-risk areas/districts where women have not been sufficiently reached by routine immunization activities. The need for multiple doses has operational implications if consideration is being given to integrating with other campaigns such as measles. There are no reports of any TT SIAs which have been integrated with ITN distribution. 2.3.4 Yellow fever As with TT campaigns above, there are no reports of any examples of integrated delivery of ITNs with a yellow fever campaign. 3. Outputs and outcomes of integrated delivery models When ITNs are delivered through health facilities to target groups, it is assumed that they will be used by these target groups, rather than sold on or used preferentially by other members of the household. Where they are not used by target groups or are sold on, output data (ITNs delivered to the target group) will not produce expected outcomes (use by the target groups). A comparative assessment of the outputs and outcomes of these different delivery systems on coverage of target groups with ITNs is not possible as the scale of implementation has varied, and evaluations have been infrequent. Therefore data from specific examples where data are available are presented. ITN coverage encompasses indicators for both household ownership and use amongst target groups, and these indicators are not used consistently amongst the programmes reviewed. The recommended indicator for assessing use amongst target groups is “the proportion of the target group (pregnant women/children under five) who slept under a ITN the night preceding the survey”. There is also lack of clarity on the definition of an ITN among different programmes. An ITN may be a net that was “ever treated” with insecticide or one that is considered as “currently treated”. Currently treated may be
defined as treated within the last six (or 12) months, a pre-treated net bought within the last six (or 12) months, or an LLIN. 3.1 Routine systems 3.1.1 ANC/MCH Data on the total number of ITNs delivered through these systems across SSA is not available. However, the average number of monthly sales in 10 countries where delivery is supported by PSI is 411,500, ranging from 4500 in Rwanda to 150,000 in Kenya. The ANC model of delivery of ITNs in Malawi went to national level between June and December 2002. Between March and December 2004, 1 million ITNs were sold to pregnant women and children under five through ANC in Malawi. A national level household survey was conducted in February 2004 (8) which provides evidence on the outcome of the programme (see Box below).
A household survey undertaken in Malawi in 2004 found that 31.4% of pregnant
women and 35.5% of children under five slept under an ITN the night preceding
the survey. This coverage varied widely between districts, with four out of the
twenty-eight districts in the country having reached the Abuja target coverage of
60% of pregnant women and children under five sleeping under an ITN the
night before the survey. The proportion of target groups sleeping under an ITN
was higher in urban than in rural areas: pregnant women, 49.1% urban and
29.0% rural; and children under five, 50.1% urban and 32.2% rural. Household
ownership (proportion of households with at least one ITN) was also higher
amongst the wealthier socioeconomic groups: 87.8% in the “wealthier”, 59.7%
in the “moderately poor” and 31.1% in the “poorest”. Coverage in target groups
by socioeconomic status was not presented.
The study estimated that 9.8% of the health facility ITNs had leaked into local
The household survey presented in the box above did not include indicators on coverage of childhood vaccinations, so it is not possible to assess the impact of integrated delivery of ITNs and immunization. In voucher schemes outputs are measured by the voucher redemption rate. The rate is calculated using two sets of data: the number of vouchers distributed to the target group and the number of these vouchers that are redeemed by the retailer in exchange for cash or more stock. The redemption rate is then calculated as the proportion of vouchers issued to pregnant women that are redeemed. Redemption rates for voucher schemes in Zambia are presented below:
In the one year pilot from September 2002 to October 2003 in two districts
of Zambia 12,707 vouchers were redeemed, which represented a
redemption rate of 74%. The voucher had a value of US$ 2.
During the roll-out phase of this scheme in 10 districts the voucher value
increased to approximately US$ 3.30. Between October 2003 and October
2004 40,037 vouchers were redeemed. This represented a redemption rate of 66.2%. During a measles campaign in June 2003 in one district of Zambia, 14,792
vouchers were distributed. These vouchers provided 100% subsidy on ITNs and the redemption rate was 99.3%.
Coverage data from the two large-scale voucher schemes in Ghana and the United Republic of Tanzania is not yet available; data from the two pilot districts in Tanzania show variable achievement between the two districts, with coverage generally higher in Kibaha district as compared with Kilosa. Coverage (proportion of the target groups who slept under an ITN the night before the survey) in Kibaha achieved 50% for pregnant women, 27.9% for children under five, and 43.6% for children less than one year (9). The equity ratio was 0.33 for pregnant women and 0.11 for children under five. This compares the proportion of children covered from the lowest socioeconomic groups with the proportion covered in the least poor. An equity ratio of 1 indicates equal coverage in the poorest and least poor, pro-rich bias increases from 1.0 to 0 and pro-poor coverage is indicated with rising values above 1.
3.1.2 Intervention Packages In the four UNICEF ACSD HIP countries baseline coverage (proportion of children under five sleeping under an ITN the night before the survey) in 2001 was compared with mid- term coverage in 2003 (10). The surveys included both specific household surveys carried out to evaluate the impact on coverage of the ACSD programme, and Demographic and Health Surveys (DHS). The ACSD districts are compared with control districts (neighbouring districts where the ACSD programme was not implemented). It is possible therefore to compare the change in coverage in the ACSD districts with that in neighbouring districts without the intervention. Increased coverage was seen in all four countries (Table 3), with both Mali and Senegal exceeding the Abuja targets for coverage in children under five within intervention districts. Table 3. Coverage of children under five with ITNs at baseline and mid-term Coverage Baseline 2001 Mid-term 2003 of under- Control (%) Control (%) fives with ITNs Benin 5.3a
Note, coverage is defined as the proportion of children under five who slept under an ITN the night before the survey.
a DHS surveys Coverage of pregnant women was lower than that of children under five in three of the four countries, the exception being Ghana (Table 4). Coverage reached the Abuja targets in Mali only.
Table 4. Coverage of pregnant women with ITNs at baseline and mid-term Coverage Baseline 2001 Mid-term 2003 of pregnant Control (%) Control (%) women with ITNs Benin 0
Where DHS surveys are used the outcomes of the intervention may be underestimated because the ACSD is district-based and may not cover all districts within a region/province. The DHS present data at the national and at regional/provincial level, while district-based data are not available. The only HIP country where DHS data are used in 2003 was Ghana, where the 21% coverage represents DHS data for Upper East Region. ACSD is operational in all districts of this region and therefore the data are representative. Does integrated delivery of ITNs with ANC increase ANC attendance? It is possible that delivery of ITNs through ANC will have a positive effect on attendance rates at ANC, on the timing of first ANC visit, and on the coverage of TT. Evidence from ACSD districts in Senegal and Mali provides some support to the hypothesis that delivery of ITNs increases attendance at ANC. However, these data are not adjusted for any of the many confounding factors that may influence these outcomes.
In ACSD districts in Mali, the proportion of pregnant women attending ANC ≥3
times per pregnancy increased from 24.9% in 2001 to 54.2% in 2003. This
was 30.7% (p<0.01) greater than the increase in control districts during the
same period (those without ACSD distribution of ITNs). ITN coverage
amongst currently pregnant women increased from 5% to 67.7% in ACSD
districts during the same time period and from <1% to 7% in control districts.
Does integrated delivery of ITNs with ANC increase TT coverage amongst pregnant women? A minimum of two doses of tetanus toxoid are required for protection against neonatal tetanus. Data on coverage with TT2 as compared with ITN coverage in pregnant women are subject to the same problems as that of ANC coverage, which is that a range of confounding factors may have a significant influence on the outcomes. Evidence from the ACSD programme in Mali, however, generally supports an increase in TT2 coverage with that of ITN coverage.
In ACSD districts in Mali coverage with TT2 increased from 22.2% in 2001 to
58.8% in 2003, in control districts coverage increased from 20.3% in 2001 to
38.9% in 2003. This represents an 18% (P=0.05) greater increase in ACSD
Assessment of the proportion of pregnant women adequately covered with sufficient doses of TT using routine data is problematic. Firstly, many women lose their ANC cards
and/or cannot recall whether they have been vaccinated. They may therefore be given several “first” doses. Secondly, women who have received a full schedule of vaccinations during a previous pregnancy will not be revaccinated. In Upper East Region of Ghana, whilst ANC attendance has risen, TT vaccination routine data showed a decrease. These problems mean that it is difficult to evaluate the impact of delivery of ITNs through ANC on TT coverage data. 3.2 Expanded routine Child Health Weeks and Child Health Days are a relatively new delivery mechanism for child survival interventions, and countries where they are currently implemented are still developing their systems for planning, implementation and monitoring. During the December 2003 CHW in Zambia 69,389 ITNs were (re)treated. This represents data from thirty-five out of the fifty-nine districts that reported on CHW activities, the remaining fourteen not reporting on ITN (re)treatment. Using distribution data from the National Malaria Control Centre (NMCC) this represents 22% (69,389/308,856) of the nets in the thirty-five districts. Administrative data on numbers of ITNs are, however, prone to problems (section 5.2.6). 3.3 Campaigns Numbers of ITNs distributed during the four experiences of integrated delivery of ITNs and immunization through measles campaigns, and the outcomes, are presented in Table 5. In Ghana where implementation involved one district only, the Abuja targets were met (see note below on indicators), in Togo and Zambia where implementation was on a larger scale, coverage through the campaign was approximately two thirds of that required to meet the Abuja targets for children under five. In Zambia coverage due to campaign plus other delivery systems increased coverage to the target levels of 60%. Results from the United Republic of Tanzania are not yet available. The coverage indicators used in the integrated distribution of ITNs with measles campaigns in Ghana and Zambia vary from the standard RBM-recommended indicators. The indicator used in these campaigns is “the proportion of households where the index child slept under an ITN the night before the survey”. The denominator is therefore based upon the number of households, whereas the standard RBM indicator uses a denominator of the number of children under five surveyed. The index child is defined as the youngest child who usually sleeps in the household who was at least six months old at the time of the campaign. During the Togo campaign standard indicators were used to evaluate the outcomes of the distribution. Table 5. Outputs and outcomes of ITN delivery integrated with measles campaigns Country Scale Number of ITNs population distributed campaign campaign (children coverage coverage with under-five with any ITN campaign ITN
Ghana and Zambia = “proportion of households where the index child slept under an ITN the night before the survey” Togo = “proportion of children under five who slept under an ITN the night before the survey” The disparity in coverage of ITNs amongst children of differing socioeconomic groups in each campaign was assessed by the equity ratio (Table 6).
Table 6. Equity across socioeconomic groups of coverage of ITNs delivered through integrated measles campaigns Country Equity ratio Pre-campaign ITN* Post campaign ITN Post campaign any net Urban Rural Total Urban Rural Total
a household ownership (all households) In Ghana pre-campaign coverage was assessed using exit interviews. The sample is therefore subject to bias which makes it not directly comparable to the representative sampling used in the post-campaign survey. In Zambia, pre-campaign ITN ownership was assessed by questioning care-takers one month post-campaign. As mentioned above coverage data from Ghana and Zambia use non-standard indicators, which means that the data on equity ratio are also not directly comparable with that from other studies. In Togo, the equity ratio was reported using household ownership, not use by children under five. During the integrated delivery of ITNs with polio NIDs in central region Ghana (October, 2004), coupons redeemable against an ITN with the addition of a 20,000 cedis top-up were delivered to 297,133 children. This compared to 554,582 children immunized. Monitoring data were used to estimate outcomes as 58.4% of children given a coupon, and 110% immunized. Findings from an evaluation are awaited. Outcome data are summarized in Table 7, and overall findings include:
¾ data on outputs and outcomes from routine delivery systems are less readily
available than that from the integrated measles campaigns;
¾ more ITNs have been delivered to date through routine integrated systems than
¾ high levels of coverage (reaching the Abuja targets) have been achieved at
district level, by routine and by campaign models;
¾ high levels of coverage (reaching the Abuja targets) have not been achieved at
the national level by any single model of integrated delivery of ITNs with immunization;
¾ current data show that integrated delivery of ITNs through measles campaigns is
achieving more equitable coverage than integrated delivery through routine systems;
¾ the earlier measles campaigns did not use the RBM standard indicators for ITN
coverage, making comparison with other delivery models impossible.
Table 7. Summary of coverage outcomes of integrated delivery models across countries Country % coverage of Coverage of pregnant Equity ratio of coverage under-fives with women with ITNs (district range) [non-standard (district range) [non-standard indicators] [non-standard indicators] indicators]
Note, these are coverage data from household surveys in areas where the specified interventions were implemented, they do not necessarily represent direct outcomes of the specified intervention alone
b “proportion of households where the index child slept under an ITN the night before the survey” where index child is the youngest child in the household who was above six months of age at the time of the survey
c “proportion of households with at least 1 ITN” not based on use
4. Considerations in selecting strategies for integration
4.1 Target groups The target groups for malaria control in most countries of SSA are those who are most vulnerable to severe outcomes of malaria, which are children under five and pregnant women. In areas of low endemicity typical of Asia and some areas (particularly highland) of SSA, all age groups are at risk of severe outcomes and target groups for malaria control may be selected based on geographical risk. Although in most of Africa the socioeconomically vulnerable are not often specifically targeted, this is usually because of the difficulty of finding effective ways in which to do so. The presentation of coverage of target groups amongst socioeconomic quintiles with the specific purpose of demonstrating that the poorest have been reached is becoming more common due to the globally driven poverty agenda. Target groups for immunization services are more varied depending upon whether these are delivered through routine services or campaigns, as described above, and also upon the delivery point (Table 8). Table 8. Comparison of target groups for malaria control and immunization services Intervention Main
Children under-five and women of childbearing age
All population EXCEPT children <9 months and
It is clear from Table 8 that the malaria target groups (pregnant women and children under five) are not represented by any single immunization delivery system, and therefore a combination of immunization systems is needed for ITN delivery in order to reach malaria target groups. As discussed previously, it may be possible to reach beyond the main target group through some of these systems (section 2.1). It is important to identify those excluded through each system in order to ensure that the target groups are covered through complementary systems. For example, if ITNs are delivered through both ANC and MCH it is possible to reach the entire ITN target group. Where ITNs are delivered through measles SIAs and ANC, children <9 months are excluded. A proportion of children under one year of age may however be reached if they sleep with their mother who has received an ITN through ANC. Are “non-targeted” target groups covered? Intra-household patterns of use of ITNs may result in those amongst the priority target group for ITNs but not specifically targeted by the delivery system being a beneficiary of the intervention. It is likely that in some instances ITNs targeted at pregnant women are used to cover children under five and vice versa. This is likely to be influenced by a range of factors and national level evidence is needed. It would be useful to assess the proportion of “non-targeted” target groups who benefited from the intervention when
evaluations are undertaken, as in the voucher tracking study in Kilosa and Kibaha districts in the United Republic of Tanzania (11). Co-use of ITNs by target groups:
Findings from a voucher tracking study carried out for the Tanzania pilot
voucher scheme found that 91.2% and 95% (Kilosa and Kibaha districts,
respectively) of the women who purchased an ITN with a voucher slept
under the ITN whilst pregnant. In these households 29.5% and 13.5% of
children under five and 51.3% and 47.3% of children under one were
One ITN per child or one ITN per caretaker? Where ITNs are delivered through routine services the target is the individual pregnant woman or child (although policy and practice may differ — see box below). During ITN delivery integrated with immunization campaigns the policy may be to target caretakers of children under five with an ITN rather than the individual child. That is one ITN per caretaker, rather than one ITN per child. In the Tanzanian and Zambian campaigns, the target was one ITN per child under five, which resulted in many households receiving several ITNs.
In Kenya, ITNs are targeted to pregnant women and children under five through ANC and CHC. Although the policy is one ITN per pregnant woman and child under five, some nurses are imposing their own criteria and limiting the ITNs to a maximum of two per household. In the 2002 integrated measles campaign in Ghana ITNs were targeted at one per caretaker of children under five, in the Zambian 2003 and Tanzanian 2005 campaigns the target was one per child under five. In the national integrated distribution in Togo 2004 there was confusion on the target group, resulting in some delivery sites giving one ITN per child and others giving one per caretaker. The health workers in less accessible areas were more likely to give one per caretaker as they were worried about running out of ITNs.
A policy decision should be taken early in planning and consensus assured amongst partners. Factors such as average household size, the size of and number of rooms within houses, and knowledge of intra-household sleeping patterns are likely to provide an evidence base for such a policy decision. Where the target group is one per child, an option is to place a limit on the number per household, with the maximum number determined depending upon the local context. Experience during the last few years with targeting of subsidies has shown that health workers frequently impose their own criteria despite policy guidelines. Such health staff-imposed systems of work are usually due to shortage of supplies or perceptions of shortage of supply due to delays in replenishment. Careful monitoring is needed to determine actual practices, and their reasons, so that remedies can be found. Can children 0–8 months be targeted with ITNs during measles SIAs? Measles campaigns target children 9–59 months or 9 months to 15 years depending on the burden of disease. Consequently children 0–8 months are excluded from receiving an ITN. Decisions on who is targeted in integrated campaigns are a balance between
intervention coverage amongst priority groups, logistical realities and practical problems such as the training of volunteers. In Kenya the measles campaign planned for 2006 will include delivery of ITNs in forty-two districts and OPV in 14 districts. OPV has a target group of 0–59 months. Delivery of ITNs to children 0–59 months is planned in the districts with integrated OPV delivery and to 9–59 months where OPV is not included. 4.2 Current coverage of ITNs and delivery systems for ITNs Coverage of ITNs is measured through household surveys. Within the context of scaling up coverage to the national level, nationally representative household surveys are needed. Demographic and Health Surveys (DHS)4 and the UNICEF Multiple Indicator Cluster surveys (MICs)5 are the main source of such data in terms of the monitoring of national progress towards global targets. However, they are undertaken infrequently, approximately every five years as discussed above. In most countries of Africa, there is still inadequate access to a supply of ITNs for target groups. Insufficient funding means that, where the commodities are available, choices have been made on where in the country they would be targeted. Consequently, although there is a wide consensus on targeting of the biologically vulnerable, with the exception of the integrated measles campaign in Togo, efforts have in practice been targeted within defined geographical areas. In Malawi, integrated delivery of ITNs through ANC has recently gone to national scale; this was achieved through expansion of a small-scale pilot programme with the availability of increased funding. Geographic targeting of ITNs within endemic areas has been a consequence of insufficient funding (with the exception of pilot projects and disaster response) and therefore scale of deployment. There have been a range of criteria upon which these selections have been made, and these have frequently been based upon poverty levels and health indicators. In order to ensure that ITNs are available to target groups throughout the country it is useful to conduct a district-by-district analysis of ITN delivery systems and the target group reached by each of these systems. Once this analysis has been carried out, then access by target groups at sub-district level may be assessed and need for any additional strategies to focus on the “hard-to-reach” determined. In this way geographical inequities may be addressed.
In Ghana twenty districts (two from each of Ghana’s ten regions) were
designated “RBM focus districts” and ITNs donated to the national malaria
control programme by partners distributed within these districts. Ghana
UNICEF selected countries for their ACSD programmes in West and
Central Africa based on high under-five mortality rates (U5MR). In Ghana,
the selection of regions in which the UNICEF ACSD package was to be
implemented was based on those with the highest levels of poverty. Within
these regions selection of districts seems to have been more historically
based. Central Region was selected for integrated ITN and polio delivery
4 ORC Macro Demographic and Health Surveys (DHS). ORC Macro, Calverton, MD. www.measuredhs.com
5 UNICEF Monitoring the Situation of Women and Children www.childinfo.org
4.3 Current coverage of EPI Although EPI is undeniably a public health success story, with more than 70% of the world's targeted population being reached with immunization6, coverage has stagnated or decreased in some countries over the last decade, mostly in the poorest countries of Africa. This is due to the weakening (and in some cases collapse) of the health services. Coverage is particularly low in some countries of West Africa, and more worrying is the disparity in coverage levels within countries. For example, in Niger coverage of all six EPI antigens in children one year of age varies between 9.9% and 76.2% in the different provinces of the country. Disparity in Ghana is less acute than in Niger, ranging from 48% in northern region to 82.3% in Volta Region, but is still highly inequitable. This will limit the reach of ITNs delivered through the same system unless ITNs become valued by the population to the extent where ITN delivery increases coverage levels of EPI (section 3.1). 5. Issues in planning, implementation, monitoring and evaluation
Many of the opportunities and challenges presented by integrated delivery of ITNs and immunization cross-cut the various delivery models, and some are specific. Cross- strategy issues in planning and implementation are presented in section 5.2. Although the issues and experiences presented are by no means exhaustive, they outline some important considerations when planning and implementing integrated delivery of ITNs and immunization through the various models. 5.1 Opportunities and challenges of specific models Opportunities and challenges of specific models are summarized in Table 9. 5.1.1 Routine delivery Maintaining a constant supply of ITNs has been a major problem in all integrated routine delivery systems. Factors contributing to this include problems in planning, forecasting, and monitoring but also in insufficient and poor continuity of funding. Evidence of the effect of stock-outs of ITNs on the delivery of interventions with which it is integrated, (such as EPI) has not been documented. However, there is anecdotal evidence from Kenya that attendance at both ANC and MCH is increased at facilities where ITNs are available and decreased at those where they are not. ITNs are not available in all health facilities in malaria risk areas in Kenya. Supplies of ITNs are dependent upon the health facility purchasing an initial seed stock of ITNs. Where facilities have done so, they report that people are travelling further to attend these facilities rather than attending those that are closer, but without subsidized ITNs. The ANC delivery model for ITNs is relatively new in Kenya but it should be possible after a longer period of implementation to use routine data on ITNs and attendance figures to assess the impact of availability of subsidized ITNs on attendance for ANC and EPI. There are many examples of health staff devising their own policies and imposing their own eligibility criteria for ITNs. Experience from the voucher scheme in Volta Region, Ghana, shows that enhanced training can effectively deal with some of these problems. The important issue is to ensure that regular monitoring is in place to identify and respond to problems.
6 See GIVS pages 4 and 5 for the latest coverage summaries www.who.int/vaccines/givs
In the Volta Region pilot voucher scheme, Ghana, 50% of health facility staff
interviewed three months post-implementation had decided at least once not to
give a voucher to a pregnant woman because she was “not ready with money”.
This was addressed by retraining and increased supervision and was not
identified as a problem in the monitoring surveys six months later.
Current experience is that financing strategies may vary within different areas of a country and also over time within the same area. In Ghana the pricing strategy of the ACSD programme does not follow that of the national programme. The national programme delivers ITNs in at least two districts per region across all ten regions of the country. These ITNs are sold from ANC clinics to pregnant women for 20,000 cedis. There are joint ACSD and National Malaria Control Programme (NMCP) districts, where these are clearly non-complementary strategies. The NMCP ITNs are sold by ANC staff from the health facilities, whereas the policy of the ACSD programme (and of the two regions in which they are implemented) is that health facility staff provide ITNs for free. Competing pricing policies have the capacity to impact not only on the uptake and use of ITNs, but also, where integrated with the delivery of other health interventions, on the uptake of other interventions. In addition, differing financing models and incentive schemes may also impact upon the motivation of health staff in delivering them.
In Kenya, ITNs are sold to health facilities for approximately US$ 0.40, they are
then sold by health staff to pregnant women and children under five for US$ 0.66.
The health facility uses US$ 0.40 to restock and then is able to accumulate the
remaining US$ 0.26 from each ITN sold to increase their stock levels or for
general needs within the facility. Uses have included transport of vaccines for
CHCs in those facilities with no refrigeration facilities, and for transport of
emergency referral patients. There are suggestions that Kenya will move towards
a policy of free ITNs for all target groups and, under the current circumstances, a
change in strategy to delivery of free ITNs to the target group is likely to create
problems in terms of health staff motivation.
Difficulties in implementation of voucher schemes have been encountered in health facilities where ANC staff have previously sold ITNs (12) This is mainly due to the fact that selling ITNs has included an incentive for the midwife, whereas the current practice of delivering vouchers does not. Direct product or voucher? As both direct product and voucher systems for delivering ITNs on a large scale are relatively new, there are still many questions to be answered. However, the basic assumption in selecting between the strategies is:
¾ Direct product delivery through ANC is most appropriate where ITN commercial
sector supply and distribution is underdeveloped.
¾ Voucher system delivery is most appropriate where there is an active (or the
likely potential for an active) commercial sector for ITNs.
The advantages of voucher systems over those of direct product distribution through health facilities are that: 1) there is less of a burden on health staff, particularly in comparison to where health staff sell ITNs; 2) the retail sector is not “crowded-out” by
direct public sector distribution; 3) the retail sector is “crowded-in” resulting in greater accessibility of ITNs for non-target groups. Disadvantages may include: 1) supply may not have the capacity to meet demand; 2) supply relies on the pull created by demand; 3) changes in markets and currency fluctuations may result in increased or decreased value of the voucher; 4) redemption systems need managing; 5) there is potential for fraud. Also, the effort required to design and implement a national voucher scheme should not be underestimated.
Table 9. Opportunities and challenges of models of integrated delivery of ITNs and immunization Model Category Opportunities Challenges
- reaching the geographically and economically
- provides a constant source for target - maintaining constant supply of ITNs group (depending upon constant
- collaboration between Malaria, and Reproductive
Health Departments - health staff imposing their own eligibility criteria - consolidating financing systems
- reaching the geographically and economically
- provides a constant source for target disadvantaged group (depending upon constant
- collaboration between Malaria, EPI, IMCI and Child Health Departments
- increasing EPI coverage - building an integrated sustainable system - capitalizing on contact opportunities - provides a constant source for target group (depending upon constant supply)
- monitoring and evaluation of the impact
- maintaining constant supply of ITNs and vaccines - monitoring and supervision of activities
- reaching children born in the 3 to 5 years
- balance of timing of the campaign to optimize
both malaria and immunization outcomes - target group confusion where multiple interventions are delivered - insufficient training of volunteers - collaboration between EPI and malaria - timely supply and distribution of ITNs
- transportation of ITNs by volunteers conducting
Where vouchers have been used to deliver ITN subsidies during campaigns they provide a means of logistical ease for delivery of subsidy compared to the ITNs themselves, which are bulky and heavy. Where vouchers are exchanged for an ITN at a health facility, the positive impact on the commercial sector, one of the objectives behind the design of voucher schemes, is lost.
5.1.2 Expanded routine In most of the countries where CHWs are currently implemented they are a relatively new delivery model, such that systems are still being developed and lessons being learned. They generally attempt to enhance the delivery of interventions, health messages and social mobilization over a short period of time in a similar fashion to that of campaigns. However, in both Ghana and Zambia the extra funding available during the initial CHWs has now stopped and districts need to fund the activities through their district budgets. This has implications for the level of human resources that can be mobilized to increase the concentrated delivery required. Countries are attempting to effectively institutionalize and sustain CHWs within their current budgets, and are therefore striving for a sustained system. Monitoring systems have been developed in Zambia, including the use of tally sheets similar to those for EPI. Outputs for each of the interventions are recorded on the tally sheets. This is equivalent to the number of nets (re)treated or the number of treatment kits distributed. During CHW in Ghana, some facilities treat nets with insecticide and others distribute treatment kits. In both Ghana and Zambia some health facilities have distributed ITNs during CHW and the decisions have been made at district level. 5.1.3 Campaigns Even in countries where malaria is endemic there may be considerable seasonality in transmission and levels of transmission based upon local climate conditions such that malaria transmission may vary substantially from year to year. Malaria control activities can also substantially alter the transmission season. Increased transmission is usually closely linked to the rainy season. The use of mosquito nets is largely driven by mosquito abundance and biting intensity, so that ITN use is generally linked to the rainy season when mosquitoes breed. Although both polio and measles can occur throughout the year, they each have high and low transmission seasons. In tropical climates the high transmission season for polio is usually during the rainy season, and, in temperate climates, during summer and early autumn. In tropical climates, measles transmission increases after the rainy season. Both polio and measles campaigns are usually conducted during low transmission seasons, that is, during the dry season when roads are accessible. The optimum time for delivery of ITNs is just before the rainy season. In countries where measles transmission is relatively low, as is becoming the case in a growing number of countries, it should be possible to be able to interrupt transmission during the ‘high season’. Therefore considerations for timing of the campaign would be dominated by logistical considerations balanced by the optimum time for delivery of ITNs. Under these circumstances the risk would then be the negative impact of the advent of cases after the campaign because the virus had already been incubating. As polio NIDs in Ghana involve house-to-house visits and ITNs are both bulky and heavy, a coupon was given with immunization rather than a direct product. The coupon
entitled the recipient to an ITN for 20,000 cedis (approximately US$ 2.40) at health facilities and other ITN depots. One coupon was given per caretaker of children under five. Efforts were made to ensure that hard-to-reach communities were able to access ITNs with their coupons. Where caretakers reported that they had lost their coupons, they were told that they could buy an ITN for 20,000 cedis anyway. The coupons were therefore not valued in the same way that they are in the Volta Region voucher scheme. The vouchers represent actual money, and they provide a discount. There were reports that some of the volunteers were asking caretakers if they could pay the 20,000 cedis before giving them a coupon, that is as in Volta Region, where they ‘ready with money’. The volunteers not only move from house to house during polio NIDs, but also more generally within the community, taking any opportunity possible to immunize a child. Some children are immunized whilst unaccompanied by an adult (schools, marketplace), so these children were not supplied with coupons as the policy was that coupons be given to caretakers. 5.2 Cross-strategy issues in planning, implementation, monitoring and evaluation 5.2.1 Forecasting and procurement Obtaining and maintaining an adequate supply of ITNs has been a problem for most if not all integrated programmes and delivery systems, both routine and campaign, and through both the public and private sector. This has been amplified with the introduction of, and preference for, long-lasting insecticidal nets (LLINs). UNICEF Ghana intended to deliver LLINs through the ACSD programme and MoH through the Central Region polio NIDs, but switched to pre-treated and bundled nets due to a global shortage in LLIN supply. Procurement of ITNs requires a lead time of several months up to one year, to allow shipment and customs clearance. Supply, particularly of LLINs, cannot be assured on a short time scale. This is in part due to increased global demand but also to the often complex importation procedures. Some manufacturers supply regular lead time information to their distributors and other partners. For routine systems where funding is assured over a period of time, forecasting and regular standing orders with manufacturers can help to reduce stock-out problems. Bulk institutional orders such as those from countries receiving Global Fund for AIDS, Tuberculosis and Malaria (GFATM) funding can cause problems in maintaining availability on the global market. In the current situation where demand is higher than supply capacity, decisions will be taken on which country gets the available LLINs and which ones do not. In Ghana, the pilot ITN voucher scheme in Volta Region experienced stock-outs of approximately three months due to the large demand for ITNs from a scheme in urban areas of other regions. This was partly due to much higher demand for ITNs than had been anticipated, but also to inadequate financial and procurement planning by the commercial partners. Supply of ITNs has been a problem during integrated campaigns. This has sometimes been due to late securing of funding for the ITNs, such as in Zambia, and sometimes simply due to late planning. Late supply or non-availability of ITNs should not delay the implementation of immunization campaigns. In Ghana, the ITNs for the integrated distribution with NIDs arrived after the campaign. The coupon recipients had to wait approximately two weeks before they could exchange their coupons for an ITN. Experiences from the integrated measles campaigns in Ghana, Zambia and Togo
suggest that it takes six to eight weeks for the ITNs to get from the manufacturer to the port. In Tanzania, the ITNs were procured from a national supplier by UNICEF.
5.2.2 Planning All partners need to be involved from the beginning of the planning process; this facilitates consensus building and determining the most appropriate roles of each partner organization based on institutional strengths. Strong partnerships at the planning phase are more likely to facilitate a pooling of resources, both financial and human. Where programmes involve a large number of partners a co-ordinating committee will help to facilitate consensus building, but this committee must be linked to a final decision-making authority in the MoH. A policy decision on targeting of each child under five with an ITN, or targeting caretakers of children under five, was unclear amongst those delivering the ITNs in the Togo integrated measles campaign. This caused confusion to the extent that some volunteers gave an ITN per child and others an ITN per caretaker, the decision often being dependent upon perceptions of supply levels. Partnership issues need to extend to include issues in integrating and institutionalizing new initiatives into routine systems. Initially, the UNICEF ACSD programme in the two regions of Ghana was perceived by regional and district health staff to be separate from other routine activities because it was funded separately. Over the last couple of years the package has been integrated to the extent that it is perceived as a component of the routine system. CHW was implemented for the first time in Ghana during May 2004. Planning at the national level involved the managers of several programmes including EPI, Nutrition, Child Health, and Malaria Control. The planning period at the national level was sufficient; however, the time remaining before implementation for training at the regional and district levels was not long enough to allow for adequate advocacy and institutionalizing of the intervention at these levels. This resulted in different understanding, interpretation and implementation of the interventions between regions and between districts. The issue for the second year of CHW is to institutionalize the programme. Partnership experiences and lessons learned are discussed in greater detail in section 6. 5.2.3 Effective targeting and tracking Strategies to deliver ITNs to pregnant women through ANC, as with all targeting strategies, need to be evaluated for targeting efficiency. Careful tracking in Kenya found that 20% of 70,000 ITNs distributed to District Health Management Teams (DHMTs) in thirty-five districts, for delivery to pregnant women through ANC, went to non-target groups (13). Eighty percent of the leakage (defined as not delivered to pregnant women) of nets from the system in this study was at the DHMT level, that is, before the ITNs reached ANC. Minimising leakage is likely to be significant in ensuring complementarity of alternative delivery systems, particularly, but not exclusively, where the commercial sector is involved. The RBM Strategic Framework for Scaling-up ITNs recommended the use of sustained and targeted subsidies. Untargeted subsidies are not recommended. In Ghana untargeted subsidies are provided by the agents who sell ITNs door-to-door in the ACSD districts. The subsidy delivered through the sales agents to the general population is the same as that reserved for target groups (pregnant women and children under five) by the MoH. In Zambia, blanket subsidies are delivered by malaria agents
through the Community Based Malaria Prevention and Control Programme (CBMPCP), the level of subsidy being greater than that given to pregnant women through subsidies at ANC. Delivery of highly subsidized ITNs to non-target groups and leakage onto open markets have two negative impacts: 1) they divert resources available for subsidies from the identified vulnerable groups; 2) they undercut commercial market prices and has a negative impact on developing an enabling environment for the development of sustainable private sector distribution systems. The use of sales agents in the ACSD programme has the capacity to increase sales, but, where incentives are included with each ITN sold, they also have the potential to increase the proportion of sales to non-target groups.
5.2.4 Geographical boundaries As mentioned above, geographic targeting of ITNs can introduce inequities which have a significant impact on implementation. During integrated ITN and immunization campaigns, careful forecasting is vital to ensure sufficient supplies of vaccines and ITNs. In Zambia, attendance at the measles campaign from other non-target districts caused logistical problems in supplies of vaccines. Cross-national border attendance at the integrated measles campaign was also seen in Togo from Volta Region, Ghana. Monitoring of the voucher scheme in Ghana found that some pregnant women returned their vouchers to midwives as they had already received an ITN from the campaign in Togo. The impact of ITN stock-outs due to underestimates of needs or to inflated attendance from outside of the geographical target area becomes more serious where it negatively impacts on immunization coverage. At some of the distribution points in Zambia vaccinations stopped when the stock of ITNs ran out. In Upper East Region of Ghana some communities (in Bawku West) refused immunizations because they wanted to receive ITNs with immunization in the same way that their neighbours in Togo had. These are major issues to take into consideration for Kenya where ITNs will not be delivered in all districts with the measles campaign in 2006, but just in forty-two malaria endemic and epidemic districts. 5.2.5 Communication messages Where target groups vary between interventions there may be problems in combining messages so that the population are clear on their entitlement. In Zambia the messages needed to effectively communicate that measles vaccination was for those aged 9 months to 15 years; vitamin A was for under-fives (6–59 months), mebendazole for those two to five years and ITNs to under-fives. There was ‘civil unrest’ in some of the campaign sites in Zambia when those attending who were five years and over were not given an ITN. There were reports that some people started to change the birth date on their child’s EPI card. Clearly the range of interventions delivered and the consequent diversity of ages for each intervention creates difficulties in designing effective communication messages. Perhaps more importantly, this will also impact upon the amount of training needed for the volunteers who are administering the interventions. 5.2.6 Monitoring and evaluation Household surveys such as the DHS and the MICs assess national and regional/provincial coverage; however, they do not incorporate tools to assess coverage
achieved by alternative delivery systems. Where there is more than one delivery system in operation, and where ITNs are also delivered through the private sector, it is difficult to accurately assess the proportion of any coverage recorded that was due to a specific delivery system. It is however possible to do this with the addition of a simple question on “source of the ITN” to survey tools, together with careful tailoring of response categories on the type of ITN acquired. Coverage of interventions in the DHS and MICs surveys is assessed at the national level, but the sampling scheme enables breakdown of coverage data to regional/provincial level. District level comparisons are not possible, which means that where programmes are delivered on a district level scale (such as the UNICEF ACSD programmes in West and Central Africa) it is difficult to measure their impact in terms of coverage through these surveys unless all districts within a province/region are involved. Thus, the impact of the programme will be diluted when assessed at the regional/provincial level. The addition of ITN coverage questions to EPI cluster surveys has been suggested both to enable assessment of coverage at the district level and to increase the pool of surveys providing ITN coverage data. A question on ITNs was added to a national EPI cluster survey in Ghana and a question on mosquito nets on the EPI cluster survey in the Republic of Timor Leste (December 2004) which enabled the collection of data on coverage to district level. It is vital, however, that where such integrated monitoring is undertaken standardized ITN questions (such as those used by the DHS) are used, as experience has highlighted a host of problems and misinterpretations due to lack of standardization of survey tools between countries. Where routine data are used to estimate coverage the denominator is likely to introduce inaccuracies. Population estimates may present problems when calculating vaccination coverage, but on the whole the numerator is less open to inaccuracies. When a child is vaccinated then they have received the vaccine. This is not so when a pregnant woman or a child under five is given an ITN. ITNs are easily transferable and may not necessarily be used regularly even when they are kept. This problem is compounded further when estimating (re)treatment rates. In Ghana UNICEF were disappointed with the (re)treatment rates obtained through the ACSD programme in Upper East Region. On the basis of calculations of the number of ITNs sold through the scheme and the number of (re)treatments undertaken, UNICEF perceived that the programme was not successful. However, data from the 2003 DHS indicates that in this region 24.1% of children under five slept under a net the night before the survey and that 21.0% of children under five slept under an ITN the night before the survey. Therefore 87.1% of the children sleeping under a net in the region were sleeping under ”currently treated” nets. This suggests a very successful (re)treatment service is in operation. The reason for the apparent lack of success of the (re)treatment service when measured through monitoring data is likely to be due to an overestimation of the number of nets in the region. The very high subsidy on ITNs through the ACSD scheme has resulted in an organized system of transfer of ITNs to the open markets in other regions of the country. Most household surveys such as the DHS and MICs are conducted during the dry season, mainly for logistical reasons, as access to remote places is easier. As the use of
ITNs is more common during and just after the rainy season, this will result in an under- estimate of use. It is not clear whether this may also affect reported ownership.
Table 10. Summary issues and suggested actions in planning and implementation Subject Issue Suggested
- Include coverage of “non-targeted” target group within evaluations (e.g. ITN coverage of < 9 months by measles campaigns)
season compared with those that do not, in terms of retention rates, use of ITNs, maintenance of insecticidal effectiveness, and the longevity of the net itself.
systems already in place Routine systems offer promise of
of equity of coverage Pricing policies of different delivery
funding - early procurement for campaigns
the targeting is only loosely implemented, there is increased capacity for antagonistic interaction with other delivery systems
Subject Issue Suggested
Health staff impose eligibility criteria
absolutely no other eligibility criteria should be imposed
the district, regional, or national level
diverse target age groups effective communication messages are difficult to design
in the community, this may be done in proportion of households where this the absence of a caretaker, where
this is the case the caretaker does not receive an ITN
6. Partnerships 6 .1 Key partnerships spearheading integrated campaigns Partnerships for integrated immunization and malaria programmes in the review countries have without exception been broad and complex, involving international, regional, national and sub-national partnerships, including both public and private sectors, typically involving several government departments, development partners, funding agencies, NGOs, commercial partners and the community. 6.1.1 Global level The two main global partnerships behind the integration of immunization and malaria control are the Roll Back Malaria7 and Measles Partnership.8 WHO and UNICEF are members of both partnerships and have signed a joint statement formalizing the strategy for integrated programming for malaria control and immunization. The two organizations are reviewing experiences to inform the development of a common framework for action for discussion with national programmes and partners and will request financial and planning institutions to consider this initiative in health sector planning activities at country level. The aim is to integrate a holistic package of malaria control interventions
7 RBM founding partners are WHO, UNICEF, World Bank and UNDP
8 The Measles Partnership is composed of the American Red Cross (ARC), CDC, the UN Foundation, WHO and UNICEF in collaboration with IFRC www.measlesinitiative.org
with both routine and campaign-style immunization activities. UNICEF is already leading the ACSD initiative in several countries in West Africa, which provides ITNs to infants at the time of routine immunization. The Measles Partnership has spearheaded the integration of ITN delivery with measles campaigns, which have so far been undertaken in Ghana, Zambia, Togo and the United Republic of Tanzania. While both these partnerships are well established — RBM was established in 1998 and the Measles Partnership in 2001 — this is the first time that the two partnerships are working together towards a common goal. The impetus for combining the delivery of ITNs with immunization has come from the commitment to accelerate scale-up of malaria control interventions to reach the Abuja targets. Global partners now urgently need to develop a coherent partnership approach which integrates malaria control (i.e. more broadly than ITNs) with both routine and campaign immunization efforts, based on a rational analysis of risk, target groups and delivery strategies, including mechanisms for partnership support and coordination.
6.1.2 Regional level These global partnerships are mirrored at the regional level. Both WHO and IFRC have offices in Harare, and UNICEF has regional offices in Nairobi and Dakar, and all have provided technical assistance to countries that have undertaken integrated child health initiatives. To a large extent the technical assistance delivered by each organization has been complementary rather than a duplication of effort, but as integrated malaria and immunization initiatives are rolled out across Africa there should be a clear strategy and plan for technical assistance from all partners to avoid overwhelming countries. The organization by WHO of the regional workshop to develop a framework for integrating child survival interventions with immunization activities in Harare, 2–5 May 2005, is a key milestone in formalizing the partnership strategy for this approach. In addition to the international partners, including WHO, ARC, Canadian RC, IFRC, GFATM, RBM, CDC, UNICEF and bilateral funding agencies, the workshop also involved national programme managers for EPI, IMCI and malaria from 8 African countries. The workshop provided an excellent opportunity to discuss partnership planning and coordination in the African region. A follow-up workshop on Monitoring and Evaluation of Integrated Distribution of ITNs with Other Child Survival Interventions was undertaken in August 2005 in Harare. 6.1.3 Country level At country level, partnerships have evolved on the basis of a common commitment to integrated approaches to immunization and malaria control, following the earlier reasonably successful demonstration projects in Ghana in 2002 and Zambia in 2003. An increasing number of partners have come forward to lend their support to the integrated campaigns. For example, there were more than twenty-five different partners involved in Togo’s National Integrated Child Health Campaign, including six departments of the Ministry of Health (Public Health, Primary Health Care, Epidemiology, Malaria Family Health and Health Education), five national Red Cross societies (American, Canadian, New Zealand, Norwegian and Togo), two UN agencies (UNICEF and WHO), IFRC, GTZ, four NGOs (Plan Togo, Freedom from Huger, PSI, RAM), CDC, three academic institutions (LSTM, LSHTM and Geneva University), the commercial sector (Vestergaard-Frandsen, DHL and Sanofi-Synthelab) and at least four funding agencies (CIDA, NORAD, GFATM and the UN Foundation (UNF)). This large number of partners
Accessed from 18.104.22.168 by nEwp0rt1 on Thu Nov 24 23:35:39 EST 20113600 Isotretinoin / Official Monographs sponses. Calculate the per centage of isotretinoin dissolved by Assay— [NOTE—Protect the System suitability solution, the Stan- dard preparation, the Assay stock preparation, and the Assay Diluent— Heat 0.1 N sodium hydroxide to about 60 ° to 70 °. Cool it to
Research Note: Price Discrimination as an University of Chicago, 1101 East 58th Street, Chicago, IL 60637 Sloan School of Management, M.I.T., E56-305, 38 Memorial Drive, Cambridge, Massachusetts 02139 [email protected] • [email protected] Abstract We would normally expect this type of offer to increase de-mand or, at a minimum, leave demand unchanged. How-Firms often sear