7/29/2019 Child Intervention
1/13
Objective To estimate the additional resources required to scale up interventions to reduce child mortality and morbidity withinthe context o the ourth Millennium Development Goals aim to reduce mortality among children aged < 5 years by two-thirds by2015.Methods A costing model was developed to estimate the nancial resources needed in 75 countries to scale up priority interventionsthat address the major causes o mortality among children aged < 5 years, including malnutrition, pneumonia, diarrhoea, malariaand key newborn causes o death such as sepsis. Calculations were made using bottom-up and ingredients-based approaches; thisallowed nancial costs to be estimated or each intervention, country and year. Costs refect WHO guidelines on inputs and deliverystrategies and encompass the delivery o interventions at community and acility levels. These costs also include programme-specicinvestments needed at national level and district level.FindingsThe scale-up scenario predicts that an additional US$ 52.4 billion will be required or the period 20062015. This representsan increase in total per-capita health expenditure in the 75 countries o US$ 0.47 in 2006; this is projected to increase to US$ 1.46
in 2015. Projected costs in 2015 are equivalent to increasing the average total health expenditure rom all nancial sources in the75 countries by 8% and raising general government health expenditure by 26% over 2002 levels. (The latest data available at thetime o the study were or 2002.) The scale-up scenario indicates that countries with weak health systems may experience dicultiesmobilizing enough domestic public unds.Conclusion While the results are approximate estimates, they show a substantial investment gap that low- and middle-incomecountries and their development partners need to bridge to reach the ourth Millennium Development Goal.
Bulletin o the World Health Organization 2007;85:305-314.
Une traduction en ranais de ce rsum gure la n de larticle. Al nal del artculo se acilita una traduccin al espaol.
A fnancial road map to scaling up essential child healthinterventions in 75 countriesKarin Stenberg,a Benjamin Johns,b Robert W Scherpbier a & Tessa Tan-Torres Edejer c
IntroductionGlobal reductions in child mortalityhave stagnated, and annually almost 11million children die beore their thbirthday, mainly rom preventable ill-nesses.1 In many countries, a reversal inprogress highlights the ailure to reachchildren and to provide known and e-ective interventions.
Te ourth Millennium Develop-ment Goal to reduce child mortalityby two-thirds by 2015 is ambitious
but achievable.2
Te World health report2005identies technical strategies to im-prove the health o mothers, newbornsand children.3 With only a decade letto reach the Millennium DevelopmentGoal targets, policies more rmly relatedto implementation are needed. Budgetassessments are a necessary step in mov-
.
a Department o Child and Adolescent Health and Development, World Health Organization, 20 avenue Appia, 1211 Geneva 27, Switzerland. Correspondence toKarin Stenberg (e-mail: [email protected]).
b Department o Health Systems Financing, WHO, Jakarta, Indonesia.c Department o Health Systems Financing, WHO, Geneva, Switzerland.doi: 10.2471/BLT.06.032052
ing rom vision to implementation ostrategic plans.
Previous attempts have been madeto estimate the costs o scaling up deliv-ery o child health interventions includ-ing the United Nations MillenniumProjects needs assessments. Findingsrom case studies rom ve countriesestimated the total investments in childhealth needed per capita in 2015: theseranged rom US$ 3.80 in Ghana to
US$ 6.80 in Uganda.4
Te Commis-sion on Macroeconomics and Healthestimated the economic costs o scalingup health interventions that addresschildhood-related illnesses in 83 coun-tries, projecting as ar as 2015. Tiscommission ound that an additionalUS$ 11.9 billion would be needed
per year above current expenditures.5Further, members o the Bellagio StudyGroup on Child Survival assessed the run-ning costs o providing 23 interventionsto improve child survival in 42 countries;the cost was estimated to be US$ 5.1billion in new resources annually.6
Our study was conducted becausenone o the pre-existing cost estimatesorecast the additional nancial undsrequired year-by-year or scale-up. Esti-
mates presented by the UN MillenniumProject represent total rather than incre-mental costs. Te estimates made by theCommission on Macroeconomics andHealth reect an economic perspectiverather than nancial expenditures. Teprice tag presented by members o theBellagio group estimate the additional
7/29/2019 Child Intervention
2/13
ResearchA fnancial road map or child survival Karin Stenberg et al.
running costs once interventions havebeen successully scaled up to ull cover-age rather than the increasing costs oscaling up health service delivery overtime. We believe that yearly estimates onancial needs are essential to guide the
advocacy or raising unds to reach theourth Millennium Development Goal.Tis study provides the rst estimateo the incremental nancial resourcesrequired over 10 years to scale up childhealth interventions to universal cover-age by 2015, and it includes programmeinvestments as well as running costs.
We acknowledge that child survivaldepends on maternal survival and preg-nancy care as well as actors outside thehealth sector (e.g. womens education,access to clean air and water, and im-provements in inrastructure). However,or this study, we limited the list o inter-ventions to those generally alling underthe responsibility o what are known aschild health programmes. As such,these estimates are intended to providea nancial road map that is useul orchild health practitioners, managers onational and international organizationsand programmes providing interven-tions that are key to childrens survival,and donors considering unding interven-tions to attain this goal.
Methods, assumptions anddata sourcesTis section summarizes the methodsused. For greater detail on the key as-sumptions underlying the models,please see the relevant technical work-ing paper.7
Te development o a nancialroad map or scaling up essential childhealth interventions required us toselect countries or the analysis, to iden-tiy priority interventions, to estimate
the population in need o each service,to dene current and target coverageo interventions, and to collect coun-try-specic costs associated with thedelivery o interventions.
For interventions that have an im-pact on the major causes o mortalityamong children aged < 5 years (mal-nutrition, pneumonia, diarrhoea andconditions aecting neonates), a modelwas constructed that incorporated allinputs. Pre-existing models developedby WHOs Department o Immuniza-
tion, Vaccines and Biologicals 8 and theRoll Back Malaria Partnership 9 wereused to assess costs or immunization
malaria among children aged < 5 years.Te models use similar methods anddata sources. In addition, costs or anti-retroviral prophylaxis and replacementeeding were assessed using the resourceneeds model developed by Constella
Futures (ormerly known as the FuturesGroup).10 Te models were applied to 75countries and harmonized or deliverystrategies and target coverage.
Te projected costs cover the incre-mental scale-up required in order to closethe coverage gap rom current coveragelevels to 95% coverage. Given the shorttime rame, the modelling assumed thatno major changes to the health system
were made, and we used a simpliedimplementation model that allowed orthe delivery o interventions within exist-
ing (non-nancial) constraints. Ongoingcare provided to children aged < 5 yearswas assumed to continue at current ratesusing current practices; this assumption
was also made or other interventionsnot included in our cost estimate.
Countries includedCountries were selected in parallel withanother WHO costing exercise orscaling up care or mothers and new-borns.11 Te main selection criterionor countries was that they have high
mortality among mothers and children,both in terms o rates and gross numberso maternal deaths, neonatal deaths anddeaths among children. Eorts weremade to include low-income countriesrom all regions. All countries where themortality rate or children aged < 5 yearsis > 100 are represented, as well as allcountries experiencing reversal or stagna-tion in the reduction o mortality ratesamong this age group. Te 75 countriesincluded have a total population o 4.6billion and account or 94% o global
deaths among children in this age group.Te countries included are listed inAnnex 1 (available at http://www.who.int/bulletin).
Health services includedInterventionsTere is sound evidence that high-im-pact, low-cost interventions could leadto a major reduction in child mortality.able 1 lists the 16 key intervention setsthat were costed; these were selectedon the basis o their potential impacts
on mortality and morbidity and theeasibility o delivery.1,12 Te estimatespresented here cover only those inter-
Costs or other essential interventionsaddressing neonatal mortality, in par-ticular those having an impact on as-phyxia and prematurity, are estimatedin the parallel exercise or maternal andnewborn care. Te two projects were
careully coordinated to prevent double-counting. Interventions delivered duringpregnancy and at birth, and interven-tions linked to complications resultingrom pregnancy and birth, have been as-signed to maternal and newborn healthcosts, whereas the treatment o neonatalinections is included in the costs oproviding child health care.
Delivery strategiesTe costs presented here reect theperspective o a public health system
and represent a continuum o care run-ning rom the household, communityand acility levels up to the nationalprogramme-management level. In-tervention-specic delivery strategiesare based on WHOs guidelines orprevention and care. Te IntegratedManagement o Childhood Illness(IMCI) strategy has proven to be anexcellent delivery channel or manyo the interventions included in thiscosting exercise.1315 Otherwise, dataon delivery channels is sparse, and the
models utilize assumptions as to wherethe incremental scale up o interven-tions would be most easible within ashort time rame (able 1).
We assumed that it is possible toreach high coverage in all countries by2015 through a combination o ex-tending existing health-care networks,stepping up outreach and relying ontrained community health workers. Weurther assumed that rapid scale-up willrequire a greater deployment o com-munity health workers in rural settings
compared with urban settings, owingto the poor availability and utilizationo ormal health-care systems in manyrural environments.
Estimating populations in needTree parameters were used to calculatethe incremental population in need oeach intervention.
Population data on children aged < 5years and the expected number o birthsper country and year: Tese data weregathered rom the UN Population
Divisions 2002 medium variant pro-jections and were adjusted or thenumber o lives saved as interven-
7/29/2019 Child Intervention
3/13
ResearchA fnancial road map or child survivalKarin Stenberg et al.
Table 1. Sixteen core interventions to improve child survival and their major point(s) o delivery, as costed in the model
Interventions Population in need Current coverage Delivery channels used to apportionincremental delivery
Source odata
Range oestimates
used
Source odata
Range oestimates
used
National Firstreerral
Primaryacility
Commu-nity
Preventive interventions
Counselling or promotiono exclusive and continuedbreasteeding
Expert opinion 100% oinants
Expert opinion 520% X X X
Counselling or improvedcomplementary eeding
Expert opinion 100% oinants
Expert opinion 2% X X
Implementation o theInternational Code oMarketing o Breast MilkSubstitutes
Expert opinion All 75countries arein need oadditionalinvestmentsat pro-gramme level
InternationalBaby FoodActionNetwork,expert opinion
None(additionalcosts wereincludedor allcountries)
X
Immunizations (BacilleCalmetteGurin,diphtheriatetanuspertussis,measles, yellow ever, polio,Haemophilus infuenzaetype b, hepatitis B)a
Expert opinion 100% ochildren
WHO 0100% X X
Insecticide treated bednetsb By country,the proportiono people inendemic areaswas determinedusing climatic
and environ-mental model-ling or clinicalreporting oincidence
0100% ochildren
Demographicand HealthSurveys,regionalaverages
Demo-graphicand HealthSurveys
0-23%
X X
Routine vitamin Asupplementation to childrenaged < 5 years
Expert opinion 100% ochildren aged> 6 months
WHO 0100% X X
Universal salt iodization WHO dataon iodinedeciency
0100% ochildren
WHO 098% X
Prevention o mother-to-child HIV transmission by
antiretroviral prophylaxis andinant eeding and counselling
Reer to assumptions in the Constella Futures resource needsmodel10
X
Treatment interventions
Case management odiarrhoea
Demographicand HealthSurveys, regionalaverages
2.46.0episodes perchild/year
Demographicand HealthSurveys, regionalaverages
1281% X X X
Antibiotic treatment ordysentery
Demographicand HealthSurveys, regionalaverages andexpert opinion
5% odiarrhoeaincidence
Diarrhoeacoverageused as proxyindicator
1281% X
Case management o
pneumonia
Demographic
and HealthSurveys,regionalaverages
0.220.34
episodes perchild/year
Demographic
and HealthSurveys,regionalaverages
1675% X X X
7/29/2019 Child Intervention
4/13
ResearchA fnancial road map or child survival Karin Stenberg et al.
Interventions Population in need Current coverage Delivery channels used to apportionincremental delivery
Source odata
Range oestimates
used
Source odata
Range oestimates
used
National Firstreerral
Primaryacility
Commu-nity
Case management o severemalnutrition
WHO GlobalDatabaseon ChildGrowth andMalnutrition
0.24.3per 100children/year withweight-or-height 12months
Estimatedrom WHOdata
094% X X
a Following assumptions in the model produced by WHOs Department o Immunization, Vaccines and Biologicals.b Following assumptions in the model produced by the Roll Back Malaria Partnership.
(Table 1, cont.)
Incidence or prevalence o a condition
or risk: Tese data were based ondata rom WHO or experts opin-ions. (Te expert committees werechosen rom sta at WHO known tobe specialists in the area o ocus, eacho whom has access to a wider net-work o global experts.) Te modelassumes that epidemiological risk willremain constant up to 2015, apartrom reductions in the incidence omeasles and Haemophilus infuenzaetype b brought about by improvedvaccine delivery.
Incremental scale-up o coverage re-quired: Tis was estimated using theuniversal coverage target o 95%
country and intervention. Te cur-
rent coverage o interventions isbased on the best available estimatesrom 1995 or later. For countrieswithout coverage data, average re-gional values were used. For inter-ventions with little available coveragedata, estimates are based on expertopinion.
Scale-up ratesAn index developed by the Commis-sion on Macroeconomics and Healthclassies countries health systems into
our dierent levels o strength based onconstraints other than lack o nance,such as those related to demand and
broader economic and political actors.16
Tis classication is used here in twoways: (i) to model country-specic andintervention-specic scale-up curvesbased on current coverage and growth-rate scenarios applied to each group inthe index, and (ii) to determine the needor additional investment required tostrengthen the health systems capacityto deliver interventions. Countries witha low index (those rated 1) are seen ashaving greater need or incremental in-vestments at the health system level, aswell as acing a slow start-up beore rapid
scale-up in the end phase. Countrieswith a high index (rated 4) are expectedto be able to reach ull coverage or most
7/29/2019 Child Intervention
5/13
ResearchA fnancial road map or child survivalKarin Stenberg et al.
Table 2. Components o patient costs and programme costs or child health interventions.7
Category Costs included
Patient costsCommodities Drugs, vaccines, laboratory tests, medical supplies
Service delivery costs Costs or consultation time include salaries o multipurpose health workers and locally procured goods, such asoverhead costs or electricity, running water and buildings
Programme costs Activities Costs includeda
Community health workers Honoraria and equipment or community health workers delivering child healthinterventions
E, H, S
Programme management Deployment o national-level and provincial-level programme sta; review o policy;development o strategic plans; coordination work; development o district plans;recruitment and management o community health workers
CD, MC, PD, PM, S,SS, TC
Training Upgrading pre-service training programmes: Implementing introductory training orcommunity health workers; implementing in-service training or sta at rst-levelacilities (IMCI training) and rst reerral level (specialized training or managemento severely ill children); training programme managers on childrens rights and the UNConvention on the rights o the child
CD, MC, PD, PM, S,SS, TC
Supportive supervision Supervision at district level, rst reerral care-level hospitals, primary-level healthcentres, and o community health workers
CD, MC, PD, S, SE,SS, TC
Externally unded technicalassistance/capacity building
Consultants ees CD
Inormation, education andcommunication
Hiring o social mobilization ocers; ormative research and development oinormation, education and communication materials; community mobilization to raiseawareness through media (radio, TV) and printed material (posters, fiers)
E, PD, PM, S, SS, TC
Monitoring and evaluation Hiring o sta (epidemiology and data entry); strategic review meetings; maintenanceo inormation technology inrastructure; community-based and acility-based surveys
CD, E, M, MC, PD,SS, TC
Inrastructure Existing hospitals upgrade equipment to standards; existing health centres upgradeequipment to standards, upgrade selected acilities to hospital standard in regionswhere there are no hospitals available (includes upgrading equipment and hiringadditional sta); provide vehicles
E, M, SS
Advocacy Development and review o advocacy strategy; provision o advocacy materials andimplementation o advocacy activities
MC, PD, PM, S, TC
Laws, policy, regulation Drating relevant legislation and policies (including legislation and policy related tothe International Code o Marketing o Breast Milk Substitutes, iodine orticationand the UN Convention on the rights o the child) monitoring activities related toimplementation o legislation and policy
MC, PD, S, TC
Programme costs orimmunization
Supervision, provision and maintenance o cold chain, purchase and maintenance ovehicles
E, M, PD, S, SS, TC
CD, consultant days; E, equipment; H, honoraria; IMCI, Integrated Management o Childhood Illness programme; M, maintenance; MC, meeting costs; PD, perdiems; PM, printing materials; S, supplies; SE, supportive supervision expenses; SS, incremental sta salaries or additional recruited sta; TC, travel costs.a All except PM and S are not considered ully tradable goods and are adjusted to country-specic prices.
Estimating country-specifc costsAll our models use a st andard -ized WHO ingredients approach(cost = quantity * price) to derive coun-try-specic cost estimates. Costs are clas-sied as patient costs or programmecosts (able 2). Patient costs reer tocosts at the point o delivering servicesto the client; they include outpatientvisits, inpatient bed days and the useo drugs, supplies and laboratory tests.Programme costs are expenses incurredat the administrative levels o the district,
province or country.17Te quantities o inputs needed
or patient care were dened accord-
guidelines or prevention. Te inputs es-sential to strengthening programmes anddelivery systems were based on expertsopinions o the minimum requirementsor efcient programme managementand related assumptions or health-sys-tem requirements. Estimates did not ac-count or storage, loss or waste o drugsand supplies, or the construction onew acilities or or the production oincremental multipurpose health proes-sionals to provide acility-based care.
Te cost or community health
workers honoraria, equipment and sup-port were included in programme costsbecause the need or community health
density rather than rom the number oconsultations with patients. Communityhealth workers were assumed to be ullyunded by the child health programme,with an estimated need or 1 communityhealth worker per 1000 population inrural areas and 1 per 1500 population inurban areas. Based on experts opinions,costs were adjusted or an attrition rateo community health workers o 25%per year.
Prices were derived rom WHOsChoosing Interventions that are Cost
Eective (WHO-CHOICE) database18and the Disease Control Priorities Proj-ect,19 except or drugs, where the median
7/29/2019 Child Intervention
6/13
ResearchA fnancial road map or child survival Karin Stenberg et al.
or Healths International drug priceindicator guide are used.20 Catalogueswere used to determine the prices olarger equipment.
Costs are presented in 2004 USdollars, include a 3% annual ination
rate and are estimated by input cat-egory, intervention, country and year.Costs reect a supply-side perspectiveusing public provider prices; they donot include households costs related tohealth-seeking behaviour or accessingservices.
FindingsTe total additional cost or imple-menting the scale-up scenarios wasestimated to be US$ 52.4 billion. Fig. 1
shows the breakdown o costs or eachset o interventions by year. We esti-mated the need in 2006 at US$ 2.2billion, which would increase to US$7.8 billion by 2015. While the costsor immunization, preventing mother-to-child transmission o HIV, treatingand preventing malaria and programmesupport activities remain relativelystable, the cost o scaling up nutritioninterventions, and the case managemento neonatal inections, pneumonia anddiarrhoea increase sharply over time.
Tis reects the relatively higher cur-rent levels o investment or some othe ormer interventions (immunizationand treating and preventing malaria)and poor initial coverage or some o thelatter interventions (such as nutritionalcounselling and case management oneonatal illness). Te relatively high costo diarrhoea case management comparedwith pneumonia is explained by theliberal distribution o oral rehydrationsalts and zinc in the model versus themore restricted inclusion o oxygen costsor treating pneumonia. Programmaticinvestments remain relatively stablethroughout the period, illustrating theeconomies o scale o using an integratedapproach.
Te average additional cost percapita or all 75 countries was US$ 0.47in 2006; this rises to US$ 1.46 in 2015.Te average investment required perchild in 2015 at ull coverage is an extraUS$ 12.31 (range: US$ 10.3517.78 orall index categories).
Te 13 countries in the groupwith the most avourable environmentor scale-up (those classied as index
Fig. 1. Yearly breakdown o additional costs o increasing the coverage o childhealth interventions to reach universal coverage by 2015
Programme costs
2015
US$(billions)
Year
Nutrition interventions (EBF, CF, SM)
Immunizations and vit. A suppl.
0
9
Diarrhoea and dysentery management (ORS and zinc)
Case management o pneumonia and measles
ITNs and antimalarials
Neonatal inections
PMTCT
USI and deworming
201420132012201120102009200820072006
8
7
6
5
4
3
2
1
CF, counselling or improved complementary eeding; EBF, counselling or promotion o exclusive and continuedbreasteeding; ITNs, insecticide treated bednets; ORS = Oral Rehydration salts; PMTCT, prevention o mother-to-childtransmission o HIV by antiretroviral prophylaxis and inant eeding and counseling; SM, case management o severemalnutrition; USI, universal salt iodization.
price tag (able 3). Tis group includescountries with large populations, suchas China and India. Yet on a populationbasis, the incremental investment neededis less than hal that o countries in indexcategory 1 (Fig. 2). Te more developedhealth systems, which occur in manymiddle-income countries, need to spenda relatively larger proportion o their ad-ditional expenditures on human resources.(Note that the salary costs or health
workers or all countries are likely to beunderestimated, given that no adjustments
were made or the need to raise salaries inlow-income countries to recruit and retainthe necessary human resources.)
In index categories 1 and 2, whereconditions currently are the most chal-lenging, an additional US$ 21 billionis required. Tese are low-incomecountries with high mortality rates, lowhealth-care coverage and relatively weakhealth systems. However, in these coun-tries the prices o labour and supplies
are comparatively low. Considerableinvestments will need to be made notonly in commodities but also to expand
Te additional resources needed in2015 correspond to a mean 2% increaseand a (population-weighted) average 8%increase in total expenditures on health inthe 75 countries rom all nancial sources(range 0168%). Similarly, projectedcosts in 2015 are equivalent to raising(population-weighted average) generalgovernment health expenditure by 26%over 2002 levels.21 (Te latest data availableat the time o the study were or 2002.)
Although the total resource requirementsare highest in countries in index category4, the implications or nancing are lesssevere or these countries. Te averageadditional cost (weighted by population)per capita in these countries (US$ 1.00) isless than a third o that needed in countriesin index category 1 (US$ 3.40); Fig. 3 il-lustrates that this is a manageable increasein current expenditures or higher indexcountries 3 and 4, but that countries withweaker health systems may experiencedifculties in mobilizing domestic public
unds or scaling up child health services,given the large relative increase in expendi-ture required (21% in category 2 countries;
7/29/2019 Child Intervention
7/13
ResearchA fnancial road map or child survivalKarin Stenberg et al.
Table 3. Estimated cost o scaling up child-health interventions, 20062015 (US$)
Indexcategorya
No. ocountries
Estimatedadditional
cost(millions
o US$)
% ototalcosts
% ochildren< 5 yearsincluded
% oglobal
mortalityamong
children< 5 years
Averagetotal
healthexpendi-
ture percapitab,c
Averagegeneral
governmenthealth
expenditureper capitab,c
Weightedaverage
incrementalcost per
capita
c
in2015
Averageincremental
cost per child< 5 years in
2015
1 22 11 667 22 17 32 14.62 4.60 3.41 17.782 19 8 994 17 18 20 41.09 9.51 2.01 12.773 20 5 537 11 10 8 35.49 15.54 1.51 11.834 14 26 241 50 55 35 73.01 27.82 1.01 10.35All 75countries
75 52 440 100 100 94 59.93 22.17 1.46 12.31
a The Commission on Macroeconomics and Healths index classies countries health systems into our dierent levels o strength, with lower numbers indicatinggreater need or incremental investments. See text or urther details.
b For data on total health expenditure and general government health expenditure, reer to the tables on National Health Accounts in the annexes to the Worldhealth report 2005: make every mother and child count.3 Data here are rom 2002, which was the latest data set available at the time o this study.
c Average health expenditure per capita and average incremental cost per capita and per child were weighted by the year-specic population or each country.
Fig. 2. Incremental expenditures required or each Commission on Macroeconomicsand Health index category, by cost category, 20062015 (simple averages,no population weighting)
Commodities (lab tests, drugs an supplies)aIndex category
Salaries (proessional health workers)b
01
25
20
15
10
5
Community health workersc
Programme costs
b
2 3 4
AdditionalUS$(millions)
required/millionpopulation
a Cost components are considered ully tradable. International prices have been used.b Cost components are partially tradable. Salary costs have been adjusted to country-specic prices.
DiscussionTe numbers presented here are in-tended to be approximate estimatesor donors, multilateral agencies andcountries to help them determine thenancial resources needed to scale upchild health programmes with the aimo reaching the ourth Millennium De-velopment Goal by 2015. Members othe Bellagio group estimated that scalingup health services or children, mothers
and newborns in 42 countries wouldreduce mortality among children aged