33717 Vigilance nutrition programs in developing coun- contributed significantly to large poor- tries. investing in health is investing in rich gaps in health outcomes like child IN LIGHT OF the especially high burden the better off. who tend to be the princi - and maternal mortality. The standard of disease and malnutrition among the pal users of health services. This is true assumption that the poor are the princi- poor. it is not surprising that many pol - not only of private health spending. but pal beneficiaries of publicly - funded icy makers have assumed that investing also of public programs typically under- health services is clearly incorrect. and in health is investing in poor people. taken in the name of health equity. must be replaced by vigilance to ensure Unfortunately. increasing empirical While such an orientation of public that the traditional pattern of higher evidence finds that for most health and spending is not unique to health. it has service coverage among upper- income groups does not continue. Hope THE PRESENTATIONS MADE at the RPP Conference provide a basis for hope that the current situation just described can be changed. by highlighting a number of successful efforts at reaching the poor with services. While not all the confer- ence presentations showed a pro-poor pattern of service use. a clear majority did. In the same way that the earlier research found worldwide inequalities in service use. the initial rays of hope represented in this issue. as well as by other RPP - documented studies are o Poorest BestˇOff Poorest global as well. 20% 20% Receive LESS than The first three articles provide BestˇOff important examples of programs with strong pro-poor reach in LatinAmerica. SOU(( e Deon Filmer TilL' Ill( Id('IH (' \)f puhll( ('XIWrld'tures 111 iledltll ,-Hid f'l!LJ( dtl()!1 j,1dy ;'::003 The Colombia article. by Maria- Luisa http 1'( (Jrl :.urldt)dnk or(o :.dr ,',dr)()U!llilJrdr, d r)( lid {?9478 Escobar. shows how an equity fund was created and financed by the government to increase health insurance coverage specifically for the poor. which in turn lowered financial barriers for service use. The article on Mexico's PROGRESA program. by David Coady and col - leagues. shows the effectiveness of an ambitious program that directly trans - fers cash to poor families if they seek health services. Finally. the article on community- based social funds in Honduras. Peru. and Nicaragua. by Laura Rawlings. shows how community ownership and involvement can pro - duce pro-poor health services. The next two articles provide instructive illustrations of pro - poor programs in Sub -Saharan Africa. The Antenatal Oral Full Med. AH. Med. Mid. Modem Tanzania article. by Rose Nathan and Care Rehydration Immunization Treatment Delivery Treatement Treatment Contraceptive Therapy of of Diarrhea of Fever Use (Women) colleagues. describes the successful Ac.Res.lnf. adaptation of private sector techniques Source G\'Jatklfl 0 RUtStE-'lll S JO[l!lS0!1 K Pande R \',Iagstdff A Sl't [0 ~(01101111[ dlffprenres In in marketing. called social marketing. to ilCdlttl rll,trltlOrl dlllJ Pl)P Ilat on increase the use by the poor of a critical NOTE NUfllt)er ot (0 Intrlf''J jrlrlCs frolll ')1 to )b (Jependille; LJPUII .... pr,lll ( malaria prevention product. insecti - cide - treated bed - nets. The second 4 Developmenr Ourreach WORLD BANK I NST I TUTE Africa article, by Fran~ois Diop and Damascene Butera, is on tage of the most promising opportunities and to counter those the success of community- based voluntary micro health factors that constitute the most important bottlenecks identi - insurance for rural poor population in Rwanda . By removing fied . It will be equally important to develop ways of monitor- the financial barriers from using care, the project produced a ing how well the poor have access to and use health, nutrition, substantial increase in the use of preventive and curative and population services, in order to prevent a repeat of situa - health services by the poor. tions where the rich capture the benefits and the poor remain The last four articles highlight pro - poor programs inAsia- outside looking in. two in Cambodia , one in Nepal and one in India. The first In brief, a promising beginning has been made, and that Cambodia article, by Indu Bhushan and colleagues, shows how provides important grounds for hope . But a beginning is only contracts with international NGOs that included incentives a beginning. The need for vigilance and hard work remains . ....., for reaching the poor led to sharply increased use of primary health services by the poor residents of underserved areas . Abdo S. Yazbeck is Lead Economist, Health, Nutrition, and The second Cambodia article , by Bruno Meessen and Ir Por, Population Program, World Bank Institute. describes how a pilot equity fund effectively helped address Davidson R. Gwatkin is a Consultant with the Health, Nutrition, and the need of the poor to access hospital services by decreasing Population Department, Human Development Network, World Bank. the financial barrier. The India article, by Kent Ranson and colleagues, describes how a women's workers union improved The authors, along with Adam Wagstaff are the Coordinators of the access to reproductive health services to the urban poor, by Reaching the Poor Program. addressing geographic and other barriers and by building trust between the community and the providers. The final article, by Anju Malhotra and colleagues, shows how partici - patory planning is used to ensure that youth- oriented repro - ductive health services serve the most vulnerable in Nepal. H ard work WH I LE EXAM PL ES LI KE THOSE just presented are encourag- ing, the wide array of approaches employed in them suggest the absence of a single, simple, universal solution. Rather, the array reflects the fact that poverty and inequality result from combinations of persistent factors that can differ widely from country to country and from region to region. In reviewing the growing experience of success and failure in reaching poor people with health services, a factor that appears to playa crit - ical role in developing successful policies and approaches based on a solid understanding of why the poor were not get - ting services. Moving past the arrogance of thinking we know the answers and working hard to understand the determinants of inequalities are the obvious next steps toward policies that are customized to the needs of the poor in given settings. This hard work of answering the 'why' questions and beginning to develop policies customized to different situa- tions includes a number of analytical tools that helps policy- makers understand: 1. where the poor live (critical for geographic targeting); ~. why the poor so often fail to come forward for services (because of household factors like lack of knowledge or resources, or because of community -level cultural factors like constraints on what women are allowed to do); 3. why public services typically fail to reach out to the poor (because of deficiencies in strategies, management , vision); 4. how the private (not -for- profit and for - profit) sector can be effectively harnessed to complement public servic - es in serving the neediest. But analytical work is clearly not enough. Policies have to be fashioned , tested and implemented in order to take advan - MAY 2005 Health Sector Reform in Colombia BY MARIA-LUISA ESCOBAR private providers for their health care was its cost. One out of every six individ- needs. A large portion of the Colombian uals in the first income quintile who fell BEFO RE TH E 1993 health sector reform, health care system was financed by public ill in 199~ did not seek medical care poor households in Colombia were dis- funds from the Treasury and supported a because they could not afford to pay for proportionately affected when facing large public network of hospitals and it (National Household Survey, 199~). health shocks . The health system was clinics through general taxation. The poor not only had less access to characterized by strong market segmen- However, out-of-pocket expenses were health care than the rich, but also paid tation' high inefficiency and badly tar- an important source of health care out- of-pocket for health care services geted public subsidies. The Ministry of financing, and the rest was provided by both by public and private providers, Health (MOH) was responsible for pro - insufficient and inefficient social security and paid more proportionally to their viding health care to all Colombians by schemes. The poor had three alternatives income level than the rich. constitutional mandate. In reality, only when facing illness: to try to access pub- The Colombia health care system before around ~o percent of the Colombian licly provided health care services, to go to the 1993 reform did not allocate public population were finanCially protected private providers, or not to seek medical subsidies to upper and middle - income against the risk of health shocks. care at all. Considering that pharmaceuti- groups. Of those who were hospitalized cals could be sold without a prescription, in public, government-funded hospi - Pre-1 993 myths that and deficient quality control and self- tals, only ~o percent belonged to the first supported the Colombian medication were common issues, the income quintile, while almost 60 per- health sector poor and less educated were more at risk cent were individuals in the third, when opting for self-care . fourth, and fifth quintiles. But in 199~, COLOMBIA HAD A HEALTH SYSTEM 1~ percent of hospitalizations and ~o based on myths. Policy makers believed Diffe re nces be tween rich percent of all surgeries done in the pub - that Colombia's public health subsidies and poor lic sector were received by patients in the were well targeted to the poor, publicly richest ~o percent group of the popula- provided health care services were free of THE MOST IMPORTANT BARRIER to tion. So, middle- and higher-income charge for all, and the poor did not choose health care use before the 1993 reform group individuals who could afford to use other hospital services were crowding out the poor in public facilities. Very rarely the poor received free care in public facilities . While 91 percent of the poorest income quintile who were hospitalized in public hospitals incurred out-of-pocket expenses, only 69 percent of individu- als in the richest quintile did the same. The private sector was important both in the financing and the provision of health services before the 6 Development Outreach WORLD BANK I NST I TUTE reform. According to the National Household Survey in 1992, 40 process, and one point of the contributions is allocated to percent of all health interventions and 45 percent of all hospital- finance the RS together with Treasury transfers to the territo - izations were done in the private sector, although only around 20 ries. Individuals who are eligible for affiliation to the RS, but percent of the Colombian population was insured. still uninsured, are called vinculados and should rely on public hospitals for care. T he 1993 Health Sector Reform In Every insured individual chooses freely an insurer and Colombia selects providers within the insurer's network. Both regimes LAW 100 0 F 1993 mandates the creation of a new system for the financing and delivery of health care, allocating public subsidies directly to individuals instead of institutions. The reform introduced four main elements to reach the poor: i) a proxy-means testing index to target the allocation of public subsidies in health (SISBEN- Selection System of Beneficiaries for Social Programs-Nunez 2004); ii) transformation of the traditional supply-side subsidies, which finance the public health care network, into individual insurance premiums for the poor subsidized by the system; iii) an equity fund with financial flows allowing for payroll contributions and Treasury resources to cross subsidize the insurance premium for the poor; and, iv) contracting health service delivery from both the public and private sectors. The new system is character- ized by universal health insurance coverage with two regimes. The formally employed and the inde- pendent workers who are able to contribute belong to the contribu- tory regime (Regimen ContTibutivo - RC). Contributions are collected by the insurer of choice. The poor and indigent do not make any insur- ance contribution and are covered under the subsidized regime (Regimen Subsidiado-RS). Payroll contributions go to a national health fund (Fondo de Solidaridad r Garantia -FOSYGA) with four sepa- rate accounts. The fund finances insurance premiums to all in the RC with an internal compensation MAY 2005 7 REASON S FOR NOT SEEKING MEDICAL CARE BY INCOME LEVEL UNIN SURED - ECV 2003 1 ˇˇˇˇ 1 I_ ! _ ! .. ˇˇ 1 ˇ 73 . . . . . . . ˇ II ˇ I 1 1- W _ I 69 ˇ II .461149 REA SON S FOR NOT SEEKING MEDICAL CARE BY INCOME LEVEL IN SURED - ECV 2 00 3 ˇ II ˇ II II 30 ˇ ˇ II ˇ II 11 ˇ 1 ˇˇˇˇ 1 ˇ II ˇ II II ˇ 1 ˇˇˇˇ 1 t......II.-II-I_ J 1 ˇˇˇˇ 1 1 ˇˇˇˇ 1 .J .1 . 1 H---IIˇ. - . 1 I .--Pi . ˇ .18.16. ˇ. 1 ˇ 35 ˇ ˇ ˇ ˇ ˇ 30. 6 Although the health sector in Colombia still faces impor- tant challenges to expand insurance coverage to all the poor, to improve quality of services and to provide a more integral benefit plan for the poor, there have been important accom- plishments that deserve our attention. The results have access to a basic benefit package, but the POS (Plan Obligatorio de Salud) for the contributory regime includes all THE REFORM BROUGHT more opportunities for access to levels of care, while the POSS (Plan Obligatorio de Salud health care to the poor. Differences between those insured and Subsidiado) has to be complemented with services provided by the non insured are important. Treatment rates are higher for public hospitals and financed through the existing traditional those insured than for the non-insured in both the urban and supply side subsidies. According to the Law, overtime those rural areas, as well as the utilization of preventive care services. supply- side subsidies should be transformed into demand- The reform of 1993 increased financial protection to all; side subsides in order to achieve universal insurance coverage particularly among the poor and those in the rural areas. with the same POS in both regimes. Colombia had ~3 percent of its population financially protect- 8 Development Outreach WORLD BANK INSTITUTE employment or to income level. and prenatal care. Health care expenditure as a per- Consistent with findings in other centage of income is much larger in the countries, those insured in Colombia case of those non-insured than for have higher health care utilization rates those in either the Contributory or the and seek care faster than those not Subsidized Regimes. Formal insurance insured. This is particularly important in in Colombia reduced out-of-pocket the case of childbirth and child and expenditures in ambulatory care maternal health. The Demographic and between 50 and 60 percent (Bitran et. Health Surveys DHS of Colombia indicate al. ~004). The poor in the RS spend a very important improvement in access around 4 percent of their income on to those services particularly in the rural ambulatory care, while the uninsured areas. According to the DHS (1986, 1990, poor spend more than 8 percent of their 1995 and ~ooo), there was a 66 percent income on ambulatory care (household increase on child delivery assisted by a consumption expenditure is used, see physician, 18 percent increase on institu- Quality of Life Survey, ~003). Out -of- tional delivery, and a 49 percent increase pocket expenditures on hospitalization in prenatal care use among rural women. among the uninsured poor, represent - The DHS ~ooo shows that there is an ed in ~003 more than 35 percent of astonishing difference between infant their income. The poor in the RC spend mortality rates among children whose a smaller proportion of their income in mothers had access to prenatal care (P) inpatient care than the poor in the RS. and to institutional delivery (ID) with However, the same study finds that a those whose mothers did not have access health shock requiring hospitalization to such services. brings 14 percent of those hospitalized and uninsured below the poverty line, The challenges while in the case of those in the Subsidized Regime, only 4 percent DESPITE POSITIVE results, the would fall below the poverty line when Colombian social insurance scheme facing the same shock. receives criticism from those who would The introduction of health insurance argue that the 1993 reform has not been improved access to preventive care. successful because it has not achieved While 65 percent of the insured had at universal coverage yet and because the least one consultation with a physiCian insurance for the poor scheme finances or a dentist for preventive reasons and a reduced benefit plan when compared without being sick in ~003, only 35 per- to the plan of the Contributory Regime. cent of the non-insured did the same. The slower than planned transforma- Regulation has given preference to tion of supply side subsidies, which children, single mothers, the elderly, finance public hospitals, into demand the handicapped, and the chronically ill side subsidies to finance health insur- to obtain priority access to insurance ance for the poor, has introduced rigidi- enrollment in the RS. In effect, data ties to the expansion of the contents of shows that those poor and insured are the health insurance plan in the ed from the risk of health shocks in less healthy than their uninsured coun- Subsidized Regime and difficulties for 1993; ten years later 6~ percent of the terparts' which could be confused with the extension of coverage among the Colombian population had access to adverse selection. In reality, individuals poor. Several attempts to introduce leg- health insurance. do not decide when they can be enrolled islation to change the present system to a While insurance coverage among in the RS, since annual extension of cov- government owed and government pro - those in the highest income quintile erage depends on the availability of vided health services as the one increased modestly with the reform, from financial resources. There is no longitu - Colombia had in 199~ have not been 60 percent in 1993 to 81 percent in ~003, dinal data to investigate if access to successful so far. However, the country insurance coverage among the poorest health insurance among the poor has faces important challenges for complet- quintile of income increased from 9 per- had an impact on overall health status, cent in 1993 to 48 percent in ~003. After which would be of interest for future the reform with the introduction of the research. However, some inferences HEATLH SECTOR REFORM Subsidized Regime, access to health could be made particularly in the case of c ontinued on page ~~ insurance is not necessarily tied to formal infant mortality, institutional delivery, MAY 200S 9 PROGRESA for Progress Mexicos Health, Nutrition, and Education Program DAVID P. COADY, DEON P. FILMER , grams were poorly targeted, extremely expensive, and typical - AND DAVIDSON R. GWATKIN ly inefficient, with up to one third of program costs going for administrative expenses. Further, the programs simply pro - MEXICO'S HEALTH. NUTRITION , and Education program vided handouts, and thus only temporary relief rather than (known by its Spanish acronym "PROGRESA") was intro - developing productive capacities that could help lead recipi - duced in 1997 in order to combat the country's stubbornly ents out of poverty over the longer term. high poverty rate and to replace a set of food subsidy and other Recognizing these problems, the Mexican Government poverty programs widely considered ineffective. These pro - changed course. It largely scrapped its earlier subsidy pro - 10 Development Outreach WORLD BANK I NSTITUTE grams, and instituted in their place the lies' education, health, and nutrition PROGRESA initiative that was much status. The benefits, referred to in the more carefully oriented toward the poor technical literature as "conditional cash and that replaced simple cash transfers transfers n, became tantamount to nega- with subsidies to household invest - tive user fees that paid instead of ments in human capacity development. charged program participants to attend schools and clinics. Beneficiary selection A summary of the benefits appears in the accompanying panel. With IN ORDER TO ENS URE that its benefits respect to education, monthly cash pay- flow primarily to poor people, PRO - ments are made for each child attend- GRESA beneficiary families are selected ing grades three through nine, with through a two - stage process. higher payments in higher grades and First, poor villages are identified on for girls in those grades. For health, the basis of a community score based on payments are provided if family mem- information available from national bers' especially mothers and children, census data about things like education- make a specified number of annual allevels, occupational composition, and clinic visits. Nutritional supplements housing conditions. Those villages for younger children are also available. scoring lowest in terms of these charac - The amounts due to each beneficiary teristics-but also within a certain dis - are calculated by the PROGRESA tance of education and health facilities- administration on the basis of atten- are deemed eligible for participation. dance information submitted electron- Second, poor families within the eli- ically by school teachers and health gible villages are selected on the basis personnel at the facilities where the what is often termed a "proxy-means beneficiary had registered upon enrol- test. n That is, household -level data on ment, by submitting completed forms factors closely associated with income distributed to them by PROGRESA field are collected through a special commu- staff. These amounts are paid out at nity census, and combined into a single local distribution points to which funds scale. Those households scoring lowest are transferred via the national tele- on this scale qualify for inclusion. graphic system. Beneficiaries are noti - (Originally, about 50% of households in fied when funds are available by com- the eligible villages were included. munity volunteers, elected by the ben- However, in response to local protests, eficiaries, who also perform a wide the PROGRESA administration adopted range of other liaison functions revised criteria that permitted partici- between PROGRESA's administrators pation by around 80% of the selected and benefit recipients. communities. As a result, the selection of communities ended up becoming Implementation considerably more important in deter- challenges mining PROGRESA's targeting effec - tiveness than the identification of THE IMPLEMENTATION OF A PRO - households.) GRAM that differed so radically from those that preceding it posed many Benefit determination challenges . One has already been and provision noted: the difficulty of enabling com- munity residents to understand or PROGRAM BENEFITS ARE DESIGNED accept the legitimacy of the rather com- ing timely payment of benefits. Initially, to contribute to long-term human capi- plex' technocratic procedure used to delays were encountered at several tal development and poverty alleviation identify beneficiaries. Because of this, points: at the community level in sub - as well as to immediate poverty relief. PROGRESA's administrators felt com- mitting completed forms to the PRO - Thus, women in the beneficiary families pelled to include more people than orig- GRESA authorities; and at the central are eligible to receive regular cash pay- inally envisaged, thereby diluting (but level in issuing payments once the forms ments-but, unlike in the case of a tradi- as will be seen, by no means completely had been received. A related limitation tional cash transfer program, only if they negating) its targeting effectiveness. was in the number of fund distribution act to improve their own and their fami - Another challenge has been in assur- points, which were often quite far from MAY 200 5 II EXISTING EVIDENCE SHOWS THAT WAIVERS AND EXEMPTIONS THROUGH REGULATIONS DO NOT WORK. FEW POOR CAN BENEFIT FROM IT AND MANY BENEFICIARIES ARE NOT THE POOREST. the beneficiaries' homes. Beneficiary dissatisfaction resulted The program has been evaluated using a number of house - in the establishment of a considerably larger number of distri- hold surveys undertaken just before and at regular intervals bution points than originally envisaged. over two years after its initiation, in 320 villages that had Other issues involved the complexity of keeping the list of received services and 186 villages that had not. These and eligible households up to date; and monitoring the effective - other evaluation studies suggest that the majority of program ness and integrity of the procedures used to identify and pay benefits have gone to poor families, and that the program has beneficiaries. While preliminary evaluations have not identi - made a significant contribution to health, nutrition, educa - fied any major problems in procedure effectiveness or tion' and poverty outcomes. integrity, the issue remains one of major concern to the pro - PROGRESA's record in reaching the poor is summarized in gram's administrators. the figure. As can be seen there, almost 60% of people reached by PROGRESAbelonged to the poorest 20% of Mexico's popu- Program accomplishme nts lation; 80 % of beneficiaries were in the poorest 40% of the country's population. The principal factors contributing to the DESPITE these important constraints, PROGRESA has grown highly progressive outcomes just reported were the selection of steadily, survived a landmark shift in power away from the poor villages, as noted earlier; and also the linkage of benefits political party that established it, and now serves over 20 mil - to education/health program participation by children (since lion people or approximately one -fifth of Mexico's population. poor people have many more children than do the better- off). Overall, payments provided through the program represent The focus on poor families within villages was less important around 20% ofthe income ofthe households receiving them. (because most of the families in the selected villages were poor-a consideration that would not prevail were the program to be introduced in higher-income areas). In addition, the PROGRESA program produced noteworthy improvements in outcome indicators. For instance, there was a 45% reduction in the severity of poverty, a 16% increase in the annual growth rate of children 12- 36 months; a 20 - 25 % reduction in the incidence of illness among children aged 0 - 5 years; and an increase in secondary school enrolment for girls from 67% to 75 % and for boys from 73 % to 78 %. All this has been accomplished at relatively modest admin- istrative cost. Overall, administrative expenses have been kept to under 10% ofthe program's total expenditures. ~ David P. Coady, Techn ical Assistance Advisor, Poverty Social Impact Analysis (PSIA) Group, Fiscal Affairs Department, International Monetary Fund. Deon P. Filmer, Senior Economist, Dewvelopment Research Group (DECRG), The World Bank. Davidson R. Gwatkin, Consultant with the Health, Nutrition, and Population Department, Human Development Network, World Bank. References: David Coady, "Alleviating Structural Poverty in Developing Countries: The Approach of PROGRESA in Mexico," Background Paper for the 2004 World Development Report, February, 2003. (Accessible at: http://econ.woridbank.orgifilesl27999_Coady.pdf) "Spotlight on Progresa: Conditional Cash Transfers Reduce Poverty in Mexico," World Development Report 2004: Making Services Work for Poor People (Wash ington: The World Bank, 2004), pp. 30-31. 12 Development Outreach WORLD BANK INST I TUTE Do Social Funds Reach the Poor? BY LAURA B. RAWLINGS SOCIAL FUNDS REPRESENT a depar- ture from traditional approaches to development led by the central govern- ment. They encourage communities and local institutions to take the lead in identifying and carrying out small- scale investments. generally in social infrastructure such as health clinics and water and sanitation systems. Social Funds appraise. finance . and supervise these grants. which then may be managed by a wide range of actors. including local governments. NGOs . line ministries. community groups, and local project committees. Introduced in Bolivia a little over a decade ago. social funds have now absorbed close to US$lO billion in for- eign and domestic financing globally. and represent international financial institutions most comprehensive expe - rience with community- led develop - ment initiatives. Although conceived to address the social costs of economic adjustment. social funds have more recently been used as a tool for reaching populations that public investment programs have historically underserved. Most social funds now explicitly aim to reach poor communities. though they do not target specific households. Most engage in geographical targeting. with preference for proposals from poorer communi- ties or notional allocation targets to poorer areas based on poverty maps . Many also try to limit the types of pro- grams financed to menus of sub - proj- MAY 2 0 0 5 13 ects likely to be needed by the poor and to have positive wel - and accepted evaluation methodologies. Evaluations were fare impacts. conducted of social funds in Armenia. Bolivia. Honduras. Despite the popularity of social funds . their effectiveness Nicaragua. Peru. and Zambia where their investments had as a mechanism for reaching poor populations and improving been concentrated in education. health. water and sanitation their welfare has remained largely unmeasured and hotly sub - projects. Each evaluation reviewed the social fund's debated. Many have questioned social funds ' ability to reach poverty targeting. sustainability. welfare impacts. and costs. poor communities and households given their approach that This article summarizes the findings with respect to poverty relies on demand being generated from communities. while targeting. other debates have focused on their institutional role and influence over central and local governments. Fi n di n ~s from "Evaluating Social This article presents findings regarding social funds' abil- F unds' ity to reach the poor. drawing from a World Bank study. "Evaluating Social Funds: A Cross - Country Analysis of SOCIAL FUNDS are effective at reaching the poor and Community Investments." that represents the first attempt to extremely poor communities and households. conduct a systematic. cross -country impact evaluation of To assess whether social funds reach poor communities. social funds using household and other types of survey data the six- country study reviewed the distribution of social fund investments over time and across communities ranked by their F7":'~~~'~':=-~""" ,,,,,,o;~ I. ~ -~ ~,;".,-,,-:,.:, ~~:! -, . . poverty status. The data show that :' . .; -'.!:! geographic distribution of social fund expenditures was progressive Armenia Bolivia Honduras Nicaragua Peru Zambia in all countries studied. with poor DECILE 1996-99 1991-98 1991-98 1991-98 1992-98 1991-98 districts receiving more per capita 10.7 10,8 14.5 25.3 than wealthier districts . and the very poorest districts receiving 11.4 19.2 shares exceeding their shares of the population. Moreover. geographic 10.0 10.6 8.3 8.6 targeting has improved over time in all six social funds. The high levels 7 5.7 8.0 8.0 8.3 3.2 8.4 of investment in some of the poorest 15.1 5.5 areas refute the idea that such com- munities are systematically inca- pable of accessing resources from demand - driven programs (see Table 1). Looking at household -level tar- ~. .' ''''::'I'~'''IIV"~':'"':.,- "'-".. ... .,;~ geting results . the study used household survey data to measure income or consumption levels of a .;~ representative sample of social DECILE Armenia Honduras Peru a Nicaragua Zambia funds beneficiaries then compared 10 9 9 their poverty rates to national poverty distributions based on the same metric. The study found that in most cases the overall distribu- tion of resources at the household level was mildly progressive. 12 9 9 including among the very poorest. Yet there was considerable variation 9 7 across countries. reflecting the dif- TOTAL 100 100 100 100 ferent policy orientation of the six social funds in the sample which ranged from largely urban post- earthquake reconstruction in Armenia to a concentration on poorer areas by the four Latin American social funds (see Table ~) . 14 Development Outreach WORLD BANK I NSTITUTE Within communities, social fund investments disproportionately reached poorer households, reflect- ing a demand from poorer house- HONDURAS holds even within mixed-income communities. DECILE Education Health Water Sewerage Latrines Targeting results also varied con- (Poorest) 1 14.4 6.9 19.3 2 ~----~~--~~----~~~----~1 3ˇ ~ .7------~------~ siderably by type of sub - project, as illustrated by the results from Nicaragua and Honduras presented in Table 3. Across countries, posi - tive discrimination towards poor households was best reached by latrine and health projects and rea- sonably reached by education and water projects, while sewerage proj- ects clearly benefited the better- off. NICARAGUA Finally, comparisons with other DECILE Education Health Water Sewerage Latrines programs showed that social funds' (Poorest) 1 8.1 9.4 2.9 14.4 geographic and household targeting 2 4. 13:5 t.7 generally compared favorably with 3 11.4 4.9 that of other targeted social pro- 4 &.2 grams, general social spending and 7.2 muniCipal-level transfers . In Peru U the social fund had the most pro- 7.0 poor geographic distribution of U expenditures among the three pro - grams compared: the social fund, another national social infrastruc- ture program (INFES), and a target- ed national nutrition program (PRONAA). In 1995 the social fund allocated 20 percent of its resources for educational infrastructure to the poorest district decile, ly significant reductions in infant and child mortality of close compared with about 8 percent for PRONAA and 7 percent for to 50 percent over a 4 year period in communities with social INFES. In Bolivia, where the social fund now serves as a cofi- fund-financed health investments when compared to similar nancing agent to municipal governments, the social fund had communities that did not benefit from social fund projects. a pro - poor expenditure pattern, while general municipal Positive health outcomes were also found for investments in transfers were concentrated in the better- off municipalities. water and sanitation across a range of countries and types of The poorest municipalities, accounting for 42 percent of investments. Bolivia's population, received 63 percent of social fund expenditures in 1993-99 but only 22 percent of total munici- How to reach the poor: Some lessons pal expenditures in 1996. In Armenia, several social pro- grams-including child allowances and disability THE FINDINGS ON SOCIAL FUNDS' ability to reach the poor benefits-were better targeted than social fund investments, suggest several lessons regarding poverty targeting, especially though other programs such as student stipends and humani - of community-driven development initiatives. tarian aid were less well targeted. First, the policy orientation of institutions is critical. Beyond results on poverty targeting, the research showed Whereas this lesson may seem self-evident, it is clear that that utilization increased following the social fund-financed those social funds, which actively used strategies to reach the investments in infrastructure. In health centers, increases poor, including poverty maps, tailored menus, and promotion were often found for overall levels utilization or occaSionally campaigns exhibited a more progressive distribution of social concentrated among particular types of services, notably for fund investments. This lesson calls for enhanced pro-activity maternal and infant health care. This increased utilization in terms of promoting social funds' availability in poor com- translated into a range of improved health outcomes in many, though not all, cases. In Bolivia, the only country where the SOCIAL FUNDS evaluation was able to assess the social fund's impact on mor- co ntinued on page 35 tality, the social fund investment resulted in large, statistical- MAY 2005 15 Social Marketing of Bednets in Tanzania BY ROSE NATHAN, HADJI MPONDA, AND Health Research and Development Centre (IHRDC) have HASSAN MSHINDA implemented social marketing of insecticide-treated nets in Tanzania. Here, we describe a specific social marketing pro- MALARIA IS STI LL a devastating disease in sub Saharan Africa gram, known as KlNET, for insecticide-treated nets in two where it kills at least one million people every year (UNICEF rural districts of Tanzania, Kilombero and Ulanga, imple- ~003). Children are most vulnerable to malaria attacks, which mented by IHRDC. The program aimed at achieving substan- kill more than 3,000 children every day, largely in Africa. At tial and sustainable use of insecticide-treated nets in young least 100 Tanzanian children die daily because of the disease . children and pregnant women. The purpose of the program Another high-risk group is pregnant women. However, all was well in line with the philosophy of social marketing; it was other adults are also exposed to substantial risk of malaria. In therefore envisaged as the most appropriate technique to Tanzania, control and care of malaria puts a huge burden on reach the target population with the insecticide -treated nets financial and economic costs both at micro and macro levels, as a malaria control intervention. thus translating into enormous poverty implications. As such, malaria has a potential in slOwing down the achievement of Setting up the program the MDGs in countries where it is endemic as is the case in many parts of Tanzania. THE PROGRAM IMPLEMENTED delivery of treated nets in five Many efforts and resources have been invested in search of phases. By the end of the third year of the program implemen- effective malaria control strategies. Among few of the strate- tation' a population of about five hundred thousand people in gies that have shown effectiveness are insecticide -treated nets 11~ villages had been covered. Phase one of the implementation (ITN). Such nets have been proven to be effective, feasible covered the area with a demographic surveillance system (DSS). intervention for reducing malaria morbidity and mortality. Community participation was made central right from the At the African Summit on Roll Back Malaria, African lead- designing stage of the program. Community members were ers signed the Ahuja Declaration, with the commitment to given a primary role in shaping the implementation activities. protect 60 percent of African children with a treated net by the The research team in partnership with district health man- year ~005 (RBM ~ooo). agement team held sensitization meetings with village lead- The issue of how the nets should be effectively delivered to ers. The meetings, in a form of open discussions included the poor communities has been a subject of debate. general health issues, prevention of malaria sustainability and Formulating effective and sustainable mechanisms that guar- cost-recovery. Community preference studies were conduct- antee access to ITNs by the most vulnerable has been a chal- ed to identify size, quality and color of their choice. Through lenge. Social marketing programs offer a way to increase local market research a brand name "ZUIA MBU" (a kiSwahili demand through promotion at the same time supplying nets at phrase which means prevent mosquitoes) was identified for subsidized prices. treated nets and insecticide. The concept Marketing package SOCIAL MARKETING is an approach where the experience of Pr oduct commercial marketing is applied to a product which has a The nets were dark green high quality polyester in two social benefit, with the main motivation being social sizes: (100 x 180 x 150 cm and 130 x 180 x 150 cm). The sizes improvement rather than financial gain to the marketer were suitable for the local sleeping places. Insecticide water- (Andreason, 1995). based formulation of lambdacyhalothrin (ICON TM) was Largely Population Service International (PSI) and Ifakara packed in 6 rnl sachet. 16 Development Outreach WORLD BANK I NSTITUTE Promotion The distribution system Formative research was conducted at baseline to explore The distribution network of the ITNs included retail agents community perceptions of severe childhood disease. In col- in each village and wholesalers in each division. The retailers laboration with District Health Management Team results of were chosen jointly by project staff and community members; the research were used to develop a range of promotional they included private shopkeepers. community leaders. health materials. Promotional materials included: billboards posted workers and priests. A reward system was used for retailers and along main roads. posters. leaflets. exercise books used at pri - wholesalers for reaching certain sales targets . Over time. inac- mary schools. T- shirts. umbrellas . caps. and point- of- sale tive retailers were replaced. Insecticide retailers. in the initial stickers and flags. distribution area were provided with bicycles to be able to offer Information Education and Communication (lEC) semi - door-to - door treatment services. As the area expanded. the nars were held for the sales agents. and groups of specially insecticide was sold as a dip-it-yourself kit containing a pair of recruited village resource people once in every six months. gloves and instructions. The kits were sold through shops The resource people included village leaders. village health (often were the same shops that sold treated nets). workers. primary school teachers and Maternal and Child The program relied on collaboration with public entities Health (MCH) aides. such as the district health management team and Ministry of MAY 200 5 17 Health, the private sector such as international and local sup- households into five wealth quintiles. The survey and the pliers, local businessmen, and other non-governmental analysis were done for the year of the start of the program, organizations involved in health. 1997, and three years after, ~ooo . Price The coverage, measured by percent of households with at Price was based on what the community indicated they least one net, improved from 37 per cent in 1997 to 73 per cent were willing to pay and experience from previous net projects. at the end of ~ooo . Coverage among the households catego- The prices were near to cost recovery-consumer prices were rized as poorest improved from ~o per cent to 54 per cent, set at TZS 3000 (US$ 5.00 in 1997) for a net, and TZS ~50 (US$ while among the least poor (rich) households it increased from 0 . 4~ in 1997) for a sachet of insecticide. 63 per cent to 9~ per cent. The poor- Targeting est/ least poor ratio of the coverage The program aimed at targeting the increased from 0.3 to 0.6 over the most - at - risk group, pregnant women three -year period (Nathan et al., ~004)' and young children. To achieve that, a discount system was developed . The sys- tem was based on a simple paper vouch - Conclusion ers issued through the MCH clinics. The vouchers were given to pregnant women SOCIAL MARKETING was associated when they visited MCH clinics for ante- with rapid overall improvements in net natal care as well as to mothers of chil- coverage, and the pace of change was dren under five years of age . The vouch- higher among the poorest than the least er was worth TZS 500 (approx. US$ 0.5) ; poor. It should however, be noted that therefore it allowed the beneficiary to this success happened in the presence purchase a Zuia Mbu net from a retailer of two enabling factors: the existing at a reduced price of TZS ~,500 (instead demand for mosquito nets, which was of TZS 3000). The retailers were reim- extremely high probably because of bursed on their next order with an addi- perceived mosquito nuisance, and the tion ofTZS 50 for each voucher as a han- existing active private sector for nets dling charge. Implementation of vouch- (Nathan et al ~004) . "" er scheme reflected a successful public private mix. Rose Nathan, Ifakara Research and Development Center, Tanzania. Reaching the poor Hadji Mponda, Ifakara Research and Development Center, Tanzania. KINET program had several features Hassan Mshinda, Director of Ifakara which qualified it as a pro-poor initia- Health Research and Development Center, tive. Those included: Tanzania. ˇ Adoption of social marketing as a strategy to deliver the insecticide- References: treated nets - by default SM has no Andreasen, A.R. Marketing Social Change: motive of financial profit. Changing Behavior to Promote Health, Social ˇ Use of discount system facilitated Development, and the Environment. San access to pregnant women and young Francisco: Jossey-Bass, 1995. children without excluding the poor- Armstrong Schellenberg J.R .M. et al. "The KINET est. project-An overview. Improving child survival in rural Tanzania: Insecticide treated nets for malar- ˇ The remotest rural settlements where ia control in the Kilombero valley (Unpublished), the poorest are concentrated were 2001. reached with the insecticide -treated Nathan, R. et al. Mosquito nets and the poor: nets through the established delivery Can social marketing redress inequities in system. access? Tropical Medicine and International The program assessed the extent to Health 9 , (2004) 1121- 1126. which it reached the poorest in the served population. This Roll Back Malaria. The African summit on Roll Back Malaria, Abuja, Nigeria, was done through annual household coverage surveys within April 25th, 2000. (WHO/CDS/RBM/2000.17). RBM Geneva; 2000. the DSS area. Each household was asked whether they owned a UNICEF. "Malaria Is Alive, Well & Killing more than 3000 Children a Day net and a similar question for other speCified assets. in Africa ." WHO and UNICEF call for urgent increased effort to Roll Back Using the reported ownership of assets, quality of houses, Malaria. (2003). hUp://www.unicef.org and occupation of the head of household, a statistical analysis (Principal Component Analysis) was done to categorize the 18 Development Outreach WORLD BANK I NSTITUTE Community-Based Health Insurance in Rwanda BY FRAN<;OIS PATHE DIOP AND JEAN gross domestic production (GDP) has grown at a yearly rate DAMASCENE BUTERA above 6 percent between 1995 and ~001 . and social infra- structures have been rebuilt with support from the interna- RWANDA HAS LIVED one ofthe mosttragic moment of its his- tional community. Rwanda remains. however. one of the tory with the genocide of 1994. which resulted in nearly one poorest countries in the world: per capita GDP is still under million deaths and the destruction of the social fabric of the $300; the incidence of poverty is as high as 60 percent of the country. Since 1994. however. the country is being rebuilt: population, and reaches 66 percent in the rural areas where MAY 200 5 19 tive bureaus to regulate contractual relations between members and the mutual organization. Participation in the CBHI scheme is voluntary and is based on a membership contract between the CBHI scheme and the member. In addition, CBHI schemes develop contractual relations with health care prOvider organizations (health centers, hospitals) for the purchasing of health care. Bylaws of CBHI schemes and their contracts with health care providers include measures for minimizing risks associated with health insurance (adverse selection, moral hazard, cost escalation, and fraud). The target population of individual CBHI schemes are inhabitants of the catchment's area of their partner health center: low risk events (health center package) which are included in the CBHI ben- efit package are shared at the partner health center catchment's area population. CBHI schemes in a nearly 90 percent of the population live (Ministry of Finance given health district, however, establish a federation at the and Economic Planning, ~oo~). district level which plays a risk-pooling mechanism function Mutual aid and community solidarity value systems have for high- risk events (hospital package). The district federa - remained resilient traits of Rwanda's society and continue to tion also plays social intermediation and representation roles be translated in coping strategies in the health care area. In all for individual CBHI schemes in their interactions and con- local communities, associations of hamac carry the sick to tractual relations with health care providers and external health facilities . Resources are specially collected in neigh- partners. Finally, the federation plays other support func- borhoods and cells to face emergencies; structured tontines tions, such as training, advice and support, information, for are more and more organized at the cell level in order to face individual schemes. priority needs in general, medical care needs in particular. Contributions to the CBHI scheme funds are on a yearly Little attention has been paid in the past to these cultural basis. Members have the option to sign up as a family with up traits of Rwanda's society within partnership and community to seven members, which costs US$7.6 per faInily per year. involvement frameworks in health development strategies. Payment of the yearly premium entitles covered family mem- After the 1994 war, however, mutual aid initiatives have bers to a benefit package which includes all preventive, cura- emerged in the health sector as community responses to the tive services, prenatal care, delivery care and laboratory reintroduction of user-fees in public and mission health exams, drugs on the MOH essential drug list, and ambulance facilities. Building on these community initiatives, health transport to the district hospital provided by the partner authorities and non- government organizations have moved health centers. With a health center referral, members also these emerging strategies to a deliberate strategy of building receive a liInited package at the district hospital. Sick mem- community-based health insurance schemes in the health bers pay a co-payment of US$0 .30 for each visit at the health sector. center. At the hospital, refereed members have direct access to the hospital package without any co -payment. Health cen- Communi~-based health Insurance ters playa gatekeeper function to discourage the inappropri - schemes (CBHI) ate use of hospital services (Schneider et al., ~001). Since 1998, a cumulative process of learning in the com- BUI LDING ON the experiences of mutual health organizations munity-based health insurance area, involving CBHI schemes which have emerged in the country, the Ministry of Health of the pilot districts and CBHI schemes in other districts, has (MOH) initiated in 1998 pilot experiments in the health dis - been launched in Rwanda. Such a learning environment has tricts of Byumba, Kabgayi and Kabutare, which played a key facilitated the emergence of innovative strategies for role in the design and organization of CBHI schemes in the strengthening existing CBHI schemes in pilot districts and country. It provided also a platform for the compilation of implementation of new CBHI schemes in other parts of the information to support the assessment of CBHI schemes, and country. These local initiatives, while maintaining the techni- to familiarize health sector actors and partners with the strate- cal design of the pilot phase, have built on the decentralization gies needed to support their implementation on a large scale. movement underway in the country, and partnerships CBHI schemes in Rwanda are health insurance organiza - between local adIninistrative structures, grassroots associa - tions based on a partnership between the community and tions' and micro -finance schemes (banques populaires) to health care providers. The CBHI schemes develop their strengthen local support systems of CBHI schemes and to bylaws, organizational structures including general assem- increase enrollment in the schemes. blies' board of directors, surveillance committees and execu - 20 Development Outreach WORLD BANK I NSTITUTE ciency gains in the consumption of health care services. Second, sick members pay small out-of-pocket co -payments at the health centers. Consequently, out-of- pocket payments are reduced significantly among CBHI scheme members as demon- strated by the comparison of members and non-members of CBHI schemes' out-of- pocket payments in Figure ~. Greater access of the poor to CBHI scheme benefits are being promoted through two main strategies. First, build- ing on partnerships between CBHI schemes, grassroots associations and micro-finance schemes (banques popu- laires), existing and newly formed grass- roots associations are motivated to enroll as a group in the CBHI schemes under a financing scheme where the micro- CBHI and the poor finance schemes provide small loans to the associations' members to pay for their yearly contributions to the CBHI CBHI SCHEMES have experienced an important growth dur- schemes. Such a financing scheme has boosted enrollment of ing the past five years in Rwanda. From one CBHI scheme in the poor in the CBHI schemes. In addition, it has opened 1998, to sixty in ~OOl. Starting in ~OOI, an adaptation phase opportunities for poor CBHI members for greater access to drawing on lessons learned and recommendations from the larger micro -finance loans to finance income-generating pilot phase extended the number of CBHI schemes and activities. Such financial arrangements developed as a conse- increased enrollment rates in individual schemes: conse- quence of the institutional arrangements between CBHI quently, on July ~oo3, ninety-seven CBHI schemes, covering schemes, micro -financing schemes and health centers, and half a million Rwandans, were functional in the country. The innovations introduced by local actors. development of CBHI schemes is currently in an extension Second, non-government organizations and administrative phase: in ~oo4, two hundred and fourteen CBHI schemes districts are building on the institutional bridges between the have developed all over the country as a result of the combined community, the CBHI schemes and health care providers to effects of promotional activities of central authorities finance the enrollment of the poorest, indigents and vulnera- (Ministry of Health and Ministry of Local Mfairs), provinces, ble groups (orphans, widows, people living with HNIAlDS). districts, local health personnel, local opinion leaders and Under these demand-based subsidy schemes, community non-government organizations. In mid-~oo4, national cov- leaders play administrative functions in the identification of erage of CBHI schemes is estimated at 1,7 millions Rwandans: the poorest and indigents and vulnerable groups, the CBHI about ~1 percent of the Rwanda population are currently ben- schemes manage the consumption of health care for these efiting from CBHI coverage in the health sector (N dahinyuka, groups, while the subsidies are financed by non-government Jovit. ~oo4)' organizations and administrative districts who serve as inter- As a consequence of the removal of financial barriers to mediaries for primary sources of finance (state, external aid). access to health care by CBHI schemes, members of CBHI schemes are four times more likely to seek modern health care Main lessons when sick than non- members (Diop, ~ooo). The household survey results of the pilot phase summarized in Figure 1 have WHILE THE EXTENSION ofCBHI in Rwanda is still underway, been replicated based on routine data from health centers the experience of the past five years provides valuable lessons during the pilot phase and recent results from health centers for the development of micro health insurance schemes in in the same pilot districts and results from health centers in developing countries. First, the development of CBHI in the districts which have implemented CBHI schemes between Rwanda built on an incremental approach which drew lessons ~OOI and ~oo3 (Butera, ~oo4)' CBHI schemes coverage has from internal experiences and external experiences of pre- also increased the use of reproductive health services, includ- payment schemes in Southern Mrica and mutual health ing prenatal care and delivery care; they had no effect, howev- organizations in Western Mrica. The M0 H provided the lead- er, on the use of family planning services. ership to initiate the pilot phase, and secured technical assis - As a result of their insurance function, CBHI schemes pro - tance from USAlD\Rwanda and Abt Associates Inc., which tect the income of their members against financial risks associ - improved on the technical design and organization of CBHI ated with illness through two mechanisms. First, when sick, schemes in the country. The MOH kept a respectable distance members of CBHI schemes seek care earlier resulting in effi - from the design and management of the schemes to ensure MA Y 200j 21 the autonomy and the appropriation of the schemes by com- munities and local health providers. It generated information HEALTH SECTOR R EFORM on the performance of the schemes and convened multiple c ontinu e d from pag e 9 forums for stakeholders to exchange experiences and to debate on the consequences and implications of the CBHI schemes on the Rwanda health system. Such an incremental ing the consolidation of the social insurance scheme and to not approach provided a platform for learning and drawing policy only cover all its population, and in particular all poor, but to directions for the development of CBHI in the country. improve efficiency and quality in the sector. The transforma- Second, as consensus built- up on the benefits of the CBHI tion from the system that Colombia had in 199~ to the new schemes, a multi -level leadership developed in the country to insurance based system has been difficult politically and provide support to the adaptation and extension of the administratively, as well as technically challenging. The schemes. Political leaders at the central level, starting from decentralization process that started in the early nineties, the Presidency, called for the mobilization of all actors to sup- introduced particular circumstances into the organization of port the implementation of CBHI schemes throughout the the health sector providing both advantages and disadvantages country. Local communities were motivated by the MOH sup - to the implementation process of the 1993 health reform. cA> port in designing and establishing CBHI schemes; such sup - port was boosted by the Ministry of Local Affairs involvement Maria-Luisa Escobar is Lead Health Economist, Human Development in promotion activities. At the province and district levels, Department Latin America & The Caribbean Region at the World Bank. prefects and mayors continue to playa key role in coordinat - ing promotional activities . At the grassroots levels, cell and The ideas expressed here are the author's responsibility and do not sector representatives are playing a key role in sensitization necessarily represent those of the World Bank and its affiliated activities, along with health personnel and local opinion lead - institutions. The author is grateful to Panagiota Panopoulou. P.hD .ˇ ers . Such a multi-level leadership has strengthened the legit- for her careful data analysis and research . imacy of CBHI in the country and enabled the mobilization of intersectoral support for the development of the schemes. References: Third, the involvement of decentralized entities and non- Bitran, R. et al. Risk Pooling, Ahorro y Prevenci6n: Estudio Regional de government organizations in CBHI promotion activities under a Pollticas para la protecci6n de los mas pobres de los efectos de los shocks de salud. Estudio de caso de Colombia. The World Bank, August 2004. policy environment where community development was a cen- tral theme, mobilized intersectoral action, resulting in local Committee on the Consequences of Un insurance. Hidden Costs, Values Lost initiatives which improved access of the poor to CBHI benefits. Uninsurance in America. Washington , D.C: The National Academies Press, 2003 . Partnerships between local micro-finance schemes, CBHI schemes, and grassroots associations have widened opportuni- Committee on the Consequences of Un insurance. Health Insurance Is a Family Matter. Washington, D.C: The National Academies Press, 2002. ties for the poor to access CBHI and micro-finance credit. Access of the poorest and indigents to CBHI benefits is being Colombian Nacional Department of Statistics, Encuesta de Caracterizaci6n Socioecon6mica. CASEN, 1993. strengthened, due to the use of CBHI schemes as intermediate local solidarity funds in the targeting of demand -based subsi- Colombian Nacional Department of Statistics. Encuesta Nacional de Calidad dies to the poorest and indigents in the health sector by non- de Vida. ECV, 1997. government organizations and administrative districts. .,., Colombian Nacional Department of Statistics. Encuesta Nacional de Calidad de Vida. ECV, 2003. Franc;ois PatM Diop and Jean Damascene Butera, Abt Associates Inc . Colombian Nacional Department of Statistics. Encuesta Nacional de Hogares. ENH, 1992. References: Colombian Nacional Department of Statistics. Encuesta Nacional de Butera, Jean Damascene and Francois Diop. SysMmes d 'Assurance Basee sur Hogares. ENH, 2000. la Communaute : Utilisation, CoOts et Financement des Soins de Sante de Encuesta Nacional de Demograffa y Salud. DHS , 1986, 1990 y 1995. Base au Rwanda. PRIME IllRwanda Rapport Technique (July 2004). Encuesta Nacional de Demograffa y Salud. Profamilia, 2000. Diop, Francois, Pia Schneider, Damascene Butera. Summary of Results: Escobar, M.L. and Panopoulou , P. Colombia The Economic Foundation Prepayment Schemes in the Rwandan Districts of Byumba, Kabgayi, and of Peace, Health Chapter 22. The World Bank: Washington, DC, 2003. Kabutare. Technical Report No. 59 (September 2000). Bethesda, MD: Partnerships for Health Reform Project, Abt Associates Inc. Escobar, M.L. " EI Sector Salud en Colombia , Logros y Desafios." Presentation given to the Colombian President in Auraca, Colombia. Ministry of Finance and Economic Planning, "A Profile of Poverty in Rwanda: August, 2004 A Report Based on the Results of the Household Living Standards Survey." Republic of Rwanda: February 2002. Ministry of Social Protection, Bogota, DC. Several excel files on insurance affiliation . Ndahinyuka, Jovit. Etude de cas sur les rtJles des acteurs dans Ie developpe- ment des mutuelles de sante au Rwanda. Projet AWARE-RH, 2004. Nunez, J. and S. Espinosa. " Asistencia Social en Colombia: Diagn6sticos y Propuestas." Nacional Department of Statistics. Draft unpublished, Bogota, Schneider, Pia, Fran~ois Diop, Daniel Maceira, and Damscene Butera. November 2004. Utilization, Cost, and Financing of District Health Services in Rwanda. Technical Report No. 61 (March 2001). Bethesda, MD: Partnerships for Wagstaff, A.and E. Van Doorslaer. Catastrophe and impoverishment in paying Health Reform Project, Abt Associates Inc. for health care: with applications to Vietnam 1993-1998. 22 Development Outreach WORLD BANK I N STITUTE Contracting Health Care Services for the Rural Poor The Case of Cambodia BY INDU BHUSHAN, ERIK BLOOM, were inconsistent and morale was low. The primary health care BENJAMIN LOEVINSOHN , AND system was not able to deliver an adequate level of services. J. BRAD SCHWARTZ Basic services like immunization were not being provided and the child mortality rate remained at very high levels. CONTRACTING NGOS to manage the primary health care sys - tem was found to be an effective means to increase service cov- The "coverage plan" erage and achieve a more pro-poor distribution of services in rural areas of Cambodia. In the mid- 1990s, war and political TO ADDRESS THESE ISSUES, the Ministry of Health (MOH) upheaval had left Cambodia with limited health care infra- proposed contracting NCOs to manage at the district level of structure, especially in rural areas. There were sufficient para - the public health care system using a results- based contract to medical and management staff, but training and quality of care monitor progress. The contract required the NCO to provide MA Y 200 5 23 management and technical support to help the public health tract-out, in which the contractors had complete management system efficiently, and equitably provide primary health care responsibility for service delivery, including hiring, firing and services to rural populations. Because of the innovative ness of setting wages, procuring and distributing essential drugs and the approach, this was originally done on a pilot basis. supplies, and organizing and staffing public health facilities; The MOH devised a "coverage plan" which defined a min- ii) contract-in, where the contractors worked within the MOH imum package of activities comprising preventive and cura- system to strengthen the existing district administrative struc- tive services such as immunization, family planning, antena - ture and health care personnel with government supplied tal care, and provision of micronutrients. With financing pro - drugs and consumables, and a nominal budget supplement for vided by the Asian Development Bank, MOH conducted a staff incentives and operating expenses; and iii) government, in large-scale experiment of contracting with NGOs for the which the management of services remained with the govern- delivery of these primary health care services as part of the ment District Health Management Team (D HMT). government overall coverage plan. In 1997, prior to health facility con- supplied drugs and consumables and the same nominal budg- struction and procurement of equipment, a pre-contract et supplement for staff incentives and operating expenses pro- baseline household survey was taken in twelve rural opera - vided to the contract- in districts. The three remaining candi - tional health districts. The five -year contracting experiment date districts were not contracted and not formally included in started at the beginning of 1999 and a final evaluation survey the experiment. These districts continued under government was taken at the end of ~003 . management, but did not receive a budget supplement. As The districts included in the experiment were randomly such, these three districts serve as a comparison group for the assigned to one of three health care delivery models: i) con - other nine contracted and government managed districts . MOH used a competitive bid- ding process to select NGOs based on the quality of technical proposals and cost. Precisely defined, objectively verifiable health car e service indicators were measured for all twelve dis - tricts using data from the base- line survey, and well - defined goals for improvement in service coverage and coverage of the poor were specified for all dis- tricts. Health service indicators included child immunization and vitamin A., antenatal care, delivery by a trained birth atten- dant, delivery in a health facility, knowledge and use of birth spac- ing. and use of health facilities for illness. An equity goal to tar- get services to the poorest one- half of the population also was mandated for all districts. The nine districts formally included in the contracting experiment were made up of two contracted -out, three contract- ed- in, and four government dis - tricts. Including the three dis - tricts not formally in the experi- ment' the twelve districts are spatially separated in three dif- ferent provinces, and each had a population of 100,000 to ~oo,ooo . At the beginning of the experiment in 1999, the twelve districts had a combined total of more than 1.5 million people. 24 Development Outreach WORLD BANK INSTITUTE Results SOURCE OF FUNDS CONTRACT-OUT CONTRACT-IN GOVERNMENT GOVER NM ENT-OTHER THERE WERE LARGE INCR EASES in the cover- Private out-ol-pocket 14.29 16.72 19.86 19.99 age rates of health services in all twelve districts, contracted and government managed, however TDTAL 19.12 20.19 21.60 21.69 the contracted districts achieved much higher SOUUI' S(Il'.',H'j J BrdiJ ,,)()(11 ()c,tEfh('I,trl(~'-,()ll((! ' r l ' ! fl'rJ1'1( <' ')1'\,1'-,1 coverage rates than the government districts. The CrHlltJOdlr1 ADB \Llllpul 11"i11!'d) immunization coverage rate in the contracted -out districts, for example, increased from ~5 . 3 per- cent in 1997 to 8~ percent in ~oo3 , an improvement of 56,7 per- to a greater degree in contracted districts than in government centage points (Figure 1). With only one exception (births with a districts. At the end of the experiment, private out- of- pocket trained attendant in contracted- out districts), the contracted expenditures in the contracted districts were Significantly districts achieved larger increases in coverage rates than the gov- lower than those in the government districts. Public expendi - ernment districts . Government districts increased coverage tures in contracted - out districts, for example, were US$3 .o9 rates for all health services, but these increases were smaller per capita higher than in government districts, but this high- than in contracted districts and failed to achieve the coverage er public expenditure is associated with a US$5 .57 per capita targets fo r vitamin A, antenatal care, trained birth attendant, and lower level of private out - of- pocket expenditure compared modern birth spacing. In general, the difference in the higher with government districts. For all contracted districts, on coverage rates achieved by contracted districts compared with average , a higher public expenditure of about US$~.50 per lower coverage in government districts was largest for facility- capita led to about a US$4 .50 per capita lower private out- of- based services (antenatal care, trained birth attendant, births in pocket expenditure. Moreover, total public plus private out- a facility, and use of public facilities for illness) than vertical of-pocket expenditures in contracted districts were lower public health programs (immunization, vitamin A, and use of than in the government districts . The larger substitution of modern contraceptive methods) . Independent assessments of public for private expenditures in contracted districts bene- the quality of care also indicated that the contractors improved fited those with a lower ability to pay for health care services the quality of services provided at health facilities more than the more than in government districts, and the overall efficiency government over the same period. of the health care system in contracted districts was better than in the government districts . Benefits to the Poor Conclusion CONTRACTED DISTRI CTS generally outperformed the gov- ernment districts with changes in the distribution of health IN SUMMARY. the results ofthis experiment ofNGO contract - care services toward a more equitable or pro - poor distribu - ing in rur al Cambodia indicate that while all districts tion. Before the experiment, the non- poor were more likely to increased health service coverage rates, the contracted dis - use public health care services in all twelve districts . tricts outperformed the government districts in achieving Concentration indices indicate the provision of health care higher coverage rates and providing a more pro- poor distri- services in the contracted districts became more equitable or bution of services. In addition, private out-of-pocket health more pro-poor by the end of the five-year experiment than in care expenditures in contracted districts we re lower than gov- the government districts (Figure ~). There was a change ernment districts, which clearly benefits those who can least toward a more pro- poor distribution in contracted districts affo rd to pay. NGOs appear to be more responsive to contrac- for health services with only two exceptions (vitamin A for tual obligations to effectively and equitably provide health contract - out and facility birth delivery for contract - in). care services than standard government provision of services Government districts , on the other hand, changed toward a given the same goals. Overall, the results suggest contracting mo r e pro- poor distribution only for vertical programs primary health care may be an efficient and effective means to (immunization, vitamin A, and modern birth spacing), and increase health care coverage rates and better target primary these changes were smaller than the improvements made by health care services to the poor. "'" contracted districts. Government districts moved toward an even less pro-poor distribution for facility- based services Indu Bhushan is Director, Pacific Department, Asian Development Bank. including antenatal care, trained birth delivery, birth in a Erick Bloom is Economist, Social Sectors Division , Mekong facility, and use of public facilities for illness. Department, Asian Development Bank. Not surprisingly, the annual public recurrent expenditure Benjamin Loevinsohn is Senior Public Health Specialist, South Asia per capita on NGO contracted districts was conSiderably high- Human Development Department, The World Bank. er than the public expenditure fo r government districts (Table J. Brad Schwartz is Lecturer, Department of Economics, University 1) . Technical assistance for district management provided by of North Carolina at Chapel Hill. NGOs and salaries paid to health care workers largely account for these differences . It appears that public expenditures, References: Schwartz, J. Brad , 2001. "Cost Effectiveness on Contracting however, substituted for private out - of-pocket expenditures Health Care Services in Cambodia," ADB (unpublished). MAY 200 5 25 Overcoll1ing Barriers Health Equity Fund in Cambodia BY BRUNO MEESSEN AND POR IR each of them having one public health center. The district hospital is in the small town of Damdek. It provides the full IN LOW - INCOME COUNTRIES. userfees have been promoted complementary package of activities foreseen by the national as a strategy to tap more resources to public health facilities. policy: internal medicine, pediatrics, obstetrics-gynecology But they may also constitute a barrier limiting utilization of and surgery. public health services by the poor. The barrier is particularly InI999, the Ministry of Health, Medecins Sans Frontieres, and critical for hospital care as the technicality, intensity and UNICEF agreed on a common approach to boost the activities in duration of the care delivered to an inpatient often leads to the health district. They introduced a new scheme, called the significant costs. Moreover, most of the population has no 'N ew Deal, ' which establishes strong incentives for the hospital hospital in its immediate vicinity. Transportation cost also and health center staff to deliver quality health services to the constitutes a heavy burden for the poorest. population. Income collected through user fees was expected to To tackle this equity problem, most governments have finance an important part of the motivational scheme. There decreed that the poor should be treated for free . The existing was however an obvious risk that the poorest would not benefit evidence shows that waivers and exemptions through regula- from the improvement in the service delivery. In order to avoid tion, in most cases, do not work: few poor can benefit from it the pitfalls discussed above , the three partners decided to and many beneficiaries are not the poorest. As an organiza- review the mechanisms dealing with hospital access. tion, why would the hospital accept to bear a cost without com- First, a special fund was established. Both international pensation? Indeed, every poor patient leads to more medical agencies agreed to commit to an earmarked funding for and paramedical work, more drug consumption, more cater- enhancing access by the poor to the hospital services, the so- ing and more troubles. On the other hand, fees are just one of called Health Equity Fund (HEF). Straight from the start, it the many costs for the patients. The poor traditionally live far was decided that the HEF should cover all the costs the poor from hospitals . In order to benefit from free medical care, have to overcome to access hospital care: the user fees, the they have to cover other costs such as transport. Moreover, transportation cost, items for facilitating their hospital stays there may be a lot of uncertainty about eligibility for waivers and some social care if necessary. and exemptions. The poorest may then decide to stay at home The next question was whom the HEF should be entrusted foregoing the treatment, or seeking inappropriate care. There to. Both international agencies were not interested in manag- is also a problem at the facility level. If the hospital staff has ing it: they did not feel to have enough expertise in this some freedom to decide whom to waive, one can expect that domain; they were expensive and non-sustainable solutions. people with some formal or informal connections with the The other option was to entrust the HEF to the hospital itself. hospital will manage to be among the beneficiaries. Social But some conflicts of interests were expected. In a model capital is not an attribute of the poorest. where a hospital staff manage such a fund, there are some per- verse incentives: the most lucrative strategy for the hospital is The Health Equity Fund experience III to spend the fund, as quick as possible, in paying their own Sotnikum, Cambodia user fees, whatever the profile of the beneficiaries. Incentives for targeting the poorest and for addressing external barriers SOTNIKUM HEALTH DISTRICT is a poor rural area, with are then limited. around ~~o.ooo inhabitants. It is divided into 17 health areas, Henceforth, the decision was to subcontract the manage- 26 Development Outreach WORLD BANK I NST ITUTE ment of the REF to a local social welfare non-governmental One year later, a second person was hired to improve presence organization (NGO) . Several benefits were expected from that in the hospital, follow - up of supported patients and informa - choice. First, one could expect some expertise and commit - tion sharing at community level. ment to care for the poorest from this type of organization. The recruitment of candidates for assistance was based on The expertise was necessary to be able to identify correctly the three ways: (1) referral by the hospital cashier who found that beneficiaries but also to tailor the assistance. The commit- the patients could not pay for the admission fee and those with ment was necessary to be responsive to the needs of the poor- referral letter from the community; (~) active recruitment in est, including defending their rights and dignity during their the wards by the NGO welfare worker (some patients are able stay in the hospital . Second, the local NGO was a low-cost and to pay the admission fee thanks to the assets they have sold or sustainable option. Thirdly, sponsors were concerned for the debts they have taken before coming to the hospital); (3) enforCing some accountability mechanisms upon the fund spontaneous applicants who have heard in their community manager (e.g. observation of leakage to non- poor should be about the existence of the REF. sanctioned). Subcontracting to an agent potentially replace - Identification is done at the hospital through interviews by able (managing a REF has a low entry cost, there are quite a lot using a set of questionnaire. The interviews focus on informa- of local NGOs in Cambodia) was perceived as a way to guaran- tion about food security, ownership of land and productive tee actual benefit to the poorest. assets , housing, occupation, household size and structure, as Quite some freedom was given to the local NGO in the well as social capital. Physical appearance, including clothing, development of the strategy for recruiting, identifying and often gives an indication of socio - economic status. The 'target assisting the poor. Experience has permitted to progreSSively group' of the REF consists of the extremely poor, as well as the design the best organizational set - up. Initially, a single poor who risk falling into extreme poverty. No fixed criteria employee was based in an office in the hospital compound. for eligibility are used, as poverty has many dimensions that are difficult to measure. Some room is left for subjective judg- As shown in Figure I , the HEF has turned the hospital into ment by the welfare worker. Judgment indeed matters: first to a real pro - poor health facility. On average, 40 percent of hos - entitle or not the candidate, second, to tailor the assistance pital inpatients received some assistance. Monitoring has package according to the specific needs of those who have constantly confirmed that beneficiaries were actual poor. been entitled. Leakage to non -poor is not an issue. If there is any problem, it The level of financial assistance is indeed determined on a is one of under- coverage. case -by- case basis, from partial payment of the admission fee The breakdown ofthe HEF expenditure shows that admin- to full coverage of the total cost of hospitalization, including istration costs, including management, identification and transport, food and basic items. Presence of the social worker social follow-up , are under control (Figure ~). An important in the hospital compound allows frequent visits in the ward. If share of the fund goes back to the hospital through the user necessary, the support can be readjusted. fees. This is of course a strong incentive for the hospital staff to provide good services to poor patients. It is important to Results notice that the transfer in kind to the assisted patient is much more than what is paid by the HEF. Indeed in Cambodia, a part IN SOTNI KUM HOSPITAL. the HEF assisted his first patient in of hospital costs are covered by the government or the donors September ~ooo. After four years, it has become an important through input financing (salaries, drugs, equipment and building block for the good performance of the health system. buildings) . By paying on average US$1l .5 for a poor patient, the HEFs give him an access to a benefit of more than US$48. Lessons learned IN SOTNIK UM. the introduction of a HEF managed by a local NCO appeared to effectively improve access to hospital care for the poor. As long as the services delivered in the hospital are meeting the standards, one can expect a Significant health outcome for the beneficiaries. During the first year, the HEF may have mainly reduced the cost of care for people who had already chosen to access care. The years after, the steep increase in utilization indicates that a considerable number of the 'new' patients were from poor households who would not have sought care at the hospital without financial support. It is important to note that in the Cambodian context most of a household's health expenditure takes place outside the public sector, often spent on poor quality treatment by informal private practitioners. Therefore, in terms of poverty prevention, the greatest poten- Year 1 Year 2 Year 3 Year 4 tial of the HEF does not seem to lie in financing expenditure SOIl{(P HHljPIIJc'1i p/ ,1/ JOOI 'spp RefpreJ'les' and alll'lOrs c,/Cli/a/IOIi' in the public sector, but in preventing unnecessary expendi- ture in the private sector, by encouraging the use of adequate public health services. The scheme has some limitations. A better coverage would be achieved if the poor got entitled for assistance before the episode of illness (e.g. through a 'poor card'). Uncertainty on eligibility and assistance would then be removed. Some exper- iments with such an approach are going on in Cambodia. Another observation has been that welfare workers were quite keen on being very specific in their targeting (avoid the leakage to non-poor) but quite reluctant to deliver proximity social care to the poorest, a kind of retreat into administrative tasks. The HEF model may be relevant to settings other than Cambodia. Similar approaches are being developed in China and some countries of Africa. Subcontracting to NCOs is OVERCOMING BARRIERS co ntinu e d o n pa g e 3S 28 Development Outreach WORLD BANK INSTITUTE Reproductive Health Services through Mobile Camps SEWA Experience in GUJarat BY M. KENT RANSON, PALAK JOSHI, of members from 11 of the state' s ~5 districts, the organization MITTAL SHAH, AND YASMIN SHAIKH has two main goals. First, to organize women workers to achieve full employment, i.e . work security, income security, THE SELF - EMP LOYED WOMEN ASSO CIATION (SEWA) is a food security, and social security. Second, to make women trade union of women who earn a living through their own individually and collectively self- reliant, economically inde - labor or small business, started by Ela Bhatt in 1 97~ . pendent and capable of making their own decisions . Headquartered in Ahmedabad (Gujarat, India) , and inclusive SEWA first became actively involved in the public health M A Y 2005 29 field in the early 1970S through health education and provision of maternity benefits. A focus of SEWA Health has always been to build capacity among local women, especially traditional midwives (dais), so that they become the barefoot doctors of their communities. Today, SEWA's health- related activities are many and diverse, and include: primary health care, delivered through 60 stationary health centres and mobile health camps; health education and training; capacity building among local SEWA leaders and dais; provision of high-quality low- cost drugs through drug shops; occupational and mental health activities; and production and marketing of traditional medi- cines. It has been a primary objective of these health services to provide to the very poor, particularly those living in areas not otherwise served by government or non -governmental organi- zations (NGOs). The services provided by SEWA Health are available both to SEWA members-of whom there are currently 468,000 in Gujarat state (calendar year ~004)-and non- mem- bers alike. The reproductive health mobile camps In response to demand from people in remote and under- serviced areas, SEWA Health began organizing reproductive 2511.1-------- health (RH) mobile health camps for women in 1999. RH mobile camps are carried out mainly in slum areas of 2O%r-------- Ahmedabad city and villages of three districts and are funded 15%1-------- largely by the United Nations Population Fund (UNFPA) and the Government of India. More than 35 camps are carried out 10% 1 - - - - - - - - per month, and the mean attendance per camp is 30 women, for a total of more than 1~,500 patients per annum. Health 5%r------ care at the camps is provided by empanelled physicians and 50 barefoot doctors and managers. The camps are repeated in 2 3 4 5 6 7 8 9 10 each area, on average, once per year. Activities at the RH mobile camps include health education Source M Kent Rallson Pollak J,,,I,, ~iltL'1 511dl1 dlill YaSlllll' 5ilillkll and training, examination and diagnostic tests (including Indld AsseSSing tile Reach of T!1rl't) S[I.'IA Hl'dlth Su /h t ') II'IUll) Pl(, Pour 111 Reaching the Poor G\'/dtklll .\!(l~'::>tdft dll(] Ydll '( K [ j,tor:-, cervical examination and Pap smears), treatment, referral and Forthr:ollll!lf' World Ban:" \Va"rllrl~trm )U(ll follow -up. Camps are usually held during the afternoon, and their duration is three to four hours. Those attending the camps are asked to pay a Rs. 5 (0.11 USD) contribution, and one -third of the total cost of medicines prOvided (although even these fees may be waived for those who are very poor). 158 rural women as they attended randomly selected camps. Increasingly in rural areas, SEWA Health is conducting We then compared these women to the general urban and rural these camps in collaboration with the Government of Gujarat, populations of Gujarat, using recent, representative surveys. with camps"held right at government primary health centers We found the RH mobile camps to be very effective at (PHCs), which are usually located in or near small villages. reaching poor women in Ahmedabad City. A comparison These camps differ from the standard "area" camps based on a composite SES index showed urban camp users to (described above) insofar as medicines are given for free, the be significantly poorer than the population of Ahmedabad. range of medicines available are restricted to those on the Camp users (and their families) were, for example, signifi- government's formulary, and health care is prOvided by pub - cantly less likely to possess a motorcycle or scooter (1 ~ percent lic doctors and nurses. Free transportation is provided by vs. 43 percent), were more likely to rely on public (vs. private SEWA to women living in neighboring villages. or shared) toilets (~~ percent vs. 9 percent), and were less likely to use natural gas as a source of cooking fuel (35 percent Reach among the poor vs. 66 percent). Figure 1(a) illustrates the distribution of urban camp users by deciles of the SES index score-the left- IN ORDER TO ASSESS the socio-economic status (SES) ofthe ward skew of this graph indicates that camp users were more women using the RH mobile camps, we surveyed 376 urban and likely to be from poorer segments of the general population. 30 Development Outreach WORLD BANK INSTITUTE The percentage of camp users fall ing below the 30 th decile of There are likely to be other, broader reasons underlying the SES score-which roughly approximates the poverty line in the difficulties in delivering services to the rural poor. Studies India-was 5~ percent (Figure 1) . in other SEWA departments have documented similar dis- In rural areas, the camps were less effective in reaching crepancies in the equity of utilization of rural versus urban poor women. Rural women did not differ significantly from the services. For example, the poorest rural members of SEWA's general, rural population in terms of their SES index score. insurance scheme (Vimo SEWA) have lower rates of claims Figure 1(b) indicates that the majority of rural camp users are than the less poor. Reasons for this differential include: from less-poor deciles of the population. Only 5.7 percent of ˇ Problems of geographic access, both to inpatient facilities use rs fell below the 30 th and to Vimo SEWA's grassroots workers; percentile-suggesting that SEWA ˇ Weaker "links" between members and Health's rural RH mobile camps do local Vimo SEWA representatives in rural not effectively target the very poorest. areas (i .e. the contact between members and the organization is less frequent, and What worked and why less intensive, in rural areas); ˇ Weaker capacities among Vimo SEWA FOR THE MOST PART. the urban grassroots workers in rural areas. services seem to be effectively tar- Already, SEWA Health has taken steps geting the poor. Reasons for this to improve the accessibility of the rural RH success are likely to include: mobile camps. SEWA Health waives the ˇ Services (especially RH mobile registration fee and the medicines fee for camps and women's education those who appear to be particularly sessions) are offered" right at peo- poor-typically a few women presenting to pIe's doorsteps", i.e. SEWA Health each camp. Perhaps these exemptions takes the services to the poor, could be granted more liberally, and in a rather than trying to bring the manner more objective, for example, by poor to the services; providing exemption to all those who pos- ˇ The services are delivered by (or at sess a below poverty line (BPL) card. least in part by) the poor them- It must also be remembered that failure selves; of a service to reach the poorest of the rural ˇ The services are generally com- poor does not necessarily mean that the bined with efforts to educate and service has failed in "reaching the poor." mobilize the community; for Even those households that fall in the high- example, preceding the RH mobile er deciles of the SES index in rural areas camps, SEWA Health workers go should be considered "less poor" rather door-to-door, educating people than "wealthy." Compared to their urban about the service, and educating counterparts, these rural households have people to use it; less in the way of cash reserves, material ˇ Costs are low (certainly relative to wealth, and thus economic security. the private for-profit sector); In conclusion, the findings of this study ˇ SEWA is an entity that people know suggest that delivery of services through a and trust. broad - based, development - oriented During in -depth interviews, union can facilitate equitable delivery of SEWA Health workers attributed the health care services. Government and scheme's success in reaching the poor dono rs can help to ensure that established to the fact that it treats poor people NGOs, with an interest in providing health with respect and "warmth." The fact that services are deliv- services, have the capacity and the resources to do so. ~ ered "to their doorsteps" was also seen as contributing to the success of SEWA Health services. Finally, the fact that SEWA Kent Ranson , Lecturer in the Health Policy Unit, London School Health's services are delivered largely by women, was also of Hygiene and Tropical Medicine, and a Research Coordinator, perceived as increasing the reach among poor women. Self-Employed Women's Association (SEWA) Insurance. Our in- depth interviews with SEWA Health grassroots Palak Joshi, Research Associate with SEWA Health. workers suggest that there are two main barriers that prevent Mittal Shah, Coordinator of the SEWA Health Team. poor rural women from using the RH mobile camps. First, for Yasmin Shaikh, Supervisor for Reaching the Poor project, SEWA. some, the 5 rupee registration fee prevents some from attend- ing the camps. Second, the camps may be difficult for women to attend, as they often coincide with hours of work. MAY 2 0 0 j 31 Do Participatory Prograllls Work? Improving Reproductive Health for Disadvantaged Youth in Nepal BY ANJU MALHOTRA , SANYUKTA MATHUR , have access. Yet, while this situation is recognized, little is ROHINI PANDE , AND EVA ROCA known about what works to improve youth access to reproduc- tive health, and, in particular, to improve the access of those THE WORLD CURRENTLY HAS its largest generation of youth who are most deprived. ever, with over a billion young people between the ages of 10 Development practitioners point to participatory approaches and 19, most of whom live in developing countries (UNFPA as effective in increasing empowerment and accountability, two 2°°4). Youth in many countries marry and begin families of the key factors in improving health services for the disadvan- while still in adolescence, yet they are often denied access to taged (WDR 2004) . At the community level, participatory sufficient information or appropriate services for prenatal approaches can lead to increased awareness, information, social care, delivery, and HIV, gaps that directly affect not only their support, and client power for community members in accessing lives but also the future well -being of their societies. The most appropriate health services. Community-based participatory disadvantaged- poor, rural, female youth-are least likely to approaches may be particularly effective for adolescent repro - 32 Development Outreach WORLD BANK INSTITUTE ductive health since community awareness is an important pre- FIGURE 1: DELIVERY IN A HEALTH FACILITY: FIRST cursor to improving attitudinal, normative, and institutional PREGNANCY, POOR AND RICH YOUNG MARRIED WOMEN constraints that lead to poor reproductive health among young people (Norman 2001, Mensch et alI998). Adolescents them- I_ POOR 50% ˇ RICH 50% I selves may be more likely to increase their knowledge base, crit- l00r------------------------------------, ical thinking, and decision-making abilities on intimate issues 90r-----------------------------~ related to sexual and reproductive health if approached in a con- ~r---~~------~~----~WL----~ sultative and inclusive manner (McCauley and Salter 1995; 70 Senderowitz 1998). However, to date, there have been no com- prehensive evaluations conducted on the effectiveness of a com- . ~ 60 50 munity-based participatory process for adolescent reproductive ~ 40 ~ health programs in developing countries, and, in particular, in ... 30 reaching poor and otherwise disadvantaged youth. This paper 20 addresses this gap, and reports on a study that tested the ability 10 of participatory approaches to improve reproductive health out- O~~~--_.~------._~~----~--~--~ comes for the most disadvantaged youth in Nepal. Baseline Study 1+-21 EndUn. SIUdy 18-25 Baseline Control 14-21 Emlline Control 18-25 (PoorAlich: 0.32) (0.54) (0.24) (0.35) Youth reproductive health needs are acute in Nepal. Despite a large youth population and chronically poor out- Source Malhotra. A S Mathur. R Pande, and [ Roca (2004) The Impact of a CommuOity-based Participatory Program on SoclOeconorTllc comes on a number of reproductive health indicators among Disadvantage In Youth Reproductive Health Outcomes In Nepal" Paper pre sen ted at the Reaching the Poor conference February 18 20, Wasillng/on, DC young people, this issue has received limited programmatic and policy attention. Early marriage, a strong predictor of reproductive risk, is nearly universal in Nepal: girls currently systemic barriers to youth reproductive health. Further, youth, marry at an average age of 16, and 52 percent have begun parents, and other community members were actively engaged childbearing by the age of 20. Among those giving birth, 55 in implementing study site program activities through a wide percent of girls under age 20 reported receiving any prenatal variety of community- based groups set up during the project. care, 14 percent of the births were attended by trained per- In contrast, in the control sites, project staff designed and sonnel, and only 9 percent of deliveries were in a health facil- implemented standard reproductive health interventions that ity. Less than 7 percent of married girls in the 15 - 19 age group addressed only the most immediate risk factors such as STDs reported using any method of contraception. Rural girls in or unwanted pregnancies. Finally, socioeconomic disadvan- Nepal, who are typically poorer than their urban counterparts, tages-based on gender, rural-urban residence, wealth, eth- are further disadvantaged as compared to girls in urban areas nicity, schooling status, and marital status-were a speCific (Ministry of Health, Nepal, 2002). focus of the intervention design and approach in the study sites, whereas this was not the case in the control sites. Study design We measured poverty by household asset ownership. While poverty is critical, it is not the only disadvantage that can keep THE NEPAL ADOLESCENT PROJECT (NAP) was a 5-yearproj - young people from accessing appropriate information and ect conducted from 1998 to 2003, in collaboration with an services around reproductive health. Thus, the study looked at international service delivery organization (EngenderHealth), multiple types of disadvantage among young people in addi- an international research organization (International Center tion to poverty, namely, gender, rural -urban residence, and for Research on Women), and local Nepali NGOs (New ERA education status. We chose prenatal care, delivery at a health Ltd. and BP Memorial Health Foundation). To test the effec- facility, and knowledge of HN transmission as important tiveness of participatory versus non -participatory approaches reproductive health outcomes for which to examine the to youth reproductive health, we implemented programs in impact of various types of disadvantage among young people. urban and rural study and control sites, with four sites in total. Data for this study come from baseline and endline cross-sec- In the study sites, there was a focus on involving the com- tional quantitative surveys, as well as qualitative and partici- munity and actively engaging disempowered groups, such as patory methods. The target age group at baseline was 14-21 the poor, young women, and ethnic minorities, at every stage of year old males and females, married and unmarried. At the the program. Study site program activities took into account end of the intervention program, data was collected on 14-25 broader development priorities voiced by diverse members of year olds, so as to capture all youth who could have participat - the community. Thus, interventions aimed at improving ed in or benefited from the project. We collected data from a youth-friendly services, peer education and counseling were total of 965 households at the baseline and 1003 households at linked with broader interventions prioritized by the commu- the end -point of this study. nity and aimed at improving the socio-economic environment and opportunities for youth. These included adult education Results programs, activities to address social norms, and access to eco- nomic livelihood opportunities. Consequently, the entire RESULTS SHOW that generally, the participatory approach was intervention package addressed structural, normative, and more successful in reducing the gap in reproductive health out- MAY 2 0 0 5 33 comes among youth with disadvantaged backgrounds as com- learned to enforce higher expectations from prOviders. pared to those with advantaged backgrounds. Our analyses also ˇ The study sites focused on altering not just reproductive show that for different health outcomes, different aspects of dis- health outcomes, but changing fundamental social norms advantage were important. Being from a rural area or a poor and institutions. This created an enabling environment for household were key constraints for getting prenatal care, while good reproductive health for youth by generating a better gender and lack of education were the key constraints for having understanding and new mindset in the communities, and accurate knowledge of HIV transmission. In most cases, the par- leading to a substantial increase in demand for such serv- ticipatory intervention approach was better able to overcome the ices, even among the disadvantaged. impact of these constraints on reproductive health outcomes than was the more traditional approach (Malhotra et al ~004) . Conclusions and implications ˇ Delivery in a health facility. At baseline, both the study and control sites showed substantial differences between rich OUR RESULTS SHOW that small -scale community efforts can and poor young women's access to a health facility for achieve empowerment and accountability. Specifically, par- pregnancy delivery (Figure 1) . By the endline, poor young ticipatory approaches can successfully provide youth, espe- women in the study sites were closer in their access to pre - cially disadvantaged youth, the means to negotiate for appro- natal care when compared to better- off women, but a sim- priate, accessible, and accurate information and services from ilar change was not evident in the control sites. As Figure 1 parents, providers and policy makers. shows, this is because the improvement in access to deliv- Equally critically, our study points to the need for broader ery at a health facility was entirely among the poorer 50 definitions of disadvantage. There is no dispute that poverty is a percent of the population in the study site, whereas in the key and powerful measure of disadvantage. Nonetheless, in control sites, both the rich and the poor gained. many rural communities in the developing world, those who are ˇ Prenatal care. Regression results show that before the most disadvantaged owe this disadvantage to complex and intervention, an urban young woman in the study site was interwoven interactions between various contextual factors that 16 times more likely to get prenatal care than her rural need to be considered. Analyses of poverty as a measure of dis- counterpart. By the end of the project she was only l.~ advantage need to be accompanied by analyses of rural-urban times more likely to receive prenatal care. The control sites residence, gender, and educational access as other important do not show a similar improvement of access to prenatal markers of social, cultural, and economic differentials. ,,., care among rural young women (Malhotra et al. ~004). ˇ Accurate knowledge of HW transmission. In all the sites at Anju Malhotra, Group Director, Social and Economic Development, baseline, girls were less likely to be able to correctly identify at the International Center for Research on Women. at least two modes of HIV transmission when compared with Sanyukta Mathur, Public Health Specialist in the Population and Social Transitions team at the International Center for Research boys. In the urban study site, the intervention led to such a on Women. substantial improvement in knowledge among girls that the Rohini Pande, Social Demographer in the Population and Social proportion of girls who were knowledgeable about HIV actu- Transitions team at the International Center for Research on Women. ally surpassed the proportion of boys. A similar change was Eva Roca , Research Assistant in the Population and Social Transitions not observable in the control sites. At the same time howev- team at the International Center for Research on Women. er, neither type of intervention was able to substantially References: reduce the difference in knowledge regarding HIV among Malhotra, A. , S. Mathur, R. Pande, and E. Roca . "The Impact of a the educated and the uneducated (Malhotra et al. ~004) . Community-based Participatory Program on Socioeconomic Disadvantage in youth Reproductive Health Outcomes in Nepal.' Paper presented at the Why did the participatory approach Reaching the Poor conference, February 18-20, 2004, Washington, DC . work? McCauley, A.P. and C. Salter. "Meeting the Needs of Young Adults," Population Reports. Baltimore, MD: Johns Hopkins School of Public Health, Population Information Program, 1995. Series J: 43 . OUR EVALUATION SUGGESTS that the participatory approach Mensch, B., J. Bruce, and M. Greene. "The Uncharted Passage: Girls' succeeded because its defining characteristics lent them- Adolescence in the Developing World ." New York, NY: The Population selves well to the problems of adolescent reproductive health. Council, 1998. ˇ The participatory intervention design made the young Ministry of Health (Nepal), New ERA, et al. Nepal Demograph ic and Health people active players in their own health, primarily by tap- Survey 2001. Calverton , Maryland: Family Health Division, Ministry of ping into and strengthening their existing social networks Health; New ERA; and ORC Macro, 2002. for information exchange and counseling: to an increasing Norman, Jane. "Bu ilding Effective Youth-Adult Partnerships." Transitions 14(1), 2001. extent, poor young people could rely on better informed peers rather than professional services for a number of Senderowitz, J. Involving Youth in Reproductive Health Projects. Research, Program and Policy Series. Washington , DC: FOCUS on Young reproductive health needs . AdultslPathfinder International, 1998. ˇ The participatory intervention empowered youth and adult World Development Report 2004. Making Services Work for Poor People. community members to demand accountability from Washington, DC and New York: The World Bank and the Oxford University providers and policymakers by building decision-making Press, 2003. structures and coalitions. In particular, young people NFPA. 2004. Website: http:www.unfpa.org/adolescentslfacts.htm 34 Developmenr Outreach WORLD BANK IN STI TUTE SOCIAL FUNDS O VERCOM I NG B ARRI E R S continued from pag e ' 5 c ontinu e d f rom pa ge 28 munities and setting targets for investments in poorer areas. probably not a necessary condition, but the purchaser - This is important to redressing the historical underfunding of provider split is. Donor money can be helpful for the pilot poorer areas typical of most countries. The results from stage, but the REF model is very possible with public money. Bolivia showed a correlation between improved poverty tar- It allows the government to kill two birds with the same stone: geting over time and growing fiscal decentralization, which (1) target the poor; (2) develop its public health system. For may hold lessons for other cases. the ease of targeting, it is important to acknowledge that the Second, there may be a tradeoff between improved house- model will probably be easier to implement in countries hold targeting and the use of more open menus of eligible where socio-economic differentiation is increasing. In investments. Social funds have defined menus featuring basic regions stricken by general extreme poverty, very low user services more likely to be demanded by the poor as a targeting fees (or no fees at all) are probably still the best approach . .,., tool, a strategy supported by the household targeting results showing significant differences in outcomes among different Bruno Meessen, Institute of Tropical Medicine, Antwerp, Belgium types of projects. Should menus be further restricted to Por Ir, Belgian Technical Cooperation, Siem Reap , Cambodia improve targeting results? That would eliminate some of the potentially positive attributes of greater choice fo r communi - References: ties. But more open menus may allow better- off households Hardeman, W., W. Van Damme, M. Van Pelt , P.lr, K. Heng, and B. Meessen, and communities to capture benefits. At the very least, for 2004, "Access to hea lth care for all ? User fees plus a Health Equity Fund in investments that tend to benefit the better- off, developing Sotn ikum, Cambodia, " Health Policy and Planning, vol. 19 (1) , pp. 22ˇ32 . more rigid screening criteria including introduction of data on income levels of potential beneficiary households may help to reduce leakage. Third, the re is a need to exploit complementarities among social funds, demand -side interventions, and more targeted social assistance . For example, social fund interventions could complement a demand -side subsidy enabling the very poorest parents to send their children to school as well as sup- port the provision of nutritional supplements through health centers. Indeed access to quality supply-side investments are a prerequisite to the provision most effective demand - side approaches , with the latter being able to directly target Subscribe resources to households or individuals, complementing the to a flagship magazine that is on former's provision of broad community benefits. the cutting edge in Finally, an increased emphasis on improving intra- district the field of global targeting would also be worth exploring, especially in more het- knowledge for development, erogeneous areas, such as urban centers. This initiative could reaching 20,000 explore a variety of approaches from improving poverty maps to readers in 130 allow for greater geographical disaggregation to engaging com- countries. munity members in identifying the poorer beneficiaries. Beyond the lessons on poverty targeting, the results of the Advertise social fund impact evaluations show that although this to reach a unique approach has been successful in generating a range of positive international audience of business leaders, policy makers, welfare outcomes, social funds cannot operate effectively government officials, academics, without well-coordinated social policy both at the central and economic journalists, research local level. .,., institutions, and civic organizations. Laura B. Rawlings is Country Sector Leader for Human Development for Central America, The World Bank. This article summarizes findings presented in Rawlings, Laura, Lynne Sherburne-Benz, and Julie Van Domelen " Evaluating Social Funds: A Cross-Country Analysis of Community Investments." World Bank Regional and Sectoral Studies, 2004. MAY 200 $ 35 \ ' OICES FRO\I TIlE FIELD Medicinal Value of Indigenous Plants Ensures Livelihood in South African Communities BY EMMANUEL KORO SOUTH AFRICAN FEMALE traditional leaders are taking a groundbreaking approach towards promoting rural liveli - hood and development through the management of indige - nous knowledge systems and sustainable exploitation of natural resources with nutritional and medicinal values, This project was prompted by other projects that are currently being implemented in South Africa, which com- mercialize the use of indigenous plants to alleviate poverty and promote job creation, As long as these projects contin- ue without a strategy that promotes conservation of these resources, there is a high risk of over-harvesting indige - nous plants. Already, urban dwellers are flocking to rural areas looking for herbs with medicinal and nutritional properties that suppress the impact of HN/ AIDS related illnesses. The female traditional leaders cited the over- harvesting of the African potato by the urban dwellers as a potential to lift these communities out of poverty through good case in point. to demonstrate the threat that South sustainable commercial exploitation of the plants. Africa's indigenous plants were facing. To achieve sustainable utilization of indigenous plants, The traditional leaders interviewed said that indigenous the female traditional leaders said that the management of plants played an important role in their daily socio-eco- indigenous knowledge systems should link traditional and nomic needs . The plants are used to cure diseases such as modern conservation methods. diarrhea, skin rush, rheumatism, and arthritis, and heal This project is jointly funded by the Kellogg Foundation wounds, cough, and headaches. Other plants are also used and the European Union's CODEOSUB (Conservation and as laxatives at birth. They are also sources of food for the Development Opportunities from Sustainable Use of communities, wherein lies the link between the nutrition- Biological DiverSity in Communal Lands of Southern al and medicinal values of indigenous plants. Africa) . South Africa's Centre for Scientific and Industrial Just as medical doctors sometimes tell patients what Research (CSIR) and Resource Africa, a South Africa-based kinds of food they should or should conservation agency formed a part- not eat when ill, the female tradi - nership in July ~oo4 , to jointly imple - tionalleaders said that this was sim- "Voices from the Field" provides ment the project. ilar to the way they administered first-hand insight into issues of The indigenous knowledge sys - traditional medicines in their com- current concern to the tems project also seeks to assist rural munities. It is therefore clear that development community. To communities to protect their rich while the nutritional and medicinal participate, send your stories to: knowledge on the functions of indige - values of indigenous plants are crit- devoutreach@worldbank.org. nous plants from being illegally ical to the general upkeep of rural Make your voice heard. acquired and patented by western residents, they also have a huge pharmaceutical and food companies. 36 Development Outreach WORLD BANK I NSTITUTE THE SOUTH AFRICAN GOVERNMENT RECENTlY INTRODUCED A BIODIVERSITY ACT TO PROTECT INDIGENOUS KNOWLEDGE SYSTEMS . Meanwhile, unscrupulous pharmaceutical companies have continued to make super profits from the illegal explOitation of indigenous knowledge that establishes the nutritional and medicinal values of indigenous plants and other related resources, without benefiting the sources of that knowledge. However, the initiative by South Mrica's traditional leaders to manage their indigenous knowledge system is expected to further promote benefit sharing between communities and the private sector. ~ Emmanuel Koro is President of Sub-Saharan Africa Forum for Environment Communicators (SAFE) To ensure sustainability and better management of the indigenous plants, the project will focus on the female tra- ditional leaders' role in managing the harvesting and exploitation of the plants . Through traditional fares, work- shops, and media publicity, awareness could be created in the communities on the best practices to manage and sus- tain indigenous knowledge. Female traditional leaders interviewed requested the South Mrican Government's assistance to protect their indigenous knowledge systems. They said that this could be achieved through greater enforcement of the recently introduced Biodiversity Act that requires researchers investigating the values of indigenous plants to declare their intentions to do so, and also make it clear that they would enter into benefit sharing agreements with commu - nities from where they acquire the knowledge on the values ACKNOWLEDGEMENT: This research was made possible through of indigenous plants. the Kellogg Foundation and European Union grants. The European The South Mrican Government has a good track reco rd Union is funding the project "Conservation and Development for promoting the need for benefit sharing between Opportunities from the Sustainable Use of Biological Diversity in indigenous communities and the private or public sectors, the Communal Lands of Southern Africa" (CODEOSUB). The through its development agency, CSIR. About two years CDDEOSUB Project is focused on promoting successes in ago, CSIR and the San Communities signed a benefit shar- Community Based Natural Resources Management (CBNRM) ing agreement with a US- based pharmaceutical company, projects within Southern African countries that signed the UN Pfizer. The agreement regulated the sustainable and com- Convention on Biodiversity and are implementing the CBD work mercial exploitation of an indigenous plant, the Hoodia programme within the arid and semi-arid ecosystems. The project Gordonii, which contains the compound PS7 that suppress - involves building the capacity of communities and NGOs involved. es one's appetite and helps reduce fat. Under this agree- The CODEOSUB Project targets eight SADC countries namely, ment, the San Communities (the acknowledged source of Botswana, Malawi, Mozambique, Namibia, South Africa, Tanzania, knowledge on the medicinal value of the Hoodia Gordonii) Zambia and Zimbabwe. were granted 6% of all royalties, if the product was suc- cessful. This agreement has set an important precedent that makes it unethical or morally incorrect for companies that fail to sign benefit -sharing agreements with commu- nities to add value to products. MAY 200 5 37 BO()I'SIILI.F HEALTH FINANCING FOR POOR farthest from the development goals and faces the PEOPLE: RESOURCE toughest challenges in accelerating progress. The report MOBILIZATION AND RISK finds that without rapid action to accelerate progress. the SHARING, Edited byAlexanderS. MDGs will be seriously jeopardized-especially in Sub- Preker and Guy Carrino The World Saharan Africa. which is falling short on all the goals. It Bank. 2005. calls on the international community to seize the This book stresses that community opportunities presented by the increased global attention financing schemes are no panacea to development to build momentum for the MDGs. for the problems that low-income countries face in resource WORLD DEVELOPMENT mobilization. They should be INDICATORS 2005 , The World regarded as a complement to-not as a substitute Bank. 2005. for-strong government involvement in health care World Development Indicators pres- financing and risk management related to the cost of ents the most current and accurate illness. The authors conclude by proposing concrete information on global develop - public policy measures that governments can introduce to ment on both a national level and strengthen and improve the effectiveness of community aggregated globally. This informa- involvement in health care financing. tion allows readers to monitor the progress made toward meeting the goals endorsed by the United Nations and its member MEASURING EMPOWERMENT: countries. the World Bank. and a host of partner organiza- CROSS-DISCIPLINARY tions in September 2,001 in their Millennium PERSPECTIVES, Edited by Deepa Development Goals. It includes over 80 tables and over Narayan. World Bank. 2005 800 indicators for 152, economies and 14 country groups. Building on the award -winning as well as basic indicators for a further 55 economies. The Empowerment and Poverty Reduction report contains six thematic presentations of analytical sourcebook. this volume outlines a commentary covering: World View. People. Environment. conceptual framework that can Economy. States and Markets. and Global Links. be used to monitor and evaluate programs centered on BEYOND THE CITY: empowerment approaches. It THE RURAL CONTRIBUTION presents the perspectives of 2,7 distinguished researchers TO DEVELOPM ENT, by David de and practitioners in economics. political science. Ferranti. Guillermo E. Perry. Daniel sociology. psychology. anthropology. and demography. all Lederman. Alberto Valdes and of whom are grappling in different ways with the William Foster. World Bank. 2005. challenge of measuring empowerment. The authors draw Beyond the City evaluates the from their research and experiences at different levels. contribution of rural development from households to communities to nations. in various and policies to growth. poverty regions of the world. alleviation. and environmental degradation in the rest of the economy. as well as in the rural space. This title brings together new theoretical and GLOBAL MONITORING REPORT empirical treatments of the links between rural and 2005. MILLENNIUM national development. New findings are combined with DEVELOPMENT GOALS: FROM existing literature to enhance our understanding of how CONSENSUS TO MOMENTUM, rural economic activities contribute to various aspects of The World Bank. April 2005. national development. This report is the second in an annual series assessing progress on the Millennium Development Goals and related development outcomes. This year's report has a special focus on Sub-Saharan Africa- the region that is 38 Development Outreach WORLD BANK I N STITUTE DEVELOPMENT OUTREACH IN TRANSLATION ONLINE THE POVERTY, HEALTH, The online version of Development OUTREACH NUTRITION, AND POPULATION (www.worldbank.org/ devoutreach)will soon feature DEPARTMENT at the World three back issues of the magazine in translation. Over the Bank introduces work in these years, international authorities in the field of areas recently undertaken or development have contributed articles that "put currently under way, in the knowledge to work for development," as the magazine's hope that the information will masthead claims. However, until now, that knowledge had prove useful to policy makers and analysts outside as well been accessible only to English -speaking readers. Now as within the Bank. The Department includes the Reaching three Special Reports will be available for the first time the Poor program, whose principal objective is to build in Arabic, Chinese, French, Russian, and Spanish. The upon its initial work by assessing more fully the record of special reports are: Young People Count (Spring ~oo~) , health initiatives in reaching the poor; and to draw upon Sustaining the Earth (Fall ~oo~), and The Private Sector: the assessment's findings to identify ways of designing BUilding Economic Growth (March ~oo3). They can be health programs that can reach disadvantaged groups found in the online archive together with all the back more effectively. issues of the magazine in English (1999 - ~oo5) . Visit: www.worldbank.org/povertyandhealth.html healthpop@worldbank.org WORLD VOLUNTEER WEB supports the volunteer com- , /11>, 1 THE INSTITUTE OF TROPICAL ~ MEDICINE is one ofthe munity by providing a global one - stop - shop for informa - world's leading institutes tion, resources, and organiza- for training, research and tions linked to volunteerism. assistance in tropical medi- It aims to represent the diver- cine and health care in sity of volunteerism in all of its cultural forms, bringing developing countries. The global ideals to local voluntary actions. It mobilizes indi- ITM is an inter-university but autonomous post-graduate viduals' organizations and networks to help achieve the institute for specialized training, research and services Millennium Development Goals (MDG) , a set of time - provision. It carries out a wide - ranging international bound targets to combat poverty, hunger, disease, illitera - program of capacity strengthening and participates in cy, environmental degradation and discrimination against activities and collaborations allover the world towards women, and to promote ties between civil society organi - the common goal of "Health Care for All." zations' governments and individual volunteers. Visit: www.itg.be Visit: www.worldvolunteerweb.org THE INFORMATION FOR THE CARIBBEAN EPIDEMIOLOGY DEVELOPMENT PROGRAM CENTRE (CAREC) works toward (lNFODEV) works to promote improving the health status of better understanding, and Caribbean people by advanCing effective use , of information the capabilities of member and communication tech- countries in epidemiology, nologies (lCT) as tools of laboratory technology and poverty reduction and broad -based, sustainable devel - related public health opment. Its mission is to help developing countries and disciplines through technical cooperation, service, their partners in the international community to use training, research and a well -trained motivated staff. It is information and communication technologies effective - administered on behalf of ~l Member Countries by the ly and strategically as tools to combat poverty, promote Pan American Health Organization (PARO) , the World sustainable economic growth, and empower individuals Health Organization's Regional Office for the Americas. and communities to participate more fully and creatively Visit: www.carec.org in their societies and economies. Visit: www.infodev.org MAY 200 5 39 (: \ [ . [.: '\ [) \ R MAY 2005 5 World Environment Day San Francisco, CA, USA 2-27 The 2005 Review Conference of the Parties www.unep.org to the Treaty on the Non - Proliferation of Nuclear Weapons (NPT) JULY 2005 UN Headquarters, New York www.un.org/ events 6-8 G8 Summit Gleneagles, Scoltland 11-13 Carbon Expo 2005: Global Carbon www.g8.gov.uk Market Fair and Conference Cologne, Germany 19-21 From Reaction to Prevention: Civil SOciety http://www.carbonexpo.coml Forging Partnerships to Prevent Violent Conflict and Build Peace 23-24 Annual Bank Conference on Development New York, UN Headquarters Economics 2005: Securing Development www.un.org/ events in an Unstable World Amsterdam, The Netherlands 21- 22 International Conference www.econ.worldbank.org/ abcde/ on Shared Growth in Mrica Accra, Ghana 25-27 II Technical Forum on Involuntary isserŽug.edu.gh Resettlement in Latin America and the Caribbean July 30- World Youth Congress Bogota, Colombia Aug 8 Stirling, Scotland jvillegas@worldbank.org www.scotland~oo5ˇcom JUNE 2005 SEPTEMBER 2005 1 World Bank Youth Seminar, 7-9 58th Annual DPIINGO Conference Working for a World Free of Poverty United Nations Headquarters Singapore New York, USA singoffice@Worldbank.org www.un.org/ dpi oirtreach Subscription Order Form SUBSCRIBERS FROM DEVElOPING NAME COUNTRIES WILL CONTINUE TO RECEIVE THE MAGAZINE FREE OF CHARGE. TITLE ORGANIZATION SUBSCRIPTION FEE APPLIES TO READERS IN THE FOLLOWING DEVELOPED COUNTRIES: ADDRESS _______________________________________________________ Australia, Austria , Belgium, Canada , Denmark, Finland, France, Germany, Greece, Iceland, Ireland , Italy, Japan , Kuwait, Luxembourg, CITY AND STATE OR PROVI NCE _________________________________________ Monaco, Netherlands, New Zealand, Norway, Oman, Portugal, Spain, Sweden, Switzerland, COUNTRY _______________________________________________________ United Arab Emirates, United Kingdom, Un ited States, and Vatican City. ZIPI POSTAL CODE _________________________________________________ PHONE ________________________________________________________ FAX E-MAIL _______________________________________________________ Please do not send cash. Make checks payable to Development OUTREACH/WBI. o Check no. ________ in the amount of $_ _ _ is enclosed . Mail order to: Editor, Development OUTREACH The World Bank 1818 H Street, NW, Room J2-200 Washington, DC 20433 USA ˇ Flscat Management and Accountable Public Governance this book is c:oncaned with institutional Imngements thllt fosta' In im:entille environment compatible with accountable governam:e. It provides tools to anllyze the fiscal health of a government. fiscal prudenee. fiscal stress. revenue performanee. public integrity. the culture of governance. Ind government accountability to its citizens. ˇ Measuring and Monitoring Government Performance This book provides tools of analysis for measuring government pcrformanee for the whole of government, for various orders of government as well as important individual sectors. ˇ Equity in Public Spending This book provides tools of analysis to disrovcr the orientation of the public sector and to create a scorecard on its role in safeguarding the interests of the poor, the elderly, women, and those otherwise disadvantaged. The book further provides a framework for citizen-cen- tered governance, in other words, creating an institutional design with appropriate incen- tives to make the public sector responsive and accountable to a median (or average) voter. he World Bank's overall development strategy emphasizes two pillars for long-term growth and poverty T reduction : improving the investment climate and empowering and investing in people. The Investment Climate Capacity Enhancement Program was established in 2003 jointly by the World Bank Institute (WBI) and the World Bank Private Sector Development (PSD) Vice Presidency to support the implementation of this development strategy. The program's objectives are: o To familiarize clients with the importance of investment climate to growth and poverty reduction . o To promote new thinking, share knowledge and disseminate best practices on how to incorporate investment climate issues in policy formulation. o To enhance clients' capacity in assessing and improving investment climate. o To train local trainers and researchers to build capacity for policy research and training in investment climate. o To provide direct implementation and capacity enhancement support to client countries and World Bank staff. The target audience for the program includes: policy makers, practitioners and stakeholders in client countries, trainers and local partners, representatives from the international donor community, and World Bank staff. For more information, please email icprogram@Woridbank.org. www.investmentclimate.org NOTEWORTHY TITLES FROM TH E WORLD BAN K INSTITUTE The Right to Tell: The Role of Mass Media in Economic n_M _ WHY IS A FREE PRESS SO IMPORTANT? Journalists play an important role as investigators, mediators, reformers. nbil.mn/A story- tellers, and most importantly, they act as facilitators of the free flow of The Right to Tell: The Role of Mass Media in Economic Development contains an outstanding list of contributors from Nobel Prize winner and former World Bank chief economist, Joseph Stiglitz to Robert J. Shiller author of Irrational Exuberance and novelist Gabriel Contributors to this volume explore the role of the media as a _chd~)g the corporate sector, and the policies that prevent the media Right to Tell also evaluates the media's function as transmitters Gf an essential ingredient for markets to operate efficiently. This pUllIIiClnon; damaging effects that an unethical or irresponsible press can cause to a An independent press is essential to sound and equitable economic development. The media help give voic& poor and the disenfranchised and allow for a free and transparent society. World Bank Institute Otwel.pm,nt Study. 2002. 400 pages. Stock no. A15203 (ISBN 0-8213-5203-2). Price US$35. For information about country discounts, see http://publications.worldbank.org/discounts Building State Capacity in Africa Edited by Brian Levy and Sahr Kpundeh Building the capacity of African states is at the top of the continent's development agenda. After decades of failed efforts, public sector reform and capacity building programs in the 1990s began focusing on strengthening participation, transparency, and accountability and on improving state effectiveness in policymaking and service delivery. Building State Capacity in Africa looks closely at these initiatives in more than a dozen African countries, describing a paradigm shift in how to analyze and build state capacity- from a narrow focus on organizational, technocratic, and public management approaches to a broader perspective that incorporates both the political dynamics and the institutional rules of the game within which public organizations operate. Building State Capacity in Africa presents and analyzes recent experiences with supply- side efforts to build administrative capacity (administrative reform, pay policies, budget formulation), and demand-side efforts to strengthen government accountability to citizens (role and impact of national parliaments, dedicated anticorruption agencies, political dynamics of decentralization, education decentralization). World Bank Institute Development Study. 2004. 380 pages. Stock no. 16000 (ISBN 0-8213-6000-0). Price: US$30. For information about country discounts, see http://pubHcations.worldbank.orgldiscounts To Order: World Bank Publications P.O. Box 960, Herndon, VA 20172-0960, USA Telephone: 703-661-1580 or 800-645-7247 Fax: 703-661-1501 www.worldbank.org/publications AVAILABLE AT BETTER BOOKSTORES ˇ WORLD BANK The reference of choice on development Publications