The impact of conditional cash transfers on health outcomes and the use of health services in low- and middle-income countries
Conditional cash transfers can be effective in increasing the use of health services and improving health outcomes. Such programmes have lower administrative costs than service delivery programmes and are relatively quicker to implement. In low-income settings, supply-side constraints, including weaknesses in governance, the banking system and information management can influence the effectiveness of such programmes.
RHL commentary by Huntington D
Cash payments to persons living in poverty are a long-standing element of social protection programmes. Adding a conditionality to cash payments, i.e., requiring individuals receiving cash payments to undertake a specific action – such as look for a job in order to receive unemployment benefits – has also been widely used. During the past decade, use of conditional cash transfers (CCT) has spread to the social sectors, most commonly to education, but increasingly also to health. Payments are generally linked to compliance with recommendations pertaining to childhood school attendance, childhood immunization and pregnancy care (perinatal visits and nutrition). CCT programmes that support education and health-sector goals are at present an established feature of social security in many Latin American middle-income countries. It is argued that a CCT gives some degree of freedom to the beneficiaries regarding the choice of services, while simultaneously reducing pressure on government-funded services to undertake outreach services, (i.e. generate demand). The spread of CCTs to other world regions and low-income countries has contributed to increasing familiarity with different approaches to using CCTs and the range of outcomes that are associated with their introduction (1).
This Cochrane review (1) sought to assess the effectiveness of conditional monetary transfers in low- and middle-income countries aimed at improving health outcomes of populations and their access to health-care services.
2. METHODS OF THE REVIEW
The authors performed an extensive search of 24 databases for randomized controlled trials, interrupted time series analysis or controlled before–after studies. This search yielded more than 24 000 references, out of which 29 papers were identified as "potentially relevant". The authors sought studies from low- and middle-income countries that included an objective measurement of at least one of the following outcomes: use of health-care services, expenditure on health care, specific health conditions or changes in equity (e.g. changes in use of services by disadvantaged groups). Studies involving all types of provider (private, governmental, and nongovernmental organizations) were included. The types of intervention sought in the studies included direct monetary transfers made to households on the condition that people in the household perform a particular action. The authors analysed the studies systematically for risk of bias.
3. RESULTS OF THE REVIEW
A total of 10 papers met the inclusion criteria. These papers were published between 2004 and 2005, even though the authors sought papers published up to May 2009. The included papers relate to the results of six CCT programmes conducted in Brazil, Colombia, Honduras, Malawi, Mexico, and Nicaragua. The papers focused exclusively on CCTs (omitting in-kind or non-conditional transfers). Only four of these programmes included conditionalities related to reproductive health (pre/post-natal care, nutritional supplements for pregnant or lactating women, or testing of HIV status). The interventions and reported outcomes in the included studies were too diverse to combine the results in a meta-analysis.
Impact on use of health-care services
Five studies, whose quality of evidence was judged to be low by the authors, had reported this outcome. In one study involving people who had been tested for HIV, a greater proportion of (27 percentage points) of people in the intervention group (those who had received CCTs) returned to the clinic for their results compared with controls (those who did not receive CCTs). Another study found that where CCTs were offered to the population, there was an increase of 2.09% in the number of daily outpatient visits to health-care facilities. A third study reported that use of health-care services increased significantly for pre-school children, but there was no significant increase in the uptake of antenatal care or 10-day postnatal check-ups. Another study reported increases in the use of health-care services by 19.5 percentage points at 1 year and 11 percentage points at 2 years in the proportion of infants (up to 3 months old) brought to health-care centres.
Impact on health outcomes
Three studies of moderate quality were available for this outcome. The data showed that CCT programmes could impact children’s health positively, but the effect was not consistent across all age groups.
Impact on immunization coverage
Four studies of moderate quality reported the effects of CCTs on immunization coverage. One reported mixed results for immunization coverage. The second showed an increase in relative treatment effect of 6.9 (confidence interval 1–12.8) in coverage for the first dose DTP/pentavalent vaccine among children. However, an increase was neither found for tetanus immunization in pregnant women nor for measles vaccination among children. A study of a CCT programme found an increase in the probability that 24-month-old children would comply with DPT vaccination schedule (relative treatment effect 8.9; standard error 0.047). Another study found no significant impact on vaccination coverage.
Impact on anthropometric or nutritional outcomes
Six papers of moderate quality reported outcomes on anthropometric measures and nutritional status from four different CCT programmes. A study from Colombia showed mixed results regarding the impact of CCTs on the nutritional status of children: there was a positive impact on nutritional status for children under age 24 months and an increase of 0.58 kg in newborn weight in urban areas in the study, but this effect was not seen in rural areas. In Nicaragua, the study found a reduction in the magnitude of stunting and proportion of under-weight children aged 0–5 years, but no impact was found on the proportion of wasted children aged 0–5 years. An evaluation of a Brazilian CCT programme showed no effect on height-for-age measures and even a negative impact on weight-for-age for children under 7 years old.
Finally, three studies reported findings for a Mexican CCT programme. One found a significant impact on growth for the youngest children (< 6 months old) in the poorest households, but no difference for older children (aged 6–12 months) or for the youngest children from less poor families. Another study reported a similar result on height gain. A third study found a positive effect on the height of children aged 12–36 months.
4.1 Applicability of the results
Even though the data could not be pooled for analysis, the review authors conclude that CCTs can be an effective means of increasing the use of health services and improving health outcomes and nutritional status of children, echoing findings from Lisa Fernald and colleagues' study of the long-term effects of the Oportunidades CCT programme in Mexico (2), as well as other papers that did not pass the inclusion criteria of this Cochrane review. The single most important contribution of the review is that it confirms though standardized methods what CCT experts have long believed about the usefulness of the approach in initiating behavioural change.
The review represents an important contribution to the literature and augments our understanding of social protection programmes, even as it provides an interesting example of the limitations encountered when rigorous scientific selection criteria screen-out the large body of evidence on complex interventions.
4.2 Implementation of the intervention
In their discussion of the findings, the review authors move beyond the limited ability to draw conclusions across a wide range of settings and imposed by the few selected papers and rely upon a wider body of evidence to summarize key strengths and weaknesses of CCT programmes. Such programmes tend to have lower administrative costs than service delivery programmes, are relatively quicker to implement, provide freedom of choice, etc. The ubiquitous problems of targeting the poor are being overcome by health-sector CCTs by simply providing payments to all pregnant women. The methods for calculating the amount of the payments and the timing or frequency have also evolved, and today many programmes are determining cash transfers on the basis of actual financial costs of using health-care services. These and other features are driving the interest in CCT programmes and should be considered in the implementation of this intervention.
The primary potential weaknesses in any CCT programme are the quality and availability of the service the beneficiary is required to use (i.e. supply-side constraints): low-quality services do not have much health impact and where services are simply not available CCT programmes can do little to improve the health of the population. As CCT programmes begin to expand into low-income settings, supply-side constraints – including weaknesses in governance, the banking system and information management (necessary for maintaining beneficiary data bases) – will influence the programmes' effectiveness (3).
4.3. Implications for research
Santiago Cueto's commentary in The Lancet (4) correctly points out that more evidence is needed on how conditional cash transfers effect behaviour change among participating families. In fact, much more operational evidence is needed to support the continued expansion of conditional cash transfer programmes worldwide. As this systematic review suggests, there is a very large body of literature on programmatic experiences available, though much of it is observational and descriptive. Increased use of targeted operations and implementation research studies are needed, in addition to large-scale and rigorously designed experimental research designs that can be captured in systematic reviews.
- Lagarde M, Haines A, Palmer N. The impact of conditional cash transfers on health outcomes and use of health services in low and middle income countries. Cochrane Database of Systematic Reviews 2009;Issue 4. Art. No.: CD008137; DOI: 10.1002/14651858.CD008137.
- Fernald, LCH, Gertler, PJ, Neufeld, LM. 10-year effect of Oportunidades, Mexico's conditional cash transfer programme, on child growth, cognition, language, and behaviour: a longitudinal follow-up study. The Lancet 2005;374:1997-2005; DOI:10.1016/SO140-673(09)61676.7.
- Cechini, Simone. Do CCT Programmes work in low-income countries?. New York, NY: International Policy Centre for Inclusive Growth, United Nations Development Programme; 2009 (One Pager No. 90, July 2009 Poverty Practice Bureau for Development Policy).
- Cueto S. Conditional cash-transfer programmes in developing countries. The Lancet 2009;374:1952-1953.
This document should be cited as: Huntington D. The impact of conditional cash transfers on health outcomes and the use of health services in low- and middle-income countries: RHL commentary (last revised: 1 May 2010). The WHO Reproductive Health Library; Geneva: World Health Organization.