Croke K, Garcia-Mora ME, Goldstein M, Mensah E, O'Sullivan M. Up Before Dawn: Experimental Evidence from a Cross Border Trader Training at the Democratic Republic of Congo-Rwanda Border. Economic Development and Cultural Change. Forthcoming.Abstract
Small-scale cross-border trade provides opportunities for economic gains in many developing countries. Yet cross-border traders—many of whom are women—face harassment and corruption which can undermine these potential gains. We present evidence from a randomized controlled trial that provided access to information on procedures, tariffs, and rights to small-scale traders to facilitate border crossings, lower corruption, and reduce gender-based violence along the DRC-Rwanda border. The training reduces bribe payment by 5 percentage points in the full sample and by 27.5 percentage points on average among compliers. The training also reduces the incidence of gender-based violence by 5.4 percentage points (30.5 p.p. among compliers). We assess competing explanations for the impacts using a game-theoretic model based on Hirschman’s Exit, Voice, and Loyalty framework. The effects are achieved through early border crossings at unofficial hours (exit) instead of traders’ use of voice mechanisms or reduced rent-seeking from border officials. These results highlight the need to improve governance and establish clear cross-border trade regulations, particularly on the DRC side of the border.
up before dawn preprint.pdf
Abay K, Abay MH, Berhane G, Chamberlin J, Croke K, Tafere K. Access to health services, food, and water during an active conflict: Evidence from Ethiopia. PLOS Global Public Health. 2022;2 (11) :e0001015. Publisher's VersionAbstract
Civil conflict began in Ethiopia in November 2020 and has reportedly caused major disruptions in access to health services, food, and related critical services, in addition to the direct impacts of the conflict on health and well-being. However, the population-level impacts of the conflict have not yet been systematically quantified. We analyze high frequency phone surveys conducted by the World Bank, which included measures of access to basic services, to estimate the impact of the first phase of the war (November 2020 to May 2021) on households in Tigray. After controlling for sample selection, a difference-in-differences approach is used to estimate causal effects of the conflict on population access to health services, food, and water and sanitation. Inverse probability weighting is used to adjust for sample attrition. The conflict has increased the share of respondents who report that they were unable to access needed health services by 35 percentage points (95% CI: 14–55 pp) and medicine by 8 pp (95% CI:2–15 pp). It has also increased the share of households unable to purchase staple foods by 26 pp (95% CI:7–45 pp). The share of households unable to access water did not increase, although the percentage able to purchase soap declined by 17 pp (95% CI: 1–32 pp). We document significant heterogeneity across population groups, with disproportionate effects on the poor, on rural populations, on households with undernourished children, and those living in communities without health facilities. These significant disruptions in access to basic services likely underestimate the true burden of conflict in the affected population, given that the conflict has continued beyond the survey period, and that worse-affected households may have higher rates of non-response. Documented spatial and household-level heterogeneity in the impact of the conflict may help guide rapid post-conflict responses.
Croke K, Gage A, Fulcher I, Opondo K, Nzinga J, Tsofa B, Haneuse S, Kruk M. Service delivery reform for maternal and newborn health in Kakamega County, Kenya: study protocol for a prospective impact evaluation and implementation science study. BMC Public Health. 2022;22 (1727). Publisher's VersionAbstract

Background: Maternal and neonatal mortality remain elevated in low and middle income countries, and progress is slower than needed to achieve the Sustainable Development Goals. Existing strategies appear to be insufficient. One proposed alternative strategy, Service Delivery Redesign for Maternal and Neonatal Health (SDR), centers on strength- ening higher level health facilities to provide rapid, definitive care in case of delivery and post-natal complications, and then promoting delivery in these hospitals, rather than in primary care facilities. However to date, SDR has not been piloted or evaluated.

Methods: We will use a prospective, non-randomized stepped-wedge design to evaluate the effectiveness and implementation of Service Delivery Redesign for Maternal and Neonatal Health in Kakamega County, Kenya.

Discussion: This protocol describes a hybrid effectiveness/implementation evaluation study with an adaptive design. The impact evaluation (“effectiveness”) study focuses on maternal and newborn health outcomes, and will be accom- panied by an implementation evaluation focused on program reach, adoption, and fidelity.

Bundy DAP, Campbell SJ, Chami GF, Croke K, Schultz L, Turner HC. Epidemiology and Economics of Deworming. In: Helminth Infections and their Impact on Global Public Health. Springer ; 2022. Publisher's Version
Croke K, Coville A, Mvukiyehe E, Dohou CJ, Zibika J-P, Ghib LS, Andreottola M, Lokaya YB, Quattrochi JP. Effects of a community-driven water, sanitation, and hygiene program on Covid-19 symptoms, vaccine acceptance, and non-Covid illnesses: A cluster-randomized controlled trial in rural Democratic Republic of Congo. Tropical Medicine and International Health. 2022;27 (9) :795-802. Publisher's VersionAbstract

Background: The government of the Democratic Republic of Congo responded to Covid-19 with policy measures, such as business and school closures and distribution of vaccines, which rely on citizen compliance. In other settings, prior experience with effective government programs has increased compliance with public health measures. We study the effect of a national water, sanitation, and hygiene program on compliance with COVID-19 policies.

Methods: Prior to the COVID-19 pandemic, 332 communities were randomly assigned to the Villages et Ecoles Assainis program or control. After COVID-19 reached DRC, individuals who owned phones (590/1312; 45%) were surveyed by phone three times between May 2020-August 2021. Primary outcomes were COVID symptoms, non- COVID illness symptoms, child health, psychological well-being, and vaccine acceptance. Secondary outcomes included COVID-19 preventive behavior and knowledge, and perceptions of governmental performance, including COVID response. All outcomes were self-reported. Outcomes are compared between treatment and control villages using linear models.


Results: The VEA program did not affect respondents’ COVID symptoms (-0.11, 95% CI -0.55, 0.33), non-COVID illnesses (-0.01, 95% CI -0.05, 0.03), child health (0.07, 95% CI -0.19, 0.33), psychological well-being (-0.05, 95% CI -0.35, 0.24), or vaccine acceptance (-0.04, 95% CI -0.19, 0.10). There was no effect on village-level COVID-19 preventive behavior (0.02, 95% CI -0.17, 0.22), COVID-19 knowledge (0.16, 95% CI - 0.08, 0.39), or trust in institutions.

Conclusions and relevance:
Although the VEA program increased access to improved water and sanitation, it did not increase trust in government. Accordingly, there was no evidence of increased compliance with COVID policies, and no reduction in illness.


Croke K, Goldstein M, Holla A. The Role of Skills and Gender Norms in Sector Switches: Experimental Evidence from a Job Training Program in Nigeria. Journal of African Economies . 2022 :ejac007. Publisher's VersionAbstract

Industrialisation and structural change entails shifting workers from low-skill to high- skill occupations. In emerging economies, multiple constraints may impede sectoral switches among workers, including skill and spatial mismatches, and social norms related to gender in the workplace. This study uses a job training experiment across five cities in Nigeria to estimate the overall effect of training on sectoral switches into the information and communications technology and business process outsourcing sector and to examine the role of various factors that might constrain switching. After 2 years, the treatment group was 26% more likely to work in the information and communication technology (ICT)-enabled service sector, although they were no more likely to be employed than the control group. Sector switches were higher among those with sector-relevant skills, and training magnified the skills premium in switching. Switches were also higher in some cities, despite large improvements in skills in all cities. Women who were implicitly biassed against associating women with professional attributes were two times more likely to switch into ICT than unbiassed women, suggesting that training helped overcome internalised social norms among female applicants to the program. These results suggest that training can be an effective strategy for inducing sector switches and overcoming social norms that hamper female mobility in the labour market and that narrower targeting may make such sector-specific job training programs more effective.

McConnell M, Mahajan M, Bauhoff S, Croke K, Verguet S, Castro MC, Melo Furtado K, Mehndiratta A, Farzana M, Faiz Rashid S, et al. How are health workers paid and does it matter? Conceptualising the potential implications of digitising health worker payments. BMJ Global Health. 2022;7 :e007344. . Publisher's Version
Chokotho L, Croke K, Mohammed M, Mulwafu W, Bertfelt J, Karpe S, Milusheva S. Epidemiology of adult trauma injuries in Malawi: Results from a multi-site trauma registry. Injury Epidemiology. 2022;9 (14). Publisher's VersionAbstract

Background: Large scale multi-site trauma registries with broad geographic coverage in low-income countries are rare. This lack of systematic trauma data impedes effective policy responses.

Methods: All patients presenting with trauma at 10 hospitals in Malawi from September 2018 to March 2020 were enrolled in a prospective registry. Using data from 49,241 cases, we analyze prevalence, causes, and distribution of trauma in adult patients, and timeliness of transport to health facilities and treatment.

Results: Falls were the most common mechanism of injury overall, but road traffic crashes (RTCs) were the most common mechanism of serious injury, accounting for (48%) of trauma admissions. This pattern was consistent across all central and district hospitals, with only one hospital recording <40% of admissions due to RTCs.  49% of RTC-linked trauma patients were not in motorized vehicles at the time of the crash. 84% of passengers of cars/trucks/buses and 48% of drivers of cars/trucks/buses from RTCs did not wear seatbelts, and 52% of motorcycle riders (driver and passenger) did not wear helmets. For all serious trauma cases (defined as requiring hospital admission), median time to hospital arrival is 5 hours 20 minutes (IQR: 1 hour 20 minutes, 24 hours). For serious trauma cases that presented on the same day that trauma occurred, median time to hospital arrival was 2 hours (IQR 1 hour, 11 hours). Significant predictors of hospital admission include being involved in an RTC, age >55, Glasgow Coma Score score <12, and presentation at hospital on a weekend.

Conclusions: RTCs make up almost half of hospitalized trauma cases in this setting, are equally common in referral and district hospitals, and are an important predictor of injury severity. Pedestrians and cyclists are just as affected as those in vehicles. Many of those injured in vehicles do not take adequate safety precautions. Most trauma patients, including those with serious injuries, do not receive prompt medical attention. Greater attention to safety for both motorized and especially non-motorized road users, and more timely, higher quality emergency medical services, are important policy priorities for Malawi and other developing countries with high burdens of RTC trauma.

Fox AM, Choi Y, Lanthorn H, Croke K. Health insurance loss during COVID-19 increases support for universal health coverage. Journal of Health Policy, Politics, and Law. 2021. Publisher's VersionAbstract

Context: The United States is the only high-income country that relies on employer-sponsored health coverage to insure a majority of its population. An estimated 15 to 27 million Americans have lost employer-sponsored health insurance during the COVID-19-induced economic downturn. We examine public opinion towards universal health coverage policies in this context.

Methods: Using data collected through an online survey of 1,211 Americans in June 2020, we examine the influence of health insurance loss on stated support for Medicare-for-All (M4A) in two ways. First, we examine the association of pandemic-related health insurance loss with M4A support. We consider this change in insurance status a ‘structural frame.’ Second, we make introduce a ‘situational frame,’ experimentally priming some respondents with a vignette of a sympathetic victim losing employer-sponsored health coverage during COVID-19. With this, we estimate the impacts of this ‘vicarious insurance loss’ on M4A support. In both cases, we examine how political party affiliation moderates the effect.

Findings: Nearly a quarter (22%) of respondents reported personal health insurance loss in the 6 months preceding the survey. We find that directly experiencing recent health insurance loss is strongly associated (10-15 pp, p<0.01) with greater M4A support. Experimental exposure to the vignette increases M4A support by 6 pp (p=0.05), largely by shifting respondents declaring they “don’t know” about their support into supporters. We find no moderating influence of political-party affiliation in the survey experiment. However, party identification does modify the association between insurance loss and M4A support; Republicans, who report much lower levels of baseline M4A support, are more likely than Democrats to have higher M4A approval if they have lost insurance recently.

Conclusions: In the context of the COVID-19 pandemic, situational framings can induce modest change in attitudes among individuals who do not have well-formed preferences about M4A. However, the effects of real world events such as job and health insurance loss are associated with larger changes, including among self-identified partisans. This suggests that the large scale loss of employer-based insurance offers the potential for new coalitions supportive of policies to expand health insurance coverage.
Quattrochi JP, Coville A, Mvukiyehe E, Dohou C, Esu F, Cohen B, Lokaya Y, Croke K. Effects of a community-driven water, sanitation, and hygiene intervention on water and sanitation infrastructure, access, behavior, and governance: a cluster-randomized controlled trial in rural Democratic Republic of Congo. BMJ Global Health. 2021;6 :e005030. Publisher's VersionAbstract

Introduction Inadequate water and sanitation is a central challenge in global health. Since 2008, the Democratic Republic of Congo government has implemented a national programme, Healthy Villages and Schools (Villages et Ecoles Assainis (VEA), with support from UNICEF, financed by UK’s Foreign, Commonwealth and Development Office.

Methods A cluster-level randomised controlled trial of VEA was implemented throughout 2019 across 332 rural villages, grouped into 50 treatment and 71 control clusters. Primary outcomes included time spent collecting water; quantity of water collected; prevalence of improved primary source of drinking water; and prevalence of improved primary defecation site. Secondary outcomes included child health, water governance, water satisfaction, handwashing practices, sanitation practices, financial cost of water, school attendance and water storage practices. All outcomes were self-reported. The primary analysis was on an intention-to-treat basis, using linear models. Outcomes were measured October–December 2019, median 5 months post-intervention.

Results The programme increased access to improved water sources by 33 percentage points (pp) (95% CI 22 to 45), to improved sanitation facilities by 26 pp (95% CI 14 to 37), and improved water governance by 1.3 SDs (95% CI 1.1 to 1.5), water satisfaction by 0.6 SD (95% CI 0.4 to 0.9), handwashing practices by 0.5 SD (95% CI 0.3 to 0.7) and sanitation practices by 0.3 SD (95% CI 0.1 to 0.4). There was no significant difference in financial cost of water, school attendance, child health or water storage practices.

Conclusion VEA produced large increases in access to and satisfaction with water and sanitation services, in self-reported hygiene and sanitation behaviour, and in measures of water governance.

Croke K. The Impact of Health Programs on Political Opinion: Evidence from Malaria Control in Tanzania. Journal of Politics. 2021;83 (1) :340-353. Publisher's VersionAbstract
For elections to produce accountable government, citizens must reward politicians who deliver benefits. Yet there is relatively limited causal evidence of changes in public opinion in direct response to specific government programs. This question is examined in Tanzania, which has implemented large health programs targeting diseases such as HIV/AIDS and malaria. Tanzania’s 2010–11 antimalaria campaign took place concurrently with a national household survey. Exploiting discontinuities based on interview dates to estimate the effect of these programs on the popularity of local politicians, this article shows that a bed net distribution campaign resulted in large, statistically significant improvements in approval of political leaders, especially in malaria endemic areas. Effects were largest shortly after program implementation but persisted for up to six months. These findings suggest that citizens update their evaluation of politicians in response to programs, especially those that address important problems, and that the effects decay over time, but not completely.
Virk A, Croke K, Yusoff MM, Mokhtaruddin K, Abdullah Z, Hanafiah ANM, Emira Soleha Ramli, Borhan NF, Almodovar-Diaz Y, Wei-Aun Y, et al. Hybrid Organizations in Health Systems: The Case of Malaysia's National Heart Institute. Health Systems and Reform. 2020;6 (1) :e1833639. Publisher's VersionAbstract
Health system reforms across high- and middle-income countries often involve changes to public hospital governance. Corporatization is one such reform, in which public sector hospitals are granted greater functional independence while remaining publicly owned. In theory, this can improve public hospital efficiency, while retaining a public service ethos. However, the extent to which efficiency gains are realized and public purpose is maintained depends on policy choices about governance and payment systems. We present a case study of Malaysia’s National Heart Institute (IJN), which was created in 1992 by corporatization of one department in a large public hospital. The aim of the paper is to examine whether IJN has achieved the goals for which it was created, and if so, whether it provides a potential model for further reforms in Malaysia and other similar health systems. Using a combination of document analysis and key informant interviews, we examine key governance, health financing and payment, and equity issues. For governance, we highlight the choice to have IJN owned by and answerable to a Ministry of Finance (MOF) holding company and MOF-appointed board, rather than the Ministry of Health (MOH). On financing and payment, we analyze the implications of IJN’s combined role as fee-for-service provider to MOH as well as provider of care to private patients. For equity, we analyze the targeting of IJN care across publicly-referred and private patients. These issues demonstrate unresolved tensions between IJN’s objectives and public service goals. As an institutional innovation that has endured for 28 years and grown dramatically in size and revenue, IJN’s trajectory offers critical insights on the relevance of the hybrid public-private models for hospitals in Malaysia as well as in other middle-income countries. While IJN appears to have achieved its goal of establishing itself as a commercially viable, publicly owned center of clinical excellence in Malaysia, the value for money and equity of the services it provides to the Ministry of Health remain unclear. IJN is accountable to a small Ministry of Finance holding company, which means that detailed information required to evaluate these critical questions is not published. The case of IJN highlights that corporatization cannot achieve its stated goals of efficiency, innovation, and equity in isolation; rather it must be supported by broader reforms, including of health financing, payment, governance, and transparency, in order to ensure that autonomous hospitals improve quality and provide efficient care in an equitable way.
Croke K. The Origins of Ethiopia’s Primary Health Care Expansion: The politics of state building and health system strengthening. Health Policy and Planning. 2020;35 (10) :1318–1327. Publisher's VersionAbstract

Ethiopia’s expansion of primary health care over the past 15 years has been hailed as a model in sub-Saharan Africa. A leader closely associated with the programme, Tedros Adhanom Gebreyesus, is now Director-General of the World Health Organization, and the global movement for expansion of primary health care often cites Ethiopia as a model. Starting in 2004, over 30 000 Health Extension Workers were trained and deployed in Ethiopia and over 2500 health centres and 15 000 village-level health posts were constructed. Ethiopia’s reforms are widely attributed to strong leadership and ‘political will’, but underlying factors that enabled adoption of these policies and implementation at scale are rarely analysed. This article uses a political economy lens to iden- tify factors that enabled Ethiopia to surmount the challenges that have caused the failure of similar primary health programmes in other developing countries. The decision to focus on primary health care was rooted in the ruling party’s political strategy of prioritizing rural interests, which had enabled them to govern territory successfully as an insurgency. This wartime rural governance strategy included a primary healthcare programme, providing a model for the later national pro- gramme. After taking power, the ruling party created a centralized coalition of regional parties and prioritized extending state and party structures into rural areas. After a party split in 2001, Prime Minister Meles Zenawi consolidated power and implemented a ‘developmental state’ strategy. In the health sector, this included appointment of a series of dynamic Ministers of Health and the mo- bilization of significant resources for primary health care from donors. The ruling party’s ideology also emphasized mass participation in development activities, which became a central feature of health programmes. Attempts to translate this model to different circumstances should consider the distinctive features of the Ethiopian case, including both the benefits and costs of these strategies.

Croke K, Chokotho L, Milusheva S, Bertfelt J, Karpe S, Mohammed M, Mulwafu W. Implementation of a multi-centre digital trauma registry: Experience in district and central hospitals in Malawi. International Journal of Health Planning and Management. 2020;35 (5) :1157-1172. Publisher's VersionAbstract

Background: Trauma is a rapidly growing component of the burden of disease in developing countries; yet systematic data collection about trauma in such contexts is relatively rare.

Methods: This paper describes the implementation of a trauma registry in 10 government-run hospitals in Malawi, with a focus on implementation logistics, stakeholder engagement strategies, and data quality procedures.

Results: 51 337 trauma cases were recorded over the first 14 months of registry operations. The number of cases per month, data accuracy, and the geographic coverage of the registry improved over time as data quality measures were implemented.

Conclusions: Multi-center digital trauma registries are feasible in low-resource settings. Stakeholder engagement, periodic in-person and frequent digital follow up with data teams, and regular channeling of findings back to data collection teams help to improve data quality and complete- ness over a 14month period. Financial and staffing constraints remain challenges for sustainability over time, but this experience demonstrates the feasibility of large-scale registry operations.

Croke K, Mengistu AT, O'Connell S, Tafere K. The impact of a health facility construction campaign on health service utilization and outcomes: Analysis of spatially-linked survey and facility location data in Ethiopia. BMJ Global Health. 2020;5 (8) :e002430. Publisher's Version ethiopia_bmj_final.pdf
Croke K, Yusoff MM, Abdullah Z, Hanafiah ANM, Mokhtaruddin K, Emira Soleha Ramli, Borhan NF, Almodovar-Diaz Y, Atun R, Virk AK. The political economy of health financing reform in Malaysia. Health Policy and Planning. 2019;34 (10) :732-739. Publisher's VersionAbstract
There is growing evidence that political economy factors are central to whether or not proposed health financing reforms are adopted, but there is little consensus about which political and institutional factors determine the fate of reform proposals. One set of scholars see the relative strength of interest groups in favour of and opposed to reform as the determining factor. An alternative literature identifies aspects of a country’s political institutions–specifically the number and strength of formal ‘veto gates’ in the political decision-making process—as a key predictor of reform’s prospects. A third group of scholars highlight path dependence and ‘policy feedback’ effects, stressing that the sequence in which health policies are implemented determines the set of feasible reform paths, since successive policy regimes bring into existence patterns of public opinion and interest group mobilization which can lock in the status quo. We examine these theories in the context of Malaysia, a successful health system which has experienced several instances of proposed, but ultimately blocked, health financing reforms. We argue that policy feedback effects on public opinion were the most important factor inhibiting changes to Malaysia’s health financing system. Interest group opposition was a closely related factor; this opposition was particularly powerful because political leaders perceived that it had strong public support. Institutional veto gates, by contrast, played a minimal role in preventing health financing reform in Malaysia. Malaysia’s dramatic early success at achieving near-universal access to public sector healthcare at low cost created public opinion resistant to any change which could threaten the status quo. We conclude by analysing the implications of these dynamics for future attempts at health financing reform in Malaysia.
Chukwuma A, Bossert TJ, Croke K. Health service delivery and political trust in Nigeria. Social Science and Medicine - Population Health. 2019;7. Publisher's VersionAbstract
Do improvements in health service delivery affect trust in political leaders in Africa? Citizens expect their government to provide social services. Intuitively, improvements in service delivery should lead to higher levels of trust in and support for political leaders. However, in contexts where inadequate services are the norm, and where political support is linked to ethnic or religious affiliation, there may be weak linkages between improvements in service delivery and changes in trust in political leaders. To examine this question empirically, we take advantage of a national intervention that improved health service delivery in 500 primary health care facilities in Nigeria, to estimate the impact of residence within 10 km of one or more of the intervention facilities on trust in the president, local councils, the ruling party, and opposition parties. Using difference-in-difference models, we show that proximity to the intervention led to increases in trust in the president and the ruling party. By contrast, we find no evidence of increased trust in the local council or opposition parties. Our study also examines the role of ethnicity and religious affiliation in mediating the observed increases in trust in the president. While there is a large literature suggesting that both the targeting of interventions, and the response of citizens to interventions is often mediated by ethnic, geographic or religious identity, by contrast, we find no evidence that the intervention was targeted at the president's ethnic group, zone, or state of origin. Moreover, there is suggestive evidence that the intervention increased trust in the president more among those who did not share these markers of identity with the president. This highlights the possibility that broad-based efforts to improve health services can increase trust in political leaders even in settings where political attitudes are often thought to be mediated by group identity.
Croke K, Atun R. The Long Run Impact of Early Childhood Deworming on Numeracy and Literacy: Evidence from Uganda. PLOS Neglected Tropical Diseases. 2019;13 (1) :e0007085. Publisher's Version croke_atun_deworming.pdf
Bundy DAP, de Silva N, Horton S, Patton GC, Schultz L, Jamison DT, and Group DCP-3 CAHDA. Investment in child and adolescent health and development: key messages from Disease Control Priorities, 3rd Edition. The Lancet. 2018;391 (10121) :687-699. Publisher's Version
de Neve J-W, Andriantavison RL, Croke K, Krisam J, Rajoela VH, Rakotoarivony RA, Rambeloson V, Schultz LB, Qamruddin J, Verguet S. Health, financial, and education gains of investing in preventive chemotherapy for schistosomiasis, soil-transmitted helminthiases, and lymphatic filariasis in Madagascar: a modeling study. PLOS Neglected Tropical Diseases. 2018;12 (12) :e0007002. Publisher's Version