1

London School of Hygiene and

Tropical Medicine

Mitigating Costs Incurred by Tuberculosis Patients

in Low and Middle Income Countries; the Role of

Cash Transfer Programmes

Candidate 108333 Supervised by: Dr Delia Boccia, Ms Sedona Sweeney

Submitted in part fulfilment of the requirements for the degree of MSc in Control of Infectious

Diseases

Academic Year: 2014-2015

Date of submission: 9th September 2015

Word count: 9984

Project length: Standard

2

Table of contents

Abstract ..................................................................................................................... 4

Acknowledgments. ................................................................................................... 5

1. Acknowledgement of academic support. 5

2. Acknowledgement of other support. 5

Introduction. ............................................................................................................. 6

1. The “catastrophic burden” of costs incurred by TB patients. 6

2. Social protection, and cash transfer programmes. 7

3. Potential impact of cash transfers for tuberculosis control. 8

Study aims and objectives. ................................................................................... 11

Methods. ................................................................................................................. 12

1. Literature review and secondary data analysis. 14

1a) Data sources and eligibility criteria. 14

1b) Data extraction. 15

1c) Data analysis. 16

1d) Data manipulation. 17

2. Situational analysis. 19

3. Ethical approval. 19

Results. ................................................................................................................... 20

1. Literature review and secondary data analysis. 20

1a) Inclusion of articles and cash transfer programme eligibility assessment. 20

1b) Characteristics of cash transfer programmes included in the analysis. 22

1c) Summary of final average patient costs in selected countries. 24

1d) Equivalence of annual cash transfers to average patient costs. 25

1e) Percentage point reduction in the cost burden incurred by TB patients after addition of the value

of annual cash transfers. 28

1f) Summary comparison of cost equivalence analysis, and cost burden reduction analysis. 29

2. Situational Analysis. 33

2a) “CRESIPT”, theory, design and implementation of a TB-specific cash transfer intervention. 33

2b) “Bolsa Familia”, theory, design and proposal for cross-sectoral implementation of a TB-sensitive

cash transfer programme. 34

Discussion .............................................................................................................. 36

Main results. 36

Main strengths of the study. 38

3

Study limitations. 38

Interpretation of results. 39

Policy implications and future research directions. 39

References. ............................................................................................................. 41

Appendix. ................................................................................................................ 49

4

Abstract

Introduction: Extreme tuberculosis (TB) related costs are defined as “catastrophic” when

they represent more than 20% of patient’s annual income. The global post-2015 TB

elimination strategy endorses social protection policies, including cash transfer programmes,

to eliminate TB-related “catastrophic” costs. Currently, there is limited research evaluating

the potential of cash transfer programmes to mitigate patient incurred costs.

Methods: Two definitions of cost mitigation were used;(a) potential of cash transfers to equal

100% of costs; and (b) potential of cash transfers to reduce the burden of costs below a

“catastrophic” threshold. To frame the application of these definitions two complementary

methods were used.

1. Literature review and analysis: of sixteen cash transfer programmes in low- or middle-

income countries for which published TB-related cost data was available. The mitigation

potential of these programmes was computed according to definitions (a) and (b).

2. Situational analysis: The review was complemented with in-depth case studies of

programmes in Peru and Brazil using semi-structured interviews with key informants.

Results:

1. Literature review and analysis: under definition (a) a large proportion of cash transfers is

needed to match costs fully; under definition (b) cash transfers greatly increase overall income,

but large amounts of additional income is required to reduce costs below a “catastrophic”

threshold. Across selected countries there was great inconsistency in mitigation potential for

either definition.

2. Situational analysis: In Peru, cash transfers equal to 20% of total costs are sufficient to

improve access to care. In Brazil, the national TB programme proposes a cross-sectoral

approach for providing TB-related cash transfers demonstrating that development synergies

are politically motivated as well as evidence-based.

Conclusions: This is the first attempt to summarise the literature on TB costs and cash

transfers programmes. Despite important limitations this study highlights critical gaps in the

current policy debate for achieving interim TB elimination milestones.

5

Acknowledgments.

1. Acknowledgement of academic support.

I would like to sincerely thank all those who have helped and supported me to shape this

project.

Project development: My enthusiasm for the project idea arose from my past experience as

an intern in the control of infectious disease department in the WHO-PAHO sub-regional office

in Peru. The idea to focus on cross-sectoral collaborations for implementing TB-related cash

transfers was finalised following discussions with one of my supervisors. The use of a

secondary literature review complemented by key informant interviews was proposed by my

one of my supervisors. The approach that I took to analysing the data was decided in

consultation with both of my supervisors.

Contact, input and support: I had three planning meetings with my supervisors (each about an

hour long), while developing my original proposal and getting it approved, as well as further

email exchanges. Over the course of the project I was based in London, and was in regular

email exchange with my supervisors, who provided practical advice on ways to approach the

project. Over the course of my project I had meetings with my supervisors, and experts in the

field to discuss ideas and initial findings.

Main research work: All references cited were identified through my own literature search.

Data extraction, manipulation and analysis was conducted independently. Key informants

were contacted after a formal introduction by my supervisor Dr. Delia Boccia. Aspects of the

project methods were refined with advice from my supervisor Sedona Sweeney (LSHTM

Research Fellow), who also helped to cross-check my analysis.

Writing-up: Each of my supervisors reviewed a first draft of each section, and a complete final

first draft of the project report.

2. Acknowledgement of other support.

I would like to thank my family and my girlfriend Poppy for being four elephants of

encouragement and helping me to focus my thoughts during all of this work.

6

Introduction.

“Extending social protection to all people, within countries and globally, will be a major step

towards achieving health equity within a generation.” CSDH (1)

1. The “catastrophic” burden of costs incurred by TB patients.

In 2013, nine million people developed tuberculosis (TB) disease causing 1.5 million people

to die mostly in low- and middle income countries (2). Despite exceptional achievements over

the previous 15 years, the decline in TB incidence remains slow (1⋅5% decrease in global TB

notification rates annually), and will be insufficient to achieve TB elimination by 2050 (2,3).

Whilst national TB programmes have led a relentless effort to reduce financial barriers for the

poor by strengthening “political commitment” to public funding, and promoting a shift from

hospital based care to ambulatory primary care, impoverished patients and their households

continue to incur unmanageable costs due to TB-illness and care (4,5). By aggravating

household vulnerability these costs can prevent or delay diagnosis, treatment and successful

outcome, leading to increased TB transmission, morbidity and mortality, Figure 1 (6,7).

The main types of costs faced by TB patients include direct medical costs, hidden direct “out

of pocket” costs such as for transport, symptom relieving medicines and food, and indirect

costs due loss of productivity (8–11). For those affected by TB, the total value of these costs

can often pose a “catastrophic” cost burden, that sometimes forces patients and their

households to reduce consumption of other minimum needs, or engage in dissaving

behaviours, Figure 1 (8,12–15). There is also evidence that a total value of all care related

costs greater than 20% of annual income is associated with twice the risk of adverse treatment

outcome (8). Although poorer populations have been found to incur lower costs during care,

because of lower relative income compared to the less poor, costs represent a greater cost

burden, which often results in worse adherence to treatment (7,8,10,16–19).

In order to address these inequitable access barriers part of the work of the stop-TB-

partnership has been to outline options for national TB programmes to “address the needs of

poor and vulnerable populations” (20,21). These efforts are reiterated with renewed ambition

in the post-2015 WHO TB elimination strategy, which endorses social protection and universal

health coverage as core unifying components to provide universal free access to care, and

prevent any TB patient from incurring “catastrophic” costs by 2025 (22,23). In order to monitor

progress towards this milestone the global TB programme at the WHO recently evaluated a

7

pilot tool to estimate “catastrophic” costs according to a threshold of costs greater than 20%

of annual income (24).

Figure 1: Conceptual framework describing how tuberculosis-related costs result in

increased TB transmission and morbidity, and contribute to a disease-poverty trap.

2. Social protection, and cash transfer programmes.

Social protection is defined as a wide range of policies and programmes designed to “protect

vulnerable populations against livelihood risks; and enhance the social status and rights of the

marginalised, by improving their capacity to manage needs that arise throughout the lifecycle”

(25). According to Figure 2, there are three major social protection interventions:

a) Social transfers1; e.g cash and/or food transfers.

b) Labour market measures; e.g Labour market programmes and vocational training.

c) Social insurance; e.g social health insurance.

Since the mid-1990s the number of cash transfer programmes around the world has grown

rapidly, and today they are the most popular form of social transfer policy in low- and middle-

income countries (26). By providing periodic cash to vulnerable populations, these

programmes aim to build human capital and productive assets to move households out of

extreme poverty (26,27). Whilst some programmes are unconditional, in order to further

encourage investments in human capital, conditional cash transfer programmes require

1 Social transfers are also referred to as social safety nets or social assistance (26)

8

beneficiaries to comply with specific conditioning factors such as school attendance,

immunizations, and health check-ups (26,27).

Figure 2: How social transfers fall within social protection and labour systems.

Adapted from: World Bank 2015 (26)

3. Potential impact of cash transfers for tuberculosis control.

For TB, social protection is proposed to enable equitable access to care, and support poor

patients by mitigating the financial risk they incur whilst seeking and receiving care (24). Most

of the evidence for the potential of cash transfers for disease control has been accumulated

by the HIV/AIDS community (28,29). In Zambia, evaluation of the impact of a poverty related

cash transfer programme on HIV-affected households provides evidence of the potential of

social protection to mitigate the economic impacts of AIDS on destitute HIV-affected

households (30,31). In addition to these HIV-sensitive approaches, in recent years, evidence

for the impact of more specific programmes has established the use of cash transfers

exclusively targeted to key-affected populations, as a way to achieve HIV/AIDS control

objective (28,32–36).

Drawing from this experience, two principle implementation models are considered for TB

control; TB-specific and TB-sensitive cash transfer programmes, Box 1 (28). Differently from

HIV/AIDS, currently there is limited evidence of the impact of cash transfer programmes to

improve access to care, or mitigate costs incurred by TB patients, Box 1 (37–41).

For TB control, one way to conceptualise cost mitigation might be to estimate the extent to

which cash transfers equal the value of costs incurred by TB-patients whilst seeking and

receiving care, this cancelling them out. Currently this approach is being used to evaluate the

mitigation potential of cash transfers provided by the TB-specific “CRESIPT” project in Peru,

Box 1 (43). Using this approach, for maximum potential cash transfers should equal 100% of

TB-related costs. However, this approach might not apply so well for a TB-sensitive

9

programme like “Bolsa Familia”, which is not intended for such a specific purpose. Another

approach might be to evaluate the mitigation effect that cash transfers achieve by adding to

annual income, thus reducing costs as a percentage of income, and consequently protecting

patients from incurring a “catastrophic” cost burden.

Currently no attempts have been made to explore possible approaches for evaluating the

potential mitigation effect of existing cash transfer programmes, and in the absence of any

evidence it is unknown which approach might apply best. The established link between poverty

and TB means that many existing cash transfer programmes might already indirectly support

poor and vulnerable TB-affected beneficiaries to access diagnosis and treatment (5).

However, without guidance, policy implementers may only speculate on the current situation

or potential of such cross-sectoral collaborations for TB control. With the aim of partially filling

this knowledge gap, this study evaluated two conceptual approaches for estimating the

mitigation potential of cash transfer programmes:

a) The equivalence of the value of cash transfers to costs incurred by TB patients.

b) The percentage point reduction in the burden of costs incurred by TB patients

towards a “catastrophic” threshold.

10

Box 1: Potential TB-related cash transfer implementation models.

TB-specific TB-sensitive

Definition

Specific programmes exclusively

targeted to TB-affected individuals

and/or households with the intent of

addressing a specific TB care and

prevention issues.

Social protection programmes not

specific to TB patients but that

indirectly have an impact on TB

control because they target groups

and/or people at high risk of TB.

Examples

ISIAT (before vs. after study) (40)

CRESIPT (Cluster RCT) (43)

Lutge et al. economic support to TB patients (Cluster RCT) (39,41)

Brazil: Bolsa Familia (Conditional cash transfer) (37)

South Africa: Disability Grant (Social Health Insurance) (37)

Potential to improve

TB-treatment outcome

ISIAT: Improved preventive therapy completion (p=0.001) (40)

CRESIPT: Ongoing evaluation (43)

Lutge et al. : No evidence for improved

treatment success (p=0.107) (40)

Brazil: Not evaluated

South Africa: Not evaluated

Cost mitigation potential

ISIAT: Not evaluated

CRESIPT: Ongoing evaluation(43)

Lutge et al.: Not evaluated

Brazil: Not evaluated

South Africa: Not evaluated

11

Study aims and objectives.

The overall aim of this study is to provide preliminary evidence on the capacity of cash transfer

programmes to mitigate TB-related costs.

The three objectives of this study are:

1. Document, and review characteristics of selected cash transfer programmes.

2. Estimate the potential mitigation effect of a selected list of cash transfer programmes

using two distinct approaches; and discuss the relative merits of each approach for

future impact evaluation.

3. Study in-depth the issues relating to the theory design and implementation of TB-

specific and TB-sensitive cash transfer programmes that influence how TB cost

mitigation might be conceptualised for each approach.

12

Methods.

Health policy and systems research methods are valuable for identifying interconnections

between emerging policies and their future implementation (44). To meet the research

objectives, and consistent with acclaimed best practice, this study drew on different

perspectives using two approaches (45).

1. Literature review and secondary data analysis: was used to evaluate the potential mitigation

effect of cash transfer programmes based on the two definitions of mitigation provided in the

introduction. An overview of the steps undertaken is provided in Figure 2:

2. Situational analysis: to complement the quantitative component of this study, two in-depth

case studies, were conducted. The two case studies were:

a) The TB-specific Community Randomized Evaluation of a Socioeconomic Intervention

to Prevent TB “CRESIPT” project in Peru (43).

b) The TB-sensitive conditional cash transfer programme “Bolsa Familia” in Brazil

(37,46).

13

Figure 2: Summary framework of methods used to manipulate and calculate the

potential mitigation effect of selected cash transfer programmes.

*Eligible surveys by Prado et al. Wyss et al. and Gospodarevskaya et al. were excluded from the analysis due to incompatible data for aggregation.

14

1. Literature review and secondary data analysis.

1a) Data sources and eligibility criteria.

Potentially relevant articles reporting TB patient-incurred costs were sourced from two recent

systematic literature reviews of TB-costs, one by Tanimura et al. and the other by Laurence

et al. (5,47). The recent, and systematic nature of these reviews mean that they should provide

an exhaustive list of published average costs incurred by TB patients up to February 20152.

Potentially relevant cash transfer programmes were identified using the World Bank report,

“The state of social safety nets 2015”, that summarises current data on social safety nets in

157 countries (26). In this study, cost articles and surveys were distinguished to illustrate

where an article reported cost values for two or more countries, or where two articles reported

cost values for one survey.

(i) Screening for eligible articles.

Identified articles reporting TB patient-incurred costs were eligible if they were conducted in a

low- or middle-income country, were written in English, were not overlapping, and reported

average direct and/or indirect as well as total costs incurred by TB patients, pre- and/or post-

diagnosis for drug sensitive TB. A priori this study did not attempt to extract data on costs

incurred by multi-drug resistant TB patients due to limited data availability in the literature (47).

Direct patient costs were defined as medical costs for receiving treatment, such as user fees

for health facilities, hospitalisation, diagnostic tests and drug expenditures; and non-medical

costs for transportation, food, non-TB drugs and room and board for patients not resident near

the TB treatment facility (47). Indirect costs were defined as the value of paid and unpaid

production loss due to time seeking treatment, or being ill (47).

(ii) Screening for eligible cash transfer programmes.

Cash transfer programmes were included in the review if they were targeted to “poor and/or

vulnerable populations”. Further information on cash transfer programmes was sourced from

reference sections of “The state of social safety nets 2015”, and from a wide range of ad hoc

electronic databases and comprehensive programme-related documents, published by

international and governmental agencies (26). Cash transfer programmes were excluded if

they were targeted to specific sub-populations (e.g refugees), or populations outside the age

range of 15-54 years of age most affected by TB (e.g orphans and vulnerable children or

2 The search period of the systematic review conducted by Laurence et al. was from January 1990 to February 2015.(47)

15

senior citizens). If one article reported cost values from several different countries, each survey

country was checked separately for an ongoing cash transfer programme. 7

1b) Data extraction.

(i) Tuberculosis patient-incurred costs.

Background information was extracted from selected articles, and comprised: survey country,

name of first author, year of data collection, survey objective and survey setting. Data on

average costs incurred by TB patients were extracted and stratified to the extent the data

allowed into sub totals of direct costs, and indirect costs. When an average value for all survey

subjects in a given survey was not available, an unweighted average was recalculated across

subgroups (5). Whenever possible cost components were extracted separately for the pre-

and post-TB diagnosis period. Pre-TB diagnosis treatment costs were those incurred between

the onset of symptoms and the initiation of treatment, and post-TB diagnosis costs were those

incurred from diagnosis to completion of treatment.

(ii) Characteristics of selected cash transfer programmes.

Background information on selected cash transfer programmes was extracted from

programme related databases and documents, and comprised: name of cash transfer

programme, year programme was established, targeting strategy, programme scale,

household coverage; or when household coverage was not available individual coverage;

value of annual cash transfer, frequency of disbursements, programme conditionalities, and

programme budget. Due to limited data availability maximum annual cash transfers was used

for consistency across countries. Household coverage was also extracted as a percentage of

population in households in countries for which census data was available in the United

Nations demographic yearbook (48). If individual coverage of a programme was only available

this was compared to the country’s population in individuals in World Bank online data (49).

Programme budget was calculated as a percentage gross domestic product (GDP) in

countries using current GDP in World Bank online data (50). In order to evaluate the scale and

sustainability of selected cash transfer programmes for TB control objectives, programme

characteristics were presented according to TB incidence in WHO’s global TB database (52).

16

1c) Data analysis.

Consistent with the two different definitions of mitigation provided in the introduction of this

document, the computation of the potential TB costs mitigation required two different analytical

approaches, specifically:

a) The equivalence of the value of cash transfers to costs incurred by TB patients.

b) The percentage point reduction in the burden of costs incurred by TB patients

towards a “catastrophic” threshold.

(ii) Equivalence of cash transfers to patient –incurred costs.

To evaluate the potential of cash transfers to mitigate costs fully, the value of annual cash

transfers provided by countries programmes were simply evaluated for their equivalence,

expressed as a percentage, to country level estimates of average patient TB costs, Equation

1 (Appendix). In countries where cash transfers were not sufficient to mitigate costs fully, the

additional cash necessary to equal costs was also calculated, Equation 2 (Appendix).

(iii) Percentage point reduction of cost burden incurred by patients.

To evaluate the potential of cash transfers to reduce the burden of TB costs to below a

“catastrophic” threshold less than 20% of annual income, a second more sophisticated

analysis was required. For this analysis total cost of care was reported as a percentage of

annual income, before and after addition of the value of annual cash transfers to annual

income, and the mitigation potential was represented by the percentage point (%pt) reduction

between these two values, Equation 3 (Appendix). For this analysis the estimated cost burden

was “catastrophic” if it was above 20% of annual income. In countries where cash transfers

did not mitigate the cost burden to below the “catastrophic” threshold the additional cash

necessary to achieve this objective was calculated, Equation 4 (Appendix).

17

1d) Data manipulation.

(i) Conversion of monetary variables to 2013 international dollars 3(2013 constant PPP$)

For manipulation and comparison of monetary variables extracted in different currencies and

measured in different years, average costs, cash transfers, programme budget and GDP were

all converted to 2013 international dollars (2013 PPP$). This was done by multiplying raw data

in US$, the official exchange rate with the local currency for the year of data collection, and

the local cumulative inflation rate from the year of data collection to 2013. All data was then

divided by the purchasing power parities (PPP$) conversion factor for cross-country

consistency. For patient-incurred costs, the exchange rates reported in reviewed articles were

preferentially used for the calculation and, in the absence of them, exchange rates in the final

year of data collection were taken from the World Bank’s online data (52).

(ii) Computation of average costs in selected countries.

Methods used to calculate country level average costs depended on the number of surveys

conducted in each country, and were estimated either as, a) mean or median costs as reported

by the single survey in the review country; or b) aggregated average mean or median costs

reported by multiple surveys of costs in the review country, Table 2. Wherever possible this

allowed approximation of the full value of total costs incurred by TB patients (direct and

indirect), both pre- and post-diagnosis, and was necessary as often surveys did not collect

data from both treatment phases, or report all component cost values.

Country level costs in China were stratified by geographical location according to the distinct

urban and rural components of the national cash transfer programme. Due to limited data

availability country level costs incurred by TB patients in Nigeria and Zambia refer to

aggregated median costs. Limited data availability also prevented any calculation of measures

of spread for estimated average costs in selected countries. Using these methods country

level average costs were estimated for 154 countries using 28 surveys.

3 An international dollar (PPP$) would buy in the cited country a comparable amount of goods and services as a U.S. dollar would buy in the United States. 4 Malaysia, Mexico, Brazil, China (Urban and Rural), Dominican Republic, Ecuador, Indonesia, Nigeria, Pakistan, Ghana, Yemen, Zambia, Tanzania, Ethiopia and Malawi.

18

Table 2: Methods used to calculate country level average costs in countries with

multiple surveys.

Country Survey

Survey treatment

phase coverage

Breakdown of costs pre- and

post-diagnosis

Data Manipulation

Brazil

Steffen Both The unweighted average of mean costs from surveys

that collected data during both treatment phases Costa Both

Prado* During

Rur. China

Xu Both

The unweighted average of mean costs from surveys

that collected data during both treatment phases

Jackson Both

Meng Both

Zhan Both

Pan Both

Nigeria Umar Both The unweighted sum of median cost components

published in different articles (direct + indirect) Umar Both

Zambia Aspler Both The sum of unweighted average median costs estimated

separately pre- and post-diagnosis Needham Before

Tanzania

Gospodarevskaya* During The unweighted average of mean costs from surveys

that collected data during both treatment phases Wandwalo* During

Wyss Both

Ethiopia

Datiko During The sum of unweighted average mean costs estimated

separately pre- and post-diagnosis Vassall Both

Mesfin Before

Malawi Kemp Before The sum of mean costs surveyed seperately pre- and

post-diagnosis Floyd During

*Eligible surveys by Prado et al., Wyss et al., and Gospodarevskaya et al., were excluded from the analysis due to incompatible data for aggregation.

(iii) Estimation of annual income in selected countries.

Although TB cost burden analyses often relate to household income (14,54–56), very few

articles included in this study reported household level income data, or household level TB

cost data (5,47,57). Although imperfect, per capita-GDP is a meaningful proxy for average

individual income, and has been used to estimate the cost burden of TB in previous studies

(5,57). For this study, per capita GDP in the poorest population quintile was further justified

because a) TB disproportionately affects poor and vulnerable populations; and b) poverty-

related cash transfer programmes are specifically targeted to the poorest population sub-

groups (26). Based on the methods used in Barter et al. and Tanimura et al. annual income

was calculated using “GDP per capita (current LCU)”, and ‘‘income share held by lowest 20%’’

in the World Bank’s online data in the year closest to data collection, Equation 5 (Appendix)

(5,57,58).

For country level analysis, annual income was estimated either as a) annual income as

calculated for the single survey of costs incurred by TB patients in the review country; or b)

unweighted average annual income across surveys used to calculate aggregated country level

costs, Table 2.

19

2. Situational analysis.

To better frame the concept of cost mitigation under different implementation scenarios, in-

depth case studies of two existing TB-related cash transfer initiatives were undertaken, and

were complemented with key stakeholder interviews:

a) The TB-specific Community Randomized Evaluation of a Socioeconomic

Intervention to Prevent TB “CRESIPT” project in Peru (43).

b) The TB-sensitive conditional cash transfer programme “Bolsa Familia” in Brazil

(37,46).

These projects were chosen as they represent two of the most advanced initiatives for

exploring the potential of cash transfers for TB control (37). The appraisal of these projects

included evaluating programme characteristics, the proposed function of cash transfers, their

existing mitigation potential and the challenges identified by each initiative for implementing

these programmes. In order to conduct this analysis data was collected from key informants

using semi-structured interviews. For Peru, key informant interviews were conducted with

principal and co-investigators of the “CRESIPT” project; For Brazil, key informant interviews

were conducted with the director of the Brazilian national TB programme, and national

scientists working on the social impact of “Bolsa Familia” in Brazil. For both Peru and Brazil,

key informants were recruited by email invitation, and wherever possible interviews were

conducted in person. Whenever possible interviews were also complemented with relevant

published peer-reviewed articles, and “grey literature”.

3. Ethical approval.

The internationally accredited ethical committee of the London School of Hygiene and Tropical

Medicine has approved all components of the study. All interviewed participants were asked

to give written informed consent to participate in the study.

20

Results.

1. Literature review and secondary data analysis.

1a) Inclusion of articles and cash transfer programme eligibility assessment.

Figure 4 is a flow chart of the review process. Of all articles one article had restricted access,

and could not be included in the review process (59). No information was available on the

level or regularity of transfers provided by the Zakat cash transfer programme in Sudan, so

cost data from this country also had to be excluded from the analysis. The final number of

articles included in the review was 28, representing 28 surveys in 15 countries. Details about

included surveys are provided in Table 3. A summary of the cash transfer programme

screening process is provided in the Appendix.

Figure 4: Country eligibility and inclusion flow chart.

*Yitaya et al. 2014 was excluded because of restricted access.

21

Table 3: Synopsis table: patient incurred cost surveys included in the review.

Country Survey Costs Evaluated

Survey treatment

phase coverage

Mean/ Median /Both

Direct med. costs

surveyed

Direct non-med. costs

surveyed

Indirect costs

surveyed

Breakdown of costs pre- and

post-diagnosis

Malaysia Elamin [2002] (60) Treatment in Penang

state Both Mean

Mexico Guzman-Montes

[2007/08] (61) Household costs Both Mean

Brazil Steffen [2010]

(62) HFDOT vs. SAT Both Mean

Brazil Costa [2000] (63) Treatment in

Salvador State Both Mean

Brazil Prado [2005/2006]

(64) CDOT vs. guardian

DOT During Mean

Urb. China Zou [2009/10]

(65) Costs to migrant

patients During Mean

Rur. China Xu [2002] (66) Household costs Both Both

Rur. China Jackson [2002/05]

(67) Household costs Both Mean

Rur. China Meng [2000] (68) Decentralized TB

control Both Mean

Rur. China Zhan [2000/01] (7) DOTS model vs. MoH model vs.

conventional model Both Mean

Rur. China Pan [2011] (69) Household costs Both Both

Ecuador Rouzier [2007]

(70) Household costs Both Mean

Dom. Republic

Mauch [2009] (15)

Household costs Both Both

Indonesia Mahendradhata [2004/05] (71)

Public-private partnerships

Both Mean

Zambia Aspler [2006] (72) Household costs Both Median

Zambia Needham [1995] (17) Household costs Before Both

Ghana Mauch [2009]

(15) Household costs Both Both

Pakistan Khan [1997/98]

(73) HFDOT vs. CDOT vs.

family DOT Both Mean

Nigeria Umar [2008] (74) Household costs Both Median

Nigeria Umar [2008] (14) Household costs Both Median

Yemen Othman [2008/09]

(75) Treatment in Sana'a

city Both Mean

Tanzania Gospodarevskaya

[2012] (76) Household costs During Both

Tanzania Wandwalo [2002] (77) Household costs During Mean

Tanzania Wyss [1996] (78) HFDOT vs. CDOT Both Mean

Ethiopia Datiko[2007] (79) HFDOT vs. CDOT During Mean

Ethiopia Vassall [2005] (80) TB/HIV Treatment Both Mean

Ethiopia Mesfin [2005]

(81) Household costs Before Both

Malawi Kemp [2007] (18) Household costs Before Both

Malawi Floyd [1997/98]

(82) HFDOT vs. CDOT During Mean

Total number of surveys in selected countries

28* Reported Costs 23 27 24 9

The years in which the majority of data collection took place are provided for each survey. *Two articles reported data from one survey. DOT Directly observed treatment, HFDOT Health facility DOT, SAT Self-administered treatment, CDOT Community DOT, MoH Ministry of Health, Dom. Republic Dominican Republic, Urb. China urban China, Rur. China rural China.

22

1b) Characteristics of cash transfer programmes included in the analysis.

Details about selected cash transfer programmes, ordered by country TB incidence, are

provided in Table 4. All eligible cash transfer programmes were targeted to households, only

Malaysia’s “Bantuan Raykat 1 Malaysia” programme was targeted to both households and

individuals. In China, Ethiopia, Malaysia and Pakistan programmes were all unconditional.

Programmes in Brazil, Dominican Republic, Ecuador, Ghana, Indonesia, Malawi, Mexico,

Nigeria and Tanzania were all conditional, and required beneficiary households to comply with

certain health, education, and in the case of Mexico, food conditions for beneficiaries to

continue to receive transfers. Targeting strategies used by selected programmes to select

beneficiaries included: means-tested, proxy-means tested, community-based, geographical

targeting, and dependency score approaches.

Cash transfer programmes in Brazil, China, Dominican Republic, Ecuador, Indonesia,

Malaysia, Mexico, Pakistan and Yemen are all operated at national level. Programmes in

Brazil, Mexico and Pakistan provided cash transfers to more than 4 million households, or

approximately 25% of national households respectively. Although coverage by number of

beneficiary households was lower, programmes in Dominican Republic and Ecuador still

covered a large proportion of the total population (25.6% and 15.4% of households

respectively). In China urban and rural components of the “Dibao” programme each provided

cash transfers to more than 20 million individuals, however, as a proportion of the total

population coverage was low (3.91% and 1.58% respectively). In Yemen the Social welfare

fund provided cash transfers to 1,507,093 households, which according to a monitoring survey

covers 35% of the population in households (83). In Indonesia as a proportion of all

households programme coverage was low (3.69%). In Nigeria, Ghana, Zambia, Tanzania,

Malawi and Ethiopia programmes are either being scaled up, or are pilot programmes; and in

most cases provided cash transfers to a small proportion of total households (range: 0.02% to

2.12%). Across countries programme budgets reflected the number of beneficiary households

receiving cash transfers.

Across selected countries there was great variation in annual cash transfers provided by

programmes (range: 361$ to 3067$), Figure 5. Programmes in Brazil, Dominican Republic,

Ecuador, Malaysia, Mexico, Nigeria and Yemen, provided cash transfers >1000$. Apart from

Nigeria, across all selected programmes, ones in these countries were the best established in

terms of scale and household coverage. Programmes in urban China, Ethiopia, Ghana,

Indonesia, Pakistan and Zambia all provide cash transfers between 500$ and 1000$, and

programmes in rural China, Malawi and Tanzania provide cash transfers <500$.

23

Table 4: Main characteristics of cash transfer programmes included in the review.

Country Cash Transfer Programme Year Estd.

National/ Scale-

Up/ Pilot

Cash transfer make-up

Cash transfer Disbursement

House. coverage (% House. Pop)

Budget 2013 PPP$

(% 2013 GDP)

TB Incidence ≤ 50 per 100,000

Mexico Oportunidades [CCT] (84) 2002 National Variable benefit: Educational support, food support, child

support. Bi-monthly 6,600,000 (23.44) 8,225,435,612 (0.41)

Brazil Bolsa Familia Programme [CCT] (84) 2003 National Basic benefit: only for extremely poor

Variable benefit: school attendance subsidy, pregnant women support

Monthly 14,086,199 (24.26) 13,726,708,075 (0.43)

Yemen Social Welfare Fund [CCT] (83) 1996 National Basic benefit

Variable benefit: household size Quarterly 1,507,093 197,700,000 (0.02)

50 per 100,000 < TB Incidence ≤ 100 per 100,000

Ecuador Bono de Desarollo Humano [CCT] (84) 2003 National Flat transfer Monthly 444,562 (15.44) 1,930,909,091 (1.12)

Dom. Republic

Progresando con Solidaridad [CCT] (84) 2004 National Variable benefit: School attendance subsidy, food support,

energy support Monthly 683,429 (25.57) 599,935,711 (0.47)

Ghana Livelihood Empowerment Against Poverty

[CCT] (85) 2008 Scale-up Variable benefit: Household size Bi-monthly 70,000 (1.28) 32,608,696 (0.03)

China Urban Dibao [UCT] (86) 1997 National Minimum income guarantee Monthly 21400000 (1.58)ᵅ 20,035,178,873 (0.12)

Rural Dibao [UCT] (86) 2007 National Minimum income guarantee Monthly 53100000 (3.91)ᵅ 19,738,361,409 (0.12)

Indonesia Programme Keluarga Harapan [CCT] (87) 2007 National Basic benefit

Variable benefit: child support, pregnant women support Quarterly 2,400,000 (3.69) 339,256,297 (0.01)

Nigeria In Care of the Poor [CCT] (88,89) 2007 Scale-Up Basic income guarantee: child support

Poverty reduction accelerator investment* Monthly/ Annually*

22,000 (0.08) 32,389,145 (3x10¯³)

Malaysia Bantuan Raykat 1 Malaysia [UCT] (90) 2012 National Flat transfer Quarterly 4,620,000 (72.72) 3,239,436,620 (0.47)

TB Incidence > 100 per 100,000

Malawi Social Cash Transfer [CCT] (91) 2006 Scale-up Variable benefit: household size,

school attendance subsidy Bi-monthly 28,000 (0.96) 75,928,345 (0.59)

Tanzania Social Action Fund [CCT] (92) 2000 Scale-up Basic benefit,

Variable benefit: child support, pregnant women support Bi-monthly 147,362

Ethiopia Social Cash Transfer [UCT] (93,94) 2011 Pilot Basic benefit,

Variable benefit: child support, disabled support Monthly 3,767 (0.02) 47,926,425 (0.37)

Pakistan Benazir Income Support Programme [UCT] (95)

2008 National Flat transfer Quarterly 4,800,000 (24.98) 3,379,942,900 (0.40)

Zambia Social Cash Transfer Scheme [UCT]

(96,97) 2003 Scale-up

Basic benefit Variable benefit: disabled support

Bi-monthly 61,000 (2.12) 17,322,835 (0.03)

Total number of programmes in all eligible countries 16

The conditionality of cash transfers are provided for each programme, UCT unconditional cash transfer, CCT conditional cash transfer House. Household, Estd. Established *Annual benefits are given once at the end of the year, when beneficiaries graduate from the programme as a poverty reduction accelerator initiative (PRAI) ᵅData refer to total number of individuals

24

Dom. Republic Dominican Republic, Urb. China urban China, Rur. China rural China Figure 5: Summary of annual cash transfers in selected countries. Data labels refer to the

value of cash transfers in respective countries. Countries are in order of highest to lowest average total cost.

1c) Summary of final average patient costs in selected countries.

After manipulation, final country level average costs represented total costs incurred for both

phases of treatment in all countries, except in urban China where costs were only available

after diagnosis. Final country level average total costs represented both direct and indirect

cost components in all countries, except for Mexico where they represented direct costs. For

cost components, TB patient-incurred direct costs were only estimated from direct non-

medical costs in three surveys (60,79,82). In these articles medical expenditures were

considered as provider costs. Three surveys did not collect data on TB patient-incurred indirect

costs (7,66,68).

Summaries of the value of TB patient-incurred costs in selected countries are presented in,

Figure 6 and Table 5.There was great variation in average total costs (range: 230$ to 3097$),

as well as in the relative contribution of direct and indirect cost components to total costs

across selected countries. In Brazil, Dominican Republic, Ecuador, Ghana, Nigeria and

Tanzania indirect costs represented the larger proportion of total costs (range: 58% to 88% of

total costs). Indirect costs contributed most to average total costs in the Dominican Republic,

and Malawi (88% and 85% of average total costs respectively). Direct costs represented the

25

larger proportion of total costs in urban and rural China, Ethiopia, Indonesia, Malawi, Malaysia,

Pakistan, Yemen, and Zambia (range: 53% to 84% of total costs). Direct costs represented

the largest proportion of total costs in Ethiopia and Malaysia (80% and 84% of total costs

respectively).

*aggregated costs, ᵅmedian costs, Dom. Republic Dominican Republic, Urb. China urban China, Rur. China rural China

Figure 6: Summary of final average tuberculosis patient costs in selected countries.

Data labels refer to average total cost incurred by TB patients in respective countries.

1d) Equivalence of annual cash transfers to average patient costs.

The cost equivalence of annual cash transfers across selected countries are reported in Figure

7, and Table 5. Based on their potential to equal the value of average total costs cash transfer

programmes were categorised as:

a) High potential impact: cash transfers equal to or greater than average total costs.

b) Medium potential impact: cash transfers equal to between 50% and 99% of average

total costs.

c) Low potential impact: cash transfers equal to 0% and 49% of average total costs.

For cash transfer programmes that did not provide cash transfers equal to or greater than

average total costs in respective countries, the outstanding cost after cash transfers is

reported in Table 5.

a) High Potential impact (cost equivalence %)

Eight of the selected cash transfer programmes had a high potential impact. Amongst these

programmes the equivalence of cash transfers to average total costs was greatest in Brazil,

26

Indonesia, Pakistan and Zambia (range: 186% to 513%) where costs were amongst the lowest

across all selected countries (range: 230$ to 538$). The equivalence of cash transfers to

average total costs was also very high in Malawi (171% of average total costs), where costs

were also amongst the lowest across all selected countries (267$). In contrast, in Nigeria and

Mexico, where costs were amongst the highest across all selected countries (1152$ and

1777$ respectively), cash transfers represented more than 150% of average total costs. In

Yemen, annual cash transfers were equal to 116% of costs.

b) Medium potential impact (cost equivalence %)

Three of the selected cash transfer programmes had a medium potential impact. In Dominican

Republic and Ethiopia, annual cash transfers were slightly less than average total costs (91%

and 93% of average total costs respectively), and after subtracting the value of cash transfers

provided in these countries, costs were equal to 247$ and 40$ respectively. In Malaysia,

annual cash transfers were equal to 76% of average total costs.

c) Low potential impact (cost equivalence %)

Five of the selected cash transfer programmes had a low potential impact. In Ecuador and

Ghana, cash transfers represented approximately half the value of average total costs (49%

and 47% of average total costs respectively), and after subtracting the value of cash transfers

costs were equal to 619$ and 1225$ respectively. In Tanzania cash transfers were equivalent

to 38% of costs. In urban and rural China, cash transfer programmes had the lowest mitigation

effect, and after subtracting the value of cash transfer provided by these programmes, costs

were equal to 2268$ and 749$ respectively.

27

Figure 7: Cash transfers as a proportion of average total costs incurred by tuberculosis patients.

*aggregated costs, ᵅmedian costs, Dom. Republic Dominican Republic, Urb. China urban China, Rur. China rural China

Countries are in order of highest to lowest average total cost.

28

1e) Percentage point reduction in the cost burden incurred by TB patients after addition of the

value of annual cash transfers.

Average total costs as a percentage of annual income before and after addition of value of

annual cash transfers are provided in Figure 6, and Table 5. Based on their impact on costs

as a percentage of annual income, selected cash transfer programmes were categorised as:

a) High potential impact: reduction in the cost burden by equal to or greater than 50

percentage points.

b) Medium potential impact: reduction in the cost burden by between 20 and 49

percentage points.

c) Low potential impact: reduction in the cost burden by between 0 and 19 percentage

points.

For programmes that did not provide sufficient cash to reduce the cost burden to below the

target “catastrophic” threshold of 20% of annual income, the values of additional income

necessary to meet this target is provided in Table 5.

a) High potential impact (percentage point reduction)

Seven of the selected cash transfer programmes reduced the burden of costs by more than

50 percentage points. Amongst these countries, cash transfers in Nigeria and Malawi were

closest to reducing the burden of costs to below a “catastrophic” threshold (49% and 46% of

annual income after transfers respectively). In Ecuador, Ghana and Tanzania, despite a high

percentage point reduction, costs still represented a large burden (134%, 98% and 92%, of

annual income after cash transfers respectively). Of these countries, the additional cash

necessary to reduce the burden of costs to below the “catastrophic” threshold (range: 752$ to

951$)

b) Medium potential impact (percentage point reduction)

Four of the selected cash transfer programmes reduced the burden of costs by between 20

and 50 percentage points. Of these countries cash transfers provided in Mexico and Zambia

were closest to reducing the burden of costs to below a “catastrophic threshold” (26% and

22% of annual income after cash transfers respectively). The additional cash necessary to

reduce the burden of costs to below a “catastrophic” threshold was 125$ and 1999$

respectively. In urban and rural China costs still represented a large burden even after

29

hypothetical receipt of cash transfers (105% and 71% of annual income after cash transfers

respectively).

c) Low potential impact (percentage point reduction)

Four of the selected cash transfer programmes reduced the burden of costs by less than 20

percentage points. In Brazil and Pakistan, cash transfers reduced the burden of costs to below

the “catastrophic” threshold (11% and 14% of annual income after cash transfers respectively).

In Malaysia and Yemen, despite a low percentage point reduction in the burden of costs, the

“catastrophic” burden incurred in these countries was almost mitigated (38% and 30% of

annual income after cash transfers respectively). The additional cash required to alleviate the

cost burden to below a “catastrophic” threshold in these countries was 1444$ and 6001$

respectively.

1f) Summary comparison of cost equivalence analysis, and cost burden reduction analysis.

For each selected cash transfer programme, two distinct approaches for evaluating the

potential of cash transfer programmes showed that there was great inconsistency depending

on which approach was used, Table 5. This was mostly due to great variability in the

relationship between average total costs, cash transfers and annual income in each country.

Two examples illustrate this:

a) Low cost equivalence %, but high cost burden percentage point reduction.

In countries like Ghana and Tanzania, cash transfers were only equal to a small proportion of

costs (1208$ vs. 590$, 49% equivalence; and 318$ vs. 837$, 38% equivalence respectively);

however, because annual income in these countries was very low relative to costs (640$ and

307$ respectively), even the small value of cash transfers greatly increased overall income,

resulting in a high percentage point reduction. Nevertheless, despite such a high impact on

costs as a percentage of annual income, even after cash transfers, costs still represented a

very large and “catastrophic” proportion of overall annual income (98% and 134% of annual

income after cash transfers respectively, requiring even more investment in order to mitigate

costs to below a “catastrophic” threshold (4811$ and 3561$ respectively)

b) High cost equivalence %; but low cost burden % point reduction

In countries such as, Pakistan and Yemen, cash transfers were equal to a large proportion of

costs (230$ vs. 626$, 272% equivalence; and 885$ vs.1026$, 116% equivalence

respectively); however, because annual income in these countries was very high relative to

costs (1056$ and 2302$ respectively), the value of cash transfers required to achieve a given

percentage point reduction needed to be higher. Nevertheless, even though the impact of cash

30

transfers on costs as a percentage of annual income was lower in these countries, after cash

transfers the “catastrophic” cost burden was either close to the 20% threshold or mitigated

(14% and 30% of annual income after cash transfers respectively). Nevertheless, in Yemen a

large amount of cash was still necessary to reduce costs to below a “catastrophic” threshold

(1433$).

31

*aggregated costs, ᵅmedian costs, Dom. Republic Dominican Republic, Urb. China urban China, Rur. China rural China

Figure 8: Costs as a percentage of annual income before and after addition of cash transfers. The cost burden reduction is the %pt difference

between the two proportions. In both cases proportions are compared to a “catastrophic” threshold equivalent to 20% of annual income. Countries are in order of highest to

lowest cost burden after addition of cash transfers to annual income.

32

Table 5: Summary comparison of two approaches to evaluate cost mitigation in selected countries.

Equivalence analysis Percentage point reduction analysis

Country Total Costs 2013 PPP$

(Rank*)

Cash transfers 2013 PPP$

(Rank*)

Cost equivalence %

(Impact†)

Value of additional cash

necessary to equal costs fully

2013 PPP$

Annual income (Rank*)

Cost burden % before cash

transfers (Rank*)

Cost burden % after cash

transfers (Rank*)

Cost burden %pt reduction

(Impact‡)

Value of additional cash necessary to

reduce cost burden to of 20% annual income

2013 PPP$

TB Incidence ≤ 50 per 100,000

Mexico 1777 (5) 3067 (1) 173 (High) 0 3820 (2) 47 (12) 26 (12) 21 (Medium) 1999

Brazil 538 (12) 2757 (2) 513 (High) 0 2096 (6) 26 (14) 11 (15) 15 (Low) 0

Yemen 885 (9) 1026 (7) 116 (High) 0 1965 (7) 45 (13) 30 (11) 15 (Low) 1433

50 per 100,000 < TB Incidence ≤ 100 per 100,000

Ecuador 2316 (4) 1091 (6) 47 (Low) 1225 1429 (8) 162 (4) 92 (4) 70 (High) 9061

Dom. Republic

2835 (2) 2587 (3) 91 (Medium) 247 2444 (3) 116 (8) 56 (7) 60 (High) 9142

Ghana 1208 (6) 590 (12) 49 (Low) 619 640 (12) 189 (3) 98 (3) 90 (High) 4811

Urb. China

3097 (1) 829 (8) 27 (Low) 2268 2110 (5) 147 (5) 105 (2) 41 (Medium) 12546

Rur. China 1110 (8) 361 (15) 32 (Low) 749 1192 (9) 93 (9) 71 (5) 22 (Medium) 3996

Indonesia 253 (14) 786 (9) 311 (High) 0 2302 (4) 11 (16) 8 (16) 3 (Low) 0

Nigeria 1152 (7) 1876 (5) 163 (High) 0 904 (11) 127 (7) 41 (8) 86 (High) 2978

Malaysia 2508 (3) 1907 (4) 76 (Medium) 601 4632 (1) 54 (10) 38 (10) 16 (Low) 6001

TB Incidence > 100 per 100,000

Malawi 267 (13) 456 (14) 171 (High) 0 128 (16) 209 (2) 46 (9) 163 (High) 752

Tanzania 837 (10) 318 (16) 38 (Low) 519 307 (15) 273 (1) 134 (1) 139 (High) 3561

Ethiopia 571 (11) 532 (13) 93 (Medium) 40 418 (14) 137 (6) 60 (6) 77 (High) 1908

Pakistan 230 (16) 626 (10) 272 (High) 0 1056 (10) 22 (15) 14 (14) 8 (Low) 0

Zambia 236 (15) 614 (11) 260 (High) 0 441 (13) 53 (11) 22 (13) 31 (Medium) 125

Grey highlighting indicates where cost mitigation impact was the same according to both approaches. *Indicates the relative position of a country, for the particular variable, on an ordinal scale of size from highest to lowest (1 to 16) †High impact >100%, Medium 50% < impact < 100%, Low impact <50% ‡ High impact >50%pts, Medium 20%pts < impact < 50%pts, Low impact <20%pts

33

2. Situational Analysis.

The aim of this situational analysis was to further contextualise the concept of cost mitigation

according to two ongoing TB-related cash transfer initiatives, one TB-specific and one TB-

sensitive:

2a) “CRESIPT”, theory, design and implementation of a TB-specific cash transfer intervention.

“CRESIPT” is a TB-specific socioeconomic intervention that includes cash transfers

exclusively targeted to TB-affected households, rather than households below the poverty line.

The prject is implemented in 32 impoverished shantytown communities of Callao, Peru. This

intervention provides conditional cash transfers to incentivise and enable screening for TB /

MDR-TB, adherence to treatment or chemoprophylaxis, as well as engaging in social

activities. Conditional cash transfers provided for adherence, to treatment are stratified into

double incentives: "optimal" compliance with conditions (e.g. monthly adherence missing less

than two daily tablets), and simple incentives: "acceptable" compliance with condition (e.g.

monthly adherence in which two or more tablets are missed, but the patient has not

abandoned treatment). A “CRESIPT” pilot study was recently completed, and informants were

able to disclose preliminary results from an impact evaluation.

In terms of the “CRESIPT” pilot protocol, cost mitigation was defined as reimbursing the total

value of direct “out of pocket” expenses that patients incur during their entire TB illness (e.g.

from the start of symptoms to completion of treatment). However, there is debate among the

TB community as to whether cash transfers should also aim to mitigate both direct out-of

pocket costs and the indirect costs associated with lost income. According to “CRESIPT”

investigators it is currently being decided whether the total value of cash transfers will be

increased to equal the total value of all costs (direct costs plus lost income), incurred by

patients whilst completing their treatment. It is hoped that such an increase may enhance the

impact of the intervention in the main “CRESIPT” study.

In the pilot phase the average cash transfer amount received by each TB-affected household

over the course of the intervention was 340$, which mitigated 20% of total TB-related

household costs. Preliminary results show that compared to control households, confirmed TB

cure in TB patients and chemoprophylaxis completion in household contacts under 20 years

old was significantly higher in households that received the intervention. However, compared

to control households there was no evidence that intervention households incurred fewer

“catastrophic” costs. Despite initial success, the pilot intervention identified a number of

challenges for providing TB-related cash transfers. Hidden bank charges and delays in cash

34

transfers eroded participants’ confidence in the intervention, and “high risk” patients in more

urban communities were difficult to engage (especially the formerly-incarcerated, drug- or

alcohol-dependent, and gang members).

2b) “Bolsa Familia”, theory, design and proposal for cross-sectoral implementation of a TB-

sensitive cash transfer programme.

As part of the unified social protection system in Brazil, the “Bolsa Familia” programme

provides conditional cash transfers to families living in poverty5, and extreme poverty6, to

promote immediate poverty relief and family development. For this programme, explicit

conditioning factors are designed to guarantee access of beneficiaries to basic social services,

including health, education and social assistance. For monitoring of non-compliance, “Bolsa

Familia” uses a “soft” form of conditionality, by which sanctions are enforced gradually,

beginning with a warning followed by temporary suspension and only after repeat episodes of

non-compliance definitive removal from the programme.

The targeting criteria, scale and institutionalisation of “Bolsa Familia” as a lead programme

within the national ministry of social development in Brazil means that it is currently considered

as a gold-standard for conceptualising TB-sensitive cash transfer programmes. Preliminary

studies to explore the potential of this programme, currently estimate that approximately 15%

of TB patients in Brazil are beneficiaries of this programme. Despite this, it remains unknown

what potential impact this programme might have on national TB control objectives, including

mitigating costs experienced by poor TB patients in this context (98,99).

In accordance with the post 2015 WHO TB elimination strategy, the Brazilian national TB

programme is proposing that “Bolsa Familia” extend an additional TB-related conditional cash

transfer equal to 25$ monthly, over and above current transfers, to beneficiaries that have a

formal diagnosis of TB infection. The aim of the conditioning factor associated with this cash

transfer would be to improve treatment completion rates by incentivising care. According to

the proposal cash transfers would be delivered to patient’s CAIXA account7, “virtually”

throughout the treatment period, following confirmation of adherence by community health

agents of the national “Family health programme”.

5 Per capita income between 44 PPP$ and 87 PPP$ 6 Per capita income of no more than 44 PPP$ 7 CAIXA is a public financial institution headquartered in Brasilia and linked with the Finance Ministry that is responsible for the implementation of federal government initiatives (100).

35

According to key informants, at this stage of the proposal the major challenge for any formal

collaboration between the Brazilian National TB programme and the “Bolsa Familia”

programme is political commitment. As a result of deterioration of the financial economy in

2015, all discussions regarding proposals to incorporate TB control objectives into “Bolsa

Familia” have been postponed until 2016.

36

Discussion

Social protection is a key component of the new post-2015 TB elimination strategy (38,101).

Cash transfer programmes, in particular, are considered powerful tools to enhance TB control

by improving TB prevention, supporting TB treatment adherence and mitigating catastrophic

costs (24,102). Nonetheless, very little evidence is currently available to support this

assumption, and inform the new post-2015 TB elimination strategy (37). By providing some

preliminary evidence on the potential of cash transfers to mitigate TB-related costs, this is the

first study to partially fill this knowledge gap. The key conclusions of this work are summarised

in Box 2, and throughout this chapter.

Main results.

Review of selected cash transfer programmes in this study highlighted that there is great

variability in the design, and implementation of programmes across distinct contexts.

Importantly, this study showed that even in some high TB incidence settings, cash transfer

programmes are well established and have great potential to incorporate current TB-control

objectives.

According to the cost equivalence analysis, cash transfers provided by programmes in ten

countries were either approximately equal, or much greater than estimated total costs incurred

by TB patients in corresponding countries (range: 91% to 513%). Nevertheless, in urban and

rural China, Ecuador, and Ghana the outstanding value of costs incurred by TB patients after

accounting for cash transfers was still very high (range: 619$ to 2268$). However, whilst cash

transfer programmes in these settings might not have the potential to mitigate the full value of

costs, other forms of social protection might still be available to patients. For example in Ghana

where cash transfers provided by the “Livelihood Empowerment against Poverty” programme

were only equal to 49% of average total costs, the Government is planning to have programme

beneficiaries covered automatically by the “National Health Insurance Scheme” (103).

According to the percentage point reduction analysis, cash transfer programmes have the

greatest mitigation potential in settings where costs are lowest, and cash transfers add most

to overall income. For example, in Malawi where costs were 267$ but still represented 209%

of annual income, cash transfers equal to 456$ greatly reduced the burden of costs incurred

by patients (-163 percentage points). However, due to the relationship between these

37

variables8, a further investment of 752$ (580 times the value of annual income) would be

required to reduce the cost burden to below a “catastrophic” threshold. This result stresses

that in addition to extending coverage of cash transfer programmes to TB patients, future

efforts to reduce the value of costs incurred by TB patients will also be essential to eliminate

“catastrophic” costs.

Comparison of the mitigation potential of selected cash transfer programmes using two

approaches highlighted that in most cases, according to which approach is used, the

estimated impact of programmes is very different. For example, in Tanzania where the

equivalence of the cash transfers relative to costs is low (318$ vs. 837$, 38%), the estimated

percentage point reduction achieved by cash transfers is high (-139 percentage points). This

result demonstrates that depending on how the objectives of cash transfers are interpreted

then their impact might vary greatly.

Case studies of existing TB-related cash transfer programmes in Peru and Brazil provided

some key evidence for ways to conceptualise TB-related cost mitigation. In particular,

evidence from the CRESIPT project in Peru suggests that cash transfers equal to 20% of costs

incurred by patients might be sufficient to enable better access to care. According to this it is

possible that a much smaller proportion of cash transfers provided by programmes like “Bolsa

Familia” might be sufficient to support patient’s access to TB care. Nevertheless, this will

depend on how reliant beneficiaries are of cash transfers received from cash transfer

programmes, as it is unethical to expect beneficiaries to sacrifice income necessary for “basic

needs”; or their right to receive transfers as a result of non-compliance with programme

conditionalities.

In Brazil, recognising that cash transfers provided by “Bolsa Familia” are mostly provided for

broader objectives, the national TB programme proposes extending a supplementary cash

transfer to programme beneficiaries affected by TB. However, due to limited funding capacity,

this initiative cannot expect to provide cash transfers equal to the full value of costs incurred

by TB patients. Therefore patients will still have to rely on a certain proportion of annual income

to mitigate the full value of costs. Finally, whilst this proposal provides evidence of the

commitment of national TB programmes to mitigate costs, it also highlights that successful

cross-sectoral collaborations between agencies are reliant on politics as well as a strong

evidence base.

8 This study highlighted that for any given value of costs and income there are diminishing returns in

burden reduction per unit increase in the value of cash transfers.

38

Main strengths of the study.

This is the first study attempt to summarise the literature on TB costs and cash transfers

programmes. Thanks to this, national TB programmes in selected countries will have a better

understanding of the coverage of cash transfer programmes in their own country, the kind of

cash transfers that are provided to beneficiaries, and the extent to which these programmes

might mitigate costs incurred by TB patients. This information should provide an incentive for

further efforts to establish cross-sectoral collaborations between national TB programmes and

non-health sectors around the world. Also, this study is the first to attempt to evaluate of the

concept of cost mitigation. In the absence of quantifiable data, this preliminary study relied on

secondary data, but despite this, the project raised important questions/issues that have been

so far neglected in the policy debate. These include the importance of how we conceptualise

the impact of cash transfer programmes, and the need to consider the relationship between

patient-incurred costs, cash transfers and patient’s income in each context. By doing so, this

study can be considered a fundamental starting point for future debate and action in this field.

Study limitations.

Obviously conclusions need to be drawn cautiously as this study has several important

limitations. First, use of maximum cash transfers provided by selected programmes

overestimates their potential impact on TB costs. In the majority of contexts the range of cash

transfers available to beneficiaries is extremely wide, illustrated in Brazil where the aggregated

value of annual cash transfers ranged from approximately 240$ to 2535$. However, by

evaluating the maximum potential mitigation effect of cash transfers, in countries where cash

transfers have a low mitigation potential this project highlights the need for further research to

evaluate other ways to mitigate TB-related costs. Second, inclusion bias due to use of only

one reference source, and limited availability of articles with patient incurred cost data, limits

the number of TB-sensitive cash transfer programmes documented by this study.

For patient-incurred cost data, the use of mean cost values limits the representativeness of

the findings, as it is well documented that this type of data is skewed, with some patients

incurring very high costs, whilst the majority incur very low costs (5). In terms of diagnostic

costs, surveys only ever included people who have been diagnosed with TB, and costs for

those ill with TB but that never get diagnosed are not represented. Patient-incurred costs might

also have been biased as many articles only surveyed certain specific costs, and also because

some used purposive sampling to evaluate costs incurred by a specific sub-populations.

Further, there is also inclusion bias for cost articles with regards to some publication

languages, and restricted access to certain journals.

39

Manipulation of the data to estimate average total costs over the full duration of treatment

further limits the accuracy of the cost data. However, this was the only way to gain an

approximate estimate of the value of total costs over full duration of care. Also, due to limited

availability this study had to use an estimate of income based on per capita GDP, and it is

very likely that TB patients’ average income varies substantially from the national average.

While these are important limitations, there is currently no better source of data, highlighting

important data gaps that need to be filled for future planning and research in this field.

Finally, the biggest limitation of this project is perhaps the inconsistency of the results

depending on the definition and analytical approach used. This latter limitation is the most

relevant hampering the capacity of this study to provide any recommendation to country

programmes. While this could be due to the distinct mathematical approaches used to

conceptualise cost mitigation, it is also possible that through these different definitions the

distinct objectives of TB-specific and TB-sensitive programmes are truly represented. If so,

the consistency between the two measurements is less important.

Interpretation of results.

Despite its limitations, this study shows that in order to obtain a true representation of the

mitigation effect of cash transfer programmes it will be critical to account for the definition

used, which in turn is likely to depend on the programme’s implementations strategy (TB-

specific vs TB-sensitive). From a TB-specific perspective, like that adopted by “CRESIPT” in

Peru, cash transfers are intended with the sole purpose of improving TB outcomes, perhaps

indicating that a cost equivalence approach might be most appropriate. Conversely, for TB-

sensitive programmes like “Bolsa Familia” that are intended for a broader range of objectives,

the second approach used in this study might be more appropriate, as it considers cash

transfers as simply a boost to annual income. With the right instrument, both approaches might

equally be applied for evaluating the mitigation potential of cash transfer programmes;

however, further research is needed to inform the development of either instrument.

Policy implications and future research directions.

The new TB elimination strategy emphasises social protection as a non-negotiable component

of the TB response to eliminate “catastrophic” costs by 2025. However, as yet there is still no

consensus on how to define or evaluate “catastrophic” costs. This study went beyond this and

added another stream of thought by highlighting a lack of consensus on the definition of cost

mitigation. The result that emerged from this project is key for highlighting the need for future

discretion in deciding which definition should be used to evaluate the potential of cash transfer

40

programmes to contribute to TB-related cost mitigation, and to inform future research in this

field.

Across the world, and especially in sub Saharan Africa, there is increasing momentum to

provide adequate and targeted welfare support using policies like cash transfer programmes.

This was highlighted recently in Zambia where the Government increased funding for the

national “Social Cash Transfer Scheme” by an unprecedented 700 per cent (96). Policy

interventions such as these suggest that it might be a perfect time to design “natural

experiments”9 to evaluate what the potential impact of cash transfers provided by cash transfer

programmes might be on costs incurred by TB patients (104). Efforts to generate a good

evidence base for the potential of social protection will be essential to promote, and sustain a

more holistic and effective approach to TB control (37,42).

Box 2: Key conclusions.

Key Conclusions

A number of cash transfer programmes seem to have already great potential to

mitigate costs.

In countries where this mitigation is not achieved additional strategies may be

required such as the integration of other forms of social protection (such as micro-

insurance)

The observed costs mitigation results seem to be subject to the way we

conceptualise mitigation and consequently measure it.

This study demonstrates the importance of reaching consensus on the following

a) How do we define and measure costs mitigation?

b) How do we define and measure costs elimination?

c) How these concepts should be applied depending on cash transfer

implementation strategy.

This consensus is essential to inform better designed studies to rigorously assess

the cost mitigation potential of cash transfer programmes, and other forms of social

protection.

9 Natural experiments are observational studies which can be undertaken to assess the outcomes and impacts of policy interventions.

41

References.

1. WHO | Closing the gap in a generation - how? [Internet]. WHO. [cited 2015 Aug 16]. Available from: http://www.who.int/social_determinants/thecommission/finalreport/closethegap_how/en/index1.html

2. WHO | Global tuberculosis report 2014 [Internet]. WHO. [cited 2015 Aug 13]. Available from: http://www.who.int/tb/publications/global_report/en/

3. Lönnroth K, Castro KG, Chakaya JM, Chauhan LS, Floyd K, Glaziou P, et al. Tuberculosis control and elimination 2010-50: cure, care, and social development. Lancet Lond Engl. 2010 May 22;375(9728):1814–29.

4. WHO | The five elements of DOTS [Internet]. WHO. [cited 2015 Aug 14]. Available from: http://www.who.int/tb/dots/whatisdots/en/

5. Tanimura T, Jaramillo E, Weil D, Raviglione M, Lönnroth K. Financial burden for tuberculosis patients in low- and middle-income countries: a systematic review. Eur Respir J. 2014 Jun;43(6):1763–75.

6. Lawn SD, Shattock RJ, Griffin GE. Delays in the diagnosis of tuberculosis: a great new cost. Int J Tuberc Lung Dis Off J Int Union Tuberc Lung Dis. 1997 Oct;1(5):485–6.

7. Zhang T, Tang S, Jun G, Whitehead M. Persistent problems of access to appropriate, affordable TB services in rural China: experiences of different socio-economic groups. BMC Public Health. 2007 Feb 8;7:19.

8. Wingfield T, Boccia D, Tovar M, Gavino A, Zevallos K, Montoya R, et al. Defining catastrophic costs and comparing their importance for adverse tuberculosis outcome with multi-drug resistance: a prospective cohort study, Peru. PLoS Med. 2014 Jul;11(7):e1001675.

9. Lönnroth K, Jaramillo E, Williams BG, Dye C, Raviglione M. Drivers of tuberculosis epidemics: The role of risk factors and social determinants. Soc Sci Med. 2009 Jun;68(12):2240–6.

10. Rajeswari R, Balasubramanian R, Muniyandi M, Geetharamani S, Thresa X, Venkatesan P. Socio-economic impact of tuberculosis on patients and family in India. Int J Tuberc Lung Dis. 1999 Oct 1;3(10):869–77.

11. Ukwaja KN, Modebe O, Igwenyi C, Alobu I. The economic burden of tuberculosis care for patients and households in Africa: a systematic review [Review article]. Int J Tuberc Lung Dis. 2012 Jun 1;16(6):733–9.

12. Farmer. The consumption of the poor. In Infections and Inequalities: The modern Plagues. Updated Edition. University of California Press; 2001.

13. Blakely T, Hales S, Kieft C, Wilson N, Woodward A. The global distribution of risk factors by poverty level. Bull World Health Organ. 2005 Feb;83(2):118–26.

14. Umar NA, Fordham R, Abubakar I, Bachmann M. The indirect cost due to pulmonary Tuberculosis in patients receiving treatment in Bauchi State—Nigeria. Cost Eff Resour Alloc. 2012 May 11;10(1):6.

42

15. Mauch V, Bonsu F, Gyapong M, Awini E, Suarez P, Marcelino B, et al. Free tuberculosis diagnosis and treatment are not enough: patient cost evidence from three continents. Int J Tuberc Lung Dis Off J Int Union Tuberc Lung Dis. 2013 Mar;17(3):381–7.

16. Munro SA, Lewin SA, Smith HJ, Engel ME, Fretheim A, Volmink J. Patient Adherence to Tuberculosis Treatment: A Systematic Review of Qualitative Research. PLoS Med [Internet]. 2007 Jul [cited 2015 Aug 13];4(7). Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1925126/

17. Needham DM, Godfrey-Faussett P, Foster SD. Barriers to tuberculosis control in urban Zambia: the economic impact and burden on patients prior to diagnosis. Int J Tuberc Lung Dis. 1998 Oct 1;2(10):811–7.

18. Kemp JR, Mann G, Simwaka BN, Salaniponi FM, Squire SB. Can Malawi’s poor afford free tuberculosis services? Patient and household costs associated with a tuberculosis diagnosis in Lilongwe. Bull World Health Organ. 2007 Aug;85(8):580–5.

19. Laokri S, Weil O, Drabo KM, Dembelé SM, Kafando B, Dujardin B. Removal of user fees no guarantee of universal health coverage: observations from Burkina Faso. Bull World Health Organ. 2013 Apr;91(4):277–82.

20. Stop TB Partnership | The Global Plan to Stop TB | The Global Plan to Stop TB 2006-2015 [Internet]. [cited 2015 Aug 13]. Available from: http://www.stoptb.org/global/plan/main/default.asp

21. WHO | Addressing Poverty in TB Control. Options for National TB Control Programmes<br />(WHO/HTM/TB/2005.352) [Internet]. WHO. [cited 2015 Aug 14]. Available from: http://www.who.int/tb/challenges/poverty/en/

22. WHO | WHO End TB Strategy [Internet]. WHO. [cited 2015 Aug 14]. Available from: http://www.who.int/tb/post2015_strategy/en/

23. Raviglione M, Zumla A, Marais B, Horton R, Motsoaledi A. A sustainable agenda for tuberculosis control and research. The Lancet. 2012 Mar 30;379(9821):1077–8.

24. Lönnroth K, Glaziou P, Weil D, Floyd K, Uplekar M, Raviglione M. Beyond UHC: Monitoring Health and Social Protection Coverage in the Context of Tuberculosis Care and Prevention. PLoS Med [Internet]. 2014 Sep 22 [cited 2015 Aug 30];11(9). Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4171373/

25. Transformative Social Protection [Internet]. [cited 2015 Aug 14]. Available from: http://www.ids.ac.uk/publication/transformative-social-protection1

26. World Bank. The State of Social Safety Nets 2015 [Internet]. The World Bank; 2015 [cited 2015 Aug 13]. Available from: http://elibrary.worldbank.org/doi/book/10.1596/978-1-4648-0543-1

27. Fiszbein A, Schady N, Ferreira FHG, Grosh M, Keleher N, Olinto P, et al. Conditional cash transfers : reducing present and future poverty [Internet]. The World Bank; 2009 Feb [cited 2015 Aug 14] p. 1–384. Report No.: 47603. Available from: http://documents.worldbank.org/curated/en/2009/01/10298306/conditional-cash-transfers-reducing-present-future-poverty

43

28. Discussion Paper: Cash Transfers and HIV Prevention [Internet]. UNDP. [cited 2015 Aug 13]. Available from: http://www.undp.org/content/undp/en/home/librarypage/hiv-aids/discussion-paper--cash-transfers-and-hiv-prevention.html

29. Understanding and acting on critical enablers and development synergies for strategic investments [Internet]. UNDP. [cited 2015 Aug 26]. Available from: http://www.undp.org/content/undp/en/home/librarypage/hiv-aids/understanding-and-acting-on-critical-enablers-and-development-sy.html

30. FINAL EVALUATION REPORT KALOMO SOCIAL CASH TRANSFER SCHEME [Internet]. Lusaka, Zambia: Ministry of Community Development and Social Services (MCDSS) German Technical Cooperation (GTZ); 2007 Sep. Available from: http://health.bmz.de/good-practices/GHPC/Cashing_in/Toolset_2_Evaluations_and_data/Final_evaluation_report_for_Kalomo_2007.pdf

31. Cashing in: How cash transfers shore up Zambian households affected by HIV [Internet]. Federal Ministry for Economic Cooperation and Development. [cited 2015 Aug 27]. Available from: http://www.bmz.de/en/healthportal/good-practices/GHPC/Cashing_in/index.jsp

32. 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 Syst Rev. 2009;(4):CD008137.

33. Michelle Adato LB. Social protection and cash transfers: To strengthen families affected by HIV and AIDS [Internet]. Washington, DC: International Food Policy Research Institute; 2012 [cited 2015 Aug 27]. Available from: http://www.ifpri.org/publication/social-protection-and-cash-transfers

34. Heise L, Lutz B, Ranganathan M, Watts C. Cash transfers for HIV prevention: considering their potential. J Int AIDS Soc [Internet]. 2013 Aug 23 [cited 2015 Aug 30];16(1). Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3752431/

35. Baird SJ, Garfein RS, McIntosh CT, Özler B. Effect of a cash transfer programme for schooling on prevalence of HIV and herpes simplex type 2 in Malawi: a cluster randomised trial. The Lancet. 2012 Apr;379(9823):1320–9.

36. Pettifor A, MacPhail C, Nguyen N, Rosenberg M. Can money prevent the spread of HIV? A review of cash payments for HIV prevention. AIDS Behav. 2012 Oct;16(7):1729–38.

37. Social Protection Interventions for Tuberculosis Control: The Impact, the Challenges and the Way Forward [Internet]. Chatham House. [cited 2015 Aug 22]. Available from: http://www.chathamhouse.org//node/2305

38. Boccia D, Hargreaves J, Lönnroth K, Jaramillo E, Weiss J, Uplekar M, et al. Cash transfer and microfinance interventions for tuberculosis control: review of the impact evidence and policy implications. Int J Tuberc Lung Dis Off J Int Union Tuberc Lung Dis. 2011 Jun;15(Suppl 2):S37–49.

39. Lutge E, Lewin S, Volmink J, Friedman I, Lombard C. Economic support to improve tuberculosis treatment outcomes in South Africa: a pragmatic cluster-randomized controlled trial. Trials. 2013 May 28;14(1):154.

44

40. Rocha C, Montoya R, Zevallos K, Curatola A, Ynga W, Franco J, et al. The Innovative Socio-economic Interventions Against Tuberculosis (ISIAT) project: an operational assessment. Int J Tuberc Lung Dis Off J Int Union Tuberc Lung Dis. 2011 Jun;15(Suppl 2):S50–7.

41. Lutge E, Lewin S, Volmink J. Economic support to improve tuberculosis treatment outcomes in South Africa: a qualitative process evaluation of a cluster randomized controlled trial. Trials. 2014 Jun 19;15:236.

42. WHO | Eliminating the Catastrophic Economic Burden of TB: Universal Health Coverage and Social Protection Opportunities [Internet]. WHO. [cited 2015 Aug 27]. Available from: http://www.who.int/tb/uhc_socialprotection/en/

43. Wingfield T, Boccia D, Tovar MA, Huff D, Montoya R, Lewis JJ, et al. Designing and implementing a socioeconomic intervention to enhance TB control: operational evidence from the CRESIPT project in Peru. BMC Public Health. 2015 Aug 21;15(1):810.

44. Sheikh K, Gilson L, Agyepong IA, Hanson K, Ssengooba F, Bennett S. Building the Field of Health Policy and Systems Research: Framing the Questions. PLoS Med. 2011 Aug 16;8(8):e1001073.

45. Gilson L, Hanson K, Sheikh K, Agyepong IA, Ssengooba F, Bennett S. Building the Field of Health Policy and Systems Research: Social Science Matters. PLoS Med. 2011 Aug 23;8(8):e1001079.

46. Bolsa Família Program: a decade of social inclusion. Executive Summary [Internet]. Institute for Applied Economic Research; Available from: http://www.sae.gov.br/wp-content/uploads/BF-Sumex_WEB-2.pdf

47. Laurence YV, Griffiths UK, Vassall A. Costs to Health Services and the Patient of Treating Tuberculosis: A Systematic Literature Review. PharmacoEconomics. 2015 May 5;

48. Population Censuses’ Datasets (1995 - Present) [Internet]. United Nations Statistics Division; [cited 2015 Aug 22]. Available from: http://unstats.un.org/unsd/demographic/products/dyb/dybcensusdata.htm

49. Total Population (in number of people) [Internet]. The World Bank; [cited 2015 Aug 22]. Available from: http://data.worldbank.org/indicator/SP.POP.TOTL

50. GDP (current US$) [Internet]. World Bank; Available from: http://data.worldbank.org/indicator/NY.GDP.MKTP.CD

51. GDP, PPP (current international $) [Internet]. World Bank; Available from: http://data.worldbank.org/indicator/NY.GDP.MKTP.PP.CD

52. WHO | Tuberculosis country profiles [Internet]. WHO. [cited 2015 Aug 31]. Available from: http://www.who.int/tb/country/data/profiles/en/

53. Official exchange rate (LCU per US$, period average) [Internet]. World Bank; [cited 2015 Aug 22]. Available from: http://data.worldbank.org/indicator/PA.NUS.FCRF

54. Wingfield T, Boccia D, Tovar M, Gavino A, Zevallos K, Montoya R, et al. Defining Catastrophic Costs and Comparing Their Importance for Adverse Tuberculosis Outcome with Multi-Drug Resistance: A Prospective Cohort Study, Peru. PLoS Med. 2014 Jul 15;11(7):e1001675.

45

55. Russell S. The Economic Burden of Illness for Households in Developing Countries: A Review of Studies Focusing on Malaria, Tuberculosis, and Human Immunodeficiency Virus/Acquired Immunodeficiency Syndrome. Am J Trop Med Hyg. 2004 Aug 1;71(2 suppl):147–55.

56. Ranson MK. Reduction of catastrophic health care expenditures by a community-based health insurance scheme in Gujarat, India: current experiences and challenges. Bull World Health Organ. 2002;80(8):613–21.

57. Barter DM, Agboola SO, Murray MB, Bärnighausen T. Tuberculosis and poverty: the contribution of patient costs in sub-Saharan Africa – a systematic review. BMC Public Health. 2012 Nov 14;12(1):980.

58. GDP per capita (current LCU) [Internet]. World Bank; [cited 2015 Aug 22]. Available from: http://data.worldbank.org/indicator/NY.GDP.PCAP.CN

59. Yitayal M, Aseffa A, Andargie G, Wassie L, Abebe M. Assessment of cost of tuberculosis to patients and their families: a cross-sectional study at Addet Health Center, Yilmana Densa District, Amhara National Regional State. Ethiop Med J. 2014 Jan;Suppl 1:23–30.

60. Elamin EI, Ibrahim MIM, Sulaiman SAS, Muttalif AR. Cost of illness of tuberculosis in Penang, Malaysia. Pharm World Sci PWS. 2008 Jun;30(3):281–6.

61. Guzmán-Montes GY, Heras-Ovalles R, Laniado-Laborín R. Indirect patient expenses for antituberculosis treatment in Tijuana, Mexico: is treatment really free? J Infect Dev Ctries. 2009 Nov 5;3(10):778–82.

62. Steffen R, Menzies D, Oxlade O, Pinto M, de Castro AZ, Monteiro P, et al. Patients’ Costs and Cost-Effectiveness of Tuberculosis Treatment in DOTS and Non-DOTS Facilities in Rio de Janeiro, Brazil. PLoS ONE [Internet]. 2010 Nov 17 [cited 2015 Aug 22];5(11). Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2984447/

63. Costa JG, Santos AC, Rodrigues LC, Barreto ML, Roberts JA. Tuberculosis in Salvador, Brazil: costs to health system and families. Rev Saúde Pública. 2005 Jan;39(1):122–8.

64. do Prado TN, Wada N, Guidoni LM, Golub JE, Dietze R, Maciel ELN. Cost-effectiveness of community health worker versus home-based guardians for directly observed treatment of tuberculosis in Vitória, Espírito Santo State, Brazil. Cad Saude Publica. 2011 May;27(5):944–52.

65. Zou G, Wei X, Witter S, Yin J, Walley J, Liu S, et al. Incremental cost-effectiveness of improving treatment results among migrant tuberculosis patients in Shanghai. Int J Tuberc Lung Dis. 2013 Aug 1;17(8):1056–64.

66. Xu B, Dong HJ, Zhao Q, Bogg L. DOTS in China – removing barriers or moving barriers? Health Policy Plan. 2006 Sep 1;21(5):365–72.

67. Jackson S, Sleigh AC, Wang G-J, Liu X-L. Poverty and the economic effects of TB in rural China. Int J Tuberc Lung Dis. 2006 Oct 1;10(10):1104–10.

68. Meng Q, Li R, Cheng G, Blas E. Provision and financial burden of TB services in a financially decentralized system: a case study from Shandong, China. Int J Health Plann Manage. 2004 Dec;19 Suppl 1:S45–62.

46

69. Pan H-Q, Bele S, Feng Y, Qiu S-S, Lü J-Q, Tang S-W, et al. Analysis of the economic burden of diagnosis and treatment of tuberculosis patients in rural China. Int J Tuberc Lung Dis Off J Int Union Tuberc Lung Dis. 2013 Dec;17(12):1575–80.

70. Rouzier VA, Oxlade O, Verduga R, Gresely L, Menzies D. Patient and family costs associated with tuberculosis, including multidrug-resistant tuberculosis, in Ecuador. Int J Tuberc Lung Dis. 2010 Oct 1;14(10):1316–22.

71. Mahendradhata Y, Probandari A, Ahmad RA, Utarini A, Trisnantoro L, Lindholm L, et al. The Incremental Cost-Effectiveness of Engaging Private Practitioners to Refer Tuberculosis Suspects to DOTS Services in Jogjakarta, Indonesia. Am J Trop Med Hyg. 2010 Jun;82(6):1131–9.

72. Aspler A, Menzies D, Oxlade O, Banda J, Mwenge L, Godfrey-Faussett P, et al. Cost of tuberculosis diagnosis and treatment from the patient perspective in Lusaka, Zambia. Int J Tuberc Lung Dis. 2008 Aug 1;12(8):928–35.

73. Khan MA, Walley JD, Witter SN, Imran A, Safdar N. Costs and cost-effectiveness of different DOT strategies for the treatment of tuberculosis in Pakistan. Health Policy Plan. 2002 Jun 1;17(2):178–86.

74. Umar NA, Abubakar I, Fordham R, Bachmann M. Direct costs of pulmonary tuberculosis among patients receiving treatment in Bauchi State, Nigeria. Int J Tuberc Lung Dis. 2012 Jun 1;16(6):835–40.

75. Othman GQ, Ibrahim MIM, Raja’a YA. Costs associated with tuberculosis diagnosis and treatment in Yemen for patients and public health services. East Mediterr Health J Rev Santé Méditerranée Orient Al-Majallah Al-Ṣiḥḥīyah Li-Sharq Al-Mutawassiṭ. 2012 Apr;18(4):393–8.

76. Gospodarevskaya E, Tulloch O, Bunga C, Ferdous S, Jonas A, Islam S, et al. Patient costs during tuberculosis treatment in Bangladesh and Tanzania: the potential of shorter regimens. Int J Tuberc Lung Dis. 2014 Jul 1;18(7):810–7.

77. Wandwalo E, Robberstad B, Morkve O. Cost and cost-effectiveness of community based and health facility based directly observed treatment of tuberculosis in Dar es Salaam, Tanzania. Cost Eff Resour Alloc CE. 2005 Jul 14;3:6.

78. Wyss K, Kilima P, Lorenz N. Costs of tuberculosis for households and health care providers in Dar es Salaam, Tanzania. Trop Med Int Health. 2001 Jan 1;6(1):60–8.

79. Datiko DG, Lindtjørn B. Cost and Cost-Effectiveness of Treating Smear-Positive Tuberculosis by Health Extension Workers in Ethiopia: An Ancillary Cost-Effectiveness Analysis of Community Randomized Trial. PLoS ONE. 2010 Feb 17;5(2):e9158.

80. Vassall A, Seme A, Compernolle P, Meheus F. Patient costs of accessing collaborative tuberculosis and human immunodeficiency virus interventions in Ethiopia. Int J Tuberc Lung Dis. 2010 May 1;14(5):604–10.

81. Mesfin MM, Newell JN, Madeley RJ, Mirzoev TN, Tareke IG, Kifle YT, et al. Cost implications of delays to tuberculosis diagnosis among pulmonary tuberculosis patients in Ethiopia. BMC Public Health. 2010 Mar 30;10(1):173.

47

82. Floyd K, Skeva J, Nyirenda T, Gausi F, Salaniponi F. Cost and cost-effectiveness of increased community and primary care facility involvement in tuberculosis care in Lilongwe District, Malawi. Int J Tuberc Lung Dis. 2003 Sep 1;7(9):S29–37.

83. Yemen National Social Protection Monitoring Survey (NSPMS): 2012-2013 Final Report [Internet]. UNICEF; 2014 Dec. Available from: http://www.ipc-undp.org/Yemen%20National%20Social%20Protection%20Monitoring%20Survey%20(NSPMS)%3A%202012-2013

84. Database of Conditional Transfer Programmes (Programas de Transferencias Condicionadas) [Internet]. ECLAC (Economic Commission for Latin America and the Caribbean of the United Nations); Available from: http://dds.cepal.org/bdptc/en/

85. International Labour Office. Rationalizing social protection expenditure in Ghana consolidated version: draft for comments. Geneva; 2014. 196 p.

86. Wang D, Umapathi N, O’Keefe P. Eligibility thresholds for minimum living guarantee programs : international practices and implications for China [Internet]. The World Bank; 2013 Nov [cited 2015 Aug 29] p. 1–48. Report No.: 83118. Available from: http://documents.worldbank.org/curated/en/2013/11/18600397/eligibility-thresholds-minimum-living-guarantee-programs-international-practices-implications-china

87. Bank TW. Program Keluarga Harapan (PKH) conditional cash transfer [Internet]. The World Bank; 2012 Feb [cited 2015 Aug 29] p. 1–38. Report No.: 67309. Available from: http://documents.worldbank.org/curated/en/2012/02/15893876/program-keluarga-harapan-pkh-conditional-cash-transfer

88. Rebecca Holmes JH-Z. Social Protection in Nigeria. Overseas Development Institute;

89. Rebecca Holmes MS, Wendy Magoronga BA, Jenny Morgan. The potential for cash transfers in Nigeria. Overseas Development Institute; 2012 Feb.

90. BR1M Online | Bantuan Rakyat 1 Malaysia [Internet]. BR1M Online. [cited 2015 Aug 29]. Available from: http://www.br1m.info/

91. Covarrubias K, Davis B, Winters P. From protection to production: productive impacts of the Malawi Social Cash Transfer scheme. J Dev Eff. 2012 Mar 1;4(1):50–77.

92. Tanzania Social Action Fund Condtional Cash Transfer [Internet]. [cited 2015 Aug 30]. Available from: http://tasaf.org/index.php/tasaf-components/supported-interventions/conditional-cash-transfer

93. Qualitative research and analyses of the economic impacts of cash transfer programmes in Sub-Saharan Africa [Internet]. Food and Agriculture Organization of the United Nations. [cited 2015 Aug 18]. Available from: http://www.fao.org/documents/card/en/c/b4d42f56-5a55-4bac-8e6f-16e2c38c5bfb

94. FAO. The Impacts of the Social Cash Transfer Pilot Programme on Community Dynamics in Tigray, Ethiopia. 2014 Aug 1 [cited 2015 Aug 29]; Available from: http://r4d.dfid.gov.uk/Output/199830/Default.aspx

95. Benazir Income Support Programme [Internet]. Available from: http://www.bisp.gov.pk/news.aspx

48

96. Stanfield Michelo. Social Cash Transfer Scale-up for Zambia [Internet]. International Policy Centre for Inclusive Growth; 2015. Available from: http://www.ipc-undp.org/pub/eng/OP287_Social_Cash_Transfer_Scale_Up_for_Zambia.pdf

97. Tesliuc CM. Using social safety nets to accelerate poverty reduction and share prosperity in Zambia [Internet]. The World Bank; 2013 Mar [cited 2015 Aug 18] p. 1–157. Report No.: 68245. Available from: http://documents.worldbank.org/curated/en/2013/03/17583491/using-social-safety-nets-accelerate-poverty-reduction-share-prosperity-zambia

98. de Albuquerque M de FPM, Ximenes RA de A, Lucena-Silva N, de Souza WV, Dantas AT, Dantas OMS, et al. Factors associated with treatment failure, dropout, and death in a cohort of tuberculosis patients in Recife, Pernambuco State, Brazil. Cad Saúde Pública. 2007 Jul;23(7):1573–82.

99. Belo MTCT, Selig L, Luiz RR, Hanson C, Luna AL, Teixeira EG, et al. Choosing incentives to stimulate tuberculosis treatment compliance in a poor county in Rio de Janeiro state, Brazil. Med Sci Monit Int Med J Exp Clin Res. 2006 May;12(5):PH1–5.

100. CAIXA [Internet]. About CAIXA. [cited 2015 Apr 9]. Available from: http://www.caixa.gov.br/Paginas/home-caixa.aspx

101. Hargreaves JR, Boccia D, Evans CA, Adato M, Petticrew M, Porter JDH. The Social Determinants of Tuberculosis: From Evidence to Action. Am J Public Health. 2011 Apr;101(4):654–62.

102. Lönnroth K, Castro KG, Chakaya JM, Chauhan LS, Floyd K, Glaziou P, et al. Tuberculosis control and elimination 2010–50: cure, care, and social development. The Lancet. 2010 May 28;375(9728):1814–29.

103. International Labour Office. Rationalizing social protection expenditure in Ghana consolidated version: draft for comments. Geneva; 2014. 196 p.

104. Craig P, Cooper C, Gunnell D, Haw S, Lawson K, Macintyre S, et al. Using natural experiments to evaluate population health interventions: new Medical Research Council guidance. J Epidemiol Community Health. 2012 Dec;66(12):1182–6.

49

Appendix.

Screening for eligible cash transfer programmes.

Countries Eligible conditional

poverty related cash transfer programme

Eligible unconditional poverty related cash transfer programme

Programme name

Bangladesh

Benin

Botswana

Brazil Bolsa Familia Programme

Burkina Faso

Cambodia

China

Dibao Minimum Living Standards Scheme

Dom. Republic

Progresando con Solidaridad

Ecuador Bono de Desarollo Humano

Egypt

Ethiopia Pilot Social Cash Transfer

Ghana Livelihood Empowerment Against Poverty

Program

Haiti

India

Indonesia Program Keluarga Harapan

Kenya

Malawi Social Cash Transfer

Malaysia

Bantuan Rakyat 1 Malaysia

Mexico Oportunidades

Myanmar

Nepal

Nigeria Eradication of Extreme Poverty and Hunger

Pakistan Benazir Income Support Programme

Sierra Leone

South Africa

Sudan

Syria

Tajikistan

Targeted Social Assistance

Tanzania Tanzania Social Action Fund

Thailand

Uganda

Ukraine

Vietnam

Yemen

Social Welfare Fund

Zambia Social Cash Transfer Scheme

Zimbabwe

50

Equation 1: Potential of cash transfers to mitigate costs.

𝐶𝑜𝑠𝑡 𝑚𝑖𝑡𝑖𝑔𝑎𝑡𝑖𝑜𝑛 𝑒𝑓𝑓𝑒𝑐𝑡 =𝐴𝑛𝑛𝑢𝑎𝑙 𝑐𝑎𝑠ℎ 𝑡𝑟𝑎𝑛𝑠𝑓𝑒𝑟

𝑇𝐵 𝑐𝑜𝑠𝑡𝑠× 100

Equation 2: Additional cash necessary to mitigate costs fully.

𝐴𝑑𝑑𝑖𝑡𝑖𝑜𝑛𝑎𝑙 𝑐𝑎𝑠ℎ 𝑛𝑒𝑐𝑒𝑠𝑠𝑎𝑟𝑦 𝑡𝑜 𝑚𝑖𝑡𝑖𝑔𝑎𝑡𝑒 𝑐𝑜𝑠𝑡𝑠 𝑓𝑢𝑙𝑙𝑦 = 𝑇𝐵 𝑐𝑜𝑠𝑡𝑠 − 𝐴𝑛𝑛𝑢𝑎𝑙 𝑐𝑎𝑠ℎ 𝑡𝑟𝑎𝑛𝑠𝑓𝑒𝑟

Equation 3: Potential of cash transfers to alleviate cost burden.

𝐶𝑜𝑠𝑡 𝑏𝑢𝑟𝑑𝑒𝑛 𝑎𝑙𝑙𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 𝑒𝑓𝑓𝑒𝑐𝑡

= (𝑇𝐵 𝑐𝑜𝑠𝑡𝑠

𝐴𝑛𝑛𝑢𝑎𝑙 𝑖𝑛𝑐𝑜𝑚𝑒× 100) − (

𝑇𝐵 𝑐𝑜𝑠𝑡𝑠

𝐴𝑛𝑛𝑢𝑎𝑙 𝑖𝑛𝑐𝑜𝑚𝑒 + 𝐴𝑛𝑛𝑢𝑎𝑙 𝑐𝑎𝑠ℎ 𝑡𝑟𝑎𝑛𝑠𝑓𝑒𝑟× 100)

Equation 4: Additional cash necessary to alleviate cost burden to 20% threshold.

𝐴𝑑𝑑𝑖𝑡𝑖𝑜𝑛𝑎𝑙 𝑐𝑎𝑠ℎ 𝑛𝑒𝑐𝑒𝑠𝑠𝑎𝑟𝑦 𝑡𝑜 𝑎𝑙𝑙𝑒𝑣𝑖𝑎𝑡𝑒 𝑐𝑜𝑠𝑡 𝑏𝑢𝑟𝑑𝑒𝑛 𝑡𝑜 20% 𝑡ℎ𝑟𝑒𝑠ℎ𝑜𝑙𝑑

= (𝑇𝐵 𝑐𝑜𝑠𝑡𝑠

0.2) − (𝐴𝑛𝑛𝑢𝑎𝑙 𝑖𝑛𝑐𝑜𝑚𝑒 + 𝐴𝑛𝑛𝑢𝑎𝑙 𝐶𝑎𝑠ℎ 𝑡𝑟𝑎𝑛𝑠𝑓𝑒𝑟)

Equation 5: Annual individual income.

𝐴𝑛𝑛𝑢𝑎𝑙 𝑖𝑛𝑑𝑖𝑣𝑖𝑑𝑢𝑎𝑙 𝑖𝑛𝑐𝑜𝑚𝑒 =(𝐺𝐷𝑃 × 𝑖𝑛𝑐𝑜𝑚𝑒 𝑠ℎ𝑎𝑟𝑒 ℎ𝑒𝑙𝑑 𝑏𝑦 𝑡ℎ𝑒 𝑙𝑜𝑤𝑒𝑠𝑡 20%)

(𝑇𝑜𝑡𝑎𝑙 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑠𝑖𝑧𝑒 × 0.2)