1. Introduction

HIV and AIDS have remained a public health concern since the beginning of its epidemic ^{[1, 2]}. With approximately 37 million people currently living with HIV, the disease creates a serious socioeconomic problem, with medical and treatment costs accounting for a large percentage of the economic burden ^{[3, 4, 5]}. Costs of implementing programs is another heavy burden—UNAIDS estimated that US$20.2 billion was spent on HIV programs in 2014, US$8.64 billion of which were from international investments, a 2% increase from 2013 estimates ^[6]. Although the rates of new HIV infection has been declining due to significant improvements in interventions and treatments, these estimates from UNAIDS suggest that health expenditures are expected to continue to increase, indicating a need for greater allocation of resources in HIV programs worldwide ^[6].

Many studies have attempted to demonstrate the impact of community-based HIV prevention, treatment, and support programs on prevention, care, and treatment outcomes in different populations. In Rwanda, a community-based program on intensive treatment support for antiretroviral therapy (ART) proved successful in retaining low rates of mortality of people living with HIV (PLHIV) ^[7]. Haiti’s nationwide HIV prevention programs, which focused on improving knowledge, safe sex practices, and access to condoms, have contributed to significant declines in HIV prevalence among their general population ^[8]. In Cambodia, the 100% condom use program has greatly slowed its HIV epidemic; by 2014, HIV prevalence in the general population had fallen to 0.28% from a peak of over 2% a decade before ^{[9, 10]}. Outstanding national leadership and commitment was recognized through the Millennium Development Goal (MDG) Award in 2010, when Cambodia reached its universal access target for ART, and halted and started to reverse the spread of HIV ^[11].

Cambodia’s success in reducing its HIV prevalence now enables the focus to shift to ensuring the most effective use of resources, whilst maintaining a strong impact at the national level. KHANA, the largest national, non-governmental organization (NGO) providing HIV prevention, care, and support services in Cambodia, has been committed to community health and development issues. In 2007, KHANA implemented a European Commission (EC) supported HIV and AIDS program with two main program components: Focused Prevention (FP) for key populations, including female entertainment workers, men who have sex with men, transgender women, and people who inject drugs, and Integrated Care and Prevention (ICP) for PLHIV and orphans and vulnerable children (OVC). The ICP project supports groups of PLHIV and OVC families to establish savings groups, using a village saving scheme model similar to those introduced by CARE International and other organizations ^[12]. Through the ICP component, KHANA provided community-based care and support, and have taken a holistic approach to the varied needs of individuals and communities by facilitating access to HIV care and treatment, providing socio-economic support to PLHIV, OVC and their families, and reducing stigma and discrimination in the communities ^[13].

However, with the major efforts and investments to reduce HIV globally, the importance of “value for money” has been increasingly recognized by the development sector in national and donor agendas, as upwards and downwards accountability and effective use of resources becomes more transparent. Most impact-evaluation studies have not included cost-benefit analyses of HIV intervention programs to include the social impacts of the programs on its population. To address this issue, the International HIV/AIDS Alliance adapted and simplified a Social Return on Investment (SROI) methodology for use at the community level. SROI is a form of cost-benefit analysis that evaluates the social, health, environmental, and economic impacts of a program in order to monetize program outputs and outcomes, which may otherwise be difficult to quantify ^{[14, 15]}.

India and Zambia, part of the International HIV/AIDS Alliance, have used the SROI methodology to assess the social and economic impacts of their community-based HIV intervention programs with promising results. A pilot study of the Alliance India’s OVC home-based care and support program (CHAHA) found their SROI ratio to be 1:4.0 based on their work in Maharastra State, meaning for every $1 invested into the CHAHA program, there would be a generated $4 return of social, health, and financial value ^[16]. The Alliance Africa’s Regional Program Phase 2 (ARP 2) partner capacity building and anti-stigma working with PLHIV found their program’s SROI ratio to be 1:21.20 in Mumbwa and 1:13.75 in Mazabuka ^[17].

KHANA has applied the SROI methodology to its ICP project to determine the value generated by the program through social, health, and economic support it provides. The SROI study was the first of its kind conducted in Cambodia—it is a timely and relevant study that uses an applicable approach in line with the donors’ current emphasis on cost-efficiency and effectiveness, value for money, and “doing more with less” in terms of cost-benefit. The objective of this study is to use the SROI methodology to assess costs, benefits, and impacts of the ICP project on community-based responses to HIV prevention, care, and treatment in communities reached by the program.

2. Methods

The SROI methodology is a participatory, beneficiary-led approach that measures and accounts for the values created by a program or series of initiatives ^{[14, 15]}. Values, defined by the program beneficiaries themselves during community consultation exercises, were mapped to common outcomes of the project, as outlined by the SROI methodology guide. Stories of change (i.e. participants’ living and health conditions) were captured during the discussions.

At first, consultative workshops and key informant interviews (KIIs) using a guided focus group discussion (FGD) approach, involved a total of 134 participants recruited through convenience sampling, and were representatives of the program’s target population: 62 PLHIV, 62 OVC caregivers, two village chiefs, four community service volunteers (CSVs), three project staff, and one health center staff. The FGDs allowed the research team to identify the most common and relevant outcomes of the ICP project. Questions and guidance for the FGDs were discussed at an initial briefing between research, monitoring, and finance staff at KHANA.

Secondly, the research team examined the extent to which the outcomes had been achieved by mapping the beneficiary-defined outcomes against relevant indicators collected in the end-line evaluation. During the SROI research period, KHANA also conducted an end-line survey for the ICP project to capture outcome changes across a range of indicators. Data collection for the end-line evaluation took place in November and December 2011 in three provinces of Cambodia, and was based on more than 1,600 interviews. For the ease of logistical planning and time constraints, the research team conducted the beneficiary consultations in one province that was already being covered by the end-line evaluation.

For this study, five stages of the SROI process were generated for analyses: (1) identifying key stakeholders; (2) mapping project outcomes using theory of change; (3) identifying common outcomes and proxies; (4) establishing project impact; and (5) calculating inputs to the project.

Key stakeholders in the ICP community-based care and support component were identified, and a decision was made over whether to include them in the analysis (Table 1).

Table 1. Stakeholders Included in the SROI analysis

Download as

PowerPoint Slide

Larger image(png format)

• Tables index

Veiw figure

View current table in a new window

View next table

The relationship between the project activities, outputs, outcomes, and impact were assessed using the theory of change and represented through impact maps. Mapping project outcomes showed how a series of programmatic activities of community action on HIV prevention, care, and impact mitigation could lead to tangible positive and negative changes in the lives of beneficiaries.

The impact maps were reconstructed with the inputs from the beneficiary consultations. Beneficiaries were consulted in groups of about 15 respondents. KHANA’s secondary qualitative research from the baseline and mid-term reviews of the ICP project were also instrumental in informing and framing the outcomes from the perspective of beneficiaries, allowing a better understanding of outcome-level changes in their lives ^[18].

Common outcomes, indicators, and financial proxies of the ICP project were identified for each beneficiary group (Table 2). A financial proxy is a means of placing a financial value on a non-financially related outcome ^{[14, 15]}. As of yet, there are few financial proxy data banks in developing countries.

A list of 18 emerging outcomes was identified for the beneficiaries as the unit of analysis. Note that not all outcomes identified in Table 2 were used towards the end calculation of the SROI ratio.

The SROI approach allowed researchers to isolate the impact of a particular activity on the project outcomes. To do this, the research team analyzed the outcomes by measuring a) attribution, b) deadweight, and c) drop-off ^[15].

• Attribution - an assessment of how much of the outcome was caused by the contribution of others (organizations or people): “who else contributed and what is their claim in achieving the outcome?”

• Deadweight - a measure of the outcome that would have happened even if the project activity had not taken place. This determines the percentage of the outcome that would have happened without any intervention.

• Drop-off - measures the effect of an outcome after the project has finished; i.e. the value that is forecast to continue for a period of time into the future. For example, when the ICP project ends, it is likely that some of the benefits realized through the income generating activity (IGA) will continue, meaning that the drop-off estimate would be low for this outcome.

Resources invested into the ICP included the direct program budget, complimentary food support from the World Food Programme (WFP), and economic costs captured within the community (i.e. self-help groups, caregivers, volunteers, and opportunity costs).

The WFP monthly food support for vulnerable PLHIV and OVC households covered 30 kg rice, 1 kg cooking oil, and 0.5 kg iodine salt.

The ICP budget accounted for 73% of the total EC budget of US$3,295,741, and covered staff costs, equipment and supplies, administrative costs, travel, training, and sub-grants to implementing partners.

Table 2. Outcomes, indicators, and financial proxies of each beneficiary group

Download as

PowerPoint Slide

Larger image(png format)

• Tables index

Veiw figure

View current table in a new window

View previous table

View next table

Financial proxies of the ICP outcomes were adjusted with a discount rate and Purchasing Power Parity (PPP) to take into account any changes (positive or negative) in the value of money over time.

The net present value (NPV) accounts for the value of money over time and includes inflation and a comparable return. When calculating the present value of an investment or input, the NPV uses a rate that discounts the cost of capital and deducts the sum from the investment level (i.e. discounting). A positive result would indicate a good investment in financial terms. There is no specific discount rate for evaluation of health projects in Cambodia. For this study, a discount rate of 5.3% was applied to all projected benefits for sensitivity analysis. This rate was used for two premises. First, according to the Cambodia Central Bank, a discount rate for general commodities was 5.25% in 2007 ^[19]. Second, for impact evaluation of health issues, a discount rate could range between 3% to 7% for sensitivity tests ^[20]. Thus, we approximated the discount rate of 5.3% to compute social return in our study.

The net present value was converted to its international dollar equivalent using a GDP-based PPP rate of 2.58, according to the trade economics Cambodia PPP rate conversion factor in 2010 ^[19].

The SROI model estimated values generated by the ICP program over the 5-year project period through the steps above. Amounts were converted from Cambodian riels into international units. The return on investment (ROI) was calculated by dividing the total value generated with the total investment.

Uncertainties may occur in models where variables are not known with exact precision, and where the relationship between inputs and outcomes are not straightforward (e.g. human behavior and interaction, wellbeing, support). Therefore, to test the robustness of the SROI model, financial proxies, attribution, and drop-off were used to assess the sensitivity of the economic model.

This study was approved by the National Ethics Committee for Health Research, Ministry of Health, Cambodia (Reference no.169 NECHR). Participants provided written informed consent after a detailed description of the study objectives and procedures were provided. They were also informed that participation in the study was voluntary, and they could refuse or discontinue their participation at any time for any reason. Participants received food and personal items such as soaps equivalent to $1.25, as compensation for their time. Transportation supports were provided to those who needed it. Removing all personal identifiers and using code numbers ensured participant confidentiality.

3. Results

The SROI for the ICP project, after accounting for the total inputs invested into the program, was an additional 96% on top of returns—for every $1 invested, $1.96 was generated in social, health, and economic value. In other words, a combined investment of approximate US$2,406,000 from the EC plus US$1,957,934 (WFP and community inputs) generated total benefits worth US$10,894,835, after adjusting for the discount rate and PPP.

The SROI model shows that the proportion of value created varies considerably according to beneficiary type (Figure 1). An unexpected value generated for caregivers of PLHIV (2%), came about in possible two ways: (1) through the opportunity to return to productive work, as the burden of caring for PLHIV in the household was reduced due to health gains; and (2) through the opportunity to support and take better care of family members, and the opportunity to be supported by the ICP program.

Download as

• PPT
  PowerPoint Slide
• PNG
  Larger image(png format)

Veiw figureFigures index

•  NEW
  View larger figure in new window

Figure 1. Proportion of value created according to beneficiary type

Table 3 shows ICP outcome values created according to beneficiary type. The three highest outcome values for PLHIV, constituting 71% of the total value created, were: i) higher quality caregiver support to PLHIV (37%); ii) improved family wellbeing and feelings of greater economic security resulting from decreased debt and decreased need to sell-off key family assets (18%); and iii) higher levels of self esteem (16%). The two livelihood (or economic productivity) outcomes represented 2.1% and 2.3% of the total value created for PLHIV.

The food aid provided to PLHIV/OVC households generated a huge value for the beneficiaries. The outcome was defined by beneficiaries in the FGDs as “less stress and greater peace of mind in relation to food security.” The indicator ‘number reporting that food support is helping them with their family daily living’ was considered a valid representation of outcome change for families of PLHIV and OVC, and could be directly linked to the provision of food support through the WFP. The value created under this outcome was $10,822,771. However, the value generated in relation to food aid is short term, and drop off would, therefore, be immediate by the nature of this emergency relief.

The highest outcome value for OVC was from ‘greater understanding and ability of caregivers to support their OVC family members,’ and represented an important positive change for OVC within their households. The OVC outcome of ‘greater feeling of positivity’ was also linked to this, and generated a high value. The self-help groups and peer support components of ICP would have had a large effect on these two outcomes. The outcome of ‘greater feeling of positivity’ was expected to be more, however the baseline for this indicator was conservatively estimated at 50% ^{[21, 22, 23]} i.e. half of OVC already felt some level of positivity about their lives at the baseline.

The IGA scheme targeting OVC households only reached a relatively small number of households, so a strong outcome for OVC was not expected.

Wider community and health services outcomes quantified the avoided costs for people reached by the ICP project by enabling them to access HIV voluntary confidential counseling and testing (VCCT) and treatment. The results showed that the value for communities in terms of avoiding late diagnosis and health-related costs were very high, confirming the information gathered during the consultations with project beneficiaries. The outcome value ‘avoidance of HIV transmission from mother to child’ did not produce substantial value, but was important to test whether the SROI approach could have been used to account for these broader outcomes i.e. the cost-benefit of the ICP project’s influence on PMTCT.

Table 3. ICP outcome values created according to beneficiary type

Download as

PowerPoint Slide

Larger image(png format)

• Tables index

Veiw figure

View current table in a new window

View previous table

Halving the financial proxy for the outcome ‘avoided costs resulting from improved family wellbeing as a result of protection of key assets’ decreased the SROI ratio from 1:1.96 to 1:1.90, showing a low sensitivity to this change. Similarly, halving the expected income generated from an increase in earning potential through income generating projects decreased the SROI ratio to 1:1.95.

Halving the financial proxy for ‘equivalent cost in care training for caregivers of PLHIV’ reduced the SROI ratio to 1:1.84 indicating moderate sensitivity to this indicator.

Halving the attribution of the project (from 80% to 40%) for health outcomes through referrals, treatment literacy, and adherence support resulted in a decrease in the SROI ratio to 1:1.92.

For the outcome value created for food security, halving the attribution of the project from 90% to 45% resulted in a decrease in the SROI ratio to 1:1.46, indicating moderate to high sensitivity here. For OVC, halving the attribution portion for the outcome ‘greater understanding and ability of caregivers to support OVC family member’ from 70% to 35% reduced the SROI to 1:1.91.

Highest value-generating outcomes were analyzed in order to test the sensitivity of the SROI model to drop-off. By increasing the annual drop-off for the outcome ‘improved understanding and ability to support PLHIV family members’ to 70% in year 3 and 90% in year 4, the SROI decreased marginally to 1:1.95, indicating that the model is not very sensitive to drop-off for this indicator. Similarly, increasing the drop-off each year for ‘improvement in livelihood status’ to 40%, 60%, and 80% were insufficiently sensitive to changes in the SROI ratio.

4. Discussion

The present SROI study found the ICP program delivered a cost to benefit ratio of 1:1.96. This suggests that investment costs put into a community-based HIV prevention, care, and treatment program in Cambodia can lead to moderate returns generated by the program, even after adjusting for the discount rate, PPP, financial proxies, attribution, and drop off.

The highest outcome value created for PLHIV and OVC was ‘greater understanding and ability of caregivers to support PLHIV and OVC.’ This contributed to increased quality of life for PLHIV/OVC, and illustrates the substantial contribution of caregivers to the ICP program.

Similarly, the ARP 2 ‘training of trainers’ was shown to generate substantial value by increasing the understanding of HIV, resulting in significant behavior change across an entire community to help support PLHIV and OVC long after the ARP 2 project ended ^[17]. Consistent with these results from ICP and ARP 2, some systematic reviews have indicated the potential cost-effectiveness of community health workers in improving health outcomes for populations in low and middle-income countries ^{[24, 25]}.

As indicated by KHANA’s survey on children affected by HIV and AIDS in 2000, 40% of OVC were without sufficient food, clothes, and basic necessities ^[26]. The ICP projects targeting caregivers of OVC for sensitization, family counseling, and food support have resulted in a better quality of life for OVC within their family environment. These results are consistent with India’s CHAHA program where substantial value was created for ‘improved health status of OVC’ (i.e. better health for OVC due to less need for treatment of opportunistic infections and other illnesses, which were significant financial burdens to families) ^[16].

A high outcome value was also noted from ‘avoidance of health costs resulting from late diagnosis.’ Validated secondary research consistently showed the potentially devastating impact of late HIV diagnosis, due to the health-related expenses incurred ^{[27, 28, 29]}. This outcome generated by the ICP project contributes to preventing households affected by HIV from slipping into health-related debt and poverty.

The ‘better livelihood prospects through IGA’ component of the ICP (i.e. build skills and increase incomes for PLHIV/OVC households) did not generate as strong a return as expected. However, this was most likely due to the limited scale of reach, since the component was a late addition to the program and had small initial grants, which resulted in a small budget allocation. An important consideration would be to assess the sustained value of the ICP project in more details by determining what the longer term socioeconomic impact of the IGAs and self-help financing loans has on PLHIV and their families. For example, substantial value was indicated for the livelihoods component in India’s CHAHA program, and the program suggests that consideration be put on the size of initial grants issued for this component since it potentially generates a large benefit return ^[16]. This would allow KHANA to ensure that their livelihood generating activities are continuing to provide value after the project activities end.

FGDs highlighted the issue of continuation of emergency household food support after the end of the WFP project in 2012. The SROI indicated a high outcome value of decreased stress related to food security, and showed a high sensitivity to the food security attribution. This component generated by the ICP project has been supported by several studies linking food insecurity and HIV, where one factor exacerbates the severity of the other ^{[30, 31, 32]}. Food security, on the other hand, has been shown to have a positive impact on PLHIV, especially in low-income countries where households may prioritize food over healthcare needs, leading to poorer health outcomes for PLHIV ^{[33, 34]}. KHANA and implementing partners should gather further evidence on the topic of food security to advocate with donors and the Cambodian government.

Though substantial, a 1:2 SROI is not an unexpected return if one takes into account the level and period of investment. KHANA’s financial commitment to a longer-term (5-year) program strategy has created an enabling environment for working towards the sustainability of the project. Without such levels of commitment, it is unlikely that the project would have generated enough of an impact to sustain social, health, environmental, and economic benefits for an estimated 3-5 years after completion. More importantly, the exercise has provided an answer to the question of “whose value counts?” Whilst by no means a perfect science, it is important to note that all monetary values, or financial proxies used to represent a program outcome should be informed by program beneficiaries. Establishing financial proxies was a major part of the consultation process with beneficiary groups and key informants. Where it was difficult to find a financial proxy through consultation, secondary sources of data were used (i.e. the cost of a bicycle). The methodology of SROI clearly focused on the outcomes identified by beneficiaries, and the value they have received from the project with each step of the process being led by beneficiary-inputs.

The key limitation in this research was time. Time constraints limited discussion between the research team and health service staff; discussions with direct beneficiaries such as PLHIV, OVC, family members, caregivers, and CSVs were given priority. In addition, the ICP project covered three provinces, but the beneficiary consultations for the scope of the SROI were captured in one province due to time constraints—most of the data on outcomes came from Kampong Speu, but additional data were recorded across all three provinces in the end-line survey and used in the SROI study. However, the program staff assumed that there would not be significant variations in results between the provinces, as the ICP program was the same across all provinces and implementing partners.

In the absence of SROI standards and a robust method of auditing an organization’s ‘claims’ to the value it creates, detailed assumptions, processes, and measurement of outcomes were implemented in order to fully document how figures were calculated or estimated. A comprehensive costing of community input is necessary to give a true reflection of how much the community invests through its involvement with the ICP. The true costs of this have not been completely captured in the present SROI study due to lack of reliable secondary information to validate financial proxies. The study was also not able to quantify certain values: government poverty cards, costs incurred by the government, financial burdens to health services of untreated PLHIV. Lastly, attribution percentages were estimated as a range rather than an exact amount, making calculations more complicated and less precise.

5. Conclusions

SROI has been used to evaluate the economic and social impacts of HIV and AIDS programs. The methodology is well suited for evaluating public health programs because it incorporates the social, health, environmental, and economic costs and benefits, making it a more holistic approach compared to traditional frameworks ^[35]. SROI is a useful tool for quantifying values of a program, and should be used as a forecasting and evaluative tool at the mid-point and end of programs to increase ownership of the program amongst beneficiaries and implementing partner organizations.

Whilst a SROI ratio of 1:1.96 is a meaningful and positive result of this study, KHANA and partners should not be complacent about ensuring maximum opportunities for identifying and effecting cost saving measures. Additionally, the implications of this SROI study would be of wider interest, rather than solely to HIV and AIDS program implementers, since several international initiatives that have helped shape, but would also benefit from, the results of this study. For example, the UNAIDS Investment Framework ^[36] emphasizes community mobilization as a critical enabler in the HIV and AIDS response, yet the definitions of what this means in relation to a range of different programmatic interventions such as behavior change, communication, prevention with key populations, and treatment and care, are not clear. The SROI approach would help to quantify the importance of community mobilization in different interventions.

Statement of Competing Interests

The authors have declared that no competing interests exist.

Author’s contributions

Conceived and designed the experiments: TS LT HS. Performed the experiments: TS LT HS. Analyzed the data: TS LT HS. Contributed reagents/materials/analysis tools: TS LT HS ML PC KP TK CN SY. Wrote the paper: TS ML SY. Commented on manuscript and interpretation of results: LT HS PC KP CN SY. Collected data relevant for modeling: TS HS LT. Designed and implemented the model used in the analysis: LT.

Acknowledgments

We gratefully acknowledge KHANA’s Program Team for providing guidance and technical support for this study. We would also like to express our appreciation and gratitude to our implementing partners, research assistants, and community support officers, whose dedicated efforts ensured the quality of the data and timeliness of the study. Lastly, we thank the participants who took their busy time to participate in this study.

References

[1]	UNAIDS, Joint United Nations Program on HIV/AIDS. Global AIDS response progress reporting 2014: construction of core Indicators for monitoring the 2011 United Nations political declaration on HIV and AIDS. Geneva, Switzerland: UNAIDS; 2014.
	In article
[2]	Lozano, R., Naghavi, M., Foreman, K., Lim, S., Shibuya, K., Aboyans, V., et al. “Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010”. Lancet 2012; 380(9859): 2095-128.
	In article		View Article
[3]	Trapero-Bertran, M., Oliva-Morena, J. “Economic impact of HIV/AIDS: a systematic review in five European countries”. Health Economic Review 2014; 4(15).
	In article		View Article
[4]	Hutchinson, A.B., Farnham, P.G., Dean, H.D., Ekwueme, D.U., Rio, C., Kamimoto, L., et al. “The economic burden of HIV in the United States in the era of highly active antiretroviral therapy: evidence of continuing racial and ethnic differences”. Journal of Acquired Immune Deficiency Syndrome 2006; 43(4): 451-7.
	In article		View Article  PubMed
[5]	UNAIDS, Joint United Nations Program on HIV/AIDS. How AIDS changed everything. Geneva, Switzerland: UNAIDS; 2015.
	In article
[6]	The Henry J. Kaiser Family Foundation, UNAIDS. Financing the response to HIV in low- and middle-income countries: International assistance from donor governments in 2012. Menlo Park, California: The Henry J. Kaiser Family Foundation; 2013.
	In article
[7]	Rich, M.L., Miller, A.C., Niyigena, P., Franke, M.F., Niyonzima, J., Socci, A., et al. “Excellent clinical outcomes and high retention in care among adults in a community-based HIV treatment program in rural Rwanda”. Journal of Acquired Immune Deficiency Syndromes 2012; 59(3): 35-42.
	In article		View Article  PubMed
[8]	Koenig, S., Ivers, L.C., Destine, R., Leandre, F., Grandpierre, R., Mukherjess, J., et al. “Successes and challenges of HIV treatment programs in Haiti: aftermath of the earthquake”. HIV Therapy 2010; 4(2): 145-160.
	In article		View Article  PubMed
[9]	National Center for HIV/AIDS, Dermatology, and STDs (NCHADS). Estimation of the HIV Prevalence among General Population in Cambodia, 2014. Phnom Penh, Cambodia: NCHADS; 2015.
	In article
[10]	National Center for HIV/AIDS, Dermatology, and STDs (NCHADS). Annual report 2014. Phnom Penh, Cambodia: NCHADS; 2015.
	In article
[11]	UNAIDS, Joint United Nations Program on HIV/AIDS. Cambodia takes MDG prize for excellence in its AIDS response. Geneva, Switzerland: UNAIDS; 2010.
	In article
[12]	KHANA. KHANA Network Household Economic Livelihood Survey Analysis. Phnom Penh, Cambodia: KHANA; 2010.
	In article
[13]	UNDP. The Socio-Economic Impact of HIV/AIDS Epidemic at the Household Level in Cambodia. Phnom Penh, Cambodia: UNDP, Sanigest International, Center for Advance Studies; 2010.
	In article
[14]	New Economics Foundation (NEF). Social Return on Investment: Valuing What Matters, London: New Economics Foundation; 2004.
	In article
[15]	Cabinet Office. A guide to Social Return on Investment. London, UK: Cabinet Office, Office of the Third Sector; 2009.
	In article
[16]	Biswas, K., Kummarikunta, G., Biswas, A., Tong, L. Social return on investment: CHAHA programme. Hove, UK: International HIV/AIDS Alliance; 2010.
	In article
[17]	Brady, R. The true cost of stigma: evaluating the social return on investment of the stigma and discrimination component of the Alliance’s Africa Regional Programme II. Hove, UK: International HIV/AIDS Alliance; 2011.
	In article
[18]	Sopheab, H., Chhea, C., Tuot, S. Midterm review of the integrated care and prevention project regarding PLHIV and OVC. Phnom Penh, Cambodia: KHANA; 2010.
	In article
[19]	World Bank. Trading economics, Cambodia GDP per capita PPP US dollar, 2010. World Bank, 2010.
	In article
[20]	Harrison, M. Valuing the Future: the social discount rate in cost-benefit analysis, Visiting Researcher Paper, Productivity Commission, Canberra; 2010.
	In article
[21]	Heng, S., Chhea, C., Tuot, S. Mid-term review of the Integrated Care and Prevention Project (ICP) regarding people living with HIV (PLHIV) and orphans and vulnerable children (OVC). Phnom Penh, Cambodia: KHANA; 2010.
	In article
[22]	Heng, S., Tuot, S. End of project evaluation: Change in HIV Integrated Prevention, Care and Impact Mitigation Efforts from 2009-2011. Phnom Penh, Cambodia: KHANA; 2012.
	In article
[23]	The National Multi-sectoral Orphans and Vulnerable Children Task Force. Orphans, children affected by HIV and other vulnerable children in Cambodia: A situation and response assessment in 2007. Phnom Penh, Cambodia: National Multi-sector OVC Task Force; 2008.
	In article
[24]	Vaughan, K., Kok, M.C., Witter, S., Dieleman, M. Costs and cost-effectiveness of community health workers: evidence from a literature review. Human Resources for Health 2015; 13(71).
	In article		View Article
[25]	Viswanathan, M., Kraschnewski, J., Nishikawa, B., Morgan, L.C., Thieda, P., Honeycutt, A., et al. “Outcomes of community health worker interventions”. Evidence Report/Technology Assessment 2009; 181
	In article
[26]	KHANA. Children affected by HIV/AIDS: Appraisal of needs and resources in Cambodia. Summary report. Phnom Penh, Cambodia: KHANA; 2000.
	In article
[27]	Zhang, X., Li, M., Feng, Y., Bu, K., Fu, G., Liu, X., et al. “Study on economic burden caused by antiretroviral treatment for people living with HIV/ADIS and influencing factors in Nanjing”. Zhonghua Liu Xing Bing Xue Za Zhi 2015; 36(5): 440-444.
	In article		PubMed
[28]	Barennes, H., Frichittavong, A., Gripenberg, M., Koffi, P. “Evidence of high out of pocket spending for HIV care leading to catastrophic expenditure for affected patients in Lao People’s Democratic Republic”. PLoS One 2015; 10(9).
	In article		View Article
[29]	Perelman, J., Alves, J., Miranda, A.C., Mateus, C., Mansinho, K., Antunes, F., et al. “Direct treatment costs of HIV/AIDS in Portugal”. Review. Revista de Saúde Pública. 2013; 47(5).
	In article		View Article  PubMed
[30]	World Health Organization. Nutrition and HIV/AIDS. Geneva, Switzerland: WHO; 2005.
	In article
[31]	Anema, A., Vogenthaler, N., Frongillo, E.A, Kadiyala, S., Weiser, S.D. “Food insecurity and HIV/AIDS: current knowledge, gaps, and research priorities”. Current HIV/AIDS Reports 2009; 6(4): 224-231.
	In article		View Article  PubMed
[32]	Ivers, L.C., Cullen, K.A., Freedberg, K.A., Block, S., Coates, J., Webb, P. “HIV/AIDS, undernutrition and food insecurity”. Clinical Infectious Disease 2009; 49(7): 1096-102.
	In article		View Article  PubMed
[33]	Weiser, S.D., Tsai, A.C., Gupta, R., Frongillo, E.A., Kawuma, A., Senkungu, J., et al. “Food insecurity is associated with morbidity and patterns of healthcare utilization among HIV-infected individuals in a resource-poor setting”. AIDS 2012; 26(1): 67-75.
	In article		View Article  PubMed
[34]	Weiser, S.D., Fernandes, K.A., Brandson, E.K., Lima, V.D., Anema, A., Bangsberg, D.R., et al. “The association between food insecurity and mortality among HIV-infected individuals on HAART”. Journal of Acquired Immune Deficiency Syndrome 2009; 52(3): 342-349.
	In article		View Article  PubMed
[35]	Banke-Thomas, A.O., Madaj, B., Charles, A., van den Broek, N. “Social return on investment (SROI) methodology to account for value for money of public health interventions: a systematic review”. BMC Public Health 2015; 12(582).
	In article		View Article
[36]	Schwartlander, B., Stover, J., Hallett, T., Atun, R., Avila, C., Gouws, E., et al. “Towards an improved investment approach for an effective response to HIV/AIDS”. Lancet 2011; 377(9782): 2031-41.
	In article		View Article