Benefit-cost analysis of nonprofit cataract surgery services in Latin America: a multinational evaluation through an SROI lens

Abstract

Cataract surgery is a transformative procedure, rapidly restoring vision and improving quality of life with minimal complications. This study evaluates the social return on investment (SROI) of bilateral cataract surgeries performed by different nonprofit organizations across Latin America. By analyzing the experiences of patients and caregivers –including increased autonomy, confidence, and reduced caregiving burdens – this research provides an estimate of the associated social and economic benefits of these interventions. The average SROI ratios under the baseline scenario are 12:1 for patients, 11:1 for caregivers, and 9:1 for patient-caregiver pairs. Sensitivity analysis reinforces these results, with SROI ratios ranging from 3:1 to 18:1 under various assumptions about outcome duration, discount rates, and valuation of time, among other scenarios. Reduced caregiving responsibilities may create conditions for young female caregivers to pursue economic activities, potentially enhancing labor force participation. These findings suggest that nonprofits may play an important part in addressing healthcare gaps and complementing public services, particularly in underserved regions. Furthermore, by helping to address healthcare access disparities and reduce economic burdens, nonprofit cataract surgeries may represent a promising approach to improving lives across Latin America. While the study focuses on seven nonprofit organizations and a specific intervention, its results demonstrate the potential for broader applications in nonprofit healthcare delivery. Lastly, the research highlights the importance of understanding and valuing both direct and indirect benefits of healthcare interventions.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12962-026-00723-2.

Introduction

Cataracts are a leading cause of preventable blindness, particularly in low- and middle-income countries, and they impose a significant global economic burden. Annual productivity losses far exceed the estimated resources needed to address these unmet needs [1]. In Latin America, cataracts remain a major cause of blindness, with suboptimal surgical coverage and outcomes below target levels [2]. Cataract surgery is a proven and reliable treatment for restoring vision; however, it frequently entails substantial non-clinical costs for patients and caregivers, including travel, time devoted to care, and forgone productive activities, that are often overlooked in conventional evaluations. Effective policy design and resource allocation require consideration of the broader economic effects, including the hidden costs borne by caregivers and the loss of productivity for both stakeholders. Patient selection for surgery typically involves evaluating visual acuity, visual function, and the impact on quality of life [3]. Research shows that cataract surgery improves both [4].

Scarcity of resources in Latin American public healthcare systems often poses substantial challenges. Research conducted for a variety of medical interventions in Mexico, for instance, estimated average waiting times of 14 weeks for surgeries, with an additional 11 weeks for diagnostic procedures [5]. Delays of this nature can significantly affect patient well-being, highlighting the urgent need for increased human and physical resources to tackle the issue. Nonprofit organizations can play a pivotal role in this matter [6, 7], whether by addressing quality or quantity gaps in the provision of medical services [8, 9], expanding service availability [10, 11], promoting training and research [12], or by offering alternative solutions to market failures [13, 14]. Nonprofits often compensate for the shortage of services due to the free rider problem in public healthcare –a situation where individuals benefit from services without contributing to their costs, leading to underfunding and inefficiencies – and they help to mitigate information asymmetry by offering transparent, high-quality care.

In this study we extend the work in Aleman-Castilla et al. [15] and employ a Social Return on Investment (SROI) perspective to evaluate cataract surgery services at several nonprofit healthcare institutions in Latin America. SROI is a widely adopted approach designed to measure and demonstrate the value of diverse non-financial outcomes generated by activities or interventions [16]. This framework helps quantify the value created in relation to the investments made. The former is typically categorized into social benefits (i.e., improvements in the quality of life for direct and indirect beneficiaries), environmental benefits (e.g., those arising from improved infrastructure or natural resource efficiency), and economic benefits (which encompass economic gains, job creation, and reduced costs in the production or provision of goods and services). Organizations often use SROI to better understand and communicate the broader effects of their work, extending beyond conventional financial metrics.

Although the SROI methodology has been widely used across various topics and contexts (e.g. [17–20]), Aleman-Castilla et al. [15] were the first to apply it to assess cataract surgery services offered by a nonprofit healthcare organization, and to our knowledge, this is in turn the first study to provide a cross-country assessment for Latin America. Our analysis draws on data from over 1600 interviews with patients and their primary caregivers across five countries, all of whom underwent bilateral cataract surgery in 2023. The institutions that participated in this study are APEC Hospital de la Ceguera, in Mexico City, Mexico (APEC); Fundación Visión, in Departamento Central, Paraguay (FV); Instituto de la Visión at La Carlota Hospital of Montemorelos University, in Nuevo León, Mexico (MM); Visualiza Hospital Oftalmológico, in Guatemala, Guatemala (VZ); Instituto Mexicano de Oftalmología, in Querétaro, Mexico (IMO); Hospital Cristiano de Especialidades Fundamisc, in San Francisco de Milagro, Ecuador (HCE); and Instituto Nacional de Oftalmología, in Lima, Peru (INO).

Similar to Aleman-Castilla et al. [15], we observe that patients who invested time and money into the surgical procedure experienced a return in terms of regaining autonomy and confidence to perform daily instrumental tasks. Likewise, their primary caregivers benefited from the patients’ increased independence, leading to reduced stress and improved well-being. This dual benefit highlights how these interventions can enhance the quality of life for both groups of stakeholders across Latin America. These results confirm the collaborative potential between the public sector and nonprofit healthcare organizations in delivering essential public goods and services cost-effectively to vulnerable populations.

Theory of change

Beyond improving vision-related quality of life, cataract surgery has been shown to generate broader social and economic effects, including changes in labor force participation and productive activities at the household level [21–23]. These effects, however, unfold within a context shaped by multiple factors that may affect patients and caregivers differently, such as the costs of surgery, medical examinations and consultations, the severity and progression of cataracts, the time required for treatment, and additional expenses related to transportation, food, or lodging. Considering this, Fig. 1 presents a stylized theory of change that serves as a conceptual framework to organize the main pathways through which bilateral cataract surgery may generate outcomes for patients and caregivers. It is intended to guide the identification of relevant inputs, and the valuation choices within the SROI framework.

Fig. 1.

Theory of change from nonprofit cataract surgery

As illustrated, patients –often accompanied by a family member or primary caregiver – seek visual health services and invest time and financial resources to access care. Visual health services are provided through clinical evaluations and diagnostic tests, leading to cataract surgery, which restores clarity to the optical system and improves vision. This intervention may translate into a range of intermediate outcomes: Patients may regain autonomy, mobility, and control over their daily activities, while also improving emotional well-being, reducing limitations associated with visual impairment, and facing lower risks of accidents. Caregivers, in turn, may also benefit from reduced stress and greater independence, as the need for constant assistance diminishes. The long-term effects of this process may be transformative. Patients, now with restored vision, could re-engage in productive activities, whether through paid employment or non-market contributions such as household production and caregiving for others. Additionally, with fewer accidents and reduced stress levels for both patients and caregivers, their improved vision may lead to an overall increase in lifespan and quality of life.

Overall, this theory of change provides a simplified representation of the potential pathways through which cataract surgery can generate social and economic value. While the diagram in Figure 1 refers to the predominant populations served by participating nonprofit organizations, it is not meant to exclude the possibility that individuals from other socioeconomic backgrounds may also benefit from such interventions. Rather, it highlights the main mechanisms through which vision-restoring procedures may contribute to improved well-being and economic participation across diverse settings.

Data and methods

Data

The data for this study were collected through interviews with patients and caregivers who underwent bilateral cataract surgery between January and December 2023 across the participating institutions. All interviews were conducted during the first quarter of 2024, following completion of surgery and routine post-operative follow-up schedules. We gathered a total of 1667 observations: 380 from APEC, 78 from MM, 237 from IMO, 311 from FV, 388 from VZ, 140 from HCE, and 205 from INO. Patients represent 55% of the sample (910 observations), and caregivers represent the other 45% (757 observations). Complete patient-caregiver pairs account for 73% of the sample (1,214 observations), and incomplete pairs represent the other 27% (453 observations). The sample size was designed to be statistically representative of the total number of bilateral cataract surgeries performed at each participating institution in 2023, based on standard sample size calculations for finite populations ⁱ. This was achieved for all cases, except for MM.

Our inclusion criteria considered patients aged at least 40 years, accompanied by their primary caregivers during the surgery and the recovery period. Our data collection process faced different challenges nonetheless, such as limited resources for conducting surveys, which lead us to rely on telephone-based interviews; difficulties in recontacting patients or caregivers, many of whom reside in remote areas; and some refusals to participate in the study, due to personal security concerns. All respondents consented to the interviews, and we ensured their confidentiality through data anonymization. As highlighted by Aleman-Castilla et al. [15], there are several advantages of focusing on bilateral cataract surgery cases. This approach ensures consistency across patients, as they all undergo the same procedure, and often leads to more significant improvements in vision and quality of life. It also reduces confounding variables that could arise when comparing patients who had surgery on one eye versus both, aligning with the common practice of performing bilateral surgeries in advanced cataract cases.

Inputs, outcomes and duration of benefits

The typical cataract surgery process begins with the patient’s initial consultation, in which ophthalmologists assess visual acuity, refraction, and pupil dilation. Subsequent consultations are scheduled to confirm the diagnosis and determine if the patient is a surgical candidate. Pre-surgery studies and clinical analyses are also conducted. After the first-eye surgery, there are up to three follow-up appointments. The same process is repeated for the second-eye surgery, with some analyses being optional if performed within 1 month of the first-eye surgery. Both patients and caregivers contribute time and money to the intervention.

• Time: Based on administrative records and operational experience reported by the participating institutions, patients make, on average, 11 visits to medical facilities and dedicate an additional 14 days to post-surgery recovery (7 days for each eye), leading to a total of 25 days devoted to the process ⁱⁱ. Caregivers are required to accompany patients at all times, so they dedicate the same number of days to the process. To value the stakeholders’ time, we used the 2023 average minimum wage set by each country’s corresponding authority, expressed in international dollars (Int $) using the corresponding purchasing power parity (PPP) conversion factors of the World Development Indicators database [24]. This is a proxy for the opportunity cost of time, representing the baseline income forgone by stakeholders while engaged in the intervention.

• Pecuniary costs: In addition to the cost of consultations, medical examinations, and surgical procedures, stakeholders may incur in food, lodging and transportation expenses as they undergo cataract surgery. To estimate the first two items, they were asked how much they had spent on food and accommodation during their medical visits. For transportation costs, they were asked their place of residence and the means of transport used, which then made it possible to estimate a Generalized Travel Cost (GTC) ⁱⁱⁱ. When both patients and caregivers reported having contributed to these expenses, the pecuniary costs were divided equally between them to reflect a shared financial responsibility. These costs were also restated in international dollars.

To identify the outcomes experienced by patients and caregivers after cataract surgery, we used open-ended and Likert-scale questions to gather their stories of change. Like the findings in Aleman-Castilla et al. [15], three main outcomes emerged: Patients regained assurance and confidence from improved vision and motor abilities, caregivers experienced reduced stress levels regarding their patients, and both patients and caregivers regained freedom to engage in personal activities. Specifically:

• Autonomy. Patients gained freedom and independence to perform instrumental activities of daily living (e.g., household chores, reading, embroidery, etc.), while caregivers were no longer required to provide assistance for these tasks. This outcome was validated using five and six Likert-scale questions, respectively, that assessed the patient’s dependency on assistance for various home or daily activities before the surgery and the level of improvement experienced afterward.

• Assurance. Patients experienced an increase in confidence to perform basic activities of daily living such as walking, climbing stairs, and avoiding accidents due to improved vision. This outcome was validated using eight Likert-scale questions that assessed the mobility difficulties and discrimination experienced by the patient before the surgery, and the level of improvement felt after it.

• Alleviation. Caregivers experienced reduced stress related to their patients, attributed to their increased independence and optimism following surgery. This outcome was assessed using four Likert-scale questions that measured the caregiver’s level of concern and reluctance to leave the patient alone prior to the surgery, as well as the relief felt after the medical procedure.

In our baseline scenario, outcome relevance was weighted by the share of associated Likert-scale questions for which stakeholders reported “agree” or “totally agree”. For example, autonomy was assigned an 80% salience if patients agreed or totally agreed with 4 of 5 related items, and alleviation a 75% salience if caregivers agreed or totally agreed with 3 of 4 items. This weighting reflects relative outcome salience only and does not affect the monetary value assigned to outcome proxies. The questions and response distribution are reported in the Appendix.

We then proceed to value these outcomes. For autonomy experienced by patients, we estimate its value as the lowest income they could earn if they worked a maximum of 48 hours per week at the prevailing general minimum wage rate, adjusted by the probability of participating in the labor force, estimated by the 2022 labor force participation rate by age, sex and country [25]. For caregivers, the value of their autonomy is also approximated by the lowest income they might earn if they were compensated as domestic personal care workers, also adjusted by their labor force participation rate. Based on INEGI [26, 27] we assume that caregivers engage in this type of work 3.6 days a week.

For assurance and alleviation, we monetize these outcomes using country-specific out-of-pocket health expenditure per capita for 2021 [28], defined as direct payments made by individuals to health care providers at the time-of-service use. These expenditures changes serve as conservative avoided-cost proxies, rather than as direct measures of these outcomes. For patients, assurance assumes that in the absence of surgery, health-related expenses would likely remain stable or increase due to worsening visual impairment and comorbidities [29], so the opportunity cost reflects potential savings from surgery. For caregivers, alleviation captures reductions in health expenditure associated with stress-related health effects of caregiving. For the baseline scenario, and in the absence of better data, we conservatively assume an annual reduction equivalent to one month of out-of-pocket health expenditure, applied only when respondents indicated that surgery lowered their health-related spending.

Lastly, we established the duration of these outcomes based on life expectancy. For patients, their remaining years of life were estimated by subtracting their age from the World Health Organization’s life expectancy by age group and country [30]. For caregivers, their remaining years of life were compared to that of their patients. If caregivers were expected to live longer, their duration of outcomes was set equal to that of their patients. Alternatively, if their estimated remaining years of life were lower, their duration of outcomes was set equal to their own remaining life expectancy.

Deadweight and drop-off

Stakeholders were asked about their ability to seek cataract surgery from alternative healthcare providers; 25% of patients and 31% of caregivers indicated that this could have been possible. However, given significant waiting times, the probability of achieving comparable vision improvement was set at 90% compared to their respective nonprofit institutions [31, 32]. Additionally, economic benefits from cataract surgery may diminish gradually over time due to aging, comorbidities, changes in labor market participation, or mortality. Thus, future outcome flows were adjusted by a constant annual percentage reduction equal to the 2021 (most recent year available) crude death rates per 1000 inhabitants by country [33], which are equal to 0.9% for Mexico, 0.8% for Paraguay and Peru, and 0.7% for Guatemala and Ecuador.

SROI ratios

For stakeholder in country , the total benefit from cataract surgery is the present value of outcomes, computed as a decreasing annuity [34]:

1

were is the total value created from cataract surgery at year 0 when the intervention takes place, is the negative of the drop-off rate, is duration of outcomes, and is the discount rate, set equal to a conventional social discount rate of 10% typically used in the evaluation of investment programs and projects in many Latin American countries [35–37]. The individual SROI ratio is then calculated as:

2

where is the value of inputs incurred by stakeholder in country . Alternatively, the SROI ratio can be calculated for each matched patient-caregiver pair :

3

Results

Sample characteristics

Figure 2 shows the distribution of cases by place of residence of the patients (i.e., Mexican states for APEC, MM, and IMO; Paraguayan departments for FV; Guatemalan departments for VZ; Ecuadorian provinces for HCE; and Peruvian departments for INO); and Fig. 3 summarizes the individual characteristics of interviewees: Women represent 64% of patients and 68% of caregivers (panel a); their mean ages (panel b) are 69 for patients (range 43–96) and 50 for caregivers (range 16- 90); 75% of caregivers are either children or spouses to their patients (panel c); and most of the patients perceive themselves as economically active (panel d), as they contribute to their household income.

Fig. 2.

Places of residence of patients

Fig. 3.

Demographic characteristics of stakeholders

Sentiment analysis

An important innovation in this study consists of the analysis of the feelings of patients and caregivers, based on the answers to the open-ended question “What changed in your life after cataract surgery?” Figure 4 presents the most salient results. Panel (a) contains a word cloud of the most mentioned substantive words d, such as autonomy, good, activities, security, better, and improved, most of them with a positive connotation. Panel (b) presents a word classification based on the Sentiments and Emotions Lexicon developed by the National Research Council of Canada, which categorizes words into eight specific emotions, and two sentiment polarities [38, 39] ^v. The analysis uses the Spanish-language version of this lexicon and is applied to responses originally collected in Spanish. This approach is widely used in sentiment analysis and natural language processing to assess emotional tone in text. Given its lexicon-based nature, results are interpreted descriptively and at an aggregate level and are intended as a complementary input to the analysis. In this case, positive substantive words outnumber negative ones 2 to 1, and emotions such as joy, fear, trust, and anticipation are more frequent than anger, disgust, or surprise. Importantly, negative words identified in the sentiment classification do not necessarily reflect negative perceptions of the post-operative outcome. Rather, they are predominantly used by respondents to describe their pre-operative condition and life circumstances, within a narrative contrasting life before and after cataract surgery.

Fig. 4.

Sentiment analysis: “what changed in your life after cataract surgery?”. Notes: sentiment analysis based on words extracted from narratives contrasting life before and after cataract surgery in response to the question “what changed in your life after cataract surgery?”. Negative sentiment words mainly describe pre-operative conditions, while positive words capture perceived post-operative improvements. Counts refer to words rather than individuals

Inputs, outcomes and duration of benefits

Table 1 provides summary statistics of the value of inputs by stakeholders, divided into four categories: Treatment, GTC, Food & Lodging, and Time. The first panel shows that the mean treatment cost for patients is higher than for caregivers (Int$ 1049.0 and Int$ 591.3 respectively), with important variability as indicated by large standard deviations and maximum values reaching up to Int$ 10,309.3 for patients and Int$ 8247.4 for caregivers.

Table 1.

Summary statistics of value of inputs by the stakeholders

	Patients (885 complete observations)					Caregivers (739 complete observations)
	Mean	Std. Dev.	Minimum	Median	Maximum	Mean	Std. Dev.	Minimum	Median	Maximum
Treatmenta	1,049.8	1,676.8	0.0	0.0	10,309.3	591.3	1,211.1	0.0	0.0	8,247.4
APEC	2,386.3	2,400.9	0.0	2,935.1	10,309.3	1,358.4	1,972.1	0.0	0.0	8,247.4
MM	1,544.0	1,500.7	0.0	1,649.5	5,360.8	1,092.1	1,318.5	0.0	386.6	4,123.7
IMO	1,279.7	1,809.4	0.0	0.0	6,597.9	479.9	1,184.0	0.0	0.0	8,041.2
FV	975.0	1,279.1	0.0	0.0	4,761.4	432.7	887.4	0.0	0.0	4,166.2
VZ	217.9	374.3	0.0	0.0	2,018.2	346.7	417.8	0.0	0.0	2,118.2
HCE	983.8	1,459.4	0.0	0.0	7,300.0	593.4	1,046.9	0.0	0.0	3,700.0
INO	354.9	928.7	0.0	0.0	4,277.8	189.1	703.2	0.0	0.0	5,400.0
Generalized Travel Cost (GTC)b	884.1	1,570.0	8.6	239.5	9,643.9	569.0	1,169.9	6.1	137.6	9,643.9
APEC	623.6	1,210.3	8.6	223.0	6,338.2	324.4	916.4	8.6	86.0	9,179.6
MM	2,469.5	2,655.2	18.0	1,336.5	7,800.0	1,061.1	1,543.4	6.1	386.2	5,469.5
IMO	531.8	977.7	14.7	113.9	5,235.1	315.6	679.2	14.7	86.0	4,914.1
FV	1,258.1	1,869.1	33.5	405.2	7,131.5	743.1	1,447.1	33.5	307.7	6,422.6
VZ	670.4	1,440.8	35.7	137.6	9,643.9	836.9	1,499.8	35.7	137.6	9,643.9
HCE	1,196.3	1,241.4	198.0	247.5	4,813.9	751.4	951.9	198.0	233.8	4,813.9
INO	638.8	1,392.8	23.5	286.5	7,363.4	237.6	457.4	23.5	63.5	3,924.8
Food & Lodgingc	82.9	301.3	0.0	0.0	3,855.7	43.2	161.6	0.0	0.0	2,097.9
APEC	101.4	342.0	0.0	0.0	2,494.8	21.7	86.5	0.0	0.0	567.0
MM	263.1	652.6	0.0	0.0	3,855.7	207.9	477.5	0.0	0.0	2,097.9
IMO	104.4	390.0	0.0	0.0	3,492.8	22.0	69.5	0.0	0.0	464.9
FV	52.4	188.5	0.0	0.0	1,841.3	9.1	37.1	0.0	0.0	204.6
VZ	83.3	267.0	0.0	0.0	2,594.9	101.7	239.9	0.0	0.0	1,410.3
HCE	49.5	101.3	0.0	0.0	550.0	31.9	92.5	0.0	0.0	500.0
INO	20.0	66.2	0.0	0.0	388.9	10.4	64.5	0.0	0.0	555.6
Timed	689.3	97.2	474.5	694.8	830.9	685.3	103.1	474.5	694.8	830.9
APEC	669.9	0.0	669.9	669.9	669.9	669.9	0.0	669.9	669.9	669.9
MM	669.9	0.0	669.9	669.9	669.9	669.9	0.0	669.9	669.9	669.9
IMO	669.9	0.0	669.9	669.9	669.9	669.9	0.0	669.9	669.9	669.9
FV	830.9	0.0	830.9	830.9	830.9	830.9	0.0	830.9	830.9	830.9
VZ	694.8	0.0	694.8	694.8	694.8	694.8	0.0	694.8	694.8	694.8
HCE	750.0	0.0	750.0	750.0	750.0	750.0	0.0	750.0	750.0	750.0
INO	474.5	0.0	474.5	474.5	474.5	474.5	0.0	474.5	474.5	474.5
Total costse	2,706.1	2,712.1	498.0	1,080.7	12,150.4	1,888.8	1,973.8	498.0	947.5	12,120.8
APEC	3,781.2	3,160.2	678.5	4,085.5	12,150.4	2,374.4	2,349.6	678.5	803.8	10,167.1
MM	4,946.5	3,684.5	722.7	4,188.1	11,183.3	3,031.0	2,769.1	676.1	1,771.7	9,400.2
IMO	2,585.8	2,440.7	684.6	964.7	9,959.6	1,487.4	1,734.7	684.6	755.9	10,682.4
FV	3,116.3	2,762.0	864.4	1,440.2	11,979.8	2,015.8	1,997.5	864.4	1,210.2	10,062.4
VZ	1,666.5	1,762.4	730.5	908.9	11,176.7	1,980.2	1,857.3	730.5	1,071.9	12,120.8
HCE	2,979.6	2,529.2	948.0	997.5	10,622.5	2,126.7	1,984.4	948.0	983.8	7,245.0
INO	1,488.2	1,835.1	498.0	864.3	11,032.5	911.6	885.5	498.0	538.0	6,384.1

Notes: All figures are expressed in international dollars. The statistics presented in each panel exclude outlier observations located 3 interquartile ranges below the first quartile and above the third quartile of the distribution for total costs.^a Includes consultations, medical examinations, and bilateral cataract surgery expenses. ^b Includes the value of time spent traveling and vehicle operating costs for 11 visits to the institution. ^c Ancillary expenses as reported by stakeholders. ^d Includes the value of time devoted to the medical procedure and recovery.^e Considers treatment expenses, GCT, food & lodging, and the value of time devoted to the medical procedure and recovery

The GTC panel, covering travel costs to medical facilities, averages Int$ 884.1 for patients and Int$ 569.0 for caregivers, with a maximum of Int$ 9643.9 in both cases, indicating the relevant burden of travel expenses. GTC mean values are almost as large as treatment costs, which constitute less than half of the total costs. Excluding stakeholder costs from economic evaluations, often conducted solely from the provider perspective, can thus lead to distorted results.

The Food & Lodging costs suggests that patients generally bear a higher share of these expenses, as reflected in the higher mean and maximum values for this category. The Time panel, representing the opportunity cost of time spent on treatment-related activities, reveals that both patients and caregivers allocate considerable time, with means of Int$ 689.3 and Int$ 685.7, respectively.

Lastly, the Total Costs panel indicates that the average cost for patients is Int$ 2706.1, which is 43.3% higher than the Int$ 1888.8 average for caregivers. The maximum values for both groups are nevertheless high, exceeding Int$ 12,000 and suggesting that the overall economic burden of cataract treatment can be substantial.

Table 1 also reveals considerable variability in costs across institutions. APEC and MM exhibit the highest mean treatment costs for both patients and caregivers, while VZ and INO show the lowest, suggesting a more affordable experience. Travel costs also vary significantly, with MM showing the highest GTC mean (Int$ 2469.5) for patients, possibly due to longer travel distances, whereas IMO and APEC have the lowest, likely due to a more localized patient base. Food & Lodging costs are relatively low across institutions, although MM reports higher averages, hinting at more extended stays or higher local accommodation costs. Time costs are higher for FV and HCE, reflecting the greater opportunity cost of time spent in treatment, as measured by the PPP-adjusted minimum wage. Overall, total costs are highest for MM and APEC, indicating a heavier burden on patients and caregivers at these institutions. Conversely, INO presents the lowest total costs, mainly benefiting from lower travel demands, lower time costs, and institutional or governmental support to cover for treatment expenses. The substantial variation in costs across institutions reflects underlying differences in country-level price structures, service delivery models, geographic accessibility, and patient travel patterns. All institutions reported costs using a harmonized definition of inputs, but the observed dispersion highlights meaningful institutional and regional heterogeneity. For this reason, pooled statistics are presented to characterize the overall distribution of stakeholder costs, while institution-specific results are reported separately to preserve contextual differences.

Figure 5 presents a summary of the total cost of inputs, the total value of outcomes in the first year and the duration of benefits for patients and caregivers, associated with bilateral cataract surgery across the different healthcare providers. Panel (a) contains box plots that illustrate the distributions of input costs. The mean cost is Int$ 2331 with a standard deviation of Int$ 2435. Panel (b) contains box plots showing the distributions of value of outcomes. The mean value of outcomes is Int$ 1765 with a standard deviation of Int$ 1578. The histograms in panels (c) and (d) display the distributions of the duration of benefits. The median duration stands at 16 years for patients and 19 years for caregivers.

Fig. 5.

Cost of inputs, value of outcomes, and duration of benefits

SROI ratios

Figure 6 presents the distribution of individual SROI ratios for patients in panel (a) and caregivers in panel (b). On average, for every international dollar invested, patients and caregivers get 12 (or 12:1) and 11:1, respectively, in the form of increased autonomy, assurance, and alleviation. It is important to mention that among patients, 59% of SROI ratios above the mean correspond to women aged 45–88 years. Likewise, among caregivers, 65% of SROI ratios greater than the mean corresponds to women aged 20–77 years, who mostly are daughters or spouses to their patients. Although not reported in the figure, the average individual SROI ratio among both patients and caregivers is equal to 12:1, with a standard deviation of 14.

Fig. 6.

Distribution of SROI ratios by group of stakeholders

Panel (c) of Figure 6 illustrates the distribution of SROI ratios for patient-caregiver pairs for the cases in which both stakeholders provided full input and outcome information. On average, each stakeholder pair generates a return of 9:1 international dollars.

Lastly, Table 2 presents the summary statistics for the distribution of the estimated SROI ratios across institutions and for each stakeholder group. INO shows the highest mean SROI ratios for patients (26:1), caregivers (23:1) and patient-caregiver pairs (22:1), suggesting potentially higher social returns on investment at this location. VZ displays the highest maximum SROI value for patients (94:1), while APEC registers the highest maximum value for caregivers (58:1). Institutions like IMO HCE, and MM show more moderate SROI ratios, with lower mean and maximum values. Overall, this table illustrates substantial differences in outcomes across institutions and stakeholder groups, highlighting the importance of regional context in evaluating intervention effects.

Table 2.

SROI ratios for patients, caregivers and patient-caregiver pairs, by institution

	Patients					Caregivers					Patient-Caregiver pairs
	Mean	Std Dev	Min	Median	Max	Mean	Std Dev	Min	Median	Max	Mean	Std Dev	Min	Median	Max
Total	12.3	14.8	0.0	6.9	94.0	11.4	12.1	0.0	6.4	57.7	9.5	10.0	0.1	5.7	62.5
APEC	6.6	11.1	0.0	2.9	76.7	8.9	12.1	0.0	3.1	57.7	4.7	5.2	0.2	3.0	26.2
MM	7.9	11.7	0.0	2.9	48.4	8.7	9.4	0.3	5.3	38.3	5.1	4.9	0.2	3.4	20.0
IMO	5.9	9.0	0.0	1.6	41.7	9.1	12.2	0.1	2.5	54.7	6.7	7.6	0.1	2.5	28.2
FV	11.7	15.8	0.1	4.1	62.0	7.0	9.3	0.0	2.1	41.7	7.4	8.8	0.2	4.0	41.3
VZ	15.0	12.6	0.4	12.0	94.0	10.6	9.1	0.0	7.9	47.7	12.0	9.2	0.7	9.5	52.4
HCE	12.0	11.0	0.0	9.3	47.8	12.5	10.8	0.0	10.9	41.2	11.3	9.8	0.1	7.3	40.9
INO	25.5	20.3	0.0	18.9	87.7	23.2	13.6	0.0	21.7	54.2	22.1	14.1	1.0	20.1	62.5

Notes: Social Return on Investment (SROI) ratios are defined as the ratio of the present value of outcomes to the value of inputs. For example, a value of 12.3 indicates that, on average, each international dollar invested generates 12.3 international dollars in social value

Sensitivity analysis

To assess the robustness of these estimates with respect to key parameters and assumptions, and for comparability purposes, we consider alternative scenarios similar to those presented in Aleman-Castilla et al. [15]. Table 3 reports the average SROI ratios for the different groups of stakeholders (patients, caregivers, pooled patients and caregivers, and patient-caregiver pairs) under the baseline and alternative sensitivity scenarios, computed in the same manner as in the previous section but recalculated for each case:

• Outcome experience thresholds. The baseline scenario weights the stakeholders’ experience of a particular outcome by the proportion of corresponding Likert-scale questions with which they declared ‘‘agree’’ or ‘‘totally agree.’’ For instance, if a stakeholder agreed or totally agreed with 4 out of 5 related Likert-scale questions, the outcome was considered to have an 80% salience level. In this alternative scenario, we adopt stricter criteria, requiring that an outcome is considered experienced only when the stakeholder agrees or totally agrees with 100% of its corresponding indicators. The results, presented in Table 3 case (a), indicate that the SROI ratios decrease on average by 65.7% relative to the baseline scenario.

• Reference wages. As explained in the Inputs section above, the time allocated by stakeholders to the cataract surgery procedure and the autonomy recovered after the intervention are valued at the 2023 average minimum wage for each country. We now adjust this reference value by halving or multiplying it by up to four times. The results are shown in case (b) of Table 3. Modifying the opportunity cost of time in this sense produces average changes of −7.2% up to +20.2% in the SROI ratios, relative to the baseline scenario.

• Value of assurance and alleviation. We also mentioned before that the annual values of these two outcomes are estimated as a reduction equivalent to one month of the 2021 out-of-pocket health expenditure per capita for each country. We now increase this reduction to six and twelve months. Results are reported in case (c). When assurance and alleviation are valued higher, the average SROI ratios increase by up to 5.7% with respect to the baseline scenario.

• Drop-off rate. The baseline scenario used the 2021 crude death rates per 1000 inhabitants as drop-off rates for future benefits. We eliminate and double these rates in case (d) of Table 3. SROI ratios change on average by +5.1% and −4.7%, respectively.

• Duration of outcomes. In case (e) we restrict the duration of outcomes to six years (i.e., the first year of intervention plus five additional years of outcomes). This may align better with funding cycles and results communication needs of nonprofit institutions. On average, SROI ratios change by −43.6% relative to the baseline scenario.

• Discount rate. Finally, we assess the effect of modifying the discount rate, from its baseline value of 10% to 5, 15, and 20%. The results are reported in case (f). The average changes in SROI ratios are +45.2%, −25.0% and −40.5%, respectively.

Table 3.

Sensitivity analysis: average SROI ratios under alternative assumptions

Case	Variable, parameter or criterion	Change	Average SROI ratiosa				Mean difference w.r.t. baseline scenario (%)c
			Patients	Caregivers	All (pooled patients and caregivers)b	Patient-caregiver pairsb
	Baseline scenario (as in Table 2)		12.3	11.4	11.9	9.5
(a)	Full thresholds for experiencing outcomes (individual “agrees” or “strongly agrees” with all indicators)	Stakeholders “agree” or “strongly agree” with all outcome indicators	4.8	3.2	4.1	3.4	−65.7
(b)	Reference wage (affects the valuation of time as an input, and of autonomy as an outcome)	Half a minimum wage	11.6	10.9	11.3	8.2	−7.2
		Two minimum wages	13.4	12.1	12.8	11.0	9.7
		Three minimum wages	14.1	12.6	13.4	12.0	15.9
		Four minimum wages	14.6	12.9	13.8	12.6	20.2
(c)	Post-surgery annual reduction of out-of-pocket health expenditure (reference value for assurance and alleviation)	6 months	12.6	11.8	12.2	9.7	2.6
		12 months	12.9	12.2	12.6	10.0	5.7
(d)	Drop-off rate	Eliminate	12.9	12.1	12.5	9.9	5.1
		Duplicate	11.7	10.9	11.3	9.1	−4.7
(e)	Duration of outcomes	6 years maximum	7.0	6.3	6.7	5.5	−43.6
(f)	Discount rate	5%	17.7	17.0	17.4	13.5	45.2
		15%	9.2	8.5	8.9	7.2	−25.0
		20%	7.3	6.7	7.1	5.7	−40.5
Average (across cases)			11.6	10.6	11.1	9.1	−6.3
Standard deviation			3.4	3.4	3.4	2.9	30.2
Minimum			4.8	3.2	4.1	3.4	−65.7
Maximum			17.7	17.0	17.4	13.5	45.2

Notes

^a All figures report average Social Return on Investment (SROI) ratios, defined as the ratio of the present value of outcomes to the value of inputs. Each cell shows the mean SROI ratio for the corresponding group of stakeholders (patients, caregivers, pooled patients and caregivers, or matched patient–caregiver pairs) under the specified scenario. For example, a value of 12.3 indicates that, on average, each international dollar invested generates 12.3 international dollars in social value. Values are directly comparable to those reported in Table 2, but are recalculated for each alternative assumption described in the sensitivity analysis

^b All (pooled patients and caregivers) refers to individual-level observations pooled across stakeholder types, whereas “patient–caregiver pairs” are computed at the matched-pair level

^c Percentage change in the average SROI ratio relative to the baseline scenario, computed across all stakeholder groups. This measure summarizes the overall sensitivity of the results to each alternative assumption. Positive (negative) values indicate higher (lower) average SROI ratios compared to the baseline scenario

In sum, the sensitivity analysis reveals that SROI ratios are most sensitive to the outcome experience thresholds, the duration of outcomes, the discount rate, and the reference wages, the latter used to value the time devoted to the medical procedure and the regained autonomy by both patients and caregivers. This is consistent with Aleman-Castilla et al. [15]. The average individual SROI ratio across all stakeholders and scenarios is 11:1, ranging between 4:1 and 17:1. For patient-caregiver pairs it is 9:1, with a minimum of 3:1 and a maximum of 14:1. The average across all cases (including the baseline scenario) is 11:1, with values between 3:1 and 18:1.

Discussion

This study builds upon Aleman-Castilla et al. [15] by offering an expanded analysis of the social return on investment (SROI) derived from nonprofit bilateral cataract surgeries across diverse Latin American contexts. The findings reveal that these interventions deliver substantial social and economic benefits, reflected in average SROI ratios of 12:1 for patients, 11:1 for caregivers, and 9:1 for patient-caregiver pairs. These results highlight how cataract surgeries not only restore vision and improve autonomy for patients, but also contribute to alleviating caregiving burdens, generating meaningful emotional and economic benefits for caregivers.

The multidimensional approach adopted here –combining sentiment analysis, valuation of inputs and outcomes, baseline benefit-cost ratios, and sensitivity analyses – provides a solid foundation for interpreting the effects of bilateral cataract surgery and generating actionable policy insights. The sentiment analysis of patients and caregivers reveals predominantly positive emotional responses, characterized by feelings of joy, trust, and anticipation. This suggests that cataract surgery may contribute to improvements in psychosocial well-being. The alignment of these findings with existing literature confirms the broader value of vision-restoring interventions.

The analysis also identifies substantial variability in the costs incurred by stakeholders, which may have implications for equity in healthcare access. For instance, high travel and treatment costs at certain institutions may emphasize the need for targeted interventions to reduce financial and geographic barriers to care. At the same time, the baseline SROI ratio estimates emphasize the transformative potential of cataract surgeries in low- and middle-income countries. These high returns are consistent with prior evidence on vision-restoring procedures and suggest that nonprofit providers can play a meaningful role in delivering high-value healthcare solutions to underserved populations.

An important consideration, nonetheless, concerns the extent to which the social returns documented in this study could also be achieved within national public or private healthcare systems. The benefits captured by our framework –such as improved autonomy, reduced caregiving burdens, and productivity gains – are not intrinsic to nonprofit provision per se, but rather to the restoration of vision itself. Similar social returns could therefore be realized by in-country health systems if comparable levels of accessibility, timeliness, geographic coverage, and affordability were ensured. At the same time, while nonprofits often fill service gaps where public provision is constrained, their presence may generate unintended effects if it leads to unfair competition with local providers or undermines long-term health system development. It is therefore crucial to view nonprofit organizations as complements to, rather than substitutes for, national health systems.

Our sensitivity analysis provides additional insight into the robustness of these findings by illustrating how SROI ratios respond to alternative assumptions regarding key parameters. For instance, variations in reference wages –where wages were halved or multiplied up to four times – showed that SROI ratios fluctuated from a −7% decrease to a + 20% increase. This indicates that the way time is economically valued can have an important effect on the returns on investment, as lower opportunity costs lead to reduced calculated benefits from time savings, while higher wages amplify these perceived benefits, reflecting the importance of valuing caregiver time appropriately, particularly in the context of vulnerable population groups in low- and middle-income countries. Similarly, variations in discount rates significantly influence the valuation of long-term benefits, with lower rates amplifying SROI ratios. These results offer valuable guidance for designing and evaluating similar interventions, to ensure their scalability and sustainability.

Despite its strengths, this study acknowledges several limitations. First, the analysis focuses on seven nonprofit organizations, which may limit the generalizability of the findings to other contexts. Although these organizations represent a diverse range of settings and operational scales within Latin America, they do not capture the full heterogeneity of nonprofit, public, or private healthcare providers. Second, the analysis relies on post-operative self-reported data from patients and caregivers, based on retrospective assessments of change rather than direct pre–post measurements. Although interviews were conducted within a relatively short period following surgery, this approach may still be subject to recall bias, particularly with respect to pre-operative conditions. Nonetheless, this design is consistent with the SROI framework, which places central emphasis on stakeholders’ own accounts of the changes experienced as a result of an intervention, and is common in studies conducted in settings where baseline data collection is constrained by logistical, financial, or access limitations. Third, the valuation of non-market outcomes such as autonomy and emotional relief relies on proxy measures, including opportunity costs of time and out-of-pocket health expenditures, which may not fully capture individual-level welfare gains. Lastly, the geographic focus on Latin America, while providing critical insights into this region, limits the generalizability of the findings to other global contexts. Socioeconomic conditions and healthcare system structures differ substantially across regions, such as Sub-Saharan Africa or Southeast Asia, where nonprofit healthcare models may operate under distinct challenges and opportunities.

Future research could extend this analysis by incorporating a broader range of healthcare providers, including public and private institutions, to assess the comparability of social returns across different delivery models. In addition, longitudinal study designs with pre- and post-intervention measurements would allow for a more precise assessment of changes attributable to cataract surgery and help address limitations associated with retrospective self-reported data. Such approaches could also provide deeper insights into the long-term effects of vision-restoring interventions on labor market participation, caregiving dynamics, and healthcare utilization. Moreover, identifying institutional practices associated with higher SROI ratios could inform strategies aimed at optimizing healthcare delivery while supporting broader health system strengthening objectives.

Conclusion

This study provides evidence of the social and economic value associated with nonprofit bilateral cataract surgeries in Latin America. The average SROI ratios of 12:1 for patients, 11:1 for caregivers, and 9:1 for patient-caregiver pairs reflect the potential broader benefits of these interventions. Sensitivity analysis further supports these findings, showing a range of returns on investment from 3:1 to 18:1 depending on key parameters. These outcomes suggest that nonprofit healthcare providers may play a relevant complementary role alongside public health systems, particularly in addressing service gaps affecting underserved populations. Furthermore, by alleviating caregiving burdens, bilateral cataract surgeries may help create conditions that facilitate greater labor force participation or non-market productive activities, especially among young female caregivers. Nevertheless, the focus of this study on seven Latin American nonprofit organizations and a specific intervention limits the generalizability of the results. Expanding future research to include additional healthcare services, regions, and long-term results will provide a deeper and better understanding of the role and potential of nonprofit healthcare initiatives in improving lives and strengthening communities.

Note

ⁱSample sizes were calculated separately for each institution to ensure representativeness of the total number of bilateral cataract surgeries performed in 2023, assuming a 95% confidence level, a 5% margin of error, and a 50% heterogeneity level.

ⁱⁱBilateral cataract surgeries are typically performed sequentially to minimize risks such as infection, optimize outcomes based on feedback from the first eye, and enable postoperative adjustments for the second eye. Simultaneous surgery may be considered in exceptional cases, such as logistical challenges or urgent rehabilitation needs, but requires stringent protocols.

ⁱⁱⁱThe GTC is widely used in transport economics to measure monetary and non-monetary cost of journeys [40]. For each person , , where is the value of time spent traveling, and is the vehicle operating cost. is calculated by multiplying the minutes spent in land transportation, from the person’s place of residence to the corresponding institute, by the 2023 minimum hourly wage in international dollars. is obtained by multiplying the distance traveled by a Basic Vehicle Operating Cost (BVOC) of Int$ 0.77 per vehicle-kilometer, corresponding to the Mexican baseline scenario for light vehicles and an International Regularity Index of 6 m/km [41]. Travel times and distances were estimated using Google Maps. For the BVOC, we used the Mexican parameters since, to our knowledge, there are no publicly available data for the other countries studied.

^ivRegarding emotions, joy is characterized by happiness and contentment, creating a sense of well-being; sadness involves feelings of sorrow and unhappiness, often linked to loss or disappointment; anger represents displeasure or hostility, usually triggered by perceived injustices; fear is an emotional response to perceived threats, causing anxiety and apprehension; trust reflects confidence and reliability, essential for building strong relationships; disgust is a feeling of aversion or repulsion towards something unpleasant; surprise arises from unexpected events, leading to feelings of astonishment; and anticipation involves excitement and expectation about future events, often accompanied by eagerness. Regarding sentiment polarities, positive refers to words that convey favorable, optimistic, or affirmative emotions, while negative includes words that express unfavorable, pessimistic, or adverse emotions.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Biographies

Ellery Lopez-Star

Managing Director at IMO. Fellow in Vitreo-Retinal Surgery at UNAM. He has a master’s degree in Hospital Management from the UOC and a master’s in business administration from Anahuac University.

Benjamin Aleman-Castilla

Economics consultant, author of various research papers and technical reports. He holds a PhD in Economics from the London School of Economics and Political Science (LSE, 2007).

Luis Andrés Ochoa Ramírez

Medical researcher and ophthalmology resident at IMO, with a master’s in medical research from Autonomous University of Queretaro. Published researcher in virtual reality surgical skills, genetic knowledge, and diagnostic imaging techniques.

Valeria Sánchez-Huerta

Managing Director of APEC and President of the Mexican Center for Cornea and Refractive Surgery. She is also professor of High Specialty and Ophthalmology at UNAM.

Pedro Arnulfo Gómez Bastar

Chairman Ophthalmology Instituto de la Visión, Universidad de Montemorelos. CBM Medical Advisor Latinoamerica. Profesor Ophthalmology Medical School Universidad de Montemorelos.

Olivia Gómez Portillo

Commercial engineer, with amaster’s degree in public health management and hospital administration.Director of the Vision Program. Vice-president of CONAVIP, which is the National Commission of Vision in Paraguay.

Mariano Yee Melgar

Ophthalmologist from Universidad de San Carlos de Guatemala, Seva Foundation Board Member, accredited researcher by MSPAS, andfounder of Visualiza Eye Care System in Guatemala.

Rafael Arias Guerrero

Commercial engineer with a specialization in financial administration and a master’s in public administration focusing oninstitutional development, graduated from the State University of Milagro,Ecuador. Currently serving as Administrative Director of HCE Fundamisc.

Félix Torres Cotrina

Managing Director of INO with master’s degree inpublic health with emphasis on Occupational Health and Environment at theUniversidad San Martin de Porres.

Ana Cristina Dahik Loor

Professor of political and social environment and Director of the Social Responsibility Research Center at IPADE. Currently undergoing her PhD in Management at Bayes Business School, City University of London.

Author contributions

All authors contributed to the study conception and design. Material preparation and data collection were performed by Ellery López-Star, Luis Andrés Ochoa Ramírez, Valeria Sánchez-Huerta, Pedro Arnulfo Gómez Bastar, Olivia Gómez Portillo, Mariano Yee Melgar, Rafael Arias Guerrero and Félix Torres Cotrina. Analysis and interpretation of the data were performed by Benjamin Aleman-Castilla, Valeria Sánchez-Huerta, Luis Andrés Ochoa Ramírez and Ellery López Star. The first draft of the manuscript was written by Benjamin Aleman-Castilla, Ana Cristina Dahik Loor and Luis Andrés Ochoa Ramírez, and all authors commented on previous unfinished versions of the manuscript. All authors read and approved the final manuscript.

Funding

The authors declare that they have no funding, financial, non-financial or personal relationships with other people or organizations that could inappropriately influence this work. The authors also declare that they have no conflict of interest regarding the findings and conclusions of this study.

Data availability

The data used in this study are not publicly available due to ethical, legal, or other concerns.

Declarations

Human ethics and consent to participate declarations

This study was conducted in accordance with the Declaration of Helsinki for medical research involving human subjects. The research protocol was reviewed and approved by the Research Ethics Committee and the Research Committee of APEC Hospital de la Ceguera (protocol EPI-23–01) and subsequently ratified by all participating institutions. The study was classified as minimal risk, involving no clinical interventions and relying exclusively on questionnaires, telephone interviews, and retrospective data collection. Participation was voluntary, and informed consent was obtained from all participants prior to inclusion.

Competing interests

The authors declare no competing interests.