The capacity building intervention focused on strengthening organizational management, leadership, team building, equitable access and use of services, financial cost-recovery, and quality of care. How this transpired depended upon the local context; while general principles were followed in all settings considerable variation in setting of priority interventions was noted. In most of the South Asian settings the focus was on providing high quality surgery to all sectors of the population, introducing a tiered patient paying system and reorganizing patient flow for maximum efficiency. In African settings priority interventions focused on strengthening basic organizational, personnel, and financial management systems and designing practical outreach programs. Because the services were provided equitably and sustainably through service fees, they met most of the criteria of effective  and systematic  capacity building.
The capacity building intervention itself varied substantially from mentee to mentee and mentor to mentor, depending upon the local context and needs. Some mentee facilities only needed minimal interventions such as re-orientation and team building strategies, others required physical renovations and equipment, while most required outreach programs and clinical and management training. Nevertheless, in all settings, raising production targets and changing institutional attitudes to growth was a big part of the capacity building program. However, target setting for growth was more of an internal exercise to increase supplies, training, and efficiency, not an explicit attempt to meet population need. In all of the settings the need for large increases in eye care services, including cataract surgery, was simply accepted as a background reality in the planning process.
Hospitals and eye units worldwide use cataract surgical volume (CSV) as an indicator of overall eye care activity level. It reflects hospital performance and efficiency including patient choice and acceptance of surgery as well as the quality of ophthalmic professionals, patient experience, equipment and supplies. CSV directly depends on the number of patients attending hospital outpatient departments and community outreach activities, often termed diagnostic/screening ‘camps’. Attendance at outpatient and outreach, in turn, reflects interaction with the service population regarding eye diseases and their awareness of treatment options. While mentee hospitals and eye units gathered and used these statistics for their own improvement, the complexity and variability was too great to include in this initial, broad level, assessment study.
The capacity building intervention was systematic  it resulted in an initial decrease in cataract surgical volume in 11 of 25 institutions due to key clinical staff undergoing training off site, physical alterations to the hospital building and installation and training on new equipment, as well as service populations learning to accept paying fees for previously free services (though free services were still provided for patients too poor to pay). Despite widely diverse population density, disease prevalence, and eye care infrastructure [13, 14] the capacity building intervention show a substantial increase in cataract surgical volume by year 4 in Africa (164%) Asia (66%) and Latin America (136%).
The proportional increase in CSV observed exceeded the change in cataract surgical rate (CSR) over the same time period in all but three jurisdictions [6,7,8,9,10]. CSR is a widely used, population-based measure, while CSV is simply the number of operations per year in an institution, without the population denominator. Nevertheless, the comparison provides a reasonable way to assess mentee growth rate to background eye care system growth in the same geographical area, thereby controlling for broad political and economic changes during the study period. CSR remains unchanged or declines in settings such as Madagascar and Tanzania where most eye care services are part of government hospitals where eye care is a low priority, equipment is not affordable, and active outreach does not occur. In settings such as India, CSR is steadily rising, albeit much slower than the mentee institutions studied here, because of substantial government and private investment in establishing high quality eye care services and broad distribution of eye care services to the primary health care level.
Surgical productivity (defined herein as cataract operations per ophthalmologist or cataract surgeon) also increased in all regions by year 4. This reflects improvement in a range of internal hospital features including training, clinical protocols, patient flow within the hospital and operating room, and reallocating less-skilled tasks from the ophthalmologist to the rest of the eye care team . Surgical productivity is also dependent upon the burden of disease in the population and there is strong evidence that the incidence of cataract (age and sex adjusted) in many African populations is two to four times lower than in other populations .
Direct, walk-in patients who pay for some or all of their care reflect a number of different conditions, some hospital related and some related to the local context. Where transport systems are efficient and effective “direct-paying patients” or their family members have chosen to seek care at the mentee facility over other local and distant treatment options and to use their own money to pay for services. This choice indicates good hospital reputation regarding quality, cost and efficiency of services. Eye care institutions such as AECS in India achieve a cash surplus with only 30–40% of their cataract patients as direct paying patients . As has been shown in different settings in Africa, direct paying patients are generally fewer in number as compared to South Asia. Poor transport systems, long distances, advanced age at development of cataract, and inadequate social support all contribute to a relatively low proportion of direct paying patients [17,18,19].
Of the 14 hospitals that reported direct paying patient information, more than half reported that paying patients made up 15% or less of total cataract surgical patients prior to the capacity building intervention. Many of these hospitals were founded as charity hospitals that provided services to the poor and therefore did not seek payment from patients leaving the hospitals vulnerable to changes in external funding. In the African countries, government hospitals (4 of the 7) had minimal funding for eye care, mostly consisting of covering salaries. In most instances, early discussions between mentors and mentees included conversations regarding implementation of fee structures and payment options suitable to the hospital context and fiscal demographics of their catchment area. Restructuring patient fees to provide a tiered payment system (particularly in South Asia and Latin America) includes taking into account what top, middle and low income earners would be able to pay for surgical intervention based on the most up-to-date demographic information available for the hospital catchment area and building pricing structures on that while still providing an option for free surgery for those who cannot afford the least expensive option.
The assessment of the return on investment of this capacity building process would suggest that the South Asian sites had the best value-for-money. As noted by Lewallen and Thulsiraj  there are a considerable number of factors that limit the ability of directly apply practices from India to different settings in Africa. Comparisons within similar settings may be a more valuable approach to assessing value-for-money. Nevertheless, in all of these settings capacity building proved to be a relatively inexpensive form of increasing the number of people who received cataract surgery.
The capacity building intervention was associated with increased cataract surgical volume and productivity in diverse settings. However, prospective controlled studies are needed to prove the impact on individual institutions in each region. State and national CSR data were used for historical comparison. Although crude indicators, they provide the best available way to capture broad trends in national and state eye health system development
A much better assessment of mentee impact would be captured by measuring cataract surgical coverage (operated cataract as a proportion of operable plus operated cataract) and the more recent ‘effective’ cataract surgical coverage (operated cataract and a good outcome as a proportion of operable cataract plus operated cataract) as applied in Ramke, et al. . While not possible herein with these early mentees, this approach is becoming more and more feasible with the increasing number of Rapid Assessment of Avoidable Blindness survey studies in all of these settings.
Data on the quality of cataract surgery, although gathered by all the mentee hospitals, were not reported in this study. In this program evaluation, high quality surgery was considered a necessary variable in higher level indicators such as the number of services provided and the number and proportion of patients willing to pay for services.