Monitoring and evaluating cataract intervention in India

Editor,—I read the article by Limberg et al 1 with interest. I am reminded of ‘Confusion of goals and perfection of means characterise the age’ by Albert Einstein. It is a well tried attempt to infuse quality initiatives in a blindness control programme. However, it is evident that the Indian experience is no different from that of many other organisations in designing, monitoring, and evaluating process indicators.2I do have some serious concerns about possible (ill) use of the indicator sight restoration rate (SRR) as a variable be to included in a mathematical model for assessing the impact of interventional strategies. A study under the aegis of reputed agencies assumes tremendous prestige and conclusions become a weighty verdict. The potential to provide serious misinformation leading to decisions in slowing down the programme is of interest.

(1) It is difficult to understand the rationale of operating in one eye only in a bilaterally blind person. Probably the authors chose to ignore that by denying the operation would make one permanently blind due to complications of hypermature cataract.

(2) I sincerely feel that the indicator SRR is not suitable to monitor qualitative aspects of the programme. It would be evident that the overall sight restoration is not alarmingly low when the first two groups (6/6–6/18 and 6/18–6/60) are taken into consideration. It would have been better if the groups were considered individually rather than treating them as separate groups preoperatively and combining them at the postoperative stage (Table 6). In this way SRR improves dramatically. Suitably modified instruments like ADVS and SF-363 are appropriate to measure vision related quality.

(3) There are 3000 non-operating ophthalmologists without any surgical facilities. No measures have been suggested to include them in the programme. The target of 700/OS/year seems unattainable with the actual figure at 440/OS/year. Any increase above the present actual numbers might dilute the quality of cataract surgery (‘Focus presently is on achieving the targets than focusing on prevention of blindness’). It is well established that any targets, let alone increasing the present levels of targets, are detrimental to the programme objectives altogether.

(4) 1.32 to 2.1 million cases (at 440 cases/surgeon/year to 700 cases/surgeon/year respectively by 3000 non-operating ophthalmic surgeons) can be operated by induction of 3000 non-operating surgeons, with available resources like staff, facilities, and supplies of material, which the authors claim are going unused and with the number of cases increasing through demand generation and case finding. Their suggestion to encourage ophthalmologists to work in areas of low cataract surgery utilisation is wishful thinking. In a health sector where primary health centres remain unmanned for years, the suggestion is impracticable. A model with paradigm shift towards optimum utilisation of available resources efficiently and effectively is needed. The ‘Arvind eye hospital model’4 with suitable modifications to suit the different geographic locations is one of the alternatives.

(5) The suggestion of selecting better cases is like ‘improving the indicator rather than the programme objectives and performance’. This is in contravention of the objectives given in the document prepared by DGHS, Government of India.5 An effectiveness indicator is valid only when it truly serves as a measure of goal achievement.6 By careful selection of ‘proper’ cases, there is a possibility of denying a chance of restoring whatever vision possible with cataract surgery. Hence, selection criteria are not in tune with the doctrine of equity and justice inherent in any national health programmes. The policy document of the World Bank assisted cataract blindness control project (1.4.75) laying down quality related guidelines is not a justification to link funding of cataract surgery to NGOs and private surgeons with sight restoration rate.

(6) In programme implementation, in developing countries with limited resources, the question of equity and justice matter more than quality which is being experimented on with limited success in some countries. No doubt effectiveness is as important as efficiency. But when it comes to quantifying effectiveness on arbitrary rates, caution should be exercised before proper survey research methodology is adopted.

(7) The projected population figure of 200 million in +50 year population by 2011 is higher by 20 million as 5 year groups were not considered individually in the calculation. Ideally, 1 year grouping is preferred to project future population, as artificially high projected population figures necessitate huge resource allocations in planning the programmes (at present both 1 year break up and 5 year break up figures are available7). This would be unrealistic in a country with limited resources.

I hope the readers and policy makers bear in mind the above observations while considering diverse aspects of the National Programme for Control of Blindness.