Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
I. Darnton-Hill, F. Sibanda, M. Mitra, M. M. Ali, A. E. Drexler, H. Rahman, and M. A. Samad Khan
Introduction
Vitamin-A deficiency is widespread in Bangladesh. Because of this, the prevalence of night blindness and xerophthalmia is well in excess of the levels at which WHO considers there to be a public health problem [1]. It is likely that, in conjunction with high levels of infectious diseases, vitamin-A deficiency contributes to the high infant and child mortality rates [2].
Three national nutrition surveys have shown very low levels of vitamin A in the diet [3-5]. The most recent, in 19811982, showed that the average intake was only 38% of the recommended daily allowance, and that 88% of families had intake below the requirement [5]. Other dietary studies have confirmed such findings.
Although vitamin-A deficiency had been identified as a problem earlier, it came to notice particularly after the liberation war in 1971. At that time WHO recognized a need to prevent blindness in children by providing high-potency vitamin-A supplements as an emergency measure until dietary intake improved. In 1973 the government of Bangladesh instituted a scheme for distributing high-dose vitamin-A capsules (200,000 IU) to all rural children under six years old.
The capsules, donated through UNICEF, now also contain 40 IU of vitamin E. The distribution within the country is the responsibility of the Nutritional Blindness Prevention Programme of Bangladesh (BPPB) within the Institute of Public Health Nutrition and Dietetics (IPHN), which is in turn a part of the Ministry of Health and Family Planning.
In the absence of improvement in diets, and indeed with probable deterioration in vitamin-A intake, the capsule-distribution programme has continued to the present.
The distribution is carried out by primary-healthcare workers, officially including both health assistants (male) and family welfare assistants (female). The two groups of health workers are administratively separate except at the levels of the upazila (subdistrict) health and family planning office and of the Ministry of Health and Family Planning. The health-delivery system is shown in figure 1.
The vitamin-A capsule distribution programme has three main components: (1) delivery and collection of capsules for distribution to rural areas, (2) distribution of the capsules to recipients and reporting back or monitoring of this distribution, and (3) assessment (internal monitoring). This paper describes each of these components and also looks at coverage of the targeted children, the impact of the programme, constraints affecting coverage, and possible future directions.
Delivery and collection of the capsules
The distribution rounds are now planned to take place every six months, in April-May and October November. Actual distribution times, however, have been very variable. The steps in the delivery and collection of the vitamin-A capsules are as follows (see fig.. 2).
The capsules arrive at Chittagong port, where they are cleared. The requirement for Chittagong Division is retained in Chittagong, and the rest are sent to the Central Medical Stores Depot in Dhaka.
An allotment list is prepared at IPHN on the basis of an estimate of the number of children 6 months to 6 years old in each district, calculated from the 1981 population figures from the Bureau of Statistics with allowance for a 2. 36% annual increase and the assumption that this age group constitutes 22% of the population. The existing balance of capsules on hand in each district is also taken into account; if the stock position is not known, IPHN writes to the district civil surgeon to obtain the information before preparing the allotment list. (Information collated from the reporting forms from the previous distribution round is also available but is not used for calculating the distribution requirements.)
A month before the distribution round is due to begin, the allotment list is sent out with a letter from IPHN to the civil surgeons, instructing them to collect the capsules for their districts. The civil surgeon sends in a request for the number of capsules needed, and the BPPB issues a letter of authorization to the Central Medical Stores to give the capsules to the person carrying the letter. Reporting forms (BPP-1) are issued from IPHN.
The capsules and forms are brought to each district headquarters and stored at the district reserve stores. From there they are distributed to the upazilas under the direction of the district health superintendent. The number of capsules released is based on an estimate of the number of children in the age group in each upazila, prepared by the district health superintendent.
The upazila health and family planning officers, on receiving instructions from the civil surgeon, send messengers with letters of authorization to the district headquarters to take delivery of the capsules and reporting forms from the district reserve stores. These are then stored at the upazila health complex, from where the health inspector or the storekeeper distributes them to the health workers.
FIG. 2. Vitamin-A capsule distribution logistics and procurement
Distribution of the capsules and reporting of coverage
Distribution
Having received the capsules, the health assistants and family welfare assistants distribute them to eligible recipients. The recipients are all children 6 months to 6 years old, and any child 7-15 years old who is night-blind. All children found to be suffering from measles, diarrhoea, or severe malnutrition should also receive the capsules in the course of the usual primary health care.
The designated procedure is for the health worker to snip the top of the capsule and express the oily contents into the child's mouth. Older children can swallow the capsules whole. Children under one year old should receive half a dose. In practice it would appear that the capsules are often left with a male relative or the mother. Among the reasons given for this are that the male health assistants culturally should not be dealing with the mothers, that their work loads are unrealistic, and even that they do not have scissors to snip off the top of the capsules.
Reporting
Data are collected by the health assistants and family welfare assistants from the field on the BPP-1 reporting forms and submitted to the assistant health inspector, who consolidates them into weekly statements, which are in turn submitted to the health inspector. The health inspector consolidates these weekly statements and prepares a statement for his upazila, which is forwarded to the civil surgeon at the district headquarters, where the district health inspector prepares a district statement from the upazila data. The district statements are forwarded to the BPPB headquarters in Dhaka for further processing by the statistician and the monitoring and supervising officer, who prepare the final statement for that round.
The old BPP-1 form asked for quite a lot of information, and the information that was returned is considered to be generally unreliable. Figures for coverage from this source are routinely around 80% which is higher than the assessment figure from the internal monitoring and considerably higher than the figure from the Bangladesh Nutritional Blindness Study of 1982-1983 [1], which showed an average coverage of less than 46%, with wide variation from district to district (15.7%-84.9%).
To reduce this discrepancy, the BPP-1 form has been simplified and now asks only five questions: the total number of children 6 months to 6 years odd, the number of these children who are night-blind, the number of capsules given for this age group, the number of older children (7-15 years old) who are night-blind, and the number of capsules given for this group. There was also a form BPP-1A for a list of people not receiving vitamin-A capsules, but, as nothing was being done with this information, it is to be discontinued, further lightening the load of the health workers.
A statement with a flow chart of the reporting procedures has been prepared, entitled "Recommended Monitoring System," which essentially restates what should be happening (fig. 3). If followed, it should improve the quality of the return. The consolidated reports are now entered into a newly acquired computer at IPHN, and, while this will not directly improve the accuracy of the data, it will allow clearer and faster feedback of reports to the field, which, it is anticipated, will have a motivating effect. Previously, coverage reports took two years to emerge from the time of the distribution round.
Internal monitoring
The programme has an internal monitoring system (the assessment). On the instructions of the director of IPHN, the civil surgeons are responsible for assessing the success of each round. The procedure involves surveying 30 households in one "JL number" (a local, village-level administrative unit) in each upazila of the district, selected according to specific instructions from a list of numbers determined randomly at the central level. Having identified the JL number, district authorities visit 30 households serially, enquiring about receipt of capsules and other questions related to vitamin A.
The questions include whether any children in the household up to 15 years old are blind or suffer from night blindness and whether they received vitamin-A capsules, how the capsules were administered (if received), whether the mother knows what the capsules are for, whether members of the household have seen the vitamin-A flip chart, and whether the respondent can name any foods that may help in preventing blindness and, if so, what foods.
Again, there is considerable scepticism about the information received from this assessment, as the district personnel responsible for carrying it out are, in effect, monitoring and assessing themselves. The findings are reported to Dhaka, and action is rarely taken on them by the local supervising officer. Spot checks suggest that in fact many mothers do not know what the capsules are for and what foods can help in preventing blindness.
The assessment forms are again complicated and require the assessing officer to be able to use codes on the spot. It appears likely that some questions on the forms may be completed after the survey, in which case the data are less likely to be accurate. The assessment figure for coverage is usually about 10% less than that derived from the field workers' reports. A new, simplified form has been designed that requires check marks only, the information from which could be more easily computerized, but there is concern that it would increase costs and the amount of paper required for forms. Consequently, the feasibility of simplifying the existing form is also being examined.
The monitoring and supervising officer and the statistician at BPPB analyse the assessment reports and compare them with the field reports. They also make spot checks in the field. UNICEF anticipates providing funding for an external organization to do monitoring to assist with problem points.
FIG. 3. Monitoring system of the vitamin-A capsule distribution programme of the Nutritional Blindness Prevention Programme of Bangladesh (BPPB)
Health Assistant/Family Welfare Assistant
- number of children reached by the vitamin-A capsule programme,
- number of children with night blindness 6 months to 6 years old,
- number of children with night blindness 7-15 years old.
Assistant Health Inspector/Health Inspector
Upazila Health and Family Planning Officer
Civil Surgeon
Nutritional Blindness Prevention Programme of Bangladesh headquarters, Dhaka
Director General of Health Services
Coverage of targeted children
The most recent capsule distribution at the time of writing was the twenty-third round, which took place in November-December 1986. The reports from the twenty-second and twenty-third rounds were the first to be analysed by computer. The estimated coverage in the twenty-third round on the basis of the field reports was 87.4% (and by the subsequent assessment 82.4%), with rates of night blindness of 0.76% for children 6 months to 6 years old and 0.44% for those 1-15 years old.
The first mass distribution was held from February to June 1973, and the second in June-July of the following year, during severe floods and famine. The distributions have occurred at different times and have been modified in various ways over the years since they started. In January-February 1979 two new target groups were introduced: 7-15-year-olds with night blindness, and lactating and pregnant mothers; the latter group was excluded after 1981. After 1976, semi-annual distribution was undertaken except in 1978 and 1982. High coverage was reported for the first round, which may be explained by the initial resources and enthusiasm. Because of the unreliability of the data, however, comparisons or trends are unlikely to be completely valid.
The government's reporting system consistently indicated that around 60% of all eligible children received capsules up to 1982 (except in the first round, when the figure was 78.2%), and 75% or more since then. There is abundant anecdotal evidence, however, that the system is not functioning appropriately, and the number of targeted children who have actually received vitamin A is thought to be considerably lower.
For the sixteenth to the twenty-third rounds (i.e. since 1982), the lack of congruence between the coverage as reported by the health workers and the sample assessment has been variable, ranging from -3.3% to 15.2% (table 1). Taking all eight rounds, 39%-66% of mothers were reported as knowing what the capsules were for, and 15%21% as having seen the flip chart. (Flash cards will replace the flip charts from the twenty-sixth round [1988] on.) It was reported that 25%-48.5% of the mothers could name vitamin-A-rich fruits or vegetables.
TABLE 1. Differences between reported coverage (as percentage of estimated target population) and assessed coverage. 1982-1987
Round |
Report (%) |
Assessment |
Difference |
16 |
78.5 |
68.8 |
9.7 |
17 |
81.2 |
66.0 |
15.2 |
18 |
75.0 |
70.1 |
4.9 |
19 |
79.5 |
73.9 |
5.6 |
20 |
82.9 |
73.6 |
9.3 |
21 |
89.0 |
81.5 |
7.5 |
22 |
75.5 |
79.0 |
-3.3 |
23 |
87.4 |
82.3 |
5.1 |
The percentage of totally blind children between birth and six years of age was reported as 0.27% in 1973, but prevalences reported from 1979 on ranged from 0.016% to 0.065%, and to 0% in the most recent report. The national nutritional blindness survey in 1982-1983 found a figure of 0.06% [1].
Figures for the fifteenth round (June-July 1982) [6], undertaken at approximately the same time as the nutritional blindness survey, show 70.1% coverage of children from birth through six years of age as reported by the health system and 62.4% by the follow up assessment, compared with 46% by the survey. The prevalence of night blindness in the same age group was reported as 1.10% by the health system, 0.77% by the assessment, and 3.6% by the survey. The percentage of mothers who knew what the vitamin capsules were for was 34% according to the health system, but less than 20% according to the survey. Only 17% were said to be able to name a vitamin-A-rich food at the time of the fifteenth round.
Impact of vitamin-A capsule distribution
It is surprising perhaps, for such a large programme, that its actual effect has been evaluated only once in the 15 years, using the prevalence of xerophthalmia and blindness as end points. As there were no baseline data with which to compare the findings, however, the results were cross-sector findings and the impact was calculated. UNICEF, IPHN, and Helen Keller International (HKI) have discussed the possibility of repeating the nutritional blindness study in 1992-1993, which would be 10 years after the first one and would therefore show trends.
The principle of high-dosage vitamin-A capsule distribution is to boost liver stores against a period of later deficiency where intakes are chronically inadequate [7]. It is assumed in a country such as Bangladesh, where breast-feeding is nearly universal, that vitamin A will be supplied to infants in their first half-year of life through breast milk, though the vitamin-A status of the mother is also a factor. In fact, in Bangladesh the major source of vitamin A for the first 20 months of life has been found to be breast milk [8].
There has now been 12 or more years of experience with the provision of vitamin-A supplements to children living in areas where deficiency is endemic in India, Bangladesh, and Indonesia. More recently programmes (not always national) have been instituted in Brazil [9], Haiti [10], El Salvador [11], and parts of the Philippines [12].
The distribution of capsules containing 200,000 IU of vitamin A in Indonesia began on a pilot basis in 1972 and was evaluated on the island of Lombok in 1977 [13]. Distribution in Indonesia is by four channels: three ongoing programmes and one vertical, single-purpose activity using volunteers to distribute capsules twice a year. Lombok, which had been identified as a high-risk area and was using three of these channels plus radio messages, was evaluated during 1977-1982. The prevalence of Bitot's spots dropped significantly from 1.6% in 1977 to 0.24% in 1983, and the frequency of corneal xerophthalmia dropped from 0.21% to 0.04%. There was no control group, but in areas of Indonesia where there was no vitamin-A distribution, the prevalence of xerophthalmia had remained stable or even increased [13]. A more recent study in Aceh, Indonesia, showed at least 70% of targeted children receiving at least one capsule in the previous year, and villages identified as having high. medium, and low coverage consistently had low, medium, and high frequencies, respectively, of signs of xerophthalmia [14].
In India the distribution has been by auxiliary nurses and midwives to their allotted children, or by the woman health visitor or family planning health assistant to other children. The programme is planned and supervised by the Directorate General of Health Services, and the Maternal and Child Health and Family Welfare Organization is responsible for its administration [ 13]. The programme uses a semiannual distribution of 200,000 IU of vitamin A in 2 ml of flavoured arachis oil given by spoon to each preschool child.
Periodic massive dosing was first suggested in 1964 [15]. In a recent comprehensive review, West and Sommer [10] concluded that programmes that are well supervised and sustain at least 65% coverage of their at-risk populations can expect to have a measurable impact on the occurrence of xerophthalmia among those children.
This coverage threshold figure, similar to the 60% found for Indian conditions, suggests a problem for Bangladesh, where, although monitoring figures for coverage range from 56.6% to 89.0°/o (with an average of 69.8% over 12 years), the external evaluation gave a figure of 45%. A recent preliminary report (of an entirely different project and not yet cleared for citing) found figures of less than 50% (42.2%-45.1%) for the coverage of households that had been visited by any vitamin-A-related health worker over three years. An even more recent survey of vulnerable group development mothers by the World Food Programme found that 39% of their children had received at least one vitamin-A capsule in the preceding six months. The authors were informed that in the twenty-fourth round capsules were not distributed in several districts.
In the national nutritional blindness survey, about one-fourth of all rural sites had less than 25% coverage [16]. In the cities, which are not covered by the health system for the capsule-distribution programme, the average was 22%. In Chittagong four out of the five slums visited had not received vitamin A at all, and the average for Dhaka urban slums was a mere 2.4% [16]. In the country as a whole, the prevalence of night blindness ranged from 4.34% for sites with 0%-24% coverage to 2.75% where coverage was over 75% (P < .001), although the prevalence of Bitot's spots remained essentially unchanged. West and Sommer [10] state that "no measurable effect on xerophthalmia [is observed] when universal coverage is less than 20-25%."
Nevertheless, the potential efficacy of mass programmes has been established [10, 13, 17]. The evidence includes evaluations showing success in Indonesia [18] and India [19-21], as well as early Filipino [12], Nepalese [22], Indian [23, 24], and north-east Brazilian [9] evaluations of pilot or restricted schemes, all of which show positive effects. Sinha and Bang [19] found newly incident cases of xerophthalmia in West Bengal to be 8.8% in controls compared with 0.8% in children who received supplements (P c .0004). This gives a degree of efficacy of 91%, which is similar to that observed in the Indonesian evaluation and the two other cited Indian experiences [10].
An evaluation of a twice-yearly mass-distribution scheme in El Salvador showed no apparent effect [11]. This appears to have been because the programme was limited to children l-4 years old, which meant that, if the pattern of xerophthalmia found in hospitalized children applied in the community, probably half the affected children would have been excluded. Given an estimated coverage of 80%, the "maximum potential effectiveness of the programme was therefore limited to 40% before distribution had even begun." The prevalence was also found to be much less than expected.
In the most recent Indian evaluation, in Hyderabad/Secunderabad, the investigators computed an odds ratio of 12.5 (95% CI: 3.2,49.5) as the risk of developing corneal xerophthalmia for children not receiving vitamin-A capsules [20].
West and Sommer [10] concluded that a universal vitamin-A distribution system that achieves reasonable coverage of its target group (e.g., 65% or better) can have a favourable impact on mild xerophthalmia, and estimated 60%80% reductions in disease from the baseline associated with one or more semi-annual distribution cycles. They further concluded: "Evidence from controlled and other supportive clinical studies suggests a 90% or better prophylactic efficacy against developing mild xerophthalmia for at least four to six months among children who are known to have received the supplement. A similar high degree of protection appears to be conferred against corneal disease as well."
An evaluation in Bangladesh in 1979-1980 was a special impact survey in known high-risk areas for xerophthalmia where 100% coverage was reported for two consecutive distribution cycles. Despite this apparent complete coverage rate, the programme was still unable to reduce night blindness to substantially less than 1%, reflecting 73% efficacy associated with a semi-annual dosing schedule in these areas [25].
This 1979-1980 evaluation used a figure of approximately 60% coverage since 1975 with a distribution interval of about six months, and concluded that there had been a considerable impact on night blindness and perhaps the prevalence of Bitot's spots but little effect on corneal disease. Cohen et al. [26] also expressed disappointment at the relatively poor efficacy of the Bangladesh programme in terms of its not being able to reduce night blindness below 2% or Bitot's spot prevalence below 0.7%, even in areas where more than 75% of children were reportedly given vitamin A. They also referred to the relatively disappointing results in the earlier Bangladesh evaluation but noted that this was contrary to the experience of neighbouring West Bengal [19], as well as the Philippines, Nepal, probably Indonesia, and lately Brazil. In Nepal a 67% reduction in the prevalence of night blindness and a decline of about half that in the prevalence of Bitot's spots with conjunctival xerosis was demonstrated [22].
In Bangladesh no consistent relationship was seen between xerophthalmia prevalence and vitamin-A capsule distribution when assessed by district figures extrapolated from site figures of the national nutritional blindness study [1]. There was, however, a significant correlation between rates of night blindness and coverage (Pearson's r = .4).
Nutritional status (assessed by mid-upper-arm circumference) was also related: significantly lower numbers of poorly nourished children were receiving the capsules, which suggests that it is those at highest risk who are being missed [1], a recognized problem with massive-dosing programmes [14, 27]. The study also found that "larger houses with better socioeconomic status had slightly better coverage" [16]. Although the most powerful discriminator turned out to be the source of household drinking water, coverage did not appear to be markedly influenced by distance and travel time to a permanent road, health centre, or market. Coverage improved with age up to three years, with any sex difference in favour of boys being insignificant [1].
Cohen et al. [1], using a computerized simulation concept, estimated that the distribution programme in Bangladesh in 1982-1983, with a 70% coverage (which is what is claimed but is probably generous), was preventing about 9,000 children from becoming blind each year. Other assumptions made were that one-third of the children with active severe corneal lesions would become blind and 50% of these would die, that effective intervention was 75%, and that the annual incidence of cases was 6 per 1,000. From the graph shown in their report, it appears that at 45% coverage the figure would be nearer 6,000 than 9,0110, although it is unclear whether this would cover geographical clustering and other factors [16]. Continuing the modelling techniques used, they found by linear regression that a 25% increase in existing coverage, to 60% (from the 45% reported in the study), would produce a further one-third fall in the rate of night blindness, from 3.6% to 2.4%. In a more recent article, however, Cohen et al. [7] suggest that "consistently reaching a far higher proportion of children is probably unrealistic."
The strength of the capsules and the interval between doses represent a necessary compromise between cost, practical constraints of the delivery system, and possible toxicity. The 200,000 IU dose represents about three to four times the total liver stores of a well-nourished 15-kg child, about 150 to 200 times the RDA for children 1-6 years old, and 2.5 times the maximum recommended by the International Vitamin A Consultative Group (IVACG) for daily prophylaxis [28]. The high-dose capsule (with 40 IU vitamin E) is given only twice a year, although the IVACG has suggested that distribution every four months would maintain body stores more consistently. This would clearly also complicate logistics and increase the cost.
At present, the government distribution programme covers only rural areas. It is planned that non-governmental agencies will distribute vitamin-A capsules in the four divisional cities of Dhaka, Rajshahi, Chittagong, and Khulna in 1988. Personnel are currently receiving training to undertake the distribution.
West and Sommer [10] noted that "a sparsity of cost data has been reported from ongoing vitamin A delivery programs." Evaluations of cost-effectiveness must use indicators for both delivery and biological cost-effectiveness [10]. A complex calculation would be required that includes ongoing costs, foreign exchange considerations, and health workers' time, as well as the cost of the vitamin A, the capsules, and logistics of distribution. Of the total cost, the major proportion is the delivery and not the nutrient [27. 29]. Mason et al. [29] have pointed out that integration of the distribution component into existing health-care delivery infrastructures would help to reduce the marginal cost of distribution.
In 1974 distribution costs were calculated at about 20 times the value of the capsules. In 1984 the cost of the liquid preparation used in India was about half that of the capsules [13]. The cost is currently about US$0.014 per capsule, including shipping to Chittagong but not distribution, storage, monitoring, etc. A recent IVACG/WHO/UNICEF task force estimated the cost per capsule to be less than US$0.02, although rates of foreign exchange would cause this to fluctuate [30]. A handling fee of about 4% and sea freight (about 15%) are additional, and the cost of delivery to recipients within the country clearly depends on local costs and on the degree to which the distribution is incorporated into the existing health structure. West and Sommer [10] calculated the approximate cost per dose for a semi-annual universal distribution programme with 75%-80% coverage at approximately US$0.44. Despite all the limitations mentioned above on the accuracy of this figure, they concluded that such programmes are both "socially correct and economically sound to pursue."
Reports of transient toxicity symptoms are infrequent in studies from India, and toxicity is almost certainly not a problem in a vitamin-A-deficient country such as Bangladesh. On the other hand, vitamin-A deficiency currently blinds almost 100 pre-school-age children every day, or at least 30,000 each year. More than half of afflicted children die within a few weeks of the blinding episode [26]. The present target population in Bangladesh stands at about 17 million children.
Formerly the health workers providing the vitamin-A capsules were some 14,000 male family welfare workers (now called health assistants), who were trained originally for malaria eradication [1]. More recently the responsibility for distribution has been divided between the male general health assistants and female family welfare assistants from the family planning sector. The actual involvement of the latter is generally recognized to have been disappointingly low. A small proportion of rural households are included in health programmes of non-government agencies. In urban slums, municipal health workers are responsible for distribution, and there is a higher proportion of work by non-government agencies in these poor areas. Some children may be given vitamin A as part of treatment during visits to clinics or other health facilities.
Constraints of the programme
It appears to be generally accepted that the system (distribution, monitoring, and assessment) is not working well. One of the problems identified is the relatively low priority given the programme, especially compared with competing demands on the health system's time. This has also been found in India [17]. Too few staff, apparently low motivation and awareness, insufficient local supervision, and lack of transportation have all been cited as constraints. Recent anecdotal reports suggest, for example, that where the Expanded Programme of Immunization is taking place, vitamin-A distribution has been effectively halted.
Health workers have many tasks and often are unaware of the reason for the vitamin-A distribution despite the fact that their role in this programme is to distribute the high-potency vitamin-A capsules and educate people about their purpose, use, and administration; to motivate people to grow foods rich in vitamin A and eat them and feed them to their children; and to collect data while distributing the capsules. Of course, they also have their other primary-healthcare responsibilities. Basic difficulties of implementation appear to be similar in all countries with mass-distribution programmes; they include problems of logistics, co-ordination, supervision, supply shortages, record-keeping, dosing techniques, worker incentives, and community compliance [10]. Education of the community has been identified as one of the weakest links [17]. One of the likely effects of these problems is that the children who are most difficult to reach and who may also be those most at risk are the ones most likely not to be reached, as has been observed in Indonesia [14] and India [17].
The health workers were trained once, at the beginning of the programme in 1973, since when no further formal training courses have been held. In late 1987 and early 1988 a massive training programme on the prevention of nutrition-related blindness has been initiated by IPHN, UNICEF, and HKI for all categories of health personnel, although there have been some problems in implementing this.
In theory there is a regular refresher system at the upazila level, in which all health workers come to the upazila health complex at least once a month to discuss problems, including those related to their responsibilities for distributing vitamin-A capsules. Also, before the beginning of every round, the health workers should theoretically be given detailed instructions and suggestions concerning their programme responsibilities. In this meeting, health workers are reminded of the distribution procedures and of the emphasis on health education. In this way, memories are supposedly refreshed each year, but this does not appear actually to happen. However, Rahman [31], who was director of IPHN at the time, reported that district-level reorientation meetings providing evaluation feedback to lower-level programme supervisors were associated with subsequent increases in reported coverage.
Another problem that has been identified is the apparent lack of supervision or accountability. At the upazila level the health and family planning officer is in overall charge of the distribution of the capsules. The health inspectors and assistant health inspectors are expected to go on regular field visits to supervise the work of the health assistants. At the more central level, the project director can write to the civil surgeon and issue letters of authorization but has no administrative power to take action. Evaluation of the programme in Hyderabad showed monitoring and supervision to be the weakest links [17].
One undoubted problem, frequently cited by the people in the field, is a lack of funds and personnel. Many sanctioned positions are not filled: only 55.7% of health positions and 76.8% of family-planning positions are currently filled (table 2).
TABLE 2. Field Staff positions in the Health Services and in Family Planning sanctioned and filled, Bangladesh, 1987
| Sanctioned | Filled | ||
| Number | % | ||
| Health Services a | |||
| health inspector | 1,400 | 895 | 64 |
| assistant health inspector | 4,200 | 2,540 | 61 |
| health assistant | 21,000 | 11,390 | 54 |
| Family Planning b | |||
| FWV (field) | 3,920 | 3,720 | 95 |
| family planning assistant | 4,500 | 4,378 | 97 |
| family welfare assistant | 18,500 | 12,587 | 68 |
a. Information from Information Unit. Directorate General of Health Services.
b. Information from Directorate General of Family Planning.
Possible future directions
Monitoring and evaluation
IPHN now has a trained monitoring officer and a computer. In the past 18 months UNICEF, HKI, and IPHN have been actively involved in upgrading delivery and monitoring. A simplified form has been prepared for the health workers' report and will be used for the twenty-sixth round (April-May 1988).
A recording procedure has been developed so that reported coverage can be separated by upazila (each of about 220,000 people). This is now in place, and the reports of the twenty-second and twenty-third rounds have just been completed using this facility.
It has been proposed by UNICEF that each household in which a child actually receives a capsule should be identified in some way, with an indication of the total number of eligible children (or recipients of the capsule). The current suggestion is that either a rubber stamp or a sticker will be placed somewhere in or on the house, perhaps at the top of the inside door jamb. This mechanism may be field-tested during the twenty-sixth round.
A major training programme is currently in place through IPHN with assistance from UNICEF and HKI. It is anticipated that this will promote improved distribution, reporting, and supervision.
Monitoring and evaluation procedures are being modified as described and will eventually include an annual external evaluation. As mentioned previously, another method of evaluation would be a repetition (after a 10-year interval) of the Bangladesh Nutritional Blindness Study survey. This would assess the impact of the programme on such end points as the prevalence of night blindness and xerophthalmia. Also, UNICEF sees this as a time to reassess (and perhaps target more closely) the capsule-distribution programme.
Distribution, targeting, and related strategies
The time of distribution has varied in the past, sometimes occurring once a year (e.g. in 1973 and 1982), and at other times once in the first part of the year (ranging from January to June) and again six months later (ranging from July to December). The official distribution time for a number of years was May-June and November-December, although distribution in many areas is known to have taken place a month or so behind schedule. Because of observed and reported peaks in xerophthalmia occurrence [32] and seasonal shortages of vitamin-A intake [33], the distribution was changed to coincide with the pre-monsoon time of April-May and the pre-harvest time of October-November, starting in 1986 with the twenty-fifth round.
Using other avenues for distribution (e.g. non-governmental agencies, immunization-programme workers) may be a possible direction. As with the current government programme, sustainability is of concern, and experience with this is limited [29].
A suggestion arising from the national survey was the preferential targeting of areas, households, or children identified as being most at risk. Although this may well cause additional problems in the logistics of distribution, xerophthalmia is known to occur in geographical clusters, and targeting would reduce the number of capsules required. Geographical clustering in Bangladesh results in a more than sixfold difference in the prevalence of mild xerophthalmia [1]. On this basis, it has been observed that "100% target group coverage with vitamin A capsules in only the seven highest prevalence districts would theoretically reduce the national incidence of mild xerophthalmia by two thirds while reducing the number of capsules by one third" [34]. Another advantage of preferential targeting would be that, although it would still be difficult to evaluate actual effects, there is '`more [analytical] power when applied to high-risk sub populations, such as urban slums" [35].
Strategies that should be implemented at the same time as vitamin-A distribution include overall social and economic improvement, nutrition education, home gardens, more appropriate child-feeding practices, establishment of blindness prevention and eye care as priorities for health, possible fortification of appropriate foods, and changes in government food and health policy. Policies to effect the concurrent reduction in protein-energy malnutrition, measles, and acute diarrhoeal and respiratory infections must be actively pursued. In our opinion, there is no current alternative to the vitamin-A capsule distribution programme. In the battle to save the sight and lives of Bangladeshi children, however, it should be complementary to a range of social and public-health activities.
References
1. Helen Keller International/lnstitute of Public Health Nutrition, Government of Bangladesh. Bangladesh nutritional blindness study, 1982-1983. New York: Helen Keller International, 1985.
2. Sommer A, Tarwotjo 1, Djunaedi E et al. Impact of vitamin A supplementation on childhood mortality: a randomized controlled community trial. Lancet 1986; 1:1169-73.
3. US Department of Health, Education and Welfare, Public Health Services. Nutrition survey of East Pakistan, March 1962-January 1964. Washington, DC, 1966.
4. Ahmad K, Huda N, Saber A et al. Nutrition survey of rural Bangladesh, 1975-76. Dacca: Institute of Nutrition and Food Science, University of Dacca, 1977.
5. Ahmad K, Hassan N. Nutrition survey of rural Bangladesh, 1981-82. Dhaka: Institute of Nutrition and Food Science, University of Dhaka, 1983.
6. Jalil MA. Final report of 15th round distribution of high potency vitamin A capsules to the Director General of Health Services, Government of the People's Republic of Bangladesh. Memo no. NOG/1/BPPB/82.831192. 1983. Mimeo.
7. Cohen N, Rahman H, Mitra M et al. Impact of massive doses of vitamin A on nutritional blindness in Bangladesh. Am J Clin Nutr 1987;45:970-76.
8. Brown KH, Black RE, Becker S, Nahar S, Sawyer J. Consumption of foods and nutrients by weanlings in rural Bangladesh. Am J Clin Nutr 1982;36:878-89.
9. Dricot-d'Ans Ch, Dricot JM, Rains Mariath JG. Vitamin A periodic dosing in the state of Paraiba, Northeast Brasil. Xeroph Club Bull 1987;35:3-4.
10. West KP, Sommer A. Delivery of oral doses of vitamin A to prevent vitamin A deficiency and nutritional blindness. Food Rev Int 1985;1:355-418.
11. Sommer A. Assessment of xerophthalmia and the mass vitamin A prophylaxis program in El Salvador. Ministry of Health, Government of El Salvador, and American Foundation for the Overseas Blind. Sept 1973-Dec 1974.
12. Solon F, Fernandez TL, Latham MC. An evaluation of strategies to control vitamin A deficiency in the Philippines. Am J Clin Nutr 1979;32: 1445-53.
13. World Health Organization. Prevention and control of vitamin A deficiency, xerophthalmia and nutritional blindness: proposal for a ten-year programme of support to countries. Nut/4.5. Geneva: WHO, 1984. Mimeo.
14. Indonesian Nutritional Blindness Prevention Program. Impact of vitamin A capsule distribution program on xerophthalmia and mortality in Aceh Province, Indonesia, 1982-1986: final report. Ministry of Health, Government of Indonesia, 1987.
15. McLaren DS. Nutr Rev 1964;22:289. (Cited in West and Sommer [10].)
16. Cohen N, Mitra M. Bangladesh nutritional blindness study, 1982-83. Vitamin A dosing. Dhaka: Helen Keller International, and Institute of Public Health and Nutrition, 1986.
17. Vijayaranghavan K, Reddy V. Delivery of large doses of vitamin A. In: West KP, Sommer A, eds. Delivery of oral doses of vitamin A to prevent vitamin A deficiency and nutritional blindness: a state-of-the-art review. ACC/SCN State-of-the-Art Series, Nutrition Policy Discussion Paper no. 2. United Nations, 1987;109-13.
18. Tarwotjo I, Gunawan S, Reddy S, ten Doesschate J, House E, Pettis S. An evaluation of the vitamin A deficiency prevention pilot project in Indonesia, 1973-75. New York: American Foundation for the Overseas Blind, 1975.
19. Sinha DP, Bang FB. The effects of massive doses of vitamin A on the signs of vitamin A deficiency in preschool children. Am J Clin Nutr 1976;29: 110.
20. Vijayaraghavan K, Rameshawan Sarma KV, Rao NP, Reddy V. Impact of massive doses of vitamin A on incidence of nutritional blindness. Lancet 1984;2:149-51.
21. Reddy V. Periodical distribution of large doses of vitamin A in India. Presented at a joint WHO/USAID/IVACG/HKI meeting on Vitamin A Deficiency and Xerophthalmia, Jakarta, Indonesia, 13-17 Oct 1980.
22. Upadhyay MP, Pillai KK, Gurung BJ, Nepal BP. Xerophthalmia in Nepal, 1980-81. His Majesty's Government of Nepal, undated.
23. Gafar MA. Report on vitamin A prophylaxis programme in Mysore State. SEA/NUT Xerophthalmia Meeting 4. New Delhi: WHO Regional Office for Southeast Asia, 1972.
24. Nair NV. Report on vitamin A prophylaxis programme in Kerala State. SEA/NUT Xerophthalmia Meeting 5. New Delhi: WHO Regional Office for South-east Asia, 1972.
25. Jalil MA. Vitamin A deficiency prevention programme in Bangladesh. In: Report of IVACG meeting, Nairobi, Kenya, November 9- 11, 1981. (Cited in West and Sommer [10].)
26. Cohen N, Rahman H, Sprague J, Jalil MA, Leemhuis de Regt E. Mitra M. Prevalence and determinants of nutritional blindness in Bangladesh children. World Health Stat Q 1985;38:317
27. Arroyave G, Bauernfeind JC, Olson JA, Underwood BA. Selection of intervention strategies. In: Guidelines for the eradication of vitamin A deficiency and xerophthalmia. IVACG report, part I. 1977.
28. Bauernfeind JC. The safe use of vitamin A: a report of the International Vitamin A Consultative Group (IVACG). New York: Nutrition Foundation, 1980.
29. Mason JB, Eastman SJ, Lotfi M. Introduction and policy implications. In: West KP, Sommer A, eds. Delivery of oral doses of vitamin A to prevent vitamin A deficiency and nutritional blindness: a state-of-the-art review. ACC/SCN State-of-the-Art Series, Nutrition Policy Discussion Paper no. 2. United Nations, 1987:2-17.
30. IVACG/WHO/UNICEF task force. Guidelines for the use of vitamin A supplements in the treatment and prevention of vitamin A deficiency and xerophthalmia: a guide for health administrators and programme managers. Mimeo. Geneva: WHO, 1988:1-21.
31. Rahman M. Periodical distribution of large doses of vitamin A in Bangladesh. Presented at a joint WHO/USAID/IVACG/HKI meeting on Vitamin A Deficiency and Xerophthalmia, Jakarta, Indonesia, 13-17 Oct 1980.
32. Kabir I, Khanum S, Rahman H. Xerophthalmia as seen at an out-patient department of an urban nutrition unit, Dhaka. Bangladesh J Child Health 1987;11:6-12.
33. Hassan N, Ahmad K. Seasonal and regional intakes of vitamin A in rural Bangladesh.
National Planning Meetings for Vitamin A Programme proceedings. FAO, 1987;57-64.
34. Northrup RS. Vitamin A capsule distibution in Bangladesh: current situation-alternative strategies. Reports: Occasional Paper Series. Dhaka: Helen Keller International, 1982.
35. Milton RC. Evaluation of the efficacy of programs for the control of severe xerophthalmia. Am J Clin Nutr 1982;35: 140-45.
![FIG. 1. Structure of the Bangladesh health-delivery system (a. Numbers in parentheses indicate the number of units; e.g., there are four divisions in the country [Bureau of Statistics Year Book, 1986]. h. Institute of Public Health Nutrition. c. Approximately equivalent to "JL number" used for land demarcation.)](8F103E03.gif){kind=link}
