Prevalence and causes of blindness and visual impairment in Mongolia: a survey of populations aged 40 years and older.

Prevalence and causes of blindness and visual impairment in Mongolia: a survey of populations aged 40 years and older J. Baasanhu,1 G.J. Johnson,2 G. Burendei,3 & D.C. Minassian4 The survey was conducted in 3 out of the 18 administrative regions (aimaks); 4345 people aged .40 years were examined, which represented 95.7% of the proposed sample. The prevalences of blindness and low vision in the sample were 1.5% (95% Cl, 0.8-2.3%) and 8.1% (95% Cl, 5.5-10.7%), respectively, from which the prevalences of blindness and low vision in the Mongolian population aged 40 years and older were estimated to be 1.4% and 7.7%, respectively. The prevalence of climatic droplet keratopathy was high (ranging from 15% to 50%) in this population, which included a large number of semi-nomadic cattle breeders, and was responsible for 7.2% of the blind- ness and 19.3% of the low vision. Cataract and glaucoma were the commonest blinding disorders, each accounting for around 35% of the blindness. Trauma accounts for a high proportion of those monocular- ly blind. Trachoma and xerophthalmia were not found. Introduction The prevalence of severe blindness (defined as no light perception, or inability to see hand movements or count fingers at less than 25 cm) in various regions of Mongolia has been estimated at 0.2%; no figures are available on the basis of the present WHO definition (i.e., less than 3/60 vision in the bet- ter eye) (1). There appears to be a high prevalence of climatic droplet keratopathy (CDK) and glaucoma but relatively few cataracts, which is a different pat- tem from most countries in a similar economic situa- tion. In order to direct a future prevention of blind- ness programme and to allocate resources to the best advantage, it was considered important to obtain population-based figures for the prevalence and causes of blindness. The population of 2.1 million is widely scattered over an area equal to that of Western Europe. Despite the long distances and high cost of transport, the health infrastructure is well developed. Excellent census data are available and are updated frequently. I Head, Department of Ophthalmology, First Clinical Hospital, Ulaanbaatar, Mongolia. 2 Director, International Centre for Eye Health (ICEH), Institute of Ophthalmology, Bath Street, London EC1V 9EL, England. Requests for reprints should be sent to Professor Johnson at this address. 3 Programmer, Software Group, Medical Research Institute, Ulaanbaatar, Mongolia. 4 Senior Lecturer and Honorary Consultant, Moorfields Eye Hos- pital, International Centre for Eye Health (ICEH), Institute of Ophthalmology, London, England. Reprint No. 5527 The country is composed of 18 aimaks (regions), each divided into 17 somons (districts). Each district is made up of a centre and 3 to 4 "brigades" (an organization of widely scattered tent houses), the total population varying between 2000 and 4000. Materials and methods Populations were sampled from three aimaks, which were selected to represent the three main geographi- cal regions of the country: Gobi desert, steppe, and mountainous. Three to five somons were selected from each aimak, representing their geographical and ethnic features. The aim was to examine every per- son aged .40 years in the selected districts, as many as possible being seen in the hospital at the somon centre; the others were seen in their homes (gers). The examinations were carried out and recorded mainly by one ophthalmologist in each aimak; four ophthalmologists were involved and were trained together for a standardized approach. During the dif- ferent stages of the survey, the principal investigator visited each aimak during the examinations and was responsible for the overall quality of the records. The timing of the survey in each region, between late May and October 1991 and from July to August 1992, was determined largely by the seasonal activ- ities of the semi-nomadic people. Visual acuity, with correction, was tested using the Cyrillic alphabet or Landolt's C chart. If vision was worse than 6/18, a pinhole was used to test for improvement. Blindness was defined as visual acuity of less than 3/60, and low vision as visual acuity in the range of 6/24 to 3/60 in the better eye with the Bulletin of the World Health Organization, 1994, 72 (5): 771-776 © World Health Organization 1994 771 J. Baasanhu et al. best available correction. The external eye, cornea, anterior chamber and anterior lens were examined with a Zeiss hand-held slit lamp. CDK or chronic actinic keratopathy was graded according to the cri- teria set out in Table 1. Pterygium, provided it was clearly extending onto the cornea, was recorded if present, regardless of size. Cataract was graded by viewing with an oph- thalmoscope from a distance of 20 cm against the red glow of the fundus through undilated pupils. How- ever, in the presence of Grades 3 or 4 keratopathy which obscured the view of the lens, the pupils were dilated with 1% tropicamide so that a clear view of the lens could be obtained over the dark band. Estima- tion of anterior chamber depth was made by side- light test prior to pupillary dilation. Tropicamide was not used in eyes suspected of being at risk of angle closure. The criteria for grading of cataract (Table 1) were based on a method that has been shown to have good inter-observer agreement in surveys (2). In addition to the systemic assessment of CDK, cataract and pterygium in all subjects, those with impaired vision of less than 6/18 in either eye were examined in detail to ascertain the principal disorder causing the visual impairment, and when possible to record the underlying etiology. The WHO protocol was followed for this purpose.a When there were two disorders, one being secondary to the other, the primary disorder was recorded as the cause of the visual loss. When there were two or more co- existing primary disorders, the disorder most readily curable - or if not curable, most easily preventable - was selected as the cause of the poor vision. The detailed eye examination included fundoscopy after dilation of the pupils. All the findings were entered on the WHO/PBL Eye Examination Record (version III). The causes of low vision or blindness were then entered for each eye, and the principal disorder for the person marked. In each case, a decision was made on the underlying cause. Finally, current action needed was indicated. Data entry into a computer database was carried out in Mongolia and analysis was performed in London. For internal comparisons, age-standardized prev- alence data were used, taking the age structure of the whole sample as the standard. An alternative stan- dard might have been the population of Mongolia, but there was no breakdown after age 70 years in the census data available to us at the time of analysis, which would have been a serious disadvantage. Con- fidence limits for the prevalence of blindness and a Coding Instructions for the WHO/PBL Eye Examination Record (Version 11I). Unpublished WHO document, PBU88.1, 1988. Table 1: Grading of climatic droplet keratopathy (CDK) and cataract Grade Description CDK: 0 No CDK 1 Droplets restricted to nasal and temporal bands of cornea 2 Droplets extending into the pupillary area, but not dense enough to interfere with vision 3 Continuous band across the pupil, dense enough to reduce vision 4 Definite raised nodules, yellowish or clear Cataract: 0 Clear red reflex 1 Trace opacity present 2 Opacity occupies less than half of red glow 3 Opacity occupies half or more of pupil space, but some red glow still visible 4 Completely opaque; black, no red glow low vision were computed, taking the 'design effect' (cluster random sampling) into account. This gives less precise estimates compared to simple random sampling, whereby individual people rather than clusters are selected at random. The 'design effect' for prevalence of blindness in this survey was large at 4.66, making the sampling error about 2 times larger compared to simple random sampling. Results The number of people aged .40 years required for the survey in the 12 sampled districts in the three aimaks and the numbers actually examined are shown in Table 2; an average of 95.7% responded. The age distribution of the sample corresponded closely to that of the population of Mongolia for both males and females aged 40 years or older. The prevalences of blindness and low vision in the sampled populations were 1.5% (95% CI, 0.8-2.3%) and 8.1% (95% CI, 5.5-10.7%), respec- tively (see Table 3). The age-specific prevalences are shown in Fig. 1. Using the sample age-specific prev- alence data, the prevalences of blindness and low vision in the population aged .40 years were esti- mated to be 1.4% and 7.7%, respectively. Table 4 shows the age-standardized prevalences of blindness, low vision, cataract, and climatic drop- let keratopathy. Umnugobi aimak showed the lowest prevalence of blindness, but the highest proportion of people with cataract. Conversely, Domod had the highest blindness rate and highest prevalence of CDK, but the lowest percentage with cataract. 772 WHO Bulletin OMS. Vol 72 1994 Blindness and visual impairment in Mongolia Table 2: Size of required sample, aged .40 years, and response rate in the 3 almaks, 1991-92 No. of somons Sample No. Coverage Aimak sampled required examined (%) Umnugobi 4 1 501 1 500 99.93 Dornod 5 1 501 1 403 93.47 Arkhangai 3 1 540 1 442 93.64 Total 12 4 542 4 345 95.66 The causes of blindness, low vision and monocu- lar blindness and their distribution in the 3 aimaks are presented in Fig. 2-4. Glaucoma and cataract are important causes of blindness in this population. Cli- matic droplet keratopathy is an important cause of blindness (7.2%) and low vision (19.3%), and is the major cause of comeal blindness and low vision 5/6 (83%) and 71/89 (80%) respectively. Among the underlying causes of monocular blindness was trau- ma (41%), which accounted for a large proportion of the cases of pththisis as well as some of the other categories. The distribution of the grades of climatic droplet keratopathy in the sample is shown in Table 5 (see page 775); grades 3 and 4 CDK were more prevalent in males in each age group except the oldest (.80 years). The relationship in each aimak can be deduced from the data in Table 4. The numbers requiring various eye care services in each aimak, as estimated by the examining oph- thalmologist, are given in Table 6. For example, the need for cataract extraction is much higher in Umnu- gobi than in the other regions. If the sampled aimaks, despite the variations between them, can be consid- ered as representative of the country as a whole, the eye care services needed by the whole population of Mongolia aged 40 years and over, which is 363 000, can be estimated. This gives a figure of approximate- ly 22 000 persons requiring cataract surgery, 2600 for glaucoma surgery, 3000 for lid surgery and, including other treatment, a total of 49 500 requiring eye care. Fig. 1. Age-specific prevalences of blindness and low vision in .40-year-olds, 1991-92. 50 - .45- 40- 35- 9 30- 0 25 - 1 20 - aL 15- 10 - 5- 0. El Blindness E] Low vision 32.3 9.1 20.529 Lm- _... 40-49 50-59 10. - - - 46.2 60-69 70-79 80+ Age group (years) Discussion The very good response rate (95.7%) in this survey was due to the very good census data, the well- developed health infrastructure, and the outstanding cooperation of the health officials and other adminis- trators. Climatic droplet keratopathy was found to be very common and more severe than in Labrador (Canada), although not as pronounced as in Somalia and the Red Sea coastal areas. Its frequency and severity could have been expected in the cattle- breeders, who all their lives are exposed to the bright glare of the sun reflected from the snow, which in some areas is present for up to 6 months of the year. There are 260 sunny days a year in Mongolia and the climate is dry; there are also strong winds. The high prevalence of CDK, in the presence of a low rate of cataract, deserves comment. There is strong circum- stantial evidence that CDK is due to exposure to reflected UV light, possibly made worse by particu- late injury from wind-blown sand or snow and ice particles. The findings therefore raise questions as to whether UV light is likely to be a major etiological Table 3: Prevalences of blindness and low vision in the samples, 1991-92 No. with No. with No. with monocular Sample Aimak blindness low vision blindness (%) total Umnugobi 16 (1.07)a 128 (8.53) 24 (1.60) 1 501 Dornod 34 (2.27) 174 (11.59) 51 (3.40) 1 501 Arkhangai 19 (1.23) 66 (4.29) 40 (2.60) 1 540 Total: 69 (1.52) 368 (8.10) 115 (2.53) 4 542 Upper limitb (2.29) (10.70) (3.15) Lower limitb (0.75) (5.51) (1.92) a Figures in parentheses are percentages. b 95% confidence limits for overall prevalence. WHO Bulletin OMS. Vol 72 1994 . -& Bn s _ _ I IsA- 773 J. Baasanhu et al. Table 4: Percentage age-standardized prevalences of blindness, low vision, cataract, and climatic droplet keratopathy (CDK) In the 3 almaks, 1991-92 Age-standardized" prevalences (%) in: Umnugobi Domod Arkhangai Blind 1.08 1.97 1.39 Low vision 8.98 10.05 4.77 Blind + low vision 10.05 12.02 6.16 Cataract: Grade 1-4 16.17 5.09 8.38 Grade 2-4 8.49 3.93 5.02 CDK present 18.60 49.71 15.31 (grade 1-4) a For age-standardization, the age structure of the whole sample was used as the standard. factor for cataract in this setting. The correlation between these two conditions will be the subject of a separate analysis and further research. The substantial number of people requiring lid surgery, in the absence of trachoma, is explained by the fact that spastic entropion is common in old age among the cattle-breeders. Many of them have no protective glasses in the severe continental climate. What are the measures to be taken within a pro- gramme for preventing blindness? Because of the very large distances between settlements in Mongo- lia and the difficulty and cost of transport, the key to progress appears to be development of preventive and curative services at the level of the aimaks. At present, each of these has one to three ophthalmolo- gists. Upgrading of diagnostic and microsurgical facilities, and additional training in microsurgery for ophthalmic surgeons would mean that most cases of cataract, trauma and comeal disease could be treated Fig. 2. Causes of blindness, 1991-92 (CDK = climatic droplet keratopathy). Fig. 3. Causes of low vision, droplet keratopathy). Other (7.6%) Phthisis (1.6%) Glaucoma (5.4%) Refractive (5.7%) Macular (0.8%) Unknown (9.5%) 1991-92 (CDK = climatic Cataract (45.1%) (includes uncorrected aphakia ~*0.6%) Corneal (24.2%) (CDK 19.3%, other corneal 4.9%) Fig. 4. Causes of monocular blindness, 1991-92 (CDK = climatic droplet keratopathy). Other (5.2%) Anterior uveitis (7.8%) _*. Unknown (5.2%) Optic atrophy (7.8%) Comeal (10.4%) (CDK 4.3%. other comeal 6.1%) Macular (2.6%) 'V Glaucoma (12.2%) Phthisis (28.7%) _ * B_ * * w_ ...._ @..... ....... _..... ....... v ..., .. \ ., ....\. ..... .... ..... ]....... ........... s....... ........... t ...., .. t ...., .. wE *\ ........ ......... . ...... . . ....... K s8@e@@sss cs........ .......... 4E e @ 9 @ * @ - - _ * @ X X @ - - - @ ar @ @ @ @ @ * - - m @ @ @ @ @ @ @ @ ............ _ @ @ @ * * * si. . . j; B S Z=S , P Cataract (20.056) (indudes uncorrected aphakia O.9%) locally and promptly within the aimak. Considera- tion may be given to a trial of prophylactic treatment for second eyes and cases at high risk of acute angle- Other (7.2%) Macular (2.9%) Phthisis (4.3%) Optic atrophy (5.8%) Corneal (8.7%) (CDK 7.2%. other corneal 1 Cataract (36.2%) (including uncorrected aphakia 2.9%) Glaucoma (34.8%) Table 6: Numbers needing basic eye care services among .40-year-olds, 1991-92 Estimated number in: Umnugobi Dornod Arkhangai Mongolia Total population 7 732 14 244 16 326 363 000 aged >40 years Need cataract surgery 979 323 530 22 004 Need glaucoma surgery 67 95 95 2 563 Need lid surgery 113 19 138 2 959 Need spectacles 118 104 233 4 473 Need medication 227 380 64 7 249 Need other treatments 340 399 201 10 207 Total needing basic eye care 1 844 1 320 1 261 49 455 WHO Bulletin OMS. Vol 72 1994774 Blindness and visual impairment In Mongolia Table 5: Distribution of grades (0 to 4) of climatic droplet keratopathy (CDK) in the sample, by age group and sex, 1991-92 Age group CDK 0 CDK 1 CKD 2 CDK 3 CDK 4 Total 40-49 years: Males 599 52 17 6 3 677 (88.5)a (7.7) (2.5) (0.9) (0.4) Females 834 39 17 0 0 890 (93.7) (4.4) (1.9) (0.0) (0.0) 50-59 years: Males 484 101 49 25 14 673 (71.9) (15.0) (7.3) (3.7) (2.1) Females 517 78 38 8 2 643 (80.4) (12.1) (5.9) (1.2) (0.3) 60-69 years: Males 226 78 72 39 36 451 (50.1) (17.3) (16.0) (8.6) (8.0) Females 219 61 77 29 7 393 (55.7) (15.5) (19.6) (7.4) (1.8) 70-79 years: Males 76 45 43 31 24 219 (34.7) (20.5) (19.6) (14.2) (11.0) Females 111 55 43 34 8 251 (44.2) (21.9) (17.1) (13.5) (3.2) >80 years: Males 30 13 13 11 1 68 (44.1) (19.1) (19.1) (16.2) (1.5) Females 31 7 16 21 5 80 (38.8) (8.8) (20.0) (26.3) (6.3) Total Males 1 415 289 194 112 78 2 088 (67.8) (13.8 (9.3) (5.4) (3.7) Females 1 712 240 191 92 22 2 257 (75.9) (10.6) (8.5) (4.1) (1.0) Both 3 127 529 385 204 100 4 345 (72.0) (12.2) (8.9) (4.7) (2.3) a Figures in parentheses are percentages (in the row). closure glaucoma. A long-term clinical trial for the prevention of CDK with protective sunglasses is also proposed. Extrapolating our results to the national population aged 40 years and over, we see that approximately 22 000 people require cataract surgery and some 50 000 need some kind of eye service or treatment. With 80 ophthalmologists and 60 ophthal- mic nurses already trained and available in the coun- try, serving these people is an achievable target. Acknowledgements We thank Dr Erdenelebeq and Dr Orsoo for their examina- tion of patients in two aimaks. This project was supported by the WHO South-East Asia Regional Office and by the Voller Sisters through Fight for Sight, London. Resume Pr6valence et causes de la c6cit6 et des alt6rations visuelles en Mongolie: enquete chez les 40 ans et plus Cette population de 2,1 millions de personnes est diss6minee sur un territoire comparable a l'Europe occidentale et est desservie par des services de sante bien d6veloppes, alors que le revenu par habitant est assez faible. Une proportion impor- tante de la population a conserv6 un mode de vie traditionnel d'eleveurs de b6tail semi-nomades. Des rapports ont signal6 une forte prevalence du glaucome et de la keratopathie en gouttelettes, mais aucune donnee sOre n'est venue confirmer WHO Bulletin OMS. Vol 72 1994 775 J. Baasanhu et al. I'ampleur ni les causes de la c6cite et de la bais- se de vision. On a choisi trois des dix-huit regions adminis- tratives (aimaks) pour representer les principales zones geographiques du pays: desert de Gobi, steppe et montagne. Chaque aimak est subdivise en somons (districts) de 2000 a 4000 habitants. Le but de 1'enquete etait d'examiner tous les sujets de 40 ans et plus dans 3 a 5 somons de chacun des 3 aimaks choisis. Au total, 4 345 per- sonnes ont 6t6 examinees, ce qui represente 95,7% de I'echantillon souhaite. La cecite et la baisse de vision ont ete notifiees en fonction de criteres de I'acuite visuelle, apres tentative de meilleure correction inferieure a 3/60 et comprise entre 6/24 et 3/60, respectivement. On a examine la cornee des sujets avec une lampe a fente (manuelle) afin d'etablir une cotation de la kerato- pathie en gouttelettes conform6ment aux criteres admis. Pour la cataracte, le degre d'atteinte a 6te 6tabli a I'aide d'un ophtalmoscope, analysant le reflet pupillaire par observation du cristallin a 20 cm de distance. La pr6valence de la cecite et de la baisse de vision dans la population echantillonnee a ete de 1,5% (intervalle de confiance a 95%, 0,8- 2,3%) et de 8,1% (IC a 95%, 5,5-10,7%), respec- tivement. Le glaucome (34,8%) et la cataracte non operee (33,3%) sont des causes importantes de cecite. Avec 8,7%, les atteintes corn6ennes constituent la troisieme cause de cecite, les k6ra- topathies en gouttelettes representant 7,2% de ces dernieres. Les causes de baisse de vision sont la cataracte (45,1 %) et les atteintes cor- neennes (24,2%, dont 19,3% de keratopathies en gouttelettes). Un traumatisme est a l'origine de 41% des cas de cecite monoculaire. On n'a observe aucun cas de trachome, de xerophtalmie ni de lepre. La keratopathie en gouttelettes est commune, alors que la cataracte est moins frequente qu'on ne le pensait. Les eleveurs sont, du fait de leur mode de vie, expos6s au rayonnement intense du soleil, reflete par la neige jusqu'a 6 mois par an. II y a 260 jours de soleil par an en Mongolie et le climat y est sec. On pense que la forte exposition potentielle aux UV explique le taux eleve de cas de keratopathie en gouttelettes. Si l'on projette le r6sultat de cette petite enquete chez les 40 ans et plus au plan national, il apparait qu'environ 22 000 personnes devraient etre operees de la cataracte et que quelque 50000 personnes ont besoin de soins ou de traitements oculaires. Avec les 80 ophtalmologistes et 60 infirmieres en oph- talmologie qualifiees dont il dispose, ce pays devrait etre en mesure de prendre en charge tous ces cas. References 1. Methods of assessment of avoidable blindness. Geneva, World Health Organization, 1980 (WHO Offset Publication No. 54). 2. Mehra V, Minassian DC. A rapid method of grading cataract in epidemiological studies and eye surveys. British journal of ophthalmology, 1988, 72: 801-803. 776 WHO Bulletin OMS. Vol 72 1994