- Email: [email protected]
Active trachoma in children aged three to nine years in rural communities in Ethiopia: prevalence, indicators and risk factors
Transactions of the Royal Society of Tropical Medicine and Hygiene (2005) 99, 120—127
Active trachoma in children aged three to nine years in rural communities in Ethiopia: prevalence, indicators and risk factors Phillippa Cumberlanda,∗, Girum Hailub, Jim Todda a b
London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK Future International Consulting Agency, Addis Ababa, Ethiopia
Received 5 November 2003 ; received in revised form 23 February 2004; accepted 3 March 2004 Available online 3 October 2004 KEYWORDS Trachoma; Prevalence; Risk factors; Ethiopia
Summary Trachoma is the most important infectious cause of blindness worldwide. In two rural populations in Ethiopia, a programme of preventative and treatment measures was initiated in May 2002. A baseline survey was conducted to evaluate the effect of this programme on the prevalence of active trachoma. A total of 1960 children aged 3—9 years, from 915 households in 40 communities, were examined for the presence of active trachoma. Demographic and household information was collected using questionnaires and household amenities and environmental conditions were observed. Overall, 72% of children had active trachoma. Twenty percent of children aged nine years had trachomatous scarring. In children, discharge in the eyes and flies on the eyes were significant indicators of trachoma (odds ratio [OR] = 3.0, 95% CI 1.94—4.55 and OR = 3.4, 95% CI 2.37—4.88, respectively). Frequency of washing children, a clean environment and hygienic disposal of excrement were significant preventative factors for active trachoma. Prevalence of active trachoma varies widely between and within districts. Risk factors comprise a mix of individual characteristics and household factors. This study demonstrates the need for broad target interventions and a flexible approach to the prevention of trachoma in high prevalence endemic rural populations. © 2004 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved.
1. Introduction * Corresponding author. Present address: Centre of Paediatric Epidemiology and Biostatistics, Institute of Child Health, 30 Guilford Street, London WC1N 1EH, UK. Tel.: +44 20 7813 8396; fax: +44 20 7905 2381. E-mail address: [email protected] (P. Cumberland).
Trachoma is the most important infectious cause of blindness worldwide. An estimated 150 million people have active disease and six million people are blind as a result of the infection (Thylefors et al., 1995). Blindness from trachoma is a result
0035-9203/$ — see front matter © 2004 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.trstmh.2004.03.011
Prevalence of trachoma in rural Ethiopia of repeated episodes of active trachoma, which usually occur in childhood (Grayston et al., 1985). A high prevalence of active trachoma has been reported in children in countries of the Sahel, East Africa, Ethiopia and India. In recent studies, active trachoma was seen in 39% of children aged less than 10 years in Mali (Schemann et al., 2002), in 36.5% of children aged 2—6 years in Menofiya, Egypt (Ezz-alArab et al., 2001) and 51% of children aged less than 10 years in Ethiopia (Bejiga and Alemayehu, 2001). Trachoma is a disease of poverty and tends to cluster within the poorest families (Mecaskey, 1998). Risk factors that have been implicated in the transmission of trachoma include the amount of water available to the household, poor hygiene and the presence of flies (Marx, 1989). In 1997, WHO launched the Global Alliance for the Elimination of Trachoma (GET) for the control of trachoma based on the ‘SAFE’ strategy (WHO, 1997). The strategy has four components: Surgery, Antibiotics, Face washing and Environmental sanitation. The public health challenge is in the delivery of the SAFE strategy to rural communities, often with inadequate water supply, to reduce the prevalence of active trachoma. Since the introduction of the SAFE initiative there have been encouraging signs in the uptake of the programme (WHO, 2000). The International Trachoma Initiative (ITI) has a global intervention programme which includes improved trichiasis surgery, mass administration of azithromycin and the distribution of public health messages on trachoma disease and prevention using radio, video and printed materials. Evaluation of the effect of the programme on the prevalence of trachoma is important (Bailey and Lietman, 2001). This evaluation includes the process indicators for the initiative, changes in behaviour associated with the disease, and trachoma infection. This paper reports on the prevalence of trachoma in 40 communities in rural Ethiopia prior to the implementation of the SAFE initiative. This study provides the baseline prevalence of trachoma and identifies factors associated with trachoma in children in these communities, which will be used to improve and inform the intervention.
2. Materials and methods This study was undertaken in three zones of Ethiopia: Gurage, Oromia and South Welo. Oromia and South Welo are adjacent zones about 300 km northeast of Addis Ababa, a highland area inhabited predominately by people of the Oromia
121 ethnic group. The Gurage ethnic group is predominant in the Gurage zone, a lowland area, about 160 km southwest of Addis Ababa. The survey was conducted in three districts where the SAFE intervention strategy was to be implemented, Enemore & Ener in the Gurage zone and Artuma Farsi and Dawa Chefa in the Oromia zone, and in two similar adjacent districts, Cheha in the Gurage zone and Kalu in South Welo zone. Thirty communities (kebeles) were selected by cluster randomization from within the intervention area and ten from the adjacent districts. A kebele consists of 5—10 villages and forms the basic administrative unit in Ethiopia with a population of between 2000 and 10 000 people. Villages were selected at random within each kebele for the survey. The survey took place from March to May 2002, before the planned start of SAFE intervention activities. This is the dry season in Ethiopia when trachoma prevalence is high. The survey team included interviewers and ophthalmic nurses taken from the local area, who were familiar with the local language. All interviewers had completed secondary education to the age of 18 years and were trained in delivering a standard questionnaire. The ophthalmic nurses, experienced in community eye health prevention and treatment according to Ethiopian guidelines, received additional training from a community ophthalmologist for the survey. In each kebele, 25—30 households were visited using survey methods similar to a study in Singida, Tanzania (Paxton, 2001). In each household, one adult carer was identified and asked to consent to a structured interview and an examination of all children aged 3—9 years in the household. The questionnaire included demographic details of both respondent and household, practical arrangements for washing, cooking and the care of livestock, and knowledge of health-related issues. The questionnaire was supplemented by observation of the household, including the toilet facilities and the presence of flies in and around the house. All children in the study were seen by an ophthalmic nurse. The nurse observed the presence of flies on the child’s eyes, and the cleanliness of the child using a standardized protocol. Discharge in the eyes and nose was noted and whether the child’s hands and hair were clean. Hair was considered unclean if the scalp was evidently unwashed. To examine the eye, the nurse inverted the upper eyelid of each child and scored for the presence of active trachoma and scarring sequelae using the simplified WHO criteria (Thylefors et al., 1987). In these criteria, follicular trachoma (TF) is identified if there are five or more follicles on the surface of the upper tarsal conjunctiva, and intense
72.0% (70—74%) a
Active trachoma infection defined as follicular trachoma or intense trachoma. 9 children with no information on type of trachoma.
77.1% (72—82%) 86.6% (84—89%) 50.9% (45—57%) 64.9% (60—69%)
62.2% (56—68%)
27—100% 68%—88% 73—100% 44—73% 27—70% 40—83%
102 (5.2%) 40 27 (11.1%) 5 64 (9.1%) 15 3 (1.0%) 6 6 (1.4%) 9
2 (0.8%) 5
1412 (36%) 192 (36%) 631 (36%) 149 (36%) 279 (36%)
161 (36%)
561 (36%) 71 (36%) 214 (36%) 89 (36%) 124 (36%)
63 (36%)
249 121 (36%) 729 414 (36%) 259 97 (36%) 293 56 (36%) 430 154 (36%)
No. of children Follicular trachoma infection Intense trachoma inflammation Active trachoma infectiona Trachomatous scarring No. of communities in survey Range of trachoma prevalence Mean trachoma prevalence (95% CI)
Cheha Artuma Fusi
Dawa Chefa
Kalu
Enemore & Ener
Table 1
trachoma (TI) if there is inflammatory thickening of the upper tarsal conjunctiva that obscures more than half of the normal, deep conjunctival vessels. Trachomatous scarring (TS) is indicated by scarring in the upper tarsal conjunctiva. Signs were recorded as being present if they were seen in either or both eyes. All children with evidence of active follicular trachoma (TF and/or TI) were treated with 1% topical tetracycline eye ointment. The first dose was administered immediately and instructions were left with the responsible adult on its subsequent, twice daily administration for six weeks and advice was given on face washing. Questionnaire data were entered using EpiInfo, version 6 (CDC, Atlanta, GA, USA) and translated into Stata Statistical Software, version 7 (Stata Corp., College Station, TX, USA) for analysis. The primary outcome was the presence of active trachoma in children. Individual risk factors for active trachoma in children were compared using logistic regression taking into account the community survey structure (Hayes et al., 2000). Adult responses, based on household characteristics, were categorized and the prevalence of responses compared across villages using the χ2 test. A household with one or more children with active trachoma was defined as positive for active trachoma. Household risk factors for the presence of active trachoma were assessed using logistic regression. The study was approved by the ethics committee of the London School of Hygiene and Tropical Medicine, UK. Permission for the study was given by the Ministry of Health, Addis Ababa, Ethiopia.
Prevalence of trachoma among children, aged 3 to 9 years old, by district
Figure 1 Prevalence of follicular trachoma, intense trachoma and trachomatous scarring by age among study children in Ethiopia.
1960 842 (36%)
P. Cumberland et al.
Total
122
Prevalence of trachoma in rural Ethiopia
123
3. Results 3.1. Characteristics of respondents and their households Across the five districts (woreda), 915 households were visited, 872 (95%) of whom had children aged 3—9 years in the household. The majority, 748 (82%), of the respondents were aged 25—49 years, with 400 (44%) being the head of household and 462 (50%) the spouse of the head of household. Education of head of household and ethnic group of the respondents reflected the rural setting, with 271 (77%) in Oromia zone, 74 (70%) in the South Welo zone and 256 (56%) in Gurage zone being unable to read and write. Sixty-six percent of respondents were subsistence farmers and 25% housewives.
3.2. Prevalence of trachoma in children In the 915 households there were 1987 eligible children, aged 3—9 years, of whom 1960 (98.6%) had parental consent for examination by the survey team. The median number of children per household was two (range 1—7), with 93% of households having three or fewer children. Overall, 842 (43%) of the children were noted to have TF and 561 (28%)
to have TI, but this varied by district (Table 1). The presence of TS was recorded for 102 (5.2%) of the children, 24 of whom had TF, 46 had TI and 32 of whom had no evidence of active trachoma. The prevalence of trachoma by age is shown in Figure 1. The prevalence of trachoma in the 40 kebeles varied from 27 to 100%, with a mean prevalence of 72% (95% CI 70—74%). Factors associated with cleanliness of the children were associated with active trachoma (Table 2). Clean hair and faces were found to be protective. Discharge in the eyes and flies on the eyes were both significant indicators of trachoma, odds ratio (OR) = 3.0 (95% CI 1.94—4.55) and OR = 3.4 (95% CI 2.37—4.88) respectively, after adjustment for age and clean hair. The joint effect of these indicators showed the effect of discharge in the eyes and flies on the eyes to be multiplicative, OR = 8.3 (4.94—13.90).
3.3. Household characteristics and risks for trachoma Access to water differed across the five districts (Table 3). In Kalu district 71 (67%) of households had access to tap water, with all of the remainder having access to other protected water sources. In Enemore & Ener district 238 (70%) of households
Table 2 Association between active trachoma (follicular trachoma and intense trachoma) and observed characteristics of study children in rural Ethiopia Factor Age (years) 3 4 5 6 7 8 9
No. of children (n = 1960) 359 271 330 252 267 214 267
No. with trachoma (%) (n = 1412) (72) 310 216 260 195 184 116 131
(86) (80) (79) (77) (69) (54) (49)
Univariate odds ratio (95% CI)a 1 0.62 0.59 0.54 0.35 0.19 0.15
(0.38—1.02) (0.38—0.90) (0.34—0.85) (0.23—0.52) (0.12—0.30) (0.09—0.27)
Multivariate odds ratio (95% CI)b 1 0.75 0.74 0.70 0.47 0.32 0.22
(0.44—1.26) (0.48—1.15) (0.43—1.14) (0.30—0.72) (0.19—0.56) (0.12—0.40)
Discharge from nose Yes No
700 1258
555 (79) 855 (68)
1 1.80 (1.29—2.52)
Clean hair Yes No
1090 867
700 (64) 709 (82)
1 2.50 (1.66—3.76)
1 1.72 (1.16—2.56)
1 3.3 (2.13—5.15) 4.3 (2.99—6.04) 12.3 (7.19—20.95)
1 3.0 (1.94—4.55) 3.40 (2.37—4.88) 8.30 (4.94—13.90)
Discharge from eyes and flies on eyes No discharge or flies 627 Discharge but no flies 103 Flies but no discharge 682 Discharge and flies 539 a b
296 77 540 495
(47) (75) (79) (92)
Adjusted for kebele. Adjusted for kebele, age, clean hair, discharge from eyes and flies on eyes.
124
Table 3
P. Cumberland et al.
Characteristics of adults and households in five study districts in rural Ethiopia District Artuma Fusi
No. of households Access to watera Protected water Unprotected water River, stream, pond
210
Dawa Chefa 144
Kalu
Enemore & Ener
106
342
Cheha
Total
113
915
125 (60) 49 (24) 34 (16)
84 (58) 12 (8) 48 (33)
106 (100)
92 (27) 12 (3.5) 238 (70)
75 (66) 3 (3) 35 (31)
482 (53) 76 (8) 355 (39)
Time to access water source (min)b <10 22 (11) 10—30 99 (47) >30 83 (40)
7 (5) 76 (53) 61 (42)
32 (30) 73 (69) 1 (1)
24 (7) 128 (38) 186 (54)
33 (29) 44 (39) 36 (31)
118 (13) 420 (46) 367 (40)
152 (73)
101 (70)
94 (89)
148 (43)
26 (23)
521 (57)
11 (6) 176 (94)
18 (14) 113 (86)
13 (12) 92 (88)
45 (14) 80 (25) 198 (61)
37 (34) 34 (31) 37 (34)
124 (14) 495 (54) 235 (26)
Proximity to overnight cow pensd In house 8 (4) Outside of house 121 (58) No animals 80 (38)
5 (4) 89 (62) 49 (34)
14 (13) 68 (64) 24 (23)
303 (89) 16 (5) 23 (7)
100 (89) 2 (2) 11 (10)
430 (47) 296 (32) 187 (20)
20 (14) 117 (81) 6 (4)
3 (3) 101 (95) 1 (1)
0 (0) 43 (13) 293 (86)
3 (3) 45 (40) 64 (57)
74 (8) 454 (50) 372 (41)
Adequate water c
Defecation site Pit latrine Farmland/away from house Back garden/close to house
Fliese No flies Few flies Many flies
48 (23) 148 (71) 8 (4)
Values are numbers and percentage. a Data missing for two households in Artuma Fusi. b Data missing for six households in Artuma Fusi and four in Enemore & Ener. c Data missing for 23 households in Artuma Fusi, 13 in Dawa Chefa, one in Kalu, 19 in Enemore & Ener, and five in Cheha. d Data missing for one household in Artuma Fusi and Dawa Chefa. e Data missing for six households in Artuma Fusi and Enemore & Ener, and one in Dawa Chefa, Kalu and Cheha.
obtained their water from a river, with only 92 (27%) of households having access to protected water. The time taken to access water differed across the five districts (Table 3) with a third of households over 30 minutes from water taking at least an hour to reach their primary water source. Overall, 161 (18%) of households were observed to have a pit latrine, but 37 (23%) of these contained no faeces and the enclosure was clean, so were probably not used. In the two districts in Gurage zone, the majority of households used a defecation site in the garden, close to the house (Table 3). Association between household factors potentially associated with active trachoma was estimated. The likelihood of a household containing a child with active trachoma increased three-fold with each additional child in the household. Other social factors were not significantly associated, but factors relating to cleanliness and hygiene were (Table 4), as was the respondent’s knowledge of tra-
choma. Flies around the house were associated with trachoma, with houses surrounded by most flies being nine times more likely to have a child with active trachoma. Disposal of waste, presence of cows and defecation site were all highly correlated with the presence of flies in and around the house. The flies are present largely as a consequence of these other factors so this variable was not included in the multivariate model. Frequent washing of children, a clean environment and hygienic disposal of excrement were significant preventative factors for active trachoma.
4. Discussion This survey reports the prevalence of trachoma by year of age for children aged 3—9 years. The prevalence of TF and TI in these communities is 86% in children aged three years, and it then
Prevalence of trachoma in rural Ethiopia
125
Table 4 Association between active trachoma (follicular trachoma and intense trachoma) in at least one child in a household and reported household characteristics in rural Ethiopia Factor
No. of households
No. with trachoma (%)
Univariate odds ratio (95% CI)a
Multivariate odds ratio (95% CI)b
No. of children 1 2 3 ≥4
170 392 248 62
121 352 236 61
2.99 (2.06—4.35)c
3.37 (2.38—4.76)c
Children washed Weekly Daily
372 383
350 (94) 314 (82)
1 0.29 (0.17—0.49)
1 0.41 (0.18—0.96)
Time to water source (min) <10 10—29 ≥30
42 476 345
35 (83) 410 (86) 317 (92)
1 1.20 (0.58—2.48) 2.20 (0.98—4.93)
Quality of water source Unprotected Protected
398 472
363 (91) 405 (86)
1 0.58 (0.32—1.05)
Defecation site Pit latrine Away from house Near house
116 481 218
95 (82) 417 (87) 212 (97)
1 1.44 (0.78—2.67) 7.81 (2.66—22.1)
Cows None/near or around house In house
469 401
384 (82) 384 (96)
1 5.00 (2.93—8.54)
Animal faeces near house No Yes
222 640
183 (82) 579 (91)
1 2.12 (1.39—3.25)
Flies None Few Many
73 439 348
54 (74) 370 (84) 335 (96)
1 1.89 (1.24—2.87) 9.07 (4.12—19.95)
Disposal of waste Burn/bury Away from house Near house
177 111 582
136 (76) 98 (88) 535 (92)
1 2.35 (1.14—4.77) 3.63 (2.06—6.39)
Respondents’ knowledge of trachoma No 222 Yes 368 Not sure 266
208 (94) 331 (90) 216 (81)
1 0.54 (0.30—0.98) 0.27 (0.14—0.53)
a b c
(71) (90) (95) (98)
1 2.03 (0.96—4.32) 5.49 (1.73—17.48)
1 1.75 (1.07—2.85)
Allowing for kebele. Allowing for kebele, and adjusting for frequency of washing, disposal of waste and animal faeces. No. of children used as a continuous variable.
declines with age. There is also evidence of TS in 20% of children aged 9 years, indicating repeated episodes of active trachoma during childhood. Previous studies in Ethiopia have also shown a high prevalence of trachoma in children aged less than 10 years (Bejiga and Alemayehu, 2001; Zerihun, 1997), which suggests that in some areas of Ethiopia every child has trachoma infection
at some time in their childhood (Melese et al., 2003). The survey relied on the clinical examination of children for evidence of trachoma using WHO guidelines (Thylefors et al., 1987). The ophthalmic nurses who carried out this examination are experienced in the detection and treatment of trachoma in the health facilities. There were two teams of
126 nurses to cover the two separate study areas, as there were different languages spoken in the two zones, so there may have been variation between teams, although each team received the same training. Risk factors for active trachoma comprise a mix of individual characteristics and household factors (Hsieh et al., 2000; Schemann et al., 2002). In our study there was a significant reduction in the prevalence of trachoma among children who had clean hair, no nasal discharge, no eye discharge and did not have flies on their face. Previous studies in Tanzania have shown the risk of trachoma is reduced by facial cleanliness and personal hygiene, and the risk can be reduced by face washing (Taylor et al., 1989; West et al., 1991, 1995). The presence of human and animal faeces is associated with flies and an increased risk of active trachoma in children. Children had a 2.5 times higher risk of active trachoma if they lived in households that reported defecation close to the house, or kept cows close to the house. Almost 44% of households were seen with large numbers of flies in the household, and this was associated with human and animal faeces near the house. Flies were associated with a five-fold increase in the risk of active trachoma, which is consistent with Emerson’s report of a four-fold decrease in trachoma transmission after fly control in Gambian villages (Emerson et al., 1999). The conditions conducive to trachoma transmission are complex and therefore there are many potential routes to the interruption of transmission. The rural population of Ethiopia is heterogeneous as regards to possible associations with trachoma disease and prevention. Prevalence varies across communities in the Oromia zone from 27 to 83% and from 73 to 100% in the Gurage zone. Gurage zone being a lowland area would be expected to have a higher prevalence but both areas have a high level of endemic trachoma. We showed that prevalence of trachoma is associated with cleanliness, hygiene and awareness of knowledge about trachoma. In the Oromia area, 68% have access to clean water, 9% have pit latrines, 76% are unable to read and write and only 16% have access to a radio. In Gurage area, 37% have access to clean water, 18% have pit latrines and 52% use a defecation site in close proximity to the house, 56% are unable to read and write and 52% have access to a radio. This indicates that historically the focus for use of resources has varied. As a result of this, the two areas have different strengths and weaknesses in terms of infrastructure that will support trachoma prevention. This highlights the need for flexibility
P. Cumberland et al. of approach to interventions across communities at risk, building on the existing strengths of each area, with all aspects of the SAFE strategy being considered. Fly control strategies should be considered in conjunction with relocation of defecation sites and changes in behaviour regarding disposal of rubbish and the location of animals at night. Further improvement in basic water and sanitation amenities are important measures against all infectious disease. The effectiveness of mass administration of antibiotics as a tool in trachoma elimination needs further investigation in high prevalence areas (Melese et al., 2003). This survey has demonstrated the need for a broad target approach in high prevalence rural populations and further surveys will evaluate mass interventions, both medical and educational, in the community. The impact of these on prevalence of trachoma in Ethiopia is important. The evaluation of the impact of improved knowledge of disease and prevention, through mass media, will have implications for the prevention of trachoma and other infectious diseases in many rural populations, globally. Conflicts of interest statement The authors have no conflicts of interest concerning the work reported in this paper.
Acknowledgements We would like to thank the Minister of Health in Ethiopia for allowing this study to be conducted, and the regional health teams in assisting the survey work. We would also thank the Zonal Health Departments in Gurage, Oromia and South Welo for their co-operation and help, along with the work being done by ORBIS International, World Vision and BBC World Service Trust to prevent trachoma in these zones. The funding for this study was provided by the International Trachoma Initiative to whom we are grateful, and we acknowledge the coordination of ITI work in Ethiopia by Ato Assefa Cherinet. We thank the project staff and survey teams for their dedication and hard work, especially the Dr Fhami Mohammed, Dr Workayehu Kebede, Dr Amir Bedri and Prof. Fikru. We acknowledge the patience of the people in the communities who took part in the survey, and the children who were examined for trachoma infection. Finally, thanks to Dr Allen Foster and Dr Anthony Solomon for their advice and guidance on the study and this paper.
Prevalence of trachoma in rural Ethiopia
References Bailey, R., Lietman, T., 2001. The SAFE strategy for the elimination of trachoma by 2020: will it work? Bull. World Health Organ. 79, 233—236. Bejiga, A., Alemayehu, W., 2001. Prevalence of trachoma and its determinants in Dalocha District. Central Ethiopia. Ophthalmic Epidemiol. 8, 119—125. Emerson, P.M., Lindsay, S.W., Walraven, G.E.L., Faal, H., Bogh, C., Lowe, K., Bailey, R.L., 1999. Effect of fly control on trachoma and diarrhoea. Lancet 353, 1401— 1403. Ezz-al-Arab, G., Tawfik, N., El-Gendy, R., Anwar, W., Courtright, P., 2001. The burden of trachoma in the rural Nile Delta of Egypt: a survey of Menofiya governate. Br. J. Ophthalmol. 85, 1406—1410. Grayston, J.T., Wang, S.P., Yeh, L.J., Kuo, C.C., 1985. Importance of reinfection in the pathogenesis of trachoma. Rev. Infect. Dis. 7, 717—725. Hayes, R.J., Alexander, N.D.E., Bennett, S., Cousens, S.N., 2000. Design and analysis issues in cluster-randomized trials of interventions against infectious diseases. Stat. Methods Med. Res. 9, 95—116. Hsieh, Y.H., Bobo, L.D., Quinn, T.O., West, S.K., 2000. Risk factors for trachoma: 6 year follow up of children aged 1 and 2 years. Am. J. Epidemiol. 152, 204—211. Marx, R., 1989. Social factors and trachoma: a review of the literature. Soc. Sci. Med. 29, 23—34. Mecaskey, J., 1998. Public Policy, Priority Setting and Trachoma. Third meeting of the WHO Alliance for the Global Elimination of Trachoma, Ouarzazate, Morocco, 19—20 October 1998. World Health Organization, Morocco [Online]. [Accessed 21 August 2004]. Available from the World Wide Web: http://whqlibdoc.who.int/hq/1999/WHO PBD GET 99.3.pdf.
127 Melese, M., Alemayehu, W., Gaynor, B., Yi, E., Whitcher, J.P., Lietman, T.M., 2003. What more is there to learn about trachoma? Br. J. Ophthalmol. 87, 521—522. Paxton, A., 2001. Rapid assessment of trachoma prevalence in Singida Tanzania. A study to compare assessment methods. Ophthalmic Epidemiol. 8, 87—96. Schemann, J.F., Sacko, D., Malvy, D., Momo, G., Traore, L., Bore, O., Coulibaly, S., Banou, A., 2002. Risk factors for trachoma in Mali. Int. J. Epidemiol. 31, 194—201. Taylor, H., West, S.K., Mmbaga, B.B.O., Katala, S.J., Turner, V., Lynch, M., 1989. Hygiene factors and increased risk of trachoma in families. Arch. Ophthalmol. 107, 1821—1825. Thylefors, B., Dawson, C.R., Jones, B.R., West, S.K., Taylor, H.R., 1987. A simple system for the assessment of trachoma and its complications. Bull. World Health Organ. 65, 477—483. Thylefors, B., Negrel, A.D., Pararajasegraram, R., Dadzie, K.Y., 1995. Global data on blindness. Bull. World Health Organ. 73, 115—121. West, S., Congdon, N., Katala, S., Mele, L., 1991. Facial cleanliness and risk of trachoma in families. Arch. Ophthalmol. 109, 855—857. West, S.K., Munoz, B., Lynch, M., Kayongoya, A., Chilangwa, Z., Mmbaga, B.B., Taylor, H.R., 1995. Impact of face-washing on trachoma in Kongwa, Tanzania. Lancet 345, 155—158. WHO, 1997. Report of the First Meeting of the WHO Alliance for the Global Elimination of Trachoma. World Health Organization, Geneva [Online]. [Accessed 21 August 2004]. Available from the World Wide Web: http://wholibdoc.who.int/hq/1997/WHO/PBL/GET/97.1.pdf. WHO, 2000. Report of the Fourth Meeting of the WHO Alliance for the Global Elimination of Trachoma.. World Health Organization, Geneva. Zerihun, N., 1997. Trachoma in Jimma zone, south-western Ethiopia. Trop. Med. Int. Health 2, 1115—1121.
Active trachoma in children aged three to nine years in rural communities in Ethiopia: prevalence, indicators and risk factors Phillippa Cumberlanda,∗, Girum Hailub, Jim Todda a b
London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK Future International Consulting Agency, Addis Ababa, Ethiopia
Received 5 November 2003 ; received in revised form 23 February 2004; accepted 3 March 2004 Available online 3 October 2004 KEYWORDS Trachoma; Prevalence; Risk factors; Ethiopia
Summary Trachoma is the most important infectious cause of blindness worldwide. In two rural populations in Ethiopia, a programme of preventative and treatment measures was initiated in May 2002. A baseline survey was conducted to evaluate the effect of this programme on the prevalence of active trachoma. A total of 1960 children aged 3—9 years, from 915 households in 40 communities, were examined for the presence of active trachoma. Demographic and household information was collected using questionnaires and household amenities and environmental conditions were observed. Overall, 72% of children had active trachoma. Twenty percent of children aged nine years had trachomatous scarring. In children, discharge in the eyes and flies on the eyes were significant indicators of trachoma (odds ratio [OR] = 3.0, 95% CI 1.94—4.55 and OR = 3.4, 95% CI 2.37—4.88, respectively). Frequency of washing children, a clean environment and hygienic disposal of excrement were significant preventative factors for active trachoma. Prevalence of active trachoma varies widely between and within districts. Risk factors comprise a mix of individual characteristics and household factors. This study demonstrates the need for broad target interventions and a flexible approach to the prevention of trachoma in high prevalence endemic rural populations. © 2004 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved.
1. Introduction * Corresponding author. Present address: Centre of Paediatric Epidemiology and Biostatistics, Institute of Child Health, 30 Guilford Street, London WC1N 1EH, UK. Tel.: +44 20 7813 8396; fax: +44 20 7905 2381. E-mail address: [email protected] (P. Cumberland).
Trachoma is the most important infectious cause of blindness worldwide. An estimated 150 million people have active disease and six million people are blind as a result of the infection (Thylefors et al., 1995). Blindness from trachoma is a result
0035-9203/$ — see front matter © 2004 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.trstmh.2004.03.011
Prevalence of trachoma in rural Ethiopia of repeated episodes of active trachoma, which usually occur in childhood (Grayston et al., 1985). A high prevalence of active trachoma has been reported in children in countries of the Sahel, East Africa, Ethiopia and India. In recent studies, active trachoma was seen in 39% of children aged less than 10 years in Mali (Schemann et al., 2002), in 36.5% of children aged 2—6 years in Menofiya, Egypt (Ezz-alArab et al., 2001) and 51% of children aged less than 10 years in Ethiopia (Bejiga and Alemayehu, 2001). Trachoma is a disease of poverty and tends to cluster within the poorest families (Mecaskey, 1998). Risk factors that have been implicated in the transmission of trachoma include the amount of water available to the household, poor hygiene and the presence of flies (Marx, 1989). In 1997, WHO launched the Global Alliance for the Elimination of Trachoma (GET) for the control of trachoma based on the ‘SAFE’ strategy (WHO, 1997). The strategy has four components: Surgery, Antibiotics, Face washing and Environmental sanitation. The public health challenge is in the delivery of the SAFE strategy to rural communities, often with inadequate water supply, to reduce the prevalence of active trachoma. Since the introduction of the SAFE initiative there have been encouraging signs in the uptake of the programme (WHO, 2000). The International Trachoma Initiative (ITI) has a global intervention programme which includes improved trichiasis surgery, mass administration of azithromycin and the distribution of public health messages on trachoma disease and prevention using radio, video and printed materials. Evaluation of the effect of the programme on the prevalence of trachoma is important (Bailey and Lietman, 2001). This evaluation includes the process indicators for the initiative, changes in behaviour associated with the disease, and trachoma infection. This paper reports on the prevalence of trachoma in 40 communities in rural Ethiopia prior to the implementation of the SAFE initiative. This study provides the baseline prevalence of trachoma and identifies factors associated with trachoma in children in these communities, which will be used to improve and inform the intervention.
2. Materials and methods This study was undertaken in three zones of Ethiopia: Gurage, Oromia and South Welo. Oromia and South Welo are adjacent zones about 300 km northeast of Addis Ababa, a highland area inhabited predominately by people of the Oromia
121 ethnic group. The Gurage ethnic group is predominant in the Gurage zone, a lowland area, about 160 km southwest of Addis Ababa. The survey was conducted in three districts where the SAFE intervention strategy was to be implemented, Enemore & Ener in the Gurage zone and Artuma Farsi and Dawa Chefa in the Oromia zone, and in two similar adjacent districts, Cheha in the Gurage zone and Kalu in South Welo zone. Thirty communities (kebeles) were selected by cluster randomization from within the intervention area and ten from the adjacent districts. A kebele consists of 5—10 villages and forms the basic administrative unit in Ethiopia with a population of between 2000 and 10 000 people. Villages were selected at random within each kebele for the survey. The survey took place from March to May 2002, before the planned start of SAFE intervention activities. This is the dry season in Ethiopia when trachoma prevalence is high. The survey team included interviewers and ophthalmic nurses taken from the local area, who were familiar with the local language. All interviewers had completed secondary education to the age of 18 years and were trained in delivering a standard questionnaire. The ophthalmic nurses, experienced in community eye health prevention and treatment according to Ethiopian guidelines, received additional training from a community ophthalmologist for the survey. In each kebele, 25—30 households were visited using survey methods similar to a study in Singida, Tanzania (Paxton, 2001). In each household, one adult carer was identified and asked to consent to a structured interview and an examination of all children aged 3—9 years in the household. The questionnaire included demographic details of both respondent and household, practical arrangements for washing, cooking and the care of livestock, and knowledge of health-related issues. The questionnaire was supplemented by observation of the household, including the toilet facilities and the presence of flies in and around the house. All children in the study were seen by an ophthalmic nurse. The nurse observed the presence of flies on the child’s eyes, and the cleanliness of the child using a standardized protocol. Discharge in the eyes and nose was noted and whether the child’s hands and hair were clean. Hair was considered unclean if the scalp was evidently unwashed. To examine the eye, the nurse inverted the upper eyelid of each child and scored for the presence of active trachoma and scarring sequelae using the simplified WHO criteria (Thylefors et al., 1987). In these criteria, follicular trachoma (TF) is identified if there are five or more follicles on the surface of the upper tarsal conjunctiva, and intense
72.0% (70—74%) a
Active trachoma infection defined as follicular trachoma or intense trachoma. 9 children with no information on type of trachoma.
77.1% (72—82%) 86.6% (84—89%) 50.9% (45—57%) 64.9% (60—69%)
62.2% (56—68%)
27—100% 68%—88% 73—100% 44—73% 27—70% 40—83%
102 (5.2%) 40 27 (11.1%) 5 64 (9.1%) 15 3 (1.0%) 6 6 (1.4%) 9
2 (0.8%) 5
1412 (36%) 192 (36%) 631 (36%) 149 (36%) 279 (36%)
161 (36%)
561 (36%) 71 (36%) 214 (36%) 89 (36%) 124 (36%)
63 (36%)
249 121 (36%) 729 414 (36%) 259 97 (36%) 293 56 (36%) 430 154 (36%)
No. of children Follicular trachoma infection Intense trachoma inflammation Active trachoma infectiona Trachomatous scarring No. of communities in survey Range of trachoma prevalence Mean trachoma prevalence (95% CI)
Cheha Artuma Fusi
Dawa Chefa
Kalu
Enemore & Ener
Table 1
trachoma (TI) if there is inflammatory thickening of the upper tarsal conjunctiva that obscures more than half of the normal, deep conjunctival vessels. Trachomatous scarring (TS) is indicated by scarring in the upper tarsal conjunctiva. Signs were recorded as being present if they were seen in either or both eyes. All children with evidence of active follicular trachoma (TF and/or TI) were treated with 1% topical tetracycline eye ointment. The first dose was administered immediately and instructions were left with the responsible adult on its subsequent, twice daily administration for six weeks and advice was given on face washing. Questionnaire data were entered using EpiInfo, version 6 (CDC, Atlanta, GA, USA) and translated into Stata Statistical Software, version 7 (Stata Corp., College Station, TX, USA) for analysis. The primary outcome was the presence of active trachoma in children. Individual risk factors for active trachoma in children were compared using logistic regression taking into account the community survey structure (Hayes et al., 2000). Adult responses, based on household characteristics, were categorized and the prevalence of responses compared across villages using the χ2 test. A household with one or more children with active trachoma was defined as positive for active trachoma. Household risk factors for the presence of active trachoma were assessed using logistic regression. The study was approved by the ethics committee of the London School of Hygiene and Tropical Medicine, UK. Permission for the study was given by the Ministry of Health, Addis Ababa, Ethiopia.
Prevalence of trachoma among children, aged 3 to 9 years old, by district
Figure 1 Prevalence of follicular trachoma, intense trachoma and trachomatous scarring by age among study children in Ethiopia.
1960 842 (36%)
P. Cumberland et al.
Total
122
Prevalence of trachoma in rural Ethiopia
123
3. Results 3.1. Characteristics of respondents and their households Across the five districts (woreda), 915 households were visited, 872 (95%) of whom had children aged 3—9 years in the household. The majority, 748 (82%), of the respondents were aged 25—49 years, with 400 (44%) being the head of household and 462 (50%) the spouse of the head of household. Education of head of household and ethnic group of the respondents reflected the rural setting, with 271 (77%) in Oromia zone, 74 (70%) in the South Welo zone and 256 (56%) in Gurage zone being unable to read and write. Sixty-six percent of respondents were subsistence farmers and 25% housewives.
3.2. Prevalence of trachoma in children In the 915 households there were 1987 eligible children, aged 3—9 years, of whom 1960 (98.6%) had parental consent for examination by the survey team. The median number of children per household was two (range 1—7), with 93% of households having three or fewer children. Overall, 842 (43%) of the children were noted to have TF and 561 (28%)
to have TI, but this varied by district (Table 1). The presence of TS was recorded for 102 (5.2%) of the children, 24 of whom had TF, 46 had TI and 32 of whom had no evidence of active trachoma. The prevalence of trachoma by age is shown in Figure 1. The prevalence of trachoma in the 40 kebeles varied from 27 to 100%, with a mean prevalence of 72% (95% CI 70—74%). Factors associated with cleanliness of the children were associated with active trachoma (Table 2). Clean hair and faces were found to be protective. Discharge in the eyes and flies on the eyes were both significant indicators of trachoma, odds ratio (OR) = 3.0 (95% CI 1.94—4.55) and OR = 3.4 (95% CI 2.37—4.88) respectively, after adjustment for age and clean hair. The joint effect of these indicators showed the effect of discharge in the eyes and flies on the eyes to be multiplicative, OR = 8.3 (4.94—13.90).
3.3. Household characteristics and risks for trachoma Access to water differed across the five districts (Table 3). In Kalu district 71 (67%) of households had access to tap water, with all of the remainder having access to other protected water sources. In Enemore & Ener district 238 (70%) of households
Table 2 Association between active trachoma (follicular trachoma and intense trachoma) and observed characteristics of study children in rural Ethiopia Factor Age (years) 3 4 5 6 7 8 9
No. of children (n = 1960) 359 271 330 252 267 214 267
No. with trachoma (%) (n = 1412) (72) 310 216 260 195 184 116 131
(86) (80) (79) (77) (69) (54) (49)
Univariate odds ratio (95% CI)a 1 0.62 0.59 0.54 0.35 0.19 0.15
(0.38—1.02) (0.38—0.90) (0.34—0.85) (0.23—0.52) (0.12—0.30) (0.09—0.27)
Multivariate odds ratio (95% CI)b 1 0.75 0.74 0.70 0.47 0.32 0.22
(0.44—1.26) (0.48—1.15) (0.43—1.14) (0.30—0.72) (0.19—0.56) (0.12—0.40)
Discharge from nose Yes No
700 1258
555 (79) 855 (68)
1 1.80 (1.29—2.52)
Clean hair Yes No
1090 867
700 (64) 709 (82)
1 2.50 (1.66—3.76)
1 1.72 (1.16—2.56)
1 3.3 (2.13—5.15) 4.3 (2.99—6.04) 12.3 (7.19—20.95)
1 3.0 (1.94—4.55) 3.40 (2.37—4.88) 8.30 (4.94—13.90)
Discharge from eyes and flies on eyes No discharge or flies 627 Discharge but no flies 103 Flies but no discharge 682 Discharge and flies 539 a b
296 77 540 495
(47) (75) (79) (92)
Adjusted for kebele. Adjusted for kebele, age, clean hair, discharge from eyes and flies on eyes.
124
Table 3
P. Cumberland et al.
Characteristics of adults and households in five study districts in rural Ethiopia District Artuma Fusi
No. of households Access to watera Protected water Unprotected water River, stream, pond
210
Dawa Chefa 144
Kalu
Enemore & Ener
106
342
Cheha
Total
113
915
125 (60) 49 (24) 34 (16)
84 (58) 12 (8) 48 (33)
106 (100)
92 (27) 12 (3.5) 238 (70)
75 (66) 3 (3) 35 (31)
482 (53) 76 (8) 355 (39)
Time to access water source (min)b <10 22 (11) 10—30 99 (47) >30 83 (40)
7 (5) 76 (53) 61 (42)
32 (30) 73 (69) 1 (1)
24 (7) 128 (38) 186 (54)
33 (29) 44 (39) 36 (31)
118 (13) 420 (46) 367 (40)
152 (73)
101 (70)
94 (89)
148 (43)
26 (23)
521 (57)
11 (6) 176 (94)
18 (14) 113 (86)
13 (12) 92 (88)
45 (14) 80 (25) 198 (61)
37 (34) 34 (31) 37 (34)
124 (14) 495 (54) 235 (26)
Proximity to overnight cow pensd In house 8 (4) Outside of house 121 (58) No animals 80 (38)
5 (4) 89 (62) 49 (34)
14 (13) 68 (64) 24 (23)
303 (89) 16 (5) 23 (7)
100 (89) 2 (2) 11 (10)
430 (47) 296 (32) 187 (20)
20 (14) 117 (81) 6 (4)
3 (3) 101 (95) 1 (1)
0 (0) 43 (13) 293 (86)
3 (3) 45 (40) 64 (57)
74 (8) 454 (50) 372 (41)
Adequate water c
Defecation site Pit latrine Farmland/away from house Back garden/close to house
Fliese No flies Few flies Many flies
48 (23) 148 (71) 8 (4)
Values are numbers and percentage. a Data missing for two households in Artuma Fusi. b Data missing for six households in Artuma Fusi and four in Enemore & Ener. c Data missing for 23 households in Artuma Fusi, 13 in Dawa Chefa, one in Kalu, 19 in Enemore & Ener, and five in Cheha. d Data missing for one household in Artuma Fusi and Dawa Chefa. e Data missing for six households in Artuma Fusi and Enemore & Ener, and one in Dawa Chefa, Kalu and Cheha.
obtained their water from a river, with only 92 (27%) of households having access to protected water. The time taken to access water differed across the five districts (Table 3) with a third of households over 30 minutes from water taking at least an hour to reach their primary water source. Overall, 161 (18%) of households were observed to have a pit latrine, but 37 (23%) of these contained no faeces and the enclosure was clean, so were probably not used. In the two districts in Gurage zone, the majority of households used a defecation site in the garden, close to the house (Table 3). Association between household factors potentially associated with active trachoma was estimated. The likelihood of a household containing a child with active trachoma increased three-fold with each additional child in the household. Other social factors were not significantly associated, but factors relating to cleanliness and hygiene were (Table 4), as was the respondent’s knowledge of tra-
choma. Flies around the house were associated with trachoma, with houses surrounded by most flies being nine times more likely to have a child with active trachoma. Disposal of waste, presence of cows and defecation site were all highly correlated with the presence of flies in and around the house. The flies are present largely as a consequence of these other factors so this variable was not included in the multivariate model. Frequent washing of children, a clean environment and hygienic disposal of excrement were significant preventative factors for active trachoma.
4. Discussion This survey reports the prevalence of trachoma by year of age for children aged 3—9 years. The prevalence of TF and TI in these communities is 86% in children aged three years, and it then
Prevalence of trachoma in rural Ethiopia
125
Table 4 Association between active trachoma (follicular trachoma and intense trachoma) in at least one child in a household and reported household characteristics in rural Ethiopia Factor
No. of households
No. with trachoma (%)
Univariate odds ratio (95% CI)a
Multivariate odds ratio (95% CI)b
No. of children 1 2 3 ≥4
170 392 248 62
121 352 236 61
2.99 (2.06—4.35)c
3.37 (2.38—4.76)c
Children washed Weekly Daily
372 383
350 (94) 314 (82)
1 0.29 (0.17—0.49)
1 0.41 (0.18—0.96)
Time to water source (min) <10 10—29 ≥30
42 476 345
35 (83) 410 (86) 317 (92)
1 1.20 (0.58—2.48) 2.20 (0.98—4.93)
Quality of water source Unprotected Protected
398 472
363 (91) 405 (86)
1 0.58 (0.32—1.05)
Defecation site Pit latrine Away from house Near house
116 481 218
95 (82) 417 (87) 212 (97)
1 1.44 (0.78—2.67) 7.81 (2.66—22.1)
Cows None/near or around house In house
469 401
384 (82) 384 (96)
1 5.00 (2.93—8.54)
Animal faeces near house No Yes
222 640
183 (82) 579 (91)
1 2.12 (1.39—3.25)
Flies None Few Many
73 439 348
54 (74) 370 (84) 335 (96)
1 1.89 (1.24—2.87) 9.07 (4.12—19.95)
Disposal of waste Burn/bury Away from house Near house
177 111 582
136 (76) 98 (88) 535 (92)
1 2.35 (1.14—4.77) 3.63 (2.06—6.39)
Respondents’ knowledge of trachoma No 222 Yes 368 Not sure 266
208 (94) 331 (90) 216 (81)
1 0.54 (0.30—0.98) 0.27 (0.14—0.53)
a b c
(71) (90) (95) (98)
1 2.03 (0.96—4.32) 5.49 (1.73—17.48)
1 1.75 (1.07—2.85)
Allowing for kebele. Allowing for kebele, and adjusting for frequency of washing, disposal of waste and animal faeces. No. of children used as a continuous variable.
declines with age. There is also evidence of TS in 20% of children aged 9 years, indicating repeated episodes of active trachoma during childhood. Previous studies in Ethiopia have also shown a high prevalence of trachoma in children aged less than 10 years (Bejiga and Alemayehu, 2001; Zerihun, 1997), which suggests that in some areas of Ethiopia every child has trachoma infection
at some time in their childhood (Melese et al., 2003). The survey relied on the clinical examination of children for evidence of trachoma using WHO guidelines (Thylefors et al., 1987). The ophthalmic nurses who carried out this examination are experienced in the detection and treatment of trachoma in the health facilities. There were two teams of
126 nurses to cover the two separate study areas, as there were different languages spoken in the two zones, so there may have been variation between teams, although each team received the same training. Risk factors for active trachoma comprise a mix of individual characteristics and household factors (Hsieh et al., 2000; Schemann et al., 2002). In our study there was a significant reduction in the prevalence of trachoma among children who had clean hair, no nasal discharge, no eye discharge and did not have flies on their face. Previous studies in Tanzania have shown the risk of trachoma is reduced by facial cleanliness and personal hygiene, and the risk can be reduced by face washing (Taylor et al., 1989; West et al., 1991, 1995). The presence of human and animal faeces is associated with flies and an increased risk of active trachoma in children. Children had a 2.5 times higher risk of active trachoma if they lived in households that reported defecation close to the house, or kept cows close to the house. Almost 44% of households were seen with large numbers of flies in the household, and this was associated with human and animal faeces near the house. Flies were associated with a five-fold increase in the risk of active trachoma, which is consistent with Emerson’s report of a four-fold decrease in trachoma transmission after fly control in Gambian villages (Emerson et al., 1999). The conditions conducive to trachoma transmission are complex and therefore there are many potential routes to the interruption of transmission. The rural population of Ethiopia is heterogeneous as regards to possible associations with trachoma disease and prevention. Prevalence varies across communities in the Oromia zone from 27 to 83% and from 73 to 100% in the Gurage zone. Gurage zone being a lowland area would be expected to have a higher prevalence but both areas have a high level of endemic trachoma. We showed that prevalence of trachoma is associated with cleanliness, hygiene and awareness of knowledge about trachoma. In the Oromia area, 68% have access to clean water, 9% have pit latrines, 76% are unable to read and write and only 16% have access to a radio. In Gurage area, 37% have access to clean water, 18% have pit latrines and 52% use a defecation site in close proximity to the house, 56% are unable to read and write and 52% have access to a radio. This indicates that historically the focus for use of resources has varied. As a result of this, the two areas have different strengths and weaknesses in terms of infrastructure that will support trachoma prevention. This highlights the need for flexibility
P. Cumberland et al. of approach to interventions across communities at risk, building on the existing strengths of each area, with all aspects of the SAFE strategy being considered. Fly control strategies should be considered in conjunction with relocation of defecation sites and changes in behaviour regarding disposal of rubbish and the location of animals at night. Further improvement in basic water and sanitation amenities are important measures against all infectious disease. The effectiveness of mass administration of antibiotics as a tool in trachoma elimination needs further investigation in high prevalence areas (Melese et al., 2003). This survey has demonstrated the need for a broad target approach in high prevalence rural populations and further surveys will evaluate mass interventions, both medical and educational, in the community. The impact of these on prevalence of trachoma in Ethiopia is important. The evaluation of the impact of improved knowledge of disease and prevention, through mass media, will have implications for the prevention of trachoma and other infectious diseases in many rural populations, globally. Conflicts of interest statement The authors have no conflicts of interest concerning the work reported in this paper.
Acknowledgements We would like to thank the Minister of Health in Ethiopia for allowing this study to be conducted, and the regional health teams in assisting the survey work. We would also thank the Zonal Health Departments in Gurage, Oromia and South Welo for their co-operation and help, along with the work being done by ORBIS International, World Vision and BBC World Service Trust to prevent trachoma in these zones. The funding for this study was provided by the International Trachoma Initiative to whom we are grateful, and we acknowledge the coordination of ITI work in Ethiopia by Ato Assefa Cherinet. We thank the project staff and survey teams for their dedication and hard work, especially the Dr Fhami Mohammed, Dr Workayehu Kebede, Dr Amir Bedri and Prof. Fikru. We acknowledge the patience of the people in the communities who took part in the survey, and the children who were examined for trachoma infection. Finally, thanks to Dr Allen Foster and Dr Anthony Solomon for their advice and guidance on the study and this paper.
Prevalence of trachoma in rural Ethiopia
References Bailey, R., Lietman, T., 2001. The SAFE strategy for the elimination of trachoma by 2020: will it work? Bull. World Health Organ. 79, 233—236. Bejiga, A., Alemayehu, W., 2001. Prevalence of trachoma and its determinants in Dalocha District. Central Ethiopia. Ophthalmic Epidemiol. 8, 119—125. Emerson, P.M., Lindsay, S.W., Walraven, G.E.L., Faal, H., Bogh, C., Lowe, K., Bailey, R.L., 1999. Effect of fly control on trachoma and diarrhoea. Lancet 353, 1401— 1403. Ezz-al-Arab, G., Tawfik, N., El-Gendy, R., Anwar, W., Courtright, P., 2001. The burden of trachoma in the rural Nile Delta of Egypt: a survey of Menofiya governate. Br. J. Ophthalmol. 85, 1406—1410. Grayston, J.T., Wang, S.P., Yeh, L.J., Kuo, C.C., 1985. Importance of reinfection in the pathogenesis of trachoma. Rev. Infect. Dis. 7, 717—725. Hayes, R.J., Alexander, N.D.E., Bennett, S., Cousens, S.N., 2000. Design and analysis issues in cluster-randomized trials of interventions against infectious diseases. Stat. Methods Med. Res. 9, 95—116. Hsieh, Y.H., Bobo, L.D., Quinn, T.O., West, S.K., 2000. Risk factors for trachoma: 6 year follow up of children aged 1 and 2 years. Am. J. Epidemiol. 152, 204—211. Marx, R., 1989. Social factors and trachoma: a review of the literature. Soc. Sci. Med. 29, 23—34. Mecaskey, J., 1998. Public Policy, Priority Setting and Trachoma. Third meeting of the WHO Alliance for the Global Elimination of Trachoma, Ouarzazate, Morocco, 19—20 October 1998. World Health Organization, Morocco [Online]. [Accessed 21 August 2004]. Available from the World Wide Web: http://whqlibdoc.who.int/hq/1999/WHO PBD GET 99.3.pdf.
127 Melese, M., Alemayehu, W., Gaynor, B., Yi, E., Whitcher, J.P., Lietman, T.M., 2003. What more is there to learn about trachoma? Br. J. Ophthalmol. 87, 521—522. Paxton, A., 2001. Rapid assessment of trachoma prevalence in Singida Tanzania. A study to compare assessment methods. Ophthalmic Epidemiol. 8, 87—96. Schemann, J.F., Sacko, D., Malvy, D., Momo, G., Traore, L., Bore, O., Coulibaly, S., Banou, A., 2002. Risk factors for trachoma in Mali. Int. J. Epidemiol. 31, 194—201. Taylor, H., West, S.K., Mmbaga, B.B.O., Katala, S.J., Turner, V., Lynch, M., 1989. Hygiene factors and increased risk of trachoma in families. Arch. Ophthalmol. 107, 1821—1825. Thylefors, B., Dawson, C.R., Jones, B.R., West, S.K., Taylor, H.R., 1987. A simple system for the assessment of trachoma and its complications. Bull. World Health Organ. 65, 477—483. Thylefors, B., Negrel, A.D., Pararajasegraram, R., Dadzie, K.Y., 1995. Global data on blindness. Bull. World Health Organ. 73, 115—121. West, S., Congdon, N., Katala, S., Mele, L., 1991. Facial cleanliness and risk of trachoma in families. Arch. Ophthalmol. 109, 855—857. West, S.K., Munoz, B., Lynch, M., Kayongoya, A., Chilangwa, Z., Mmbaga, B.B., Taylor, H.R., 1995. Impact of face-washing on trachoma in Kongwa, Tanzania. Lancet 345, 155—158. WHO, 1997. Report of the First Meeting of the WHO Alliance for the Global Elimination of Trachoma. World Health Organization, Geneva [Online]. [Accessed 21 August 2004]. Available from the World Wide Web: http://wholibdoc.who.int/hq/1997/WHO/PBL/GET/97.1.pdf. WHO, 2000. Report of the Fourth Meeting of the WHO Alliance for the Global Elimination of Trachoma.. World Health Organization, Geneva. Zerihun, N., 1997. Trachoma in Jimma zone, south-western Ethiopia. Trop. Med. Int. Health 2, 1115—1121.