Trachoma, caused by ocular infection with Chlamydia trachomatis, is hyperendemic on the Bijagós Archipelago of Guinea Bissau. An understanding of the risk factors associated with active trachoma and infection on these remote and isolated islands, which are atypical of trachoma-endemic environments described elsewhere, is crucial to the implementation of trachoma elimination strategies.
A cross-sectional population-based trachoma prevalence survey was conducted on four islands. We conducted a questionnaire-based risk factor survey, examined participants for trachoma using the World Health Organization (WHO) simplified grading system and collected conjunctival swab samples for 1507 participants from 293 randomly selected households. DNA extracted from conjunctival swabs was tested using the Roche Amplicor CT/NG PCR assay. The prevalence of active (follicular and/or inflammatory) trachoma was 11% (167/1508) overall and 22% (136/618) in 1–9 year olds. The prevalence of C. trachomatis infection was 18% overall and 25% in 1–9 year olds. There were strong independent associations of active trachoma with ocular and nasal discharge, C. trachomatis infection, young age, male gender and type of household water source. C. trachomatis infection was independently associated with young age, ocular discharge, type of household water source and the presence of flies around a latrine.
In this remote island environment, household-level risk factors relating to fly populations, hygiene behaviours and water usage are likely to be important in the transmission of ocular C. trachomatis infection and the prevalence of active trachoma. This may be important in the implementation of environmental measures in trachoma control.
Trachoma, caused by ocular infection with Chlamydia trachomatis, is the leading infectious cause of blindness worldwide. The World Health Organization elimination strategy includes community mass treatment with oral antibiotics, education regarding hygiene and facial cleanliness and environmental improvements. Population-based trachoma prevalence surveys are essential to determine whether community interventions are required. Knowledge of risk factors associated with trachoma and C. trachomatis infection in a particular setting may help prioritise trachoma elimination activities. We conducted a trachoma prevalence survey to establish the prevalence of active (follicular and/or inflammatory) trachoma and C. trachomatis infection on the Bijagós Archipelago of Guinea Bissau. We also collected household risk factor data from survey participants. Active trachoma prevalence was 11% overall and 22% in children aged 1–9 years. C. trachomatis infection prevalence was 18% overall and 25% in children aged 1–9 years. Active trachoma and the presence of C. trachomatis infection were strongly correlated. Risk factors for disease and infection were similar. In this environment, measures of facial cleanliness (ocular and nasal discharge) and household-level risk factors relating to fly populations, hygiene behaviours and water usage are likely to be important in C. trachomatis transmission. This may have implications in the implementation of trachoma elimination activities.
Citation: Last AR, Burr SE, Weiss HA, Harding-Esch EM, Cassama E, Nabicassa M, et al. (2014) Risk Factors for Active Trachoma and Ocular Chlamydia trachomatis Infection in Treatment-Naïve Trachoma-Hyperendemic Communities of the Bijagós Archipelago, Guinea Bissau. PLoS Negl Trop Dis 8(6): e2900. doi:10.1371/journal.pntd.0002900
Editor: Julius Schachter, University of California San Francisco, United States of America
Received: December 28, 2013; Accepted: April 11, 2014; Published: June 26, 2014
Copyright: © 2014 Last et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the Wellcome Trust [grant number 097330/Z/11/Z]. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Trachoma is caused by ocular infection with Chlamydia trachomatis and is the leading infectious cause of blindness worldwide. It manifests as distinct clinical syndromes beginning with an acute self-limiting keratoconjunctivitis, which following repeated episodes may progress to a more chronic inflammatory and immunofibrogenic process leading to conjunctival scarring and blinding sequelae.
Trachoma is endemic in 50 countries, with 325 million people at risk of blinding disease . Trachoma is responsible for visual impairment in 1.2 million people and 3% of blindness globally . The highest prevalence of active trachoma (trachomatous inflammation-follicular (TF) and/or trachomatous inflammation-intense (TI)) is in sub-Saharan Africa and the distribution of disease is heterogeneous .
Ocular C. trachomatis is probably transmitted between individuals through direct spread from eye to eye during close contact, direct or indirect spread of infected nasal or ocular secretions on fingers or cloths (fomites) and indirect passive transmission by eye seeking flies. There is no known animal reservoir of C. trachomatis in endemic environments, the primary reservoir being young children.
Blinding trachoma is usually found in hot, arid, dusty regions. A recent systematic review examined studies reporting higher trachoma prevalence in savannah areas and areas of lower rainfall, and found weak but consistent evidence supporting anecdotal findings that trachoma is associated with semi-arid environments .
This study was conducted on the Bijagós Archipelago, a remote group of islands off the coast of Guinea Bissau with a total population estimated at 24,000 , where trachoma is hyperendemic. The climate and environment are not typical of trachoma-endemic areas. The islands are covered with subtropical forest and altitude does not exceed 50 m. The climate is tropical, hot and humid. The islands are surrounded by mangroves and mudflats. There is significant rainfall (average 400 mm/month) from May to November .
Many studies have suggested that the prevalence of trachoma is associated with environmental risk factors such as poor sanitation, access to water and latrine use , . Eye-seeking flies (Musca sorbens) have also been associated with trachoma as passive vectors  but significant disease exists in areas where fly populations are scarce and are therefore less likely to contribute to trachoma transmission . M. sorbens preferentially breeds in human faeces and there may be association between fly populations and lack of latrine access or use , . Social risk factors such as migration events and crowded living conditions have also been shown to be important in transmission of C. trachomatis and the appearance of active trachoma , .
Clustering of disease at the community, household and bedroom levels has been noted and is likely to reflect the dynamics of transmission between family members with prolonged close contact , –. Most transmission events have been shown to occur at the household level with more gradual spread within the community .
The World Health Organization (WHO) advocates the implementation of the SAFE strategy (Surgery for trichiasis, Antibiotics for active infection, Facial cleanliness to prevent disease transmission and Environmental improvement to increase access to water and sanitation) for trachoma elimination. The WHO recommends annual mass treatment of entire communities with oral azithromycin for three years if the prevalence of TF in 1–9 year olds within a district or community exceeds 10%. Mass antibiotic treatment aims to clear infection from communities such that transmission ceases to be a public health concern . Following this, an assessment is made of A, F and E interventions and a decision is taken to continue or cease treatment .
Despite their inclusion in the SAFE strategy, local environmental factors are not well understood, though many are potentially modifiable risk factors for infection and disease. The relative importance of these risk factors is not clear and may differ between communities. Fewer studies have investigated risk factors for disease and infection simultaneously –. Understanding risk factors associated with trachoma and C. trachomatis infection may increase our understanding of disease and transmission dynamics allowing for optimization of community-specific interventions.
We examined household and individual-level risk factor associations with ocular C. trachomatis infection and active trachoma in this unique environment, where trachoma is a significant public health problem. Prior to these surveys, these communities were treatment-naïve and had not been exposed to any trachoma control interventions.
This study was conducted in accordance with the declaration of Helsinki. Ethical approval was obtained from the Comitê Nacional de Ética e Saúde (Guinea Bissau), the LSHTM Ethics Committee (UK) and The Gambia Government/MRC Joint Ethics Committee (The Gambia). Verbal consent was obtained from community leaders. Written informed consent was obtained from all study participants or their guardians on their behalf if participants were children. A signature or thumbprint is considered an appropriate record of consent in this setting by the above ethical bodies.
Study Design and Population
We conducted a cross-sectional population-based trachoma prevalence survey on four islands of the Bijagós Archipelago of Guinea Bissau (Bubaque, Canhabaque, Soga and Rubane) in January 2012. Trachoma survey methodology has been described previously –. We randomly sampled one in five households, representing a one stage probability sample design satisfying desired criteria for population-based prevalence surveys , . A sample size of 1500 ensured adequate power with conservative correction (using a design effect of 4) to account for anticipated household clustering. The sample size provides over 90% power to detect an odds ratio (OR) of 2 associated with a risk factor found in 20% of subjects without disease or infection, or an OR of 3 for a risk factor present in 5% of subjects without disease or infection with 95% confidence. The sample size also provides good precision for an estimated TF prevalence of >25% in 1–9 year olds on the four islands of Bubaque and Canhabaque (±4%), Soga (±6%) and Rubane (±10%), which is adequate to determine whether these communities require mass drug treatment with azithromycin in line with WHO policy.
A census of persons resident in randomly selected households was conducted prior to the household survey. Residency was defined as living within the household for longer than the preceding month or intending to stay resident in the household for longer than one month. This was updated to reflect the de facto population (those present in the household on the previous night) to limit absenteeism.
Household Risk Factor Survey
Demographic, socio-economic and environmental information was collected at household and individual levels. Household-level risk factor data were obtained using questionnaires administered to the household head or an appropriate responsible adult and included items on the level of education of the household head, their socio-economic status, whether the household had been exposed to any health education or promotion within the community, household access to and use of latrines, access and use of water and measures of sanitation, waste and presence of flies in the environment. The questionnaire was supported through observational data collected on water use, latrine use and environmental sanitation. Household size (measured as number of members of all ages) and number of children under the age of 10 years within the household was recorded. Researchers were masked to trachoma status of household members at the time of the household survey.
Trachoma Prevalence Survey
Following the household risk factor survey all individuals from study households were invited to attend for clinical examination and conjunctival sampling. Individuals' age, sex and ethnic group and data on facial cleanliness (the presence of ocular and/or nasal discharge and whether or not there were flies on the face) were collected at the time of examination.
A single trained examiner assessed each participant using the WHO simplified grading system where TF (trachomatous inflammation – follicular) and/or TI (trachomatous inflammation – intense) constitute active trachoma and TS (trachomatous scarring), TT (trachomatous trichiasis) and CO (corneal opacity) are trachomatous sequelae which may lead to blindness . A trachoma grade was assigned to the upper tarsal conjunctivae of each consenting participant using adequate light and a 2.5× binocular magnifying loupe.
Two sequential samples were taken from the left upper tarsal conjunctiva of each participant with Dacron swabs (Fisher Scientific, UK) using a standardised procedure , . The first swab was collected into transport medium for other studies. The second dry swab was collected into a microcentrifuge tube (Simport, Canada) and used in this study. Previous work using the Roche Amplicor CT/NG assay (Roche Molecular Systems, NJ USA) in a population-based study has shown that there was good agreement between first and second swabs with respect to C. trachomatis DNA positivity by PCR . Swabs were kept on ice in the field and frozen to −80°C within 8 hours of collection.
Measures were taken to avoid cross-contamination in the field. Control swabs (pre-marked swabs drawn at random from the swab dispenser and passed 10 cm in front of the eye ensuring no contact between the swab tip and participant) were taken to ensure field and laboratory quality control.
Community Mass Treatment
After survey completion all communities on the study islands were treated with a single height-based dose of oral azithromycin in accordance with WHO and national protocols.
Each swab was suspended in 400 µl sterile phosphate buffered saline (PBS) after thawing at room temperature. DNA was extracted from the swab/PBS suspension using an adapted whole blood protocol on the QIAxtractor (Qiagen, Crawley, UK) automated instrument and eluted into a final volume of 50 µl DX Elution Buffer (Qiagen).
Detection of Infection with C. trachomatis
C. trachomatis DNA was detected using the Roche Amplicor CT/NG assay (validated for use with ocular swabs ). Required reaction buffer conditions were obtained as described previously and used in the standard assay . Positive and negative samples were assigned according to the manufacturer's instructions. In this study, C. trachomatis infection is defined as the presence of C. trachomatis DNA by Amplicor PCR.
Data were double entered into a customised database (MS Access 2007). Discrepancies were resolved through reference to original data forms. Data were further cleaned prior to analysis in STATA 13 (Stata Corporation, College Station, Texas USA).
Random effects logistic regression models were used to assess the variability between villages and households assuming a three tier hierarchy to the data (at village, household and individual levels). Null models were used to examine the effect of cluster variables on the outcome using the likelihood ratio test (LRT), which if significant, provided strong evidence that between-village and household variance was non-zero. The log likelihood and the LRT were used to compare models.
Univariable associations with active trachoma (TF/TI) and infection with C. trachomatis were examined using two-level hierarchical random effects logistic regression, accounting for between-household variation. Covariates associated with active trachoma or C. trachomatis infection with p<0.10 (using the Wald test) were sequentially added to the multivariable model after a priori adjustment for age and gender (as categorical variables). Covariates were retained in the final model if the Wald p-value≤0.05 unless otherwise specified. Further exploration of environmental predictor variables was conducted using logistic and hierarchical random effects logistic regression models as appropriate using the same criteria. As C. trachomatis infection is on the causal pathway between several risk factors and active trachoma, models with and without C. trachomatis infection were fitted. The model including C. trachomatis infection provides estimates of independent associations of other risk factors with active trachoma which are not mediated through C. trachomatis infection.
All statistical analyses were carried out using STATA 13. Statistical significance was determined at the 5% level.
Study Population Characteristics
From an estimated total rural population of 5,613 inhabitants on the four study islands , 1,511 individuals from 293 randomly selected households across 39 villages were enrolled. Of these, 1,508 had an ocular assessment and conjunctival swabs were obtained from 1,507. The median age of participants was 13 years (range 1 month–88 years) and 57% were female. The majority of participants were of the Bijagós ethnic group (Table 1).
Prevalence of Active Trachoma and C. trachomatis Infection
The prevalence of active trachoma in 1–9 year olds was 22.0% (95% Confidence Interval (CI) 18.9–25.5%) (136/618). The prevalence of active trachoma was highest in children under the age of 5 years (27.3% (95% CI 23.1–31.9%) (113/416)). Overall, 11.1% (95% CI 9.4–12.6%) (167/1508) of the study population had active trachoma. The relationship between trachoma and infection is shown in Table 2. C. trachomatis DNA was detected in 18.0% overall (269/1507) and 25.4% of 1–9 year olds (157/618). All 15 (∼1% of total) control swabs were negative for C. trachomatis DNA.
Multilevel Clustering of Active Trachoma and C. trachomatis Infection
Null models for both active trachoma and C. trachomatis infection adjusted for age and gender showed significant clustering at island, village and household levels. For active trachoma, the variance estimated due to between-household clustering was 1.11 (standard error (SE) 0.17, p<0.0001). The between-village clustering variance was 0.75 (SE 0.16, p<0.0001) and between-island clustering variance was 0.50 (SE 0.28, p = 0.0100). For C. trachomatis infection, the variance estimated due to between-household clustering was 1.37 (SE 0.15, p<0.0001), between-village clustering was 0.89 (SE 0.14, p<0.0001) and between-island clustering was 0.40 (SE 0.18, p = 0.0005). The clustering effect was strongest at household level and models adjusting for clustering at household level only were a better fit than those including adjustment for village and island clustering. Adjusting for clustering at household level significantly improved the model versus standard logistic regression analyses (p<0.0001). Two-level hierarchical regression models with adjustment for household level clustering are presented in this analysis.
Factors Associated with Active Trachoma
Univariable associations with active trachoma are presented in Table 3. The final multivariable model showed that active trachoma was strongly independently associated with C. trachomatis infection (OR = 11.2 (95% CI 6.9–18.1)), ocular (OR = 2.0 (95% CI 1.0–4.0)) and nasal (OR = 2.5 (95% CI 1.5–4.3)) discharge, male gender (OR = 1.9 (95% CI 1.2–2.9)) and being aged 0–5 years (OR = 10.2 (95% CI 5.1–20.4) compared to being >15 years of age) (Model 2, Table 4). There was also a strong independent association between household water access and active trachoma, such that households with access only to a traditional natural spring as a water source had an increased risk of active trachoma compared to households with access to multiple water sources (OR = 1.9 (95% CI 0.9–3.9)). The model without C. trachomatis infection shows stronger associations, indicating that some effect of these factors is mediated through C. trachomatis infection (Table 4). Comparison of the two models suggests that some of the effect of younger age and water source is partly mediated through C. trachomatis infection, but these remain independently associated with trachoma beyond this effect.
Factors Associated with C. trachomatis Infection
Univariable associations with C. trachomatis infection are presented in Table 5. In the final multivariable model C. trachomatis infection was strongly independently associated with being aged ≤10 years. The presence of ocular discharge (OR = 2.3 (95% CI 1.3–4.4)) and household access only to a traditional natural spring (OR = 6.6 (95% CI 2.8–15.2)) and or access to a single water source only (OR = 3.9 (95% CI 1.9–8.0)) (rather than households who had access to multiple water sources) were strongly associated with infection (Table 6). The presence of flies around a latrine was also independently associated with infection (OR = 2.1 (95% CI 1.1–3.8)). The presence of flies around a latrine were strongly associated with the presence of flies in the environment surrounding the household (OR = 8.3 (95% CI 5.4–12.7), p<0.0001) and the presence of visible faeces within the latrine (OR = 46.7 (95% CI 28.5–76.6), p<0.0001). There was no association between flies in the environment (OR = 1.1 (95% CI 0.4–3.0), p = 0.91) nor flies around the latrine (OR = 0.5 (95% CI 0.1–2.6), p = 0.43) with flies on the face at the time of examination.
We have shown that there is clustering of both active trachoma and infection at household, village and island levels. We found that the most substantial clustering was present at household-level, thus the random effects introduced by this are adjusted for using two-level hierarchical mixed effects models. Clustering at household level has been described in other populations , , . We found C. trachomatis infection to be more strongly clustered at the household level than active trachoma, which has also been reported in Tanzanian communities .
Consistent with results from other studies, we found that being aged 0–5 years , , , –, the presence of ocular C. trachomatis infection  and the presence of ocular and nasal discharge , , ,  were strongly associated with active trachoma. We also found that being female was associated with a reduced odds of active trachoma, but not infection. The finding of an association between gender and active trachoma are contradictory in the literature –. In this analysis gender was included in all models due to a priori assumptions.
At the household level, the only household-level factor independently associated with both active trachoma and C. trachomatis infection was access to type and number of water sources, where household access to a traditional natural spring as the only water source was more strongly associated with active trachoma and C. trachomatis infection than access to another (single) source or multiple water sources of various types. This association is likely to be mediated through the presence of C. trachomatis infection. A similar association was shown in Ethiopia, where a piped water supply was associated with reduced risk of active trachoma compared to spring or river/lake water sources . This may have been related to increased distance to the non-piped water sources  or to volume of water used . In Bijagós communities, every household has access to a water source within 15 minutes walking distance. It is likely that this association may reflect more subtle hygiene and water use behaviours related to allocation of water collected for specific hygiene practices. A greater volume of water can be obtained with greater ease from an improved well or borehole than is possible from a natural spring. Although we found no statistically significant association between the volume of water stored in the household and active trachoma, it may be that the volume of water specifically allocated for hygiene practices is important, as shown in The Gambia , and that this finding is a proxy for water use allocation.
Children aged 0–5 years are at increased risk of ocular infection with C. trachomatis. In 6–10 year olds the risk is similarly elevated. In contrast, children aged 0–5 years are at much greater risk of active trachoma than 6–10 year olds. This supports suggestions that young children are the main reservoir of infection , ,  and demonstrates that this reservoir persists beyond pre-school age children in this population. The difference found in associations between age and active trachoma and infection may be explained by the presumed acquisition of immunity in childhood, which may result in reduced prevalence of clinically active trachoma despite the remaining strong association with infection. Duration  and intensity  of active trachoma and C. trachomatis infection have been shown to be age-dependent. It may also be that clinical signs present in 0–5 year olds are less specific for C. trachomatis infection and that other bacteria such as Streptococcus pneumoniae and Haemophilus influenzae contribute to the clinical phenotype, such that they induce a recall of the follicular response in children previously exposed to C. trachomatis, as hypothesised in cross-sectional studies showing an association between follicular trachoma and these bacteria in Tanzania and The Gambia , . Overall, there is a diminishing association with age, indicating that age may also represent a proxy for close contact and specific hygiene behaviours, acquired immunity and duration of infection. Whilst there were strong univariable associations between infection and ocular and nasal discharge, the association with nasal discharge is not significant in the multivariable model. Ocular discharge may more specific for C. trachomatis infection than nasal discharge, likely representing a consequence of C. trachomatis infection rather than a risk factor for it.
Latrine access was not associated with active trachoma or C. trachomatis infection, despite the fact that 60% of individuals do not have access to a latrine. Latrine access does not translate into latrine use . We assessed latrine use through observation of the latrines in addition to collecting reported data. Although there is high reported use of latrines within a household with latrine access, there may be household members that do not use the latrine. In these populations children aged 1–9 years are often discouraged from using latrines due to caregiver concerns about their safety when using latrines (Thompson et al., unpublished data).
The presence of flies around a household pit latrine was strongly associated with infection in the final multivariable model, although there was no association between flies on the faces of individuals at the time of examination with either active trachoma or infection. The presence of flies on faces has been well-documented as a risk factor for active trachoma in some environments , ,  and there is evidence that the eye-seeking flies Musca sorbens (and possibly other domestic muscidae) are passive vectors in C. trachomatis transmission –. One study that examined fly density around used household pit latrines in The Gambia found that the majority of flies emerging from the latrine were Chrysomya albiceps, rather than M. sorbens . We do not have data on the species of flies around pit latrines in Bijagós communities, however it is possible that there is ovipositioning by M. sorbens in or around latrines in these communities. We found a strong association between flies present in the environment and flies around the latrine, although there was no association with flies on the face at the time of examination. With decreasing age, there was a trend in association between flies on the face and flies around the latrine, but this association was non-significant.
M. sorbens have been shown to preferentially breed in human excreta, in addition to domestic animal faeces . Although the presence of animal and human excreta within 15 metres of the household were not risk factors for active trachoma or C. trachomatis infection, they were ubiquitous in the villages. There was a strong association noted between flies around a latrine and the presence of visible faeces within a latrine. These findings suggest that flies may be an important in the transmission of trachoma in this environment. Our study was conducted in the cooler dry season, possibly explaining the lack of association between flies on faces and in the environment and C. trachomatis infection in the regression analysis. Fly populations and density have been shown to have seasonality elsewhere , .
The cross-sectional design of this study is a limitation in understanding determinants of disease and infection. There may also be some limitations in using stepwise regression methods such that models may be overfitted or unstable . Risk factors are often difficult to define and despite the presence of statistically significant associations, it is likely that the relationship between socio-behavioural and environmental risk factors and disease and infection is complex and not fully explained by this analysis. It may be that intra- and inter-familial transmission occurs in these communities as a consequence of specific hygiene practices and complex socio-behavioural factors. These factors are often difficult to define and require longitudinal studies beyond the scope of the current study. Further epidemiological studies examining the species and contribution of flies to trachoma transmission and water use and specific hygiene behaviours in relation to trachoma and C. trachomatis infection in this population may be of importance in successful trachoma control activities.
We have described individual and household-level risk factor associations with active trachoma and ocular infection with C. trachomatis on the Bijagós Archipelago to improve our understanding of the relationship between disease and infection in this remote treatment-naïve trachoma-hyperendemic population.
These data suggest that in this environment household-level risk factors relating to fly populations, hygiene behaviours and water usage are likely to be important in the transmission of ocular C. trachomatis infection. Education about cleanliness, sanitation and hygiene practices is likely to be important in reducing transmission of infection in these communities. Ensuring the provision of water sources which allow adequate water to be allocated for hygiene may assist this, and further studies examining specific hygiene practices may be useful. Reducing fly populations around the latrines where they exist may be of benefit. These findings may be important in the implementation of the F and E components of SAFE in this population.
In order to fully understand the factors associated with active trachoma and ocular C. trachomatis infection in these communities, further epidemiological studies examining transmission and clustering of C. trachomatis infection are required. These studies should focus on pathogen factors such as the role of infection intensity and strain diversity, and socio-behavioural factors such as specific hygiene behaviours.
Azithromycin was donated for treatment of trachoma on the islands by Pfizer Inc. through application via the International Trachoma Initiative. We extend our thanks to our colleagues in the Programa Nacional de Saúde de Visão in Bissau and to the communities of the Bijagós Archipelago, our field research team and all study participants in Guinea Bissau. We acknowledge the contribution from KM Thompson's unpublished data to our discussion.
Conceived and designed the experiments: ARL EMHE SEB MJH RLB. Performed the experiments: ARL EC MN. Analyzed the data: ARL HAW. Contributed reagents/materials/analysis tools: ARL SEB MN MJH. Wrote the paper: ARL RLB MJH DCM. Obtained the funding: ARL RLB MJH DCM. Coordinated the field project: ARL EC MN SEB. Interpreted the data: ARL HAW RLB EMHE.
- 1. World Health Organization (2012) Prevention of blindness and visual impairment (Priority Eye Diseases): Trachoma. Available: https://www.who.int/blindness/cause/priority/en/index2.html.
- 2. Solomon AW, Peeling RW, Foster A, Mabey DCW (2004) Diagnosis and Assessment of Trachoma. Clin Micro Rev 17 (4) 982–1011. doi: 10.1128/cmr.17.4.982-1011.2004
- 3. Ramesh A, Kovats S, Haslam D, Schmidt E, Gilbert CE (2013) The impact of climatic risk factors on prevalence, distribution, and severity of acute and chronic trachoma. PLoS Negl Trop Dis 7 (11) e2513. doi: 10.1371/journal.pntd.0002513
- 4. National Population Census (Bolama-Bijagós Region) (2010) Instituto Nacional de Estatística, Bissau, Guiné-Bissau.
- 5. World Bank (2013) From the Climatic Research Unit (CRU) at the University of East Anglia (UEA). Climate Change Knowledge Portal: Guinea Bissau. Available: http://sdwebx.worldbank.org/climateportal/index.cfm?page=country_historical_climate&ThisRegion=Africa&ThisCCode=GNB
- 6. Burton MJ, Holland MJ, Faal N, Aryee EA, Alexander ND, et al. (2003) Which members of a community need antibiotics to control trachoma? Conjunctival Chlamydia trachomatis infection load in Gambian villages. Invest Ophthalmol Vis Sci 44: 4215–4222. doi: 10.1167/iovs.03-0107
- 7. Emerson PM, Cairncross S, Bailey RL, Mabey DC (2000) Review of the evidence base for the ‘F’ and ‘E’ components of the SAFE strategy for trachoma control. Trop Med Int Health 5: 515–527. doi: 10.1046/j.1365-3156.2000.00603.x
- 8. Emerson PM, Lindsay SW, Alexander N, Bah M, Dibba SM, et al. (2004) Role of flies and provision of latrines in trachoma control: cluster randomized controlled trial. Lancet 363: 1093–1098. doi: 10.1016/s0140-6736(04)15891-1
- 9. Taylor HR, Velasco FM, Sommer A (1985) The ecology of trachoma: an epidemiological study in southern Mexico. Bull World Health Organ 63: 559–567.
- 10. Bailey R, Osmond C, Mabey DC, Whittle HC, Ward ME (1989) Analysis of the household distribution of trachoma in a Gambia village using a Monte Carlo simulation procedure. Int J Epidemiol 18: 944–951. doi: 10.1093/ije/18.4.944
- 11. Burton MJ, Holland MJ, Makalo P, Aryee EA, Alexander ND (2005) Re-emergence of Chlamydia trachomatis infection after mass antibiotic treatment of a trachoma-endemic Gambian community: a longitudinal study. Lancet 365: 1321–8. doi: 10.1016/s0140-6736(05)61029-x
- 12. West SK, Munoz B, Turner VM, Mmbaga BB, Taylor HR (1991) The epidemiology of trachoma in central Tanzania. Int J Epidemiol 20: 1088–1092. doi: 10.1093/ije/20.4.1088
- 13. Blake IM, Burton MJ, Bailey RL, Solomon AW, West SK (2009) Estimating household and community transmission of ocular Chlamydia trachomatis. PLoS Negl Trop Dis 3: e401. doi: 10.1371/journal.pntd.0000401
- 14. Kuper H, Solomon AW, Buchan J, Zondervan M, Foster A, et al. (2003) A critical review of the SAFT strategy for the prevention of blinding trachoma. Lancet Infect Dis 3: 372–381. doi: 10.1016/s1473-3099(03)00659-5
- 15. WHO (2004) Report of the eighth meeting of the WHO Alliance for the Global Elimination of Blinding Trachoma, Geneva: Switzerland. Available: http://www.who.int/blindness/publications/GET_8_Report.pdf
- 16. Abdou A, Nassirou B, Kadri B, Moussa F, Muñoz B, et al. (2007) Prevalence and risk factors for trachoma and ocular Chlamydia trachomatis infection in Niger. Br J Ophthalmol 91: 13–17. doi: 10.1136/bjo.2006.099507
- 17. West SK, Rapoza P, Muñoz B, Katala S, Taylor HR (1991) Epidemiology of ocular chlamydial infection in a trachoma-hyperendemic area. J Infect Dis 163: 752–756. doi: 10.1093/infdis/163.4.752
- 18. Harding-Esch E, Edwards T, Mkocha H, Muñoz B, Holland MJ, et al. (2010) Trachoma prevalence and associated risk factors in The Gambia and Tanzania: Baseline results of a cluster randomized controlled trial. PLoS Negl Trop Dis 4 (11) e861. doi: 10.1371/journal.pntd.0000861
- 19. Edwards T, Harding-Esch EM, Hailu G, Andreason A, Mabey DC, et al. (2008) Risk factors for active trachoma and Chlamydia trachomatis infection in rural Ethiopia after mass treatment with azithromycin. Trop Med Int Health 13: 556–565. doi: 10.1111/j.1365-3156.2008.02034.x
- 20. Ngondi J, Reacher M, Matthews F, Brayne C, Emerson P (2009) Trachoma survey methods: A literature review. Bull World Health Organ 87: 143–51. doi: 10.2471/blt.07.046326
- 21. Bennett S, Woods T, Liyanage WM, Smith DL (1991) A simplified general method for cluster-sample surveys of health in developing countries. World Health Stat Q 44: 98–106.
- 22. Wright HR, Vu H, Taylor HR (2005) How to assess the prevalence of trachoma. Br J Ophthalmol 89: 526–7. doi: 10.1136/bjo.2005.066183
- 23. Thylefors B, Dawson CR, Jones BR, West SK, Taylor HR (1987) A simple system for the assessment of trachoma and its complications. Bull World Health Organ 65 (4) 477–83.
- 24. Keenan JD, Lakew T, Alemayehu W, Melese M, Porco TC (2010) Clinical activity and polymerase chain reaction evidence of chlamydial infection after repeated mass antibiotic treatments for trachoma. Am J Trop Med Hyg 82 (3) 482–7. doi: 10.4269/ajtmh.2010.09-0315
- 25. Stare D, Harding-Esch E, Munoz B, Bailey R, Mabey D (2011) Design and baseline data of a randomised trial to evaluate coverage and frequency of mass treatment with azithromycin: The Partnership for the Rapid Elimination of Trachoma (PRET) in Tanzania and The Gambia. Ophthalmic Epidemiol 18 (1) 20–9. doi: 10.3109/09286586.2010.545500
- 26. Michel CE, Roper KE, Divena MA, Lee HH, Taylor HR (2011) Correlation of clinical trachoma and infection in Aboriginal communities. PLoS Negl Trop Dis 5 (3) e986. doi: 10.1371/journal.pntd.0000986
- 27. Leslie DE, Azzalo F, Ryan N, Fyfe J (2003) An assessment of the Roche Amplicor Chlamydia trachomatis/Neisseria gonorrhoeae multiplex PCR assay in routine diagnostic use on a variety of specimen types. Commun Dis Intell Q Rep 27 (3) 373–9.
- 28. Holland MJ, Jeffries D, Pattison M, Korr G, Gall A (2010) Pathway focused arrays reveal increased matrix metalloproteinase-7 (Matrilysin) transcription in trachomatous trichiasis. Invest Ophthalmol Vis Sci 51 (8) 3893–3902. doi: 10.1167/iovs.09-5054
- 29. Hagi M, Schemann JF, Mauny F, Momo G, Sacko D (2010) Active trachoma among children in Mali: Clustering and environmental risk factors. PLoS Negl Trop Dis 4 (1) e583. doi: 10.1371/journal.pntd.0000583
- 30. Schemann JF, Sacko D, Malvy D, Momo G, Traore L, et al. (2002) Risk factors for trachoma in Mali. Int J Epidemiol 31: 194–201. doi: 10.1371/journal.pntd.0000583
- 31. Alemayehu W, Melese M, Fredlander E, Worku A, Courtright P, et al. (2005) Active trachoma in children in central Ethiopia: association with altitude. Trans R Soc Trop Med Hyg 99: 840–3. doi: 10.1016/j.trstmh.2005.06.013
- 32. Cumberland P, Hailu G, Todd J (2005) Active trachoma in children aged three to nine years in communities in rural Ethiopia: prevalence, indicators and risk factors. Trans R Soc Trop Med Hyg 99: 120–7. doi: 10.1016/j.trstmh.2004.03.011
- 33. Polack S, Kuper H, Solomon AW, Massae PA, Abuelo C, et al. (2006) The relationship between the prevalence of active trachoma, water availability and its use in a Tanzanian village. Trans R Soc Trop Med Hyg 100: 1075–1083. doi: 10.1016/j.trstmh.2005.12.002
- 34. Baggaley RF, Solomon AW, Kuper H, Polack S, Massae PA, et al. (2006) Distance to water source and altitude in relation to active trachoma in Rombo district, Tanzania. Trop Med Int Health 11: 220–7. doi: 10.1111/j.1365-3156.2005.01553.x
- 35. Mesfin MM, de la Camera J, Tereke IG, Ananual G, Araya T, et al. (2006) A community-based trachoma survey: Prevalence and risk factors in the Tigray region of northern Ethiopia. Ophthalmic Epidemiol 13: 173–81. doi: 10.1080/09286580600611427
- 36. Harding-Esch EM, Edwards T, Sillah A, Sarr-Sissoho I, Aryee EA, et al. (2008) Risk factors for active trachoma in The Gambia. Trans R Soc Trop Med Hyg 102: 1255–1262. doi: 10.1016/j.trstmh.2008.04.022
- 37. Golovaty I, Jones L, Gelaye B, Tilahun M, Belete H, et al. (2009) Access to water source, latrine facilities and other risk factors of active trachoma in Ankober,. Ethiopia 4 (8) e6702. doi: 10.1371/journal.pone.0006702
- 38. Bejiga A, Alemayehu W (2001) Prevalence of trachoma and its determinants in Dalocha District, Central Ethiopia. Ophthalmic Epidemiol 8: 119–125. doi: 10.1076/opep.188.8.131.5268
- 39. Bailey R, Downes B, Downes R, Mabey D (1991) Trachoma and water use: A case-control study in a Gambian village. Trans R Soc Trop Med Hyg 85: 824–8. doi: 10.1016/0035-9203(91)90470-j
- 40. Bailey R, Duong T, Carpenter R, Whittle H, Mabey D (1999) The duration of human ocular Chlamydia trachomatis is age dependent. Epidemiol Infect 123 (3) 479–86. doi: 10.1017/s0950268899003076
- 41. Gambhir M, Basanex MG, Burton MJ, Solomon AW, Bailey RL, et al. (2009) The development of an age-structured model for trachoma transmission dynamics, pathogenesis and control. PLoS Negl Trop Dis 3 (6) e462. doi: 10.1371/journal.pntd.0000462
- 42. Burton MJ, Hu VH, Massae P, Burr SE, Chevallier C, et al. (2011) What is causing active trachoma? The role of non-Chlamydial pathogens in a low prevalence setting. Invest Ophthalmol Vis Sci 52 (8) 6012–7. doi: 10.1167/iovs.11-7326
- 43. Burr SE, Hart JD, Edwards T, Baldeh I, Bojang E, et al. (2013) Association between ocular bacterial carriage and follicular trachoma following mass azithromycin distribution in The Gambia. Plos Negl Trop Dis 7 (7) e2347. doi: 10.1371/journal.pntd.0002347
- 44. O'loughlin R, Fentie G, Flannery B, Emerson PM (2006) Follow-up of a low-cost latrine promotion programme in one district of Amhara, Ethiopia: characteristics of early adopters and non-adopters. Trop Med Int Health 11 (9) 1406–15. doi: 10.1111/j.1365-3156.2006.01689.x
- 45. Brechner RJ, West S, Lynch M (1992) Trachoma and flies. Individual vs environmental risk factors. Arch Ophthalmol 110: 687–9. doi: 10.1001/archopht.1992.01080170109035
- 46. Emerson PM, Bailey RL, Walraven GE, Lindsay SW (2001) Human and other faeces as breeding media of the trachoma vector Musca sorbens. Med Vet Entomol 15 (3) 314–20. doi: 10.1046/j.0269-283x.2001.00318.x
- 47. Emerson PM, Simms VM, Makalo P, Bailey RL (2005) Household pit latrines as a potential source of the fly Musca sorbens – a one year longitudinal study from The Gambia. Trop Med Int Health 10 (7) 706–9. doi: 10.1111/j.1365-3156.2005.01432.x
- 48. Miller K, Pakpour N, Yi E, Melese M, Alemayehu W, et al. (2004) Pesky trachoma suspect finally caught. Br J Ophthalmol 88 (6) 750–1. doi: 10.1136/bjo.2003.038661
- 49. Taye A, Alemayehu W, Melese M, Gayid A, Mekonnen Y, et al. (2007) Seasonal and altitudinal variations in fly density and their association with the occurrence of trachoma in the Gurage zone of central Ethiopia. Ann Trop Med Parasitol 101 (5) 441–8. doi: 10.1179/136485907x176544
- 50. Da Cruz L, Dadour IR, McAllister IL, Jackson A, Isaacs T (2002) Seasonal variation in trachoma and bush flies in north-western Australian Aboriginal communities. Clin Experiment Ophthalmol 30 (2) 80–3. doi: 10.1046/j.1442-6404.2002.00491.x
- 51. Steyerberg EW, Eijkemans MJ, Habberna JD (1999) Stepwise selection in small data sets: a simulation study of bias in logistic regression analysis. J Clin Epidemiol 2 (10) 935–42. doi: 10.1016/s0895-4356(99)00103-1