Article Text
PDF
Abstract
Background/Aims Population-based surveys on diabetes and diabetic retinopathy (DR) are necessary to increase awareness and develop screening and therapeutic programmes. The aim was to estimate the prevalence of DR in older adults of different ethnic backgrounds in Suriname.
Methods Fifty clusters of 60 people aged ≥50 years were randomly selected with a probability proportional to the size of the population unit. Eligible people were randomly selected through compact segment sampling and examined using the Rapid Assessment of Avoidable Blindness plus Diabetic Retinopathy (RAAB + DR) protocol. Participants were classified as having diabetes if they: were previously diagnosed with diabetes; were receiving treatment for glucose control; had a random blood glucose level >200 mg/dL. These participants were dilated for funduscopy, assessed for DR following the Scottish DR grading protocol and evaluated for ethnicity and DR ophthalmic screening frequencies.
Results A total of 2806 individuals was examined (response 93.6%). The prevalence of diabetes was 24.6%. In these patients any type of DR and/or maculopathy occurred in 21.6% and sight-threatening DR in 8.0%. Of the known diabetics, 34.2% never had an eye examination for DR and in 13.0% the last examination was >24 months ago. The prevalence of diabetes was significantly higher in Hindustani people compared with other major ethnic groups.
Conclusions The prevalence of diabetes and diabetics without regular DR control in people aged ≥50 years in Suriname was higher than expected. The uptake for special services for DR has to be expanded to decrease patient delay and DR-induced blindness.
- Epidemiology
- Public health
- Retina
Statistics from Altmetric.com
Introduction
It is estimated that in 2014, there were 387 million people with diabetes and that by 2030, this number will increase to 438 million people globally.1 Since nearly 80% of all people with diabetes live in low-income and middle-income countries,2 this increase will largely occur in those countries, including those in South and Latin America.3 Diabetic retinopathy (DR), a severe eye complication of diabetes, is responsible for 1.0% of blindness and visual impairment (VI) worldwide.4 Even in high-income countries, it is the leading cause of blindness in the working-age population.2 As other causes of blindness are likely to decrease due to Vision 2020 programmes, the number of blind people due to posterior segment disease including DR, will increase.2 Population-based surveys on diabetes and DR are scarce, but necessary to provide an up-to-date assessment of the problem, to improve awareness and to develop effective intervention programmes.
The Rapid Assessment of Avoidable Blindness (RAAB) survey method is a simple methodology to assess the prevalence and causes of blindness in people aged ≥50 years in a specific geographic area.5 RAAB was developed to focus primarily on the prevalence of avoidable causes of VI and blindness such as cataract, refractive errors and corneal scarring.5 Due to awareness of the increasing burden of DR worldwide, a new method has been developed to estimate the prevalence of DR within the RAAB survey. So far, the inclusion of DR in RAAB has been successfully undertaken in 11 countries, with available data from Mexico, Moldova, Saudi Arabia and Jordan.3 ,6–8 In contrast to earlier studies using different study methodologies, the use of this standardised method makes it possible to compare data from different countries. However, significant gaps still exist in reliable population-based DR data from low-income and middle-income nations.2
In Suriname, the prevalence of diabetes is expected to be around 20% in patients aged ≥50 years.1 Recently, data were reported on the prevalence and causes of blindness and VI in people aged ≥50 years in Suriname.9 The sample prevalence of bilateral blindness was estimated at 2.3% with cataract as the most frequent cause (54%), followed by glaucoma (23.8%) and other posterior segment disease (7.9%).9 Despite high-quality ophthalmic care including laser therapy and vitreoretinal surgery, DR was shown to be the fourth most frequent cause of bilateral blindness (3.2%) and VI (3.3%).9 More detailed data on the prevalence of diabetes and DR in Suriname are lacking and urgently needed for appropriate planning of DR services.
The following study aimed to identify the prevalence of DR in Suriname including the number of diabetics on regular ophthalmic controls, using the RAAB + DR method. Since the population composition of Suriname consists of a multicultural society,10 ethnic differences in diabetes and DR prevalence were also investigated. The data obtained will be used for the development of preventive and therapeutic diabetes and DR programmes, to prevent patient delay and to reduce the burden of DR-induced blindness in Suriname.
Materials and methods
Rapid Assessment of Avoidable Blindness
The RAAB + DR survey was carried out between August 2013 and November 2014 in accordance with the codes of conduct of the Declaration of Helsinki. Ethical approval for the study was obtained from the Surinamese Ministry of Health. The RAAB only includes individuals aged ≥50 years as this age group has the highest prevalence of blindness.5 Giving a target population of 101 000 (census 2011), the sample size calculation module of the RAAB software indicated that a sample size of 3000 subjects would provide enough power to estimate an expected prevalence of blindness of 2.3%, with a variation of 32.5%, a design effect of 1.6 for a cluster size of 60 and taking a non-response of 7% into account.9 Based on an expected prevalence of diabetes of 20% in people aged ≥50 years and 25% prevalence of DR among diabetics, the required sample size for DR was 2300, which would be achieved in this sample. Fifty census enumeration areas were randomly selected with a probability proportional to the population in the area. Within each selected enumeration area 60 residents aged ≥50 years were randomly selected through compact segment sampling and examined in their own house using the standard RAAB protocol.5 Informed written or thumb-printed consent was obtained from all participants. Presenting distance visual acuity (VA) was tested using a Snellen tumbling E chart in full daylight. The primary cause of blindness and VI was assessed by an ophthalmologist in people with VA less than 6/18 in either eye. Care was taken that all people examined during the survey who required medical attention were provided with medical care.
Diabetes and DR assessment
Participants were classified as having diabetes if they were previously diagnosed with diabetes (known diabetics), if they were receiving treatment for glucose control (known diabetics) and if they were not earlier diagnosed with diabetes, but had a random blood glucose level of >200 mg/dL (11.1 mmol/L, random blood glucose test (Bayer)).3 Known diabetics were asked about age at diagnosis, treatment type and whether they had any eye examination because of their diabetes. All participants classified as having diabetes were dilated for indirect funduscopy by a trained ophthalmologist. Findings were graded during examination using the Scottish DR grading system for which teams were specially trained.11
Statistical analysis
Automated analyses included in the RAAB software package were used to analyse the data. Non-respondents were excluded from statistical analyses.9 The cluster sampling design was taken into account when CIs were calculated for sample prevalence of diabetes and DR. P values <0.05 were considered statistically significant.
Logistic regression analyses were performed in SPSS V.22 (IBM SPSS statistics) to estimate the age and sex standardised ORs for diabetes or DR prevalence. To compare diabetes and DR prevalence between ethnic groups, logistical regression analyses were performed in R (V.3.2.0) and quasi-SEs were used to calculate 95% CIs.12
Results
A total of 2998 subjects aged ≥50 years was numerated in the survey, of whom 2806 (93.0%) took part in the RAAB (table 1).9 Of the 2806 included RAAB participants, 2738 (97.6%) completed the DR module.
Age and gender composition of Suriname and sample population
Diabetes and DR
Overall, 689 (24.6%; 95% CI 21.8% to 27.3%; table 2) people had diabetes according to the predefined criteria. Prevalence of diabetes was lower among those ≥80 years (11.6%; 7.8% to 15.3%) compared with other age groups (range 24.4–28.0%; table 2). There was no significant difference in the prevalence of diabetes by gender. Of all patients with newly diagnosed diabetes, 58.3% were found among people aged 50–59 years (table 2).
Prevalence of diabetes by age group and by gender
Hindustani people were significantly more likely to have diabetes compared with other major ethnic groups (figure 1). Prevalence of diabetes was higher in urban compared with rural areas (OR 4.69, 2.9 to 7.5, p<0.001). The estimated prevalence of any DR in the full survey population (including non-diabetic subjects) was 5.3% (4.4 to 6.3), whereas 0.9% (0.5 to 1.3) had proliferative DR and 1.4% (0.9 to 1.9) had referable diabetic maculopathy (table 3). The prevalence of blindness and VI among diabetics was 1.9% (0.8 to 2.9) and 4.4% (2.6 to 6.1), respectively, versus 2.4% (1.3 to 3.5) and 7.1% (5.5 to 8.8) in people without diabetes.
Prevalence of DR in diabetics and in the entire sample using the Scottish DR grading system
Comparison of diabetes and diabetic retinopathy (DR) prevalence between different ethnic groups. The plot shows estimates and CIs (based on quasi-SEs) to compare diabetes and DR prevalence between different ethnic groups. ORs against the reference group (Hindustani, estimate set at zero) can be calculated per ethnicity by taking the exponential of the estimates.
Among people with diabetes, 617 (89.6%) were previously diagnosed. The mean duration of diabetes was 12.3 years (SD 11.1 years). Of these, 211 (34.2%) never had an eye examination, 248 (40.2%) had an eye examination <1 year ago, 78 (12.6%) between 1 and 2 years earlier and 80 (13.0%) >2 years ago. Seventy-two people (10.4%) were assessed as ‘new cases’ (random blood sugar >200 mg/dL). Of the ‘known diabetics’, >40% had a random blood glucose of >200 mg/dL and only 58.5% was considered to be well controlled (random blood sugar <200 mg/dL). About 77.3% of diabetics were only using tablets to control the disease, 15.6% were using insulin and 6.9% were not using any medical treatment. Diabetes in women was significantly better controlled compared with men (mean 191 mg/dL vs 212 mg/dL; p<0.001).
Of the 689 people classified as having diabetes, 49 (7.1%) declined to undergo pupillary dilation. Of the 640 (92.9%) examined people, 21.6% had any sign of DR (and/or maculopathy), 12.3% had any sign of maculopathy and 8% had sight-threatening DR (STDR, table 3). The prevalence of DR and/or maculopathy was not significantly related to age (table 4), but considerably higher among known compared with new cases (OR 4.65; 1.7 to 13.0, p=0.003; table 3).
Prevalence of diabetic retinopathy and/or maculopathy by age and by gender
Diabetes duration (OR 1.06, 1.0 to 1.1, p<0.001) and use of insulin (OR 4.82, 3.0 to 7.7, p<0.001) were also significantly associated with diagnosis of DR. Prevalence of DR was not significantly related to ethnicity (figure 1) or blood sugar level (OR 1.0, p=0.080).
Discussion
This was the first survey on blindness, VI and DR in the Republic of Suriname and, to the best of the authors’ knowledge, the first RAAB survey in South America which included the DR module. Overall, the prevalence of diabetes was 24.6% of which 21.6% had any sign of DR or maculopathy. Eight per cent of people with diabetes had STDR and more than one-third never had any eye examination.
The prevalence of diabetes was slightly higher than expected. This could be explained by the fact that the hypothesised prevalence was based on the prevalence of diabetes in the total population,1 since no specific data were available for the population aged ≥50 years. Prevalence of diabetes was average compared with earlier RAAB + DR surveys performed in other countries. In these surveys, prevalence of diabetes ranged from 11.4% (10.4–12.4) in the Republic of Moldova to 29.7% (28.1–31.4) in Taif, Saudi Arabia.6–8 In Chiapas, Mexico, the geographically closest country having RAAB + DR data available, the prevalence of diabetes was 21.0% (19.5–23.1).3
In Suriname, prevalence of diabetes varied between different ethnic and different age groups. The population in Suriname is different from other countries in South America. It has a more multicultural society consisting of Hindustanis (27%), Creoles (16%), Javanese (14%), Maroons (22%), as well as mixed (13%), Chinese (1%) and various other minorities.10 In the current survey, diabetes was most prevalent in the Hindustani and least prevalent in the (rural) Maroon population. Prevalence of diabetes was also higher in the urban and coastal areas compared with the rural population. The higher prevalence of diabetes in Hindustani people is consistent with a population-based study performed among Surinamese participants in The Netherlands showing Hindustani origin to be one of the most important predictors for diabetes.13 Although a different genetic profile may, in part, explain these differences, environmental and modifiable factors related to diet and lifestyle may also play an important role.14 ,15 In people aged ≥80 years, prevalence of diabetes was low, which could be due to a higher mortality rate in patients with complicated diabetes due to cardiovascular diseases at younger age. Since, more than 50% of patients with newly diagnosed diabetes were found in people aged 50–59 years, screening for diabetes could be effective in this specific age group.
The prevalence of DR among diabetics in Suriname was quite low compared with other countries. In 1999, the prevalence of DR in Latin America was estimated to be 40.2%.16 Furthermore, in an overview of the global pattern of DR, prevalence estimates ranged from 10% to 61% in known diabetics and from 1.5% to 31% in new cases.2 In the same overview, DR prevalence was shown to be higher in low-income and middle-income countries.2 Unfortunately, both studies included people of different age categories, and different screening methods were used which makes it difficult to compare. However, the estimated prevalence of DR and STDR in Suriname was still lower compared with the reported prevalence in other RAAB + DR surveys, ranging from 36.8% (17.5% STDR) in Taif, Saudi Arabia to 55.9% (14.6% STDR) in the Republic of Moldova.3 ,6–8 Apart from differences in eye care organisation, these differences could be due to population (ethnicity) and socioeconomic differences.2 Furthermore, the prevalence of DR in Suriname may be underestimated compared with RAAB + DR surveys where a fundus camera was used, which was found to be a more sensitive diagnostic tool for evaluation of DR than clinical examination in the field.3
In Suriname, several factors were associated with a higher prevalence of DR. Although DR prevalence was not significantly related to ethnicity, it was estimated to be highest among Hindustani and Creole diabetics and lowest among Maroons. In general, it is suggested that differences in ethnicity may play a role in DR prevalence, although relatively few studies have investigated this.2 Available data from other multiethnic populations showed prevalence of DR to be the highest in Africans/Afro-Caribbeans compared with South Asians or white Europeans.17 Since most population groups in Suriname have the same access to ophthalmic care, genetic susceptibility could play a role. On the other hand, the interior Maroons have the same West African origin as the urbanised Creole population, but no DR was found in the interior population. These findings suggest additional environmental influences on the risk of developing DR.
In our survey, the use of insulin and long duration of diabetes were both significantly associated with a higher risk of developing DR. Although not shown in our survey, in general, good control of blood sugar level is the most important factor in reducing incidence and progression of DR.2 ,18 ,19 For Suriname, the number of poorly controlled diabetics was high, but lower when compared with other RAAB + DR surveys.3 ,6 ,7 Over 40% of known diabetics in Suriname had a random blood sugar >200 mg/dL and required a better adjustment of their diabetes. It was remarkable that diabetes in women was significantly better controlled than in men, which could be explained by the fact that women in Suriname are more self-sufficient. While duration of diabetes was significantly related to DR, age was not, which was most likely due to the small sample size or survival of patients with diabetes without comorbidity.
Another critical point of concern is the number of patients with diabetes who were never examined by an ophthalmologist. Regular ophthalmic screening is important in preventing or moderating the visual consequences of DR. Suriname already offers a great range of screening and treatment possibilities for DR. These include fluorescence angiography, optical coherence tomography, laser therapy, intravitreal injections and vitreoretinal surgery, needed for an optimal DR screening and treatment programme.16 Reducing delay in diagnosis and treatment of DR could tremendously reduce the risk of becoming blind.16 Overall, the relatively low prevalence of DR, STDR, poor controlled diabetics and newly diagnosed diabetics in our survey compared with other RAAB + DR surveys may suggest relatively good diagnostic and treatment services for DR and diabetes in Suriname. Still, optimisation of diabetic (eye)care could improve these numbers even more. In the current survey, all newly diagnosed diabetics and those diagnosed with DR were provided with a full ophthalmic examination and follow-up in the Suriname Eye Centre.
This study had some limitations related to the selection of the sample. The older age groups were over-represented in our sample, which may have caused a slight underestimation of the prevalence of diabetes. Estimates of prevalence of diabetes may also have been influenced by the rapid methods required in the RAAB + DR protocol to test glucose levels: only one measurement of non-fasting glucose levels was performed. Oral glucose tolerance test, fasting glucose or repeated measurements may have provided more reliable data. Furthermore, only people aged ≥50 years were examined and our survey offers no data about the prevalence of diabetes in younger age groups.
In conclusion, the estimated prevalence of diabetes in people aged ≥50 years in Suriname is higher than expected with unacceptably high proportions of uncontrolled diabetics and patients who never had an ophthalmological examination. To decrease the prevalence of STDR and DR-induced blindness, the uptake of special services for DR has to be expanded by increasing the coverage of regular DR controls and the proportion of patients with a well-adjusted blood sugar level.
Acknowledgments
The authors thank all crew and staff members of the Suriname Eye Centre for participating in the field during data collection and data entry, the Medical Mission for providing logistic support in the interior of Suriname and the General Bureau of Statistics for the census data and the enumeration area maps.
References
Footnotes
Contributors JM, JCP, A-MTBdM-V and HL have contributed to all aspects of the planning, conduct and reporting of the work described. HCIT, MRS and CMF-P contributed to the conduct of this survey. RMAvN, DRAM and ACM contributed to the design of the work, the analysis/reporting of data and revised the work critically for important intellectual content. All authors agree with the final version of the manuscript and agree to be accountable for all aspects of the work.
Funding Academic Hospital Paramaribo, Paramaribo; Worldwide Access to Medical Advances Foundation, Amsterdam; Nelly Reef Fund, Amsterdam; Stichting Blindenhulp, Den Haag; Stichting tot Verbetering van het Lot der Blinden, Huizen; Rudolph en Barbara Hoppenbrouwers Fonds, Amsterdam; Stichting Nederlands Oogheelkundig Onderzoek, Nijmegen; Stichting Wetenschappelijk Onderzoek Binkhorst, Nijmegen.
Disclaimer The Academic Hospital Paramaribo or other funding organisations had no role in the design or conduct of this research.
Competing interests None declared.
Patient consent Obtained.
Ethics approval Ministry of Health of Suriname.
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- At a glance