Community screening for undiagnosed dysglycaemia in a semi-urban locality in Sri Lanka

Abstract

Introduction: Previous studies on the prevalence of pre-diabetes and diabetes in Sri Lanka have shown conflicting results. We studied the prevalence of pre-diabetes and diabetes using fasting plasma glucose (FPG) and 75g oral glucose tolerance (OGTT) test among a representative sample of adults in a semi-urban locality in Southern Sri Lanka. Sensitivity and specificity of FPG over OGTT in diagnosing pre-diabetes and diabetes were calculated. Methods: Healthy adults, aged 20-75 years, selected by multi-stage random sampling from one medical officer of health area in Southern Sri Lanka were screened with FPG and 75g OGTT. Subjects known to have diabetes were excluded. Results: Mean (SD) age of men (n=234) and women (n=593) were 44 (10.1) and 43(10.8) years, respectively. With OGTT, 157 (19.0%) subjects had impaired glucose tolerance while another 60(7.3%) had diabetes. WithFPG, 270 (32.6%) had impaired fasting glycaemia while another 51(6.2%) had diabetes. Prevalence of prediabetes or diabetes was not different among men and women by OGTT; 55 (23.5%) vs 162 (27.3%), p=0.293while it was significantly different by FPG; 107 (45.7%) vs 214 (36.1%), p=0.011. Prevalence of pre-diabetes or diabetes was significantly higher in age group ≥45 yrs compared to age <45 yrs by both tests. FPG showed low sensitivity (44.7%) but higher (76.1%) specificity of detecting dysglycaemia. FPG showed sensitivity of 53.3% and specificity of 97.5% in detecting undetected diabetes. Conclusions: Prevalence of pre-diabetes is higher than in previous studies in Sri Lanka. Significantly higher proportion of males compared to females, had pre-diabetes or diabetes by FPG test. Compared to subjects age <45 years, the prevalence of pre-diabetes or diabetes was significantly higher among those ≥45 years. FPG lacks adequate sensitivity and specificity to detect those with diabetes or dysglycaemia.