Prevalence of chronic kidney disease and risk factors for its progression: A cross-sectional comparison of Indians living in Indian versus U.S. cities

Abstract

Background: While data from the latter part of the twentieth century consistently showed that immigrants to high-income countries faced higher cardio-metabolic risk than their counterparts in low- and middle-income countries, urbanization and associated lifestyle changes may be changing these patterns, even for conditions considered to be advanced manifestations of cardio-metabolic disease (e.g., chronic kidney disease [CKD]).

Methods and findings: Using cross-sectional data from the Center for cArdiometabolic Risk Reduction in South Asia (CARRS, n = 5294) and Mediators of Atherosclerosis in South Asians Living in America (MASALA, n = 748) studies, we investigated whether prevalence of CKD is similar among Indians living in Indian and U.S. cities. We compared crude, age-, waist-to-height ratio-, and diabetes- adjusted CKD prevalence difference. Among participants identified to have CKD, we compared management of risk factors for its progression. Overall age-adjusted prevalence of CKD was similar in MASALA (14.0% [95% CI 11.8-16.3]) compared with CARRS (10.8% [95% CI 10.0-11.6]). Among men the prevalence difference was low (prevalence difference 1.8 [95% CI -1.6,5.3]) and remained low after adjustment for age, waist-to-height ratio, and diabetes status (-0.4 [-3.2,2.5]). Adjusted prevalence difference was higher among women (prevalence difference 8.9 [4.8,12.9]), but driven entirely by a higher prevalence of albuminuria among women in MASALA. Severity of CKD--i.e., degree of albuminuria and proportion of participants with reduced glomerular filtration fraction--was higher in CARRS for both men and women. Fewer participants with CKD in CARRS were effectively treated. 4% of CARRS versus 51% of MASALA participants with CKD had A1c < 7%; and 7% of CARRS versus 59% of MASALA participants blood pressure < 140/90 mmHg. Our analysis applies only to urban populations. Demographic--particularly educational attainment--differences among participants in the two studies are a potential source of bias.

Conclusions: Prevalence of CKD among Indians living in Indian and U.S. cities is similar. Persons with CKD living in Indian cities face higher likelihood of experiencing end-stage renal disease since they have more severe kidney disease and little evidence of risk factor management.

Fig 1

Prevalence of CKD according to demographic correlates in (A) men and (B) women In MASALA, men with income in the lower tertiles had higher CKD prevalence than men with income in the top tertile. In CARRS, men with no college education had higher CKD prevalence than men with college education. Across studies, men with income in the lower tertiles in CARRS had higher CKD prevalence than men with income in the lower tertiles in the MASALA. Women in the MASALA study had significantly higher prevalence of CKD across nearly all demographic correlates compared with women in CARRS. * denotes statistically significant difference within each study, # denotes statistically significant difference between studies.

Fig 2

Prevalence difference in CKD in the MASALA study from the CARRS study (A) men and (B) women. We present prevalence difference in CKD 1. Unadjusted, 2. Adjusted for age, and 3. Adjusted for age, waist-to-height ratio, and diabetes. Prevalence difference in overall CKD and albuminuria among men in CARRS and MASALA was negligible in all three models; prevalence of eGFR < 60 ml/min/1.73m² was slightly higher in men in CARRS. Unadjusted prevalence in overall CKD and albuminuria among women in MASALA was 11.1% and 11.8% higher respectively compared with CARRS; adjusting for diabetes and waist-to-height ratio did not attenuate this prevalence difference.

Fig 3. Prevalence of risk factors for adverse events, and evidence of their management among participants with CKD.

Of the 558 and 122 participants with CKD in CARRS and MASALA respectively, 430 (77%) and 119 (98%) had complete data on prevalence of risk factors for progression of CKD and/or cardiovascular events. While 43% of participants with CKD in CARRS had diabetes, only 17% were on medications and only 2% (i.e., 4% of those with CKD and diabetes) had A1c < 7.0.