Leveraging the South African Diabetes Prevention Programme to screen for chronic kidney disease: an observational study

Abstract

Objective

To evaluate the viability of leveraging an existing screening programme (the South African Diabetes Prevention Programme (SA-DPP)) to screen for chronic kidney disease (CKD), by assessing the yield of CKD cases among those participating in the programme.

Design

Observational study conducted between 2017 and 2019.

Setting

16 resource–poor communities in Cape Town, South Africa.

Participants

690 participants, aged between 25 and 65 years, identified as at high risk for type 2 diabetes mellitus (T2DM) by the African Diabetes Risk Score.

Primary outcome measure

The prevalence of CKD among those participating in the SA-DPP.

Results

Of the 2173 individuals screened in the community, 690 participants underwent further testing. Of these participants, 9.6% (n=66) and 18.1% (n=125) had screen-detected T2DM and CKD (defined as an estimated glomerular filtration rate (eGFR) of<60 mL/min/1.73 m² and/or albumin-to-creatinine ratio >3 mg/mmol), respectively. Of those with CKD, 73.6% (n=92), 17.6% (n=22) and 8.8% (n=11) presented with stages 1, 2 and 3, respectively. Of the participants with an eGFR <60 mL/min/1.73 m², 36.4% had no albuminuria and of those with normal kidney function (eGFR ≥90 mL/min/1.73 m²), 10.2% and 3.8% had albuminuria stages 2 and 3, respectively. Of those with T2DM and hypertension, 22.7% and 19.8% had CKD, respectively.

Conclusion

The fact that almost one in five participants identified as high risk for T2DM had CKD underscores the value of including markers of kidney function in an existing screening programme. By using an opportunistic approach to screen high-risk individuals, those with CKD can be identified and appropriately treated to reduce disease progression.

Correspondence to Dr Cindy George; [email protected]

STRENGTHS AND LIMITATIONS OF THIS STUDY

• The strength of our study is that both estimated glomerular filtration rate (eGFR) and albuminuria were used to define chronic kidney disease (CKD), unlike most other population-based CKD prevalence studies in South Africa and Africa in general which rely on eGFR only.

• Due to the self-selection approach of recruitment and the disproportionate female participation, our study findings may not be generalisable.

• The small proportion of participants with chronic kidney disease (CKD) in this study resulted in reduced statistical power when analysis was stratifying by CKD stage.

• CKD was defined based on a single time point serum and urinary creatinine and albumin assessment and not on repeated measurements, at least 3 months apart, as per guidelines.

Introduction

Chronic kidney disease (CKD) is a leading cause of morbidity and mortality globally,¹ affecting more than 840 million individuals worldwide.² The increasing burden of CKD is demonstrated by its ascent in ranking among the global causes of disability-adjusted life-years, rising from 29th in 1990 to 18th in 2019 overall, and from 14th to 8th in the older aged groups (aged ≥50 years).³ However, despite being a global problem, the prevalence of CKD is increasing most rapidly in low-income and middle-income countries (LMICs) where the burden of disease is more pronounced.⁴ This is worrisome as the healthcare systems in most LMICs are already under pressure, and options for kidney replacement therapy are not frequently available or affordable.^{5 6} Given the inequity in access to healthcare services, which disproportionally affects disadvantaged populations, and the costs of kidney replacement therapies, early detection of CKD followed by low-cost treatments should be encouraged.⁷

Early-stage CKD presents with no or non-specific symptoms and is commonly diagnosed opportunistically from screening tests for other diseases, or when the disease has progressed, and symptoms appear.⁸ Therefore, screening for CKD plays an important role in early detection, as implementing treatment on diagnosis can slow the rate of kidney function loss and reduce morbidity and mortality.^{9 10} However, there is often a strong argument against community-based CKD screening due to the potential harm arising from screening and the cost implications of such an undertaking. According to a recent study, community-based CKD screening is unlikely to be effective or cost-effective anywhere in the world.¹¹ In contrast, community-based screening for CKD risk factors like hypertension and type 2 diabetes mellitus (T2DM) are deemed effective. Community-based screening programmes for hypertension and T2DM provide an opportunity to incorporate screening for CKD. Certainly, using the screening of hypertension and T2DM, which are common risk factors for CKD, as a gateway for CKD screening in clinical settings will involve minimal additional costs. Furthermore, (1) the yield of screen-detected cases is likely to be high, considering the high prevalence and incidence of CKD in the presence of these risk factors; (2) awareness of the presence of CKD with hypertension or T2DM can prompt the intensification or modification of treatments to enhance kidney protection and prevent CKD progression and (3) a large proportion of people with CKD likely have a combination of suboptimal risk factors with raised levels of blood pressure and/or glucose that fall below the threshold for disease classification. These individuals with pre-diabetes and/or prehypertension are not generally targeted for CKD screening in routine practice but may already have CKD. The opportunistic incorporation of CKD testing in hypertension or T2DM screening programmes can therefore identify CKD that may otherwise be missed if only those with established hypertension or T2DM are screened for the condition.

The aim of this study was to evaluate the viability of CKD screening when incorporated into an existing disease screening programme. The yield of CKD cases in the South African Diabetes Prevention Programme (SA-DPP) was determined by assessing markers of kidney function (serum and urinary creatinine levels and urinary albumin) among participants at high risk for T2DM.

Material and methods

Study population and setting

The SA-DPP is a ‘real-world’ randomised implementation trial, of a structured lifestyle intervention programme, adapted from programmes previously shown to be effective in Finland,¹² Australia¹³ and India.¹⁴ The SA-DPP uses an open-labelled cluster randomised control design, conducted across 16 resource–poor communities in Cape Town, South Africa. Participants were recruited by self-selection approaches, by raising awareness of the study with flyers distributed in the community or through local councillors’ offices, churches and schools. Interested participants were invited to predetermined venues in their community for community-based risk screening. In the current study, baseline data were obtained from black and mixed ancestry participants, aged between 25 and 65 years, who were at high risk for T2DM.¹⁵ The data included in this study were collected between 2017 and 2019 and the details have been previously described.¹⁵

Community-based screening to identify high-risk individuals

For the community-based risk screening, the African Diabetes Risk Score (ADRS),¹⁶ which is a validated African screening tool comprising non-laboratory-based variables including age, waist circumference (WC) and the presence of hypertension, was used to identify adults at high risk for T2DM. Trained fieldworkers administered a brief questionnaire, which included age, gender, population group, and measured anthropometry and blood pressure. Standard anthropometric methods were used to measure weight, height and WC.¹⁷ Body weight (nearest 0.1 kg) was measured with a calibrated Omron digital scale, with the participant in light clothing and without shoes. A stadiometer was used to measure the participant height (nearest cm), with the participant standing in an upright position, on a flat surface. WC was measured using a non-elastic tape measure at the level of the umbilicus. Blood pressure measurements were taken in a seated position after 5 min of rest. The systolic and diastolic blood pressures (SBP and DBP, respectively) were recorded three times at 2 min intervals, using an appropriately sized cuff and an automated blood pressure monitor (Omron 711, Omron Health Care, Hamburg, Germany). An average of the last two readings was used in the analyses.

Clinic-based assessments of high-risk participants

Participants deemed at high risk, based on the ADRS, were invited for further clinical and biochemical assessments. At the clinic, trained fieldworkers administered questionnaires to obtain information on participant sociodemographic and personal and family medical history. Anthropometric and blood pressure measurements were repeated using standardised techniques as described above.

As per the WHO’s guidelines,¹⁸ blood samples were collected after a 10-hour overnight fast by a qualified nurse for the oral glucose tolerance test (OGTT). Following the administration of 75 g anhydrous glucose dissolved in 250 mL, blood samples were taken 2 hours later. Biochemical analyses were conducted at an ISO accredited laboratory (PathCare Laboratories, Cape Town, South Africa). Plasma glucose was determined by the glucose oxidase method (Glucose Analyzer 2, Beckman Instruments, Fullerton, California, USA), serum insulin, determined by a Microparticle Enzyme Immunoassay (AxSym Insulin Kit, Abbot, Illinois, USA) and glycated haemoglobin (HbA1c) was analysed with high-performance liquid chromatography (Biorad Variant Turbo, BioRad, Johannesburg, South Africa). Vitamin D (25(OH)D3) was measured using liquid chromatography mass spectrometry and enzymatic colorimetric methods were used to measure serum calcium, phosphate, aspartate aminotransferase (AST), alanine aminotransferase (ALT) and gamma-glutamyl transferase (GGT). Full blood counts, including total red blood cells (RBC), total white blood cells, haemoglobin, haematocrit and platelets, were measured on a Coulter LH 750 haematology analyser (Beckman Coulter, South Africa).

For the current study, we used the blood and urine samples in the SA-DPP biobank to conduct secondary laboratory analyses. To determine the levels of serum and urinary creatinine, the modified Jaffe-Kinetic method (calibrated to isotope dilution mass spectrometry standards) (Beckman AU, Beckman Coulter, South Africa) was used, and the colorimetric (using bromocresol purple) method (Beckman AU, Beckman Coulter, South Africa) was used to determine the level of urine albumin.

Classification of kidney function and comorbidities

Kidney function was estimated using the serum creatinine-based CKD Epidemiology Collaboration 2009 equation¹⁹, with the race correction factor omitted. CKD was defined as an estimated glomerular filtration rate (eGFR) of <60 mL/min/1.73 m² and/or urinary albumin-to-creatinine ratio (uACR) >3 mg/mmol. CKD staging was based on the Kidney Disease Improving Global Outcomes (KDIGO) guidelines²⁰ as, stage 1 (eGFR ≥90 mL/min/1.73 m² and uACR >3 mg/mmol), stage 2 (eGFR 60–89 mL/min/1.73 m² and uACR >3 mg/mmol) and stage 3 (eGFR<60 mL/min/1.73 m²). Albuminuria (stage 2) was defined as uACR between 3 and 30 mg/mmol and albuminuria (stage 3) as >30 mg/mmol.²¹

Given that GFR declines with healthy ageing without any overt signs of kidney damage, CKD was also defined by an age-adapted definition, as an eGFR <75 mL/min/1.73 m² for participants younger than 40 years, eGFR <60 mL/min/1.73 m² for participants aged between 40 and 65 years and eGFR <45 mL/min/1.73 m² for participants aged greater than 65 years.²² Additionally, the age-standardised prevalence of CKD was calculated, using the standard world population distribution as projected by the WHO for 2000–2025.²³

Body mass index (BMI) was calculated as weight in kilograms divided by height in metres squared (kg/m²). This was categorised as normal weight (BMI ≤24.9 kg/m²), overweight (BMI 25.0–29.9 kg/m²) and obese (BMI ≥30 kg/m²). Hypertension was defined as SBP ≥140 mm Hg and/or DBP ≥90 mm Hg,²⁴ or taking antihypertensive medications. We further categorised our study participants into four groups related to the level of blood pressure control, namely, (1) normotensive (defined as no use of antihypertensive medication and SBP/DBP <140/90 mm Hg), (2) treated and controlled blood pressure (defined as use of antihypertensive medication and SBP/DBP <140/90 mm Hg), (3) treated but uncontrolled blood pressure (defined as use of antihypertensive medication but SBP/DBP ≥140/90 mm Hg) and (4) newly detected hypertension (defined as no use of antihypertensive medication and SBP/DBP ≥140/90 mm Hg). Normal and dysglycaemia categories, based on the OGTT, were defined according to WHO criteria¹⁸ as: (1) normal glucose tolerance (fasting glucose (FG) <6.1 mmol/L and 2-hour glucose <7.8 mmol/L) or (2) pre-diabetes including impaired FG (6.1≤FG<7.0 mmol/L and 2-hour glucose <7.8 mmol/L), impaired glucose tolerance (FG <7.0 mmol/L and 7.8≤2 hours glucose<11.1 mmol/L) and (3) T2DM (FG ≥7.0 mmol/L and/or 2 hours glucose ≥11.1 mmol/L). High GGT was defined as levels >38 IU/L, and based on the laboratory (PathCare, South Africa) reference standards. Liver fibrosis was classified based on the fibrosis-4 (FIB-4) index, where FIB-4 index was calculated using the formula: (age (years) × AST (IU/L))/ (platelet (10⁹ /L) × √ALT (IU/L)).²⁵ Low risk for advanced fibrosis was defined a FIB-4 score <1.30, intermediate risk as a value between 1.30 and 2.67, and high risk as FIB-4 >2.67.²⁶ Anaemia was defined using the National Kidney Foundation Kidney Disease Outcome Quality Initiative guidelines as haemoglobin level <135 g/L for men and <120 g/L for women.²⁷

Statistical analysis

The SA-DPP sample size was calculated based on the following assumptions: (1) a cumulative incident diabetes rate of 13.6% at 2–3 years, as observed in our Bellville South cohort,²⁸ (2) an expected relative risk of 0.51, which is the pooled effect estimate of efficacy trials comparing lifestyle intervention to usual care in diabetes prevention studies,²⁹ (3) an intracluster correlation coefficient for FG of 0.02,³⁰ (4) a significance level of 5% with a type II error risk of 20% and (5) an estimated 36 months lost to follow-up of 20–25%.

Due to the non-Gaussian distribution of most variables, the participant characteristics were summarised as median (25–75th percentile) or counts and percentages. Group comparisons were analysed by χ² tests, Wilcoxon rank-sum and Kruskal-Wallis tests. The Dunn’s test was used as non-parametric pairwise multiple-comparison post hoc test when the Kruskal-Wallis test was rejected. All statistical analyses were performed using STATA V.17 (Statcorp) and statistical significance was based on a p<0.05.

Patient and public involvement

Participants and/or the public were not involved in the design, or conduct, or reporting or dissemination plans of this research.

Results

Of the 2173 individuals screened in the community, 690 participants, deemed at high risk of T2DM based on the ADRS, presented at our research clinic for an OGTT and other assessments (online supplemental file). The sociodemographic, clinical and biochemical characteristics are summarised by CKD status in table 1. Among the 690 participants included in this study, 80.9% were female, with a group median age of 52 years. Of these participants, 9.6% had screen-detected T2DM and 18.1% had CKD, with 2.2% presenting with both CKD and T2DM. A similar CKD prevalence rate was observed with age-adapted eGFR thresholds (18.1%); however, the age-standardised prevalence of CKD was lower, at 14.6%. Furthermore, there were high rates of obesity (77.1%), hypertension (55.0%), raised GGT levels (45.8%), intermediate risk of advanced liver fibrosis (21.4%) and anaemia (14.2%) among participants in this study. There were no significant differences in the sociodemographic and anthropometric variables between participants with and without CKD. However, SBP (128.0 vs 123.5 mm Hg; p=0.004) and DBP (86.0 vs 83.0 mm Hg; p=0.014) were higher in participants with CKD compared with those without. Although hypertension prevalence was not significantly different by CKD status (p=0.215), uncontrolled hypertension on treatment was significantly higher in those with than without CKD (42.7% vs 23.4%). The median levels of GGT (47.0 vs 35.0 IU/L; p=0.008), AST (26.0 vs 23.0 IU/L; p=0.004) and FIB-4 index (1.0 vs 0.9; p=0.016), were higher in participants with CKD compared with those without CKD, while RBC count (4.5 vs 4.6×10¹²/L; p=0.046) was lower in CKD compared with those with normal kidney function. The prevalence of high GGT (p=0.008) and anaemia (p=0.042) were significantly higher in participants with CKD compared with those without CKD. All other biochemical variable were similar between groups.

Table 1

Sociodemographic, clinical and biochemical characteristics presented in the overall sample and by CKD status

Sociodemographic variables	Total (n=690)	Without CKD (n=565)	CKD (n=125)	P value
Age (years)	52 (45–59)	52 (45–59)	53 (47–60)	0.241
Gender (n,% female)	558 (80.9)	460 (81.4)	98 (78.4)	0.438
African Diabetes Risk Score	2.3 (1.7–3.4)	2.3 (1.7–3.4)	2.4 (1.8–3.4)	0.882
Anthropometry
Weight (kg)	91.0 (79.6–103.6)	92.2 (80.4–104.6)	88.0 (76.1–101.3)	0.050
Waist circumference (cm)	102.7 (95.3–111.1)	103.4 (95.7–111.1)	101.3 (93.4–111.1)	0.242
Hip circumference (cm) (n=632)	112.6 (103.2–121.7)	113.0 (104.3–122.4)	111.3 (102.1–118.3)	0.067
Body mass index (kg/m2)	35.6 (30.5–40.5)	35.7 (30.6–40.6)	33.9 (29.4–39.9)	0.185
Body mass index categories (n, %)				0.316
Normal	29 (4.2)	23 (4.1)	6 (4.8)
Overweight	129 (18.7)	100 (17.7)	29 (23.2)
Obese	532 (77.1)	442 (78.2)	90 (72.0)
Blood pressure (BP)
Systolic blood pressure (mm Hg)	124.5 (113.5–137.0)	123.5 (113.5–135.0)	128.0 (116.0–145.5)	0.004
Diastolic blood pressure (mm Hg)	83.0 (77.0–91.5)	83.0 (77.0–90.3)	86.0 (78.5–94.5)	0.014
Hypertension	379 (55.0)	304 (53.9)	75 (60.0)	0.215
Among participants with hypertension (n=379):				<0.0001
Treated and controlled BP	143 (37.7)	127 (41.8)	16 (21.3)
Treated and uncontrolled BP	103 (27.2)	71 (23.4)	32 (42.7)
Screen-detected HPT	133 (35.1)	106 (34.9)	27 (36.0)
Biochemical
Fasting blood glucose (mmol/L)	5.0 (4.6–5.5)	5.0 (4.6–5.5)	5.0 (4.6–5.6)	0.691
2-hour glucose (mmol/L) (n=688)	6.0 (4.9–7.4)	6.0 (4.9–7.3)	6.3 (5.1–7.6)	0.205
Glucose categories (n, %) (n=688)				0.600
Normoglycaemia	520 (75.6)	428 (76.0)	92 (73.6)
Pre-diabetes (IFG/IGT)	102 (14.8)	84 (14.9)	18 (14.4)
Type 2 diabetes	66 (9.6)	51 (9.1)	15 (12.0)
HbA1c (%) (n=685)	5.8 (5.6–6.1)	5.8 (5.6–6.1)	5.9 (5.6–6.2)	0.740
Fasting insulin (IU/L)	8.8 (6.2–12.6)	8.5 (5.9–12.1)	11.1 (7.2–14.8)	0.144
Vitamin D (ng/mL)	6.1 (5.0–7.8)	6.0 (5.0–7.7)	6.2 (5.1–8.1)	0.222
Calcium (mmol/L) (n=688)	2.3 (2.3–2.4)	2.3 (2.3–2.4)	2.4 (2.3–2.4)	0.644
Phosphate (mmol/L) (n=688)	1.1 (1.0–1.2)	1.1 (1.0–1.2)	1.1 (1.0–1.2)	0.981
Gamma-glutamyl transferase (IU/L) (n=688)	36.0 (24.0–61.0)	35.0 (24.0–55.0)	47.0 (26.0–78.0)	0.008
High gamma-glutamyl transferase (n=688)	315 (45.8)	245 (43.4)	70 (56.5)	0.008
Aspartate aminotransferase (IU/L) (n=688)	24.0 (20.0–29.0)	23.0 (20.0–29.0)	26.0 (21.0–34.0)	0.004
Alanine aminotransferase (IU/L) (n=646)	22.0 (16.0–32.0)	22.0 (16.0–32.0)	22.0 (17.0–33.0)	0.372
AST/ALT ratio	1.1 (0.9–1.4)	1.1 (0.9–1.4)	1.2 (0.9–1.5)	0.110
Fibrosis-4 index (n=644)	0.9 (0.7–1.3)	0.9 (0.7–1.3)	1.0 (0.8–1.4)	0.016
Liver fibrosis (n, %)				0.065
No risk	497 (77.2)	413 (78.4)	84 (71.8)
Intermediate risk	138 (21.4)	109 (20.7)	29 (24.8)
High risk	9 (1.4)	5 (0.9)	4 (3.4)
Red blood cells (×1012/L)	4.6 (4.2–4.9)	4.6 (4.3–4.9)	4.5 (4.2–4.8)	0.046
White blood cells (×109/L)	23.0 (18.0–28.0)	23.0 (18.3–28.0)	23.0 (17.0–28.0)	0.270
Platelet count (×109/L)	276 (235–325)	276.0 (234.5–322.5)	276.0 (235.0–333.0)	0.705
Haematocrit (volume %)	0.4 (0.4–0.4)	0.4 (0.4–0.4)	0.4 (0.4–0.4)	0.442
Haemoglobin (g/L)	135 (126–143)	135 (127–143)	134 (124–144)	0.491
Anaemia, n (%)	103 (14.9)	77 (13.6)	26 (20.8)	0.042

Data are presented as median (25–75th percentiles) or count and percentages.

AST/ALT ratio, aspartate aminotransferase to alanine aminotransferase ratio; CKD, chronic kidney disease; HbA1c, glycated haemoglobin; HPT, hypertension; IFG, impaired fasting glucose; IGT, impaired glucose tolerance.

The prevalence of CKD in the overall sample and grouped by glucose and blood pressure categories are shown in figure 1. In those with pre-diabetes, T2DM and hypertension, 17.6%, 22.7% and 19.8% had CKD, respectively. Of the participants with hypertension, the prevalence of CKD was highest in those on antihypertensive treatment but with uncontrolled blood pressure (31.1%), while 20.3% of those newly identified with hypertension and 11.2% of those on treatment with controlled blood pressure had CKD.

Figure 1. Prevalence (%) of chronic kidney disease overall and by glucose and blood pressure categories. Data presented as percentages. BP, blood pressure; HPT, hypertension; T2DM, type 2 diabetes mellitus.

The stages of CKD according to eGFR and albuminuria following KDIGO classification are presented in figure 2. Of the 11 participants with an eGFR <60 mL/min/1.73 m², 4 (36.4%) had no albuminuria, with 36.4% (n=4) and 27.3% (n=3) presenting with moderate (uACR: 3–30 mg/mmol) and severe albuminuria (uACR: >30 mg/mmol), respectively. Furthermore, of the those with normal kidney function (eGFR ≥90 mL/min/1.73 m²), 67 (10.2%) and 25 (3.8%) had moderate and severe albuminuria, respectively.

Figure 2. Stages of chronic kidney disease according to estimated glomerular filtration rate and albuminuria following Kidney Disease Improving Global Outcomes (KDIGO) classification. displayed are number of patients (%) within each category. The colour code indicates risk category according to KDIGO 20 : green ‘low risk’, yellow ‘moderate risk’, orange ‘high risk’ and red ‘very high risk’. CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate.

Table 2 describes the participant characteristics by CKD stage. The majority of individuals with CKD presented with stage 1 CKD (73.6%), with 17.6% and 8.8% presenting with stages 2 and 3, respectively. Participants with stage 3 CKD were older than those with normal kidney function and stage 1 CKD (p=0.030 for both). Levels of AST were significantly higher with stage 2 CKD compared with stage 3 CKD (p=0.042). SBP and DBP did not differ by stages of CKD but differed between those with normal kidney function and those with CKD as follows: normal kidney function vs CKD stage 1 (SBP: p=0.007 and DBP: p=0.010), stage 2 (SBP: p=0.039) and stage 3 (DBP: p=0.013).

Table 2

Sociodemographic, clinical and biochemical characteristics in participants by CKD stages

Sociodemographic variables	No CKD (n=565)	Stage 1 (n=92)	Stage 2 (n=22)	Stage 3 (n=11)	P value
Age (years)	52 (45–59)*	52 (45–59)*	56 (51–61)	57 (52–63)	0.029
Gender (n,% female)	460 (81.4)	75 (81.5)	15 (68.2)	8 (72.7)	0.408
African Diabetes Risk Score	2.3 (1.7–3.4)	2.4 (1.8–3.1)	2.2 (1.7–4.8)	2.8 (1.9–3.9)	0.865
Kidney function
Serum creatinine (µmol/L)	57.0 (48.0–67.0)	54.0 (46.5–62.0)	78.5 (72.0–88.0)	122.0 (96.0–160.0)	0.0001
eGFR (ml/min/1.73 m2)	103.0 (95.0–114.0)	106.0 (98.0–117.5)	79.5 (75.0–83.0)	49.0 (32.0–57.0)	0.0001
uACR (mg/mmol)	0.6 (0.4–1.0)	6.0 (4.1–14.1)	6.5 (3.6–17.3)	3.9 (0.8–43.2)	0.0001
Anthropometry
Weight (kg)	92.2 (80.4–104.6)	89.1 (77.8–101.7)	84.4 (70.6–95.3)	78.7 (63.2–102.4)	0.117
Waist circumference (cm)	103.4 (95.7–111.1)	101.6 (93.9–111.4)	97.2 (93.1–109.7)	100.6 (93.4–107.0)	0.497
Hip circumference (cm) (n=632)	113.0 (104.3–122.4)	112.7 (102.3–120.9)	110.4 (99.4–117.9)	108.6 (96.4–108.9)	0.085
BMI (kg/m2)	35.7 (30.6–40.6)	34.7 (30.5–40.7)	31.6 (26.9–39.5)	31.9 (27.2–36.9)	0.121
BMI categories (n, %)					0.039
Normal	23 (4.1)	2 (2.2)	2 (9.1)	2 (18.2)
Overweight	100 (17.7)	19 (20.7)	8 (36.4)	2 (18.2)
Obese	442 (78.2)	71 (77.2)	12 (54.5)	7 (63.6)
Blood pressure (BP)
SBP (mm Hg)	123.5 (113.5–135.0)	129.5 (115.0–145.5)†	126.5 (123.5–153.0)‡	127.5 (106.5–156.0)	0.031
DBP (mm Hg)	83.0 (77.0–90.3)	86.5 (78.3–94.0)§	80.8 (75.0–94.5)	90.5 (82.5–105.5)¶	0.017
Hypertension	304 (53.9)	54 (58.7)	12 (54.5)	9 (81.8)	0.263
Among participants with hypertension (n=379):					0.010
Treated and controlled BP	127 (41.8)	10 (18.5)	3 (25.0)	3 (33.3)
Treated and uncontrolled BP	71 (23.4)	23 (42.6)	5 (41.7)	4 (44.4)
Screen-detected HPT	106 (34.9)	21 (38.9)	4 (33.3)	2 (22.2)
Biochemical
FBG (mmol/L)	5.0 (4.6–5.5)	5.0 (4.6–5.6)	4.9 (4.4–5.6)	4.8 (4.7–5.3)	0.886
2-hour glucose (mmol/L) (n=688)	6.0 (4.9–7.3)	6.3 (5.1–7.6)	6.3 (4.7–8.5)	6.4 (5.6–7.2)	0.624
Glucose categories (n, %) (n=688)					0.543
Normoglycaemia	428 (76.0)	70 (76.0)	13 (59.1)	9 (81.8)
Pre-diabetes (IFG/IGT)	84 (14.9)	11 (12.0)	6 (27.3)	1 (9.1)
Type 2 diabetes	51 (9.1)	11 (12.0)	3 (13.6)	1 (9.1)
HbA1c (%) (n=685)	5.8 (5.6–6.1)	5.9 (5.6–6.2)	5.7 (5.3–6.2)	5.7 (5.6–6.2)	0.591
Fasting insulin (IU/L)	8.5 (5.9–12.1)	11.1 (6.4–15.5)	11.0 (8.7–13.2)	–	0.334
Vitamin D (ng/mL)	6.0 (5.0–7.7)	6.2 (5.0–7.8)	6.7 (5.9–8.1)	6.8 (5.2–10.6)	0.361
Calcium (mmol/L) (n=688)	2.3 (2.3–2.4)	2.3 (2.3–2.4)	2.4 (2.3–2.4)	2.3 (2.3–2.4)	0.794
Phosphate (mmol/L) (n=688)	1.1 (1.0–1.2)	1.1 (1.0–1.2)	1.1 (1.0–1.2)	1.2 (0.9–1.3)	0.777
GGT (IU/L) (n=688)	35.0 (24.0–55.0)	45.0 (26.0–81.0)	46.5 (25.0–64.0)	49.0 (24.0–122.0)	0.071
High GGT (n=688)	245 (43.4)	51 (56.0)	13 (59.1)	6 (54.5)	0.071
AST (IU/L) (n=688)	23.0 (20.0–29.0)	26.0 (21.1–34.0)	26.5 (22.0–34.0)**	21.0 (20.0–28.0)	0.009
ALT (IU/L) (n=646)	22.0 (16.0–32.0)	23.0 (17.0–33.0)	21.0 (18.0–31.0)	18.5 (15.5–37.5)	0.799
AST/ALT ratio	1.1 (0.9–1.4)	1.2 (0.9–1.5)	1.3 (1.1–1.4)	1.3 (0.9–1.5)	0.413
Fibrosis-4 index (n=644)	0.9 (0.7–1.3)	1.0 (0.8–1.3)	1.1 (0.9–1.5)	1.3 (0.7–1.6)	0.063
Liver fibrosis (n, %)					0.124
No risk	413 (78.4)	66 (75.0)	14 (66.7)	4 (50.0)
Intermediate risk	109 (20.7)	19 (21.6)	6 (28.6)	4 (50.0)
High risk	5 (0.9)	3 (3.4)	1 (4.8)	0 (0)
Red blood cells (×1012/L)	4.6 (4.3–4.9)	4.5 (4.2–4.9)	4.5 (4.2–4.6)	4.7 (4.5–5.1)	0.071
White blood cells (×109/L)	23.0 (18.3–28.0)	22.0 (17.0–28.0)	26.0 (16.0–31.9)	25.0 (19.0–26.0)	0.550
Platelet count (×109/L)	276.0 (234.5–322.5)	276.5 (235.0–333.5)	271.0 (244.0–335.0)	261.0 (217.0–325.0)	0.956
Haematocrit (volume %)	0.4 (0.4–0.4)	0.4 (0.4–0.4)	0.4 (0.4–0.4)	0.4 (0.4–0.5)	0.433
Haemoglobin (g/L)	135 (127–143)	133 (123–145)	135 (133–144)	137 (129–158)	0.390
Anaemia, n (%)	77 (13.6)	22 (23.9)	2 (9.1)	2 (18.2)	0.063

Data are presented as median (25–75th percentiles) or count and percentages.

*p=0.030 (CKD stage 3 vs no CKD; CKD stage 3 vs CKD stage 1).

†p=0.007 (no CKD vs CKD stage 1).

‡p=0.039 (no CKD vs CKD stage 2).

§p=0.010 (no CKD vs CKD stage 1).

¶p=0.013 (no CKD vs CKD stage 3).

**p=0.042 (CKD stage three vs CKD stage 2).

ALT, alanine aminotransferase; AST, aspartate aminotransferase; BMI, body mass index; CKD, chronic kidney disease; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; FBG, fasting blood glucose; GGT, gamma-glutamyl transferase; HbA1c, glycated haemoglobin; IFG, impaired fasting glucose; IGT, impaired glucose tolerance; SBP, systolic blood pressure; uACR, urinary albumin-to-creatinine ratio.

Discussion

To our knowledge, this is the first study to show that by using an opportunistic approach, CKD can be detected early, allowing for timely referral for specialised testing to confirm diagnosis and subsequent care. This was achieved through leveraging the information already collected in an existing screening programme that targeted individuals at high risk for T2DM and included a few additional kidney-related biochemical markers to the variables for testing. The yield of screen-detected cases was high for a low investment which cost ZAR237.80 (US$14.59) per person and highlights the potential cost-effectiveness of such a strategy.

By including a minimal number of markers of kidney function (namely serum and urinary creatinine, and urinary albumin) to the scope of markers already collected, we found that 18.1% of those at high risk for developing T2DM had CKD with the majority (73.6%) having mild CKD (CKD stage 1). The CKD burden, at 22.7%, was even higher in participants with newly diagnosed T2DM, which underscores the need for frequent screening of individuals at high risk for T2DM to avoid T2DM presenting with complications at diagnosis. Therefore, using T2DM as a gateway for CKD screening through existing screening programmes is justified as such an approach, together with diagnosing new T2DM, simultaneously identified those with complications, that is, CKD. The newly diagnosed T2DM may receive comprehensive care with tight control of both their T2DM and CKD. This intensification of treatment could contribute to a delay in CKD progression and consequently help reduce the risk of developing end-stage kidney disease (ESKD) or CVD-related complications.³¹ Further support for CKD screening in individuals at high risk for T2DM was the substantial CKD burden in pre-diabetes (17.6%). Notably, if screening for CKD was initiated only after the development of T2DM, the identification of CKD in individuals with pre-diabetes, which generally fall below the threshold for disease management in clinical practice, would have been missed. This would then have been a lost opportunity to identify and manage CKD early and delay progression of the disease in this high-risk group.

Our study also highlights the importance of screening for albuminuria as 91.2% of those with CKD would have gone undetected if CKD were based on eGFR alone. Guidelines recommend albuminuria testing using ACR, like we did in our study, however, this is not always possible in many low-resource settings. In these instances, low-cost semiquantitative methods, like urinary dipsticks, can be used to measure albuminuria with subsequent confirmation of positive dipstick result with a quantitative laboratory test to confirm CKD diagnosis.²⁰ Or repeated dipstick assessments can be employed to reduce the possibility of false-negative results as this could delay the timely diagnosis and management of CKD.

Given that this is the first study to report the prevalence of CKD in people at high risk for developing T2DM, based on the ADRS, the prevalence estimates cannot be directly compared with other studies as no similar data have been published. Nevertheless, at a similar median age (52 vs 53 years), the prevalence of CKD in those with pre-diabetes in our study was comparable to that reported in a large representative sample in the USA (17.6% vs 17.7%, respectively).³² Also, although an older population (median age of 68 years) with a higher prevalence of advanced CKD (stages 3–5), a South African study found that the prevalence of CKD in those with pre-diabetes was 19.8%.³³ The similarly high CKD prevalence in pre-diabetes across several studies suggests that perhaps there should be regular CKD screening for all individuals with pre-diabetes.

A likely contributor to the substantial CKD burden in this study is the high prevalence of hypertension, which at 55% is higher than the 44%–46% reported for South Africa.³⁴ While the high reported prevalence of hypertension is consequent to the score used to identify high-risk individuals, a larger proportion of the participants with hypertension had CKD compared with those with normal blood pressure (19.8% vs 16.1%, respectively). The prevalence of CKD may be related to the delayed detection of hypertension or the suboptimal control of blood pressure in treated hypertension, as reported in the current study and in several South African studies.^{34 35} Indeed, a high proportion of participants with treated but uncontrolled hypertension had CKD (31.1%) in this study as did participants with newly detected hypertension (20.3%). This further highlights the benefit of screening high-risk individuals for CKD. Notably, adequate blood pressure control is fundamental to slowing the progression of CKD^{36 37} and timeous treatment with antihypertensive medication can improve both kidney and cardiovascular outcomes^{38 39} thereby preventing the progression to ESKD and reducing the risk of all-cause and cardiovascular mortality.^{38 40 41}

Elevated GGT and the FIB-4 index, which are commonly used markers of liver injury and non-alcoholic fatty liver disease,⁴² have been linked to increased CKD risk in various populations.^43–46 In our study, 56.5% of the participants with CKD presented with higher-than-normal GGT levels, compared with 43.4% of participants without CKD. Also, a significant proportion of people with CKD presented with intermediate and high risk for advanced liver fibrosis, based on the FIB-4 index, compared with those without CKD (28.2% vs 21.6%). Early recognition and interventions directed at reducing the risk of liver injury among individuals with CKD could reduce CKD progression.

Anaemia was prevalent in our study population (14.9% of total sample), with nearly twice as many participants with CKD having anaemia compared with those without CKD, as shown in other studies as well.^{47 48} Although the overall prevalence of anaemia in this study was not uncommon for South Africa,⁴⁹ the prevalence in participants with CKD is concerning. While erythropoiesis stimulating agents and iron supplementation to treat anaemia are unlikely to be prescribed to people in the early stages of CKD, anaemia can accelerate the decline in kidney function by causing kidney haemodynamic alterations and tissue hypoxia.⁸ It is strongly predictive of all-cause and cardiovascular mortality,^{50 51} and should thus be closely monitored.

Although lifestyle interventions addressing unhealthy diets, physical inactivity, tobacco smoking and alcohol misuse are advocated to reduce the growing global burden of non-communicable diseases,^{52 53} little is known about the impact of reducing unhealthy lifestyle behaviours on kidney health. The SA-DPP intervention, implemented in individuals with pre-diabetes, will provide a unique opportunity to examine the effects of improving lifestyle behaviours on changes in CKD status.

This is the first study to show that using an opportunistic approach, through leveraging the information already collected in an existing screening programme is advantageous to screen for CKD. However, our study does have limitations. The SA-DPP study included participants at high risk of T2DM and our findings might not be reproducible across other non-communicable diseases screening programmes. The small number of participants identified with CKD in this study reduced the statistical power of our analyses when stratifying by CKD stage. Based on the self-selection approaches used to recruit participants, the disproportionate greater number of females, the varying socioeconomic status, lifestyle behaviours and disease prevalence (hypertension and T2DM) across provinces and by urban–rural residence in South Africa,³⁴ our study findings cannot be generalised. Another limitation is that CKD was defined based on a single time point serum and urinary creatinine and albumin assessment and not on repeated measurements, at least 3 months apart, as per KDIGO guidelines.²⁰ However, a strength of our study is that both eGFR and albuminuria were used to define CKD, unlike most other population-based CKD prevalence studies in South Africa and Africa in general which rely on eGFR only for CKD classification. Finally, as for all studies using eGFR to characterise CKD, instead of the gold standard of measured GFR, the overestimation or underestimation of the estimate cannot be excluded.

Conclusion

The fact that almost one in five participants identified as high risk for T2DM had CKD underscores the value of including markers of kidney function in existing disease screening programmes. Our findings provide support for key stakeholders and policy makers to adapt current strategies for hypertension and T2DM screening to include screening for CKD. Indeed, by using an opportunistic approach to screen high-risk individuals, those with early-stage CKD can be identified and appropriately managed to reduce disease progression. Existing cardiovascular or non-communicable disease screening programmes should perhaps explore including markers for CKD evaluations to maximise limited resources without compromising on effectiveness.

We would like to acknowledge the South African Medical Research Council (SAMRC) for infrastructure and support. The authors would also like to acknowledge the entire South African Diabetes Prevention Programme team and collaborators for their continued support.

Data availability statement

Data are available on reasonable request. The dataset depicted in this manuscript are available from the corresponding author on reasonable request.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

This study involves human participants and was approved by Research Ethics Committee of the South African Medical Research Council (SAMRC) (approval no. EC018-7/2015). Participants gave informed consent to participate in the study before taking part.

Footnote

Contributors Conceived and/or designed the work that led to the submission (CG, JH, NP, APK, UN), acquired data (JH) and/or played an important role in interpreting the results (CG), drafted (CG) or revised the manuscript (all authors), and approved the final version (all authors). All authors accept full responsibility for the work and/or conduct of the study, had access to the data and controlled the decision to publish.

Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

Competing interests None declared.

Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

Provenance and peer review Not commissioned; externally peer reviewed.

Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

References

¹ GBD 2017 Causes of Death Collaborators. Global, regional, and national age-sex-specific mortality for 282 causes of death in 195 countries and territories, 1980-2017: a systematic analysis for the global burden of disease study 2017. Lancet 2018; 392: 1736–88. doi:10.1016/S0140-6736(18)32203-7 http://www.ncbi.nlm.nih.gov/pubmed/30496103

² Jager KJ, Kovesdy C, Langham R, et al. A single number for advocacy and communication-worldwide more than 850 million individuals have kidney diseases. Nephrol Dial Transplant 2019; 34: 1803–5. doi:10.1093/ndt/gfz174 http://www.ncbi.nlm.nih.gov/pubmed/31566230

³ GBD 2019 Diseases and Injuries Collaborators. Global burden of 369 diseases and injuries in 204 countries and territories, 1990-2019: a systematic analysis for the global burden of disease study 2019. Lancet 2020; 396: 1204–22. doi:10.1016/S0140-6736(20)30925-9 http://www.ncbi.nlm.nih.gov/pubmed/33069326

⁴ Ameh OI, Ekrikpo UE, Kengne A-P. Preventing CKD in low- and middle-income countries: a call for urgent action. Kidney Int Rep 2020; 5: 255–62. doi:10.1016/j.ekir.2019.12.013 http://www.ncbi.nlm.nih.gov/pubmed/32154447

⁵ GBD Chronic Kidney Disease Collaboration. Global, regional, and national burden of chronic kidney disease, 1990-2017: a systematic analysis for the global burden of disease study 2017. Lancet 2020; 395: 30045–3. doi:10.1016/S0140-6736(20)30045-3 http://www.ncbi.nlm.nih.gov/pubmed/32061315

⁶ Stanifer JW, Muiru A, Jafar TH, et al. Chronic kidney disease in low- and middle-income countries. Nephrol Dial Transplant 2016; 31: 868–74. doi:10.1093/ndt/gfv466 http://www.ncbi.nlm.nih.gov/pubmed/27217391

⁷ Stanifer JW, Von Isenburg M, Chertow GM, et al. Chronic kidney disease care models in low- and middle-income countries: a systematic review. BMJ Glob Health 2018; 3: e000728. doi:10.1136/bmjgh-2018-000728 http://www.ncbi.nlm.nih.gov/pubmed/29629191

⁸ Webster AC, Nagler EV, Morton RL, et al. Chronic kidney disease. Lancet 2017; 389: 1238–52. doi:10.1016/S0140-6736(16)32064-5 http://www.ncbi.nlm.nih.gov/pubmed/27887750

⁹ Katz IJ, Gerntholtz TE, van Deventer M, et al. Is there a need for early detection programs for chronic kidney disease? Clin Nephrol 2010; 74 Suppl 1: S113–8. http://www.ncbi.nlm.nih.gov/pubmed/20979975

¹⁰ Levin A, Stevens PE. Early detection of CKD: the benefits, limitations and effects on prognosis. Nat Rev Nephrol 2011; 7: 446–57. doi:10.1038/nrneph.2011.86 http://www.ncbi.nlm.nih.gov/pubmed/21712852

¹¹ Tonelli M, Dickinson JA. Early detection of CKD: implications for low-income, middle-income, and high-income countries. J Am Soc Nephrol 2020; 31: 1931–40. doi:10.1681/ASN.2020030277 http://www.ncbi.nlm.nih.gov/pubmed/32839279

¹² Absetz P, Valve R, Oldenburg B, et al. Type 2 diabetes prevention in the "real world": one-year results of the GOAL Implementation Trial. Diabetes Care 2007; 30: 2465–70. doi:10.2337/dc07-0171 http://www.ncbi.nlm.nih.gov/pubmed/17586741

¹³ Aziz Z, Absetz P, Oldroyd J, et al. A systematic review of real-world diabetes prevention programs: learnings from the last 15 years. Implement Sci 2015; 10: 172. doi:10.1186/s13012-015-0354-6 http://www.ncbi.nlm.nih.gov/pubmed/26670418

¹⁴ Thankappan KR, Sathish T, Tapp RJ, et al. A peer-support lifestyle intervention for preventing type 2 diabetes in India: a cluster-randomized controlled trial of the Kerala diabetes prevention program. PLoS Med 2018; 15: e1002575. doi:10.1371/journal.pmed.1002575 http://www.ncbi.nlm.nih.gov/pubmed/29874236

¹⁵ Hill J, Peer N, Jonathan D, et al. Findings from community-based screenings for type 2 diabetes mellitus in at risk communities in Cape town, South Africa: a pilot study. Int J Environ Res Public Health 2020; 17: 2876. doi:10.3390/ijerph17082876 http://www.ncbi.nlm.nih.gov/pubmed/32326364

¹⁶ Mayige M. Derivation and validation of a simple risk score for undiagnosed diabetes for Tanzania and other African populations. Newcastle, UK: Newcastle University, 2014.

¹⁷ Alberti KGMM, Zimmet P, Shaw J. Metabolic syndrome--a new world-wide definition. A Consensus Statement from the International Diabetes Federation. Diabet Med 2006; 23: 469–80. doi:10.1111/j.1464-5491.2006.01858.x http://www.ncbi.nlm.nih.gov/pubmed/16681555

¹⁸ Alberti KG, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a who consultation. Diabet Med 1998; 15: 539–53. doi:10.1002/(SICI)1096-9136(199807)15:7<539::AID-DIA668>3.0.CO;2-S http://www.ncbi.nlm.nih.gov/pubmed/9686693

¹⁹ Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009; 150: 604–12. doi:10.7326/0003-4819-150-9-200905050-00006 http://www.ncbi.nlm.nih.gov/pubmed/19414839

²⁰ Kidney Disease: Improving Global Outcomes (KDIGO) CKD Work Group. KDIGO 2012 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int Suppl 2013; 3: 1–150.

²¹ Levey AS, de Jong PE, Coresh J, et al. The definition, classification, and prognosis of chronic kidney disease: a KDIGO controversies conference report. Kidney Int 2011; 80: 17–28. doi:10.1038/ki.2010.483 http://www.ncbi.nlm.nih.gov/pubmed/21150873

²² Delanaye P, Jager KJ, Bökenkamp A, et al. Ckd: a call for an Age-Adapted definition. J Am Soc Nephrol 2019; 30: 1785–805. doi:10.1681/ASN.2019030238 http://www.ncbi.nlm.nih.gov/pubmed/31506289

²³ Ahmad O, Boschi-Pinto C, Lopez A, et al, eds. Age standardization of rates: a new WHO standard. Geneva: World Health Organization, 2001.

²⁴ World Health Organization. A global brief on hypertension: silent killer, global public health crisis, 2013.

²⁵ Sterling RK, Lissen E, Clumeck N, et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology 2006; 43: 1317–25. doi:10.1002/hep.21178 http://www.ncbi.nlm.nih.gov/pubmed/16729309

²⁶ Shah AG, Lydecker A, Murray K, et al. Comparison of noninvasive markers of fibrosis in patients with nonalcoholic fatty liver disease. Clin Gastroenterol Hepatol 2009; 7: 1104–12. doi:10.1016/j.cgh.2009.05.033 http://www.ncbi.nlm.nih.gov/pubmed/19523535

²⁷ KDOQI, National Kidney Foundation. KDOQI clinical practice guidelines and clinical practice recommendations for anemia in chronic kidney disease. Am J Kidney Dis 2006; 47: S11–145. doi:10.1053/j.ajkd.2006.03.010 http://www.ncbi.nlm.nih.gov/pubmed/16678659

²⁸ Matsha TE, Soita DJ, Hassan MS, et al. Three-Year's changes in glucose tolerance status in the Bellville South cohort: rates and phenotypes associated with progression. Diabetes Res Clin Pract 2013; 99: 223–30. doi:10.1016/j.diabres.2012.10.018 http://www.ncbi.nlm.nih.gov/pubmed/23199814

²⁹ Gillies CL, Abrams KR, Lambert PC, et al. Pharmacological and lifestyle interventions to prevent or delay type 2 diabetes in people with impaired glucose tolerance: systematic review and meta-analysis. BMJ 2007; 334: 299. doi:10.1136/bmj.39063.689375.55 http://www.ncbi.nlm.nih.gov/pubmed/17237299

³⁰ Parker DR, Evangelou E, Eaton CB. Intraclass correlation coefficients for cluster randomized trials in primary care: the cholesterol education and research trial (CEART). Contemp Clin Trials 2005; 26: 260–7. doi:10.1016/j.cct.2005.01.002 http://www.ncbi.nlm.nih.gov/pubmed/15837446

³¹ Stevens PE, Levin A, Kidney Disease: Improving Global Outcomes Chronic Kidney Disease Guideline Development Work Group Members. Evaluation and management of chronic kidney disease: synopsis of the kidney disease: improving global outcomes 2012 clinical practice guideline. Ann Intern Med 2013; 158: 825–30. doi:10.7326/0003-4819-158-11-201306040-00007 http://www.ncbi.nlm.nih.gov/pubmed/23732715

³² Plantinga LC, Crews DC, Coresh J, et al. Prevalence of chronic kidney disease in US adults with undiagnosed diabetes or prediabetes. Clin J Am Soc Nephrol 2010; 5: 673–82. doi:10.2215/CJN.07891109 http://www.ncbi.nlm.nih.gov/pubmed/20338960

³³ George C, Matsha TE, Korf M, et al. The agreement between fasting glucose and markers of chronic glycaemic exposure in individuals with and without chronic kidney disease: a cross-sectional study. BMC Nephrol 2020; 21: 32. doi:10.1186/s12882-020-1697-z http://www.ncbi.nlm.nih.gov/pubmed/32000712

³⁴ National Department of Health, Statistics South Africa, South African Medical Research Council, ICF. South Africa demographic and health survey 2016, 2019.

³⁵ Peer N, Steyn K, Lombard C, et al. A high burden of hypertension in the urban black population of Cape town: the cardiovascular risk in black South Africans (CRIBSA) study. PLoS One 2013; 8: e78567. doi:10.1371/journal.pone.0078567 http://www.ncbi.nlm.nih.gov/pubmed/24250798

³⁶ Ruggenenti P, Cravedi P, Remuzzi G. Mechanisms and treatment of CKD. J Am Soc Nephrol 2012; 23: 1917–28. doi:10.1681/ASN.2012040390 http://www.ncbi.nlm.nih.gov/pubmed/23100218

³⁷ Maione A, Navaneethan SD, Graziano G, et al. Angiotensin-Converting enzyme inhibitors, angiotensin receptor blockers and combined therapy in patients with micro- and macroalbuminuria and other cardiovascular risk factors: a systematic review of randomized controlled trials. Nephrol Dial Transplant 2011; 26: 2827–47. doi:10.1093/ndt/gfq792 http://www.ncbi.nlm.nih.gov/pubmed/21372254

³⁸ Schmieder RE, Schutte R, Schumacher H, et al. Mortality and morbidity in relation to changes in albuminuria, glucose status and systolic blood pressure: an analysis of the ONTARGET and TRANSCEND studies. Diabetologia 2014; 57: 2019–29. doi:10.1007/s00125-014-3330-9 http://www.ncbi.nlm.nih.gov/pubmed/25037746

³⁹ de Zeeuw D, Remuzzi G, Parving H-H, et al. Albuminuria, a therapeutic target for cardiovascular protection in type 2 diabetic patients with nephropathy. Circulation 2004; 110: 921–7. doi:10.1161/01.CIR.0000139860.33974.28 http://www.ncbi.nlm.nih.gov/pubmed/15302780

⁴⁰ Matsushita K, Coresh J, Sang Y, et al. Estimated glomerular filtration rate and albuminuria for prediction of cardiovascular outcomes: a collaborative meta-analysis of individual participant data. Lancet Diabetes Endocrinol 2015; 3: 514–25. doi:10.1016/S2213-8587(15)00040-6 http://www.ncbi.nlm.nih.gov/pubmed/26028594

⁴¹ Adler AI, Stevens RJ, Manley SE, et al. Development and progression of nephropathy in type 2 diabetes: the United Kingdom prospective diabetes study (UKPDS 64). Kidney Int 2003; 63: 225–32. doi:10.1046/j.1523-1755.2003.00712.x http://www.ncbi.nlm.nih.gov/pubmed/12472787

⁴² Niemelä O, Alatalo P. Biomarkers of alcohol consumption and related liver disease. Scand J Clin Lab Invest 2010; 70: 305–12. doi:10.3109/00365513.2010.486442 http://www.ncbi.nlm.nih.gov/pubmed/20470213

⁴³ Ryu S, Chang Y, Kim D-I, et al. Gamma-Glutamyltransferase as a predictor of chronic kidney disease in nonhypertensive and nondiabetic Korean men. Clin Chem 2007; 53: 71–7. doi:10.1373/clinchem.2006.078980 http://www.ncbi.nlm.nih.gov/pubmed/17110470

⁴⁴ Targher G, Kendrick J, Smits G, et al. Relationship between serum gamma-glutamyltransferase and chronic kidney disease in the United States adult population. findings from the National health and nutrition examination survey 2001-2006. Nutr Metab Cardiovasc Dis 2010; 20: 583–90. doi:10.1016/j.numecd.2009.05.012 http://www.ncbi.nlm.nih.gov/pubmed/19699624

⁴⁵ Kunutsor SK, Laukkanen JA. Gamma-Glutamyltransferase and risk of chronic kidney disease: a prospective cohort study. Clinica Chimica Acta 2017; 473: 39–44. doi:10.1016/j.cca.2017.08.014

⁴⁶ Lee DY, Han K, Yu JH, et al. Gamma-Glutamyl transferase variability can predict the development of end-stage of renal disease: a nationwide population-based study. Sci Rep 2020; 10: 11668. doi:10.1038/s41598-020-68603-0 http://www.ncbi.nlm.nih.gov/pubmed/32669624

⁴⁷ George C, Matsha TE, Erasmus RT, et al. Haematological profile of chronic kidney disease in a mixed-ancestry South African population: a cross-sectional study. BMJ Open 2018; 8: e025694. doi:10.1136/bmjopen-2018-025694 http://www.ncbi.nlm.nih.gov/pubmed/30391922

⁴⁸ Akinsola A, Durosinmi MO, Akinola NO. The haematological profile of Nigerians with chronic renal failure. Afr J Med Med Sci 2000; 29: 13–16. http://www.ncbi.nlm.nih.gov/pubmed/11379460

⁴⁹ Maluleke T, Shisana O, Shisana O, et al. The South African National health and nutrition examination survey (SANHANES-1), 2013.

⁵⁰ Babitt JL, Lin HY. Mechanisms of anemia in CKD. J Am Soc Nephrol 2012; 23: 1631–4. doi:10.1681/ASN.2011111078 http://www.ncbi.nlm.nih.gov/pubmed/22935483

⁵¹ Levey AS, Coresh J. Chronic kidney disease. Lancet 2012; 379: 165–80. doi:10.1016/S0140-6736(11)60178-5 http://www.ncbi.nlm.nih.gov/pubmed/21840587

⁵² Galaviz KI, Narayan KMV, Lobelo F, et al. Lifestyle and the prevention of type 2 diabetes: a status report. Am J Lifestyle Med 2018; 12: 4–20. doi:10.1177/1559827615619159 http://www.ncbi.nlm.nih.gov/pubmed/30202378

⁵³ Alouki K, Delisle H, Bermúdez-Tamayo C, et al. Lifestyle interventions to prevent type 2 diabetes: a systematic review of economic evaluation studies. J Diabetes Res 2016; 2016: 1–14. doi:10.1155/2016/2159890 http://www.ncbi.nlm.nih.gov/pubmed/26885527