Background
Methods
Reporting
Literature review
Validation cohort
Outcome
Data source
Study population
Predictors and missing data
Data analysis
Sensitivity analyses
Results
CKD prediction models included for external validation
Authors [ref] Publication year | Study design/Study context Study period | Ethnicity Age range | Population size Number (%) of CKD cases | Type of models Time horizon | Handling of missing values Method of internal validation | Definition of CKD | Predictors in model |
---|---|---|---|---|---|---|---|
Bang et al. [54] 2007 | Cross-sectional population-based survey/Screening programme 1999-2002 | US, mixed 20–85 years | 8530 601 (7.5 %) | Logistic 2 years | Excluded Random split-sample | At least one eGFR measurement < 60a | Age, sex, anaemia, proteinuriaa, hypertension, diabetes mellitus, history of cardiovascular disease, history of heart failure, peripheral vascular disease |
Chien et al. [51] 2010 | Prospective cohort study/ Secondary care 2003 | Taiwan, Chinese 51.2 years (mean) | 5168 190 (3.7) | Cox 4 years | NR NR | At least one eGFR measurement < 60a | Age, BMI, diastolic blood pressure, type 2 diabetes, history of stroke |
Hippisley-Cox and Coupland (QKidney®) [36] 2010 | Prospective cohort population based/Primary care 2002-2008 | UK, mixed 35–74 years | 1,591,884 23,786 (1.5 %) | Cox 5 years | Multiple imputation Random split-sample | At least one eGFR measurement < 45a, kidney transplant; dialysis; nephropathy diagnosis; proteinuria | Age, ethnicity, deprivation, smoking, BMI, systolic blood pressure, diabetes mellitus, rheumatoid arthritis, cardiovascular disease, treated hypertension, congestive cardiac failure, peripheral vascular disease, NSAID use, and family history of kidney disease |
Kshirsagar et al. [53] 2008 | Prospective cohort study/ Community-based 1987-1989 | US, white and black 45–64 years | 9470 1605 (16.9 %) | Logistic 9 years | NR Random split sample | At least one eGFR measurement < 60a | Age, sex, anaemia, hypertension, type 2 diabetes mellitus, history of cardiovascular disease, history of heart failure, peripheral vascular disease |
Kwon et al. [55] 2012 | Cross-sectional survey/ Population-based 2007-2009 | Korean, Asian ≥19 years | 6565 100 (1.5 %) | Logistic 1 year | Excluded Split sample | At least one eGFR measurement < 60a | Age, sex, anaemia, proteinuriaa, hypertension, type 2 diabetes mellitus, history of cardiovascular disease |
O’Seaghdha et al. [52] 2011 | Prospective cohort study/ Population-based 1995-2008 | US white 45–64 years | 2490 229 (9.2 %) | Logistic 10 years | Excluded Bootstrap | At least one eGFR measurement < 60a | Age, hypertension, diabetes mellitus |
Thakkinstian et al. [56] 2011 | Cross-sectional survey/ Community-based NR | Thailand-Asian ≥ 18 years | 3459 606 (17.5 %) | Logistic 1 year | NR Bootstrap | At least one eGFR measurement < 90a | Age, hypertension, diabetes mellitus, kidney stones |
Parameters | No CKD | CKD | |||||
---|---|---|---|---|---|---|---|
Patients with complete follow-up | Patients with incomplete follow-up | ||||||
Missing | Missing | Missing | |||||
Included patients | 156,615 | None | 172,361 | None | 6038 | None | |
Died before developing CKD | 719 (0.5) | None | 6941 (4) | None | / | None | |
Follow-up (mean, SD) | 5.6 (0.2) | None | 5.4 (0.7) | None | 2.6 (1.7) | None | |
Age (mean, SD) | 42.1 (16.7) | None | 42.7 (17.3) | None | 70.3 (12.5) | None | |
Female sex | 82,883 (52.9) | None | 89,389 (51.9) | None | 3452 (57.2) | None | |
Townsend index (mean, SD)e | 1.6 (3.5) | 2900 (1.9) | 1.6 (3.4) | 3244 (1.9) | 1.4 (3.4) | 47 (0.8) | |
Ethnicity | Not recorded | 55,586 (35.6) | Not applicabled | 61,220 (35.6) | Not applicabled | 2014 (33.4) | Not applicabled |
White | 90,443 (57.8) | 99,243 (57.7) | 3889 (64.5) | ||||
Other | 10,586 (6.8) | 11,898 (6.9) | 135 (2.2) | ||||
Smokinge | Non-smoker | 66,769 (48.8) | 19,901 (12.7) | 72,137 (48.4) | 23,296 (13.5) | 2167 (37.7) | 292 (4.8) |
Ex-smoker | 29,980 (21.9) | 33,097 (22.2) | 2475 (43.1) | ||||
Light smoker (1–9 cg/day) | 11,072 (8.1) | 12,128 (8.1) | 344 (6) | ||||
Moderate smoker (10–19 cg/day) | 16,951 (12.4) | 18,472 (12.4) | 413 (7.2) | ||||
Heavy smoker (≥ 20 cg/day) | 11,942 (8.7) | 13,231 (8.9) | 347 (6) | ||||
BMI, kg/m2 (mean, SD)e | 26.6 (6) | 33,717 (21.5) | 28 (6.1) | 38,628 (22.4) | 28.4 (6) | 518 (8.6) | |
Diastolic blood pressure, mmHg (mean, SD)e | 76.9 (9.8) | 75,616 (48.3) | 78.9 (10.2) | 85,075 (49.4) | 75.8 (10.2) | 1164 (19.3) | |
Systolic blood pressure, mmHg (mean, SD)e | 128.2 (15.8) | 75,602 (48.3) | 130.5 (16.7) | 85,058 (49.3) | 136.3 (16.7) | 1166 (19.3) | |
eGFR, mL/min/1.73 m2 (mean, SD) | 83.7 (9.4) | 118,912 (75.9) | 82.5 (9.4) | 131,103 (76.1) | 69.4 (11.3) | 1828 (30.3) | |
Hb, g/dLe | 13.9 (1.6) | 110,723 (70.7) | 13.8 (1.6) | 122,430 (71) | 13.4 (1.7) | 2530 (41.9) | |
Proteinuriaa,b | 751 (0.5) | 149,234 (95.3) | 18 (0.2) | 164,097 (95.2) | 236 (3.9) | 4665 (77.3) | |
Quantitative albuminuriab,c | 129 (0.1) | 152,266 (97.2) | 4 (0) | 167,482 (97.2) | 62 (1) | 5167 (85.6) | |
HDL cholesterol levelb, mg/dL (mean, SD) | 25.9 (7.9) | 122,477 (78.2) | 26.7 (7.9) | 135,066 (78.4) | 25.7 (7.8) | 2413 (40) |
Study population characteristics
CKD risk factors | No CKD | CKD (n = 6038) | |
---|---|---|---|
Patients with complete follow-up (n = 156,615) | Patients with incomplete follow-up (n = 172,361) | ||
Hypertensiona | 22,074 (14.1) | 24,971 (14.5) | 3554 (58.9) |
Hypertensive treatmentb | 22,122 (14.1) | 24,769 (14.4) | 3655 (60.5) |
Type 1 diabetes mellitusa | 703 (0.4) | 740 (0.4) | 36 (0.6) |
Type 2 diabetes mellitusa | 5574 (3.6) | 6383 (3.7) | 1221 (20.2) |
History of cardiovascular diseasea | 11,096 (7.1) | 13,407 (7.8) | 2182 (36.1) |
History of heart failurea | 743 (0.5) | 1088 (0.6) | 387 (6.4) |
History of strokea | 1875 (1.2) | 2538 (1.5) | 509 (8.4) |
Peripheral vascular diseasea | 2127 (1.4) | 2532 (1.5) | 331 (5.5) |
Kidney stonesa | 751 (0.5) | 814 (0.5) | 64 (1.1) |
Rheumatoid arthritisa | 1321 (0.8) | 1512 (0.9) | 142 (2.4) |
Systemic lupus erythematosusa | 99 (0.1) | 104 (0.1) | 8 (0.1) |
Family history of kidney diseasea | 25 (0) | 28 (0) | 3 (0) |
NSAID useb | 5101 (3.3) | 5389 (3.1) | 402 (6.7) |
Acute kidney injury in the last 2 years | 1975 (1.3) | 2633 (1.5) | 413 (6.8) |
Prostatic hypertrophya | 967 (0.6) | 1143 (0.7) | 173 (2.9) |
Haematuriaa | 3176 (2) | 3574 (2.1) | 341 (5.6) |
Lithium useb | 150 (0.7) | 219 (0.1) | 52 (0.9) |
Tacrolimus useb | 4 (0) | 5 (0) | 2 (0) |
Cyclosporin useb | 12 (0.1) | 20 (0) | 6 (0.1) |
External validation
Study | Patients with complete follow-up (n = 162,653) | Full validation cohort (n = 178,399) | ||||
---|---|---|---|---|---|---|
AUC (95 % CI) | MAPE (SD)a | Calibration slope (CI) | c-index (95 % CI) | MAPE (SD)a | ||
Models | Bang et al. [54] | 0.899 (0.895–0.903) | 0.063 (0.162) | 0.97 (0.96–0.98) | NA | NA |
Chien et al. [51]b | 0.898 (0.895–0.901) | 0.081 (0.162) | 0.65 (0.64–0.65) | 0.888 (0.885–0.892) | 0.085 (0.166) | |
QKidney® [36]b | 0.910 (0.907–0.913) | 0.05 (0.166) | 1.02 (1.01–1.04) | 0.900 (0.897–0.903) | 0.052 (0.165) | |
Kshirsagar et al. [53] | 0.896 (0.892–0.900) | 0.068 (0.164) | 1.74 (1.72–1.76) | NA | NA | |
Kwon et al. [55] | 0.899 (0.895–0.902) | 0.086 (0.158) | 0.68 (0.67–0.69) | NA | NA | |
O’Seaghdha et al. [52] | 0.907 (0.904–0.911) | 0.089 (0.169) | 0.53 (0.52–0.53) | NA | NA | |
Thakkinstian et al. [56] | 0.892 (0.888–0.985) | 0.179 (0.161) | 0.44 (0.43–0.45) | NA | NA | |
Simplified Scores | Bang et al. [54] | 0.895 (0.891–0.899) | NA | NA | NA | NA |
Chien et al. [51] | 0.880 (0.876–0.883) | NA | NA | NA | NA | |
Kshirsagar et al. [53] | 0.891 (0.887–0.895) | NA | NA | NA | NA | |
Kwon et al. [55] | 0.895 (0.891–0.898) | NA | NA | NA | NA | |
Thakkinstian et al. [56] | 0.869 (0.864–0.873) | NA | NA | NA | NA |
Study | Threshold (SD) | PPV (SD) | Sensitivity (SD) | Specificity (SD) | |
---|---|---|---|---|---|
Bang et al. [54] | Proposed | 4 | 0.146 (0.002) | 0.865 (0.004) | 0.805 (0.001) |
Best | 4 | 0.146 (0.002) | 0.865 (0.004) | 0.805 (0.001) | |
Chien et al. [51] | Proposed | 7 | 0.106 (0.001) | 0.916 (0.003) | 0.701 (0.001) |
Best | 8 | 0.133 (0.002) | 0.863 (0.004) | 0.783 (0.001) | |
QKidney® [36] | Proposed | NR | NA | NA | NA |
Best | 0.017 (0.002) | 0.147 (0.006) | 0.870 (0.012) | 0.805 (0.012) | |
Kshirsagar et al. [53] | Proposed | 3 | 0.143 (0.002) | 0.872 (0.004) | 0.799 (0.001) |
Best | 3 | 0.143 (0.002) | 0.872 (0.004) | 0.799 (0.001) | |
Kwon et al. [55] | Proposed | 4 | 0.147 (0.002) | 0.862 (0.004) | 0.807 (0.001) |
Best | 4 | 0.147 (0.002) | 0.862 (0.004) | 0.807 (0.001) | |
O’Seaghdha et al. [52] | Proposed | NA | NA | NA | NA |
Best | 0.086 (0.010) | 0.138 (0.007) | 0.885 (0.015) | 0.786 (0.015) | |
Thakkinstian et al. [56] | Proposed | 5 | 0.071 (0.001) | 0.936 (0.003) | 0.529 (0.001) |
Best | 6 | 0.140 (0.002) | 0.861 (0.004) | 0.796 (0.001) |
Study | Development dataset | Validation dataset, patients with complete follow-up (n = 162,653) | ||
---|---|---|---|---|
Mean linear predictor (from summary statistics) | Mean linear predictor (from summary statistics) | Mean linear predictor (SD) (from individual patient data) | ||
Models | Bang et al. [54] | −3.9 | −4.2 | −4.2 (1.4) |
Chien et al. [51] | 0.1 | −0.5 | −0.5 (1.5) | |
QKidney® [36] | −0.1 | −0.3 | −0.1 (1.9) | |
Kshirsagar et al. [53] | −3.0 | −3.5 | −3.5 (0.8) | |
Kwon et al. [55] | −3.0 | −3.4 | −3.3 (1.2) | |
O’Seaghdha et al. [52] | −1.6 | −1.8 | −1.8 (0.9) | |
Thakkinstian et al. [56] | −2.3 | −3.8 | −3.8 (1.9) |