Skip to main content

Table 2 Age-gender–specific 6-year cumulative incidence of diabetes mellitus among the Nakuru eye disease cohort study participants

From: The incidence of diabetes mellitus and diabetic retinopathy in a population-based cohort study of people age 50 years and over in Nakuru, Kenya

 

Male

Female

Overall

Age group (years)

N (Cases / at risk)

Risk per 1000/6 years (95% CI)a

N

Risk per 1000/6 years (95% CI)a

N

Risk per 1000/6 years (95% CI)a

Diabetes Mellitus (N = 2056)

 50–59

24 / 393

63.7(43.0,93.3)

37 / 544

69.6(49.2,97.7)

61 / 937

67.1(52.2,85.8)

 60–69

19 / 328

62.3(39.2,97.6)

22 / 326

67.0(42.6103.8)

41 / 654

64.7(46.3,89.5)

 70–79

8 / 180

48.2(22.2101.5)

9 / 151

57.5(29.9107.8)

17 / 331

52.6(33.0,82.7)

 80+

3 / 66

40.3(12.3123.9)

1 / 68

11.9(1.6,84.9)

4 / 134

25.8(9.6,67.5)

 All ages

54 / 967

58.6(44.7,76.4)

69 / 1089

63.0(48.8,81.1)

123 / 2056

61.0(50.3,73.7)

Diabetic Retinopathy – among those without DM and without DR at baseline (N = 1421)

 50–59

5 / 297

24.6(8.5,68.9)

7 / 394

20.0(7.8,50.3)

12 / 691

22.0(11.0,43.4)

 60–69

5 / 237

22.9(9.8,53.0)

1 / 229

3.9(0.5,28.5)

6 / 466

13.3(5.4,32.8)

 70–79

2 / 123

15.2(3.6,61.4)

0 / 89

2 / 212

8.6(2.1,34.8)

 80 + b

0 / 29

0 / 23

0 / 52

 All ages

12 / 686

20.5(10.9,38.2)

8 / 735

11.5(4.8,27.1)

20 / 1421

15.8(9.5,26.2)

Diabetic Retinopathy – among those with DM at baseline, but without DR at baseline (N = 44)

 50–59

3 / 8

400.4(83.3830.7)

3 / 14

198.9(46.8556.8)

6 / 22

278.3(111.7541.7)

 60–69

4 / 10

409.8(130.2763.1)

0 / 4

4 / 14

268.8(78.8612.2)

 70–79

1 / 4

175.5(2.7943.6)

0 / 2

1 / 6

126.8(6.3770.0)

 80 + b

0 / 1

0 / 1

0 / 2

 All ages

8 / 23

337.7(162.1573.3)

3 / 21

118.8(29.7372.5)

11 / 44

224.7(116.9388.2)

  1. aEstimated using inverse probability weights to account for loss to follow up
  2. bNo-one with DR at follow up among 80+ group
  3. Risk of the outcome in the 6-year follow up, adjusted for loss to follow up using inverse probability weightings
  4. Sample sizes are small for the DR analyses, so estimates have a wide confidence interval