Mar 19, 2018 - mortality at ages 35-74 years from a vascular, renal or infectious cause, by A) ... with no diagnosed or undiagnosed diabetes at recruitment. 10.
Supplementary appendix This appendix formed part of the original submission and has been peer reviewed. We post it as supplied by the authors. Supplement to: Herrington W G, Alegre-Díaz J, Wade R, et al. Effect of diabetes duration and glycaemic control on 14-year cause-specific mortality in Mexican adults: a blood-based prospective cohort study. Lancet Diabetes Endocrinol 2018; published online March 19. http://dx.doi.org/10.1016/S2213-8587(18)30050-0.
Effect of diabetes duration and glycaemic control on 14year cause-specific mortality in Mexican adults: a bloodbased prospective cohort study Supplementary Material, Table of Contents Supplementary figures Sex-specific relevance of previously-diagnosed and, separately, undiagnosed diabetes to mortality at ages 35-74 years from a vascular, renal or infectious cause, by A) duration of diabetes; and B) glycaemic control Age-, sex-, diabetes-duration- and glycaemic-control-specific relevance of previouslydiagnosed and, separately, undiagnosed diabetes to: ‘Other’ (i.e., not vascular, renal or infectious) causes of death except acute diabetic crises
Page 3
4-5
All-cause mortality Relevance of previously-diagnosed diabetes to mortality at ages 35-74 years from vascular, renal and infectious causes:
6-9
by duration of diabetes and age at risk by glycaemic control and age at risk by other characteristics by duration of diabetes and glycaemic control Relevance of baseline HbA1c to cause-specific mortality at ages 35-84 among participants with no diagnosed or undiagnosed diabetes at recruitment
10
Mortality rates at ages 35-74 years for each cause of death among participants with previously-diagnosed diabetes, undiagnosed diabetes and no diabetes at recruitment
11
Supplementary Tables Number of vascular, renal and infectious deaths at ages 35-74 years by underlying cause (ICD-10 code), among 133 662 participants with no other chronic disease (other than diabetes) at recruitment
12
Death rate ratios associated with previously diagnosed and undiagnosed diabetes for mortality from vascular, renal, and infectious causes (separately and combined) at ages 35–74 years, by A) duration of diabetes and B) HbA1c - Comparison of ‘conventional’ 95% confidence intervals with ‘group-specific’ 95% confidence intervals
13
Baseline characteristics among men and women aged 75-84 without chronic disease at recruitment (other than diabetes)
14
Page 2 of 14
Webfigure 1: Sex-specific relevance of previously-diagnosed and, separately, undiagnosed diabetes to mortality at ages 35-74 years from a VASCULAR, RENAL or INFECTIOUS cause by A) duration of diabetes; and B) glycaemic control (A) 19.1
Previously-diagnosed diabetes:
(B)
- Men
16
16
- Men 365
- Women 7.3
395 5.0
346
594 427
4
4
553
223
1.7
1.7
2
2
RR (95% CI)
173 4.7
- Women 264 7.1
8
8
6.7
11.3
9.4
509
449
328
11.0
12.0
10.9
Previously-diagnosed diabetes:
16.5
Men: No diabetes
Men: No diabetes
794
794
1.0
1.0
Women: No diabetes
1
1
Women: No diabetes 915
0
915
5
10
15
Duration of diabetes, years
20
5
6
7
8
9
10
11
12
13
HbA1c, %
Rate ratios exclude data from any participants who had previously diagnosed chronic disease other than diabetes (chronic kidney disease, ischaemic heart disease, stroke, cirrhosis, cancer or emphysema) and are adjusted for age at risk, smoking status, district, educational level, height, weight, and waist and hip circumferences. In addition, the rate ratio estimates for men and women with previously-diagnosed diabetes (black filled squares) are also adjusted for the effect on risk of any A) HbA1c differences or B) diabetes duration differences between those groups. The 95% confidence intervals reflect the variance of the log risk in that 1 group, taking into account the variance of the log risk in the reference group without diabetes. The rate ratio and confidence interval estimates for those with undiagnosed diabetes are not plotted, but would be shown at 3.0 (2.6-3.5) for women with undiagnosed diabetes and 5.3 (4.5-6.1) for men with undiagnosed diabetes. In each panel, the vertical distance between the two horizontal dashed lines is about the same as the vertical distance between the two solid lines, demonstrating that the relative risk for those with previously-diagnosed diabetes versus no diabetes was about the same for men and women, and that this was true irrespective of diabetes duration or glycaemic control. Page 3 of 14
Webfigure 2: Relevance of previously-diagnosed and, separately, undiagnosed diabetes to mortality from OTHER causes except acute diabetic crises by: A) age and sex; B) duration of diabetes; and C) glycaemic control
(C) Deaths at ages 35-74 years
1.1 (1.0–1.4) 0.8 (0.7–1.0) 1.0 (0.9–1.1)
190/498 110/522 300/1020 1
2
4
RR
8
1.3
1.6
Previouslydiagnosed diabetes
2
2
Previouslydiagnosed diabetes 1.3
1.4
1.1 158 167
188 162 269
1.0 2187
0
1.5
1.3
1.1
5
Undiagnosed diabetes No diabetes
10
15
Duration of diabetes, years
20
1.0
276
1
75-84 Women Men Both
1.4 (1.2–1.6) 1.2 (1.0–1.4) 1.3 (1.1–1.4)
257/645 168/590 425/1235
1
60-74 Women Men Both
1.7 (1.3–2.1) 1.7 (1.3–2.2) 1.7 (1.4–2.0)
91/538 73/414 164/952
RR (95% CI)
35-59 Women Men Both
4
RR (95% CI) for previouslydiagnosed diabetes vs no diabetes (diagnosed or undiagnosed) 4
Age at No. deaths risk with/without (years) diabetes
8
(B) Deaths at ages 35-74 years 8
(A) Deaths at ages 35-84 years
188
2187
5
6
7
8
146
Undiagnosed diabetes No diabetes
9
10
11
12
HbA1c, %
See Figure 1 for details of exclusions, analyses and conventions. The rate ratios for participants with undiagnosed diabetes (ie, no previous diagnosis but baseline HbA1c ≥ 6.5%) compared with participants without diabetes were 1.5 (1.21.9) at ages 35-59, 1.0 (0.81.2) at ages 60- 74 and 1.2 (1.01.4) at ages 75-84, and were similar in men and women. Page 4 of 14
13
Webfigure 3: Relevance of previouslydiagnosed and, separately, undiagnosed diabetes to ALLCAUSE mortality by: A) age and sex; B) duration of diabetes; and C) glycaemic control
(B) Deaths at ages 3574 years
RR (95% CI) for previously diagnosed diabetes vs no diabetes (diagnosed or undiagnosed)
3559 Women Men Both
516/897 440/701 956/1598
5.8 (5.2–6.6) 6.0 (5.2–6.8) 5.9 (5.4–6.4)
6074 Women Men Both
1291/1228 897/1118 2188/2346
3.6 (3.3–3.9) 3.3 (3.0–3.6) 3.5 (3.3–3.7)
7584 Women Men Both
905/1273 570/1188 1475/2461
2.1 (2.0–2.3) 1.9 (1.7–2.1) 2.0 (1.9–2.2)
8
Previously diagnosed diabetes
6.4
1125 4.0
4
3.9 3.2
1
2
4
RR
8
5.9
1110
4
Age at No. deaths risk with/without (years) diabetes
(C) Deaths at ages 3574 years
Previously diagnosed diabetes
8
(A) Deaths at ages 3584 years
1184
Undiagnosed diabetes
595
2
2.1
2
2.1
601
Undiagnosed diabetes
595
1.0
1
1.0
1
RR (95% CI)
835
1433
2.9
No diabetes
3944
0
5
10
15
Duration of diabetes, years
20
No diabetes
3944
5
6
7
8
9
10
11
12
HbA1c, %
See Figure 1 for details of exclusions, analyses and conventions. The rate ratios for participants with undiagnosed diabetes (ie, no previous diagnosis but baseline HbA1c ≥ 6.5%) compared with participants without diabetes were 2.5 (2.22.9) at ages 3559, 1.8 (1.62.0) at ages 6074 and 1.4 (1.21.5) at ages 7584, and were similar in men and women. Page 5 of 14
13
Webfigure 4: Relevance of previously-diagnosed diabetes to mortality at ages 35-74 years from VASCULAR, RENAL and INFECTIOUS causes, by duration of diabetes and age at risk
(B) RENAL
100.9 60-74 yrs 88.9 285 35-59 yrs
64
64
(A) VASCULAR
53.3 34.6
38.3
60-74 yrs 21.3
16
60-74 yrs: No diabetes
101
4
703 3.9
54
8
62
5.6
180
82
35-59 yrs
6.5
261
18.1
12.3
296
16 8
265
131
4
RR (95% CI)
18.4
32
32
32.8
118
3.5 60-74 yrs: No diabetes 127
2
2
46
1.0
1
35-59 yrs: No diabetes
35-59 yrs: No diabetes
1
1.0 348
97
0
5
10
15
20
5
10
15
20
64
(d) VASCULAR, RENAL OR INFECTIOUS
64
(C) INFECTIOUS
0
41.6 24.2
16
35
8
8
35-59 yrs
77
5.2
4
4
35
25.4
685
261
11.8
6.8
358
128
48 6.3 4.3
109
19.0
16
15.4
11.9 9.3
25.0
60-74 yrs
16.2
RR (95% CI)
32
32
60-74 yrs 659 35-59 yrs
215
135 60-74 yrs: No diabetes
1093
60-74 yrs: No diabetes
2
2
263
1.0
35-59 yrs: No diabetes
616
171
0
1.0
1
1
35-59 yrs: No diabetes
5
10
15
Duration of diabetes, years
20
0
5
10
15
20
Duration of diabetes, years
Exclusions, analyses and conventions as for panel B of Figure 1, with the exception that rate ratios are now shown separately for deaths at ages 35-59 and deaths at ages 60-74. The age- specific death rate ratios and CIs corresponding to undiagnosed diabetes are not shown in this plot, but the average RRs at ages 35-74 are given in panel B of figure 1 and panels A-C of figure 2. Page 6 of 14
Webfigure 5: Relevance of previously-diagnosed diabetes to mortality at ages 35-74 years from VASCULAR, RENAL and INFECTIOUS causes, by glycaemic control and age at risk
104.3
(B) RENAL
60-74 yrs
48.8 227
196
21.3 82
202
16
16
35-59 yrs 161 31.6
198
19.3 292
74
91 60-74 yrs: No diabetes
4.3
703
8
35-59 yrs
5.6
6.5
3.5
4
8
8.4
4
RR (95% CI)
60-74 yrs
32
32
32.3 26.6
240 62.6
60.8
64
64
(A) VASCULAR
55
60-74 yrs: No diabetes
63
2
2
127
1.0
1.0
35-59 yrs: No diabetes
1
1
35-59 yrs: No diabetes
348
5
97
6
7
8
9
10
11
12
13
6
7
8
9
10
11
12
13
64
(d) VASCULAR, RENAL OR INFECTIOUS
64
(C) INFECTIOUS
5
42.3 60-74 yrs
35-59 yrs 352
10.1
8
16 8
18.5
7.9
65
51
442
35-59 yrs
168 188
5.2
4.3
529
634
16
79 6.9
31
60-74 yrs: No diabetes
1093
4
4
RR (95% CI)
91 12.9
13.7 115
22.4
60-74 yrs
18.3
7.6
29.6
32
32
25.4
60-74 yrs: No diabetes
2
2
263
1.0
1.0 616
171
5
35-59 yrs: No diabetes
1
1
35-59 yrs: No diabetes
6
7
8
9
10
HbA1c, %
11
12
13
5
6
7
8
9
10
11
12
13
HbA1c, %
Exclusions, analyses and conventions as for panel C of Figure 1, with the exception that rate ratios are now shown separately for deaths at ages 35-59 and deaths at ages 60-74. The age- specific death rate ratios and CIs corresponding to undiagnosed diabetes are not shown in this plot, but the average RRs at ages 35-74 are given in panel C of figure 1 and panels A-C of figure 3. Page 7 of 14
Webfigure 6: Relevance of previouslydiagnosed diabetes to mortality at ages 3574 years from a VASCULAR, RENAL or INFECTIOUS cause, by particular characteristics Median HbA1c with/without diabetes, %
No. deaths with/without diabetes
Death RR (95% CI) for previously diagnosed diabetes vs no diabetes
35−59
9.5/5.4
708/616
11.5 (10.0–13.1)
60−74
8.4/5.6
1605/1093
5.5 (5.1–5.9)
Women
9.0/5.4
1326/915
7.2 (6.6–7.9)
Men
8.8/5.4
987/794
6.7 (6.1–7.4)
Coyoácan
8.8/5.4
694/618
7.2 (6.5–8.1)
Iztapalapa
9.0/5.4
1619/1091
6.8 (6.3–7.4)
Never
8.9/5.4
1062/726
7.1 (6.4–7.8)
Former
8.7/5.4
544/357
6.8 (5.9–7.8)
Current
9.2/5.4
707/626
6.9 (6.2–7.8)
University
8.6/5.4
147/194
8.0 (6.4–9.9)
High school
9.0/5.4
358/370
8.1 (6.9–9.3)
Elementary
9.0/5.5
1440/906
7.1 (6.5–7.7)
Other
8.9/5.5
368/239
5.0 (4.2–5.9)
I (23.9)
9.7/5.4
829/404
10.7 (9.5–12.1)
II
9.2/5.4
570/347
8.9 (7.8–10.2)
III IV (34.5)
8.6/5.4
446/405
5.9 (5.2–6.8)
8.3/5.6
468/553
4.5 (3.9–5.1)
2313/1709
7.0 (6.7–7.3)
Variable Age
Sex
District
Smoking
Education
BMI (kg/m2)
Overall
1
2
4
8
16
RR
Rate ratios exclude data from any participants who had previouslydiagnosed chronic disease other than diabetes (chronic kidney disease, ischaemic heart disease, stroke, cirrhosis, cancer, or emphysema) and are adjusted for or, where appropriate, stratified by, age at risk, sex, district, smoking status, educational level and body mass index (split into 4 equallysized groups). Rate ratios are provided with their 95% confidence intervals and compare participants with previouslydiagnosed diabetes with participants with no previous diagnosis and baseline HbA1c
