Skip to main content

Table 3 Sensitivity analyses of the association between statin use and sepsis among patients with T2DM

From: Optimal statin use for prevention of sepsis in type 2 diabetes mellitus

Subpopulation or exposure

No. of patients

Sepsis

No. of Sepsis

aHR*

95% CI

P value

Age group, y

  ≤ 50

283,159

19,762

0.34

(0.31, 0.35)

 < 0.0001

 51–60

218,267

21,908

0.35

(0.32, 0.36)

 < 0.0001

 61–70

164,508

26,945

0.38

(0.36, 0.40)

 < 0.0001

  ≥ 71

146,486

40,870

0.39

(0.35, 0.42)

 < 0.0001

Sex

 Female

380,418

51,342

0.37

(0.35, 0.38)

 < 0.0001

 Male

432,002

58,142

0.36

(0.34, 0.37)

 < 0.0001

Income levels (NTD)

 1. Low income

12,041

3265

0.36

(0.31, 0.40)

 < 0.0001

 2. Financial dependent

251,412

55,673

0.37

(0.35, 0.44)

 < 0.0001

 3.  ≤ 20 000

385,432

5389

0.36

(0.34, 0.40)

 < 0.0001

 4. 20 001–30 000

76,087

2998

0.38

(0.34, 0.41)

 < 0.0001

 5. 30 001–45 000

55,238

1357

0.35

(0.30, 0.40)

 < 0.0001

 6.  > 45 000

32,210

40,803

0.41

(0.34, 0.49)

 < 0.0001

Urbanization

 Rural

230,663

38,083

0.36

(0.34, 0.37)

 < 0.0001

 Urban

581,757

71,402

0.37

(0.35, 0.38)

 < 0.0001

Types of antidiabetic drugs used

 Zero

294,033

30,139

0.40

(0.36, 0.45)

 < 0.0001

 One type

203,068

26,415

0.39

(0.36, 0.42)

 < 0.0001

 Combined two types

203,069

28,121

0.36

(0.35, 0.40)

 < 0.0001

 Combined three types

81,498

16,909

0.32

(0.29, 0.36)

 < 0.0001

  ≥ 4 types

30,752

7902

0.29

(0.27, 0.34)

 < 0.0001

Antidiabetic drugs

 Insulin

104,152

21,872

0.44

(0.40, 0.48)

 < 0.0001

 Metformin

378,934

68,208

0.28

(0.26, 0.35)

 < 0.0001

 SU

393,213

82 574

0.37

(0.35, 0.40)

 < 0.0001

 AGI

49,355

9,874

0.36

(0.33, 0.38)

 < 0.0001

 TZD

35,201

7,043

0.37

(0.33, 0.39)

 < 0.0001

 DPP4i

637

144

0.41

(0.31, 0.51)

 < 0.0001

 SGLT2i

6,157

1,354

0.42

(0.30, 0.49)

 < 0.0001

 Others

46,509

21,500

0.39

(0.34, 0.41)

 < 0.0001

aDCSI score

 0

427,080

38 752

0.40

(0.33, 0.43)

 < 0.0001

 1

172,323

20,273

0.38

(0.35, 0.40)

 < 0.0001

 2

119,930

23,183

0.34

(0.32, 0.38)

 < 0.0001

  ≥ 3

93,087

27,276

0.32

(0.30, 0.34)

 < 0.0001

CCI score

 0

427,080

38,752

0.38

(0.36, 0.40)

 < 0.0001

  ≥ 1

366,467

61,426

0.34

(0.33, 0.41)

 < 0.0001

Coexisting comorbidities

 Hypertension

377,673

65,991

0.37

(0.35, 0.38)

 < 0.0001

 Rheumatoid arthritis

24,373

4497

0.35

(0.30, 0.39)

 < 0.0001

 Ankylosing spondylitis

11,821

1905

0.36

(0.30, 0.41)

 < 0.0001

 Psoriasis

6003

1037

0.32

(0.25, 0.39)

 < 0.0001

 Psoriatic arthritis

553

134

0.14

(0.06, 0.31)

 < 0.0001

 Crohn’s disease

11,478

1863

0.35

(0.29, 0.40)

 < 0.0001

 Ulcerative colitis

1745

287

0.36

(0.23, 0.53)

 < 0.0001

 COPD

154,478

30,107

0.36

(0.33, 0.38)

 < 0.0001

 Chronic liver disease

183,748

25,026

0.32

(0.30, 0.36)

 < 0.0001

 Chronic kidney disease

15,948

5053

0.33

(0.31, 0.36)

 < 0.0001

 Heart failure

44,693

12,089

0.36

(0.33, 0.38)

 < 0.0001

 Coronary artery disease

167,000

32,251

0.38

(0.35, 0.40)

 < 0.0001

 Stroke

96,745

24,094

0.33

(0.31, 0.35)

 < 0.0001

 Coagulopathy

1321

330

0.17

(0.12, 0.24)

 < 0.0001

 Dementia

17,003

5743

0.31

(0.27, 0.34)

 < 0.0001

 Psychosis

1656

372

0.48

(0.34, 0.68)

 < 0.0001

 Ankylosing spondylitis

11,821

1905

0.34

(0.30, 0.41)

 < 0.0001

 SLE

14,718

3224

0.32

(0.28, 0.36)

 < 0.0001

 Cancer

33,684

6,733

0.30

(0.25, 0.39)

 < 0.0001

DDD

  ≤ 1

764,110

104,395

0.33

(0.32, 0.37)

 < 0.0001

  > 1

48,309

5089

0.57

(0.45, 0.69)

 < 0.0001

  1. DDD defined daily dose, AIDS acquired immunodeficiency syndrome, CCI Charlson comorbidity index, COPD chronic obstructive pulmonary disease, SLE systemic lupus erythematosus, NTD New Taiwan Dollar, aDCSI adapted Diabetic Complication Severity Index, aHR adjusted hazard ration, CI confidence interval, SU Sulfonylureas, AGI Alpha glucosidase inhibitors, TZD Thiazolidinedione, DPP4i Dipeptidyl peptidase 4 inhibitors, SGLT2i Sodium-glucose cotransporter-2 inhibitors
  2. *The aHR was derived from the inverse probability-weighted Cox model considering statin use as a time-dependent covariate, and the model was adjusted for age groups, sex, income levels, urbanization, types of antidiabetic drugs used, antidiabetic drugs, diabetic severity (aDCSI score), coexisting comorbidities, medication use, and CCI scores