Table 3 Sensitivity and sub-cohort analyses for the event rate and hazard ratio (95% confidence interval) of study outcomes for intermediate-acting human insulin versus long-acting insulin analogue (reference group).

From: A nationwide cohort study for comparative vascular safety of long-acting insulin analogue versus intermediate-acting human insulin in type 2 diabetes

 

Event rate per 1000 person-years (no. of events)

Adjusted HRc (95% CI)

IAHI

LAIA

Sensitivity analysis 1: intention-to-treat analysis

(nb = 8479)

(nb = 8479)

 

Composite CVDs

4.92 (158)

2.94 (103)

1.73 (1.35–2.23)

Three-point MACEa

3.19 (103)

1.90 (67)

1.73 (1.27–2.36)

Microvascular diseases

569.57 (6344)

657.86 (6988)

0.88 (0.85–0.91)

 Nephropathy

182.47 (3947)

188.15 (4305)

0.93 (0.89–0.97)

 Retinopathy

226.19 (4208)

273.15 (5063)

0.85 (0.81–0.88)

 Neuropathy

116.58 (2737)

116.64 (2978)

0.97 (0.92–1.02)

Hospitalized hypoglycemia

27.29 (836)

25.33 (849)

1.06 (0.96–1.17)

All-cause mortality

2.62 (85)

1.56 (55)

1.68 (1.19–2.38)

 Fatal CVD

1.29 (42)

0.76 (27)

1.78 (1.09–2.90)

Sensitivity analysis 2: 3 years of maximum observational time

(nb = 8479)

(nb = 8479)

 

Composite CVDs

7.53 (46)

3.46 (36)

2.12 (1.35–3.32)

Three-point MACEa

3.92 (24)

1.34 (14)

2.68 (1.37–5.26)

Microvascular diseases

1188.53 (3978)

1116.59 (5049)

0.88 (0.84–0.91)

 Nephropathy

423.51 (2021)

339.86 (2560)

0.89 (0.84–0.94)

 Retinopathy

419.23 (1967)

434.59 (3000)

0.85 (0.80–0.90)

 Neuropathy

252.57 (1322)

194.30 (1667)

0.94 (0.87–1.01)

Hospitalized hypoglycemia

38.09 (229)

19.87 (204)

1.81 (1.49–2.20)

All-cause mortality

2.45 (15)

0.48 (5)

5.02 (1.79–14.08)

 Fatal CVD

1.31 (8)

0.19 (2)

6.18 (1.28–29.93)

Sensitivity analysis 3: lag-time analysis

(nb = 8479)

(nb = 8479)

 

Composite CVDs

7.43 (48)

4.56 (56)

1.77 (1.18–2.64)

Three-point MACEa

3.71 (24)

2.52 (31)

1.70 (0.97–2.96)

Microvascular diseases

1001.51 (3742)

944.64 (4917)

0.90 (0.86–0.94)

 Nephropathy

384.40 (1969)

293.92 (2573)

0.92 (0.87–0.98)

 Retinopathy

354.65 (1783)

371.27 (2904)

0.84 (0.79–0.89)

 Neuropathy

220.58 (1232)

164.60 (1655)

0.94 (0.87–1.01)

Hospitalized hypoglycemia

34.33 (218)

18.76 (227)

1.79 (1.48–2.17)

All-cause mortality

2.32 (15)

1.87 (23)

1.73 (0.87–3.42)

 Fatal CVD

1.24 (8)

1.62 (20)

1.16 (0.49–2.75)

Sensitivity analysis 4: excluding matched sets with a difference in propensity score of > 0.1

(nb = 7451)

(nb = 7451)

 

Composite CVDs

7.59 (45)

4.91 (53)

1.77 (1.17–2.67)

Three-point MACEa

3.70 (22)

2.78 (30)

1.64 (0.93–2.91)

Microvascular diseases

1104.57 (3483)

1061.14 (4510)

0.86 (0.82–0.90)

 Nephropathy

386.91 (1777)

304.02 (2322)

0.90 (0.85–0.96)

 Retinopathy

394.09 (1753)

409.54 (2724)

0.83 (0.78–0.88)

 Neuropathy

1104.57 (3483)

1061.14 (4510)

0.86 (0.82–0.90)

Hospitalized hypoglycemia

35.58 (207)

18.81 (200)

1.83 (1.49–2.23)

All-cause mortality

2.02 (12)

2.03 (22)

1.53 (0.74–3.18)

 Fatal CVD

1.18 (7)

1.85 (20)

1.53 (0.74–3.18)

Sub-cohort analysis: excluding the patients with cardiovascular or microvascular disease history

(nb = 3779)

(nb = 3779)

 

Composite CVDs

2.84 (9)

0.56 (3)

3.99 (1.02–15.56)

Three-point MACEa

1.57 (5)

0.00 (0)

N/A

Microvascular diseases

493.36 (1074)

452.65 (1488)

0.90 (0.83–0.97)

 Nephropathy

132.61 (375)

100.49 (487)

1.10 (0.96–1.26)

 Retinopathy

257.70 (653)

247.19 (959)

0.86 (0.78–0.95)

 Neuropathy

100.44 (292)

91.06 (441)

0.88 (0.76–1.03)

Hospitalized hypoglycemia

28.57 (89)

10.04 (53)

2.49 (1.76–3.53)

All-cause mortality

0.00 (0)

0.19 (1)

N/A

 Fatal CVD

0.00 (0)

0.00 (0)

N/A

  1. IAHI, intermediate-acting human insulin; LAIA, long-acting insulin analogue; HR, hazard ratio; CI, confidence interval; CVD, cardiovascular disease; MACE, major adverse cardiovascular events.
  2. aThree-point MACE included non-fatal myocardial infarction, non-fatal stroke, and death due to cardiovascular diseases.
  3. b“n” refers to the number of stable use sets.
  4. cThe variables adjusted in these analyses were age, gender, hospital grade, history of retinopathy, and cerebrovascular disease, which are shown to be statistically different between IAHI and LAIA users at baseline (in terms of standardized mean difference value of > 0.1) in Table 1.