Skip to main content
. 2020 Dec 17;3:559927. doi: 10.3389/frai.2020.559927

TABLE 3.

Q-Finder results on the detection of predictive factors describing patients with a higher risk than the others in experiencing hypoglycemia under Premixed insulin than under Basal insulin (with or without Prandial insulin).

Subgroup Ranking a Subgroup description Coverage Discovery/Test Adjusted odds-ratios for differential treatment effect (IC95%) Discovery b p-value for differential treatment effect Discovery Adjusted odds-ratios for differential treatment effect (IC95%) Test b p-value for differential treatment effect Test Adjusted p-value for differential treatment effect Test c
S1 Statins for dyslipidemia = Yes AND Device used for insulin: Vials and syringes = No AND Total # of anti-diabetics agents 1 28%/31% 3.04 [1.9; 5.0] 7.02E-6 2.12 [1.4; 3.2] 2.36E-4 1.18E-3
S2 Device used for insulin: Disposable pen = Yes AND Smoking habits = Never AND Total # of anti-diabetics agents 1 25%/26% 3.31 [2.0; 5.6] 7.13E-6 1.93 [1.3; 2.9] 2.04E-3 4.28E-3
S3 Total # of anti-diabetics agents 1 AND # of different devices used by the patient 1 48%/61% 2.71 [1.8; 4.2] 9.55E-6 2.59 [1.7; 4.0] 1.92E-5 1.92E-4
S4 Treated for other form of dyslipidemia = Yes AND Times seen by a diabetologist in the past 3 months 1 AND Device used for insulin: Vials and syringes = No 33%/38% 3.55 [2.0; 6.3] 1.26E-5 1.93 [1.2; 3.0] 5.02E-3 7.17E-3
S5 Receives oral glycaemic lowering drugs = Yes AND Times seen by a diabetologist in the past 3 months = 0 AND Device used for insulin: Vials and syringes = No 29%/34% 2.98 [1.8; 4.9] 2.40E-5 1.86 [1.2; 2.8] 2.14E-3 4.28E-3
S6 Statins for dyslipidemia = Yes AND Total # of anti-diabetics agents 1 AND Age at diagnosis (year) 56 30%/33% 2.74 [1.7; 4.4] 2.64E-5 2.04 [1.4; 3.0] 4.08E-4 1.34E-3
S7 Treated for other form of dyslipidemia = Yes AND Times seen by a diabetologist in the past 3 months 1 AND # of different devices used by the patient 1 33%/44% 3.37 [1.9; 6.0] 2.79E-5 2.05 [1.2; 3.4] 4.58E-3 7.17E-3
S8 Statins for dyslepidemia = Yes AND Device used for insulin: Vials and syringes = No AND HDL serum cholesterol (mg/dL) 58.0 27%/30% 3.22 [1.9; 5.6] 2.82E-5 1.05 [0.7; 1.7] 8.21E-1 8.21E-1
S9 Statins for dyslipidemia = Yes AND Visits diabetes websites = No AND Duration of insulin therapy (year) 4 34%/32% 2.59 [1.7; 4.1] 3.12E-5 1.14 [0.8; 1.7] 5.09E-1 5.65E-1
S10 Other form of dyslipidemia = Yes AND Visits diabetes websites = No AND Duration of insulin therapy (year) 4 40%/37% 2.56 [1.6; 4.0] 3.22E-5 1.25 [0.9; 1.8] 2.48E-1 3.10E-1
a

Subgroup ranking is based on p-value for differential treatment effect on discovery dataset

b

Odds-ratios are adjusted for confounding factors through multiple regression model

c

Adjusted p-values for multiple testing are based on a Benjamini-Hochberg procedure on the test dataset