Table 3.
Within-sibling associations between cardiometabolic outcomes and ADHD-PGS (unique individuals N = 17 692).
| Condition | Prev. (%) / Mean (SD) | OR/Beta (95%CI) | % change log(OR)/Beta | Pbetween-within sib |
|---|---|---|---|---|
| Metabolic disease | 3926 (25.47%) | 1.108 (1.023, 1.199)a | −1.72 | 0.03a |
| Cardiovascular disease | 556 (3.58%) | 1.055 (0.879, 1.266) | +19.06 | 0.31 |
| Biomarkers | ||||
| Glucose (mmol/L) | 4.86 (0.50) | −0.005 (−0.031, 0.020) | −129.41 | 0.90 |
| HbA1c (%) | 5.47 (0.31) | −0.005 (−0.029, 0.019) | −50.00 | 0.02a |
| Haematocrit (v/v) | 0.42 (0.03) | 0.015 (−0.006, 0.036) | −21.05 | 0.15 |
| Haemoglobin (mmol/L) | 8.76 (0.77) | 0.016 (−0.005, 0.036) | −11.11 | 0.05a |
| Total cholesterol (mmol/L) | 5.02 (1.01) | 0.001 (−0.022, 0.025) | −93.33 | 0.08 |
| LDL cholesterol (mmol/L) | 3.19 (0.93) | −0.001 (−0.024, 0.022) | −105.88 | 0.26 |
| HDL cholesterol (mmol/L) | 1.48 (0.37) | −0.022 (−0.045, 0.001) | −40.54 | 7.30e-8a |
| Triglycerides (mmol/L) | 1.08 (0.56) | 0.022 (−0.003, 0.048) | −42.11 | 7.36e-3a |
| Apolipo B100 (g/L) | 0.92 (0.24) | 0.032 (−1.017, 0.165) | −25.71 | 0.96 |
| ALT (U/L) | 21.86 (11.16) | 0.002 (−0.039, 0.043) | −88.23 | 0.66 |
| Alkaline Phosphatase (U/L) | 61.36 (16.57) | 0.018 (−0.023, 0.059) | −48.57 | 0.15 |
| Gamma-GT (U/L) | 23.79 (14.37) | 0.026 (−0.015, 0.066) | −36.58 | 0.02a |
| Creatinine (µmol/L) | 72.90 (11.87) | −0.011 (−0.031, 0.009) | −31.25 | 9.12e-3a |
| Creatinine 24 h urine (mmol/L) | 8.23 (3.78) | 0.023 (−0.001, 0.047) | −8.00 | 2.99e-8a |
| Uric acid (mmol/L) | 0.29 (0.07) | −0.001 (−0.036, 0.033) | −105.26 | 0.65 |
| hsCRP (mg/L) | 2.08 (2.45) | −0.011 (−0.062, 0.039) | −142.31 | 0.11 |
| Leucocytes (10^9/L) | 5.97 (1.50) | 0.031 (0.004, 0.058) | −22.50 | 7.30e-8a |
| Systolic BP | 124.96 (15.69) | −0.008 (−0.031, 0.015) | −140.00 | 0.31 |
| Diastolic BP | 73.39 (9.95) | 0.003 (−0.021, 0.028) | −87.50 | 0.85 |
Fixed effects models controlling for genotyping chip, 8 PCs, PC*chip, sex, age.
Average attenuation in within-sibling (compared to between-sibling) models for cardiometabolic measures: −56%.
The within-sibling analyses are performed for outcomes that showed significant association with ADHD-PGS in main analyses (Figs. 1, 2). Did not include apolipoproteins due to low sample sizes.
BP Blood pressure, hsCRP High-sensitivity C-reactive protein, Apolipo Apolipoprotein, Gamma-GT Gamma-glutamyl transferase, ALT Alanine transaminase, HDL high-density lipoprotein, LDL low-density lipoprotein, HbA1c Haemoglobin A1c.
aStatistically significant (p < 0.05) FDR-adjusted p value. Pbetween-within sib = statistical comparison of ADHD-PGS estimate in between-sibling versus within-sibling model.