Table 1.
Data sources on metabolic and lifestyle factors and drug targets
| Exposure | Instrumental variables | Unit | Participants included in analysis | Sample overlap (%) | Adjustments |
| Metabolic factor | |||||
| Body mass index33 | 312 | SD | 806 834 individuals of European descent | 4.1 | Age, sex, and genetic 1-5 principal components |
| Waist-to-hip ratio33 | 581 | SD | 697 734 individuals of European descent | 4.7 | Age, sex, and genetic 1-5 principal components |
| Type 2 diabetes34 | 497 | One unit in log-transformed odds | 228 499 type 2 diabetes cases and 1 178 783 non-cases of multi-ancestries | 0.0 | Age, sex, and the first 10 genetic principal components |
| Fasting insulin35 | 38 | log pmol/L | Up to 196 991 individuals of European descent | 0.0 | BMI, study specific covariates, and principal components |
| Fasting glucose35 | 71 | mmol/L | Up to 196 991 individuals of European descent | 0.0 | BMI index, study specific covariates, and principal components |
| Systolic blood pressure36 | 228 | 10 mm Hg | Up to 1 006 863 individuals of European descent | 3.3 | Age, sex, BMI, genotyping chips |
| HDL cholesterol37 | 473 | SD | 403 943 individuals of European descent | 8.2 | Age, sex, and genotyping chips |
| LDL cholesterol37 | 199 | SD | 440 546 individuals of European descent | 7.5 | Age, sex, and genotyping chips |
| Triglycerides37 | 392 | SD | 441 016 individuals of European descent | 7.5 | Age, sex, and genotyping chips |
| Lifestyle factor | |||||
| Smoking initiation38 | 314 | SD in log-transformer odds | 1 232 091 individuals of European descent | 2.7 | Age, sex, and the first 10 genetic principal components |
| Age of smoking initiation38 | 7 | SD | Up to 262 990 individuals of European descent | 12.5 | Age, sex, and the first 10 genetic principal components |
| Cigarettes per day38 | 19 | SD | Up to 263 954 individuals of European descent | 12.5 | Age, sex, and the first 10 genetic principal components |
| Lifetime smoking index9 | 126 | SD change of lifetime smoking index | 462 690 individuals of European descent | 7.1 | Genotyping chip and sex |
| Alcohol drinking38 | 84 | SD increase of log-transformed alcoholic drinks/week | 941 280 individuals of European descent | 3.5 | Age, sex, and the first 10 genetic principal components |
| Coffee consumption39 | 12 | 50% change | 375 833 individuals of European descent | 8.8 | Age, sex, BMI, total energy, proportion of typical food intake, and 20 genetic principal components |
| Caffeine consumption40 | 24 | SD | 362 316 individuals of European descent | 9.1 | Age, sex, genotyping array, and the first 30 genetic principal components |
| Strenuous sports41 | 6 | ≥ 2–3 v 0 days/week | 350 492 individuals of European descent | 9.4 | Age, sex, genotyping chip, first 10 genomic principal components, centre, and season |
| Television watching42 | 112 | Hours/day | 422 218 individuals of European descent | 7.8 | Age squared, age, sex, age-sex interaction, the first 30 principal components |
| Drug target | |||||
| Lipid lowering target10 | One for HMGCR, two for LDLR, one for NPC1L1, two for PCSK9 |
10 mg/dL LDL cholesterol | 80 959 to 295 826 individuals of European descent | 0 | Age, sex, study sample, and five principal components |
| Antihypertensive target11 | One for ACEi, six for β blockers, 23 for CCBs |
10 mm Hg systolic blood pressure | 757 601 individuals of European descent | 4.4 | Age squared, age, sex, BMI, genotyping chips |
Sample overlap was calculated by the sample size of the outcome (n=32 974) divided by the sample size of the exposure whose data sources included UK Biobank. ACEi=angiotensin converting enzyme inhibitor; BMI=body mass index; CCBs=calcium channel blockers; HDL=high density lipoprotein; LDL=low density lipoprotein; SD=standard deviation.