Table 1.

Variables planned to be included in IPD meta-analysis modelling*

Type	Analysis 1	Analysis 2	Analysis 3	Analysis 4
Outcomes/DV	Androgen concentration†	Incident CVD‡	Incident deaths (cancer‡)	Incident dementia
		Incident deaths (CVD‡)	Incident cancer‡ (diagnoses)	Baseline cognition
		Incident deaths (all-cause)		Change in cognition
Focal predictor	–	Androgen concentration†	Androgen concentration†	Androgen concentration†
Covariates/IV demographic	Age	Age	Age	Age
	Education level	Education level	Education level	Education level
	Ethnicity	Ethnicity	Ethnicity	Ethnicity
	Marital status	Marital status	Marital status	Marital status
	Site	Site	Site	Site
Risk factors and  comorbidities	Alcohol consumption	Alcohol consumption	Alcohol consumption	Alcohol consumption
	BMI, waist	BMI, waist	BMI, waist	BMI, waist
	Physical activity	Physical activity	Physical activity	Physical activity
	Smoking status	Smoking status	Smoking status	Smoking status
	BP, hypertension	BP, hypertension	BP, hypertension	BP, hypertension
	General health	General health	General health	General health
		Atrial fibrillation
	Prevalent CVD	Prevalent CVD
	Prevalent cancer‡		Prevalent cancer‡
	Prevalent dementia			Prevalent dementia
	Baseline cognition
	COPD	COPD
	Diabetes	Diabetes	Diabetes	Diabetes
	Cholesterol, LDL, HDL	Cholesterol, LDL, HDL
	Creatinine level	Creatinine level
	Lipid lowering medications	Lipid lowering medications
	Anxiety			Anxiety
	Depression			Depression
	Psychotropic drug use			Psychotropic drug use

*Black font: highly desirable; Green font: only if available.

†Androgens: total testosterone, dihydrotestosterone, oestradiol, luteinising hormone, sex hormone binding globulin.

‡Subgroup analyses are also planned. For CVD outcomes: heart failure, myocardial infarction, stroke. For cancer outcomes: colorectal cancer, lung cancer, prostate cancer.

BMI, body mass index; BP, blood pressure; COPD, chronic obstructive pulmonary disease; CVD, cardiovascular disease; DV, dependent variable; IV, independent variables.