Table 3.

Sensitivity and subgroup analyses for the association between any bisphosphonate use and risk of breast cancer.

Analysis	Categories		No. of studies	RR (95% CI)	P value	P values between subgroup
Sensitivity analysis	Adjusted confounders	≥2	10	0.86 (0.80-0.94)	0.0003∗
Subgroup analysis	Study location	Western country	9	0.86 (0.79-0.94)	0.0004∗	0.49
		Eastern country	2	0.96 (0.71-1.27)	0.75
	Study design	Cohort	7	0.89 (0.81-0.98)	0.02∗	0.26
		Case-control	4	0.78 (0.64-0.96)	0.02∗
	Menopausal status	Pre- & post-	6	0.83 (0.75-0.93)	0.0008∗	0.27
		Post-	5	0.92 (0.81-1.04)	0.18
	Measure of BP use	Medical record	8	0.87 (0.80-0.96)	0.004∗	0.75
		Self-report	3	0.85 (0.70-1.01)	0.07
	Study quality	High quality	5	0.85 (0.63-1.15)	0.3	0.87
		Low quality	6	0.88 (0.82-0.93)	<0.0001∗
	No. of cases	≤1000	5	0.87 (0.66-1.14)	0.31	0.93
		>1000	6	0.88 (0.82-0.94)	0.0001∗
	No. of participants	≤20000	5	0.83 (0.63-1.08)	0.17	0.66
		>20000	6	0.87 (0.80-0.94)	0.0004∗

Abbreviations: RR: relative risk; CI: confidence interval; BPs: bisphosphonates. Note: ^∗P < 0.05.