Table 3:

Summary of the articles included in a systematic review of the associations between invasive ovarian cancer risk and use of injectable depot-medroxyprogesterone acetate (DMPA).

No.	First Author Year of publication	Study design	Time period and country of recruitment	Duration of follow-up	Sample size	Participants’ characteristics	Outcome	Exposure	Adjustment factors	Results
1	Liang et al. 1983 (9)	Retrospective cohort study of women receiving DMPA for contraception	1967–1976 in the U.S.	Until Dec 31, 1980 (4–13 years)	5,003	Black women given DMPA during 1967–1976 under FDA Investigational New Drug approval.	Not clear if invasive or borderline or both	DMPA	Age, year, race	SIR=0.86 (0.0–4.6) Observed 1 case of ovarian cancer. Compared to an expected 1.16 cases based on national rates for black women (adjusted for lost to follow-up),
2	WHO 1991 (10)	Hospital-based case-control	1979–1986 in Mexico 1979–1988 in Thailand		224 cases (9.8% used DMPA) 1,781 controls (12.9% used DMPA)	Cases and controls were matched on age, hospital and year of interview.	Invasive and borderline	DMPA	Age, hospital, year of interview, number of live births and use of COCs.	OR=0.81 (95% CI 0.36–1.8) for invasive.
3	Wilailak et al. 2012 (11)	Hospital-based case-control	2006–2008 Thailand		330 cases (17.9% used DMPA) 982 controls (25.7% used DMPA)	Cases and controls were matched on age and hospital.	Invasive	DMPA	COC use, breastfeeding, parity and family history of gynecological cancer.	OR=0.52 (95% CI 0.33–0.88)
4	Urban et al. 2012 (23)	Hospital-based case-control	1995–2006 South Africa		182 cases (5.5% ever used injectable and never used oral contraceptives) 1,492 controls (16.7% ever used injectable and never used oral contraceptives).	Black females 18–79 years old	Invasive	Injectable contraceptives	Age at diagnosis, year of diagnosis, education, tobacco smoking, alcohol consumption, parity/age at first birth, number of sexual partners, urban/rural residence, and province of birth.	OR for Ever use of injectable and Never use of oral contraceptives compared to Never use of both: 0.35 (95% CI 0.17–0.71).
5	Huang et al. 2015 (24)	Population-based cohort	1997–2000 China	12.6 years follow-up on average.	174 cases (3.4% used contraceptive shots) 70,259 participants (2.6% used contraceptive shots)	40–70 years old.	Invasive	Contraceptive shots	Age at recruitment, education, years of ovulation, irregular ovulatory cycles, first-degree family history of cancer, regular physical activity within 5 years, other contraceptive methods (never/ever)	HR= 1.33 (0.58–3.04)
6	Iversen et al. 2018 (12)	Prospective, nationwide cohort	1995–2015 Denmark	Up to 20 years.	1,249 cases/ (3 of them had used DMPA and had not switched their contraceptives) 1,879,227 participants	Aged 15–49 years.	Invasive	DMPA	Calendar year, parity, age, education, tubal sterilization, hysterectomy, endometriosis, polycystic ovary syndrome, and family history of breast or ovarian cancer	Among women who had not switched their hormonal contraceptives, hazard ratio for DMPA use compared to never use of any hormonal contraceptives HR= 6.56 (95% CI 2.11–20.40)