Article Text

Download PDFPDF

O04.2 Hormonal Contraception is Associated with a Reduced Risk of Bacterial Vaginosis: A Systematic Review and Meta-Analysis
  1. L A Vodstrcil1,2,
  2. J S Hocking1,
  3. M Law3,
  4. S Walker1,
  5. C K Fairley1,4,
  6. S Tabrizi2,5,6,
  7. C S Bradshaw1,4,7
  1. 1Melbourne School of Population and Global Health, University of Melbourne, Carlton, Australia
  2. 2Murdoch Children’s Research Institute, Parkville, Australia
  3. 3Kirby Institute, University of New South UK, Sydney, Australia
  4. 4Melbourne Sexual Health Centre, Carlton, Australia
  5. 5Department of Microbiology and Infectious Diseases, The Royal Women’s Hospital, Parkville, Australia
  6. 6Department of Obstetrics and Gynaecology, University of Melbourne, Parkville, Australia
  7. 7Department of Epidemiology and Preventative Medicine, Monash University, Melbourne, Australia


Background We conducted a systematic review and meta-analysis of published data to examine the association between hormonal contraception (HC) and bacterial vaginosis (BV).

Methods Three databases (Medline, Web of Science and Embase) were searched until the end of January 2013 and duplicate references removed. Inclusion criteria were (1) > 20 BV cases; (2) data available to derive the association between HC and BV; (3) > 10% of participants used HC; (4) accepted BV diagnostic method. Data extracted included: type of HC, BV diagnostic method, and BV outcome (prevalent, incident, recurrent). Meta-analyses were conducted to calculate overall and pooled odds/risk ratios (OR/RR), stratified by HC-type and BV outcome. This systematic review is registered with PROSPERO (CRD42013003699).

Results Of 1710 unique references identified, 328 were assessed for eligibility and 48 studies met inclusion criteria. Overall, 36 reported BV-prevalence, 12 BV-incidence and 3 BV-recurrence. Twenty three studies reported data for combined-HC, 9 for progesterone-only HC and 23 did not specify HC-type. Diagnostic methods included: Nugent’s (n = 31), Amsel’s (n = 15), Spiegel’s (n = 1) and Ison-Hay (n = 1). Pooled BV prevalence was 30.7% (95% CI: 26.7–34.7%) and ranged from 4.7%–66.7%, with > 99% of the observed variance of prevalent BV explained by heterogeneity (I2 = 99.6%). The pooled effect of any HC-use on the composite-BV outcome (prevalent/incident/recurrent) was OR = 0.67 (95% CI: 0.62–0.71). When stratified by BV outcome, any HC-use was associated with decreased risk of prevalent (OR = 0.63; 0.58–0.68), incident (RR = 0.78; 0.68–0.87) and recurrent (RR = 0.61; 0.49–0.73) BV. When prevalent BV was stratified by HC-type, combined-HC (OR = 0.68; 0.62–0.74), unspecified HC-type (OR = 0.60; 0.53–0.67), and progesterone-only methods (OR = 0.67; 0.46–0.88) were all associated with decreased risk of prevalent BV.

Conclusion HC-use is associated with a significant and consistent decreased risk of all BV outcomes, with a greater reduction in risk for prevalent than incident BV. Surprisingly, both combined and progesterone-only contraceptive methods were associated with reduced risk of prevalent BV.

  • bacterial vaginosis
  • hormonal contraception
  • meta-analysis

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.