No effect of hormonal contraception on HIV disease progression in large multi-country cohort

Polly Clayden, HIV i-Base

Some studies have suggested that hormonal contraceptive use is associated with accelerated HIV disease progression in untreated women.

In an oral presentation, Elizabeth Stringer showed data from the MTCT Plus Initiative, a family based HIV care and treatment programme with pregnancy as an entry point to care and includes 14 sites in Africa and Thailand.

Between 2003-2008, MTCT Plus enrolled and followed 7846 women. Women received 6 monthly CD4 measurements and contraceptive use was self reported and varied by site.

This analysis included women with available contraception and CD4 data, who were not yet eligible for treatment, and not pregnant (or within 90 days of delivery).

Contraception exposure was defined in three categories: progesterone only, which included implants and injectables; combined oestrogen-progesterone; and no exposure to hormonal contraception, which included no contraception and all non-hormonal forms of contraception.

The primary endpoint in this analysis was disease progression defined as eligibility for ART (according to WHO criteria) or death.

The investigators used Kaplan Meier method and Cox proportional hazard regression to estimate time to disease progression. Contraception exposure was categorised according to the method reported on entry to the cohort. The investigators also performed a separate time-varying analysis in which women who switched methods contributed person-time to each exposure category.

Dr Stringer reported that 4530/7846 women in the cohort were eligible for this analysis. Of these, 830 women were exposed to progesterone only contraception; 226 to combined oestrogen-progesterone and 3099 had no hormonal exposure. The remaining 375 women received contraception but the method was unknown.

Baseline characteristics were similar across the exposure groups and all groups had a median CD4 count of >400 cells/mm3. There was considerable variability of initial contraception method prescribed across sites.

Dr Stringer reported that during the period of analysis, 902 women overall reached a primary endpoint (see Table 1).

Table 1: Overall rate of death or ART eligibility

Events Rate* 95% CI
Death 66 1.1 0.9-1.4
ART eligible 881 17.0 15.9-18.2
Death or eligible 902 17.4 16.3-18.6

*Rate per 100 person years

When the investigators looked at time to primary endpoint according to exposure, they found no difference between the three exposure categories, p=0.42.

Furthermore, multivariate analysis controlling for age, parity, baseline WHO stage, CD4 count, BMI, Hb, condom use and site, with no exposure as the reference, revealed no difference in disease progression between the three exposure categories (see Table 2).

Table 2. Impact of contraception category on progression

Crude HR (95% CI) Adjusted HR (95% CI) Time varying AHR (95% CI)
No exposure 1.0 1.0 1.0
Progesterone only 1.0 (0.8-1.2) 1.0 (0.8-1.2) 0.9 (0.8-1.2)
Combined progesterone-oestrogen 1.0 (0.7-1.3) 0.9 (0.6-1.2) 0.8 (0.6-1.1)

Finally, when they looked at the hazard of disease progression by site, neither progesterone only nor combined progesterone-oestrogen contraception use appeared to have an impact.

Dr Stringer concluded that there was no evidence of hormonal contraception accelerating HIV disease progression in this dataset. However she noted that differences between progesterone-based methods of contraception could not be elucidated. She suggested that further research in this field is needed.


These findings are reassuring, but as the study investigators suggest, more data is needed.


Stringer et al. Effect of hormonal contraception on HIV disease progression: a multi-country cohort analysis. 16th CROI, February 2009, Montreal, Candada. Oral abstract 175.

Links to other websites are current at date of posting but not maintained.