Date Published: June 11, 2019
Publisher: Public Library of Science
Author(s): Usha Kumar, Louise Pollard, Lucy Campbell, Selin Yurdakul, Clara Cantalapiedra Calvete, Bola Coker, Tao Chen, Abdel Douiri, Elizabeth Ann Micks.
To assess the impact of specialist contraceptive support after abortion on effective contraceptive use at six months and subsequent abortions within two years.
Multicentre randomised controlled trial among women undergoing induced abortion in three London boroughs. Allocation was through electronically concealed stratified randomisation by centre, blinding clinicians and participants to arm allocation until interventions. Control group received standard care, comprising advice to follow up with their general practitioner or contraceptive clinic as needed. Intervention group additionally received specialist contraceptive support via telephone or face-to-face consultation at 2–4 weeks and 3 months post-abortion. Primary outcome was rate of effective contraceptive use at six months post-abortion. Secondary outcomes were subsequent abortions within two years.
569 women were recruited between October 2011 and February 2013, randomised to intervention (282) and control (287) groups; 290 (142 intervention, 148 control) were available for primary outcome analysis. Intention-to-treat analysis showed no significant difference between the two groups in effective contraceptive use after abortion (62%, vs 54%, p = 0·172), long-acting contraceptive use (42% versus 32%, p = 0·084), and subsequent abortion (similar rates, at 1 year: 10%, p = 0·895, between 1–2 years: 6%, p = 0·944). Per-protocol analysis showed those who received the complete intervention package were significantly more likely to use effective contraception (67% versus 54%, p = 0·048), in particular long-acting contraception (49% versus 32%, p = 0·010) and showed a non-significant reduction in subsequent abortions within 2 years (at 1 year: 5% versus 10%, p = 0·098; and between 1–2 years: 3% versus 6%, p = 0·164, respectively).
Structured specialist support post-abortion did not result in significant use of effective contraception at six months or reduction in subsequent abortions within two years. Participants engaging with the intervention showed positive effect on effective contraception at six months post-abortion. The potential benefit of such intervention may become evident through further studies with increased patient participation.
Abortion rates in England and Wales have declined in the last decade, however subsequent abortions are on the rise; in 2016, 38% of women undergoing abortions had one or more previous abortions as compared to 32% in 2006 . In London alone 42% of the abortions in 2016 were subsequent abortions , despite the widespread availability of free contraception. National population-level data on abortions within two years after an initial abortion are not available, however, hospital-level data from studies in Scotland have reported rates ranging from 12·3%  to 14·6% . Pregnancies ending in subsequent abortions within a short timeframe are likely to be unintended, reflecting unmet contraceptive needs and highlighting the gap in provision of effective post-abortion care.
1006 women were assessed for eligibility; 437 were excluded; amongst these, 352 declined and the remaining 85 were not eligible. 569 patients were randomised: 282 to the intervention (I) arm and 287 to the control (C) arm. 18 patients withdrew from the study (ten decided not to have an abortion and eight did not wish to continue with the study). A total of 290 participants (I = 142, C = 148) completed the six-month follow-up (Fig 1).
This study showed that a structured specialist support package following abortion did not increase the use of effective contraception at six months post-abortion. There was no significant difference in the incidence of subsequent abortions within two years between the intervention and control groups. However, participants who engaged in the interventions showed a positive effect on effective contraceptive use, in particular long-acting contraceptive methods. A reduction was observed in subsequent abortions within one and two years in the per-protocol population, but did not reach statistical significance.