Treatment decisions, not patient demographics, predict a positive pregnancy test after autologous IVF: a retrospective cohort study from northern India

BackgroundInfertility affects about one in six people worldwide over a lifetime, and in vitro fertilisation (IVF) is the mainstay of treatment for many of them. In India most IVF is delivered in private clinics and paid for out of pocket. The question is which factors determine whether an IVF cycle produces a pregnancy: the characteristics a patient arrives with, or the treatment decisions made once she is there.MethodsWe conducted a retrospective cohort study of 288 women who underwent IVF at a private fertility centre in Faridabad, Haryana, between April 2023 and October 2024. One record was analysed per woman, taken from her most recent cycle. The outcome was a positive serum beta-human chorionic gonadotropin (beta-hCG) test, the last outcome routinely documented at the centre. Donor-oocyte (recipient) cycles were analysed separately from autologous cycles, because success in a donor cycle depends on the donor rather than on the patient. Categorical associations were tested using Fisher’s exact test. Independent predictors were identified using multivariable binomial logistic regression.ResultsSeventy-seven cycles (26.7%) used donor oocytes and 211 (73.3%) were autologous. The positive beta-hCG rate was 70.6% in autologous cycles and 54.5% in donor cycles (p = 0.016). Within autologous cycles, two treatment decisions were associated with the outcome: intracytoplasmic sperm injection (ICSI) versus conventional IVF (76.0% versus 46.2%, p < 0.001) and frozen versus fresh embryo transfer (74.9% versus 51.3%, p = 0.006). No demographic characteristic was associated with the outcome. After adjustment (N = 210), conventional IVF carried lower odds than ICSI (adjusted odds ratio 0.16, 95% CI 0.06–0.42, p < 0.001) and frozen transfer higher odds than fresh (adjusted odds ratio 2.80, 95% CI 1.30–6.04, p = 0.009). Oocyte yield was marginally associated (adjusted odds ratio 1.11 per oocyte, 95% CI 1.00–1.22, p = 0.041). In absolute terms, ICSI corresponded to 32.5 percentage points and frozen transfer to 18.5 percentage points. Age was not associated with the outcome, although autologous cycles spanned only 22–42 years and the study was underpowered to detect age effects.ConclusionIn autologous cycles at this centre, treatment decisions rather than patient demographics predicted a positive pregnancy test. The cohort was socioeconomically narrow and drawn from a single private centre. The absence of demographic effects should therefore not be read as evidence that social position does not matter. Social position might operate earlier, in determining who reaches a fertility centre at all.