abstract
- OBJECTIVE: The clinical indications for in vitro fertilization (IVF) have expanded to include many forms of infertility in addition to tubal disease. Pregnancies in IVF cycles are noteworthy but there is frequently a spontaneous cure for infertility among similar couples. The relative merit of IVF treatment over spontaneous cure or other forms of fertility treatment has not been rigorously evaluated. DESIGN: The study was a randomized controlled clinical trial comparing the clinical pregnancy rate among couples undergoing IVF with the rate among couples awaiting an IVF treatment. PATIENTS: Patients entering a provincially funded program of IVF were randomly allocated to a period of delay prior to IVF treatment (Control n = 194) or 1 or more cycles of IVF treatment (Experimental n = 205). MAIN OUTCOME MEASURES: Clinical pregnancy rate and adjusted time to pregnancy. RESULTS: In the Control group there were 13 pregnancies. In the Experimental group there were 13 pregnancies before treatment could be arranged. There were 20 additional pregnancies in treatment cycles. The intention-to-treat analysis showed an increase in the proportion of pregnancies from 8% to 17.4% and parturition from 4.9% to 11.6%. Substantially more patient-time (due to IVF) was required to achieve this increase. There was no difference between groups when time-to-event was considered by survival analysis, although a long-term trend in favor of the Experimental group was suggested. Low-event frequency and broad confidence intervals in Control patients prior to censor and transfer to IVF treatment prevented a conclusive assessment of the long-term benefits of IVF treatment. Generalizing these findings, improved effectiveness may be evident with delayed access to treatment (longer waiting lists), suitable candidates with appropriate primary clinical diagnoses and durations of infertility and higher rates of treatment over time (larger clinics). CONCLUSIONS: IVF treatment is effective in increasing, proportionally, the numbers of pregnancies, live births, and parturitions, but this occurred with significantly longer patient commitment.