Impact of the "physician factor" on pregnancy rates in a large assisted reproductive technology program: do too many cooks spoil the broth?

Objective: To determine whether in one program with unified treatment proto cols, patients can expect varying treatment outcomes with different physici ans. Design: Retrospective data analysis. Setting: University-affiliated infertility center with 14 physicians. Patient(s): One thousand eight hundred fifty IVF cycles performed consecuti vely between August 1995 and June 1997. Intervention(s): The pregnancy rate and implantation rate per ET were evalu ated for individual physicians between August 1995 and June 1996 (phase I). Physicians with lower success rates underwent strict supervision from July 1996 to June 1997 (phase II). Main Outcome Measure(s): Variations in success rates between physicians. Result(s): The pregnancy rate varied among the physicians from 13.2%-37.4%, and the implantation rate varied from 4.3%-14%. Some physicians' outcomes improved between phase I and phase II of the study, whereas others' did not . The pregnancy and implantation rates varied significantly for some physic ians, depending on whether they were responsible for the choice of stimulat ion protocol, supervision of cycle monitoring, or ET in their own or other physicians' patients. Conclusion(s): Outcomes of IVF vary depending on the treating physician. Lo wer than expected pregnancy and implantation rates usually are not caused b y poor ET techniques alone, but appear to be disproportionately the consequ ences of poor cycle stimulation. (Fertil Steril(R) 1999;71:1001-9. (C) 1999 by American Society for Reproductive Medicine.).