Cost analysis for ART using a model of the effect of age and FSH levels upon pregnancy rates: the concept of a minimally acceptable pregnancy rate

Objective: It is universally accepted that increasing age and diminished ovarian reserve reduce the likelihood of pregnancy with ART. However, after deciding to pursue ART, patients often wish to proceed with IVF even when factors such as an elevated FSH and age reduce the likelihood of success to exceedingly low levels. Patient judgment in this setting is often conflicted. Ethical and philosophical differences between ART clinics prevent consensus as to which patients are permitted to proceed with ART. The objective of this study is to examine how a ’minimally acceptable anticipated pregnancy rate’ influences economic consequences of an ART stimulation in poor prognosis cases. Design: Hypothetical cost analysis and modeling. Materials and Methods: Modeling of ART success rates was based on basal FSH and age using data presented by Akande et al. (2002) and Levi et al. (2002). For analysis, cost of a fresh ART cycle was set at $10,000. Cost analysis was restricted to cost to achieve pregnancy, and did not include variables such as miscarriage (increases with age) multiple births and premature delivery, which would increase the total cost per pregnancy. For simplicity, cost of cancelled cycles was assumed to be the same as retrieval cycles. Crude cost per delivery to the economic system was calculated as equal to the total cost of patients accepted/expected delivery rate. Expense of a donor cycle was assumed to be $25,000 and with an expected pregnancy rate of 60%, would result in an estimated cost of $41,666/pregnancy. Results: Graph of pregnancy rate (y axis) versus age (x axis) with FSH values (logit transformation) is shown in Figure 1. Pregnancy rate varied according to FSH and age, consistent with the literature. Using this model, it is possible to arbitrarily set different ’minimally acceptable pregnancy rates’ and calculate the economic consequences. If the minimal acceptable rate is set very high, the cost per delivery is low. For instance if the acceptable minimum pregnancy rate is set at 25% the cost per delivery can be estimated at $40,000. However, this would exclude a significant number of patients and if such patients were referred for donor oocyte cycles would then result in excessive expenses related to donor cycles. If the minimal rate is lowered to 5%, the cost per pregnancy is $200,000. If an ART program adopts the policy to accept patients with a 1% likelihood of pregnancy, the cost per pregnancy to society is $1,000,000. It is noted that a decision to ’allow patients to try’ in cases where pregnancy is unlikely the total cost to the patient remains the same, but the overall cost to the health care system is considerably higher. Since miscarriage increases as age increases, the cost for patients with elevated FSH and advanced age, the true cost per baby may be considerably higher than $1,000,000. Conclusions: As utilization of ART services increases with legislated third party coverage in many states, the concept of a ’minimally acceptable expected pregnancy rate’ becomes relevant to national policy regarding assisted reproduction. If an ART program offers ART to patients with very low likelihood of pregnancy, the economic consequences of this decision may place heavy financial burdens upon society. This model improves upon the overly simplistic concept of a single FSH cutoff value. Supported by: Intramural NICHD, PREB