A new proposal for measuring ‘success’ in IVF aims to meet a patient’s predetermined chance of success and expectations at the lowest possible risk, costs, time and treatment burden; clinical outcome is best measured as cumulative live birth rate per intention to treat.
The measurement of ‘success’ in ART has long been a contentious subject. Pregnancy per transfer was superseded by live birth rate and later cumulative LBR per started cycle, with some registries today apparently preferring LBR per transfer. Yet whatever the measure, heterogeneity in the methods of calculation continues, such that an international consensus last year called for a ‘core outcome’ set of measurements for reporting study outcomes.(1) Publication of this initiative, known as COMET (Core Outcome Measures in Effectiveness Trials), had been preceded by a session organised by the Cochrane group at ESHRE’s 2019 annual meeting in Vienna at which statistician Jack Wilkinson counted more than 800 combinations of numerator and denominator reported as outcomes in 142 IVF RCTs published in peer-reviewed journals in just one year. Important outcomes such as clinical pregnancy were defined in more than 60 different ways or were not defined at all, thus causing ‘methodological frailty’ in the evidence base.
The question of homogeneity in the measurement of outcome has now been raised again by specialists from the GeneraLife group in Italy and elsewhere.(2) This time, the focus is not so much consistency in reporting clinical trials but more a broad definition of success for patients - which includes ‘clinical, financial and patient-centred perspectives’. Clinically, their idea of success is ‘cumulative LBR per intention to treat’, and, as in the COMET reports, they once again ask if it’s time for consensus.
Cumulative LBR per ITT, they write, represents the most valuable measure because it comprises the treatment in its entirety, from stimulation to embryo ‘manipulations’ and the sum of fresh and cryopreserved embryo transfers. Moreover, they add that, if the ideal number of oocytes is collected from each patient, this outcome cannot be improved; ‘it is mostly dependent on the competence of the resulting embryos and the receptivity of the endometrium’. They therefore highlight the significance of not impairing this embryo competence with ‘poor practice’. Thus, at any time a new policy, protocol or procedure is introduced, ‘it should be ensured that the IVF efficacy is at the very least preserved’.
The next perspective in the proposed outcome model is financial – which need not be limited merely to cost-effectiveness. Also needing consideration are accessibility and waiting time, investigation of each partner, counselling, treatment strategies, adoption of single embryo transfer, and long-term monitoring.
Among the ‘patient-centred’ factors for inclusion as a measure of cost-effectiveness are time, treatment risks and psychosocial factors. Indeed, the authors suggest, ‘any negative experience should be clearly acknowledged when defining success in IVF’. The minimisation of IVF-related risks, such as OHSS, miscarriage and multiple pregnancies, would also define the quality of treatment. In addition, the satisfaction of the patient – in terms of needs, preferences and values – should guide clinical decisions, which would be supported by accurate and reliable patient information, counselling and informed consent. Focus on Reproduction has recently described calls in the UK for transparency in patient information, notably in the public presentation of costs and success rates.(3)
Clearly, as most outcome analyses reflect, results should be presented against a background of confounders, such as patient age, duration and cause of infertility, BMI, ovarian reserve and reproductive history. Any definition of success, say the authors, is ‘misleading’ if not presented along with confounders. This may be resolved by ‘presenting the data among populations of patients clustered according to the most significant features affecting IVF outcomes’ (for example, range of maternal age).(4)
The consistent presentation of success rates has also been raised recently by publication of ESHRE’s consensus on clinical performance indicators in ART.(5) This report too advises clinics to consider four treatment decisions with regards to ART – burden, effectiveness, safety and costs – to complement their outcome measurements.
Rienzi et al acknowledge that their proposed model for calculating success is not simple. Indeed, many of the features they list to define ‘success’ seem beyond quantification or measurement. But their aim is to provide patients with ‘safe and efficient’ treatments which here converge into a ‘multidisciplinary network’ and which meet the patient’s predetermined chance of success and expectations and at the lowest possible risk, costs, time and burdens.
Just as the editors of the main reproduction journals gave their commitment to the core outcome measurements in publishing clinical studies, so the authors here call on the main societies ‘to join forces . . . to reach a consensus on the definition of success in IVF’. It’s difficult, they concede – a ‘complex task’ according to their paper’s title – but the objective is a reliable measure which recognises clinical effectiveness, acceptability and safety.
1. Duffy JMN, AlAhwany H, Bhattacharya S, et al. Developing a core outcome set for future infertility research: an international consensus development study. Hum Reprod 2020; 35: 2725–2734. doi.org/10.1093/humrep/deaa241
2. Rienzi L, Cimadomo D, Vaiarelli A, et al. Measuring success in IVF is a complex multidisciplinary task: time for a consensus? Reprod Biomed Online 2021; doi.org/10.1016/j.rbmo.2021.08.012 1472
3. See https://www.focusonreproduction.eu/article/News-in-Reproduction-Clinic-advertising
4. Correia KF, Dodge LE, Farland LV, et al. Confounding and effect measure modification in reproductive medicine research. Hum Reprod 2020; 35: 1013–1018.
5. ESHRE Clinic PI Working Group, Vlaisavljevic V, Apter S, Capalbo A, et al. The Maribor consensus: report of an expert meeting on the development of performance indicators for clinical practice in ART. HR Open 2021; doi:10.1093/hropen/hoab022