Human Reproduction, Vol.32, No.8 pp. 1556–1559, 2017 Advanced Access publication on June 13, 2017 doi:10.1093/humrep/dex214 OPINION A prognosis-based approach to infertility: understanding the role of time ESHRE Capri Workshop Group*,† *Correspondence address: P.G. Crosignani, IRCCS Ca’ Granda Foundation Maggiore Policlinico Hospital, Via M. Fanti 6, 20122 Milano, Italy. E-mail: firstname.lastname@example.org Submitted on January 30, 2017; resubmitted on May 9, 2017; accepted on May 19, 2017 ABSTRACT: The current deﬁnition of infertility acknowledges the importance of duration of pregnancy seeking but fails to recognize the prevalent negative impact of female age. In fact, the diagnosis of unexplained infertility increases with women’s age because of our incapacity to discern between age-related infertility and real unexplained infertility. Physicians’ response to the pressures of increased female age has been to take prompt refuge in assisted reproduction despite the lack of robust evidence and the inherent risks and costs of these procedures. Moreover, the prioritization of immediate health gains over those in the future, preference for accessing active treatment rapidly and reluctance to wait for spontaneous pregnancy expose patients to additional risks of overtreatment. Solutions are not simple to ﬁnd but an alternative and innovative vision of infertility based on prognosis may be a valid solution. The availability of validated dynamic models based on real-life data that could predict both natural and ART-mediated conceptions may be of beneﬁt. They could facilitate patients’ counselling and could optimize the chances of success without exposing patients to unnecessary, expensive and demanding treatments. Key words: aging / subfertility / infertility / fecundity / age-related infertility / unexplained infertility Introduction The dictionary deﬁnition of infertility implies a state of sterility. However, it is the absence of conception over a speciﬁc time horizon of 12 months despite regular sexual intercourses which underpins the International Committee for Monitoring Assisted Reproductive Technology and the World Health Organization deﬁnitions of infertility (Zegers-Hochschild et al., 2009). The main purpose of deﬁning infertility in this manner is to identify a population of couples who have a poorer prognosis (rather than absolute infertility) and merit further investigation and treatment. Unlike most other medical conditions, the term ‘infertility’ (sometimes called ‘subfertility’) represents a prognosis, rather than a diagnosis (Dunson et al., 2004). The current deﬁnition of infertility acknowledges the inﬂuence of duration of pregnancy seeking as a prognostic factor but fails to accommodate the prevalent negative impact of increased female age on fecundity (ESHRE Capri Workshop Group, 1996). To date, age has not been part of the diagnostic criteria for infertility (Gurunath et al., 2011). Given the profound impact of female age on prognosis (Hunault et al., 2004), it would seem logical that the diagnosis of infertility should incorporate not just duration of pregnancy seeking but also female age. Noteworthy, age-related fertility decline is a universal pattern and age at last birth is extremely similar across historical ages and cultures (Eijkemans et al., 2014). Unexplained infertility and age-related infertility In at least one in four infertile couples (Evers, 2002; Practice Committee of the ASRM, 2006) a cause cannot be identiﬁed. Consequently, couples in this heterogeneous group are all grouped under a single heterogeneous category of ‘unexplained’ infertility (NICE, 2013; Practice Committee of the ASRM, 2015). While the general tendency in older women is to expedite access to assisted reproduction, it may be worth disentangling unexplained infertility from age-related infertility, as illustrated in a recently published mathematical model of human conception according to a woman’s age (Broer et al., 2011; Somigliana et al., 2016). This model shows that the rate of false positive diagnosis of unexplained infertility (i.e. infertility due to unknown causes but excluding age) after 2 years of regular intercourse rises from 10% for women trying for a pregnancy before 35 years to 80% for those initiating such attempts at age 40. In † The list of The ESHRE Capri Workshop Group contributors is given in the Appendix. © The Author 2017. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please e-mail: email@example.com Time and infertility fact, the reliability of a diagnosis of truly unexplained infertility (as opposed to age-related infertility) falls dramatically with age. Accordingly, epidemiological evidence showed that the incidence of unexplained infertility increases with women’s age (Maheshwari et al., 2008). Theoretically, age-related infertility and unexplained infertility may beneﬁt from different therapeutic approaches. As ART does not improve oocyte quality, one may expect ART to be of little efﬁcacy in women with age-related infertility (Van Voorhis, 2007). In contrast, some forms of unexplained infertility could be overcome by IVF, such as occult ovulation defects, subtle disorders of tubal function that cannot be identiﬁed with the traditional tests of tubal patency, cervical abnormalities and male factor problems that cannot be identiﬁed by standard semen analysis. However, ﬁrmly distinguishing between unexplained infertility and age-related infertility in older women is currently impossible. Noteworthy, hormonal measurements of the ovarian reserve is of scant utility in this context (Somigliana et al., 2016). In fact, despite the weakness of the evidence and the lack of a robust rational, there is a general agreement on the need to accelerate the diagnostic process in older women. Tests necessary to exclude any barriers to natural conception and the possibility to opt for expectant approach are generally deemed of scant interest because of the imperative to access treatment early in these women, given their rapidly narrowing window of opportunity. NICE (2013) suggests that investigation and treatment should be considered after 6 months of trying for a baby in women over 36 years of age in whom prognosis is relatively poor. However, there is a real danger that this approach could lead to overtreatment. The role of values and preferences in informing treatment choices Freedom to choose between healthcare options and the ability to take an active role in decision-making are emerging as important outcomes in healthcare (Entwistle et al., 2012), suggesting that it is important to look beyond the most obvious clinical endpoints of fertility treatment (pregnancy) to ﬁnd relevant indicators of success. Immediate health beneﬁts are valued higher by patients than those realized at a future date. This has led to the practice of discounting future health gains in clinical and monetary terms (Tinghög, 2012). Couples seeking help for infertility may actually have overriding preferences for a pregnancy which occurs sooner rather than later. Approximately a third of Dutch couples with unexplained subfertility, many of whom had a realistic chance of a spontaneous pregnancy, found expectant management unacceptable (Kersten et al., 2015), possibly because they associated early treatment with an early pregnancy and/or valued treatment as a desirable endpoint in its own right. A pregnancy leading to live birth at any point (over a given time horizon) may have been the accepted currency of treatment success in the past, but time to pregnancy is becoming the preferred outcome for the couple. Are we over-medicalizing reproduction? This emphasis on minimizing time to a desired pregnancy and the overwhelming belief in the power of active treatment makes IVF very 1557 attractive for all couples including those with no obvious barrier to conception. This is in marked contrast to the origins of IVF and ICSI, which were developed to treat patients with ‘severe’ tubal factor infertility and ‘severe’ male factor infertility (Evers, 2016). The efﬁcacy of IVF for these indications is unquestionable. Thereafter, however, we have found ourselves trapped in a ‘slippery slope’ situation where there is a temptation to expand the use of ICSI for mild male factor infertility and IVF for one-sided tubal pathology, unexplained infertility, endometriosis, poor ovarian reserve and, most importantly, advanced maternal age. In the UK, the number of IVF cycles increased from 35 450 to 60 473 between 2000 and 2011, and >50% of the increase was accounted for by unexplained subfertility (including also agerelated infertility) (HFEA, 2012). Our response to the pressures of increased female age and a diminishing reserve of time has been to seek refuge in assisted reproduction. We have created a therapeutic illusion on a grand scale (Casarett, 2016; Evers, 2016). A prognosis-based approach using clinical prediction models to ensure timely treatment We have established that in most couples infertility is not absolute but time has a negative impact on prognosis (through both duration and female age) which we are unable to reverse. We have also emphasized the inappropriateness of a systematic leaning towards immediate use of assisted reproduction and the potential clinical impact of hooking up unexplained infertility and age-related infertility into a unique entity. A possible way to tailor management strategies for couples that can optimize their chances of success without leading to unnecessary, expensive and demanding treatments could be to use prognostic models based on clinical and laboratory parameters. These models should estimate chances of treatment-independent as well as treatment-related chances of pregnancy so that couples and clinicians can make sensible decisions about when to initiate treatments. They should be pragmatic and thus developed based on real-life clinical data. This approach could consent to overcome some yet obscure aspects of our discipline, such as in particular the above-mentioned elusive difference between unexplained infertility and age-related infertility. The main challenges in making prediction models that can estimate the beneﬁt of treatment of no treatment are (i) uniform data collection of both no treatment period and treatment periods and (ii) random allocation of treatment/correcting for non random allocation in observational data. Existing fertility prediction models are able to compute pregnancy outcomes within a ﬁxed time horizon. For example, the validated Hunault model that has entered clinical practice (Hunault et al., 2004; Leushuis et al., 2009; van Loendersloot et al., 2010) estimates the chance of spontaneous pregnancy leading to live birth over a period of 1 year in couples attending a fertility clinic for the ﬁrst time. It can be used to decide whether a couple should receive fertility treatment straightaway or undergo a period of expectant management. However, it is a static model that does not adapt to time. It does not provide any individualized decision support regarding when to initiate active treatment in a couple who have been advised expectant management. Since a couple’s prognosis decreases over time, it would be beneﬁcial to have an individualized and dynamic prediction model that could estimate this changing prognosis at different points in time. Van Eekelen et al. (2017) 1558 recently developed such a dynamic model with promising results but external validation is still required. On the other hand, in order to make decisions regarding timing and type of treatment, we should also be aware that the impact of treatment may not be identical in all women with similar chances of spontaneous pregnancy (McLernon et al., 2014). There are current intense efforts in developping universal models for the prediction of ART success, with most recent models being more and more reﬁned. However, none has yet been incorporated within routine clinical practice (Leushuis et al., 2009; van Loendersloot et al., 2010; McLernon et al., 2016). Ideally, we need a model which can provide estimates of the change in the absolute beneﬁt of treatment versus no treatment over time such as a dynamic prediction model with landmarking. This involves ﬁtting a series of separate regression models from each consecutive month (landmark) since registration at the fertility clinic, in order to predict the chances of pregnancy leading to live birth over, say, a follow-up window of 1 year (van Houwelingen and Putter, 2012). In a couple with a good prognosis who has been initially advised expectant management, this approach could be used to decide when, in the future, the absolute beneﬁt of treatment is likely to trump their chance of spontaneous pregnancy such that treatment should begin (McLernon et al., 2014). It could be argued that a prognosis-based approach may not be consistent with a patient-centered approach as it ignores patients’ potential preferences for active versus expectant treatment. However, patient-centred care is not about patients choosing treatments based on their intuitive beliefs. Instead, it implies focus on the patient as a person (rather than the disease) and shared control of treatment decisions, which are best made in the presence of clear and accurate prognostic information. Conclusions Time is an essential but intricate aspect of infertility care. It inﬂuences diagnosis, prognosis and success of treatments. Handling the faceted and sometimes contrasting aspects of time for infertile couples in everyday clinical practice is sometimes demanding. This situation may actually favour irrational and potentially harmful clinical attitudes. On the other hand, keeping decision-making simple is also important and patients’ values and wishes deserve utmost consideration. Physicians have currently to frequently deal with this intricate scenario. Developing and validating comprehensive dynamic models for a tailored and rational use of assisted reproduction may be of beneﬁt in this context. These models could optimize resource allocation, prevent risks and improve the quality of shared clinical decision-making. Acknowledgements The secretarial assistance of Mrs Simonetta Vassallo is gratefully acknowledged. Authors’ roles All lecturers and chairs contributed to the preparation of the ﬁnal manuscript. ESHRE Capri Workshop Group Funding The meeting was organized by the European Society of Human Reproduction and Embryology with an unrestricted educational grant from Institut Biochimique S.A. (Switzerland). Conﬂict of interest None declared. Appendix A meeting was organized by ESHRE [September 1–2, 2016] to discuss human infertility. The lecturers included: D.F. Albertini (Director, Division of Laboratories, Center For Human Reproduction, New York, USA); R. Anderson (Head of Section, Obstetrics and Gynaecology, University of Edinburgh, Edinburgh, UK); S. Bhattacharya (Professor of Reproductive Medicine, Head of Division of Applied Health Sciences and Director Institute of Applied Health Sciences, School of Medicine and Dentistry, University of Aberdeen, Aberdeen Maternity Hospital, Foresterhill, Aberdeen, UK); J.L.H. Evers (Department of Obstetrics and Gynecology, Maastricht University Medical Centre, Maastricht, The Netherlands); D.J. Mclernon (CSO Postdoctoral Fellow, Medical Statistics Team, Division of Applied Health Sciences, University of Aberdeen, Aberdeen, UK); S. Repping (Universiteit van Amsterdam, Academisch Medisch Centrum, Centrum voor Voortplantingsgeneeskunde, Amsterdam, The Netherlands); E. Somigliana (Clinica Ostetrica e Ginecologica, IRCCS Ca’ Granda Foundation, Maggiore Policlinico Hospital, Milano, Italy). The chairs included: D.T. Baird (Centre for Reproductive Biology, University of Edinburgh, UK); P.G. Crosignani (IRCCS Ca’ Granda Foundation, Maggiore Policlinico Hospital, Milano, Italy); K. Diedrich (Klinik für Frauenheilkunde und Geburtshilfe, Universitätsklinikum Schleswig-Holstein, Campus Lübeck, Germany); R. G. Farquharson (Liverpool Women’s Hospital, Department of OB/ GYN, Liverpool, UK); K. Lundin (Reproductive Medicine, Sahlgrenska University Hospital, Gothenburg, Sweden); J.S. Tapanainen (Department of Obstetrics and Gynaecology, University of Helsinki, Helsinki University Hospital, Helsinki, Finland); A. Van Steirteghem (Centre for Reproductive Medicine, Universitair Ziekenhuis Vrije Universiteit Brussel, Belgium). References Broer SL, Eijkemans MJ, Scheffer GJ, van Rooij IA, de Vet A, Themmen AP, Laven JS, de Jong FH, Te Velde ER, Fauser BC et al. Anti-mullerian hormone predicts menopause: a long-term follow-up study in normoovulatory women. J Clin Endocrinol Metab 2011;96:2532–2539. Casarett D. The science of choosing wisely—overcoming the therapeutic illusion. N Engl J Med 2016;374:1203–1205. Dunson DB, Baird DD, Colombo B. Increased infertility with age in men and women. Obstet Gynecol 2004;103:51–56. Eijkemans MJ, van Poppel F, Habbema DF, Smith KR, Leridon H, te Velde ER. Too old to have children? Lessons from natural fertility populations. Hum Reprod 2014;29:1304–1312. Entwistle V, Firnigl D, Ryan M, Francis J, Kinghorn P. Which experiences of health care delivery matter to service users and why? A critical interpretivesynthesis and conceptual map. J Health Serv Res Policy 2012;17: 70–78. Time and infertility ESHRE Capri Workshop Group. Guidelines to the prevalence, diagnosis, treatment and management of infertility, 1996. Hum Reprod 1996;11: 1775–1807. Evers JL. Female subfertility. Lancet 2002;360:151–159. Evers JL. Santa Claus in the fertility clinic. Hum Reprod 2016;31:1381–1382. Gurunath S, Pandian Z, Anderson RA, Bhattacharya S. Deﬁning infertility— a systematic review of prevalence studies. Hum Reprod Update 2011;17: 575–588. HFEA, Human Fertilisation and Embryology Authority. (2012) Historical UK fertility treatment data and ﬁgures: 1991–2011. www.hfea.gov.uk (30 May 2016, date last accessed). Hunault CC, Habbema JDF, Eijkemans MJC, Collins JA, Evers JLH, te Velde ER. Two new prediction rules for spontaneous pregnancy leading to live birth among subfertile couples, based on the synthesis of three previous models. Hum Reprod 2004;19:2019–2026. Kersten FA, Hermens RP, Braat DD, Hoek A, Mol BW, Goddijn M, Nelen WL, Improvement Study Group. Overtreatment in couples with unexplained infertility. Hum Reprod 2015;30:71–80. Leushuis E, van der Steeg JW, Steures P, Bossuyt PM, Eijkemans MJ, van der Veen F, Mol BW, Hompes PG. Prediction models in reproductive medicine: a critical appraisal. Hum Reprod Update 2009;15:537–552. Maheshwari A, Hamilton M, Bhattacharya S. Effect of female age on the diagnostic categories of infertility. Hum Reprod 2008;23:538–542. McLernon DJ, te Velde ER, Steyerberg EW, Mol BWJ, Bhattacharya S. Clinical prediction models to inform individualized decision-making in subfertile couples: a stratiﬁed medicine approach. Hum Reprod 2014;29: 1851–1858. McLernon DJ, Steyerberg EW, Te Velde ER, Lee AJ, Bhattacharya S. Predicting the chances of a live birth after one or more complete cycles of in vitro fertilisation: population based study of linked cycle data from 113 873 women. BMJ 2016;355:i5735. 1559 NICE – National Institute for Health and Clinical Excellence. National Collaborating Centre for Women’s and Children’s Health. Fertility: Assessment and Treatment for People with Fertility Problems. London (UK); 2013 Feb. 63 p. (Clinical guideline; no. 156). Practice Committee of the American Society for Reproductive Medicine. Effectiveness and treatment for unexplained infertility. Fertil Steril 2006; 86:S111–S114. Practice Committee of the American Society for Reproductive Medicine. Diagnostic evaluation of the infertile female: a committee opinion. Fertil Steril 2015;103:44–50. Somigliana E, Paffoni A, Busnelli A, Filippi F, Pagliardini L, Vigano P, Vercellini P. Age-related infertility and unexplained infertility: an intricate clinical dilemma. Hum Reprod 2016;31:1390–1396. Tinghög G. Discounting, preferences, and paternalism in cost-effectiveness analysis. Health Care Anal 2012;20:297–318. Van Eekelen R, Scholten I, Tjon-Kon-Fat RI, van der Steeg JW, Steures P, Hompes P, van Wely M, van der Veen F, Mol BW, Eijkemans MJ et al. Natural conception: repeated predictions over time. Hum Reprod 2017; 32:346–353. Van Houwelingen HC, Putter H. Dynamic Prediction in Clinical Survival Analysis, 1st edn. Boca Raton, USA: CRC Press, 2012. van Loendersloot LL, van Wely M, Limpens J, Bossuyt PM, Repping S, van der Veen F. Predictive factors in in vitro fertilization (IVF): a systematic review and meta-analysis. Hum Reprod Update 2010;16:577–589. Van Voorhis BJ. Clinical practice. In vitro fertilization. N Engl J Med 2007; 356:379–386. Zegers-Hochschild F, Adamson GD, de Mouzon J, Ishihara O, Mansour R, Nygren K, Sullivan E, van der Poel S. The International Committee for Monitoring Assisted Reproductive Technology (ICMART) and the World Health Organization (WHO) Revised Glossary on ART Terminology, 2009. Hum Reprod 2009;24:2683–2687.