Abstract

Groundwork: It is well known that miscarriage risk increases with age. Even so, studies usually investigate only maternal historic period effects. We investigated both maternal historic period and paternal age furnishings on miscarriage risk to provide insight into this frequent reproductive failure. METHODS: The terminal planned pregnancies (n = 3174) that ended in a nascence or miscarriage were analysed in a retrospective population-based written report on women aged 25–44 years in Denmark, Germany, Italian republic and Espana. Maternal and paternal ages were analysed together, using a unmarried variable `couple age' in a multivariate logistic regression analysis, with couples equanimous of a adult female and a homo both anile 20–29 years forming the reference group. RESULTS: After aligning for various factors (e.g. reproductive history, country), we plant that the risk of miscarriage was higher if the woman was aged ≥35 years, every bit has already been reported in a number of studies. Notwithstanding, the increase in risk was much greater for couples composed of a woman aged ≥35 years and of a human aged ≥40 years. Potential source of bias (especially `reproductive bounty') are discussed. CONCLUSIONS: The gamble of an adverse pregnancy outcome is highest if both partners are avant-garde in age.

Introduction

The proportion of women giving nascency after 35 years of age has risen considerably in industrialized countries. In the Usa, the birth rate (per g) of women aged 35–39 years increased from xix.0 in 1976 to 37.4 in 1998 ( Ventura et al., 1988; Guyer et al., 1999). Over the same period, several studies accept concluded that women aged ≥35 years take a higher frequency of various adverse reproductive events: infertility, spontaneous abortion, pregnancy complications (such every bit Caesarean section, pre-eclampsia), congenital abnormalities, maternal mortality and perinatal mortality, than practice younger women (Schwartz and Mayaux, 1982; Newcomb et al., 1991; Nybo Andersen et al., 2000).

One of the most frequent agin reproductive events is spontaneous ballgame, with upwardly to 10% of recognized conceptions lost during the pregnancy ( Nybo Andersen et al., 2000). Nigh studies concerning the chance factors for spontaneous abortion have concluded that the predominant negative furnishings are those of advanced maternal age (with a articulate increment in risk after 35 years) and previous spontaneous abortion (Leridon, 1976; Regan et al., 1989; Coste et al., 1991; Fretts et al., 1995; Nybo Andersen et al., 2000; Osborn et al., 2000). Conflicting results have been obtained for other spontaneous ballgame risk factors such every bit maternal consumption of tobacco, maternal psychological problems, interval between pregnancies, or previous employ of the contraceptive pill ( Risch et al., 1988; Coste et al., 1991; Dominguez et al., 1991). Thus about of these studies analysed the effect of female factors on spontaneous abortion. Male factors may also affect spontaneous ballgame run a risk, but have been niggling analysed. Virtually studies on male person factors have focused on the outcome of professional exposure of the man to toxic substances (such as ethylene oxide, rubber chemicals, solvents, thiocarbamates, carbaryl) on spontaneous abortion chance in his partner ( Lindbohm et al., 1991; Savitz et al., 1997). Paternal age has just rarely been considered: simply five studies (published before 1980) take investigated the run a risk of spontaneous abortion co-ordinate to both maternal and paternal age (Yerushalmy, 1939; Warburton and Fraser, 1964; Woolf, 1965; Resseguie, 1976; Selvin and Garfinkel, 1976). Four of these studies analysed medical certificates for fetal deaths occurring after 20 weeks gestation (Yerushalmy, 1939; Woolf, 1965; Resseguie, 1976; Selvin and Garfinkel, 1976). Three of these iv studies concluded that paternal historic period had an effect, after adjusting for maternal age (Yerushalmy, 1939; Woolf, 1965; Selvin and Garfinkel, 1976); the fourth did not (Resseguie, 1976). Withal, only one of these studies used a multivariate model to accept confounding effects into account but, unfortunately, in this model, the effect of age was assumed to be linear (Selvin and Garfinkel, 1976). A fifth study analysed the run a risk of spontaneous abortion in mothers with a child presenting a possible genetic defect (Warburton and Fraser, 1964). Using a multiple regression model, the authors were unable to assess the relative effects of paternal and maternal age. Even so, they suggested that paternal age had a greater effect than maternal age.

We addressed the question of an upshot of paternal age on the risk of miscarriage by studying data recorded in the European Multicenter Written report on Infertility and Subfecundity conducted between 1991 and 1993, which took the couple as the unit of measurement of assay although only the women were interviewed. We studied maternal and paternal ages together, using a single categorical variable, `couple age', in a multivariate model of miscarriage risk.

Materials and methods

The data used in this report are from the retrospective European Study of Infertility and Subfecundity, conducted betwixt Baronial 1991 and February 1993. Fourteen population-based samples of women aged 25–44 years were randomly selected from census registers and electoral rolls in towns in Denmark, Federal republic of germany, Italian republic and Kingdom of spain. The same protocol and questionnaire were practical to all samples. The questionnaire was adult in English, translated into national languages then translated back into English language to check the translations. The aim of the European Written report of Infertility and Subfecundity was to evaluate the frequency of, and take chances factors for, subfecundity and infertility in these countries. The `unit of measurement of analysis' was the couple, although only women were questioned, by trained women, in a personal interview lasting xxx–90 min. Questions concerned sociodemographic characteristics (male and female), reproductive history, time taken to reach the showtime and final pregnancies other than those catastrophe in induced abortion, utilise of contraception, the seeking of medical advice by couples with fertility problems, recent occupational exposures and working conditions of the human being and adult female and the lifestyle (booze and java consumption) of the human being and woman.

We analysed terminal planned pregnancies ending in miscarriage or birth. We excluded other causes of death (induced abortion, ectopic pregnancy, stillbirth and death within 7 days post-obit the birth), which may take aetiologies other than miscarriage. We likewise excluded pregnancies that were not completed at the time of interview. We restricted our analysis to planned pregnancies, as unplanned pregnancies ending in induced abortion may exist alleged by the woman as a miscarriage. Moreover, in the case of unplanned pregnancy, the notion of `time to pregnancy' is meaningless.

Constitution of historic period classes

Previous studies ( Coste et al., 1991; Nybo Andersen et al., 2000; Osborn et al., 2000) have shown no meaning effect of maternal age on spontaneous abortion between the ages of 20 and of 29 years. We therefore used the xx–29 yr historic period grouping equally the reference for both women and men. As we had very few couples including a woman or homo aged <twenty years and equally nosotros were interested in the effects of ageing, nosotros decided to exclude these younger couples. For couples aged ≥30 years, we stratified data into v year classes. Every bit we found only 20 women aged ≥40 years and 61 men aged ≥45 years, nosotros pooled these `older' individuals with the previous age group so the oldest historic period group was 35–44 years for women and 40–64 years for men. Thus, maternal historic period was divided into three age classes: xx–29 (reference age group), 30–34 and 35–44 years; and paternal age was divided into four age classes: 20–29 (reference age group), thirty–34, 35–39 and 40–64 years. Maternal and paternal ages were significantly correlated (r = 0.64, P < 0.0001). Every bit a result of this correlation, analysis of maternal historic period and paternal age in the same multivariate model posed bug of colinearity. To bargain with this, nosotros divers a new variable, `couple age', combining the three maternal age groups with the iv paternal age groups. We defined a total of 12 classes for this variable `couple historic period' (presented in Figure 1a ). The `couple historic period' variable made it possible to study maternal and paternal ages.

The potential confounding factors considered included the land (Kingdom of denmark, Federal republic of germany, Spain, and Italy); the time to pregnancy (TTP, classed every bit `TTP <half-dozen months', `6 to <12 months' and `TTP ≥12 months'); the smoking habits of the woman and of the man (recorded as `yes/no' responses: the woman/the man smoked/did not smoke at the fourth dimension they started having intercourse without using birth command); the past reproductive history of the woman recorded by four dichotomous variables: number of the pregnancy (= one/>1), history of miscarriage (yes/no), history of ectopic pregnancy (aye/no), history of induced abortion (yes/no).

We carried out unadjusted and adjusted logistic regression analyses (using SAS®) according to the various risk factors for miscarriage. We tested goodness of fit by the Hosmer–Lemeshow exam (Hosmer and Lemeshow, 1989).

Results

The European Study of Infertility and Subfecundity included 6188 women. Of these, we showtime excluded 2958 subjects because they did not satisfy analysis inclusion criteria: women who had never been pregnant (n = 1518); women whose pregnancies had all ended in induced abortion (n = 226); women whose last pregnancy was not a miscarriage or a live birth, i.due east. the last pregnancy was ectopic (north = 55), a stillbirth (n = 16), a death within seven days (n = viii), or was not nonetheless completed (northward = 204); the final pregnancy was unplanned (due north = 879); the woman or her partner was <20 years of age (n = 52). Another 56 subjects were excluded because of missing information. The remaining 3174 pregnancies constituted the report population, for which 4.9% ended in a miscarriage. The characteristics of the study population [number/pct/unadjusted odds ratio (OR)] are shown in Table I : this population consisted predominantly of couples equanimous of a woman and a human being both aged xx–29 years (32.0%) who took <six months (70.two%) to achieve a pregnancy that was not the first (69.3%).

We analysed age outcome by estimating OR with logistic regression and assessed the goodness of fit of our model by the Hosmer–Lemeshow test (P = 0.76). Table Ii shows the adjusted miscarriage OR for the 12 couple-age classes (taking couples of women and men both 20–29 years old as the reference grouping), land, number of the pregnancy, TTP, female and male smoking, history of miscarriage, history of ectopic pregnancy and history of induced abortion. The results for the couple-age classes of Table II are also presented visually in Figure 1a . In that location was a significant increase in the risk of miscarriage if the woman was 30–34 years of age and the man was xl–64 years of age, or if the woman was 35–44 years of age, whatever the age of the human. This increase was much more marked for couples composed of a woman aged ≥35 years and of a man aged ≥forty years: for these couples the OR was 6.73 [95% confidence interval (CI): 3.50–12.95], versus 3.38 (95% CI: 1.76–6.47) for couples composed of a woman anile ≥35 years and of a man aged 35–39 years. The OR for couples composed of a woman aged ≥35 years and of a man aged 20–29 years was also loftier (OR = 9.18) but this was probably due to the small number of couples in this group (n = 10), the confidence interval being very large (95% CI: i.lxxx–46.66). To summarize these trends, we regrouped couple-historic period classes co-ordinate to the OR obtained in Table Ii (and reported in Figure 1b ) into three `zones' (Effigy 1c ): `standard zone' (no significant increase in miscarriage risk), `high zone' (significant increment in miscarriage run a risk), and `highest zone' (major increase in miscarriage adventure). Figure 1c shows the adjusted miscarriage OR for these three couple-historic period zones: the adventure of miscarriage was significantly higher in the zones of `high' ( OR: two.87; 95% CI: one.86–4.45) and `highest' (5.65; 3.twenty–9.98) risk than in the `standard take a chance' zone (Figure 1c ). The OR was too significantly higher in the `highest risk' zone (i.97; 1.03–3.77) than in the `loftier risk' zone.

The frequency of miscarriage (Tabular array II ) was as well significantly higher if TTP was between 6 and 12 months (OR: two.27; 95% CI: ane.44–3.60) or >12 months (two.33; one.58–3.43). A history of miscarriage (ii.29; 1.53–3.42) and a history of ectopic pregnancy (3.98; 1.42–11.18) were likewise linked to miscarriage. No increase in miscarriage risk was associated with land, number of the pregnancy, female person and male smoking and history of induced abortion.

Word

Analysis of the terminal pregnancy outcome for 3174 women interviewed in a large European written report showed a clear effect of maternal historic period and paternal age on the risk of miscarriage, with iii trends, every bit shown in Effigy 1c . If the adult female was twenty–29 years old, the risk of miscarriage was not significantly high whatever the age of the homo. If the woman was 30–34 years old, the hazard of miscarriage was higher if the human was aged ≥40 years. If the woman was aged ≥35 years, the hazard of miscarriage increased whatever the age of the man. However, if the woman was aged ≥35 years and the homo was aged ≥xl years, the risk of miscarriage was essentially higher with an OR of i.97 (1.03, 3.77) when comparison the `highest' and the `high' risk zones (Figure 1c ). We checked that our results were not due to the inclusion of the oldest individuals (≥xl years onetime for women and ≥45 years one-time for men) by restricting our analysis to women aged xx–39 years and to men aged 20–44 years. When doing this, like results were obtained. In addition, in that location may be a residual confounding issue betwixt maternal and paternal ages: in one age class, the women were one year older on average if the man was xl–64 years old than if the man was 20–29 years old. For example, in the form in which the women were 35–44 years old and the men were 20–29 years onetime, the mean historic period of the women was 36.4 years whereas in the form in which the women were 35–44 years old and the men were 40–64 years old, the mean historic period of the women was 37.1 years. Still, we estimated that this residual misreckoning event could not account for the large increase in miscarriage hazard for couples equanimous of a woman aged ≥35 years and of a man aged ≥forty years. We recognize that studying historic period in large classes (≥5 years) limits the analysis to threshold effects only (Effigy 1 ) and that age probably has a much less clear-cutting effect, having a gradual deleterious impact. Our assay was also limited by the small numbers in some age groups (specially for couples composed of a adult female aged ≥35 years and of a man aged 20–29 years or 30–34 years).

Several studies take shown an increment in the risk of spontaneous abortion in women aged ≥35 years ( Dominguez et al., 1991; Nybo Andersen et al., 2000; Osborn et al., 2000). Multiple regression analysis in a cross-sectional unmatched case–command report ( Dominguez et al., 1991) has shown that maternal age begins to have an result simply after the historic period of 35 years. Our results are consistent with this conclusion but, taking paternal age into account, we establish that (i) if the man was ≥forty years of historic period, maternal age began to take a negative effect before, from the age of 30 years; (2) if the man was ≥40 years of age, the increase in miscarriage take a chance was much higher if the woman was aged ≥35 years. In previous studies, the effect of paternal age has rarely been analysed. The about relevant study (Selvin and Garfinkel, 1976) assessed, in a multivariate logistic model, the risk of spontaneous ballgame based on >ane.v×x6 birth and fetal death certificates recorded in New York State from 1959 to 1967. The authors concluded that maternal and paternal ages have furnishings of like significance. All the same, in their logistic model, they considered that maternal and paternal age effects were linear, a controversial supposition: reproductive age patterns are usually represented as J- or U-shaped curves ( Nybo Andersen et al., 2000). Nosotros used a more advisable methodological approach and found a clear negative effect of paternal age after 40 years.

Various hypotheses have been put forwards to business relationship for the increment in adverse reproductive outcomes with age. In women, a link between increasing age and a higher incidence of chromosomal abnormality has been established ( Boue et al., 1975; Cowchock et al., 1993). However, conflicting conclusions have been fatigued concerning the possible effect of age on oocyte quality and uterus senescence ( Levran et al., 1991; Abdalla et al., 1993; Cano et al., 1995). In men, studies using a fluorescence in-situ hybridization approach have suggested that the probability of producing aneuploid offspring is college for older fathers (≥50 years former) than for younger fathers (<thirty years old) ( Griffin et al., 1995). A recent newspaper comparing men aged 23–39 years with men aged 59–74 years also concluded that in that location was a higher frequency of sperm chromosome aberrations in older men ( Sartorelli et al., 2001). These results are presumably related to the higher risk of birth defects in the offspring of older fathers, specially for those linked to new dominant mutations (Penrose, 1955; Savitz et al., 1991; McIntosh et al., 1995), giving a general pattern of increasing relative risk of adverse reproductive outcomes with paternal age. This paternal age result led to fixing of the upper age limit for semen donors at forty years (Bordson and Leonardo, 1991). In addition, if sexual activity declines with male age, this may lead to an increment in the risk of miscarriage linked to fertilization of an `one-time' (time elapsed between ovulation and fertilization) oocyte (Guerrero and Rojas, 1975). Withal, this hypothesis is controversial ( Gray et al., 1995).

Our report confirmed the importance of a history of miscarriage, a well-documented miscarriage risk cistron (Leridon, 1976; Risch et al., 1988; Regan et al., 1989; Coste et al., 1991). In this study, a history of ectopic pregnancy was also a gamble factor for miscarriage, consistent with the results of other studies (Honore, 1979; Coulam et al., 1989; Fedele et al., 1989). We also found that the risk of miscarriage was significantly higher for couples taking ≥6 months to excogitate. This association between a long time being required for formulation and a college risk of miscarriage is less well documented, and has been reported in some studies ( Tietze et al., 1950; Rachootin and Olsen, 1982; Strobino et al., 1986; Schaumburg and Boldsen, 1992; Baird et al., 1993; Joffe and Li, 1994; Hakim et al., 1995; Greyness and Wu, 2000). In retrospective studies, reproductive effect observations depend on the statement of the woman, and subclinical abortions are completely omitted. As estimates of the subclinical ballgame rate range from 8 to 78% ( Wilcox et al., 1988; Modvig et al., 1990; Zinaman et al., 1996), these unnoticed spontaneous abortions may be an important factor in the time to conception declared by women. The observed higher miscarriage chance in couples taking longer to conceive may be due partly to an clan with subclinical abortions. The three risk factors found in our analysis suggest that there may exist a common crusade for such unlike adverse reproductive events: delay in conception, ectopic pregnancy, miscarriage, and perhaps too infertility. This notion of the aforementioned `central reproductive disorder' giving `various adverse reproductive outcomes' requires further investigation.

Bias and limits

The data were collected from retrospective declarations by the women, so two limitations must be considered concerning pregnancy outcomes. Commencement, it has been estimated that only three-quarters of spontaneous abortions are recalled in this blazon of study (Wilcox and Horney, 1984), but, every bit our analysis concerned the effect of the terminal pregnancy, women were probably less likely to have forgotten. 2nd, the notion of miscarriage was not precisely defined, every bit information technology would have been in medical information. Women may have been confused to some extent concerning the type of death. However, data were collected by means of interviews by trained women who could aid women to differentiate between the types of deaths if needed. Induced abortion may also take been alleged equally miscarriage ( Simonds et al., 1998; Houzard et al., 2000). We restricted our assay to planned pregnancies to limit this possible bias. Furthermore, the frequency of miscarriage observed in our study (iv.9%) is consistent with that of ~v–6% reported in other studies of last pregnancy outcomes ( Weinberg et al., 1994). This is almost half that for starting time pregnancies, which is 10–15% ( Weinberg et al., 1994). As suggested by Weinberg et al., such a low rate for last pregnancies may reduce the statistical power of the study ( Weinberg et al., 1994). The low miscarriage rate for last pregnancy outcomes is due to the couple'due south attitude: couples tend to `supplant' a miscarriage by another pregnancy until they attain a live birth (a phenomenon called `reproductive bounty'). The run a risk of succeeding in replacing a miscarriage depends on the time required to conceive. Information technology had been concluded that this time could increment with maternal age and paternal age ( Ford et al., 2000) simply this age event was afterwards called into question (Sallmen and Luukkonen, 2001). Such factors may consequence in overestimation of the couple-age effect in our analyses. Even so, our results on historic period effect in the `high-risk zone' (OR = 2.87) are consequent with those of other studies on maternal age. For instance, an OR of 3.13 (95% CI: one.56, six.26) was institute for women aged ≥35 years in a cross-sectional example–control study ( Dominguez et al., 1991). Nosotros thus observed no major age overestimation for these ages linked to the analysis of final pregnancies. We planned to adjust for the effect of a possible `replacement mental attitude' by studying commencement pregnancy outcomes; ten.five% of offset pregnancies concluded in miscarriage in the European Written report of Infertility and Subfecundity. Still, first pregnancy outcomes were not suitable for analysing the age effect because nearly all the couples concerned were young (for showtime pregnancies, 96.5% of couples were in the `standard risk' zone in Effigy 1c , 2.half dozen% in the `high adventure' zone and 0.nine% in the `highest chance' zone).

Finally, the multicentre European study was carried out in urban areas. Nevertheless, in our opinion, there is little or no difference betwixt rural and urban areas because no conclusive results have been produced concerning this issue (Hemminki and Forssas, 1999).

In summary, we demonstrate here the importance of taking into account both maternal and paternal age to measure miscarriage risk: if the woman is ≥35 years of age and the human is ≥40 years of historic period, the take a chance of miscarriage is substantially college than for couples of other age combinations. The idea of major reproductive take chances for couples combining negative male and female factors has already emerged in studies of couples seeking medical advice ( Emperaire et al., 1982; Spira, 1986; Thonneau et al., 1991). Thus, a French multicentre survey conducted in 1988–1989 on couples who consulted a doctor for infertility showed that in 39% of cases, both the homo and the woman presented reproductive disorders and concluded that `infertility is essentially a `couple'southward concern" ( Thonneau et al., 1991). More generally, this work raises questions concerning the level of miscarriage risk for couples in assisted reproduction programmes: the turn down in assisted reproduction success rate with the historic period of the adult female is well known ( Rosenwaks et al., 1995). In Australia, 26% of the women who became pregnant past IVF procedures during the 1979–1986 period were ≥35 years of age. Among these women, the charge per unit of spontaneous abortion was 34.8 versus 21.v% amid women anile <35 years (Saunders and Lancaster, 1989). In the United states of america, 8.iv% of the women who became pregnant by IVF procedures during 1991 were ≥xl years of age. For these women, the rate of spontaneous abortion was 33.nine versus 18.half-dozen% among women aged <40 years (American Fertility Social club, 1993).

In conclusion, older couples (and specially those asking for assisted reproduction), i.e. couples composed of a woman aged ≥35 years and of a man aged ≥40 years, should be informed that they have a high risk of miscarriage. The importance of investigating the paternal historic period effect was pointed out by R.Jeffrey Chang, Md, President of ASRM at the 2000 almanac meeting: "The impact of historic period on reproductive health is a vital result for the 21st Century. We have known that historic period was an of import factor for women'due south reproductive health. The question of declining fertility as men historic period is an important 1 that needs boosted research to resolve".

Table I.

Characteristics of the written report population used for analysis of miscarriage risk, European Study on Infertility and Subfecundity, 1991–1993 (n = 3174)

Variable northo. % Rough OR (95% CI)
aReference grouping.
OR = odds ratio; CI = confidence interval.
Couple age (years)
    Maternal Paternal
    xx–29 20–29a 1016 32.0
    20–29 thirty–34 719 22.7 1.05 (0.61, 1.82)
    20–29 35–39 169 v.3 1.37 (0.threescore, 3.17)
    20–29 40–64 46 1.5 2.22 (0.65, 7.54)
    30–34 20–29 77 ii.4 two.21 (0.83, 5.85)
    30–34 30–34 474 14.9 i.77 (i.03, 3.03)
    30–34 35–39 302 9.five one.xx (0.60, 2.42)
    30–34 forty–64 82 2.6 3.44 (1.53, 7.74)
    35–44 20–29 10 0.3 vii.94 (1.62, 38.96)
    35–44 xxx–34 32 1.0 4.54 (one.50, thirteen.73)
    35–44 35–39 150 4.7 four.06 (ii.19, vii.54)
    35–44 40–64 97 three.1 7.74 (iv.18, 14.33)
Country
    Denmarka 527 16.6
    Germany 858 27.0 ane.13 (0.68, one.84)
    Spain 471 xiv.9 0.88 (0.48, 1.63)
    Italy 1318 41.5 one.14 (0.71, ane.84)
Pregnancy rank
    1a 974 30.7
    >1 2200 69.iii 1.33 (0.92, 1.93)
Fourth dimension to pregnancy
    <6 monthsa 2227 70.2
    6 to <12 months 375 11.8 2.37 (1.52, 3.69)
    ≥12 months 572 18.0 2.71 (1.87, 3.92)
Woman smoking
    Noa 1879 59.2
    Yes 1295 40.8 1.27 (0.92, 1.76)
Man smoking
    Noa 1318 41.5
    Yeah 1856 58.5 1.10 (0.79, 1.53)
History of miscarriage
    Noa 2671 84.ii
    Yes 503 15.8 ii.37 (1.65, three.39)
History of ectopic pregnancy
    Noa 3142 99.0
    Yes 32 1.0 3.69 (1.40, nine.73)
History of induced ballgame
    Noa 2879 ninety.7
    Yes 295 9.iii 1.76 (1.xi, ii.79)
Variable no. % Crude OR (95% CI)
aReference group.
OR = odds ratio; CI = confidence interval.
Couple age (years)
    Maternal Paternal
    20–29 20–29a 1016 32.0
    20–29 30–34 719 22.7 1.05 (0.61, one.82)
    20–29 35–39 169 v.3 one.37 (0.60, 3.17)
    20–29 40–64 46 i.5 ii.22 (0.65, vii.54)
    xxx–34 xx–29 77 2.four 2.21 (0.83, v.85)
    30–34 thirty–34 474 xiv.9 1.77 (1.03, 3.03)
    thirty–34 35–39 302 ix.5 1.20 (0.60, ii.42)
    30–34 40–64 82 two.6 three.44 (1.53, vii.74)
    35–44 20–29 x 0.iii 7.94 (one.62, 38.96)
    35–44 30–34 32 i.0 4.54 (one.l, xiii.73)
    35–44 35–39 150 4.vii iv.06 (2.19, 7.54)
    35–44 40–64 97 3.1 7.74 (iv.xviii, 14.33)
Country
    Denmarka 527 16.six
    Germany 858 27.0 i.13 (0.68, 1.84)
    Spain 471 14.nine 0.88 (0.48, 1.63)
    Italy 1318 41.five ane.fourteen (0.71, 1.84)
Pregnancy rank
    1a 974 30.7
    >1 2200 69.3 1.33 (0.92, ane.93)
Fourth dimension to pregnancy
    <vi monthsa 2227 70.two
    6 to <12 months 375 11.8 2.37 (1.52, 3.69)
    ≥12 months 572 18.0 two.71 (1.87, iii.92)
Adult female smoking
    Noa 1879 59.2
    Yes 1295 twoscore.eight 1.27 (0.92, ane.76)
Man smoking
    Noa 1318 41.5
    Yeah 1856 58.v 1.ten (0.79, 1.53)
History of miscarriage
    Noa 2671 84.2
    Aye 503 15.8 2.37 (1.65, 3.39)
History of ectopic pregnancy
    Noa 3142 99.0
    Yes 32 1.0 3.69 (one.forty, 9.73)
History of induced abortion
    Noa 2879 90.vii
    Yes 295 9.3 one.76 (one.11, 2.79)

Table I.

Characteristics of the study population used for analysis of miscarriage risk, European Study on Infertility and Subfecundity, 1991–1993 (northward = 3174)

Variable northo. % Rough OR (95% CI)
aReference group.
OR = odds ratio; CI = conviction interval.
Couple age (years)
    Maternal Paternal
    xx–29 xx–29a 1016 32.0
    20–29 xxx–34 719 22.7 1.05 (0.61, i.82)
    twenty–29 35–39 169 5.three 1.37 (0.60, 3.17)
    20–29 forty–64 46 ane.5 2.22 (0.65, seven.54)
    thirty–34 20–29 77 2.4 2.21 (0.83, 5.85)
    30–34 30–34 474 14.9 1.77 (1.03, 3.03)
    30–34 35–39 302 9.5 1.20 (0.60, 2.42)
    thirty–34 40–64 82 2.half dozen 3.44 (ane.53, 7.74)
    35–44 20–29 10 0.3 vii.94 (one.62, 38.96)
    35–44 30–34 32 1.0 four.54 (1.50, 13.73)
    35–44 35–39 150 four.vii 4.06 (ii.19, 7.54)
    35–44 40–64 97 three.1 7.74 (4.18, 14.33)
Country
    Denmarka 527 16.half dozen
    Deutschland 858 27.0 one.xiii (0.68, 1.84)
    Kingdom of spain 471 14.9 0.88 (0.48, 1.63)
    Italia 1318 41.5 1.14 (0.71, 1.84)
Pregnancy rank
    anea 974 30.vii
    >1 2200 69.three 1.33 (0.92, 1.93)
Time to pregnancy
    <half dozen monthsa 2227 70.ii
    vi to <12 months 375 11.8 ii.37 (ane.52, 3.69)
    ≥12 months 572 xviii.0 2.71 (1.87, 3.92)
Woman smoking
    Noa 1879 59.2
    Yep 1295 40.8 one.27 (0.92, 1.76)
Homo smoking
    Noa 1318 41.5
    Yep 1856 58.5 1.x (0.79, one.53)
History of miscarriage
    Noa 2671 84.2
    Yes 503 xv.eight 2.37 (ane.65, 3.39)
History of ectopic pregnancy
    Noa 3142 99.0
    Yeah 32 1.0 3.69 (1.40, ix.73)
History of induced abortion
    Noa 2879 90.7
    Yes 295 9.three 1.76 (one.xi, two.79)
Variable no. % Crude OR (95% CI)
aReference group.
OR = odds ratio; CI = conviction interval.
Couple age (years)
    Maternal Paternal
    twenty–29 20–29a 1016 32.0
    20–29 30–34 719 22.7 1.05 (0.61, ane.82)
    xx–29 35–39 169 5.3 1.37 (0.60, 3.17)
    20–29 forty–64 46 1.five two.22 (0.65, 7.54)
    30–34 xx–29 77 2.4 ii.21 (0.83, five.85)
    30–34 30–34 474 14.9 i.77 (ane.03, 3.03)
    30–34 35–39 302 nine.5 1.20 (0.60, 2.42)
    xxx–34 40–64 82 2.six 3.44 (1.53, vii.74)
    35–44 20–29 10 0.3 7.94 (i.62, 38.96)
    35–44 30–34 32 1.0 four.54 (1.50, thirteen.73)
    35–44 35–39 150 4.7 4.06 (2.19, 7.54)
    35–44 40–64 97 3.one 7.74 (iv.xviii, xiv.33)
Country
    Denmarka 527 16.vi
    Federal republic of germany 858 27.0 1.thirteen (0.68, 1.84)
    Spain 471 14.ix 0.88 (0.48, i.63)
    Italia 1318 41.5 1.xiv (0.71, 1.84)
Pregnancy rank
    onea 974 thirty.7
    >1 2200 69.3 1.33 (0.92, 1.93)
Time to pregnancy
    <six monthsa 2227 70.2
    6 to <12 months 375 11.8 two.37 (1.52, iii.69)
    ≥12 months 572 18.0 two.71 (one.87, 3.92)
Adult female smoking
    Noa 1879 59.2
    Yes 1295 40.8 1.27 (0.92, one.76)
Man smoking
    Noa 1318 41.5
    Yes 1856 58.5 ane.x (0.79, 1.53)
History of miscarriage
    Noa 2671 84.2
    Yes 503 15.8 two.37 (1.65, 3.39)
History of ectopic pregnancy
    Noa 3142 99.0
    Yes 32 1.0 3.69 (1.40, ix.73)
History of induced ballgame
    Noa 2879 90.vii
    Yeah 295 9.iii 1.76 (1.11, 2.79)

Tabular array II.

Adjusted associations of couple-age classes with pregnancy upshot (miscarriage/live-nascency), European Study on Infertility and Subfecundity 1991–1993 (north = 3174)

Variable Adjusted OR (95% CI)
aReference group.
OR = odds ratio; CI = confidence interval.
Couple age (years)
    Maternal Paternal
    xx–29 xx–29a
    twenty–29 30–34 1.06 (0.61, i.86)
    20–29 35–39 1.31 (0.56, iii.07)
    20–29 xl–64 1.80 (0.52, 6.24)
    30–34 20–29 1.72 (0.62, 4.74)
    xxx–34 xxx–34 1.62 (0.93, 2.82)
    thirty–34 35–39 1.06 (0.52, 2.17)
    30–34 40–64 2.ninety (1.26, 6.67)
    35–44 20–29 nine.18 (1.lxxx, 46.66)
    35–44 30–34 three.87 (1.24, 12.02)
    35–44 35–39 3.38 (i.76, half dozen.47)
    35–44 xl–64 6.73 (3.50, 12.95)
State
    Denmarka
    Germany 1.22 (0.71, 2.09)
    Kingdom of spain 1.18 (0.61, ii.27)
    Italia i.35 (0.81, ii.26)
Pregnancy rank
    anea
    >1 0.84 (0.54, 1.31)
Time to pregnancy
    <6 monthsa
    half dozen to <12 months 2.27 (1.44, 3.threescore)
    ≥12 months 2.33 (1.58, 3.43)
Woman smoking
    Noa
    Yes 1.27 (0.90, i.79)
Man smoking
    Noa
    Yes one.09 (0.77, 1.56)
History of miscarriage
    Noa
    Aye 2.29 (1.53, 3.42)
History of ectopic pregnancy
    Noa
    Yes iii.98 (1.42, xi.18)
History of induced ballgame
    Noa
    Yeah 1.56 (0.93, two.63)
Variable Adjusted OR (95% CI)
aReference group.
OR = odds ratio; CI = confidence interval.
Couple age (years)
    Maternal Paternal
    20–29 20–29a
    20–29 30–34 ane.06 (0.61, ane.86)
    20–29 35–39 1.31 (0.56, 3.07)
    20–29 40–64 1.80 (0.52, six.24)
    30–34 20–29 1.72 (0.62, 4.74)
    30–34 30–34 1.62 (0.93, two.82)
    30–34 35–39 1.06 (0.52, 2.17)
    30–34 40–64 ii.90 (one.26, vi.67)
    35–44 20–29 9.18 (one.fourscore, 46.66)
    35–44 30–34 3.87 (1.24, 12.02)
    35–44 35–39 3.38 (1.76, 6.47)
    35–44 40–64 vi.73 (3.50, 12.95)
Country
    Kingdom of denmarka
    Germany i.22 (0.71, 2.09)
    Spain 1.18 (0.61, ii.27)
    Italy ane.35 (0.81, 2.26)
Pregnancy rank
    anea
    >i 0.84 (0.54, 1.31)
Fourth dimension to pregnancy
    <6 monthsa
    half dozen to <12 months 2.27 (i.44, three.threescore)
    ≥12 months 2.33 (1.58, iii.43)
Woman smoking
    Noa
    Yep 1.27 (0.90, one.79)
Man smoking
    Noa
    Yep i.09 (0.77, 1.56)
History of miscarriage
    Noa
    Aye two.29 (1.53, 3.42)
History of ectopic pregnancy
    Noa
    Yes 3.98 (ane.42, 11.18)
History of induced abortion
    Noa
    Yeah 1.56 (0.93, 2.63)

Table Two.

Adjusted associations of couple-historic period classes with pregnancy result (miscarriage/alive-nativity), European Report on Infertility and Subfecundity 1991–1993 (n = 3174)

Variable Adapted OR (95% CI)
aReference grouping.
OR = odds ratio; CI = confidence interval.
Couple age (years)
    Maternal Paternal
    20–29 twenty–29a
    20–29 30–34 one.06 (0.61, 1.86)
    twenty–29 35–39 ane.31 (0.56, 3.07)
    20–29 40–64 1.fourscore (0.52, vi.24)
    30–34 20–29 1.72 (0.62, 4.74)
    30–34 xxx–34 1.62 (0.93, ii.82)
    thirty–34 35–39 i.06 (0.52, 2.17)
    30–34 40–64 2.90 (1.26, 6.67)
    35–44 20–29 9.xviii (1.eighty, 46.66)
    35–44 xxx–34 3.87 (1.24, 12.02)
    35–44 35–39 three.38 (1.76, 6.47)
    35–44 xl–64 vi.73 (3.50, 12.95)
Country
    Denmarka
    Germany one.22 (0.71, 2.09)
    Espana 1.xviii (0.61, 2.27)
    Italian republic 1.35 (0.81, 2.26)
Pregnancy rank
    1a
    >1 0.84 (0.54, 1.31)
Time to pregnancy
    <6 monthsa
    6 to <12 months two.27 (one.44, 3.60)
    ≥12 months 2.33 (1.58, 3.43)
Woman smoking
    Noa
    Yes 1.27 (0.90, 1.79)
Human smoking
    Noa
    Aye 1.09 (0.77, i.56)
History of miscarriage
    Noa
    Yes ii.29 (1.53, iii.42)
History of ectopic pregnancy
    Noa
    Yes three.98 (1.42, xi.eighteen)
History of induced abortion
    Noa
    Yes 1.56 (0.93, two.63)
Variable Adjusted OR (95% CI)
aReference group.
OR = odds ratio; CI = conviction interval.
Couple age (years)
    Maternal Paternal
    20–29 twenty–29a
    20–29 30–34 1.06 (0.61, i.86)
    20–29 35–39 i.31 (0.56, three.07)
    20–29 40–64 i.eighty (0.52, half dozen.24)
    30–34 twenty–29 1.72 (0.62, 4.74)
    thirty–34 30–34 1.62 (0.93, 2.82)
    30–34 35–39 ane.06 (0.52, two.17)
    30–34 40–64 2.90 (one.26, half dozen.67)
    35–44 20–29 9.xviii (1.lxxx, 46.66)
    35–44 30–34 3.87 (i.24, 12.02)
    35–44 35–39 3.38 (i.76, half-dozen.47)
    35–44 forty–64 6.73 (3.fifty, 12.95)
Country
    Kingdom of denmarka
    Germany 1.22 (0.71, 2.09)
    Spain 1.18 (0.61, 2.27)
    Italy 1.35 (0.81, 2.26)
Pregnancy rank
    onea
    >one 0.84 (0.54, 1.31)
Time to pregnancy
    <6 monthsa
    6 to <12 months 2.27 (1.44, iii.60)
    ≥12 months 2.33 (1.58, 3.43)
Woman smoking
    Noa
    Yes 1.27 (0.ninety, i.79)
Man smoking
    Noa
    Yes 1.09 (0.77, 1.56)
History of miscarriage
    Noa
    Aye ii.29 (1.53, three.42)
History of ectopic pregnancy
    Noa
    Yeah 3.98 (one.42, 11.18)
History of induced abortion
    Noa
    Yes 1.56 (0.93, 2.63)

Figure 1.

Couple age in the study of miscarriage risk, European Study on Infertility and Subfecundity, 1991–1993 (n = 3174).

Couple age in the written report of miscarriage risk, European Study on Infertility and Subfecundity, 1991–1993 (n = 3174).

Figure 1.

Couple age in the study of miscarriage risk, European Study on Infertility and Subfecundity, 1991–1993 (n = 3174).

Couple age in the report of miscarriage run a risk, European Study on Infertility and Subfecundity, 1991–1993 (n = 3174).

3

To whom correspondence should be addressed at: Research Group in Man Fertility, La Grave Hospital, Place Lange, 31052 Toulouse Cedex, France. E-mail service: thonneau.p@chu-toulouse.fr

The authors limited detail thanks to Dr Henri Leridon (National Demographic Studies Institut, Paris, France) for helpful comments and criticism. The European Study of Infertility and Subfecundity is a European Community/Cooperation in Scientific and Technical Research (COST) Activeness Inquiry Programme. The members of the project management group were South.Juul (Project Leader), West.Karmaus, J.Olsen, T.Fletcher, F.Bolumar, I.Figa-Talamanca, P.Thonneau and S.Pautelakis. The studies included in the Research Action were funded by European Community contracts MR4/0205/DK and MR4/0343/DK. Additional support was provided by funds from each participating country. Finally, we thank Julie Sappa (from Alex Edelman & Associates) for English language rereading.

References

Abdalla, H.I., Burton, Thou., Kirkland, A. et al. (

1993

) Age, pregnancy and miscarriage: uterine versus ovarian factors.

Hum. Reprod.

,

8

,

1512

–1517.

American Fertility Society (

1993

) Assisted reproductive technology in the U.s.a. and Canada: 1991 results from the Social club for Assisted Reproductive Technology generated from the American Fertility Society Registry.

Fertil. Steril.

,

59

,

956

–962.

Baird, D.D., Ragan, Northward.B., Wilcox, A.J. et al. (1993) The relationship between reduced fecundability and subsequent fetal loss. In Gray, R., Leridon, H. and Spira, A. (eds), Biomedical and Demographic Determinants of Reproduction. Clarendon Press, Oxford, pp. 329–341.

Bordson, B.L. and Leonardo, V.Southward. (

1991

) The appropriate upper age limit for semen donors: a review of the genetic effects of paternal age.

Fertil. Steril.

,

56

,

397

–401.

Boue, J., Bou, A. and Lazar, P. (

1975

) Retrospective and prospective epidemiological studies of 1500 karyotyped spontaneous human being abortions.

Teratology

,

12

,

11

–26.

Cano, F., Simon, C., Remohi, J. et al. (

1995

) Effect of aging on the female person reproductive organisation: testify for a part of uterine senescence in the decline in female fecundity.

Fertil. Steril.

,

64

,

584

–589.

Coste, J., Job-Spira, Northward. and Fernandez, H. (

1991

) Take chances factors for spontaneous abortion: a case–control study in French republic.

Hum. Reprod.

,

6

,

1332

–1337.

Coulam, C.B., Johnson, P.M., Ramsden, Yard.H. et al. (

1989

) Occurrence of ectopic pregnancy among women with recurrent spontaneous abortion.

Am. J. Reprod. Immunol.

,

21

,

105

–107.

Cowchock, F.South., Gibas, Z. and Jackson, L.G. (

1993

) Chromosome errors equally a crusade of spontaneous abortion: the relative importance of maternal age and obstetric history.

Fertil. Steril.

,

59

,

1011

–1014.

Dominguez, 5., Calle, East., Ortega, P. et al. (

1991

) Adjusting take a chance factors in spontaneous abortion by multiple logistic regression.

Eur. J. Epidemiol.

,

7

,

171

–174.

Emperaire, J.C., Gauzere-Soumireu, E. and Audebert, A.J. (

1982

) Female fertility and donor insemination.

Fertil. Steril.

,

37

,

90

–93.

Fedele, L., Acaia, B., Parazzini, F. et al. (

1989

) Ectopic pregnancy and recurrent spontaneous abortion: ii associated reproductive failures.

Obstet. Gynecol.

,

73

,

206

–208.

Ford, West.C., North, K., Taylor, H. et al. (

2000

) Increasing paternal historic period is associated with delayed conception in a large population of fertile couples: prove for failing fecundity in older men. The ALSPAC Study Team (Avon Longitudinal Written report of Pregnancy and Childhood).

Hum. Reprod.

,

15

,

1703

–1708.

Fretts, R.C., Schmittdiel, J., Mclean, F.H. et al. (

1995

) Increased maternal historic period and the risk of fetal expiry.

Northward. Engl. J. Med.

,

333

,

953

–957.

Gray, R.H. and Wu, Fifty.Y. (

2000

) Subfertility and run a risk of spontaneous abortion.

Am. J. Publ. Health

,

90

,

1452

–1454.

Grey, R.H., Simpson, J.50., Kambic, R.T. et al. (

1995

) Timing of conception and the risk of spontaneous abortion amongst pregnancies occurring during the use of natural family planning.

Am. J. Obstet. Gynecol.

,

172

,

1567

–1572.

Griffin, D.Thousand., Abruzzo, M.A., Millie, East.A. et al. (

1995

) Non-disjunction in homo sperm: evidence for an effect of increasing paternal age.

Hum. Mol. Genet.

,

iv

,

2227

–2232.

Guerrero, R. and Rojas, O.I. (

1975

) Spontaneous abortion and aging of human being ova and spermatozoa.

Northward. Engl. J. Med.

,

293

,

573

–575.

Guyer, B., Hoyert, D.Fifty., Martin, J.A. et al. (

1999

) Annual summary of vital statistics—1998.

Pediatrics

,

104

,

1229

–1246.

Hakim, R.B., Gray, R.H. and Zacur, H. (

1995

) Infertility and early pregnancy loss.

Am. J. Obstet. Gynecol.

,

172

,

1510

–1517.

Hemminki, E. and Forssas, Due east. (

1999

) Epidemiology of miscarriage and its relation to other reproductive events in Finland.

Am. J. Obstet. Gynecol.

,

181

,

396

–401.

Honore, L.H. (

1979

) A pregnant association betwixt spontaneous abortion and tubal ectopic pregnancy.

Fertil. Steril.

,

32

,

401

–402.

Hosmer, D.W. and Lemeshow, Due south. (1989) Applied Logistic Regression. Wiley, New York, 307pp.

Houzard, S., Bajos, N., Warszwawski, J. et al. (

2000

) Analysis of the underestimation of induced abortions in a survey of the general population in French republic.

Eur. J. Contracept. Reprod. Wellness Care

,

5

,

52

–lx.

Joffe, M. and Li, Z. (

1994

) Association of fourth dimension to pregnancy and the outcome of pregnancy.

Fertil. Steril.

,

62

,

71

–75.

Leridon, H. (

1976

) Facts and artifacts in the written report of intra-uterine mortality: a reconsideration from pregnancy outcome.

Popul. Stud.

,

thirty

,

319

–335.

Levran, D., Ben-Shlomo, I., Dor, J. et al. (

1991

) Aging of endometrium and oocytes: observations on conception and abortion rates in an egg donation model.

Fertil. Steril.

,

56

,

1091

–1094.

Lindbohm, Grand.50., Hemminki, K., Bonhomme, G.G. et al. (

1991

) Effects of paternal occupational exposure on spontaneous abortions.

Am. J. Public Health

,

81

,

1029

–1033.

McIntosh, Thousand.C., Olshan, A.F. and Baird, P.A. (

1995

) Paternal age and the gamble of birth defects in offspring.

Epidemiology

,

6

,

282

–288.

Modvig, J., Schmidt, L. and Damsgaard, One thousand.T. (

1990

) Measurement of total risk of spontaneous abortion: the virtue of conditional risk interpretation.

Am. J. Epidemiol.

,

132

,

1021

–1038.

Newcomb, W.West., Rodriguez, M. and Johnson, J.W. (

1991

) Reproduction in the older gravida. A literature review.

J. Reprod. Med.

,

36

,

839

–845.

Nybo Andersen, A.One thousand., Wohlfahrt, J., Christens, P. et al. (

2000

) Maternal age and fetal loss: population based register linkage study.

Br. Med. J.

,

320

,

1708

–1712.

Osborn, J.F., Cattaruzza, M.S. and Spinelli, A. (

2000

) Risk of spontaneous abortion in Italy, 1978–1995, and the effect of maternal age, gravidity, marital status, and education.

Am. J. Epidemiol.

,

151

,

98

–105.

Penrose, 50.S. (1955) Parental age and mutation.

Lancet

,

ii

,

312

–313.

Rachootin, P. and Olsen, J. (

1982

) Prevalence and socioeconomic correlates of subfecundity and spontaneous ballgame in Kingdom of denmark.

Int. J. Epidemiol.

,

eleven

,

245

–249.

Regan, L., Braude, P.R. and Trembath, P.L. (

1989

) Influence of past reproductive performance on risk of spontaneous ballgame.

Br. Med. J.

,

299

,

541

–545.

Resseguie, L.J. (

1976

) Paternal historic period, stillbirths and mutation.

Ann. Hum. Genet.

,

40

,

213

–219.

Risch, H.A., Weiss, Northward.S., Clarke, E.A. et al. (

1988

) Risk factors for spontaneous abortion and its recurrence.

Am. J. Epidemiol.

,

128

,

420

–430.

Rosenwaks, Z., Davis, O.Thousand. and Damario, K.A. (

1995

) The role of maternal age in assisted reproduction.

Hum. Reprod.

,

10

(Suppl. 1),

165

–173.

Sallmen, M. and Luukkonen, R. (

2001

) Is the observed association increasing paternal age and delayed conception an artefact?

Hum. Reprod.

,

16

,

2027

–2028.

Sartorelli, Eastward.K., Mazzucatto, Fifty.F. and De Pina-Neto, J.M. (

2001

) Issue of paternal age on human sperm chromosomes.

Fertil. Steril.

,

76

,

1119

–1123.

Saunders, D.M. and Lancaster, P. (

1989

) The wider perinatal significance of the Australian in vitro fertilization information collection program.

Am. J. Perinatol.

,

half-dozen

,

252

–257.

Savitz, D.A., Schwingl, P.J. and Keels, M.A. (

1991

) Influence of paternal age, smoking, and booze consumption on congenital anomalies.

Teratology

,

44

,

429

–440.

Savitz, D.A., Arbuckle, T., Kaczor, D. et al. (

1997

) Male person pesticide exposure and pregnancy event.

Am. J. Epidemiol.

,

146

,

1025

–1036.

Schaumburg, I. and Boldsen, J.L. (

1992

) Waiting fourth dimension to pregnancy and pregnancy event among Danish workers in the textile, clothing, and footwear industries.

Scand. J. Soc. Med.

,

20

,

110

–114.

Schwartz, D. and Mayaux, M.J. (

1982

) Female person fecundity as a office of historic period: results of bogus insemination in 2193 nulliparous women with azoospermic husbands. Federation CECOS.

North. Engl. J. Med.

,

306

,

404

–406.

Selvin, South. and Garfinkel, J. (

1976

) Paternal age, maternal age and nascency society and the risk of a fetal loss.

Hum. Biol.

,

48

,

223

–230.

Simonds, W., Ellertson, C., Springer, 1000. et al. (

1998

) Abortion, revised: participants in the U.S. clinical trials evaluate mifepristone.

Soc. Sci. Med.

,

46

,

1313

–1323.

Spira, A. (

1986

) Epidemiology of human reproduction.

Hum. Reprod.

,

one

,

111

–115.

Strobino, B., Fox, H.Due east., Kline, J. et al. (

1986

) Characteristics of women with recurrent spontaneous abortions and women with favorable reproductive histories.

Am. J. Publ. Health

,

76

,

986

–991.

Thonneau, P., Marchand, Southward., Tallec, A. et al. (

1991

) Incidence and master causes of infertility in a resident population (1,850,000) of three French regions (1988–1989).

Hum. Reprod.

,

6

,

811

–816.

Tietze, C., Guttmacher, A.F. and Rubin, S. (

1950

) Unintentional ballgame in, 1,497 planned pregnancies.

J. Am. Med. Assoc.

,

142

,

1348

–1350.

Ventura, S.J., Taffel, Due south.M. and Mosher, Due west.D. (

1988

) Estimates of pregnancies and pregnancy rates for the United States, 1976–85.

Am. J. Public Wellness

,

78

,

506

–511.

Warburton, D. and Fraser, F.C. (

1964

) Spontaneous abortion risks in homo: data from reproductive histories collected in a medical genetics unit.

Am. J. Hum. Genet.

,

16

,

one

–25.

Weinberg, C.R., Baird, D.D. and Wilcox, A.J. (

1994

) Bias in retrospective studies of spontaneous ballgame based on the outcome of the near recent pregnancy.

Ann. NY Acad. Sci.

,

709

,

280

–286.

Wilcox, A.J. and Horney, L.F. (

1984

) Accuracy of spontaneous abortion recall.

Am. J. Epidemiol.

,

120

,

727

–733.

Wilcox, A.J., Weinberg, C.R., O'Connor, J.F. et al. (

1988

) Incidence of early on loss of pregnancy.

Due north. Engl. J. Med.

,

319

,

189

–194.

Woolf, C.M. (

1965

) Stillbirths and parental historic period.

Obstet. Gynecol.

,

26

,

1

–8.

Yerushalmy, J. (

1939

) Age of father and survival of offspring.

Hum. Biol.

,

11

,

342

–356.

Zinaman, M.J., Clegg, E.D., Brown, C.C. et al. (

1996

) Estimates of human fertility and pregnancy loss.

Fertil. Steril.

,

65

,

503

–509.

© European Society of Human Reproduction and Embryology

© European Order of Human Reproduction and Embryology