Article Text

Download PDFPDF

Women’s age and total motile normal morphology sperm count predict fecundability: a prospective cohort study
  1. Mei Ting Lam1,2,
  2. Hang Wun Raymond Li1,2,3,
  3. Ching Yin Grace Wong1,3,
  4. William Shu Biu Yeung1,
  5. Pak Chung Ho1,
  6. Ernest Hung Yu Ng1,2
  1. 1 Department of Obstetrics and Gynaecology, The University of Hong Kong, Queen Mary Hospital, Pokfulam, Hong Kong
  2. 2 Department of Obstetrics and Gynaecology, Kwong Wah Hospital, Yau Ma Tei, Hong Kong
  3. 3 The Family Planning Association of Hong Kong, Wan Chai, Hong Kong
  1. Correspondence to Dr Mei Ting Lam, Department of Obstetrics and Gynaecology, Kwong Wah Hospital, Yau Ma Tei, Hong Kong; lmt256{at}


Objective This study investigated the role of women’s age, serum anti-Müllerian hormone (AMH) level and semen parameters in predicting fecundability.

Methods This was a prospective cohort study on couples attending for preconceptional health check. Occurrence of conception at 1 year after ceasing contraception and time to pregnancy were noted by telephone follow-up. The women’s age, serum AMH level and total motile normal morphology sperm count (TMNC) were compared between those who conceived and those who did not after 1 year; their independent predictive value on conception at 1 year was analysed by logistic regression. Among those conceiving within 1 year, Spearman’s correlations between time to pregnancy and the clinical parameters were studied.

Results Of the 100 couples analysed, we found younger age of the women (p=0.008), higher serum AMH level (p=0.038) and higher TMNC (p=0.015) in those that conceived within 1 year. Multivariate logistic regression found that women’s age (OR 0.867, 95% CI 0.761 to 0.988, p=0.032) and TMNC (OR 1.089, 95% 1.001–1.185, p=0.047), but not serum AMH level, significantly predicted conception within 1 year. Among those that conceived within 1 year, none of the parameters analysed were correlated with time to pregnancy within 1 year.

Conclusions Women’s age and TNMC are significant independent predictors of conception within 1 year. No parameter was shown to predict the time to pregnancy within 1 year. This finding can aid preconceptional counselling of couples who are planning for pregnancy.

  • infertility
  • counselling
  • family planning service provision
  • natural family planning

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Key messages

  • Women’s age and total motile normal morphology sperm count, but not serum anti-Müllerian hormone level, were significant independent predictors of conception within 1 year.

  • Lower high-density lipoprotein cholesterol levels in the female and male were found in couples that conceived within 1 year, but the magnitude of difference was very small.

  • In our cohort, couples conceiving within 1 year or those who did not showed no differences in men’s age, body mass index, other lipid parameters, physical activity and stress score.


Fecundability is the probability of a pregnancy, during a single menstrual cycle in a woman with adequate exposure to sperm and no contraception, culminating in a live birth.1 Time to pregnancy is the time taken to establish a pregnancy, measured in months or in numbers of menstrual cycles.1 There is a trend towards delaying childbearing, resulting in more women facing fertility problem at the time they hope to conceive.2 If the factors affecting fertility could be identified, it might be possible to offer counselling on family planning and possible early fertility intervention. Various factors have been studied for their influence or predictive value on fecundability, particularly those related to the production and competence of gametes, such as women’s age and serum anti-Müllerian hormone (AMH) level as well as semen parameters.

AMH is a polypeptide hormone solely secreted by granulosa cells of preantral and small antral ovarian follicles in the adult female. Serum AMH level demonstrates negligible fluctuation throughout the menstrual cycle,3 and hence represents a reliable marker of the ovarian reserve.4 As ovarian ageing is one important factor leading to reduced fecundability in women, studies have tried to explore the ability of AMH measurement to predict natural fecundability in the general population with contrasting results.5–10

Conversely, although semen analysis is a standard investigation for couples presenting with subfertility, and that the fifth edition of the WHO manual (2010) revised the reference range based on a more evidence-based approach,11 its role in predicting fecundability in the general population still remains poorly defined12 and it was not adopted in many studies looking into fecundability.5 8–10

Besides, the correlation between a number of health and lifestyle factors such as coital frequency, obesity, metabolic problems and physical activity with fecundability is also of interest. Obesity as well as higher non-fasting serum free cholesterol concentrations in both men and women are associated with reduced fecundability.13–15

Other lifestyle factors may also impact on fecundability. It was reported that physical activity has an influence on fecundability, in that physical activity of any type might improve fertility among overweight and obese women, whereas lean women who substitute vigorous exercise with moderate exercise may have improved fertility.16 Perceived stress was also reported to be associated with reduced fecundability in women.17

Apparently, fecundability is subject to influence by multiple factors in both the male and the female, and yet most of the existing studies examined individual factors without taking into consideration the partner’s factors. We carried out this prospective observational study to investigate the independent effects of age, serum AMH level and semen parameters, when controlled for each other, on fecundability in couples planning for conception. The effect of coital frequency, obesity, metabolic factors, stress level and physical activity were also studied. We hypothesised that these factors have a potential effect on fecundability.


Subjects and data collection

This was a prospective observational study with 1 year follow-up. Participants were recruited at the Pre-Pregnancy Check-Up Service of the Family Planning Association of Hong Kong, a primary sexual and reproductive healthcare institution. Recruitment was conducted between November 2015 and April 2017, with the last follow-up completed in March 2019. Study participants were couples who were planning to conceive, had stopped contraception for not more than 6 months or about to stop contraception, and the women were nulliparous and aged between 20 and 44 years. Exclusion criteria included history of infertility, having no sexual exposure or history of coital dysfunction, history of tuboperitoneal disease, pelvic inflammatory disease, ectopic pregnancy or endometriosis, anovulation or irregular menstrual cycles (cycle length <21 or >35 days), known endocrine disease, use of hormonal treatment which might affect ovarian or testicular functions within 3 months, use of injectable hormonal contraception within 6 months, men with azoospermia, and known medical or genetic diseases that the couple had which might affect fertility.

Written informed consent was sought from the participants. The research nurse conducted an interview with each couple to obtain demographic information, and medical and reproductive history using a standardised questionnaire. The English and Chinese versions of the International Physical Activity Questionnaire – Short Form (IPAQ-SF)18 19 were used to assess physical activity for all participants. Activity was expressed as metabolic equivalent task (MET) minutes per week. Stress level was measured by the Perceived Stress Scale score (PSS-10).20 Anthropometric measurements of the couple including body height and weight were recorded. The serum AMH level of the women was measured by the Access AMH assay (Beckman-Coulter Inc., Marseille, France).21 Glycated haemoglobin (HbA1c) and lipid profile were measured for the couple. Semen analysis was done according to the WHO manual (fifth edition). If there was an abnormal parameter(s), another semen analysis was repeated 3 months later. The sample was collected after sexual abstinence for 2–7 days. The mean value was taken if the first one was abnormal and semen analysis was repeated.

Telephone follow-up was conducted at 6 and 12 months after recruitment to obtain information on contraception, coital and pregnancy history. For couples who had regular unprotected intercourse for less than 12 months, telephone follow-up was repeated after trying for 12 months. Couples who did not have regular unprotected intercourse 12 months after recruitment were withdrawn from the study.

Outcome measures

The primary outcome measures were conception at 1 year and time to pregnancy in those who conceived. Conception was defined as a positive urine pregnancy test as reported by patients on telephone follow-up. Time to pregnancy was counted from the time when the couple started to have regular unprotected intercourse without contraception.


A minimum sample size of 81 would be adequate to determine a statistical significance between a receiver operating characteristic (ROC) curve with area under the curve of 0.75 and one with the null hypothesis value of 0.5, with power of 80% and type I error of 0.05. Allowing for a dropout rate of 30%, which would be excluded from data analysis, we intended to recruit 120 couples. This would also fulfil the recommended minimum sample size of 100 for estimating the predictive value of individual factors on infertility in a univariate logistic regression model.

Continuous variables, expressed as median (25th–75th percentile), were compared between those who conceived and those who did not after ceasing contraception for 1 year by Mann–Whitney U-test. Fisher’s exact test was used to compare categorical variables between groups. Univariate binary logistic regression was used to study the predictive value of those individual parameters on conception at 1 year. Age and serum AMH level of the women, total motile normal morphology sperm count (TMNC) and coital frequency were entered into a standard multivariate binary logistic regression model to study their predictive roles after controlling for each other. ROC curve analysis was applied to determine the predictive value on conception at 1 year. Among those who conceived within 1 year, correlation between clinical parameters and time to pregnancy was determined by Spearman’s correlation. Statistical analysis was performed using the IBM SPSS version 25.0 (IBM Corp., Armonk, NY, USA) and MedCalc Statistical Software version 15 (MedCalc Software bvba, Ostend, Belgium). A two-tailed value of p<0.05 was considered statistically significant.

Patient and public involvement

This was solely an investigator-initiated study. There was no patient or public involvement in the design or execution of this study.


A total of 112 couples were recruited, of which 100 completed the study and were eligible for analysis. Among the other 12 couples, five were lost to follow-up, five did not attempt conception by the end of the study period, one divorced and one had assisted reproductive treatment in another clinic before the end-of-study follow-up. For the 100 couples, all women were Chinese, 99 men were Chinese and one man was Caucasian. Two women were using hormonal contraception as their last method of contraception, one was using an intrauterine device, whereas the rest were using barrier or natural contraception. Characteristics of the participants are shown in table 1.

Table 1

Comparison of study participants' basic demographic and clinical parameters between those couples that conceived within 1 year and those that did not

Of the 100 analysed couples, 61 got pregnant within 1 year, with median time to pregnancy of 2 (1–5) months; 39 did not get pregnant in 1 year and fulfilled the definition of infertility. Those who conceived within 1 year, compared with those who did not, had significantly younger age and higher serum AMH level of the women, as well as total motile sperm count and TMNC (table 1).

Couples who conceived within 1 year also had lower HDL-cholesterol in both the female and male partner, although the magnitude of difference was very small. Other clinical parameters including coital frequency, age of men, body mass index (BMI), HbA1c, triglycerides, total cholesterol, low-density lipoprotein (LDL) cholesterol, PSS-10 score and physical activity (MET/week in the IPAQ-SF) of the men and women were not statistically different between the two groups (p>0.05). The results are shown in table 1.

Univariate logistic regression of factors predicting conception within 1 year by univariate analysis found that age, AMH and high-density lipoprotein (HDL) cholesterol level of the women, as well as TMNC, were significant factors predicting conception within 1 year (table 2). When the age and serum AMH level of the women, and TMNC were entered for multivariate analysis, the female’s age and TMNC, but not serum AMH level of the women, were significant independent factors predicting conception within 1 year (table 3).

Table 2

Univariate logistic regression analysis of factors predicting conception within 1 year

Table 3

Predicting conception within 1 year by multivariate logistic regression analysis where the age and serum anti-Müllerian hormone level of the women and total motile normal morphology count of the men’s semen were entered

Table 4 shows the correlation between time to pregnancy and clinical parameters in the women and men using Spearman’s correlation. No parameter was significantly correlated with the time to pregnancy within 1 year. There was no significant difference in time to pregnancy (p=0.643) between those having sexual intercourse twice per week or more and less than twice per week. Cox proportional hazards regression analysis revealed that women’s age (p=0.012), but not serum AMH nor TMNC, significantly contributed to prediction of time to pregnancy (online supplementary table 1). The survival curve showing occurrence of pregnancy with time is shown in online supplementary figure 1.

Supplemental material

Table 4

Spearman’s correlation between time to pregnancy and the clinical parameters in the participants who conceived in 1 year (n=61)

The areas under the ROC curve for women’s age and TMNC in semen in predicting conception within 1 year were 0.66 (95% CI 0.55 to 0.77) and 0.65 (95% CI 0.54 to 0.75), respectively, and that of the multivariate logistic regression model combing the two factors was 0.70 (95% CI 0.60 to 0.79).


Findings from the present study suggested that the women’s age and TMNC, but not serum AMH level, were significant factors predicting conception after trying for 1 year. A previous prospective observational study reported that 82% of couples conceived during 12 cycles.22 Another prospective study among those practising timed intercourse reported that 89.6% of women conceived in 1 year.23 However, an internet-based prospective study from Denmark reported a pregnancy rate of 69.6% only at 12 months.24 In our study, however, only 61% of couples conceived in 1 year. The apparently lower figure could be due to the relatively low coital frequency in our population, even among pregnancy planners. Less than 50% of our cohort had sexual intercourse at least two times per week. In a study of Hong Kong women attending family planning and pre-pregnancy check-up services in 2007–2009, 18.1% had intercourse less than 12 times in a year.25

Increasing women’s age was demonstrated to be a good predictor of reduced fecundability, while men’s age was not associated with fecundability after adjustment for the women’s age.26 In the 501 couples of the Longitudinal Investigation of Fertility and the Environment (LIFE) study, men’s age was associated with fecundability.27 However, the women’s age was not controlled for in that study, and it could be that older men merely had older female partners. Our study suggested that older women, but not men, were associated with reduced fecundability. Reproductive ageing in men generally occurs at the more extreme end of the age range.28 Given that majority of our study participants were in the young range (all men were aged ≤50 years) it is not unexpected that men’s age was not observed to have an independent influence on fecundability.

Semen parameters have been reported to be associated with time to pregnancy.27 29 In line with that, reduced TMNC was found to be significantly associated with reduced fecundability in our study. Obviously, this determines the number of sperms available at the site of fertilisation after the sexual act. However, this is not the sole determinant as other factors such as coital frequency would be important also.

It is worthwhile to note that fecundability is subjected to multifactorial influence. The predictive value of either the women’s age or TMNC alone or in combination was just modest, as shown by the c-statistics in the ROC curves, despite the statistical significance of the individual factors.

Steiner et al reported that AMH was associated with day-specific probability of pregnancy.5 A study of 750 women from the same group found that low AMH was not associated with reduced fecundability.9 Other studies involving younger healthy pregnancy planners found that AMH did not predict natural fertility.6 7 10 An observational study of 87 women who had a live birth from planned natural pregnancy found that AMH was not related to the time to pregnancy.8 The women in our study were younger (only 17% were aged 35+ years) and our results concurred with those studies involving younger women.

Cholesterol is a substrate for steroidogenesis and hence may impact on fertility .30 The LIFE study found that higher free cholesterol was associated with reduced fecundability independent of BMI.15 A secondary analysis in the Effects of Aspirin in Gestation and Reproduction trial in 2017 also found that reduced fecundability was associated with all lipoprotein abnormalities including low HDL-cholesterol.31 Our results revealed that lower HDL-cholesterol was significantly associated with conception at 1 year; however, the magnitude of difference was extremely small, and is unlikely to be clinically meaningful. We also studied the effect of HbA1c, a glycaemic index, on fecundability, which has not been reported in the literature. No significant effect of HbA1c on fecundability was suggested in our cohort. This could be limited by the fact that no one in our cohort had abnormal HbA1c at or above 6.0%.

A previous study reported that a total MET of 1200–1740 min/week was associated with highest fecunability.16 Although couples who conceived in 1 year in our cohort had similar MET results, we could not demonstrate a significant effect of physical activity on fecundability, probably because the majority of the individuals in our cohort were not doing vigorous physical activity.

The main strength of our study is that it was a prospective study with 1 year follow-up and both the male and female partners of the couples were closely followed up so as to avoid recall bias. The subjects were recruited among pregnancy planners in the community, in contrast to some previous studies which targeted couples attending infertility clinics. However, our sample size was only modest, and it was just adequate to confer adequate power for our main analyses. A larger further study with adequate power to establish and validate a prediction model by the women’s age and TMNC would be worthwhile. Another limitation was that semen parameters and lifestyle factors such as smoking, exercise and stress level may fluctuate or change with time. We only assessed these parameters at recruitment, which may not represent the whole course of the follow-up period. The fact that the majority of the couples were non-smokers and had normal BMIs may make the comparison of these parameters difficult and limit its predicting power, and our findings cannot be extrapolated to couples with more extreme BMIs.


In conclusion, women’s age and TMNC are significant independent predictors of conception within 1 year. No parameter can predict the time to pregnancy within 1 year. This finding can aid preconceptional counselling of couples who are ceasing contraception in preparation for pregnancy.


The authors would like to thank Ms Celia Ng and Ms Wylie Wong for co-ordinating subject recruitment and follow-up. They also want to express their gratitude to all staff members of the Family Planning Association of Hong Kong who facilitated the conduct of this study, and to Ms Benency Po-Chau Wong and Dr Angel Chan for performing the biochemical assays. Lastly, they thank all the couples who participated in this study.



  • Contributors HWRL, WSBY, PCHO and EHYN conceived and designed the study. MTL, HWRL, GCYW and EHYN supervised subject recruitment and follow-up. MTL analysed the data, interpreted the results and drafted the manuscript. All authors provided critical revisions and approved the final manuscript prior to submission.

  • Funding This study was supported by internal research funding from the Department of Obstetrics and Gynaecology, The University of Hong Kong.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Patient consent for publication Not required.

  • Ethics approval Ethics approval was obtained from both the Institutional Review Board of the University of Hong Kong and the Hospital Authority Hong Kong West Cluster. Approval was also obtained from the Health Services Subcommittee, Family Planning Association of Hong Kong.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available upon reasonable request.

Linked Articles