Occupational factors and markers of testicular function among men attending a fertility center

Abstract

STUDY QUESTION

Are occupational factors associated with markers of testicular function among men attending a fertility center?

SUMMARY ANSWER

Men working non-daytime/rotating shifts and those with physically demanding jobs have higher sperm concentration and total sperm count as well as higher estradiol and total testosterone concentrations.

WHAT IS KNOWN ALREADY

Semen quality has declined during recent decades and has been negatively correlated with higher risks of common chronic diseases and mortality, highlighting its public health importance beyond fertility and reproduction. While most of the previous epidemiology literature on male fertility has focused on environmental exposures, dietary factors, and other related variables, little attention has been paid to occupational factors.

STUDY DESIGN, SIZE, DURATION

This observational study included 377 men who were male partners in couples seeking infertility treatment at a fertility center, who enrolled in the Environment and Reproductive Health (EARTH) study between 2005 and 2019.

PARTICIPANTS/MATERIALS, SETTING, METHODS

Self-reported information on lifting/moving heavy objects, typical shift, and physical level of exertion at work was collected from a take-home questionnaire. Semen samples were analyzed following World Health Organization guidelines. Enzyme immunoassays were used to assess reproductive hormone concentrations. Linear regression models were used to evaluate the association between occupational factors and measures of testicular function, while adjusting for covariates such as age, BMI, education, race, smoking, and abstinence time, and accounting for multiple semen samples (mean = 2, min–max = 1–9) in analyses for semen parameters.

MAIN RESULTS AND THE ROLE OF CHANCE

Men had a median (interquartile range) age of 36 (33, 39) years and were predominantly Caucasian (87%). Of the men who completed the survey, 12% reported often lifting or moving heavy objects at work, 6% reported heavy physical exertion at work, and 9% reported evening or rotating shifts. Men who reported often lifting or moving heavy objects at work had 46% higher sperm concentrations (P = 0.01) and 44% higher total counts (P = 0.01) compared with men who reported never lifting or moving heavy objects at work. Similar results were found for men working in rotating shifts compared to those in day shifts, as well as for men involved in heavy levels of physical exertion compared to those with light levels at work. We also found that men involved in heavy/moderate levels of physical exertion at work had higher circulating testosterone concentrations compared to those with lighter exertion (adjusted means of 515 and 427 ng/dl, respectively, P = 0.08), and men who often moved/lifted heavy objects at work had higher estradiol concentrations, compared to those who never did (adjusted means of 36.8 and 27.1 pg/ml, respectively, P = 0.07). Men working evening/rotating shifts had 24% higher testosterone (P = 0.04) and 45% higher estradiol concentrations (P = 0.01), compared to men working day shifts. No associations were observed for ejaculated volume, total motility, morphologically normal sperm, or serum FSH and LH concentrations.

LIMITATIONS, REASONS FOR CAUTION

Due to our study design which recruited men from couples seeking fertility treatment, it may not be possible to generalize our findings to men from the general population. Also, as is the case of all studies based on self-reported questionnaires, measurement error and misclassification of the exposure are potential concerns.

WIDER IMPLICATIONS OF THE FINDINGS

Physically demanding jobs and rotating or evening shift occupations may be associated with higher testicular function in men measured as higher sperm concentrations and counts as well as higher serum testosterone and estradiol levels. Confirmation of these findings in other non-fertility clinic study populations is warranted.

STUDY FUNDING/COMPETING INTEREST(S)

NIH grants R01ES022955, R01ES009718, R01ES033651, and R01ES000002 from the National Institute of Environmental Health Sciences (NIEHS) and Legacy, Inc. R.A.G. works part time for Legacy, Inc., which provided funds to perform this analysis. There are no other conflicts of interest.

TRIAL REGISTRATION NUMBER

N/A.

Introduction

Infertility is estimated to affect about 10–15% of all couples worldwide (Mascarenhas et al., 2012; Inhorn and Patrizio, 2015), and male factor infertility, as defined according to the World Health Organization (WHO) reference values for semen quality, is the most prevalent single cause, accounting for 40% of infertility cases (Légaré et al., 2014). In addition, semen quality, which has declined during the last decades (Levine et al., 2017; Minguez-Alarcon et al., 2018), has been also negatively associated with higher risks of common chronic diseases and mortality, highlighting the public health importance of semen quality beyond fertility and reproduction (Jensen et al., 2009; Eisenberg et al., 2014, 2016; Latif et al., 2017; Choy and Eisenberg, 2018). Therefore, greater attention to modifiable factors that are related to human fertility is warranted.

While most of the previous epidemiology literature on male fertility has focused on environmental chemicals (Virtanen et al., 2017; Dziewirska et al., 2018; Zamkowska et al., 2018; Knapke et al., 2022; Mínguez-Alarcón et al., 2022), dietary factors (Salas-Huetos et al., 2017; Efrat et al., 2018; Nassan et al., 2018), and other related variables such as BMI, waist circumference, and recreational physical activity (Chavarro et al., 2010; Minguez-Alarcon et al., 2014; Bian et al., 2022), little research has been conducted on occupational factors (Sheiner et al., 2002; El-Helaly et al., 2010; Eisenberg et al., 2015). In this analysis, we investigated whether self-reported occupational factors such as moving heavy objects, type of work shifts, and level of physical exertion at work were associated with repeated measures of semen quality among men seeking fertility care in Boston, MA, USA. As a secondary analysis, we also evaluated whether these occupational exposures were related to circulating serum levels of reproductive hormones in a subset of these men.

Materials and methods

Study population

Study participants were male partners of couples enrolled in the Environment and Reproductive Health (EARTH) study, a prospective cohort study of couples seeking infertility treatment at the Massachusetts General Hospital (MGH) Fertility Center, aimed at evaluating environmental and dietary determinants of fertility (Minguez-Alarcon et al., 2018). Men between the ages of 18–56 years and without a history of vasectomy were eligible to participate, and ∼40% of those contacted by the research nurses agreed to participate and were enrolled. After the study procedures were explained, participants signed an informed consent form. Between 2005 and 2019, 511 men enrolled in the EARTH and provided at least one semen sample and 401 (78%) completed a questionnaire containing information on occupational factors. Of these, 377 (94%) men who self-reported being actively working and who contributed 950 semen samples (mean = 2, range = 1–9) were included in this analysis. Of these men, a subset of 145 provided a blood sample for measurement of certain reproductive hormones in serum. The participant’s date of birth was collected at entry, and weight and height were measured by trained study staff. BMI was calculated as weight (in kilograms) divided by height (in meters) squared. The participants completed a study staff-administered questionnaire that contained additional questions on lifestyle factors (including use of marijuana and cocaine as well as chemical exposures, e.g. aluminum, solvents, cadmium, diesel, pesticides, in the workplace), reproductive health, and medical history. Time spent in leisure physical and sedentary activities was assessed using a validated questionnaire (Wolf et al., 1994). Starting in 2007, the diet was assessed using an extensively validated food frequency questionnaire and two data-derived dietary patterns, the ‘Prudent’ and the ‘Western’ diet, were calculated based on reported food intakes using principal component analysis (Gaskins et al., 2012). The study was approved by the Human Subject Committees of the Harvard T.H. Chan School of Public Health and the MGH, and informed consent was obtained from all participants.

Occupational factors

Information on work schedule and physically demanding work was collected on a take-home questionnaire. Men reported how often they lifted or moved heavy objects in their current job with response options of ‘never’, ‘sometimes’, or ‘often’. Men also reported the level of physical exertion in their current job using the following categories: light (e.g. most time spent sitting, office work), moderate (e.g. lifting/pushing light loads, long periods of walking), and heavy (e.g. lifting, pushing heavy loads, heavy manual labor). To assess the associations with shift timing, men reported whether their typical work shift was day, evening, night, or rotating. The EARTH survey which contains the occupational questions (page 17: questions 10c, 10d, and 10e) can be found as Supplementary File S1.

Semen assessment

Semen was collected on site at the MGH via masturbation in a sterile plastic specimen cup following a recommended 48-h abstinence period of 2–5 days. Some men provided multiple semen samples for further evaluation of their semen quality and/or because their female partners underwent more than one cycle of fertility treatment. Men reported the duration of abstinence before providing the sample. Before analysis, the sample was liquefied at 37°C for 20 min after collection and analyses were conducted 10–20 min afterwards. Semen volume (ml) was measured by an andrologist using a graduated serological pipet. Sperm concentration (million/ml) and motility (% motile) were assessed using a computer-aided semen analyzer (10HTM-IVOS, Hamilton-Thorne Research, Beverly, MA, USA). To measure sperm concentration and motility, 5 ml of semen from each sample was placed into a pre-warmed (37°C) Makler counting chamber (Sefi-Medical Instruments, Haifa, Israel). A minimum of 200 sperm cells from at least four different fields were analyzed from each specimen. Motile sperm were defined according to the WHO classification (World Health Organization, 2010). Total sperm count (million/ejaculate) was calculated by multiplying sperm concentration by semen volume. Sperm morphology (% normal) was measured on two slides per specimen (with a minimum of 200 cells assessed per slide) with a microscope using an oil-immersion 100× objective (Nikon, Tokyo, Japan). Strict Kruger scoring criteria were used to classify men as having normal or below normal morphology (Kruger et al., 1988). Andrologists were blinded to the participants’ occupational information. For quality assurance/quality control purposes, the laboratory conducted weekly monitoring of sperm morphology smears. In addition, the laboratory performed quarterly competence evaluations of all technicians and proficiency testing by an outside evaluator occurred every 6 months.

Reproductive hormone assessment

In a subset of men, a non-fasting blood sample was collected. After blood samples were clotted, tubes were centrifuged at ∼1100×g for 20 min. The resulting serum samples were then aliquoted, frozen, and stored at −80°C until transfer to the Clinical and Epidemiologic (CER) Laboratory at the Boston Children’s Hospital (Boston, MA, USA) for analysis using FDA-approved assays run on a clinical analyzer. Specifically, serum concentrations of FSH, LH, total testosterone, and estradiol were measured using liquid chromatography–tandem mass spectrometry. The assay sensitivities for LH and FSH were 1.2 and 1.1 IU/l, respectively. The intra-assay coefficients of variation (CVs) for LH and FSH were <5% and <3%, respectively, and the inter-assay CVs for both hormones were <9%. The assay sensitivity for estradiol was 20 pg/ml, the within-run CV was between 3% and 11% and the total CV was between 5% and 15%. Total testosterone measurements had inter-assay and intra-assay CVs of 12% and 10%, respectively, and a sensitivity of 4 ng/dl (0.139 nmol/l). Laboratory technicians were blinded to both occupational factors and semen parameters of the study participants. Each hormone was used as a continuous outcome and we also calculated total testosterone/LH ratio as an additional outcome.

Statistical analysis

Demographic, baseline reproductive characteristics, and occupational factors of the men were presented using median (interquartile ranges (IQRs)) or percentages. Distribution of semen parameters and reproductive hormone concentrations were reported using median (IQR) and mean (SD). Kolmogorov–Smirnov tests were used to evaluate whether the testicular function outcomes were normally distributed. Occupational exposures were categorized into groups based on the questionnaire responses, although for hormone analyses, categories were collapsed to two levels due to small sample sizes. Multivariable linear mixed models with random subject effects to account for repeated semen measurements within the same man were used to estimate the association of occupational factors with continuous semen parameters. Multivariable linear models were used to estimate the association of occupational factors with continuous serum reproductive hormones. To allow for better interpretation of the results, population marginal means (Searle et al., 1980) are presented after adjusted for all the covariates in the model (at the mean level for continuous variables and at a value weighted according to their frequencies for categorical variables).

Confounding was assessed using prior knowledge on biological relevance and descriptive statistics from our study population through the use of directed acyclic graphs (Weng et al., 2009) (Supplementary Fig. S1). The variables considered as potential confounders included factors previously related to semen parameters (Sharma et al., 2013; Rooney and Domar, 2014), and factors associated with occupational factors and semen parameters in this study. Final models were adjusted for age (years), BMI (kg/m²), education (colleague degree versus other), race (white versus other), smoking status (ever and never smoked) and, for the analyses of semen parameters, abstinence time (days). Two sensitivity analyses (one for semen parameters and another for reproductive hormones) were conducted further adjusting for physical activity, use of marijuana and/or cocaine, chemical exposures in the workplace, and assessment of dietary patterns (Prudent and Western) among the subset of men with data on these variables, to address the concern of residual confounding by other potential confounders. Statistical analyses were performed with SAS (version 9.4; SAS Institute Inc., Cary, NC, USA).

Results

The 377 men with semen samples included in our analysis had a median (IQR) age of 36 (33, 39) years and were predominantly Caucasian (87%) (Table I). The median (IQR) BMI was 27 (24, 30) kg/m² and only 5% were current smokers, while 5% and 6% of men had a history of cryptorchidism or varicocele, respectively. Among the men, 12% reported often lifting/moving heavy objects at work, 6% reported heavy physical exertion at work, and 9% worked evening, night, or rotating shifts. Similar demographic characteristics, reproductive conditions, and occupational factors were found in the subset with reproductive hormones, except for varicocele, which was more prevalent among the subset of men with available hormone measurements (11% versus 6%). Correlations between the three examined occupational factors were moderate and ranged from 0.32 to 0.70.

Table I.

Demographic, reproductive characteristics, and occupational factors [median (IQR) or N (%)] of men in Environment and Reproductive Health (EARTH) study.

	N = 377 with semen analysis	N = 145 with hormones
Demographics
Age, years	35.8 (32.6, 39.2)	35.1 (32.2, 38.5)
Race, N (%)
White	329 (87)	123 (85)
Black	10 (3)	4 (3)
Asian	23 (6)	11 (8)
Other	15 (4)	7 (5)
BMI, kg/m2	26.8 (24.2, 29.5)	26.9 (24.0, 29.8)
Education, N (%)
High school/some college	55 (15)	17 (12)
College graduate	129 (34)	51 (35)
Graduate degree	193 (51)	77 (53)
Current smoker, N (%)	17 (5)	5 (4)
Reproductive history, N (%)
Undescended testes	17 (5)	14 (9)
Varicocele	24 (6)	16 (11)
Epididymitis	5 (1)	2 (1)
Prostatitis	9 (2)	5 (3)
Occupational factors, N (%)
Moving/lifting heavy objects
Never	199 (53)	85 (58)
Sometimes	132 (35)	47 (32)
Often	46 (12)	13 (9)
Typical work shift
Day	342 (91)	131 (90)
Evening	5 (1)	2 (1)
Rotating	30 (8)	12 (8)
Level of physical exertion
Light	286 (76)	109 (75)
Moderate	69 (18)	28 (19)
Heavy	22 (6)	8 (6)

The median (IQRs) sperm concentration, total sperm count, total testosterone, and estradiol concentrations were 55.2 (26.4, 98.2) million/ml, 134 (65.0, 243) million/ejaculate, 398 (324, 507) ng/dl, and 27.0 (22.0, 35.0) pg/ml, respectively (Table II). Most semen samples were above the WHO reference values for good semen quality (World Health Organization, 2010). Samples were collected, on average (standard deviation, min, max), 144 (267, −338, 1052) days after the men completed the occupational factor information in the questionnaire.

Table II.

Semen parameters and reproductive hormone levels of men in Environment and Reproductive Health (EARTH) study.

	N	Median (IQR)	Mean (SD)
Semen parameters
Ejaculated volume, ml	946	2.7 (1.9, 3.7)	2.8 (1.3)
Sperm concentration, million/ml	950	55.2 (26.4, 98.2)	70.0 (63.7)
Total sperm count, million/ejaculate	946	134 (65.0, 243)	179 (159)
Total motility, %	950	47.0 (27.0, 64.0)	45.2 (22.8)
Normal morphology, %	836	6.0 (4.0, 8.0)	6.3 (3.3)
Abstinence time, days	950	2.1 (1.8)	2.7 (3.6)
Reproductive hormones
FSH, IU/l	141	4.3 (2.9, 6.5)	5.8 (4.9)
LH, IU/l	113	4.5 (3.2, 6.0)	4.9 (2.9)
Total testosterone, ng/dl	145	398 (324, 507)	445 (250)
Estradiol, pg/ml	73	27.0 (22.0, 35.0)	31.9 (23.0)

Several self-reported occupational factors were associated with sperm concentration and total count after adjusting for age, BMI, education, race, smoking, and abstinence time (Table III). For example, men who reported often lifting or moving heavy objects at work had 46% higher sperm concentration (P = 0.01) and 44% higher total sperm count (P = 0.01) compared with men who reported never lifting or moving heavy objects at work. Similar results in sperm concentration and total sperm count were found for men working in rotating shifts compared to those in day shifts, as well as for men with a heavy level of physical exertion compared to those with light level at work. The frequency of moving heavy objects, typical work schedule, and level of physical exertion at work was not associated with ejaculated semen volume, total motility, or morphologically normal sperm (Table III).

Table III.

Adjusted^a semen quality parameters by self-reported occupational factors among 377 men contributing 950 semen samples in Environment and Reproductive Health (EARTH) study.

	Ejaculated volume, ml	Sperm concentration, million/ml	Total sperm count, million/ejaculate	Total motility, %	Normal morphology, %
Moving/lifting heavy objects
Never	2.87 (2.70, 3.04)	32.2 (23.1, 44.9)	82.7 (59.3, 115)	44.5 (41.4, 47.6)	6.32 (5.89, 6.74)
Sometimes	2.83 (2.61, 3.04)	40.4 (31.6, 51.6)	100.0 (78.6, 128)	43.1 (39.3, 47.0)	6.23 (5.69, 6.77)
Often	2.88 (2.51, 3.24)	59.6 (45.7, 77.9)*	149.0 (115, 194)*	49.1 (43.3, 55.0)	5.98 (4.93, 7.02)
Typical work shift
Day	2.87 (2.74, 3.00)	36.4 (29.4, 45.1)	92.6 (74.7, 115)	44.6 (42.2, 46.9)	6.23 (5.89, 6.57)
Evening	2.95 (2.13, 3.76)	41.2 (21.3, 79.4)	111.0 (44.7, 274)	42.2 (25.1, 59.4)	5.10 (2.95, 7.27)
Rotating	2.72 (2.26, 3.17)	53.3 (39.5, 72.0)*	126.0 (92.0, 172)	45.8 (39.6, 52.0)	6.62 (5.53, 7.71)
Level of physical exertion
Light	2.87 (2.73, 3.01)	34.3 (26.6, 44.1)	87.4 (67.9, 113)	43.6 (41.0, 46.2)	6.27 (5.91, 6.62)
Moderate	2.84 (2.56, 3.14)	45.7 (34.2, 61.1)	114.0 (85.8, 151)	47.4 (42.5, 52.3)	6.25 (5.42, 7.10)
Heavy	2.70 (2.19, 3.21)	68.1 (45.8, 44.1)*	161.0 (110, 235)*	49.2 (40.7, 57.7)	5.90 (4.56, 7.25)

^aData are presented as predicted marginal means (95% CI) unless otherwise noted, adjusted for age, BMI, education, race, smoking, and abstinence days.

^* P-value <0.05 when comparing that group to the reference group (never for moving heavy objects, day for typical work shift, and light for level of physical exertion).

Among the subset of 145 men with hormone measurements available, the typical work shift was related to serum concentrations of both total testosterone and estradiol (Fig. 1). Specifically, men usually working evening or rotating shifts had 24% higher testosterone (P = 0.04) and 45% higher estradiol concentrations (P = 0.01), compared to men working day shifts. We also found that men involved in heavy/moderate level of physical exertion at work had higher testosterone concentrations compared to those involved in light level (adjusted means of 515 and 427 ng/dl, respectively, P = 0.08). In addition, men who typically moved or lifted heavy objects at work had higher estradiol concentrations, compared to those who never did (adjusted means of 36.8 and 27.1 pg/ml, respectively, P = 0.07). No associations between these occupational factors and serum FSH and LH concentrations were observed (Fig. 1). Compared with men working day shifts, those who typically worked evening or rotating shifts had significantly higher total testosterone/LH ratios (adjusted means (95% CI) = 16.1 (6.26, 26.0) versus 1.21 (−2.35, 4.78)) (Supplementary Table SI). No associations were found for the other occupational factors we examined.

Figure 1.

Adjusted circulating blood reproductive hormones by self-reported occupational factors among 145 men in the Environment and Reproductive Health (EARTH) study. Data are presented as predicted marginal means (95% CI) unless otherwise noted and were adjusted for age, BMI, education, race, and smoking.

Lastly, we evaluated occupational factors in relation to both semen quality (Supplementary Table SII) and reproductive hormones (Supplementary Table SIII) while further adjusting for physical activity, use of marihuana and/or cocaine, chemical exposures in the workplace, and assessment of dietary patterns (Prudent and Western) among a subset of men (N = 294 and N = 113) to address the concern of residual confounding by other potential confounders. While some of the associations with semen parameters became slightly attenuated, those for reproductive hormones became stronger. These minor changes may be due, in part, to the decrease of ∼20% in the number of participants included in these analyses.

Discussion

In this observational analysis of men attending a fertility center and providing multiple semen samples and reproductive hormones in a subset, we investigated the relationship between self-reported occupational factors and several markers of testicular function. We observed that men working non-daytime/rotating shifts and those with physically demanding jobs had higher sperm concentrations and total sperm counts as well as estradiol and total testosterone concentrations, compared to men working daytime shifts and those who reported never lifting/moving heavy objects or with jobs that involved only a light level physical exertion. No associations were found for semen volume, sperm motility, or morphology or serum concentrations of FSH and LH. These findings suggest that certain occupational factors may be associated with testicular function in men and add to the existing epidemiology literature on modifiable lifestyle factors of male fertility.

In contrast with our results, work-related heavy exertion was related to lower sperm concentration and total sperm count, and shift work was not associated with semen quality among 456 men from the general population who collected one semen sample at home and participated in the Longitudinal Investigation of Fertility and the Environment (LIFE) study between 2005 and 2009 (Eisenberg et al., 2015). The major difference between the two studies is that men in the LIFE study had higher median sperm concentrations and total sperm counts than men in the EARTH study (62 versus 55 million/ml and 195 versus 134 million/ejaculate, respectively). In addition, men in the LIFE study were younger overall but also heavier than men in our study. Among 1818 couples also from the US general population who participated in the Pregnancy Study Online (PRESTO), the authors found no associations between men’s work schedules and couple’s fecundability measured as female partner’s time to conception (McKinnon et al., 2022). Men participating in PRESTO were also overall younger and heavier than men in the EARTH study. Further studies on male occupational factors in relation to semen quality as well as couples’ reproductive outcomes are clearly needed.

Men’s leisure and professional physical activity in relation to circulating reproductive hormones, including testosterone and estradiol, has been extensively investigated (Wood and Stanton, 2012; Cano Sokoloff et al., 2016). However, the epidemiologic literature on exercise at work with hormones is still scarce (Theorell et al., 1990). In a pilot study among 44 men, the authors found that plasma total testosterone was higher among men with sedentary and not physically demanding work (Theorell et al., 1990). In our EARTH study, we found that men with physically demanding jobs have higher circulating serum testosterone. Testosterone hormone regulates important physiological processes such as muscle protein metabolism, sexual and cognitive functions, secondary sex characteristics, erythropoiesis, plasma lipids, and bone metabolism (Nieschlag and Nieschlag, 2014). We also observed that occupational factors were associated with higher circulating serum concentrations of estradiol among men in the EARTH study. We hypothesize that an excess of testosterone is being converted into estrogen by aromatase enzyme (expressed in several tissues including the testis) in a process called aromatization, to maintain normal levels of both hormones (Simpson and Davis, 2001; Carreau et al., 2003). Thus, the higher estradiol results may be explained by the association of work schedules and physical demanding jobs with higher serum testosterone, rather than a positive relationship between the examined occupational factors and serum estradiol per se. In fact, the men in the current study have serum testosterone and estradiol at normal concentrations as for healthy adult men. Future studies to test this hypothesis are warranted. Of note, free testosterone levels were significantly associated with lower calcaneal speed of sound, fat-free mass, and handgrip strength among 335 Malaysian Chinese and Malay men aged 40 years and older (Chin et al., 2012). Also, both total and free testosterone levels were related to muscle mass, strength, and physical performance among 1489 older (>64 years) men (Auyeung et al., 2011).

The current study has several limitations worth noting. First, due to the study design including male partners in couples seeking fertility treatment, it may not be possible to generalize our findings to men from the general population. However, the men in the EARTH study are considered to have good semen quality (World Health Organization, 2010), with comparable semen parameters to those among young healthy men from the general US population (Mendiola et al., 2014) and Europe (Mendiola et al., 2013). Also, selection bias is a concern as we included only men who were actively working at the time of the exposure assessment, since it is possible that men who were less healthy would not be working and thus not responding to the occupation factor question. Second, as is the case in all studies based on self-reported questionnaires, measurement error and misclassification of the exposure are a concern. Related to this point, our questionnaire did not collect information on changing jobs over the study period and this also represents a limitation of the current analysis. Third, only a few participants had data on other reproductive hormones such as sex hormone-binding globulin (SHBG) and thus SHBG was not considered in this analysis. Also, our study includes total testosterone and not just free testosterone, which actually represents the active form of the hormone. The strengths of this study include the collection of multiple semen samples for most of the study participants, assessment of not only semen parameters but also other markers of testicular function such as reproductive hormones, and comprehensive adjustment of possible confounding variables due to the standardized assessment of a wide range of participant characteristics.

In conclusion, we found that work schedules and physically demanding jobs are associated with improvements in several markers of testicular function among men attending a fertility center. Further studies are needed to confirm these results in other study populations and to elucidate potential mechanisms that explain the associations.

Data availability

The data underlying this article cannot be shared publicly due to participant confidentiality and privacy concerns.

Authors’ roles

J.E.C. and R.H. were involved in the study concept and design and critical revision of the article for important intellectual content. P.L.W. was involved in the study concept and design and critical revision of the article for important intellectual content, and provided statistical expertise. L.M.-A. analyzed the data, drafted the article, and had primary responsibility for the final content. L.M.-A., P.L.W., R.A.G., R.H., and J.E.C. interpreted the data. J.B.F. and I.S. were involved in acquisition of the data. All authors approved the final article.

Funding

Funding was provided by NIH grants R01ES022955, R01ES009718, R01ES033651, and R01ES000002 from the National Institute of Environmental Health Sciences (NIEHS) and Legacy, Inc.

Conflict of interest

R.A.G. works part time for Legacy, Inc. There are no other conflicts of interest.