Plain English Summary

The time prior to conception, preconception, is widely acknowledged as an integral period whereby a woman’s health, lifestyle, and diet influence the outcomes of future pregnancy and the health of future offspring. Similarly, the influence of a man’s health, lifestyle, and diet during preconception on pregnancy and offspring outcomes must be considered. However, the male reproductive partner’s role during preconception has attracted much less researcher attention when compared to maternal exposures and outcomes and may be undervalued.

Therefore, this review explores the modifiable risk factors of males in the preconception period and how these risks influence adverse pregnancy and/or offspring outcomes. A total of 65 papers are included for review which examined risks associated with factors such as alcohol use, physical activity, stress, and nutrition. Overall, the papers identified some consistent results: paternal smoking increased risk of adverse offspring outcomes, while increased paternal body mass index was associated with higher offspring birthweight. Nevertheless, this review concludes that paternal preconception modifiable risk factors remain largely underexplored. Evidently, more high-quality research must be conducted to better understand the health, lifestyle, and diets of males in the preconception period and how various paternal modifiable risks can influence their partner’s pregnancy and the health and developmental outcomes of their offspring.

Introduction

Preconception care is defined as the provision of health interventions (behavioural, social, and/or biomedical) to women and couples prior to conception [1]. It addresses the transgenerational opportunity of enabling and optimizing health while limiting risk factors associated with both short- and long-term adverse health outcomes for women, men, and their children. There is global consensus on the key aspects of preconception care [2], yet a consistent definition and clear attributes of the preconception population remain elusive [3]. Preconception research predominately focuses on maternal modifiable risks or health behaviours associated with offspring outcomes [4] as demonstrated by a scoping review of preconception health behaviours which found only 11% of all studies included paternal modifiable risks or health behaviours [5]. Nonetheless, the research community recognizes the father or male partner’s contribution to child health and development before birth [6, 7] and the need to balance our gaze on men in preconception care [8]. This is further supported by the increasing number and diversity of publications about paternal preconception health [9] and formulation of the Paternal Origins of Health and Disease (POHaD) model [10]. As such, the preconception population may include all reproductively aged individuals in a period from their birth to the conception of their (or their partner’s) pregnancy. The care provided during this period must respond to a clear set of identified risk factors and exposures as relevant to each individual.

Indeed, when planning parenthood, males find themselves within a contentious grey zone; concurrently involved while also considered an outsider [11]. A recent survey in the UK found that men are interested in engaging in positive preconception health behaviours [7]. Of the over 500 men surveyed, 19% had visited a primary health provider for preconception health advice, and those who had received advice were more likely to adopt positive health behaviours before their partner’s pregnancy. On the other hand, general practitioners (GPs) report low confidence in their knowledge about paternal preconception health care and modifiable factors affecting male fertility [12, 13]. They describe feeling apprehensive or even sensitive to the subject matter and/or challenged by navigating the stereotypical masculine predispositions toward fertility and preconception care [14]. In general, preconception risks are not raised by GPs with male patients unless subfertility is involved and preconception discussions are often encumbered by numerous impediments including the limited time, financial constraints, and knowledge of GPs, plus in some cases, a lack of GP motivation and perceived need for health care [12]. A systematic review of preconception care guidelines found that six of the 11 guidelines included provided preconception care guidance for men [15]. Only one guideline, a position paper from the American Academy of Family Physicians, contained a dedicated section outlining recommendations on preconception interventions for men [16]. Evidently, there is an unmet need for health professionals, and men, to readily access current relevant information regarding paternal preconception health exposures and outcomes, informing clinical practice and directing health decisions.

Evidence supporting paternal preconception care considers males contribution to child health and development before conception via direct (genetic and epigenetic contributions – health and lifestyle behaviours, exposure to environmental toxins, life stressors, and neuroendocrinology) and indirect pathways (the couple’s relationship, and the influence of men on their partner’s health and health behaviours) [17]. Yet, there is a stark contrast between the magnitude of research investigating maternal preconception health risks—including body composition, lifestyle behaviours, and diet/nutrition – and the relative scarcity of research attention directed towards understanding paternal health exposures and outcomes. In direct response, this systematic review aims to advance paternal preconception research by synthesising the current evidence on associations of modifiable paternal preconception health behaviours and risk factors with pregnancy and/or offspring outcomes.

Methods

This review was prospectively registered in PROSPERO (Registration Number: CRD42021209994), and reported in line with PRISMA 2020 guidance [18] and the AMSTAR 2 critical appraisal tool [19].

Search strategy

A search was conducted on January 5th 2023, (See Supplementary File 1 Search strategy), through the following databases: 1) Medline (OVID) 2) Embase (OVID), 3) Maternity and Infant care [MIDIRS] (OVID) 4) CINAHL (EBSCO), 5) PsycINFO (EBSCO), 6) Scopus, & 7) ISI Proceedings. For each database, a date limit of 2012–2023 was set. When available, subject headings identified from the controlled vocabulary of each database were also included in the search. On January 11th 2023, a Google Scholar search was conducted for the search term ‘Paternal preconception’, applying the filter to limit articles published since 2022 and searching through to page seven, identifying any other recently published relevant papers. Google Scholar was also used to identify relevant studies citing each included paper. Reference lists of each included paper were then checked for additional relevant studies.

Selection criteria

Papers were included if they were original contemporary observational research (cross-sectional, cohort or case–control study designs) involving males in the preconception period, examining an association or correlation of a modifiable risk factor or health behaviour to pregnancy and/or offspring health and developmental outcomes. The male participants must identify as being the partner of the pregnant women and/or the biological father of the child for which pregnancy and offspring outcomes were reported (Table 1 – PICO).

Table 1 PICO (Population, Intervention, Comparison, Outcome) inclusion criteria
Full size table

Observational study designs are generally utilized to identify correlations and establish findings at the population level hence are solely considered in this review.

Papers were excluded if they were: reviews, did not report new empirical findings from original studies (i.e. commentaries, opinion-pieces and editorials), not studying humans, not examining male parent exposures, did not differentiate between maternal and paternal preconception exposures, or if the exposure examined specific illness populations. Papers were also excluded when the exposure was not assessed or retrospectively recalled during the preconception period, the outcome was not related to pregnancy or offspring health or development, or the risk factor or health behaviour was not modifiable. Google Translate was used to decipher any studies located in languages other than English.

Data extraction

Papers were imported into Covidence systematic review software [20], and duplicates removed by automation. Titles and abstracts were screened by TC, AS and DS. Full-text articles were obtained for relevant studies and reviewed based on inclusion criteria by TC who then extracted data from each included paper. AS or DS randomly reviewed the extracted data of ten included studies for accuracy and completeness. Any conflicts were resolved by consensus.

Data extracted from each paper included: the authors and year, study design and duration, location, the preconception population, total number of participants, the paternal exposures (and exposure measures), paternal outcomes (and outcome measures), any covariates considered and the main results from each association reported.

Quality assessment

The quality of each paper was assessed by TC using the Newcastle–Ottawa Scale (NOS). The NOS comprises three domains 1) selection of participants, 2) comparability of study groups, and 3) outcome of interest (cohort studies) or ascertainment of exposure (case–control studies), assigning stars in each domain to a maximum of nine stars [21]. Papers were then categorized as good quality (7–9 stars), fair quality (4–6 stars) or poor quality (0–3 stars) using groupings employed in previous research [22].

A meta-analysis was considered, but not possible due to exposure and outcome heterogeneity.

Results

A total of 65 papers were included in this review (Fig. 1 – PRISMA Flowchart) [18], comprising cohort (n = 56) and case control studies (n = 9) (Table 2 – Summary Table) & (Table 3 – Summary Table Findings). The majority of papers were conducted in the USA (n = 18), Europe and the UK (n = 19), and China (n = 17), several papers were from Australia [23,24,25,26,27,28,29] or included an Australian health centre [30,31,32,33,34]. Approximately half of all papers (n = 29) included a sample size between 370 and 2,900, while others included > 20,000 (n = 11) or ≤ 200 participants (n = 13).

Fig. 1
figure 1

PRISMA Flowchart

Full size image
Table 2 Summary table
Full size table
Table 3 Summary table of findings from included studies
Full size table

Study participants were diverse consisting of couples either intending pregnancy or pregnant (n = 25), sub-fertile and seeking fertility treatment undergoing IVF/ICSI (n = 11), or mothers and fathers of infants (n = 26). Two studies included adolescents followed into parenthood as adults [42, 80], and one study included individual respondents of a national family growth survey, actively attempting pregnancy [59].

Modifiable preconception risk factors and/or health behaviour exposures examined include paternal body composition (n = 25), alcohol intake (n = 6), cannabis use (n = 5), physical activity (n = 2), smoking (n = 20), stress (n = 3), and nutrition (n = 13) (including dietary folate intake and consumption of foods and dietary patterns). Two papers investigated multiple exposures [42, 43].

Outcomes examined include fecundability (n = 6) [38, 61, 72, 77, 78, 85], (time to) pregnancy (n = 4) [48, 54, 59, 68], IVF/ICSI ongoing pregnancy (n = 1) [84] or live birth (n = 7) [41, 49, 58, 60, 81, 82, 86, 87], offspring birthweight or adiposity (n = 10) [40, 42, 45, 47, 50, 52, 53, 62, 65, 66], including small for gestational age [SGA] [43], neonatal (n = 1) [23] and offspring body composition (n = 4) [24, 30, 39, 73]. Other outcomes examined include offspring asthma (n = 4) [25, 31, 33, 69] and lung function (n = 2) [32, 34], childhood leukemia (n = 4) [26, 29, 64, 67], childhood brain tumours (n = 2) [27, 28], and offspring behavioural issues (n = 2) [46, 55].

There was an increasing number of papers identified for inclusion in this review with the least number of papers published in 2012 and the greatest number of papers published in 2022 (see Fig. 2).

Fig. 2
figure 2

Papers included in this review

Full size image

Results below are described for papers assessed as good quality with approximately half (n = 30) rated good quality and two receiving a maximum nine-star rating [36, 54] (Table 4—Newcastle Ottawa Scale [NOS] quality assessment (Cohorts)) & (Table 5—Newcastle Ottawa Scale [NOS] quality assessment (Case controls)). Results for the fair and poor-quality papers are not further described.

Table 4 Newcastle Ottawa Scale [NOS] quality assessment (Cohorts)
Full size table
Table 5 Newcastle Ottawa Scale [NOS] quality assessment (Case controls)
Full size table

Body Composition

Twenty-five papers investigated associations between paternal BMI and various offspring outcomes. These papers were derived from studies (n = 21) conducted in the US, Europe, China, Australia, and Turkey and less than half (n = 11) rated as good quality.

Less than half of the papers (n = 10) used anthropometric assessment by the research team to determine BMI [37, 38, 40,41,42, 45, 47,48,49, 54]; heights and weights utilized to formulate BMI were determined in the preconception period, generally from males in couples undergoing IVF/ICSI [41, 49] or males in couples currently attempting pregnancy/pregnant [40, 45, 48, 54]. Of the papers validating the BMI utilizing anthropometric assessments, most were good quality and generally affirmed significant results. The remaining papers utilize retrospective reports of preconception paternal weight and height or collect paternal height and weight from medical records. Maternal reporting (n = 7) occurred at approximately 10 to 18 weeks gestation; or up to four months postpartum [35, 46]. Paternal self-reporting of their own weight and height (n = 8) occurred at approximately week 17 of gestation [39, 43, 50, 52, 53] or up to 6 months postpartum [43]. In two papers, overweight paternal status, when a child of 8 years, was reported years later through a validated drawing of silhouettes [33, 34].

The outcomes and outcome measures varied, however, ten studies assessed the association of paternal BMI with offspring BMI/bodyweight [23, 39, 40, 42, 45, 47, 50, 52, 53], and one paper assesses offspring weight and BMI changes from childhood (5 years) into adulthood (21 years) [24].

Results of associations between body composition and offspring outcomes were inconsistent. In some studies paternal preconception overweight (25.0—29.9 kg/m2) and obesity (> 30 kg/m2) were not associated with offspring birthweight [42] and paternal pregravid BMI was not an independent predictor for large for gestational age (LGA) or small for gestational age (SGA) [45]. However, other studies found that offspring birthweight increased by 10.7 g per unit increase in paternal pregravid BMI (95% CI: 0.5, 20.9, P = 0.04) [45], and each standard deviation (SD) increment of paternal BMI (approximately 3.27 kg/m2) was associated with an additional 29.6 g increase of birth weight (95% CI: 5.7, 53.5, P = 0.02) [52]. Further, compared with normal weight men, paternal pre-pregnancy overweight was associated with a significantly increased risk of preterm birth (aOR 1.34 95% CI: 1.25,1.45) and low birth weight (aOR 1.60 95% CI: 1.46–1.74) in offspring [47].

Paternal pregravid weight (P = 0.04), not height (P = 0.43), was associated with infant birth weight [45] and with increased risk of macrosomia (aOR = 1.49, [95% CI: 1.16, 1.92]) [53], while neonatal birth weight was associated with paternal height only (β = 0.08, P = 0.003) [23]. In another study, paternal pre-pregnancy BMI was only associated with offspring BMI when using absolute BMI values not BMI as a z-score [39].

Fathers’ overweight before puberty had a negative indirect effect, mediated through sons’ height, on sons’ forced expiratory volume in one second (FEV1) (beta (95% CI): − 144 (− 272, − 23) mL) and forced vital capacity (FVC) (beta (95% CI): − 210 (− 380, − 34) mL), and a negative direct effect on sons’ FVC (beta (95% CI): − 262 (− 501, − 9) mL) [34].

Male BMI ≥ 25 kg m2 was not associated with time to pregnancy (TTP) [48], yet underweight (< 18.5 kg/ m2) was associated with a longer TTP (adjusted fecundability odds ratio [aFOR] = 0.95, [95% CI: 0.94, 0.96]) compared to normal BMI (18.5—23.9 kg/m2) [54].

In couples undergoing IVF/ICSI, paternal periconceptional BMI was negatively associated with fertilization rate (β =  − 0.01 [SE = 0.004], P = 0.002]), while paternal BMI was not associated with the total motile sperm count (TMSC), the KIDScore, the embryo usage rate, a positive pregnancy, fetal heartbeat, or live birth [41].

Offspring methylation was associated with paternal BMI independent of maternal BMI (P =  < 0.05) [44]. Methylation decreased by 5% in cord blood with every 1-unit increase in paternal BMI (P = 3.13 × 10 -҆ꝰ), decreases persist at 3 years old (P = 0.002) and 7 years old (P = 0.004) [44]. Paternal BMI was associated with methylation at cg01029450 in the promoter region of the ARFGAP3 gene; methylation at this site was also associated with lower infant birthweight (β =  − 0.0003; SD = 0.0001; P = 0.03) [44].

No association was found between behavioural outcomes at pre-school age and underweight (< 18.5 kg/m2) or obesity (≥ 30 kg/m2) in fathers [36]. Equally, no associations were found between paternal BMI overweight (≥ 25 kg/m2- < 30 kg/m2), obese class I (≥ 30 kg/m2- < 35 kg/m2) and obese class II (≥ 35 kg/m2) and offspring behavioural issues or psychiatric symptoms at 7–8 years [46].

Alcohol

Six papers examined alcohol as an exposure [26, 42, 43, 55,56,57]; three rated as good quality [42, 55, 56]. Excluding one, each paper used paternal self-reports of alcohol consumption with varying definitions; one article specified units/per week [43], the others assessed consumption more broadly either as intake ≥ 1/week [56], ≥ 1/month [42] or general intake [57]. A single study presented a maternal report of paternal preconception alcohol consumption, 3 months before conception, at 12–16 weeks gestation [55].

When examining an outcome of offspring anogenital distance (AGD), in the paternal alcohol-exposed group (> 81 g/wk), male offspring had shorter mean AGDs [56]; for AGD-AP [the centre of the anus to the cephalad insertion of the penis] at birth (β =—1.73, P = 0.04) and 12 months (β = -7.29, P = 0.05), and shorter mean AGD-AS [the centre of the anus to the posterior base of the scrotum] at 6 months (β =—4.91, P = 0.02) [56]. Female offspring had shorter mean AGD-AF [ the centre of the anus to the posterior convergence of the fourchette) (β = -0.72, P = 0.02) at birth yet longer mean AGD AC [the centre of the anus to the clitoris] (β = 2.81, P = 0.04) and AGD-AF (B = 1.91, P = 0.04) at 12 months [56]. Further, the relative risks of anxiety or depression were increased by 33% (RR = 1.33 [95% CI: 1.09, 1.61]) and 37% (RR = 1.37 [95% CI: 1.02,1.84]) among girls in the exposed group at ages 4 and 6, respectively [55]. Paternal alcohol consumption greater than once per month was not associated with offspring birthweight or gestational age [42].

Cannabis

Paternal cannabis exposure was assessed in five papers [42, 58,59,60,61], two rate as good quality [42, 60]. Each paper has a sample size < 1,200 and each utilized paternal self-reporting of cannabis use broadly assessing general use, rather than specific amounts, over a pre-determined period (i.e., last 2 months or 12 months).

In sub-fertile couples undergoing IVF/ICSI, compared to males who were past or never cannabis users, couples where the male partner used cannabis at enrolment had increased probability of implantation (77.9, [95% CI: 53.5, 91.5], P =  < 0.05) and live birth (47.6, [95% CI: 32.4, 63.3], P =  < 0.05), independent of women's cannabis use [60]. Clinical pregnancy was not associated with male cannabis use [60], nor was gestational age or offspring birthweight [42].

Physical activity

The associations of paternal physical activity with offspring outcomes were assessed in two papers [42, 43], one rated as good quality [42]. This study found no association between paternal preconception bouts of physical activity per week and gestational age or offspring birthweight [42].

Smoking

The association of tobacco smoking with offspring outcomes was examined in 20 papers [27, 30,31,32, 42, 62,63,64,65,66,67,68,69,70,71,72,73,74,75, 88]; half (n = 10) were rated as good quality [31, 32, 42, 63, 64, 66,67,68, 73, 74] and nine papers adjusted for maternal smoking and/or paternal passive smoking in their analysis [32, 62, 64,65,66, 71, 73, 74].

Paternal cigarette smoking was associated with a longer TTP compared with never users (aFOR = 0.41, [95% CI: 0.24, 0.68]), while no associations were found for other tobacco products including cigars or snuff and chew tobacco [68].

Outcomes involving smoking and birth defects report that during the periconceptional period, light paternal smoking [1–9 cigarettes/day] increased the risk of isolated conotruncal heart defects (aOR = 2.23, [95% CI: 1.05, 4.73]) [63]. Medium paternal smoking [10–19 cigarettes/day] increased the risk of septal defects (aOR = 2.04, [95% CI: 1.05, 3.98]) and left ventricular outflow tract obstructions (aOR = 2.48, [95% CI: 1.04, 5.95]) [63]. Heavy paternal smoking (≥ 20 cigarettes/day) increased the risk of isolated conotruncal heart defects (aOR = 8.16, [95% CI: 1.13, 58.84]) and left ventricular outflow tract obstructions (aOR = 13.12, [95% CI: 2.55, 67.39]) [63]. Likewise, an increased risk of birth defects was found for continued-smoking (OR = 1.87, [95% CI: 1.36, 2.56], P < 0.001) and decreased-smoking groups (OR = 1.41, [95% CI: 1.10, 1.82], P = 0.007) compared with those fathers that quit smoking during early pregnancy and those who did not smoke at all during preconception [74].

Paternal preconception smoking at least one cigarette/day for one month was not associated with gestational age or offspring birthweight [42]. In contrast, a second study found sons whose fathers started smoking < 11 years, the adjusted mean differences in BMI, waist circumference, and fat mass all showed higher values at ages 13, 15, and 17 [66]. Further, the risk of childhood overweight and obesity was increased among children exposed to paternal preconception smoking compared to children without paternal smoking exposure (OR = 1.41 [95% CI: 1.17, 1.85]) [73].

Paternal preconception smoking 12 months prior to conception was associated with an increased risk of childhood acute myeloid leukemia (AML) (OR = 2.51, 95% CI: 1.21, 5.17) [64] and paternal smoking just 3 months prior to conception provided significant associations with acute lymphoblastic leukemia (ALL) (OR = 1.2 [95% CI: 1.1,1.5)] and acute myeloblastic leukemia (AML) (OR = 1.5 [95% CI: 1.0–2.3]) [67].

Paternal preconception smoking also provided significant associations with offspring lung function and asthma; fathers’ smoking initiation in prepuberty (generation G1) had a negative direct effect on their own FEV1/FVC (difference in offspring’s expected score − 0.36, 95% CI: − 0.68, -0.04) compared with fathers’ never smoking. This exposure had a negative direct effect on both offspring’s FEV1 (− 0.36, 95% CI: − 0.63, − 0.10) and FVC (− 0.50, 95% CI: − 0.80, − 0.20) (generation G2) [32]. Fathers’ smoking before age 15 years was associated with higher risk of asthma without nasal allergies in their offspring [relative risk ratio ((RRR) = 1.43 95% CI: 1.01, 2.01] [31].

Stress

Paternal stress exposure was examined in three papers [43, 76, 77]; including one rated as good quality [77]. This study found men's baseline perceived stress scale [PSS] scores were not associated with fecundability [77].

Nutrition

Papers examining paternal nutrition (n = 13) evaluated the associations of a range of nutritional exposures including paternal preconception folate, vitamins B6 and B12, and general dietary patterns with numerous offspring outcomes. These papers utilized data from several studies (n = 8) originating in the US, Norway, The Netherlands, and Australia. Approximately half of these papers (n = 7) rated as good quality.

Paternal nutritional factors explored included dietary patterns [82, 83] or specific foods groups including dairy [86], and meat [87]. IVF/ICSI-induced live birth was an outcome examined in three papers [82, 86, 87]. A positive association was found between paternal poultry intake and fertilization rate, with a higher fertilization rate among men in the highest quartile of poultry intake [78%] compared with those in the lowest quartile [65%] [87]. Men's total dairy intake was not associated with fertilization rate, implantation rate, clinical pregnancy rate, or live birth rate [86]. Also, paternal adherence to specific dietary patterns [pattern 1 = greater intake of processed foods/meats/high fat/dairy/sugar; pattern 2 = greater intake of fruit/vegetables/legumes/whole grains/nuts/fish] was not associated with fertilization rate [82] when undergoing IVF cycles.

One paper investigated dietary exposures during adolescence and subsequent neonatal health [80]. In a sample of adolescents followed into adulthood becoming fathers (n = 2,140), an extra serving of fruit per week was associated with a 2.35 g increase in offspring placenta weight [95% CI: 0.284, 4.42], P = 0.03 [80]. Further, paternal lunching regularly in adolescence was associated with an increase in offspring head circumference (β = 0.160, [95% CI: 0.001, 0.320], P = 0.05) and whole grain bread consumption was associated with a lower ponderal index (β = -0.003, [95% CI: -0.005, -0.001], P = 0.01) [80]. Birthweight was not associated with any paternal dietary exposures [80].

Generally, paternal preconception dietary patterns were collected through paternal self-reports on standardised food frequency questionnaires (FFQ) at baseline and include fast foods [42]; males eating fast food more frequently had infants born earlier than men who eat fast food less frequently (-0.16, [95% CI: − 0.32, 0.00], P = 0.04) [42].

Two papers specifically investigated paternal folate [79, 81]. In males undergoing fertility treatment, a 400 μg/day higher preconception folate intake was associated with a 2.6-day longer gestation [95% CI: 0.8, 4.3], P = 0.004 [81]. In spontaneously conceived pregnancies, a significant negative association was found between paternal red blood cell [RBC] folate status and crown-rump length (CRL) trajectories, in Quartile 2 [875–1,018 nmol/L;] (β = -0.14; [95% CI:—0.28, -0.006], P = 0.04) and Quartile 4 [1,196–4,343 nmol/L] (β =—0.19, [95% CI:—0.33, -0.04], P = 0.012) compared with the reference values in Quartile 3 [79]. A negative association was also found for embryonic volume (EV) trajectories in Quartile 4 (β =—0.12, [95% CI: -0.20, -0.05], P = 0.001) [79].

Discussion

This paper reports the first review collating literature assessing modifiable paternal health behaviours and risk factors in the preconception period and highlights clear disparity between the preconception research for women as compared to that for men. While single papers identified in our review do demonstrate adverse pregnancy and offspring outcomes associated with paternal risk factors in the preconception period, current research of paternal health behaviours and risk factors provides an emerging rather than mature evidence-base. Nevertheless, our review did identify a number of important findings.

One consistent finding of this review was the association between paternal preconception smoking and increased risk of adverse infant outcomes, including birth defects and childhood leukemia especially acute myeloid leukemia/acute myeloblastic leukemia (AML). Adverse outcomes such as birth defects are mirrored in maternal preconception smoking literature [89,90,91], yet the impact of maternal smoking on the risk of AML remains contentious [92, 93]. Smoking in the preconception period may be as perilous for males as for females, as smoking can potentially affect semen quality [94]. Many male smokers (and even more so in smoking couples) consider smoking an indispensable characteristic of their domestic, social and working lives [95] and many report a lack of motivation, willpower, and/or strength to successfully quit [96], in turn influencing female smoking patterns and family environments [95]. Paternal preconception smoking may well be contributing to the estimated 240,000 newborns dying worldwide annually due to birth defects [97]. The finding of paternal preconception smoking and the increased risk of adverse infant outcomes is altogether disconcerting considering the widespread use of tobacco, and that males are more likely than females to engage in risk-taking behaviours, including smoking [98]; the estimated global prevalence of male adolescent smokers in 133 countries is 23.29%.

The papers in this review which focus upon body composition with birthweight outcomes generally affirm positive associations between increasing paternal BMI and offspring birthweight. Indeed, this finding aligns with the literature outside this review which acknowledges that mothers and fathers with overweight or obesity are more likely to have children with overweight or obesity [99,100,101,102], compared with those with a normal weight. The positive associations between increasing paternal BMI and offspring bodyweight may, in part, be due to paternal contributions of sperm quality and potential changes to the epigenetic profiles of spermatoza [10, 103] resulting from unhealthy preconception environments and relationships with food. Food-based parenting strategies [100] and spending too much time sedentary [104] may also contribute to influencing offspring weight status. One paper in this review did chart offspring weight and BMI changes from childhood into adulthood [24], however, this reported research did not control for the offspring’s diet and physical exercise.

Nonetheless, an individual’s birthweight can influence both their body weight in childhood [105] and their body weight as they transition into adulthood [106]; external literature positively associates both a higher birthweight and childhood obesity with overweight/obesity at 15–20 years of age [107]. Frameworks to maintain healthy bodyweight, in turn promoting healthy birthweights, endure in the Global action plan on physical activity 2018–2030 [108] and in national overweight/obesity guidelines in countries such as Australia [109] and the Unites States [110].

It is important to note that most papers included in this review utilize retrospective reports (paternal self-reports or maternal reports) of anthropometric data collected at baseline. Such retrospective self-reporting is also evident in the maternal preconception literature [111, 112] and is often considered unreliable and subject to inaccuracies due to self-reporting bias or recall bias [113]. Inaccuracies and reporting bias may be present in particular in papers that utilize maternal reports of paternal preconception height and body weight at minimum 10 weeks of gestation in some papers up to 4 months postpartum. Consequently, retrospective reports of data at baseline may undermine the validity, accuracy, and therefore the reliability of BMI data used in these papers.

The majority of papers in our review report research undertaken in distinct geographical regions with the USA, Europe and the UK, and China heavily represented. As such, the implications for reduced geographical spread of the available research examining paternal preconception health exposures and outcomes must also be considered. It may be that existing region-specific idiosyncrasies of paternal health behaviours, and associated adverse health outcomes for their children, are yet to be described due to the absence of research conducted in other countries and cultures. These gaps limit the opportunities for tailored preconception care policies and interventions and constrain the broader understanding of the potential importance of paternal preconception care. Notwithstanding, such issues foster opportunities for other countries and cultures to identify, learn from and support paternal health.

While almost all papers in this review adjust for some confounders, less than half (n = 23) adjusted for the same maternal exposure (i.e., paternal BMI studies adjusting for maternal BMI). Many papers in this review did not adjust for maternal exposures and thus may present biased results and conclusions. Further, many maternal studies do not control for paternal exposures which is a limitation in the field that requires urgent research attention and refocus.

The date parameters set during the search may also represent a limitation as it may have resulted in manuscripts published before the 2012 being overlooked. However, up until recently the preconception research field has primarily focused on the effects of maternal exposures and as such it is unlikely that significant research was overlooked by this date restriction.

Further limitations of the review include the potential for missed citations due to issues with article indexing. Our search protocol did not employ search term truncations or singular synonyms in the final search string which may have resulted in some citations being missed. However, the search protocol was informed by an experienced health librarian, and additional methods – such as reference list and citation checking—were used to identify relevant manuscripts not identified through the primary search. Furthermore, previous search strings trialed for this review that used different synonyms, truncations and search term categories did not result in any additional relevant manuscripts being identified beyond those included in the final search. As such, the literature review is the most comprehensive review of the topic conducted to date.

This review is innovative in that it provides the first examination of paternal preconception risk factors and their association with adverse pregnancy and offspring outcomes. The rigour of the review is also bolstered through adhering to established systematic review reporting guidelines (PRISMA and AMSTAR).

Conclusion

Overall, this review shows that paternal preconception modifiable risk factors are largely underexplored; smoking and body composition appear to be important areas for consideration in paternal preconception care. While the current literature identifies an emerging evidence-base around paternal preconception modifiable risk factors, there is a need for further investigation to help better inform paternal preconception care and national and international preconception care guidelines. In particular, further research is necessary to identify and better understand the modifiable risk factors affecting males in the preconception period, and how these risk factors influence offspring outcomes, to inform clinical recommendations and health decisions. The future of paternal preconception care and the integration of such care into frontline health practice and policy rests with informed collaboration between clinicians, researchers and policymakers [8].