Description of studies
Details of each of the 46 studies can be found in Appendix 2. Table 1 provides an overview of studies by quantitative and qualitative designs. Nineteen papers were quantitative, 26 qualitative, and one used a mixed methods design . The mixed methods study was classified as qualitative, as the emphasis of reporting was clearly on this form of data. Thirty-nine studies were peer-reviewed papers, and seven were unpublished theses [63–69]. Although the studies were spread across regions, the majority were based in the US and Canada, followed by Australia, the UK, and Europe (majority Swedish). One study was based in the Middle East , one interviewed African-American couples , and another two interviewed Australian couples who were born in the Middle East [71, 72], with the ethnicities of samples in other studies primarily Caucasian. Of note, no studies were based in Asian or African regions.
[Insert Table 1 near here]
Most of the papers recruited heterosexual couples as the primary informant group, with another 16 focusing on heterosexual men only, and two including service providers (e.g., obstetricians/gynaecologists, midwives, grief counsellors and social workers) alongside bereaved men and women [68, 73]. No studies included gay or transgender men. Sample sizes varied widely, from one (an authoethnography)  to 131 men  in qualitative studies, and nine  to 341 men  in quantitative studies (see Appendix 2 for details).
The majority of studies investigated grief experiences following miscarriage (definitions ranging between ≤ 20–24 weeks’ gestation), 10 following stillbirth, and 15 following a combination of miscarriage, stillbirth and/or neonatal death (commonly defined collectively as ‘perinatal death’). Two papers explored experiences following termination of pregnancy for nonviable (or lethal) foetal anomalies [76, 77]. No papers focused exclusively on neonatal death. Twenty-three studies (16 quantitative and seven qualitative) focused on grief as a primary outcome; the remaining included elements of grief secondary to general explorations of experiences of loss, including ‘meaning’ , ‘impact’  and ’emotional responses’  among others [26, 64–66, 70–74, 77–87]. Two qualitative [78, 79] and two quantitative [27, 88] studies also investigated grief following pregnancy loss that continued into a subsequent pregnancy or birth of a child.
All but one of the included quantitative studies were variations of cohort designs, most commonly using structured questionnaires to assess grief. The remaining study was a RCT, examining the effectiveness of nurse-care and self-care interventions on grief following miscarriage . Qualitative studies predominantly used individual semi or unstructured interviews, however two studies used a postal  or online questionnaire , one used focus groups , and one was an autoethnography .
With the exception of four studies as noted above [31, 70–72], the majority of participants across studies were Caucasian, with those including mixed ethnicities providing little to no discussion on cultural or ethnic differences. Furthermore, all studies were conducted in high-income countries, and all male participants were heterosexual men who experienced pregnancy loss with a female partner. With the exception of six studies that did not specify men’s marital status [73, 74, 83, 84, 90, 91], the majority of male participants across studies remained in a relationship with the partner they were with at the time they experienced the loss/es, limiting knowledge regarding the experiences of single, separated or gay men. A total of 16 of the 46 included studies recruited only men [9, 28, 56, 64, 65, 67, 68, 74, 75, 80, 84–87, 90, 92]; the remaining included men in conjunction, or as a comparison to, their female partner. Thirteen studies reported age and standard deviations for male participants [26–28, 63, 67, 68, 78, 88, 92–96]. Across these, the average age of a total of 1,052 men was 33 years (pooled SD = 8.74). The remaining studies either did not report male participant ages [9, 71–74, 77, 82, 83, 86, 97], combined men’s ages with women’s [31, 79, 81, 89, 91, 98], or provided an average age and/or range [56, 62, 64–66, 69, 70, 75, 76, 80, 84, 85, 87, 90, 99–101]; with the youngest participant being 20 years , and the oldest 61 years , at the time of study participation.
Quality of included studies
An assessment of quality was undertaken for each study using CASP checklists . Study quality varied, however the overall standard was acceptable and therefore none were excluded based on poor quality. With the exception of 12 studies, whose recruitment methods were unclear or not reported [28, 99], or invited eligible bereaved parents from a certain hospital/region during a specific time period [62, 77, 85, 91, 93–95, 97, 100, 101], almost all of the studies used convenience, purposive or snowball sampling to recruit participants through social/print media or local clinics, hospitals and community settings. While ethically justified given the sensitive nature of the research, these methods may not reach men who are not linked into existing networks and may therefore be more isolated. As such, the results may not be representative of the whole population of bereaved men to pregnancy/neonatal loss (indicated by the narrow range of variability in participant characteristics). All studies adhered to appropriate ethical standards including obtaining informed consent, protecting participant confidentiality through ID numbers or pseudonyms, and offering contact details of pregnancy/neonatal loss support services to bereaved parents in case of distress. However, 10 studies did not state whether institutional ethical approval had been sought or obtained [28, 56, 83, 86, 87, 91, 92, 96, 98, 99], and two acknowledged potential conflicts of interest, relating to the first author being the developer of the intervention under investigation , and another being employed by the bereavement service under evaluation . Otherwise, no additional conflicts were declared by study authors or identified as a result of quality rating.
Qualitative studies were generally of a high standard, with methodologies and analyses—five using content [31, 56, 62, 67, 74], four thematic [68, 71–73], three grounded theory [66, 69, 77], one autoethnographic , one descriptive  and 12 phenomenological [63–65, 70, 78–80, 82–86]—clearly reported and justified in the context of ‘exploratory’ or ‘understanding lived experience’ research aims. Quantitative studies reported either correlational and regression analyses [27, 76, 81, 91, 93, 95, 97, 98], or group difference tests [26, 28, 81, 88, 92, 94, 96, 99–101], including significance testing of resulting relationships or differences. However, one small quantitative study reported only numbers and percentages of participants who endorsed a particular feeling relating to grief or service outcome , and another reported percentages of participants who had received certain support services following a loss .
With the exception of two studies which employed author-developed measures of grief and support service satisfaction [75, 87] the remaining quantitative studies employed standardised and validated measures for both predictors and grief [26–28, 76, 81, 88, 91–101]. However, there was an inconsistency in the use of grief measures and reporting grief. Although the majority of studies (n = 13) used the Perinatal Grief Scale (PGS) as a primary measure of grief, some of these reported average total grief scores [76, 93, 95, 98, 100], others average subscale scores [88, 101], both [28, 92, 96], or subscale correlations to predictor variables [27, 94]. Finally, 17 studies also grouped together different types of loss as part of the investigation of grief (e.g., miscarriage and stillbirth, stillbirth and neonatal death, or all three types together) [27, 56, 64–68, 70, 78, 84, 87, 88, 91, 93–95, 98]. As a result, outcomes specific to these different loss groups may have gone unrecognised, with only two studies discussing differences in support and subsequent grief between miscarriage and stillbirth [56, 68].
Findings relating to the grief experience
A total of 13 quantitative studies used the PGS as the primary measure of grief [27, 28, 76, 88, 92–96, 98, 100, 101]. Other grief measures used by remaining studies included the Grief Experience Inventory-Loss Version (GEI-L) , the Revised Impact of Miscarriage Scale (RIMS) [26, 81, 101], the Miscarriage Grief Inventory ; the Texas Revised Inventory of Grief (TRIG-F) , and, although primarily a measure of stress rather than grief, the Impact of Events Scale (IES) [76, 96, 99].
Of the 13 included quantitative studies that provided raw grief scores for men, outcomes varied considerably both between studies and within them, indicated by wide range and standard deviations [26, 28, 76, 81, 88, 92, 93, 96, 98–101]. Average total grief scores for men across studies using the PGS as a primary measure varied from 36  to 133.19  from a possible range of 33 to 165. However, the majority of average total PGS scores across remaining studies were between 73 and 83, with standard deviations ranging between values of 16 and 22 [28, 92, 93, 96, 98]. Population norms suggest that total grief scores above 91 for the PGS are reflective of a high degree of grief . The outcomes reported across studies here (with the exception of one study ) indicate that men typically aren’t scoring in the highly significant grief range; however, they are nevertheless scoring quite highly in general [28, 76, 88, 92, 93, 96, 98] (see Table 2 for a comparison of studies reporting total M and SD scores for the PGS). Similarly, for three studies using the RIMS as a measure of grief, outcomes also varied with subscale scores ranging from 0–57 , 5–24 , and subscale SDs up to 4.08 .
This variation in grief scores across studies may be due to inconsistencies in the timing of grief measurement. One study assessed participants’ grief scores one week following miscarriage , another into a subsequent pregnancy which was an average of nine months later , and another between one month and 32 years following the loss event . Overall, there was no clear data on the effects of time since loss on grief for men (see Table 2). However, some studies also noted that even when the losses had occurred many years in the past, participants were able to recall detailed thoughts and feelings about the loss, and their grief had not necessarily diminished with time [62, 76, 83, 87].
[Insert Table 2 near here]
In nine of 10 studies which compared men and women, men’s grief scores were found to be significantly lower or less intense than those of women [76, 99, 101], with approximately 20 points of difference on the PGS and IES [93, 96, 98], and 3 points of difference on the RIMS [26, 81]. Importantly, however, some studies noted that the use of existing grief measures (including the PGS and RIMS) might not be valid for measuring men’s grief experiences, particularly in relation to potential differences between internal versus external grieving styles [26, 88, 94, 100]. This is supported by the fact that there were mixed findings in terms of overall scale scores across similar studies looking at grief following miscarriage, with Despair (internalised grief) scores higher in men than those for Active Grief (externalised grief) in two studies [92, 100], and lower in the remaining in two [96, 101]. Across other grief measures employed by studies, men scored highly on the Devastating Event (RIMS), Denial and Social Desirability (GEI-L), and Avoidance (IES) subscales [26, 81, 96, 99, 101], which may represent some of the more inward responses to loss involved in some men’s grief experiences.
In the majority of qualitative studies (n = 14), men reported that the loss of their baby was a significant life event, regardless of gestational or infant age [9, 56, 64–69, 78, 80, 83, 85, 86, 90]. However, other men in 10 studies (some overlapping with the above 14 studies) also reported less intense reactions, including stating that their partners experienced worse grief in comparison to them [31, 63, 68, 71, 72, 78–80, 82, 83]. Regardless of grief intensity, in 15 studies, men seemed to face additional or unique tasks and challenges that complicated their experience or delayed the timing of grief. These included a sense of helplessness or powerlessness, especially during the female partner’s miscarrying or labour experience in pregnancy loss [9, 31, 80, 83, 85], and additional roles and practical responsibilities such as caring for other children, completing paperwork, organising a funeral/burial, and informing family and friends [9, 56, 64, 65, 68, 69, 84–86].
Although the grief experience was highly varied, and subsequent grieving styles across qualitative studies were mixed, there was a general trend among male participants towards instrumental grieving, and the use of active or problem-focused coping strategies [9, 62–73, 75, 78–80, 82–86]. ‘Keeping busy’ and ‘moving forward’ were common desires [62, 66, 70, 78, 82, 84], with men seeking out distractions including sporting activities or increased exercise [65, 66, 69], returning to work [64–68, 72, 79, 83, 84], completing household tasks [65, 68, 78, 85], and the use of creative, hands-on outlets such as woodworking, painting or writing [9, 64, 65]. However, men in 10 studies also reported outward emotional grief expressions such as crying, although these were frequently kept private, with many men preferring to grieve independently and alone [9, 56, 63–66, 69, 85, 86, 90].
Findings relating to predictors of men’s grief
Of the included quantitative studies, 16 included an analysis on predictors of men’s grief and/or correlations to related factors [26–28, 76, 81, 88, 91–98, 100, 101]. As part of a wider exploration of the grief experience, all of the included qualitative studies also discussed factors and/or situations that contributed (both positively and negatively) to men’s grief. Overall, a wide range of varied predictors/factors were considered by the included studies, which fell broadly into four domains or levels: (1) individual/person-level factors; (2) interpersonal factors; (3) community/sociocultural factors; and (4) public policy factors.
Attachment to the baby
One of the strongest factors found to impact upon grief across studies at the individual level was men’s attachment to the baby. Across 11 qualitative studies, men who had developed a bond with their baby throughout the pregnancy described more intense grief following a subsequent loss [56, 65, 67–69, 78, 80, 83, 85, 86, 90]. However, in five studies some men stated that they did not feel that they had a relationship with the developing baby [31, 68, 80, 83] (either because it was an early miscarriage or they described little involvement during the pregnancy), or made a conscious attempt during pregnancy not to get attached, due to previous experience of loss or diagnosis of a life-threatening condition . In these cases, grief was reported as less intense. Actions that served to increase attachment included spending time with the baby [9, 90], and attending ultrasound appointments to ‘see’ the baby and hear the heartbeat [9, 56, 68, 69, 78, 83, 86, 90]. Although estimates of grief outcome were imprecise due to a small male sample size, one quantitative study measuring grief after seeing or holding the stillborn baby identified worsened grief for men . Similarly, fathers in six qualitative studies who held or spent time with their baby following a stillbirth generally also reported high levels of grief [9, 65, 69, 70, 85, 90]. Importantly however, the cause and effect relationship here is unclear—it may be that men who spent time with their baby were already more attached, and therefore more likely to experience worsened grief.
Seven quantitative studies explored men’s attachment to the developing baby including viewing an ultrasound , vividness of visual imagery , increasing gestational age [26–28, 76, 98], and holding or seeing the baby following stillbirth . Men who viewed an ultrasound image had an average PGS total score 23 points higher than those who did not view any images , and men with a strong visual image of their baby, as measured by the Baby Vividness of Visual Imagery Questionnaire (BVVIQ; “vivid imagers”), had an average PGS total score 40 points higher than those who did not . Again, however, the causal relationship here is unclear.
Attachment could also be considered in relation to gestational age, since a longer pregnancy could result in more opportunities for bonding. In five quantitative studies, increasing gestational age was associated with higher grief scores [26–28, 76, 98]. However, qualitative studies complicated this picture—in studies inclusive of multiple loss types, men who had experienced earlier losses did not describe less intense grief than those with later losses [56, 64–66, 68]. The authors of studies which focused on miscarriage also noted that men’s grief responses were not dissimilar to the grief of men described in studies focused on stillbirth or neonatal death [69, 80, 86]. As such, the impact of gestational age on grief remains unclear.
Two studies on the same sample of bereaved parents in Australia [93, 95] investigated the relationships between perinatal grief and a general personality proneness to shame (attributing regretful actions to oneself) and guilt (considering one’s actions as regretful). Overall, shame and guilt-proneness were found to explain 63% of the variance in grief outcome (as measured by the PGS) in men, with the largest contribution being shame-proneness, accounting for 56% of the variance in men’s late grief (13 months following a stillbirth or neonatal death) . In the follow-up study , which conducted analysis within the couple, women’s self-conscious emotions/grief tendencies did not appear to influence men’s emotions/grief tendencies (although men’s impacted upon women’s). Franche  similarly explored the predictive value of self-criticism on grief after pregnancy/neonatal loss. Considered in combination with other obstetric and demographic variables, higher levels of self-criticism were found to be significantly associated with higher scores on all subscales of the PGS in men (p <.01 for the Active Grief subscale, and p <.001 for Despair and Difficulty Coping subscales).
Findings relating to the relationship between demographic factors and grief were mixed. Only one quantitative study  found age to be a significant predictor of grief outcome following miscarriage, with men aged <35 years scoring higher on the Devastating Event subscale of the RIMS. The remaining quantitative studies including age as a predictor did not find a significant association [27, 98, 100], and qualitative studies did not specifically explore or discuss the impact of age on grief. However, the majority of men who participated in qualitative studies were generally aged 28 years or over, with the exception of two studies which reported minimum ages of 20 and 21 years [56, 70].
Ethnicity did not emerge as a significant predictor of grief, although it was rarely explored. One study comparing Swedish and American couples’ experiences of miscarriage , found differences between the samples on one subscale of the RIMS (Loss of Baby); however, this difference was attributed by the authors to linguistic understanding and wording of scale questions, rather than the grief experience itself. Other quantitative studies whose samples included a small number participants identifying with cultural backgrounds other than Caucasian (e.g., African American, Asian-Australian, Hispanic, Native American) either did not comment on or examine differences [26, 93–95, 98], or did not find any significant differences in grief outcome . Five qualitative studies had mixed ethnic samples (e.g., Jamaican, African-American, Hispanic/Latino), but none reported any differences in grief outcome—although, their aim was not to do so [64, 67, 69, 78, 86]. Further, in two Australian-based studies of the same sample of participants with Middle-Eastern backgrounds, culture was not discussed as impacting upon grief outcome [71, 72]. In one qualitative study based in Israel , high drop-out rates for participation in the study were noted due to (mostly) the husband’s objection to participating in the context of a typically “closed” religious society. Finally, in a study of perinatal loss in low-income African-American parents, grief for fathers did not differ to those in other studies; however, “dealing with stressful life events”, including economic hardship and other unrelated family deaths, were found to compound the perinatal loss grief experience for both parents .
In one quantitative study , involvement in organised religious activity was inversely associated with Despair subscale scores on the PGS for men (p = 0.047). In seven qualitative studies, men who reported religious or spiritual beliefs also found this to be a source of comfort in coping with their grief; this was both from a meaning-making perspective (e.g., “what God does, He does it for the best”) , and from the additional social support that was received from religious/church communities [65, 66, 69, 70, 78, 85, 86]. However, the experience of loss for some men (and their partners) in two qualitative studies also led to questioning or challenging of their religious beliefs [9, 31].
Recurrent loss and living children
Findings relating to the impact of previous losses and number of living children on grief were also varied across studies. In one quantitative study which examined men who had experienced recurrent miscarriage, reported grief and stress scores were high on both the PGS (M = 72.23, SD = 16.85), and IES (M = 26.53, SD = 13.76) . In contrast, men with a history of loss in nine qualitative studies [56, 68, 69, 71, 72, 77, 82, 86, 87] did not report different or increased levels of grief; yet, in four studies, men reported increased worry about future pregnancies [66, 69, 80, 82].
In two quantitative studies which included subsequent pregnancy status as an indicator of grief intensity, no significant relationships were found between a group who were currently pregnant following a loss, and a group who had not had a subsequent pregnancy or child [88, 94]. However, in three qualitative studies examining experiences of grief into subsequent pregnancies/children, it was clear that men’s grief continued, along with added concerns and vigilance due to the knowledge of potential risks [78, 79, 84]. Similarly, one of three studies examining the presence of living children at the time of loss found a relationship to worsened grief in men , however for the remaining two studies including this factor, it was unrelated [91, 94]. Four qualitative studies described how living children could both enhance the reality of the developing baby (thus worsening grief), and make coming to terms with the loss easier; either through enhanced appreciation for surviving children, reassurance about the possibility of successful future pregnancies, or providing a caring role to focus on [65, 80, 82, 85].
Interpersonal factors identified as predictors of grief included quality of the partner relationship, the ‘supporter role’, and support and acknowledgement received from family, friends, and healthcare professionals.
Quality of the partner relationship
In ten qualitative studies, men noted that the relationship with their partner could be either a positive or negative contributor to the grief experience [62, 64, 66–69, 71, 75, 79, 85]. For many participants in these studies, a lack of recognition for their grief from family, friends and healthcare professionals meant their partner became their main source of interpersonal support [66, 68, 85]. Although many men reported supportive relationships with “frank and honest communication”  resulting in a stronger couple bond that buffered the grief experience, many also experienced conflict or relationship strain due to incongruent grieving styles [62, 64, 66, 68, 69, 75, 79, 80, 85]. Where dissonant grieving styles or conflict were present, men reported a sense of alienation or frustration that added to their grief experience [62, 67, 68, 79]. However, despite early conflict, where couples learned to effectively navigate one another’s grief, the relationship was ultimately strengthened [66, 69, 79]. No quantitative studies explored relationship quality as a contributor to grief.
The supporter role
Although not a factor quantified for measurement in any of the included quantitative studies, one of the most consistently reported and important elements relating to men’s grief experience across qualitative studies was that of taking on the role of being a ‘supporter’ to their female partner and family. A total of 23 of the 26 included qualitative studies identified an element of the supporter role from men’s responses [31, 56, 62–66, 68–73, 77–80, 82–86, 90]. In 21 of these, all of the male participants reported their primary role of being the supporter to their female partner, and in the remaining two, the majority of men (five of nine , and 14 of 15 ) also reported this role. For men in five studies, the need to support their partner explicitly came from a perception that she had a more intense grief reaction in comparison to themselves [31, 66, 68, 79, 83]. In 15 studies, men described having to suppress or put aside their own grief in order to take on this role [56, 64–66, 68–72, 77, 79, 80, 85, 86, 90]. As a result of doing so, many of these men reported a feeling of being ignored or unrecognised as grievers, rather seen merely as the ‘support person’ [56, 68, 84]. Although some men in three studies reported feeling as though this supporter role was helpful throughout their grief process, in that it gave them a meaningful task to focus on [31, 78, 80], for the majority of men across the remaining studies, this role ultimately served as a hindrance in allowing them to acknowledge, express and manage their grief and emotional responses [31, 56, 63–66, 68, 69, 71–73, 77, 79, 80, 83–86, 90].
Support and acknowledgement from family and friends
In sixteen studies looking at support, ten found family and/or friends to be a helpful facilitator to men’s coping and healing following the loss [9, 63, 64, 66, 72, 78, 80, 83, 85, 86, 98], especially since many men explicitly reported a preference not to engage in formal counselling  and/or support groups [63, 68]. However, across these studies, men’s experiences of support from family and friends varied greatly. In the one quantitative study that looked at family and friend support as variables, ‘talking with friends’ was associated with increased grief scores, along with ‘timing of talking to family’; although there is no description of what is meant by this . In the remaining qualitative studies, the majority of men also reported talking with either close family members or friends post-loss, and they found this meaningful and helpful most of the time [63, 64, 66, 68, 80, 83, 85, 86]. Practical support immediately following the loss (e.g., making meals) was particularly appreciated by men in three qualitative studies [68, 72, 86], and for others “subtle” gestures of care from other male friends, including sharing their own stories or scheduling time/activities post-loss, were immense comforts [9, 85, 86]. However, seven qualitative studies also reported negative—or a total absence of—interactions with family and friends [66–69, 78, 80, 84]. In two of these studies, men did not feel the need to discuss their grief with anyone other than their partners, or avoided talking to others about the loss in the hope that avoidance would reduce the impact [78, 80]. In the remaining five, men desired support from family and friends, however stated that “no one”  was available to them, due to a lack of understanding, avoidance and/or discomfort [66–69]. Where a lack of acknowledgement or support from family and friends was available, reported grief experiences were worsened [67, 68, 84].
Support and acknowledgement from healthcare professionals
Similar to support from family and friends, the role of healthcare professionals was recognised in one quantitative study  and 13 qualitative studies [56, 63, 67–69, 72–74, 77, 78, 82, 85, 86] as essential to the grief process. However, among studies that examined healthcare provider support, findings were again mixed. In 10 studies, some men reported positive experiences with healthcare staff [56, 67, 68, 74, 77, 78, 80, 82, 85, 86]. Particularly, three studies noted that providers who worked “extra hard” to provide both medical and practical information to men were valued , and parents who received the support of specialist bereavement care teams, or follow-up telephone calls from care providers, commented positively on this experience [77, 82]. However, men in one quantitative study felt excluded from services, and none were satisfied with the support they received from health professionals . Likewise, other men in 10 of both the same and different qualitative studies also reported negative interactions with healthcare staff, leading to sadness, anger or distress which worsened or prevented the grieving process [56, 68, 69, 72–75, 80, 83, 85, 86]. Common issues included insensitive language or confusing medical terminology [83, 85, 86], a lack of answers or explanations [68, 69], a lack of practical information on how they could care for their female partner or organise a funeral/burial [56, 69, 72], and failing to recognise their distress and role as a father [56, 73–75, 80, 83]. It should be noted that the majority of studies reporting negative experiences with healthcare providers or the hospital focused on miscarriages as opposed to later-term losses, with the exception of three which focused exclusively on stillbirth [73, 74, 85]. Two studies which explored healthcare support following both miscarriage and stillbirth also noted differences in care between these types of losses, with miscarriages receiving considerably less support in comparison to stillbirths [56, 68].
Disenfranchisement of grief following pregnancy/neonatal loss
A lack of community acknowledgement and understanding for grief following pregnancy loss was explicitly identified by male participants in seven qualitative studies from the US [9, 65, 73, 86], Ireland [56, 82] and Australia , but this was not explored in quantitative studies. Across these, men discussed widespread taboo, stigma and silence surrounding miscarriage and/or stillbirth which worsened their grief. Reported experiences of disenfranchisement included questioning their identity as fathers due to confusion surrounding whether their pregnancy was understood as a baby or not , only discussing their loss if/when prompted by another bereaved parent , and hurtful comments from others which minimised their grief or encouraged them to “move on” from the loss [68, 86]. Overall, this sense of disenfranchisement due to a lack of community acknowledgement for pregnancy loss led men to experience increased distress and feelings of isolation [9, 56, 65, 68, 86].
Male role expectations and attitudes toward men’s grief
Tying in closely with the ‘supporter role’ discussed above, a pressure to conform to masculine role expectations and social attitudes toward how men should grieve, was expressed by men in 19 of the included qualitative studies based in Australia [68, 72, 90], the UK , the US [9, 62–67, 69, 73, 79, 84, 86], Ireland , Sweden  and Israel . No quantitative studies explored this factor. In 14 studies, male participants specifically discussed the need to be “strong”, and a perceived expectation to hide their grief [56, 63–66, 68, 69, 72, 79, 83, 84, 86, 90]. Men reported that these expectations had a direct negative impact on their grieving process, as they felt prevented from displaying their emotions in front of others, seeking support, and/or working through their grief [9, 56, 64, 68, 73, 79, 84, 86, 90]. In turn, this expectation to hide their emotions also meant that the impact of the loss on these men was frequently disguised from family, friends and healthcare professionals. This led to a generalised lack of recognition for their grief and also a further sense of disenfranchisement, above that which already exists for grief following pregnancy/neonatal loss generally [67, 68, 86].
Public policy factors
Women-focused maternity care and support services
Similarly to the data presented on men’s experiences of support from healthcare providers, a general broader focus on women-centred care in the hospital environment, and among existing support services, was identified by nine of the included qualitative studies, but not in quantitative studies, as a factor impacting upon men’s grief [9, 56, 67, 68, 74, 75, 84–86]. A general community attitude that pregnancy and subsequent loss was primarily a “women’s experience”  was explicitly expressed by men in three of these qualitative studies [56, 67, 84]. Men reported feeling overlooked or ignored in the context of existing healthcare and support services. For example, in the hospital environment, both following loss and during subsequent pregnancies, men felt “out of place” , “marginalised”  and sometimes, as though they “barely existed” . Similar sentiments were echoed in the context of support services/groups which were delivered primarily by women and focused on “ ‘traditionally feminine’ modes of grieving” , which may not be suited to all types of grief and responses [9, 67, 68, 84]. Men in five studies expressed a desire for recognition [84–86], as well as a need for increased male involvement in care and support services [56, 68]. Indeed, in studies where other males were present and available to men, or healthcare staff sought to specifically involve and include men in services, reported grief experiences and outcomes were improved [9, 56, 67, 68, 74].
Workplace policies: bereavement leave
Another consistent theme that arose at a policy level for men was related to the workplace; particularly, surrounding the availability of paternity or bereavement leave following pregnancy/neonatal loss. Returning to work following loss was explicitly discussed by men in 11 qualitative studies [9, 64–69, 79, 83, 84, 86] and one quantitative study . For the majority of men in these studies, particularly those who described a more instrumental grieving style, work provided a welcome and helpful distraction from their loss, used as a strategy to better manage and cope with their grief [64–67, 79, 83]. However, four qualitative studies which examined men’s experience of returning to work in more depth identified varied outcomes [9, 68, 84, 86]. In three of these studies, men were not provided with the same opportunities as their female partners to take paid leave from work following their loss [9, 68, 84], leading to physical and emotional exhaustion that further added to their grief, along with difficulties in concentration and keeping up with tasks. In one quantitative study , men also reported difficulty returning to work. In contrast, the burden of grief was eased for male participants in two studies who were offered extended paid leave or extensions on work-related deadlines [68, 86].
The emerging model: a socio-ecological theory of men’s grief
Spanning the individual, interpersonal, community and public policy realms, the factors identified in this review align with a socio-ecological approach to understanding grief. We propose a preliminary model of men’s grief, adapted from Bronfenbrenner’s  ecological systems theory (see Figure 2). The original theory (focusing more broadly on development as opposed to grief) purported that an individual’s development is impacted by four interacting levels in the environment: the microsystem (the immediate environment), the mesosystem (settings in which we actively participate), the exosystem (wider social setting), and the macrosystem (culture and belief systems) . Like the original theory, the model of men’s grief proposed here acknowledges that the grief experience does not exist in isolation. Rather, it is shaped by a complex system of interacting factors and levels, including those relating to the individual, their relationships, the surrounding community, the wider context and governing policies. Each of these levels also interacts with one another in a bi-directional nature. For example, cultural norms and beliefs regarding men’s roles—particularly in pregnancy—may play a vital role in informing the women-centred focus of perinatal healthcare and bereavement leave policies (and vice versa). These norms can also impact the ways that individuals interact with one another in response to pregnancy/neonatal loss, as do these interpersonal interactions serve to support the overarching cultural norms. At the centre, the individual, their personality, knowledge, attitudes and skills are impacted by, and continually interact with, all of these contributors.
The overarching theme of this model is the concept of “double disenfranchisement”, first introduced by Cacciatore and Raffo  in their study on lesbian maternal bereavement. The authors argued that given an additional lack of societal recognition for their status as legitimate mothers, lesbian women can experience an added level of disenfranchisement in relation to their existing grief experience of pregnancy loss . In a similar way, the lack of recognition that many men cited in this review for their position as grieving fathers indicates that they may also experience a sense of added or double disenfranchisement. Consequently, in conjunction with the factors presented above, it is imperative that men’s grief following pregnancy/neonatal loss is not viewed entirely as an individual response to the event, but as part of a wider socio-ecological process.
[Insert Figure 2 near here]