Does a history of childbirth impact injury prevalence and mental health in female military members?

From September 2020 to February 2021, data for this cross-sectional study were collected via a questionnaire developed as part of the “Multi-stage approach to addressing sex-disparities in musculoskeletal injuries in military members” research project. The overarching aims of the questionnaire were to characterize MSKi risk, reproductive health, and barriers to recruitment and retention of females in the CAF. The inclusion criteria for the questionnaire were: being a member of the CAF (past or present) and between the ages of 18–65 years. Recruitment was conducted by the research team and project stakeholders via posters, social media, newsletters, and snowball methods. The questionnaire was accessed, and informed consent was provided via the online cloud-based survey development software SurveyMonkey Inc. (San Mateo, CA, USA).

A total of 2001 participants provided consent and completed the SurveyMonkey questionnaire. For the present study, the inclusion criteria were: regular or reserve, being female, and being between the ages of 18 and 65 years. Members who were male (n = 1092), intersex (n = 1), did not provide biological sex (n = 54), retired (n = 35), medically released (n = 49), or did not provide CAF employment status (n = 22) were excluded from this study. Therefore, the final n = 748 was used for this analysis. This study was approved by the local Research Ethics Board (H-04-19-3442) and the procedures were performed in accordance with the Declaration of Helsinki.

The minimum required sample size for the larger project was estimated using two strategies: (1) representation of the overall CAF population and (2) previously reported odds ratios from studies comparing MSKi rates from females vs. males in the military context. For the present study, we used the minimum sample size derived from the first strategy. The estimated number of CAF members (i.e., Regular Forces and Reserves) is 103 873 with approximately 16% being female (Jackson 2020). Considering the estimated prevalence of RSI (∼35%) and acute injuries (∼19.8%) reported in female CAF members (Guerin 2020), maximum error of ±5%, and a design effect of 1.5 (accounting for the nature of recruitment and study design), the minimal sample size was n = 508.

Parity status and pregnancy-related complications [e.g., postpartum depression, miscarriage, stillborn, or traumatic birth (use of vacuum or forceps, or episiotomy required)] were used as the independent variable, and MSKi rates (RSI and acute injuries) and MSKi-related outcomes (e.g., impact on daily activities and mental health status) were analyzed as dependent variables. Appendix A further describes all the independent variables, covariates, and outcomes used in the analysis. The original categories for response, categories, and references for analysis are also presented in the same table.

All analyses were conducted with IBM SPSS Statistics version 27 (SPSS Inc., Chicago, IL, USA) and significance was set at p <0.05. Data were presented descriptively as absolute (n) and relative (%) frequencies, as well as median and range (minimum and maximum). For continuous variables, data normality was checked using the Shapiro–Wilk test. Comparisons between groups (parous vs. nulliparous) were made using Mann–Whitney U test, according to the results of the normality test (non-normality-distributed data). χ² tests or Fisher’s Exact tests (i.e., when the assumption of having <20% of the cells, with an expected count of less than 5, was not met) were used to test the bivariate associations (i.e., 2×2) between the independent variables (parity and pregnancy-related complication) and the outcomes (MSKi and health outcomes). Independent variables not significantly associated with the outcomes (p ≥ 0.05) were not moved to the logistic regression analysis. For the significant bivariate associations, binary logistic regression models were generated and include the adjusted odds ratio (aOR) [95% confidence intervals (CIs) and p values] using age, rank, military environment, and current body mass index (BMI) classification as covariates, unless otherwise indicated (e.g., if they have not met a test assumption). The following assumptions were confirmed for the binary logistic regressions: (i) sufficiently large sample size (i.e., at least 10 observations with the least frequent outcome for each independent variable); (ii) multicollinearity for independent variables and covariates included in the model; (iii) linearity of the logit for the continuous covariate included (i.e., age); and (iv) goodness of fit (i.e., Hosmer–Lemeshow test).

Parous members reported more RSI overall compared to the nulliparous group (80.9% vs. 69.9%, p = 0.002). A comparison of RSI prevalence, stratified by body region, is displayed in Fig. 1. When adjusted for age, BMI, military environment, and rank, parous females were more likely to report an RSI (aOR = 1.573, CI: 1.025 to 2.415, p = 0.038) compared to the nulliparous group. When stratified by body region, parous females were more likely to report an RSI of the wrist (aOR = 1.628, CI: 1.09 to 2.430, p = 0.017) and foot (aOR = 1.742, CI: 1.199 to 2.530, p = 0.004).

Of the participants who sustained RSI (n = 453), 34.5% (35.1% nulliparous vs. 33.8% parous, p = 0.771) had an impacted career progression or length, and 75.1% (74.5% nulliparous vs. 75.9% parous, p = 0.831) formally reported their RSI to a medical health professional. While RSI interference with work duties was observed at similar rates between parous and nulliparous (59.6% nulliparous vs. 58.2% parous; p = 0.760), the parous group reported a higher impact on daily activities (76.2% nulliparous vs. 88.4% parous;p ≤ 0.001).

Acute injuries were reported by 59.2% of participants (58.0% nulliparous vs. 60.7% parous; p = 0.512) and no significant differences were observed when stratifying by body region or acute injury type. The most common body areas affected by acute injury were the ankle (28.7%), lower back (23.5%), and knee (20.8%). Sprains and strains were the most reported acute injury type (35.9%), followed by broken bones (13.1%), scrapes and bruises (11.6%), and concussions (9.6%).

Of the participants who sustained an acute injury (n = 354), 38.5% (39.7% nulliparous vs. 36.9% parous, p = 0.597) felt the injury impacted career progression or length, and 90.2% (87.7% nulliparous vs. 93.2% parous, p = 0.204) formally reported acute injuries to a medical health professional. No difference in daily activities (56.8% nulliparous vs. 58.0% parous; p = 0.820) or work duties (55.0% nulliparous vs. 45.0% parous; p = 0.064) being negatively impacted by acute injury were observed.

Seventy two out of 313 participants with a history of childbirth did not respond to questions pertaining to pregnancy-related complications. The following pregnancy-related complications were reported (n = 241): previous miscarriage (26.6%), preterm birth (9.1%), gestational diabetes mellitus (7.1%), preeclampsia (7.5%), gestational hypertension (3.7%), postpartum depression (22.4%), and traumatic birth (28.6%). Significant findings comparing the pregnancy-related complications with MSKi and health-related outcomes identified using χ² were carried forward for bivariate logistic regression analysis presented in Table 2.

Descriptive analysis of questions pertaining to physical training provided during pregnancy and postpartum is illustrated in Fig. 2.

The higher rate of wrist RSI reported by the parous group was expected considering carpal tunnel syndrome, an RSI involving the wrist, has been associated with pregnancy (Padua et al. 2010). A literature review performed by Padua et al. (2010) found that pregnancy-related carpal tunnel syndrome affected approximately 17% of research participants. While 30% of our parous sample reported wrist RSI, specific diagnosis was not accounted for. Padua et al. (2010) also noted that most cases of pregnancy-related carpal tunnel syndrome appear to resolve spontaneously within 1–3 years. Research exploring wrist injury in parous CAF females should consider known risk factors for pregnancy-related carpal tunnel syndrome (i.e., obesity, BMI, gestational weight gain, and relaxin) (Wright et al. 2014; Kang et al. 2017).

Other injuries associated with pregnancy are not prominent in the literature. However, pain at specific body regions throughout pregnancy, such as plantar pain, is fairly well documented (Anselmo et al. 2017). There are a number of changes identified in the foot during pregnancy (i.e., length, width, and volume), and altered plantar pressure distribution during gait has been observed (Wetz et al. 2006; Anselmo et al. 2017; Vico Pardo et al. 2018; Gimunová et al. 2020; Christopher et al. 2022). The changes in the foot can persist after childbirth and are attributed to increases in body mass and endocrine system responses to pregnancy (Segal et al. 2013; Vico Pardo et al. 2018). While it is unclear if these pregnancy related changes contribute to injury risk at the foot in female CAF members, considerations for these members are recommended when returning from maternity leave. Specifically, scheduled fitting of new personal protective equipment (i.e., footwear and uniform) after pregnancy should be prioritized to ensure the member is adequately protected when returning to duty. Physical training programs for CAF members who have given birth should be created with the understanding the wrist and feet may be more susceptible to injury. Furthermore, practitioners should assess gait after pregnancy to determine if the new movement patterns are potentially problematic.

Our review of the literature did not uncover specific research examining postpartum depression and MSKi, though associations with lumbopelvic pain have been documented previously (Gutke et al. 2007). In a cohort study conducted by Gutke et al. (2007), depressive symptoms were three times more likely in participants with lumbopelvic pain at 3 months postpartum (Gutke et al. 2007). To note, the prevalence of postpartum depressive symptoms (Edinburgh Postnatal Depression Scale screening cut-off ≥10) in Gutke et al. (2007) was 8% compared to postpartum depression diagnosis of 26.6% in this present study (Gutke et al. 2007). Disparities in depression rates between our study and Gutke et al.’s (2007) could be attributed to our data being collected via self-report questionnaire and our population of focus (military), compared to depression assessed in clinic and general Scandinavian population (Gutke et al. 2007). Furthermore, the present study did not include pain, pregnancy, or postpartum specific MSKi questions, only injuries sustained while serving in the CAF were examined. While 43% of medical releases by women from the CAF are MSKi related, 45% are attributed to mental health (Serré 2019; Dursun 2020). Additional screening and psychological support should be incorporated into the postpartum care for CAF members as they may be at additional risk for MSKi.

Our findings indicate associations between miscarriage and some MSKi. It is possible that behaviour and psychological factors are contributing to the relationship between miscarriage and these injuries. Factors such as smoking, lifestyle, and medical conditions increase risk of both outcomes (Qu et al. 2017; Sammito et al. 2021). Another factor to consider is vitamin D deficiency. Females diagnosed with vitamin D deficiency (<50 nmol/L) are at an increased risk of miscarriage compared to females who are vitamin D replete (>75 nmol/L) (Tamblyn et al. 2022). Additionally, vitamin D deficiency is correlated with increased bone fracture risk and the pathophysiology of depression (Vellekkatt and Menon 2019; Jakobsen et al. 2021). To note, in the present study, participants who had experienced miscarriage were more likely to report lower levels of mental health. Stress and depressive symptoms caused by either miscarriage or MSKi may be contributing to the correlation (Franche et al. 2009; Qu et al. 2017; Farren et al. 2020; Rangel et al. 2021; Wong et al. 2021). Specifically, depression is associated with less efficient recovery and greater disability in people with lower back pain and increased complications in patients with traumatic MSKi (Vellekkatt and Menon 2019; Jakobsen et al. 2021; Tamblyn et al. 2022).

Our findings associating pregnancy-related complications and parity status with MSKi demonstrate the need for collaboration between practitioners when supporting the female CAF member. The medical conditions included in this study often require specialized care that can be supported by physical training. Evidence-based physical training programs are effective at decreasing RSI risk in combat trained females and may be an important method of protecting a body that has undergone childbirth from MSKi (Knapik et al. 2003). The rates of CAF members not receiving specialized training guidance during pregnancy (93.4%) and postpartum (92.4%) indicated by our analysis of physical training participation (Fig. 2), and minimal support for exercise or program modifications, initiatives in this area are strongly recommended.

The cross-sectional, retrospective study design of the present study limits the level of evidence and application of our findings. As data pertaining to pregnancy as a risk factor for MSKi are lacking, this questionnaire follows the recommended injury surveillance methods proposed for both sport and military (Meeuwisse 1994; Bahr and Krosshaug 2005; Meeuwisse et al. 2007; Jones et al. 2018). The contributions of childbirth and pregnancy-related complications as possible risk factors of MSKi in the CAF justifies further investigation. Specifically, timing of injuries in relation to pregnancy and the postpartum period, injury types (e.g., carpal tunnel syndrome and diastasis recti), type of delivery (vaginal birth, planned cesarean section, and unplanned cesarean section), and pain conditions (e.g., pelvic girdle pain and low back pain) should all be examined. It is important to note the prevalence of miscarriage in this study only includes females who currently have biological children, participants who answered “no” to having children did not receive questions pertaining to pregnancy related complications. While more information is needed before a causal relationship can be established, clinicians may consider our findings in the care of a female CAF member.