The aim of the study was to broaden the knowledge about help-seeking behavior of postpartum women affected by CB-PTSD and/or gPTSD in order to improve access to appropriate services. For this purpose, differences in the likelihood of and barriers to help-seeking between postpartum omen with CB-PTSD, gPTSD, and women who were not affected by clinically relevant symptoms of these two mental health problems were examined.
The main results can be summarized as follows: there were no group differences in the likelihood of help-seeking among the groups of postpartum women with CB-PTSD, postpartum women with gPTSD, or postpartum women without clinically relevant symptoms of these two mental health problems. In addition, symptom severity of CB-PTSD or gPTSD did not predict the likelihood of help-seeking, but sociodemographic confounders did. An overall group-comparison of the number of possible barriers to help-seeking did not reveal any overall group-difference. However, post-hoc tests revealed that postpartum women affected by CB-PTSD reported more barriers than postpartum women without clinically relevant symptoms of CB-PTSD or gPTSD as shown through post-hoc tests. Additionally, women affected by CB-PTSD and women affected by gPTSD reported more instrumental barriers compared to the non-affected women, and women affected by CB-PTSD reported more barriers regarding fears about treatment and stigmatization than non-affected women.
In our sample, 3.7% of the participants screened positive for CB-PTSD, 2.5% for gPTSD, and 0.4% for comorbid CB-PTSD and gPTSD. The prevalence of CB-PTSD is above the prevalence of 2.6% found in a comparable sample of postpartum women in Germany by Weigl et al. [74]. Contrary to that study, we decided against the necessary presence of the A criterion (i.e., threat of death or serious injury to the mother or baby during labor, birth or immediately afterwards) for group allocation to CB-PTSD. As a result, women who did not report an A criterion but still fulfilled all other relevant symptoms of clinical CB-PTSD (n = 61) were assigned to the CB-PTSD group. In this way, we were able to include more postpartum women, who were distressed due to CB-PTSD symptoms, and for whom access to appropriate services should be improved. Consequently, our CB-PTSD group has less strict inclusion criteria than in Weigl et al. [74] which may have led to the higher prevalence.
Likelihood of Help-Seeking Across the Symptom Groups
Contrary to our hypothesis, the groups of women with CB-PTSD, women with gPTSD, and non-affected women in our sample did not differ in terms of their reported likelihood of help-seeking. It is remarkable that the mean values of the scale assessing the likelihood of help-seeking were relatively high across all groups of postpartum women, because it means that regardless of whether or not women had symptoms of CB-PTSD or gPTSD, they rated their likelihood of help-seeking as relatively high. A high self-reported likelihood of help-seeking has already been found among the general population in Germany [88], when participants were asked how likely they would seek treatment or counseling if they would be (or were) affected by mental health problems.
A possible explanation for the high scores of the likelihood of help-seeking indicated by all postpartum women of our sample and a lack of group differences could be that positive attitudes towards mental health problems and psychotherapeutic treatment have increased substantially among the general population in Germany over the last years [89], thus including postpartum women. Positive attitudes of help-seeking are an important indicator of actual help-seeking for mental health problems, as shown in a study by Mojtabai et al. [90] based on a sample of the general population
However, many women of our sample who were affected by CB-PTSD or gPTSD failed to translate their theoretically high likelihood of help-seeking into actual help-seeking as seen by the low number of women who were actually receiving help. Have et al. [88] suggest that this could be due to low values of the perceived effectiveness of treatment options. In addition, there are other barriers that impede the actual help-seeking process as discussed in a later section. Another explanation could be a lack of self-awareness of their symptoms. Many women who were affected by CB-PTSD or gPTSD did not report a mental health problem at all, although they were assigned to the group of women with CB-PTSD or gPTSD by validated screening instruments. Thus, it can be assumed that many affected women did not recognize their own symptoms as part of a mental health problem leading to no help-seeking behavior. This is consistent with findings of Daehn et al. [41], which indicate that a lack of self-awareness and identification of mental health problems among perinatal women leads to less help-seeking behavior. Similarly, Webb et al. [67] named lacking knowledge about mental health problems as one of the most widespread barriers to help-seeking behavior of postpartum women.
Symptom Severity of CB-PTSD or gPTSD as Predictor of the Likelihood of Help-Seeking
The severity of symptoms of CB-PTSD or gPTSD among postpartum women did not predict the likelihood of help-seeking in our sample. Instead, the likelihood of help-seeking was predicted by maternal age and income for CB-PTSD, and by income for gPTSD.
In a study by Seefeld et al. [62] that focused on postpartum women and their likelihood of help-seeking depending on the severity of symptoms of PPD and/or PAD, it was found that a higher symptom severity of PPD predicted a lower likelihood of help-seeking. Conversely, a higher symptom severity of PAD predicted a higher likelihood of help-seeking. Our results indicate that a similar relationship between the symptom severity of CB-PTSD or gPTSD and the likelihood of help-seeking does not exist. Thus, in order to improve access to treatment and counseling for women affected by symptoms of CB-PTSD or gPTSD, focusing on other aspects besides disorder-specific symptom severity such as the reduction of barriers seems to be crucial as discussed in the later section about barriers to help-seeking.
Role of confounders for likelihood of help-seeking
The mentioned analyses focused on investigating whether group allocation (CB-PTSD, gPTSD, non-affected) or the symptom severity of CB-PTSD or gPTSD predict the likelihood of help-seeking in postpartum women. However, it was not the group allocation that was associated with the likelihood of help-seeking, but the confounders maternal age and income. Older women and women with more income reported a higher likelihood of help-seeking than younger women and women with lower income. This is in line with previous research that focused on PPD and indicated that lower rates of help-seeking were associated with younger age and lower income [56, 91, 92]. The results of our study suggest that this may also apply to postpartum women affected by CB-PTSD or gPTSD. Additionally, higher age and income – not severity of symptoms - predicted a higher likelihood of help-seeking. Similarly, the analyses on symptom severity of gPTSD as a possible predictor of the likelihood of help-seeking revealed that not the severity of symptoms of gPTSD predicted the likelihood of help-seeking, but income did. The association between the likelihood of help-seeking and income found in both analyses is not surprising, as this is in line with results of the general population and their attitudes towards mental health help-seeking – which are more positive in case of higher income [88]. Thus, socio-demographic variables and their association with the likelihood of help-seeking should also be considered in future studies.
Barriers to Help-Seeking
No overall difference in the sum score of barriers across all symptom groups was found. However, post-hoc tests revealed that women who did not report clinically relevant symptoms of CB-PTSD or gPTSD showed fewer barriers to help-seeking than women affected by CB-PTSD. Seefeld et al. [62] found similar results regarding higher barriers in postpartum women with PPD and/or PAD compared to non-affected women. However, postpartum women affected by gPTSD did not report more barriers than non-affected women. Thus, postpartum women with CB-PTSD differed from postpartum women with gPTSD when compared to non-affected women. Postpartum women with CB-PTSD and gPTSD do not differ in terms of the type of symptoms, but they do differ in terms of their traumatic index event [9, 10]. Specific features of the traumatic event childbirth among women with CB-PTSD could account for particular differences in the perception of barriers to help-seeking. Since childbirth is one of very few mental health problems following an event that is socially valued as positive [11], the development of CB-PTSD without medically or objectively complicated childbirth is hard to understand for outsiders. Therefore, women who are affected by CB-PTSD may not feel the same legitimacy of their symptoms compared to women with gPTSD. This is why women with CB-PTSD could be particularly affected by stigma and shame, as described in the section about group differences for the type of barriers.
Prior to that, it should be mentioned that the socio-demographic variables maternal age, duration of residence in Germany, and income could have also affected the number of perceived barriers, as they were not homogeneously distributed across the groups. The descriptive and group-specific mean values of these confounders revealed that women affected by gPTSD had the lowest mean values in all confounders and non-affected women had the highest ones. Hence, non-affected women were on average older, were more likely to be born in Germany as indicated by the mean time of residence in Germany, and reported the highest income across all three groups. The lower scores of women with gPTSD or CB-PTSD could be due to risk factors in the development of these two mental health problems. Risk factors for developing gPTSD in the general population include young age [93], a low socio-economic status [94], and a relatively short period of residence in the respective country also due to potentially traumatic experiences during immigration [95, 96]. Young maternal age and low income are also considered to be risk factors for the development of CB-PTSD [19, 97].
Group Differences for the Type of Barriers
Women affected by CB-PTSD reported more barriers in Subscale (1) Fears about Treatment and Stigmatization than women without clinically relevant symptoms of CB-PTSD or gPTSD. The pattern is different for women with gPTSD, who did not report more barriers in this subscale compared to non-affected women. Thus, women affected by CB-PTSD appear to have specific fears regarding treatment and concerns about stigmatization. Other studies have already shown that stigma and shame are widespread among postpartum women with general mental health problems [59, 68, 98, 99]. As described by Sonnenburg et al. [100] there is an ideal image of motherhood including expectations about fulfilling and self-sacrificing care for the child, which could cause psychological distress when women perceive discrepancies between this image and their own behavior. Thus, many postpartum women affected by mental health problems are afraid of being perceived as a bad mother and experience shame and fears about stigmatization [67]. In case of CB-PTSD, where childbirth is the traumatic index event and therefore the trauma and PTSD symptoms are directly related to the child, women may be even more reluctant to seek help than women with gPTSD, where the traumatic index event is not related to childbirth. This could explain why only women with CB-PTSD reported more barriers in this subscale compared to the group of non-affected women, whereas women with gPTSD did not. Another aspect of the difference regarding this subscale can be explained by results of Kingston et al. [101] that indicate that there is little knowledge about mental health problems of postpartum women among the general population. Even among healthcare professionals, there is little awareness of CB-PTSD [102, 103], so that women affected by CB-PTSD must also be concerned that their service providers will not be able to deal with their symptom adequately. Hence, it can be assumed that stigma is a greater barrier for women with CB-PTSD than it is for women with gPTSD.
Regarding Subscale (2) Health Beliefs, no differences could be identified between women affected by CB-PTSD or gPTSD and non-affected women. One of the items of this subscale (“I would rather talk to friends or family about my problems”) was the highest rated item among all women of our sample, indicating that even non-affected women seem to prefer to talk to their social environment instead of seeking professional services for treatment and counseling, maybe due to the already mentioned overall lacking knowledge about appropriate services and their effectiveness for treating CB-PTSD and other mental health problems.
The results for Subscale (3) Instrumental Barriers showed that women suffering from CB-PTSD or gPTSD reported more instrumental barriers (e.g. lack of time for and lack of childcare during treatment and counseling) than non-affected women. These results could be interpreted as an indication that non-affected women cannot fully empathize with affected women and cannot anticipate instrumental barriers that women with CB-PTSD or gPTSD perceive since they were never really confronted with the need for actual help-seeking.
One aim of the study was to uncover possible differences between women with CB-PTSD and women with gPTSD. In the context of different types of barriers, it can be summarized that no significant group differences were found in the type of barriers between the two subgroups of postpartum PTSD, but group differences between non-affected women and women affected by CB-PTSD or gPTSD were found regarding the type of barriers. Regarding the number of barriers women affected by CB-PTSD or gPTSD reported more barriers than non-affected women. Thus, overcoming barriers, especially instrumental ones, needs to be addressed for both types of postpartum PTSD in general and overcoming barriers related to fears about treatment and stigmatization specifically for CB-PTSD.
Strengths and Limitations
To the best of our knowledge, this is the first study with a large sample comparing the likelihood of and barriers to help-seeking between postpartum women affected by postpartum PTSD and women not affected by clinically relevant PTSD symptoms that considers the two types of postpartum PTSD, namely CB-PTSD and gPTSD. Thanks to a detailed recruitment strategy and a high response rate, our sample is comparable to other samples of postpartum women in Germany and other European countries on a descriptive level [26, 39, 74, 104, 105]. Thus, valuable conclusions can be drawn from our results in order to improve access to appropriate counseling or treatment services of postpartum women who are affected by clinically relevant symptoms of CB-PTSD or gPTSD. Furthermore, the criteria for group allocation to the group of women affected by CB-PTSD allowed us to include women with highly stress causing birth experiences, even if they did not explicitly affirm the A criterion. Another strength of our study is that we primarily examined attitudes and intentions regarding help-seeking by assessing the likelihood of help-seeking. In contrast to simple comparisons of prevalence rates with utilization rates, knowledge about attitudes and intentions enables us to determine basic motivational features of help-seeking. Our study revealed that not attitudes about help-seeking hindered the process of help-seeking among postpartum women affected by clinically relevant symptoms of CB-PTSD or gPTSD, but specific barriers did. These could be reduced through targeted measures (see clinical implications).
Despite the aforementioned strengths of our study, it is important to address some limitations which may restrict the generalizability of our results. First, some requirements for statistical analyses were violated, which may restrict the generalizability of our results: Regarding the two ANCOVAs, residuals were not normally distributed and the covariates were not homogenously distributed across the symptom groups. The latter also applied to the MANCOVA. This underlines the role of the confounders. Furthermore, there was no homogeneity of error variance for Subscale (3) Instrumental Barriers.
Second, data were collected through telephone interviews and thus could be influenced by social desirability [106, 107]. Moreover, this assessment method contributes to a selection bias, as particularly severely affected women may not have had the capacity or resources to conduct a one-hour long telephone interview. Furthermore, results of the reliability analyses were low or questionable regarding the barrier subscales (2) Health Beliefs (α = 0.64) and (3) Instrumental Barriers (α = 0.58). Thus, the results of these scales should be interpreted with caution.
Research Implications
Due to the small size of the group of women who were affected by comorbid symptoms of CB-PTSD and gPTSD, future studies should examine specific characteristics regarding the likelihood of and barriers to help-seeking of postpartum women with comorbid symptoms of CB-PTSD and gPTSD. As CB-PTSD may negatively affect the mother’s and child’s life beyond the period of 6 months postpartum chosen in this study [24–26], it would also be interesting to gain knowledge about the course of symptoms over time through follow-up interviews. Our results suggest that the likelihood of help-seeking and the barriers experienced in the help-seeking process of postpartum women are linked to sociodemographic factors, which is why the relationship between sociodemographic characteristics and help-seeking behavior should be investigated in more detail in the future. A study by Bina et al. [83] on attitudes towards seeking postpartum mental health treatment among a general sample of postpartum women has already argued for the consideration of socio-demographic factors (including age and income). Furthermore, research points to the widely spread comorbidity of CB-PTSD and gPTSD with PPD [108, 109]. That is why comorbid symptoms of these two disorders in the postpartum period and their links to the likelihood of and barriers to help-seeking should be examined. Thus, further studies on postpartum women, their help-seeking behavior, and associations with socio-demographic factors and other mental health problems, especially PPD, should be conducted.
Clinical Implications
The results of our study, particularly the identified disorder-specific barriers in help-seeking behavior of women with CB-PTSD, gPTSD, and women who are not affected by clinically relevant symptoms of these two disorders, imply that a targeted reduction of barriers such as stigma, instrumental factors, and knowledge and awareness gaps regarding CB-PTSD is key to improve access to appropriate services.
This is in line with central statements of a recently published expert review by Horsch et al. [110] that offers various recommendations for the implementation of preventive measures at the level of health care professionals as well as at the level of policy making in order to improve access to treatment and counseling services for postpartum women with CB-PTSD. These measures should include awareness campaigns, e.g. through information events in the context of pregnancy, where knowledge about postpartum mental health problems, and in particular about CB-PTSD or gPTSD, could be spread among perinatal women. This would better prepare them to recognize mental health problems after childbirth. Furthermore, prevention measures should also reach out to healthcare professionals in order to train them to screen for signs of PTSD symptoms after childbirth, e.g. during postnatal check-ups. This allows an early detection of symptoms of CB-PTSD or gPTSD and could reduce the fear of stigmatization as it would be part of routine examinations without the women having to take the initiative to seek help themselves.
Since our study also provides indications that women with low income and women who were not born in Germany reported more barriers to help-seeking than women who have a higher income or were born in Germany, it is important to take this particularly vulnerable group sufficiently into account when implementing appropriate services (e.g. through multilingual services, support for travel costs, etc.). In order to further reduce stigma related barriers, offers that are incorporated in routine postpartum care examinations and integrated screenings for postpartum mental health problems seem to be crucial as also advocated for by Ayers et al. [111]. This should be combined with taking instrumental barriers into account, e.g. through integrated childcare options during maternal service utilization, sufficient offers in rural areas, or financial subsidies.
In addition, health professionals should render the birth as less traumatic as possible. To ensure that they are sufficiently trained for this, educational trainings about CB-PTSD and other postpartum health problems would be useful, as already in practice in some European countries (e.g. France, Netherlands) according to Thomson et al. [112]. With 740,000 childbirths in Germany every year [113], and therefore around 46,000 women affected by any type of postpartum PTSD, investing in preventive measures can avoid high financial burdens for society caused by treatment costs [114].