Obsessive-compulsive disorder (OCD) symptoms during pregnancy and postpartum: prevalence, stability, predictors, and comorbidity with peripartum depression symptoms

Abstract

Background

Pregnancy and postpartum are considered vulnerable periods for new parents to develop obsessive-compulsive disorder (OCD). The aim of this study was threefold: (1) to establish the prevalence of OCD symptoms and its course in the peripartum period; (2) to examine comorbidity with depressive symptoms; and (3) to investigate which sociodemographic, obstetric, and individual characteristics are predictors of OCD symptoms.

Methods

A longitudinal study included 397 women during pregnancy (T1) and 6–12 weeks postpartum (T2). Participants filled out the obstetrical and demographic sheet, Anxiety Sensitivity Index (ASI), Emotional Stability subscale from the International Personality Item Pool-50 (IPIP-50), Brief Resilience Scale (BRS) all at T1, and Yale-Brown Obsessive Compulsive Scale (Y-BOCS) and Edinburgh Postpartum Depression Scale (EPDS) at T1 and T2.

Results

In this sample, 15.1% of women reported OCD symptoms during pregnancy and 15.1% in the postpartum, with 9.8% of women who had symptoms at both time points. However, the majority of women experienced symptoms of mild severity, according to the Y-BOCS. Of the women experiencing OCD symptoms, 33% and 43% had comorbid depressive symptoms in pregnancy and the postpartum period, respectively. The level of OCD symptoms significantly decreased after childbirth. None of the sociodemographic or obstetric variables were a significant predictor of OCD symptoms during pregnancy or postpartum. After controlling for current depression symptoms, higher psychological concerns of anxiety sensitivity (but not physical and social concerns) and higher neuroticism were significant predictors of higher levels of OCD symptoms both at T1 and T2. At the same time, higher resilience was a significant predictor of lower levels of OCD symptoms only at T1.

Conclusion

One in six women has OCD symptoms in the peripartum period, with substantial comorbidity with depression symptoms. Women who are high on neuroticism and anxiety sensitivity are prone to OCD symptoms, while resilience is a significant protective factor.

Clinical trial number

Not applicable.

Background

Pregnancy and the postpartum period represent a challenging phase when individuals are particularly susceptible to experiencing mental health problems [1]. Despite the field of perinatal mental health mostly being focused on peripartum depression (PPD), anxiety disorders are either equally or more prevalent than PPD [2, 3] with substantial comorbidity between them [4]. According to a meta-analysis, one in five women is affected by anxiety disorders in the peripartum period [5], and pregnancy and postpartum are considered vulnerable periods for new parents to develop obsessive-compulsive disorder (OCD) [6]. Obsessions entail intrusive, unwanted, and persistent thoughts, images, or impulses that induce anxiety, while compulsions involve uncontrollable repetitive behaviours or mental rituals aimed at reducing discomfort and anxiety. Although OCD is no longer classified within anxiety disorders but within the separate “obsessive-compulsive and related disorders” category [7], it has a lot of shared characteristics with anxiety disorders [8].

A meta-analysis found prevalence rates of OCD diagnosis to be 2.1% during pregnancy and 2.4% in the postpartum period, which is 1.5-2 times more often than in the general population [9]. A more recent meta-analysis showed a pooled prevalence of OCD diagnosis to be 9.1% during pregnancy and 6.2% in postpartum [10]. However, studies focusing on OCD symptoms reported much higher prevalence rates in the community sample compared to OCD diagnosis [11], with the prevalence rate of OCD symptoms 6% during pregnancy and 14% in the postpartum [12].

Intrusive thoughts are commonly found in parents, and about 80% of mothers and 70% of fathers report some concerns, mainly related to the well-being of the newborn, including checking, cleaning, and hygiene maintenance, which may have a protective purpose from an evolutionary point of view during this period [13]. It is interesting to note that OCD in postpartum was often marked with aggressive thoughts (62%), which were less commonly present in OCD during pregnancy (17%) or in OCD not related to the peripartum period (23%) [14]. Although these aggressive thoughts were mainly about the infant [14], they were not associated with aggressive behaviour toward the infant [15]. It should also be emphasised that obsessive thoughts are intrusive and ego-dystonic by their nature, reflecting worries and concerns which a person wishes to avoid [16], but may be very intimidating for a parent. What distinguishes common preoccupations of new parents and individuals who develop OCD is proposed in a cognitive-behavioural model [6] by (a) abrupt increase of responsibility for vulnerable newborns, (b) misinterpretations of these intrusive thoughts as a sign of threat, and (c) change in behaviour patterns that contribute to the maintenance of the disorder. Additionally, OCD disorder, besides clinical symptoms, is marked with distress and impairment in everyday functioning.

A recent systematic review of the comorbidity of OCD in the general population showed that among adults, the most common comorbid disorders were mood disorders, with major depressive disorders prevailing [17]. In the case of peripartum OCD, comorbidity with other disorders seems to be an especially dominant feature [18]. About 60% of postpartum women with significant levels of OCD symptoms also had significant depressive symptoms [19]. However, another study found a much lower comorbidity of 15% between OCD and depression in women with pregnancy-onset OCD [20]. Therefore, the comorbidity rate between OCD and PPD is yet to be determined.

Regarding the risk factors for OCD, an umbrella review of systematic reviews did not find any strong risk factor associated with OCD [21], but some original studies considered potential environmental factors [22] and perinatal risk factors [23, 24]. However, apart from the cognitive vulnerability for OCD [6], other risk factors for peripartum OCD have been rarely examined, as noted by the review from 2011 [25], and no further extensive studies have been made in this respect. However, the latter review pointed out conflicting findings regarding parity and preterm birth, with some studies showing that primiparous women and those who had preterm birth are at risk of postpartum OCD, but other studies did not find these associations. Of the personality disorders, avoidant and obsessive-compulsive personality disorders were found as risk factors [20, 25], but other personality traits have not been examined in the peripartum period. Pregnancy can be a stressful period due to physical and psychological adaptation, anxiety about childbirth, and the demands of caring for older children, increasing the risk of mental health challenges during and after pregnancy [26]. In stressful and ambiguous situations, personality traits may play a significant role in a stress response, and neuroticism was associated with higher levels of perceived stress in a large meta-analysis [27]. Furthermore, neuroticism and anxiety sensitivity were implied in different anxiety-spectrum disorders [28, 29], but they were not examined in the context of peripartum OCD.

Anxiety sensitivity is characterised as the apprehension towards anxiety-induced symptoms [30]. It encompasses three facets: fear of bodily sensations, fear of cognitive impairment, and concerns regarding public appearances. Anxiety sensitivity has been proposed as a transdiagnostic underlying trait in several anxiety disorders. While higher levels of anxiety sensitivity have been associated with higher levels of OCD symptoms in a clinical sample [31], a meta-analysis revealed that OCD does not have unique associations with anxiety sensitivity dimensions compared to other disorders [29]. Moreover, individuals with OCD did not have higher levels of anxiety sensitivity compared to individuals with mood disorders [29]. However, it is emphasised that the anxiety sensitivity dimension concerning fear of cognitive impairment is implicated explicitly in OCD, which is in line with cognitive-behavioural models [32].

Neuroticism, identified as a tendency to consistently experience negative emotions in response to various stressors, is involved in numerous anxiety and mood disorders [28] and has a shared genetic risk with OCD symptoms [33]. Higher neuroticism was found in adults with OCD [34] but has not been examined in peripartum women, in whom neuroticism might be even more important given that women, in general, have higher levels than men [35].

When considering the development of peripartum OCD, it is crucial to consider not only the risk factors but also protective factors like resilience, which denotes the flexible ability to adapt effectively to stress and adversities [36]. In the general population, lower levels of resilience were associated with higher levels of OCD symptoms [37]. In clinical samples, it was found that individuals with OCD have lower levels of resilience compared to healthy controls and that resilience is important for better mental health outcomes in stressful periods during the COVID-19 pandemic [38] and for the treatment outcome [39]. Although resilience was found to be a protective factor when confronted with stressful situations in pregnant women [40], resilience was not examined in the context of OCD in the peripartum period, to the authors’ best knowledge.

A review on peripartum OCD pointed out the lack of prospective studies in this field and the lack of studies exploring risk factors for OCD [25]. Another review emphasised the importance of investigating not only OCD on a clinical level but also on a subsyndromal level [18]. Finally, there is a lack of evidence about the role of personality traits in OCD, especially in the peripartum population. To fill in the gap in the knowledge, the aim of the study was threefold: (1) to establish the prevalence of OCD symptoms and its course in the peripartum period; (2) to examine the comorbidity of OCD symptoms with depressive symptoms; and (3) to investigate which sociodemographic, obstetric, and individual characteristics are predictors of OCD symptoms. We expected the prevalence rate of OCD symptoms to range between 6 and 14%. We also expected to find substantial comorbidity between OCD and depressive symptoms but could not predict the exact rate given the inconsistencies in the literature. Finally, we expected that higher anxiety sensitivity and neuroticism, and lower resilience would contribute to the higher levels of OCD symptoms. Regarding sociodemographic and obstetric variables, we did not have a firm hypothesis given the insufficient or contradictory findings in the literature.

Methods

Participants and procedure

This study was a part of the longitudinal study in the peripartum period. The research received approval from the Ethics Committees of both the Catholic University of Croatia and the Clinical Hospital “Sveti Duh” (Zagreb, Croatia) in accordance with the Declaration of Helsinki. A convenient sample of pregnant women was approached during the regular prenatal check-ups (T1) at the maternity ward by their obstetrician. All pregnant women are encouraged to have at least one routine prenatal check-up at the maternity ward where they plan to give birth. Inclusion criteria were: (1) being 18 years old or older, (2) being pregnant in the second or third trimester of pregnancy, and (3) understanding the Croatian language. No exclusion criteria were applied. After signing informed consent forms, women completed questionnaires and returned them in sealed envelopes. Between 6 and 12 weeks postpartum (T2), women were contacted via email and/or text message, receiving a Google Forms link to an online questionnaire based on their due date. Participation was voluntary, and participants could withdraw from the study at any time without explanation.

At T1, we approached 573 women, with 46 (8.0%) failing to provide contact details. By T2, we reached out to 527 women, of whom 412 successfully completed online questionnaires. However, eight respondents fell outside the 6–12-week timeframe, and seven were excluded due to missing data on key variables. Consequently, our final sample comprised 397 women, accounting for 69.3% of the initial group approached. In comparison to those who did not complete the study, the women who participated at both time points were notably older (t(565) = -2.34, p =.020), more educated (χ²(2) = 16.67, p <.001), and resided in more urban areas (χ²(2) = 14.20, p =.001). However, there were no significant differences between the two groups in terms of parity, relationship status, OCD or PPD symptoms at T1, neuroticism, anxiety sensitivity, or resilience.

Women participated on average at 35.86 (Sd = 3.91; range 16–41) weeks of pregnancy at T1 and 7.86 (Sd = 1.64) weeks after childbirth at T2. In the final sample, women were, on average, 31.75 years old (Sd = 4.96, range 19–47), mostly married or cohabiting, highly educated, of average or above average socioeconomic status, and lived in urban areas (Table 1). The great majority (90.4%) denoted that they belonged to an ethnic/racial majority (White/Caucasian). Regarding mental health problems, only a minority reported having a previous diagnosis. For the majority, this was a planned pregnancy. Half of the women were primiparous and the majority of them gave birth vaginally. Only a few infants were born preterm. A minority of the infants had health problems, as reported by mothers, while around a quarter of the mothers had some medical complications.

Table 1.

Sociodemographic data of the sample (N = 397)

		n (%)
Marital status a	Married	298 (75.6)
	Cohabiting	88 (22.3)
	Other	8 (2.1)
Education a	Elementary or secondary school	108 (27.2)
	College or university	287 (72.8)
Socioeconomic status a	Below average	7 (1.7)
	Average	240 (61.1)
	Above average	146 (37.2)
Place of residence a	City (> 100,000 citizens)	317 (80.3)
	City (< 100,000 citizens)	40 (10.1)
	Rural area	38 (9.6)
Previous diagnosis of mental health problem a	Yes	14 (3.5)
	No	375 (94.7)
	I do not know	7 (1.8)
Parity	Primiparous	220 (55.8)
	Multiparous	174 (44.2)
Pregnancy plans	Planned pregnancy	251 (63.2)
	Unplanned pregnancy, but wanted	143 (36.0)
	Unplanned and unwanted pregnancy	3 (0.8)
Preterm birth (before 37 weeks of gestation)	No	389 (98.0)
	Yes	8 (2.0)
Type of birth	Vaginal	308 (77.6)
	Vaginal instrumental	1 (0.3)
	Emergency caesarean section	55 (13.8)
	Planned caesarean section	33 (8.3)
Medical complications for the baby	No complications	362 (91.2)
	Minor complications	30 (7.5)
	Major complications	5 (1.3)
Medical complications for the mother	No complications	288 (72.5)
	Minor complications	102 (25.7)
	Major complications	7 (1.8)

Note: ^a there are some missing values at T1 so the sum does not add to 397 in all cells

Instruments

The Yale-Brown Obsessive Compulsive Scale (Y-BOCS) [41] is a 10-item scale designed to measure the intensity of obsessive and compulsive symptoms during the last week in individuals with OCD, and was used as a self-report measure of the OCD symptoms as the primary outcome. Five items assess obsessions, and five items assess compulsions, but only the total score is calculated. Total scale scores range from 0 to 40, and higher results indicate more intense OCD symptoms. A score of 7 or less indicates subclinical levels, 8–15 indicates probably minor, 16–23 moderate, 24–31 severe, and 32–40 extremely severe OCD symptoms. The scale was translated to Croatian, and the validation showed that the total score can be computed as a linear combination of all ten items with high reliability of the scale [42]. In the current study, the total score was calculated with McDonald’s ω of 0.88 and 0.91 during pregnancy and postpartum, respectively.

The Anxiety Sensitivity Index (ASI) [43] assesses apprehension towards anxiety symptoms and sensations. It consists of 16 items distributed among three domains: Physical Concerns (exemplary item: Unusual body sensations scare me.), Psychological Concerns (exemplary item: When I am nervous, I worry that I might be mentally ill.), and Social Concerns (exemplary item: Other people notice when I feel shaky.). Respondents rate each item on a 5-point scale, yielding a total score ranging from 0 to 64, where higher scores indicate greater anxiety sensitivity. The scale has been translated into Croatian and validated, demonstrating a Cronbach’s α of 0.88 for the total scale [44]. In this study, McDonald’s ω was 0.88 and 0.81 for Physical and Psychological Concerns. Cronbach’s α for Social Concerns was only 0.43, so it was omitted from the subsequent analysis.

The International Personality Item Pool-50 (IPIP-50) [45] evaluates the Big Five personality traits: extraversion, emotional stability, conscientiousness, agreeableness, and intellect. Developed in Croatia, the IPIP-50 consists of ten items per subscale, each rated on a 5-point scale. In this study, only the emotional stability subscale was utilised (exemplary item: Get stressed out easily.), and the scoring was reversed, resulting in scores ranging from 10 to 50, where higher scores reflect greater neuroticism. In the original study, Cronbach’s α was 0.87, and McDonald’s ω in the current study was 0.89.

The Brief Resilience Scale (BRS) [46] is a 6-item instrument designed to assess resilience to stress, reflecting an individual’s ability to withstand or rebound from stressors (exemplary item: I tend to bounce back quickly after hard times.). Participants rate items on a 5-point scale, with three items reverse-coded. The total score, calculated as the average value, indicates higher resilience with a higher score. This scale has been translated into Croatian and demonstrated a Cronbach’s α of 0.82 [47]. In this study, McDonald’s ω was 0.84.

The Edinburgh Postnatal Depression Scale (EPDS) [48] is a 10-item scale to evaluate depressive symptoms over the last week (exemplary item: I have felt sad or miserable.). Each item has different answers rated from 0 to 3, which produces a total score ranging from 0 to 30, where a higher score indicates a higher level of depressive symptoms. The scale is valid for use both during pregnancy and postpartum [49]. The scale was translated and validated in the Croatian peripartum population [50]. A cut-off score of 13 for higher specificity was established in a large individual participant data meta-analysis [49]. In the current study, McDonald’s ω was 0.85 and 0.89 during pregnancy and postpartum, respectively.

The demographic sheet comprised questions on maternal age, relationship status, education, socioeconomic level, place of living, ethnicity/racial affiliation (Ethnic/racial majority (e.g., White in Croatia), Ethnic/racial minority, and Not sure) and previous mental health problems (with options Yes, No, and I do not know). Obstetric questions were about gestational age, parity (at T1), type of birth, gestational age at birth, and medical complications for the baby and the mother during pregnancy or childbirth (at T2).

Statistical analyses

If a participant responded to at least 70% of the items within each scale at T1, total scale scores were computed by mean imputation; however, at T2, there were no missing data due to the settings of Google Forms. Utilising G*Power [51] and considering a medium effect size, a significance level of 5%, a power of 80%, and up to 10 predictors, it was determined that a minimum of 118 participants would be required, a threshold that was surpassed. Scale reliability was assessed using McDonald’s ω coefficient. Descriptive statistics were computed, and associations were explored using Pearson r and Spearman rank coefficients. Differences between means were calculated by one-way ANOVA. The prediction of OCD symptoms was investigated through hierarchical regression analysis (enter method), with sociodemographic and obstetric variables significantly associated with OCD symptoms in the first step, current depression symptoms in the second step, and individual psychological characteristics in the third step. All assumptions were inspected, including multicollinearity. All statistical analyses were performed using IBM SPSS software for Windows version 29.0.

Results

Prevalence of OCD symptoms in peripartum

In this sample, 60/397 women (15.1%) reported OCD symptoms during pregnancy, and 60/397 women (15.1%) reported symptoms in the postpartum. However, these were not all the same women. Of the 60 women, 39 of them had OCD symptoms at both time points, which makes 9.8% of the whole sample or 65% of women reporting OCD symptoms either during pregnancy or postpartum. Of the women reporting OCD symptoms, most were minor symptoms, few of them were moderate or severe, and none were extreme (Table 2).

Table 2.

Prevalence of OCD symptoms during pregnancy (T1) and postpartum (T2) in 397 women

OCD symptoms	Result on Y-BOCS	Pregnancy n (%)	Postpartum n (%)
Subclinical level	0–7	337 (84.9)	337 (84.9)
Minor symptoms	8–15	50 (12.6)	52 (13.1)
Moderate symptoms	16–23	8 (2.0)	5 (1.2)
Severe symptoms	24–31	2 (0.5)	3 (0.8)
Extreme symptoms	32–40	0 (0)	0 (0)

Note: Y-BOCS– Yale-Brown Obsessive Compulsive Scale

Course and stability of the OCD symptoms in peripartum

When comparing mean levels of OCD symptoms from T1 to T2, there was a small but significant decrease in the level of symptoms after childbirth (t(396) = 3.22, p <.001). On the other hand, the correlation between OCD symptoms across the two time points was moderate and significant over the peripartum period (r =.67, p <.001), indicating that women who reported higher levels of OCD symptoms during pregnancy also reported higher levels of symptoms after childbirth (Table 3).

Table 3.

Correlations between personality and OCD symptoms in peripartum women (N = 397)

	OCD symptoms T1	OCD symptoms T2	Depression symptoms T1	Depression symptoms T1	ASI Total score	ASI Physical Concerns	ASI Psychological Concerns	Neuroticism	Resilience
1. OCD symptoms T1	-	0.67**	0.53**	0.40**	0.44**	0.36**	0.50**	0.53**	− 0.41**
2. OCD symptoms T2		-	0.44**	0.55**	0.36**	0.27**	0.43**	0.47**	− 0.33**
Depression symptoms T1			-	0.51**	0.41**	0.33**	0.48**	0.64**	− 0.37**
Depression symptoms T2				-	0.37**	0.32**	0.38**	0.51**	− 0.30**
3. ASI Total score					-	0.93**	0.84**	0.47**	− 0.43**
4. ASI Physical Concerns						-	0.63**	0.39**	− 0.38**
5. ASI Psychological Concerns							-	0.53**	− 0.44**
6. Neuroticism								-	− 0.49**
7. Resilience									-
M  (Sd)	3.81 (4.24)	3.23 (4.67)	7.03 (4.45)	5.88 (4.96)	16.77 (9.43)	7.99 (6.08)	4.43 (3.56)	25.08 (6.61)	3.42 (6.62)

Note: ASI– Anxiety Sensitivity Index; * p <.05, ** p <.01

Comorbidity of the OCD symptoms with peripartum depression symptoms

In the current sample, using the cut-off score of 13 on the EPDS scale for peripartum depression, 45 women (11.3%) reported PPD symptoms during pregnancy and 45 women (11.3%) in the postpartum. Of the women reporting OCD symptoms during pregnancy, 33.3% reported PPD symptoms (20/60). On the other hand, of the women reporting PPD symptoms during pregnancy, 44.4% reported OCD symptoms (20/45). After childbirth, of the women reporting OCD symptoms, 43.3% reported PPD symptoms (26/60). On the other hand, of the women reporting PPD symptoms, 57.8% reported OCD symptoms (26/45). Women with PPD symptoms had more comorbid OCD symptoms than vice versa both during pregnancy (χ²(1) = 10.28, p <.01) and after childbirth (χ²(1) = 16.68, p <.001).

Correlates and predictors of the OCD symptoms in peripartum

None of the demographic variables (age, education, employment status, place of living, socioeconomic status) were associated with OCD symptoms during pregnancy or after childbirth, except for relationship status (r_S= 0.12, p <.05), where women who were married reported lower levels of OCD symptoms in pregnancy, but the correlations were low. This was further examined by one-way ANOVA to compare married women, women living with a partner and women in other relationship constellations. This revealed that there was no effect of marital status and no significant differences between the groups (F (2, 394) = 2.90, p =.056), and was, therefore, not further discussed.

Regarding obstetric variables, parity and type of birth were not associated with OCD symptoms. However, higher levels of OCD symptoms after childbirth were associated with preterm birth (r_S= 0.10, p <.05) and medical complications of the baby (r_S= 0.13, p <.05), although both correlations were very low. Higher levels of neuroticism, Physical Concerns and Psychological Concerns, and lower levels of resilience were associated with higher levels of OCD symptoms during pregnancy and postpartum (Table 3). Furthermore, higher levels of OCD symptoms were moderately associated with higher levels of depression symptoms during pregnancy and postpartum. Also, depression symptoms were in low to moderate correlations with personal characteristics and were, therefore, controlled for in regression analyses.

Two hierarchical regression analyses were conducted to examine the predictors of OCD symptoms during pregnancy and postpartum (Table 4). In the first step, demographic (relationship status for OCD symptoms during pregnancy) and obstetric variables (preterm birth, baby’s medical complications) were entered in the first step, which accounted for 1–3% of the OCD symptoms. In the second step, current depression symptoms were entered, which explained about 27% of the OCD symptoms both in pregnancy and postpartum. In the third step, personality traits, including anxiety sensitivity, neuroticism, and resilience, were included, which explained an additional 8–11% of the OCD symptoms. Of different anxiety sensitivity dimensions, only the Psychological Concerns dimension was a significant predictor of OCD symptoms both during pregnancy and postpartum. Neuroticism was a significant predictor of OCD symptoms both during pregnancy and after childbirth, while resilience was a significant predictor of OCD symptoms only during pregnancy. After controlling for depression symptoms, higher levels of anxiety sensitivity and neuroticism and lower levels of resilience contributed to the higher levels of OCD symptoms.

Table 4.

Prediction of OCD symptoms based on personality in peripartum women (N = 397)

	OCD symptoms during pregnancy				OCD symptoms after childbirth
Predictors	β	t	p	Predictors	β	t	p
Step 1				Step 1
Constant		4.47	< 0.001	Constant		12.42	< 0.001
Relationship status a	0.12	2.32	0.021	Preterm birth b	0.13	2.60	0.010
-				Baby’s complications c	0.10	1.93	0.054
	R2 = 0.013; F(1, 393) = 5.36; p = 0.021				R2 = 0.032; F(2, 392) = 6.39; p = 0.002
Step 2				Step 2
Constant		−0.26	0.794	Constant		0.62	0.535
Relationship status a	0.04	0.995	0.320	Preterm birth b	0.05	1.24	0.217
-				Baby’s complications c	0.05	1.23	0.218
Depression symptoms at T1	0.52	12.06	< 0.001	Depression symptoms at T2	0.53	12.38	< 0.001
	R2 = 0.281; Δ R2 = 0.268; F(2, 392) = 76.42; p < 0.001				R2 = 0.304; Δ R2 = 0.272; F(3, 391) = 56.97; p < 0.001
Step 3				Step 3
Constant		0.18	0.854	Constant		−0.06	0.955
Relationship status a	0.06	1.44	0.150	Preterm birth b	0.06	1.43	0.152
-				Baby’s complications c	0.06	1.54	0.124
Depression symptoms at T1	0.25	4.69	< 0.001	Depression symptoms at T2	0.37	7.80	< 0.001
ASI Physical Concerns	0.02	0.37	0.710	ASI Physical Concerns	−0.07	−1.24	0.216
ASI Psychological Concerns	0.21	3.75	< 0.001	ASI Psychological Concerns	0.22	3.84	< 0.001
Neuroticism	0.19	3.25	0.001	Neuroticism	0.14	2.52	0.012
Resilience	−0.13	−2.69	0.007	Resilience	−0.08	−1.74	0.082
	R2 = 0.399; Δ R2 = 0.118; F(6, 388) = 43.00; p < 0.001				R2 = 0.386; Δ R2 = 0.082; F(7, 387) = 34.74; p < 0.001

Note: ^a Relationship status: 0-married, 1-cohabiting, 2-other; ^b Preterm birth: 0-no, 1-yes; ^c Baby’s complications: 0-no, 1-minor, 2-major; ASI– Anxiety Sensitivity Index

Discussion

Studies on peripartum mental health are still mainly focused on PPD, while studies on disorders from the anxiety spectrum are insufficient. To fill in the gap, this study focused on OCD symptoms, trying to establish its prevalence, comorbidity with PPD symptoms, and predictors over the peripartum period. The main findings showed that 15.1% of women reported OCD symptoms either during pregnancy or after childbirth, with a substantial overlap with PPD symptoms. Also, the mean level of OCD symptoms significantly decreased after childbirth. Significant predictors of OCD symptoms were established, specifically psychological concerns of anxiety sensitivity, neuroticism, and resilience. These results are discussed further.

One in six women, or 15.1% of women, reported OCD symptoms either during pregnancy or postpartum, which reflects point prevalence, rather than lifetime prevalence. Direct comparisons with other studies might be difficult because of different methodology (interview for establishing clinical diagnosis, use of self-report questionnaires to assess symptom intensity, or checklists for symptom presentation). Also, different scales were used in the peripartum samples with different cut-off scores. However, the prevalence established in our study is a somewhat higher percentage of OCD symptoms than the one depicted in the literature, with 6% during pregnancy and 14% postpartum [12] It should be highlighted that, in the current study, most women had minor symptoms, while moderate and severe symptoms were reported by 2.5% of women during pregnancy and 2.0% in postpartum.

It is interesting to note that of the women with OCD symptoms, for one in three women, these symptoms were present at only one time point, either during pregnancy or after childbirth, and thus probably transient. However, for two in three women, the symptoms were present at both time points. To add, although the level of OCD symptoms decreased after childbirth, the association between the OCD symptoms during pregnancy and after childbirth was moderate and showed stability. Although these symptoms were mild or moderate in most women, with none identified as extreme, they should not be ignored. These thoughts, often marked with aggressive content of harm to the infant [14], can be especially disturbing for the mother. However, it needs to be highlighted that these thoughts are ego-dystonic and reflect concerns and fears that a person wants to avoid [16] and they are not followed by aggressive behaviour toward the infant [15] but rather more avoidant behaviour, self-assurance, and cognitive and behavioural distraction [52]. One particular manifestation of OCD with aggressive thoughts is paedophilia-themed OCD, where a person experiences distressing and intrusive thoughts about sexually harming the child [53]. In some instances, these thoughts may cause the parent to avoid routine caregiving tasks, such as changing their child’s diapers, out of fear of inappropriate physical contact. Harm-related obsessive thoughts should also be distinguished from psychotic infanticidal ideas, where obsessive thoughts are ego-dystonic, unwanted, anxiety-provoking and lead to protective behaviours and avoidance [54, 55].

Regarding the comorbidity between OCD and PPD symptoms, this issue is essential, given that OCD severity increases when depressive symptoms are present [50]. In the current sample, the overlap of OCD symptoms with depressive symptoms was 33% during pregnancy and 43% after childbirth, putting the level of overlap somewhere between the previously reported 15% [20] and 60% [19]. What is more interesting is that there was an even more extensive overlap of PPD symptoms with OCD symptoms (44-57%), indicating that there might be distinctive subtypes of predominant OCD symptoms and predominant depressive symptoms with some obsessions and compulsions as integral parts of depressive symptomatology specific for the peripartum period. The latter is reflected in the recent literature on PPD manifesting with obsessive thoughts [56] and a recently developed new measure for peripartum depression comprising obsessive thoughts [57]. However, it remains open to debate whether OCD in peripartum is different to OCD in other time periods [58].

Although all aspects of anxiety sensitivity were associated with OCD symptoms on a bivariate level, only psychological concerns or fear of cognitive dyscontrol were associated with OCD symptoms both during pregnancy and after childbirth, after controlling for depression and other variables. This finding is difficult to compare as there are no studies examining anxiety sensitivity in the peripartum period. However, it is in line with studies on non-peripartum populations showing the association between anxiety sensitivity and OCD symptoms in clinical samples of women and men [31]. Also, more specifically regarding the dimension of psychological concerns, our findings align with the study by Wheaton et al. [59], which showed that psychological concerns were associated with difficulties with unacceptable thoughts in a non-clinical sample of students. People with OCD often interpret their inability to manage intrusive thoughts as a sign of mental fragility. Moreover, those experiencing frequent unacceptable thoughts may fear that their perceived mental vulnerability and diminished cognitive control could lead to a loss of control over their actions [60]. It is important to take anxiety sensitivity into account, given that higher levels of pre-treatment anxiety sensitivity may hinder the effects of cognitive-behaviour therapy for OCD [61].

Furthermore, the other trait found to be predictive of OCD symptoms both during pregnancy and postpartum was neuroticism. This finding is a valuable addition to the literature showing the association between neuroticism and OCD in the non-peripartum population [34], given that previous studies did not examine this association in the peripartum population. Neuroticism, the personality trait implicated in different forms of psychopathology [62], predisposes a person to experience negative emotions in response to various stressors, which are commonly presented in the peripartum period [57, 63]. Ormel et al. [64] highlighted neuroticism as the single most important risk factor for different types of psychopathologies, and it was confirmed as a vulnerability factor for OCD during peripartum in the current study.

In contrast with anxiety sensitivity and neuroticism as risk factors, our results showed that resilience was a significant protective factor in the context of OCD symptoms during pregnancy. Although resilience was not a significant predictor of OCD symptoms postpartum when other variables were also in the model, they were associated prospectively on a bivariate level. In other words, women with higher levels of resilience during pregnancy reported lower levels of OCD symptoms postpartum. These findings are the first report on this association in the peripartum population but align with previous studies in the non-peripartum population showing a negative association between resilience and OCD symptoms [37]. A systematic review of coping strategies during pregnancy highlighted how effective coping methods foster resilience in peripartum individuals by mitigating stress and enhancing both maternal and infant health outcomes [65]. Interventions focused on fostering resilience were found to be applicable and effective for PPD and posttraumatic stress disorder [66, 67]; however, effective programs for OCD are yet to be established.

Of obstetric factors associated with OCD symptoms, it is noteworthy to mention preterm birth, which was associated with higher levels of OCD symptoms postpartum, but not when depression and individual characteristics were taken into account. Maternal OCD is an adverse risk factor for several neonatal outcomes, among which is giving preterm birth [68]. The experience of preterm childbirth is generally a risk factor for adverse maternal mental health, especially posttraumatic stress disorder following childbirth [69]. Parents of newborns admitted to neonatal units are at higher risk for anxiety and posttraumatic stress compared to the general peripartum population [70]. Most mothers of preterm newborns reported fear for the baby’s life and expected more difficulties in the future, especially those with very preterm newborns, as compared to mothers with full-term newborns [71]. Parental concerns about the newborn’s well-being may translate into a compulsion of checking to reassure that the child is well and reduce anxiety. This obsessive-compulsive behaviour can be explained as an evolutionary response to threat, and early parenthood is a time of vulnerability [13], especially if there is also a heightened level of risk, such as in the case of preterm birth. It is interesting to note that preterm birth, among other perinatal factors, is a risk factor for developing OCD in adolescence and adulthood [23, 24]. So, it seems that the association between OCD and preterm birth is complex.

Several limitations of the study should be acknowledged. First, the study was conducted on a community peripartum sample and was not clinical. Also, due to dropout in the follow-up assessment after childbirth, it should be noted that the final sample was somewhat older, more educated, and lived in urban areas more often compared to the initial sample. Also, the sample was predominantly married/cohabiting with high socioeconomic status and may not represent the general population. Therefore, future studies should aim to recruit a more heterogeneous sample, while a clinical sample would provide better insight into OCD features during the peripartum period. The Y-BOCS was used as a self-reported measure, and a clinical interview was not administered, so we do not have exact data on the OCD symptoms onset and if the symptoms started in the peripartum or before. Also, as this was a secondary analysis of the longitudinal project on peripartum mental health, assessment of the OCD presentation was limited to the use of the Y-BOCS measuring the intensity of the obsessive and compulsive symptoms; however, we did not collect additional information on the specific content of these obsessions and compulsions through symptom checklist. Also, we did not include details on pre-existing medical complications, such as polycystic ovary syndrome found as a risk factor for OCD in general population in a meta-analysis [72], which might be included in further studies. Nevertheless, the study overcomes the gap in the literature mainly in the aspect of examining personality traits and individual characteristics for OCD symptoms in the peripartum population prospectively.

Conclusion

To summarise, one in six women experience OCD symptoms during pregnancy and postpartum, with moderate stability over this period and substantial overlap with depression symptoms. Individual characteristics that are shown as risk factors in the development of OCD were the psychological aspect of anxiety sensitivity and neuroticism, while resilience was shown to be a protective factor during pregnancy. These psychological risk factors could ease the identification of women at risk for the development of OCD symptoms in peripartum and offer them prevention programs. Although a study from 2011 [73] demonstrated the effectiveness of the prevention program for postpartum OCD based on cognitive-behaviour therapy (CBT) incorporated in prenatal classes, it was the only prevention program found in a recent systematic review [74], which generally showed effect on anxiety symptom decrease, with CBT-based programs as most effective. A very recent open-trial of a prevention program delivered over the Internet in a small sample showed initial acceptability and feasibility [75]. However, further efforts are needed to investigate effective strategies to prevent the onset or alleviate the OCD symptoms. Given the considerable percentage of women experiencing OCD symptoms, screening for peripartum mental health issues should not be restricted to PPD only but expanded to other mental health problems from the anxiety spectrum that are as prevalent [2, 5], and OCD symptomatology should not be overlooked. It is important to identify these women in a timely manner and offer them adequate and effective treatment. In this aspect, cognitive-behavioural therapy has the most evidence of its effectiveness, while psychopharmacological options still need to be proven in high-quality studies [76, 77]. Future studies should look more in-depth at the aetiology of OCD in peripartum and examine whether OCD in peripartum is distinctive to OCD in the non-peripartum period. Also, it would be necessary to investigate whether peripartum OCD is homogeneous or has some subtypes, especially if comorbidity with other peripartum mental health problems is taken into account.

Abbreviations

ASI

Anxiety Sensitivity Index

BRS

Brief Resilience Scale

CBT

cognitive-behaviour therapy

EPDS

Edinburgh Postnatal Depression Scale

IPIP

50-International Personality Item Pool-50

OCD

Obsessive-compulsive disorder

PPD

Peripartum depression

Y

BOCS-Yale-Brown Obsessive Compulsive Scale

Author contributions

Conceptualisation: SNR, MB; Data curation: MM; Formal analysis: SNR; Funding acquisition: SNR; Investigation: SNR, MB, MŽ, MM; Methodology: SNR, MB, MŽ, MM; Project administration: SNR; Supervision: SNR; Writing original draft: SNR; Writing review & editing: MB, MM, MŽ. All authors read and approved the final manuscript.

Funding

This study was funded and supported by an approved research project of the Catholic University of Croatia: “Determinants, outcomes, and interrelation of mental and physical health during pregnancy and postpartum (MumHealth)“. MŽ was supported by the Croatian Science Foundation grant DOK-2020-01-4127. MM was supported in part by the Croatian Science Foundation under the project number HRZZ- MOBODL-2023-12-6514. Funding sources had no other role other than financial support.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by the Ethics Committee of the Catholic University of Croatia (Class: 641-03/21 − 03/21; No: 498 − 16/2-22-04) and the Ethics Committee of the Clinical Hospital “Sveti Duh” (No: 012-1539).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.