Abstract
Purpose:
The purpose of this study was to examine associations between postpartum obsessive-compulsive (OC) symptoms, infant care and feeding worries, and breastfeeding experiences in a sample of postpartum women.
Materials and Methods:
Women were recruited via an online U.S. research volunteer database and were eligible if they were aged 18–47 years, their primary language was English, they had an infant 2–6 months old, and tried breastfeeding the infant at least once. Participants completed a survey to assess breastfeeding experiences and practices, OC symptoms and other mental health conditions, and demographics. Modified Poisson regression and linear regression were used to estimate associations between clinically elevated OC symptoms, reported thoughts or worries about infant feeding and care, and breastfeeding experiences, problems, and duration.
Results:
Of 232 participants, 32 (14%) had clinically elevated OC symptoms. These women had more perinatal OC symptoms (scoring 3.6 points higher on perinatal OC symptoms score [95% confidence interval {CI}: 0.4 to 6.9]), including symptoms specific to infant care and feeding [e.g., adjusted relative risk(repeated washing of baby's bottles, bowl, or plate) = 2.37, 95% CI:1.55 to 3.64], and more breastfeeding problems (adjusted β = 0.3, 95% CI: 0.0 to 0.2) than women with fewer OC symptoms. However, they did not report an overall worse breastfeeding experience (adjusted β = 0.4, 95% CI: −9.3 to 10.1). Adjusted models controlled for depressive symptoms.
Conclusions:
Obsessive thoughts and compulsions were common in this sample and extended to infant feeding activities. These symptoms were associated with more breastfeeding problems. Interactions in mother–infant dyads are critical because of lasting impacts on parent–child relationships and child development. Treating OC symptoms may foster healthier mother–infant relationships.
Keywords: obsessive-compulsive disorder, postpartum, breastfeeding, anxiety
Introduction
Obsessive-Compulsive Disorder (OCD) involves obsessions (intrusive, recurrent, and persistent thoughts) and/or compulsions (repetitive behaviors) that are time-consuming and distressing.1 The postpartum period is a particularly vulnerable time for the development of clinically significant obsessive-compulsive (OC) symptoms; in fact, approximately 0.7–2.7% of postpartum women have OCD.2,3 Subclinical OC symptoms are common as well, with one large U.S. study (N = 461) reporting rates of mild OC symptoms in ∼10% of women at 2 and 6 weeks of postpartum.4 Postpartum symptoms may be triggered by the dramatic decrease of estrogen and the resulting lack of serotonin interacting with other predisposing factors such as poor sleep, fatigue, and cognitive vulnerability due to dysfunctional OCD-related beliefs.5–11
Postpartum OC symptoms often center around fears of harming, misplacing, or contaminating the infant and fear of sudden infant death syndrome (SIDS).12 Mothers may react to their fears by avoiding or modifying their behavior in relation to their infants.9 In fact, women with postpartum OCD perceive their parenting to be negatively affected, especially their ability to enjoy and have fun with their child.13
Because OC symptoms can have pervasive negative impacts on a wide range of activities of daily living and mothers with elevated symptoms may modify their parenting behaviors, it is likely that elevated symptoms would have measurable effects that carry over to infant feeding. In a 2005 study (n = 87), women with higher scores on the Brief Symptoms Inventory obsessive compulsiveness subscale tended to adopt more restrictive feeding practices with their 1-year-old children based on the Child Feeding Questionnaire.14 This could lead to altered adiposity long-term.15
However, whether postpartum OC symptoms adversely affect breastfeeding and related infant care practices has not been thoroughly examined. Impeded breastfeeding could interfere with bonding both hormonally or psychologically, making women feel less confident in their roles as mothers. It would also mean that women and their children would be less able to gain the benefits of breastfeeding.16 A recent study reported that women with a postpartum OCD diagnosis were less likely to breastfeed at 6 months postpartum compared with a community control group (27% versus 57%); however, additional details were not reported.16
OC symptoms frequently are accompanied by elevated depressive symptoms, and postpartum depressive symptoms have been shown to be associated with reduced breastfeeding duration, lower breastfeeding confidence, and more breastfeeding problems.17–19 Thus, it is important to control for depressive symptoms to focus on the potential effect of OC symptoms on breastfeeding outcomes. Therefore, the objective of this study was to examine associations between postpartum OC symptoms, breastfeeding, and infant care experiences in a sample of U.S. women with infants.
We hypothesized that women with clinically elevated OC symptoms would be more likely to report OC symptoms specific to infant care and feeding, a poorer overall breastfeeding experience, lower breastfeeding self-efficacy, more breastfeeding problems, and shorter breastfeeding duration compared with women with lower OC symptoms, after controlling for depressive symptoms.
Materials and Methods
Recruitment and eligibility
All women 18–47 years of age who were registered as general population research volunteers on the National Institutes of Health-sponsored Research Match website (https://www.researchmatch.org) were invited through email to participate in the Life After Pregnancy Study (LAPS) between May and October 2016. The study was advertised as not only being about infant feeding but also other psychosocial aspects of motherhood and postpartum health. The invitation was also posted on several Facebook communities for mothers and on electronic message boards at a major children's hospital.
Advertising for LAPS did not specifically target participants with any particular physical or mental health conditions or infant feeding or care issues. Eligible participants were women whose primary language was English, aged 18 years or older, had an infant 2–6 months old, completed the Obsessive-Compulsive Inventory–Revised (OCI-R) as part of the study, and had tried breastfeeding their infant at least once.20 Women who completed the survey were eligible to receive one of four $100 gift cards as an incentive. This study was reviewed and approved by the institutional review board at Nationwide Children's Hospital.
Data collection
The study invitation linked to an online REDCap survey, a research survey hosting service with a local secure database.21 When participants opened the survey, informed consent was first obtained, and then, participants were screened for eligibility. The 45-minute survey assessed psychosocial well-being, mental health symptoms, lactation and infant feeding history up to the time of the study, use of breastfeeding support services, personal evaluation of breastfeeding experiences, and demographic and other personal characteristics. Among the questions about psychosocial well-being, women were asked questions about their worries and thoughts about being a mother. Women completed the instruments described below as part of the survey.
Obsessive-Compulsive Inventory–Revised
This is an 18-item self-report instrument that measures OC symptoms. The OCI-R is divided into six subscales: washing, checking, ordering, obsessing, hoarding, and neutralizing, with items presented on a Likert scale ranging 0–4 (“not at all” to “extremely”).20 Items are summed to form a total score with a possible range of 0–72. Scores at or above 21 are typically considered compatible with a diagnosis of OCD, but because of the self-reported nature of our data collection, we instead considered women who scored at or above 21 to have clinically elevated symptoms of OCD, whereas women who scored below 21 were considered to not have met the published threshold for clinically elevated OC symptoms.22
Spielberger State–Trait Anxiety Inventory
State (current) anxiety was assessed with the Spielberger State–Trait Anxiety Inventory (STAI).23 State anxiety scores were calculated by summing 20 STAI items marked on a 1–4 scale for “not at all” to “very much so” for each feeling indicative of anxiety, with higher scores indicating greater anxiety.
Edinburgh Postnatal Depression Scale
The Edinburgh Postnatal Depression Scale (EPDS) consisted of 10 items to assess postpartum depressive symptoms within the previous 7 days, using a scale of 0–3 (“no, not at all” to “yes, most of the time”).24 Depressive symptoms items were summed, and scores ≥12 were considered indicative of elevated depressive symptoms.
Parenting Stress Index, Fourth Edition, Short Form
The Parenting Stress Index, Fourth Edition, Short Form (PSI-4-SF) was used to measure maternal distress centering around parenting.25 For the PSI-SF, percentile scores were calculated for Total Stress, where higher scores indicated greater problems. Women who scored >90th percentile were considered to have high parenting stress.
Psychiatric diagnoses
Women were asked to report any current psychiatric diagnoses they had received from a medical professional.
Perinatal Obsessive-Compulsive Scale and related questions
Four questions from the Perinatal Obsessive-Compulsive Scale (POCS) were used to measure the presence or absence of obsessive thoughts or behaviors specific to the perinatal period. These were accompanied by 10 novel items developed by the study team tailored to mothers' worries and behaviors specific to infant feeding (Supplementary Table S1).26 Response options used a 5-point scale, ranging from 0 (“Never”) to 4 (“Very Often”).
The series of 14 questions related to perinatal OC symptoms exhibited very good consistency (Cronbach's alpha = 0.83) and were summed to form a continuous composite score with a maximum value of 56 for analysis. Each item was also analyzed as a binary variable where women who indicated they “Very Often” worried or had thoughts about a particular infant feeding or care concern were compared with women who had “no” or less frequent thoughts or worries.
Breastfeeding practices
Women were asked if they were currently breastfeeding, either at the breast or by pumping, and when they started and stopped (if they had stopped). Women who had ever expressed (pumped) milk to feed their infant were asked how many times per day they typically pumped. Possible responses included: never used a breast pump, less than once per day, 1–2 times per day, 3–5 times per day, 6–8 times per day, 9–12 times per day, and >12 times per day. Breast milk feeding duration (feeding baby directly from the breast, expressed milk feeding, or a combination of both methods) was calculated for those who had ceased the particular feeding practice by subtracting the age of the child when breastfeeding started from the age when breastfeeding stopped and then dichotomized at 2 months of age for analysis.
Breastfeeding Self-Efficacy Scale–Short Form
The Breastfeeding Self-Efficacy Scale–Short Form (BSES-SF) consisting of 14 items presented on a Likert scale ranging 1–4 (“not at all confident” to “always confident”) was used to measure breastfeeding self-efficacy.27 The instrument was scored according to previously published methods. Higher self-efficacy scores indicate greater self-efficacy. The BSES-SF was analyzed as a continuous variable.
Maternal Breastfeeding Evaluation Scale
The Maternal Breastfeeding Evaluation Scale (MBFES) consisting of 30 items scored on a 1–4 point scale from “strongly disagree” to “strongly agree” was used to evaluate the overall breastfeeding experience (e.g., maternal enjoyment and role attainment, infant satisfaction and growth, lifestyle, and body image).28 To evaluate the outcomes regardless of breastfeeding status at the time of the study, we altered the verb tense. The instrument was scored according to previously published methods. The sum of the MBFES formed a score for the woman's overall evaluation of the breastfeeding experience (higher scores indicated a more positive experience), with some items reverse coded. The MBFES was analyzed as a continuous variable.
Breastfeeding problems
Women were also asked if they experienced any of a list of 12 breastfeeding problems, including physical problems (e.g., sore or cracked nipples), milk flow or supply problems, and infant problems (e.g., baby had trouble sucking).29
We used descriptive statistics to characterize the sample and explore the distributions of each variable. We examined the correlation between OCI-R scores and anxiety and depressive symptoms using Pearson's correlation coefficient and logistic regression, respectively. We used modified Poisson regression to estimate associations (risk ratios) between having clinically elevated OC symptoms and having frequent thoughts or worries about each infant feeding and care concern versus having such thoughts or worries less often or not at all.
Next, we used linear regression to examine associations between having clinically elevated OC symptoms and the continuous outcomes for breastfeeding self-efficacy (BSES-SF), overall breastfeeding experience (MBFES), the sum of breastfeeding problems, the sum of the infant feeding or care concerns, and the number of times per day the woman pumped. We also adjusted all the above models for depressive symptoms (EPDS continuous score), but sparse data precluded control for additional variables. Finally, we examined the association between clinically elevated OC symptoms and whether women breastfed (at the breast, fed pumped/expressed milk, or a combination) for at least 2 months versus less than 2 months using the modified Poisson regression.
This last set of models was also adjusted for maternal age, education, employment status, and type of delivery (Cesarean section or vaginal).
Results
Of the 232 women in our sample, 54% were between the ages of 27 and 34 years (Fig. 1 and Table 1). Half of the women (51%) had some college education or a college degree, whereas 41% had a postgraduate education. Most were married or living with a partner (89%) and identified as Caucasian/White (74%). A little less than half (47%) recently delivered their first child. Sixty percent of the women had returned to work since delivery. One fourth (25%) of the women delivered their baby via Cesarean section. Overall (18%) of the sample reported a history of a psychiatric diagnosis, and 17% had elevated depressive symptoms.
FIG. 1.
Participant flow diagram.
Table 1.
Participant Characteristics by Obsessive-Compulsive Symptoms, Life After Pregnancy Study (2016, United States)
| Maternal or infant characteristic | Total sample (N = 232), n (%) | Women with subthreshold OC symptoms (n = 200, 86%), n (%) | Women with clinically elevated OC symptomsa (n = 32, 14%), n (%) |
|---|---|---|---|
| Maternal age, years | |||
| 18–26 | 44 (19) | 33 (17) | 11 (34) |
| 27–30 | 63 (27) | 54 (27) | 9 (28) |
| 31–34 | 64 (27) | 58 (29) | 5 (16) |
| ≥35 | 62 (26) | 55 (28) | 7 (22) |
| Maternal education | |||
| ≤High school/GED | 17 (7) | 16 (8) | 1 (3) |
| Some college/associate's degree | 52 (22) | 37 (19) | 15 (47) |
| College graduate | 68 (29) | 61 (31) | 7 (22) |
| Postgraduate education | 95 (41) | 86 (43) | 9 (28) |
| Marital status | |||
| Married/living with partner | 207 (89) | 182 (91) | 25 (78) |
| Single | 17 (7) | 11 (6) | 6 (19) |
| Separated, divorced, widowed, or partner not living together | 8 (3) | 7 (4) | 1 (3) |
| Maternal race | |||
| Caucasian/White | 172 (74) | 153 (77) | 19 (59) |
| African American/Black | 17 (7) | 13 (7) | 4 (13) |
| Other or multiple races | 18 (8) | 17 (9) | 1 (3) |
| Missing | 25 (11) | 17 (9) | 8 (25) |
| Maternal ethnicity | |||
| Hispanic | 11 (5) | 9 (5) | 2 (6) |
| Non-Hispanic | 199 (86) | 176 (88) | 23 (72) |
| Missing | 22 (9) | 15 (8) | 7 (22) |
| Household income | |||
| <$15,000 | 8 (3) | 7 (4) | 1 (3) |
| $15,000 to <$35,000 | 22 (9) | 15 (8) | 7 (22) |
| $35,000 to <$55,000 | 26 (11) | 24 (12) | 2 (7) |
| $55,000 to <$75,000 | 38 (16) | 33 (17) | 5 (16) |
| $75,000 to <$95,000 | 30 (13) | 25 (13) | 5 (16) |
| ≥$95,000 | 85 (37) | 80 (40) | 5 (16) |
| Missing | 23 (10) | 16 (8) | 7 (22) |
| Employment | |||
| Stay at home mother | 68 (29) | 57 (29) | 11 (34) |
| Employed full time or full-time student | 108 (47) | 91 (46) | 17 (53) |
| Employed part time | 31 (13) | 28 (14) | 3 (9) |
| Temporary maternity leave | 25 (11) | 24 (12) | 1 (3) |
| Mother returned to work | 139 (60) | 119 (60) | 20 (63) |
| Child age when returned to work/school (among those who had returned), months | |||
| 0 | 8 (6) | 8 (7) | 0 (0) |
| 1 | 36 (26) | 27 (23) | 9 (45) |
| 2 | 44 (32) | 39 (33) | 5 (25) |
| 3 | 32 (23) | 29 (24) | 3 (15) |
| 4–6 | 8 (6) | 8 (7) | 0 (0) |
| Missing | 11 (8) | 8 (7) | 3 (15) |
| Child attends day care program | |||
| Among those who were employed or in school | 102 (73) | 90 (76) | 12 (60 |
| Missing | 10 (7) | 7 (6) | 3 (15) |
| How household is able to make ends meet | |||
| With great difficulty | 10 (4) | 7 (4) | 3 (9) |
| With difficulty | 18 (8) | 14 (7) | 4 (13) |
| Just get by | 83 (35) | 69 (35) | 14 (44) |
| Easily | 70 (30) | 67 (34) | 3 (9) |
| Very easily | 28 (12) | 27 (14) | 1 (3) |
| Missing | 23 (10) | 16 (8) | 7 (22) |
| Household size | |||
| 2–3 People | 103 (44) | 92 (46) | 11 (34) |
| 4–5 People | 83 (36) | 34 (36) | 11 (34) |
| ≥6 People | 23 (10) | 20 (10) | 3 (9) |
| Missing | 23 (10) | 16 (8) | 7 (22) |
| No. of children in household | |||
| 1 | 109 (47) | 97 (49) | 12 (38) |
| 2 | 63 (27) | 58 (29) | 5 (16) |
| 3 | 26 (11) | 19 (10) | 7 (22) |
| ≥4 | 11 (5) | 10 (5) | 1 (3) |
| Missing | 23 (10) | 16 (8) | 7 (22) |
| Cesarean section | 57 (25) | 44 (22) | 13 (41) |
| Missing | 1 (0) | 0 (0) | 1 (3) |
| Infant age at time of survey, months | |||
| 2 | 51 (22) | 44 (22) | 7 (22) |
| 3 | 54 (23) | 47 (24) | 7 (22) |
| 4 | 47 (20) | 42 (21) | 5 (16) |
| 5 | 36 (16) | 34 (17) | 2 (6) |
| 6 | 44 (19) | 33 (17) | 11 (34) |
| Psychiatric problems diagnosed by a medical professionalb | |||
| Anxiety | 33 (14) | 28 (14) | 5 (16) |
| Obsessive-compulsive disorder | 3 (1) | 2 (1) | 1 (3) |
| Depression | 20 (9) | 14 (7) | 6 (19) |
| Postpartum depression | 5 (2) | 5 (3) | 0 (0) |
| Missing | 22 (9) | 15 (8) | 7 (22) |
| EPDS | |||
| ≥12 | 40 (17) | 24 (12) | 16 (50) |
| Missing | 3 (1) | 2 (1) | 1 (3) |
| PSI-4-SF >90th percentile for Total Stress | 6 (3) | 3 (2) | 3 (9) |
Women with clinically elevated OC symptoms received a score ≥21 on the OCI-R.
Percentages do not sum to 100 because only women who indicated that they have a psychiatric problem diagnosed by a medical professional were asked to indicate their particular diagnosis.
EPDS, Edinburgh Postnatal Depression Scale; OC, obsessive-compulsive; OCI-R, Obsessive-Compulsive Inventory–Revised; PSI-4-SF, Parenting Stress Index, Fourth Edition, Short Form.
Thirty-two (14%) women had clinically elevated OC symptoms based on the OCI-R. Women with clinically elevated OC symptoms were more than seven times more likely than women with low symptoms to have elevated postpartum depressive symptoms (odds ratio = 7.73, 95% confidence interval [CI]: 3.39 to 17.62). OC symptoms were positively correlated with state anxiety as well (r = 0.45, p < 0.0001).
Women with clinically elevated OC symptoms were more likely to endorse each of the four items from the POCS questionnaire, although effect estimates were slightly attenuated upon adjustment for depressive symptoms (Table 2). Of the four items, they were most likely to report being very often worried about accidentally harming their baby (adjusted relative risk = 2.01 95% CI: 1.19 to 3.41). Of the novel questions about infant care and feeding worries, 56% (n = 18) of the women with clinically elevated symptoms answered “Very Often” to repeatedly washing their babies' dishes. These women also had 2.71 (95% CI: 1.20 to 6.12) times the risk of reporting repeatedly counting or organizing containers or baggies of breast milk in their freezer.
Table 2.
Obsessive-Compulsive Symptoms Specific to Infant Care and Feeding in Relation to Scores on the Obsessive-Compulsive Inventory–Revised, Life After Pregnancy Study (2016, United States)
| Women subthreshold OC symptoms (n = 200, 86%) |
Women with clinically elevated OC symptomsa
(n = 32, 14%) |
RR (95% CI) | Adjusted RR (95% CI)b | |||
|---|---|---|---|---|---|---|
| Yes, n (%) | No, n (%) | Yes, n (%) | No, n (%) | |||
| Being criticized and/or judged as a motherc | 53 (27) | 147 (74) | 16 (50) | 16 (50) | 1.89 (1.24 to 2.86) | 1.40 (0.92 to 2.12) |
| Baby being contaminated (e.g., by germs)c | 42 (21) | 157 (79) | 14 (44) | 17 (53) | 2.14 (1.33 to 3.43) | 1.92 (1.13 to 3.25) |
| Baby being unwell at birth or having an unhealthy babyc | 37 (19) | 163 (82) | 14 (44) | 18 (56) | 2.37 (1.45 to 3.86) | 1.63 (0.99 to 2.69) |
| Accidentally harming my babyc | 30 (15) | 170 (85) | 13 (41) | 19 (59) | 2.71 (1.59 to 4.62) | 2.01 (1.19 to 3.41) |
| Your baby's breast milk or formula being contaminated | 10 (5) | 190 (95) | 3 (9) | 29 (91) | 1.88 (0.55 to 6.45) | 1.15 (0.39 to 3.34) |
| Running out of breast milk or formula for your baby | 41 (21) | 159 (80) | 11 (34) | 21 (66) | 1.68 (0.97 to 2.91) | 1.32 (0.77 to 2.28) |
| That you were not producing enough breast milk for your baby | 65 (33) | 135 (68) | 16 (50) | 16 (50) | 1.54 (1.03 to 2.29) | 1.30 (0.85 to 2.00) |
| That your baby was not growing enough | 23 (12) | 176 (88) | 8 (25) | 24 (75) | 2.16 (1.06 to 4.41) | 1.44 (0.77 to 2.68) |
| That you would make a mistake in preparing your baby's bottle and it would hurt your baby | 8 (4) | 190 (95) | 4 (13) | 28 (88) | 3.09 (0.99 to 9.68) | 1.95 (0.67 to 5.67) |
| Kept a log of how much breast milk you were pumping each time you pumped | 37 (19) | 163 (82) | 7 (22) | 25 (78) | 1.18 (0.58 to 2.42) | 1.31 (0.63 to 2.71) |
| Repeatedly washed your baby's bottles, bowl, or plate | 48 (24) | 152 (76) | 18 (56) | 14 (44) | 2.34 (1.58 to 3.47) | 2.37 (1.55 to 3.64) |
| Weighed your baby on a scale at home | 21 (11) | 179 (90) | 3 (9) | 29 (91) | 0.89 (0.28 to 2.82) | 0.95 (0.31 to 2.92) |
| Used a device (e.g., milk saver, breast shell), you placed in your bra to collect milk leaking from breast to feed your baby later | 8 (4) | 192 (96) | 4 (13) | 28 (88) | 3.13 (1.00 to 9.78) | 4.52 (1.43 to 14.31) |
| Repeatedly counted or organized containers or baggies of breast milk stored in your freezer | 16 (8) | 184 (92) | 8 (25) | 24 (75) | 3.13 (1.46 to 6.70) | 2.71 (1.20 to 6.12) |
Missing data: Women with subthreshold OC symptoms: Baby being contaminated (1), Baby not growing enough (1), Make a mistake in preparation of baby's bottle (2). Women with clinically elevated OC (OC) symptoms: Baby being contaminated (1).
Women with clinically elevated OC symptoms were those who received a score ≥21 on the OCI-R.
Adjusted for depressive symptoms.
Questions are from the POCS.
CI, confidence interval; OC, obsessive-compulsive; OCI-R, Obsessive-Compulsive Inventory–Revised; POCS, Perinatal Obsessive-Compulsive Scale; RR, relative risk.
Women in the clinically elevated OC symptoms group had more total obsessive thoughts or behaviors specific to the perinatal period (6.0 points higher on the continuous perinatal OC symptoms variable, 95% CI: 2.7 to 9.3), although this was attenuated to 3.6 points upon adjustment for depressive symptoms (95% CI: 0.4 to 6.9) (Table 3). They initially appeared to have lower breastfeeding self-efficacy than women with fewer OC symptoms (β = −7.2, 95% CI: −13.8 to −0.6), but this was somewhat attenuated and no longer statistically significant upon adjustment. However, they did not report an overall worse breastfeeding experience (adjusted β = 0.4, 95% CI: −9.3 to 10.1).
Table 3.
Associations Between Obsessive-Compulsive Symptoms and Breastfeeding and Infant Care Practices, Life After Pregnancy Study (2016, United States)
| Women subthreshold OC symptoms (n = 200, 86%) |
Women with clinically elevated OC symptomsa
(n = 32, 14%) |
β (95% CI) | Adjusted β (95% CI) | |
|---|---|---|---|---|
| Median (inter-quartile range) | ||||
| Postpartum OC symptoms related to infant care | 16 (12) | 24 (18) | 6.0 (2.7 to 9.3) | 3.6 (0.4 to 6.9) |
| BSES-SF | 55 (20) | 43 (26) | −7.2 (−13.8 to −0.6) | −3.5 (−10.0 to 2.9) |
| Overall breastfeeding experience (MBFES) | 119 (33) | 110 (37) | −5.3 (−15.1 to 4.6) | 0.4 (−9.3 to 10.1) |
| Breastfeeding problems | 5 (3) | 6 (3) | 0.6 (−0.3 to 1.6) | 0.3 (0.0 to 0.2) |
| No. of times pumped breast milkb | c | c | 0.6 (0.0 to 1.3) | 0.6 (−0.0 to 1.2) |
Missing data: Women subthreshold OCD symptoms: BSES (28), MBFES (26), breastfeeding problems (14). Women with clinically elevated OC symptoms: BSES (10), MBFES (10), breastfeeding problems (7).
Women with clinically elevated OC symptoms were those who received a score ≥21 on the OCI-R.
Number of times per day using a breast pump was a semi-continuous variable (Fig. 2).
Adjusted for depressive symptoms.
BSES-SF, Breastfeeding Self-Efficacy Scale–Short Form; CI, confidence interval; MBFES, Maternal Breastfeeding Evaluation Scale; OC, obsessive-compulsive; OCD, Obsessive-Compulsive Disorder; OCI-R, Obsessive-Compulsive Inventory–Revised.
The median number of breastfeeding problems in this sample was 5 (interquartile range = 3), and women with elevated OC symptoms had slightly more breastfeeding problems than other women (adjusted β = 0.3, 95% CI: 0.0 to 0.2) after accounting for depressive symptoms. Of the 232 women in our sample, 193 (83%) experienced at least 2 breastfeeding problems. Women with clinically elevated OC symptoms were 0.76 (95% CI: 0.58 to 0.99) times less likely to breastfeed for at least 2 months or more, but this was not statistically significant upon adjusting for participants' age, education, employment, delivery method, and depressive symptoms (Table 4).
Table 4.
Breastfeeding and Pumping, Life After Pregnancy Study (2016, United States)
| Women subthreshold of OC symptoms (n = 200, 86%) |
Women with clinically elevated OC symptomsa
(n = 32, 14%) |
RR (95% CI) | Adjusted RR (95% CI)b |
|||
|---|---|---|---|---|---|---|
| Yes, n (%) | No, n (%) | Yes, n (%) | No, n (%) | |||
| Breastfed for at least 2 months | 169 (85) | 30 (15) | 20 (63) | 11 (34) | 0.76 (0.58 to 0.99) | 0.85 (0.64 to 1.11) |
| Expressed milk fed for at least 2 months | 138 (69) | 55 (28) | 17 (53) | 13 (41) | 0.79 (0.57 to 1.10) | 0.94 (0.67 to 1.31) |
| Direct milk fed for at least 2 months | 157 (79) | 40 (20) | 18 (56) | 12 (38) | 0.76 (0.58 to 1.01) | 0.93 (0.69 to 1.25) |
Missing data: Women with clinically elevated OC symptoms: Breastfed (1), Expressed (2), Direct (2) Women subthreshold of OC symptoms: Breastfed (1), Expressed (7), Direct (3).
Women with clinically elevated OC symptoms received a score ≥21 on the OCI-R.
Adjusted for participant's age, education, employment, delivery method, and depressive symptoms.
CI, confidence interval; OC, obsessive-compulsive; OCI-R, Obsessive-Compulsive Inventory–Revised.
Their mode of breastfeeding (i.e., fed directly from the breast or pumped) did not prove to be statistically significantly different when compared with women with subthreshold OC symptoms. Of the women who were asked about how many times a day they pumped, 25% of both groups pumped 1–2 times a day. Sixteen percent (n = 5) of women with clinically elevated OC symptoms pumped 6 or more times a day. In comparison, 8% (n = 15) of women with subthreshold OC symptoms pumped 6 or more times a day (Fig. 2).
FIG. 2.
Pumping frequency among women with clinically elevated OC symptoms and women with subthreshold OC symptoms. Some women with clinically elevated OC symptoms (11) and some women subthreshold OC symptoms (53) were not asked for pumping frequency because they indicated earlier in the survey that they never pumped or hand-expressed. OC, obsessive-compulsive.
Discussion
In this study, women with clinically elevated OC symptoms had more total obsessive thoughts or behaviors specific to the perinatal period than women without clinically elevated symptoms, and these thoughts and behaviors extended to infant feeding and care activities. They also had more breastfeeding problems than women with fewer OC symptoms, but they did not self-report a poorer breastfeeding experience or poorer self-efficacy, after accounting for comorbid depressive symptoms.
The literature lacks focus on OC in relation to breastfeeding, childcare, and feeding. Some studies have focused on anxiety and fairly consistently found medium to large negative effects on breastfeeding duration, but very few have focused on OCD specifically.30 The results of the present study generally align with those of a single small study by Challacombe et al., who reported that women diagnosed with OCD were less likely to breastfeed at 6 months postpartum (27% versus 57%; chi square p = 0.01).16 Our effect estimate for the association between OC symptoms and breastfeeding duration was below the null as well, but the CI included the null upon adjustment for confounders.
Challacombe et al. hypothesized that because women with OCD were less likely to breastfeed, OCD could potentially interfere with protective effects of breastfeeding, “such as lower self-efficacy in the mothering role.” We noted that women with elevated OC symptoms also had lower scores on the BSES-SF, but upon adjustment for depressive symptoms, this effect estimate was somewhat attenuated and the CI included the null, suggesting that both OC symptoms and depressive symptoms play a role in breastfeeding self-efficacy in this sample.
This study also identified some associations between elevated OC symptoms and other aspects of infant feeding. For instance, our data suggested that women with elevated symptoms may have pumped more times per day, with a few pumping 12 or more time per day, which may interfere with self-care behaviors. Future research might focus on the possibility that excessive pumping may be a component of postpartum OC symptomatology. We observed associations with other aspects of feeding-related anxiety (e.g., repeatedly washed your baby's bottles, bowl, or plate, repeatedly counting, and organizing stored breast milk), compatible with the findings of Farrow and Blissett about food restriction.9,14,31
One limitation of this study is that all mental health measures were self-reported, which may be less objective than a diagnostic interview. For instance, the OCI-R is a self-reported instrument, and so, some women may have been misclassified. However, the OCI-R remains true to the longer version OCI in terms of psychometrics, and the OCI has been shown to have excellent discriminant validity among groups defined by diagnostic instruments and also good convergent validity with other OCD instruments.20,32 Also, the Research Match Tool precluded us from examining selection bias in our sample because it did not enable calculation of a response rate or characterization of those who declined to participate.
Another limitation is that our study was cross-sectional; thus, it was not able to examine whether OC symptoms changed across the postpartum period or to evaluate the temporality of the relationship between breastfeeding practices and experiences and the onset of OC symptoms. Also, in general, our sample had higher socioeconomic status in comparison to the general population.33 Additionally, although our sample included women with elevated OC symptoms, the overall sample size meant that our effect estimates were rather imprecise because of limited statistical power.
Finally, as is common with mental health conditions, we observed comorbidity in our sample, for instance, among elevated depressive symptoms and elevated OC symptoms. Depression also has been shown to have negative effects on a variety of breastfeeding outcomes such as duration and self-efficacy, but OCD has its own characteristic pattern of thoughts and behaviors that distinguish it from depression and so can be expected to have its own effects on breastfeeding and parenting.11,34
Despite its limitations, our sample included women from across the United States, while most studies in this field are based on single obstetric care sites. Our survey included a range of mental health and psychosocial measures that have been previously validated. Women in our study may have been more likely to disclose sensitive information (e.g., mental health) because the survey was self-administered rather than in-person, thereby potentially increasing the accuracy of our assessments. Finally, our survey consisted of detailed breastfeeding measures and included questions about pumping, which is rarely studied specifically.28,29,35
Conclusions
Given that mothers in this study who had clinically elevated OCD-related worries and compulsions also tended to have thoughts or behaviors specific to infant care and feeding and more breastfeeding problems, timely assessment and treatment of postpartum OC symptoms, and the likely accompanying depressive symptoms, may have benefits to infant care and maternal–child relationships. In the future, longitudinal research to examine these associations over time would further inform intervention approaches.
Supplementary Material
Acknowledgments
We thank the Ohio State University Center for Clinical and Translational Science; Myra George of Nationwide Children's Hospital Clinical Research Services; and Kelly Boone, Erin Shafer, Thalia Cronin, and Rachel Mason of the Center for Biobehavioral Health. We had no writing assistance.
Authors' Contributions
All authors contributed to article preparation and/or study conception and design. Material preparation, data collection, and analysis were performed by S.A.K., C.E.C., and J.L.J. The first draft of the article was written by W.L.P. and S.A.K. All authors commented on previous versions of the article and read and approved the final article.
Disclosure Statement
The authors have no conflicts of interest to disclose.
Funding Information
This study was supported by the National Institutes of Health/National Center for Advancing Translational Sciences (UL1TR001070) and internal funds of the Research Institute at Nationwide Children's Hospital.
Supplementary Material
References
- 1. American Psychiatric Association (APA). Diagnostic and Statistical Manual of Mental Disorders: DSM-5. Washington, DC: American Psychiatric Association (APA), 2013. [Google Scholar]
- 2. Navarro P, García-Esteve L, Ascaso C, et al. Non-psychotic psychiatric disorders after childbirth: Prevalence and comorbidity in a community sample. J Affect Disord 2008;109:171–176. [DOI] [PubMed] [Google Scholar]
- 3. Wenzel A, Haugen EN, Jackson LC, et al. Anxiety symptoms and disorders at eight weeks postpartum. J Anxiety Disord 2005;19:295–311. [DOI] [PubMed] [Google Scholar]
- 4. Miller ES, Chu C, Gollan J, et al. Obsessive-Compulsive symptoms during the postpartum period. J Reprod Med 2013;58:115–122. [PMC free article] [PubMed] [Google Scholar]
- 5. Sadock BJ, Kaplan HI, Sadock VA. Kaplan and Sadock's Synopsis of Psychiatry: Behavioral Sciences/Clinical Psychiatry, 10th ed. Philadelphia, PA: Lippincott Williams & Wilkins, 2007. [Google Scholar]
- 6. Biegon A, Reches A, Snyder L, et al. Serotonergic and noradrenergic receptors in the rat brain: Modulation by chronic exposure to ovarian hormones. Life Sci 1983;32:2015–2021. [DOI] [PubMed] [Google Scholar]
- 7. Cone RI, Davis GA, Goy RW. Effects of ovarian steroids on serotonin metabolism within grossly dissected and microdissected brain regions of the ovariectomized rat. Brain Res Bull 1981;7:639–644. [DOI] [PubMed] [Google Scholar]
- 8. Renner KJ, Krey LC, Luine VN. Effect of progesterone on monoamine turnover in the brain of the estrogen-primed rat. Brain Res Bull 1987;19:195–202. [DOI] [PubMed] [Google Scholar]
- 9. Sichel DA, Cohen LS, Dimmock JA, et al. Postpartum obsessive compulsive disorder: A case series. J Clin Psychiatry 1993;54:156–159. [PubMed] [Google Scholar]
- 10. Williams KE, Koran LM. Obsessive-compulsive disorder in pregnancy, the puerperium, and the premenstruum. J Clin Psychiatry 1997;58:330–334; quiz 335–336. [DOI] [PubMed] [Google Scholar]
- 11. Fairbrother N, Thordarson DS, Challacombe FL, et al. Correlates and predictors of new mothers' responses to postpartum thoughts of accidental and intentional harm and obsessive compulsive symptoms. Behav Cogn Psychother 2018;46:437–453. [DOI] [PubMed] [Google Scholar]
- 12. Abarmowitz JS, Nelson CA, Rygwall R, et al. The cognitive mediation of obsessive-compulsive symptoms: A longitudinal study. J Anxiety Disord 2007;21:91–104. [DOI] [PubMed] [Google Scholar]
- 13. Challacombe F, Salkovskis P. A preliminary investigation of the impact of maternal obsessive-compulsive disorder and panic disorder on parenting and children. J Anxiety Disord 2009;23:848–857. [DOI] [PubMed] [Google Scholar]
- 14. Farrow C, Blissett J. Is maternal psychopathology related to obesigenic feeding practices at 1 year? Obes Res 2005;13:1999–2005. [DOI] [PubMed] [Google Scholar]
- 15. Spill MK, Callahan EH, Shapiro MJ, et al. Caregiver feeding practices and child weight outcomes: A systematic review. Am J Clin Nutr 2019;109(Suppl 7):990s–1002s. [DOI] [PubMed] [Google Scholar]
- 16. Challacombe FL, Salkovskis PM, Woolgar M, et al. Parenting and mother-infant interactions in the context of maternal postpartum obsessive-compulsive disorder: Effects of obsessional symptoms and mood. Infant Behav Dev 2016;44:11–20. [DOI] [PubMed] [Google Scholar]
- 17. Dias CC, Figueiredo B. Breastfeeding and depression: A systematic review of the literature. J Affect Disord 2015;171:142–154. [DOI] [PubMed] [Google Scholar]
- 18. Tamminen T. The impact of mother's depression on her nursing experiences and attitudes during breastfeeding. Acta Paediatr Scand Suppl 1988;344:87–94. [DOI] [PubMed] [Google Scholar]
- 19. Flores-Quijano ME, Córdova A, Contreras-Ramírez V, et al. Risk for postpartum depression, breastfeeding practices, and mammary gland permeability. J Hum Lact 2008;24:50–57. [DOI] [PubMed] [Google Scholar]
- 20. Foa EB, Huppert JD, Leiberg S, et al. The Obsessive-Compulsive Inventory: Development and validation of a short version. Psychol Assess 2002;14:485. [PubMed] [Google Scholar]
- 21. Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)—A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42:377–381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Chaudron LH, Nirodi N. The obsessive–compulsive spectrum in the perinatal period: A prospective pilot study. Arch Womens Ment Health 2010;13:403–410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Spielberger C. Manual for the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press, 1983. [Google Scholar]
- 24. Murray L, Carothers AD. The validation of the Edinburgh Post-natal Depression Scale on a community sample. Br J Psychiatry 1990;157:288–290. [DOI] [PubMed] [Google Scholar]
- 25. Abidin RR. Parenting Stress Index, 4th edition (PSI-4). Lutz, FL: PAR, Inc., 2012. [Google Scholar]
- 26. Lord C, Rieder A, Hall GBC, et al. Piloting the Perinatal Obsessive-Compulsive Scale (POCS): Development and validation. J Anxiety Disord 2011;25:1079–1084. [DOI] [PubMed] [Google Scholar]
- 27. Dennis CL. The breastfeeding self-efficacy scale: Psychometric assessment of the short form. J Obstet Gynecol Neonatal Nurs 2003;32:734–744. [DOI] [PubMed] [Google Scholar]
- 28. Leff EW, Jefferis SC, Gagne MP. The development of the Maternal Breastfeeding Evaluation Scale. J Hum Lact 1994;10:105–111. [DOI] [PubMed] [Google Scholar]
- 29. Binns CW, Scott JA. Breastfeeding: Reasons for starting, reasons for stopping and problems along the way. Breastfeed Rev 2002;10:13–19. [PubMed] [Google Scholar]
- 30. Hoff CE, Movva N, Rosen Vollmar AK, et al. Impact of maternal anxiety on breastfeeding outcomes: A systematic review. Adv Nutr 2019;10:816–826. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Black DW, Gaffney G, Schlosser S, et al. The impact of obsessive-compulsive disorder on the family: Preliminary findings. J Nerv Ment Dis 1998;186:440–442. [DOI] [PubMed] [Google Scholar]
- 32. Foa EB, Kozak MJ, Salkovskis PM, et al. The validation of a new obsessive–compulsive disorder scale: The Obsessive–Compulsive Inventory. Psychol Assess 1998;10:206. [Google Scholar]
- 33. Zambaldi CF, Cantilino A, Montenegro AC, et al. Postpartum obsessive-compulsive disorder: Prevalence and clinical characteristics. Compr Psychiatry 2009;50:503–509. [DOI] [PubMed] [Google Scholar]
- 34. Slomian J, Honvo G, Emonts P, et al. Consequences of maternal postpartum depression: A systematic review of maternal and infant outcomes. Womens Health (Lond) 2019;15:1–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Dennis CL, Faux S. Development and psychometric testing of the Breastfeeding Self-Efficacy Scale. Res Nurs Health 1999;22:399–409. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.