Interoception and Eating Behavior in Participants at Risk for Hypertensive Disorders of Pregnancy

Abstract

Background:

Disordered eating during pregnancy can impact maternal and fetal health. Disordered eating has been linked to higher cardiovascular risks including hypertensive disorders of pregnancy (HDP). Interoceptive awareness, the ability to perceive and respond to bodily sensations, is reduced among people with disordered eating and may be associated with blood pressure (BP). This study tested these associations in pregnant women at risk for HDP.

Methods:

Ninety-five pregnant women at risk for HDP participated in the study. At ~18 weeks’ gestation we measured 24-hour BP, interoceptive awareness, and disordered eating. Linear regression analyses were used to test associations, adjusting for covariates (BMI, education, income, race).

Results:

Greater interoceptive awareness—specifically lower anxiety about bodily sensations (“Not Worrying”)—was significantly associated with lower daytime diastolic BP (B = −0.21, p = .05), lower uncontrolled eating (B = −0.52, p < .001), and lower emotional eating (B = −0.51, p < .001). Higher scores on the “Attention Regulation” subscale were associated with less Uncontrolled Eating (B=−.24, p=.02) and less emotional eating (B=−.25, p=.02). Cognitive Restraint was associated with higher nighttime diastolic BP (B = 0.27, p = .04) but not interoceptive awareness.

Conclusion:

Increased interoceptive awareness was associated with both healthier eating behaviors and lower BP in pregnant women at risk for HDP. Interventions that enhance interoceptive awareness may offer a promising strategy for reducing risk for cardiovascular complications in pregnancy.

INTRODUCTION

Disordered eating, defined as a spectrum of problematic eating behaviors and distorted attitudes towards food, weight, shape, and appearance [1, 2], is associated with increased risk for cardiovascular disease. In a meta-analysis of over 50,000 patients, disordered eating was associated with increased incidence of cardiovascular disease, cardiovascular events, and all-cause mortality [3]. In a sample of 2,095 participants from the Coronary Artery Risk Development in Young Adults study, disordered eating was significantly associated with poorer cardiovascular health in a composite measure including blood pressure, cholesterol, glucose, diet, physical activity, and smoking [4]. In addition, recent evidence demonstrated that disordered eating is associated with blunted cardiovascular responses to stress and less response habituation to recurring stress, suggesting that dysregulated hemodynamic stress reactivity may serve as a mechanism through which disordered eating increases risk for cardiovascular disease [5].

Pregnancy is a vulnerable period during which women are susceptible to disordered eating [6]. A study by Martinez-Olcina et al. reported that 17.3% of pregnant women experienced episodes of binge eating [7]. An additional study found that during pregnancy, women showed lower levels of dietary restraint, reported eating more, and had increased levels of dissatisfaction with their body shape compared to pre-pregnancy levels [8]. There is also a small body of evidence demonstrating that disordered eating behaviors in pregnancy are associated with poorer perinatal cardiovascular health. For example, Jouppi et al. (2024) found loss of control eating during the second trimester of pregnancy was associated with greater cardiovascular risk on the American Heart Association Life’s Essential 8 composite score (diet, physical activity, nicotine exposure, sleep health, body mass index (BMI), blood lipids, blood glucose, and blood pressure) [9]. In addition, in a large Swedish cohort, any form of disordered eating in pregnancy was associated with increased risk for antepartum hemorrhage [10]. Taken together, these studies suggest that disordered eating is common in pregnancy and may increase risk for maternal cardiovascular complications. Therefore, identifying targets for interventions to decrease risk for disordered eating and subsequent cardiovascular illness are needed in order to improve maternal health.

Interoception, an interactive process of receiving, accessing, appraising, and integrating inner bodily signals [11], is a promising target for interventions and is heightened during pregnancy given the increased focus on the body’s internal state and the growing fetus [12]. Kirk and Preston (2019) [13] found that trusting the body was correlated with greater body satisfaction and acceptance of pregnancy-related physical changes, while lower interoception has been linked to disordered eating [14]. There is also evidence that interoception is associated with cardiovascular risk; in a cross sectional study, greater interoceptive awareness was associated with higher clinic blood pressure among patients with hypertension compared to normotensive patients [15], suggesting that interoception enhances cardiovascular reactivity in non-pregnant adults. Previous research has also shown bidirectional associations between interoceptive processes and the cardiovascular and central nervous systems among non-pregnant patients with early stage hypertensive disease using a laboratory paradigm [16].

No studies, to our knowledge, have examined associations among interoceptive awareness, disordered eating, and blood pressure in pregnancy. Thus, the goals of the present research are to examine these associations in a sample of pregnant women at risk of developing hypertensive disorders of pregnancy (HDP). We hypothesized that those with higher interoceptive awareness would display less disordered eating and lower maternal blood pressure prior to the onset of HDP.

MATERIALS AND METHODS

Ninety-five pregnant women were enrolled in an ongoing clinical trial of phone-delivered prenatal mindfulness training to prevent HDP (NCT04626245). The study protocol was previously published [17]. Briefly, participants were included in the study if they were considered at risk to develop HDP according to American College of Obstetrician and Gynecologist (ACOG) guidelines [18], were 18 years of age or older, < 20 weeks’ gestation, with a singleton pregnancy, English speaking, and were not engaged in regular mindfulness practices at the time of enrollment. In the present study, we examined data collection prior to randomization in the clinical trial. Participants provided written informed consent prior to engaging in study activities and were compensated for their time. This study was approved by the Institutional Review Board (Board # 020120, 1/13/2021).

Measures

Interoceptive Awareness

Participants completed the Multidimensional Assessment of Interoceptive Awareness (MAIA) scale, a psychometric tool developed by Mehling et al. (2012) [19] to assess various dimensions of interoceptive awareness, which refers to the perception of internal bodily signals [20]. The MAIA includes eight subscales. ‘Noticing’ reflects the ability to recognize bodily sensations as they occur. ‘Not-distracting’ assesses the tendency not to avoid or suppress awareness of these sensations. ‘Not-worrying’ measures the propensity not to worry about bodily sensations. ‘Attention Regulation’ reflects the ability to maintain a present-focused awareness of bodily sensations without judgment. ‘Emotional Awareness’ captures the capacity to connect bodily sensations with emotional experiences. ‘Self-regulation’ evaluates the ability to use interoceptive information to effectively manage emotional states. ‘Body Listening’ measures the degree to which individuals actively engage with their bodily sensations. ‘Trusting’ measures the extent to which individuals view their bodily signals as reliable sources of information. Each subscale is scored on a scale of 0 to 5, with 5 indicating the greatest awareness of bodily awareness. The MAIA has demonstrated good reliability and validity across diverse populations and settings and has been utilized to explore the relationship between interoceptive awareness and various psychological outcomes, including anxiety, depression, and eating disorders [21, 22]. The Cronbach’s alpha equals 0.944, reflecting good internal consistency.

Perceived Eating Behavior/Attitudes

Participants completed the Three Factor Eating Questionnaire- Revised 18 Items (TFEQ-R18). The TFEQ-R18 is a validated scale for the study of eating behavior in people with obesity and general populations [23] and includes 18 items on a 4-point Likert scale (1 = definitely true, 2= mostly true, 3= mostly false, and 4= definitely false) [24]. The scale measures three dimensions of eating: cognitive restraint, uncontrolled eating, and emotional eating [23]. Cognitive restraint is defined by conscious restriction of food intake; uncontrolled eating is the tendency to eat more than usual due to a loss of control; emotional eating is described by a tendency to eat in response to emotional urges [25]. Higher scores indicate greater disordered eating behaviors. The TFEQ has been used numerous times during pregnancy to measure disordered eating’s effect on weight gain [26]. The Cronbach’s alpha equals 0.839, reflecting good internal consistency.

Antenatal Blood Pressure

Ambulatory Blood Pressure (ABPM) was assessed over 24 hours. BP measures were collected every 30 minutes during the day and every 60 minutes at night using Microlife WatchBP03, validated for use in pregnancy [27]. ABPM software automatically calculates average daytime and nighttime Systolic Blood Pressure (SBP) and Diastolic Blood Pressure (DBP).

Statistical Analysis

First, data was checked for significant skewness to confirm that assumptions of statistical tests were met. No variables were significantly skewed (<2). We performed descriptive statistics to describe the sample. We then tested for potential covariates that may be associated with BP including race, BMI, number of past pregnancies, education, income, and marital status. Those that were significantly associated with BP were included in our subsequent analysis. We then conducted linear regression analyses between 1) MAIA and BP, 2) MAIA and TFEQ, and 3) TFEQ and BP measures. IBM SPSS v 25 was used to examine associations. P-values < .05 were considered statistically significant.

RESULTS

We examined 95 women at risk for HDP from the RCT who completed the TFEQ measure. The average age of participants was 32 years (SD=5). Average BMI was 30.4 kg/m^2 (SD=7). Mean gestational age at enrollment was 18 weeks (SD=1). Approximately 30% of the sample reported their ethnicity as Hispanic, 19% of the sample reported their race as Black, 71% reported their race as White, 7% reported their race as Asian, 1 % reported their race as Native Hawaiian/ Pacific Islander, 2% reported their race as American Indian/Alaskan Native, and 12% reported their race as other. Around 19% of participants were primigravida (pregnant with their first baby). Nineteen percent reported an average household income of <$40K per year, and 19% reported less than a college education. Sixty-four percent of the sample reported being married or in a committed relationship (See Table 1).

Table 1.

Participant Characteristics

Maternal Characteristic	Mean	SD	%
Age (years)	32	5
Race (% Black)			19%
Ethnicity (% Hispanic)			30%
Body mass index (kg/m2)	30.4	7
Weeks of Gestation	18	1
Primigravida/Pregnant with first baby (% yes)			19%
Education (% high school or less)			19%
Income (% $40,000 or less)			19%
Relationship Status (% partnered)			64%

Next, we tested potential covariates that could be associated with maternal BP using independent samples T tests or chi-squared analyses. We found that BMI was significantly associated with higher 24 hr SBP, 24 hr DBP, daytime SBP, daytime DBP, and nighttime SBP (all p values <.05). Higher education was associated with lower nighttime SBP and DBP (p<.05). Higher income was associated with lower daytime DBP (p<.05). There were no significant associations between maternal age, marital status, ethnicity, race, or number of previous births and maternal BP (p>.05). Therefore, based on these findings BMI, education, and income were included as covariates in subsequent models.

First, to test our study hypotheses we performed linear regression analyses, adjusting for the aforementioned covariates, to examine the association between the MAIA subscales and BP measures (daytime SBP, nighttime SBP, daytime DBP, and nighttime DBP). MAIA “Not Worrying” was significantly associated with daytime DBP (B=−.21, p=.05) such that greater interoceptive awareness was associated with lower DBP. No other MAIA subscales were significantly associated with BP measures.

Next, we conducted linear regression analyses, adjusting for the covariates, to test the associations between MAIA and the three TFEQ subscales. We found that higher scores on the MAIA “Not Distracting” subscale were associated with less Emotional Eating (B=−.23, p=.04). Higher scores on the “Not Worrying” subscale were associated with less Uncontrolled Eating (B=−.52, p<.001) and less Emotional Eating (B=−.51, p<.001). Higher scores on the “Attention Regulation” subscale were associated with less Uncontrolled Eating (B=−.24, p=.02) and less emotional eating (B=−.25, p=.02). There were no significant associations among MAIA subscales and Cognitive Restraint.

Lastly, we tested associations between TFEQ subscales and BP measures using linear regression analyses. We found that higher scores on Cognitive Restraint was significantly associated with higher nighttime DBP (B=.27, p=.04). No other associations were statistically significant.

DISCUSSION

The present study aimed to explore the associations between interoception, eating behaviors, and BP in pregnant women at risk for HDP. Specifically, we examined whether interoceptive awareness was associated with disordered eating, and whether disordered eating was associated with maternal BP. We hypothesized that interoceptive awareness would predict less disrupted eating behavior and lower BP. Our results showed that interoceptive awareness was significantly associated with less uncontrolled and emotional eating and lower BP.

The “Not Worrying” subscale of the MAIA was significantly associated with lower daytime DBP suggesting that pregnant women who are less anxious about their bodily sensations may experience lower BP throughout the day. This aligns with research indicating that anxiety and subsequent increased sympathetic activation are correlated with cardiovascular risk factors, including hypertension [28, 29]. Furthermore, previous research has shown bidirectional associations between interoceptive processes and the cardiovascular and central nervous systems [16]. While other subscales of interoception did not show significant associations with BP, the link between reduced bodily worry and DBP highlights the potential role of somatic perception and appraisal in maternal cardiovascular health.

Additionally, our results suggested that lower anxiety about bodily sensations (as measured by the Not Worrying subscale) was associated with decreased emotional eating and uncontrolled eating. This finding is consistent with prior research suggesting that heightened interoceptive awareness—particularly the ability to experience bodily sensations without excessive concern—may serve as a protective factor against disordered eating patterns [30]. A meta-analysis of 41 studies confirmed large interoceptive deficits among adults with eating disorders, as well as those who have recovered from eating disorders [31]. Researchers hypothesize that deficits in recognizing bodily sensations may contribute to eating disorders by disrupting perceptions of hunger and fullness cues. Difficulties in detecting hunger cues may lead to meal skipping or food restriction, whilst difficulty sensing fullness may result in overeating or binge eating.

Interestingly, scores on the Cognitive Restraint eating behavior was associated with higher nighttime DBP, yet showed no significant association with interoceptive awareness. This may suggest that the most effective interventions to address specifically restrictive eating and decrease its impact on cardiovascular health include those that target cognitive restructuring, such as cognitive behavioral therapy (CBT) rather than those that target interoceptive awareness, such as mindfulness training. This is consistent with past evidence that CBT outperformed other forms of psychotherapy to treat eating disorders [32].

Taken together, results from this study advance the field of prenatal health by highlighting interoception as a potential target for interventions aimed to lower BP in pregnant women at risk for HDP. Intuitive eating, or eating according to the body’s natural hunger and fullness cues [33], is positively associated with interoceptive sensitivity and inversely associated with body mass index [30]. Furthermore, Mindfulness based interventions (MBIs) have been shown to improve both interoceptive awareness and blood pressure in adults [34, 35], which may impact eating behavior. In fact, an app-based mindfulness training app was shown to reduce craving-related eating behaviors by 40.21% in overweight or obese women in just one month [36]. Another systematic review found that, across 21 studies, MBIs significantly improved binge eating severity, as measured using the Binge Eating Scale, a questionnaire measuring frequency and intensity binge eating behavior as well as feelings surrounding binge episodes [37]. Thus, MBIs may be an avenue through which the pregnant population can improve interoception and subsequently eating behavior, particularly emotional and uncontrolled eating.

While previous research has established links between interoception, disordered eating, and cardiovascular health in non-pregnant individuals [16, 21], we are aware of no prior studies that have explored these relationships within the context of pregnancy. Given the profound physiological and psychological changes that occur during pregnancy—including altered cardiovascular regulation, metabolic shifts, and increased vulnerability to stress—understanding the role of interoceptive processes in this population is of particular significance [38]. The findings of this study contribute to the literature by demonstrating that greater interoceptive awareness, particularly reduced anxiety about bodily sensations, is associated with reduced instances of emotional or uncontrolled eating and lower BP. The findings further demonstrate that the increased cognitive restraint is associated with higher BP; however, cognitive restraint was not associated with interoception, suggesting that interventions focused on cognitive process would yield the most benefit for this eating behavior.

Strengths of this study include the use of 24-hour ambulatory BP measurements to ensure a more precise evaluation of hypertension and related outcomes. Additionally, the inclusion of a racially and ethnically diverse population broadens the study’s applicability, making the findings more relevant to a wider range of individuals. Furthermore, by specifically recruiting individuals at risk for HDP our study gained valuable insight into how interoceptive awareness may influence the development and progression of HDP. Results from this study should also be considered in light of several limitations. First, the sample size was relatively small, limiting the generalizability of our results. Future studies with larger samples are needed to confirm and extend these findings. Additionally, our study design was cross-sectional, preventing us from establishing causality. Future studies should focus on collecting longitudinal data analyzing changes in disordered eating and interoception throughout pregnancy. Future research should aim to examine whether interoception-based interventions, such as MBIs, mindful eating, or mindfulness-based stress reduction (MBSR) can improve eating behavior and subsequent cardiovascular outcomes in pregnant populations. As well, eating disorder diagnoses were not clinically assessed in this study.

Ultimately, our findings suggest that interoceptive awareness, specifically the ability to remain “unworried” about bodily sensations, is associated with less disordered behavior and lower daytime BP in pregnant women at risk for HDP. Future research should explore targeted mindfulness interventions in this population to determine their effectiveness in preventing HDP and improving maternal health outcomes.

LIST OF ABBREVIATIONS

ABPM

ambulatory blood pressure monitor

BP

blood pressure

BMI

body mass index

CBT

cognitive behavioral therapy

DBP

diastolic blood pressure

HDP

hypertensive disorders of pregnancy

MBI

mindfulness based interventions

MBSR

mindfulness-based stress reduction

MAIA

Multidimensional Assessment of Interoceptive Awareness scale

SBP

systolic blood pressure

TFEQ-R18

Three Factor Eating Questionnaire- Revised

Data availability:

The data that support the findings of this study are available from the corresponding author upon reasonable request.