Restricting diet for perceived health benefit: A mixed‐methods exploration of peripartum food taboos in rural Cambodia

Abstract Food taboos encompass food restrictions practiced by a group that go beyond individual preferences. During pregnancy and lactation, food taboos may contribute to inadequate nutrition and poor maternal and infant health. Restriction of specific fish, meat, fruits and vegetables is common among peripartum women in many Southeast Asian countries, but data from Cambodia are lacking. In this mixed‐methods study, 335 Cambodian mothers were asked open‐ended questions regarding dietary behaviours during pregnancy and up to 24 weeks postpartum. Descriptive statistics and content analysis were used to characterize food taboos and multiple logistic regression analyses were conducted to identify predictors of this practice. Participants were 18–44 years of age, all of Khmer ethnicity and 31% were primiparous. Sixty‐six per cent of women followed food taboos during the first 2 weeks postpartum, whereas ~20% of women restricted foods during other peripartum periods. Pregnancy taboos were often beneficial, including avoidance of sugar‐sweetened beverages, coffee and alcohol. Conversely, postpartum avoidances typically included nutrient‐dense foods such as fish, raw vegetables and chicken. Food taboos were generally followed to support maternal and child health. No significant predictors of food taboos during pregnancy were identified. Postpartum, each additional live birth a woman had reduced her odds of following food taboos by 24% (odds ratio [95% confidence interval]: 0.76 [0.61–0.95]). Specific food taboo practices and rationales varied greatly between women, suggesting that food taboos are shaped less by a strict belief system within the Khmer culture and more by individual or household understandings of food and health during pregnancy and postpartum.

significant predictors of food taboos during pregnancy were identified. Postpartum, each additional live birth a woman had reduced her odds of following food taboos by 24% (odds ratio [95% confidence interval]: 0.76 [0.61-0.95]). Specific food taboo practices and rationales varied greatly between women, suggesting that food taboos are shaped less by a strict belief system within the Khmer culture and more by individual or household understandings of food and health during pregnancy and postpartum.
K E Y W O R D S diet, dietary restriction, food and nutrition, food taboos, global health, lactation, maternal health, peripartum period 1 | INTRODUCTION Nutrition is globally recognized as an important pillar of health, especially in low-and middle-income countries (Development Initiatives, 2020;World Health Organization, 2020). Every year, three million children under 5 years of age die from malnutrition and many more suffer its irreversible consequences (UNICEF, 2020), such as impaired cognitive and motor development. Pregnancy and lactation are critical periods in which nutritional requirements are particularly high, placing women and their infants at increased risk of malnutrition. Malnourished women experience more adverse pregnancy outcomes, such as postpartum haemorrhage (Todd et al., 2019), fetal loss (Black et al., 2008;Victora et al., 2021) and having a low birth weight baby (Blossner & de Onis, 2005;Victora et al., 2021). During lactation, poor nutrition can result in maternal nutrient depletion or even specific micronutrient deficiencies in their infants (Dror & Allen, 2018), hindering maternal health and infant development (Ballard & Morrow, 2013;Black et al., 2008).
There is high peripartum engagement with the formal health system in Cambodia, with 95% of mothers attending antenatal care and 83% delivering in a health facility (National Institute of Statistics, Directorate General for Health & ICF International, 2015). Despite this, traditional practices such as ang pleung, or 'mother roasting', where postpartum women lie on a bed set over a fire for 3-7 days to restore the hot/cold balance (White, 2004), still exist. Ang pleung is discouraged by the Cambodian Ministry of Health (2012) due to the potential health consequences of this ritual (Bazzano et al., 2020), such as delayed breastfeeding initiation (White, 2004;Wren & Chambers, 2011). Nevertheless, ang pleung continues to be practiced, particularly in rural areas (Montesanti, 2011).
Another traditional peripartum practice in Cambodia is the adherence to food taboos, avoidances of certain foods that go beyond reasons of personal tastes or preferences. Food taboos are usually culturally specific customs that prohibit certain food choices and are transferred throughout generations (Iradukunda, 2020). Food taboos seem to be particularly prevalent in the peripartum period as a means of protecting the health of women and children (Köhler et al., 2019;Meyer-Rochow, 2009). However, adhering to food taboos has been seen to reduce dietary diversity (Smith et al., 2022) and alter women's nutrient intakes (Barennes et al., 2009;Koon et al., 2005), which can lead to malnutrition (Köhler et al., 2019).
Peripartum food taboos have been reported in societies across the globe (Iradukunda, 2020;Kavle & Landry, 2018;Meyer-Rochow, 2009). In Southeast Asia, common avoidances include foods from the sea and freshwater, meats, eggs (Köhler et al., 2019), fruit, water spinach, cabbage and other vegetables (Köhler et al., 2018). In Laos, up to 98% of women restrict intake of certain foods in their postpartum diet (Barennes et al., 2015). Little is known about the prevalence and impact of peripartum food taboos in Cambodia; some previously identified restrictions include spicy foods (Richman et al., 2010;Wallace et al., 2014), coconut milk, porridge (Montesanti, 2011), varieties of fish (Wallace et al., 2014;White, 2002), pig's head and buffalo meat (White, 2002). However, most reports in Cambodia stem from small, qualitative studies, so the scale and prevalence of food taboos are unknown. Given the higher nutritional needs during

Key messages
• Most food taboos were practiced early postpartum, with women restricting a wide range of foods in their diets based on differing rationales. Rather than tradition or superstition, rural Cambodian women followed food taboos most commonly to support varying aspects of maternal and infant health.
• Although food avoidances in pregnancy had potential to benefit health (e.g., sugar-sweetened beverages or alcohol), postpartum restrictions included nutrient-rich foods (e.g., fish, chicken, or beef).
• The widespread practice of food taboos in Cambodia, yet lack of cohesiveness and predictors for these practices, warrants a deeper exploration of peripartum dietary behaviours and their health impacts. pregnancy and lactation, on top of existing food insecurity and malnutrition (Boonyabancha et al., 2019;National Institute of Statistics, Directorate General for Health, & ICF International, 2015), any peripartum dietary restriction can present a risk to maternal and child health. With this, this study aimed to explore the food taboos of pregnant and lactating women in Cambodia.
Specifically, the objectives were (i) to identify foods that are intentionally avoided in the maternal diet, along with the prevalence of these practices, (ii) to describe the rationale for adherence to food taboos and (iii) to identify sociodemographic and health predictors of adherence to food taboos.

| METHODS
This study was a mixed-method, secondary analysis of data collected from the Trial of Thiamine Supplementation in Cambodia (Whitfield et al., 2019)

| Data collection
Data were collected in participants' homes using intervieweradministered questionnaires at 2, 12 and 24 weeks postpartum (Whitfield et al., 2019). At 2 weeks, information was collected on women's sociodemographic and health characteristics, as well as a retrospective collection of dietary practices in pregnancy and the first 2 weeks after childbirth (defined here as early postpartum). During the mid (2 through 12 weeks) and late (12 through 24 weeks) postpartum visits, information was collected on dietary practices since the participant's previous study visit. At each visit, participants were asked if there were any new foods/beverages they intentionally avoided in their diet during pregnancy/postpartum, along with the reasons for each dietary modification. Responses were open-ended, so multiple responses and multiple reasons for avoidances sometimes emerged. All data were collected in Khmer and then translated into English before analysis.

| Data analysis
Descriptive statistics were computed for participant characteristics and dietary practices, presented as n (%) for categorical variables and mean (SD) for continuous variables. Shapiro-Wilk test was applied to assess the normality of data distribution (Mishra et al., 2019); data with non-normal distributions are presented as median (interquartile range).
IBM SPSS v. 26.0 for Windows (IBM Corp, 2018) was used to perform quantitative data analyses, with a significance level of p < 0.05.
To illustrate the types of foods/beverages avoided in the maternal diet, each item was categorized using the minimum dietary diversity for women (MDD-W), a validated indicator for assessing population-level dietary diversity among women of reproductive age (INDDEX Project, 2018). All basic and optional MDD-W food groups were considered for this analysis. The category 'condiments and seasonings' was only used when items falling into that category were individually reported by women (e.g., chilli peppers were categorized as a condiment/seasoning, but the seasoning in soups or mixed dishes was not specifically recorded).
Content analysis was employed to describe women's rationales for adhering to peripartum food taboos (Elo & Kyngäs, 2008;Vaismoradi et al., 2013). An inductive approach was used for coding to allow codes to be derived directly from the data set. Codes were subsequently organized into larger categories representing study participants' perspectives and beliefs (Braun & Clarke, 2006;Nowell et al., 2017).
Logistic regression models with backward elimination were built to identify sociodemographic and health characteristics predictive of peripartum food avoidance (McDonald, 2015). Adherence to food taboos was presented as a dichotomous variable (i.e., yes/no).

| RESULTS
In total, 335 women were enroled in the study (see the Supporting Information: Figure for flowchart and exclusion reasons). Participants had a mean (SD) age of 28 (6) years and had experienced 2.5 (1.4) pregnancies (Table 1). All women were of Khmer ethnicity and nearly all (99%) were married. Most had <7 years of formal education (60%) and were part of the three lower relative wealth quintiles (77%).

| Description of food taboos
Although only 18% of women practiced food taboos during pregnancy, 71% practiced them postpartum (Table 2). Taboos were most common in early postpartum, with 66% of women restricting foods in their diets in the first 2 weeks after childbirth. By mid and late postpartum, food taboos were followed by only 17% and 18% of women, respectively. With this low adherence, the median number of foods avoided per woman was zero for all time points, except early postpartum (median of 1). The types of food and beverages avoided by women were far from universal. For instance, in early postpartum, 114 unique items were considered taboo by women in the study sample. The most frequently avoided foods during pregnancy were spicy foods, energy drinks and coffee, whereas postpartum avoidances were commonly fish without scales (e.g., eel, catfish), raw vegetables and fermented foods.

| MDD-W classification of food taboos
Study participants considered foods from nearly all MDD-W food groups taboo (see Table 3). During pregnancy, the most frequently avoided food group was condiments/seasonings (25%), particularly chillis, spicy food and fish paste. Sugar-sweetened beverages such as energy drinks, soft drinks, sweetened condensed milk and juices were also commonly avoided (23%), as were foods that fell in the 'other' category (22%; e.g., alcohol, coffee and betel leaf). Postpartum avoidances were primarily of meat/fish and 'other vegetables' (39% and 22% of food avoidances, respectively; the category 'other vegetables' includes vegetables that are not particularly rich sources of vitamin A). Fish without scales, chicken, beef and buffalo were among the most avoided animal products. Taboo vegetables included bamboo shoots, sponge gourd, eggplant, wax gourd and cucumbers.

| Predictors of food taboos
Food taboos were far more prevalent postpartum than in pregnancy (71% vs. 18%), with most of these avoidances occurring in the first 2 weeks after delivery. This finding is congruent with traditional beliefs in Cambodia, as women are thought to be in a particularly fragile, weak state after childbirth (Bazzano et al., 2020;Montesanti, 2011;White, 2002) and participants largely restricted foods in their diets to protect maternal health. The high prevalence of food taboos during the first 2 weeks postpartum is intriguing as it coincides with the period during which traditional postpartum practices such as ang pleung (mother roasting) are performed (Bazzano et al., 2020). Previous studies have indicated that women intentionally restrict and add foods during ang pleung (Hoban, 2002;MacLellan, 2010;White, 2002), but the relationship between postpartum dietary behaviours and traditional birthing practices has not been thoroughly explored. Some evidence suggests that women consume foods based on their hot/cold (yin/yang) properties during ang pleung, and that foods should be reintroduced with caution following this ritual (Hoban, 2002;White, 2004); however, few participants in our study referred to the former concept and none to the latter when explaining their rationales for dietary restrictions.
Our study revealed that peripartum food taboos were slightly less prevalent in Cambodia than in neighbouring Southeast Asian countries. A study in Indonesia observed 28% of women following There may have also been a cultural shift in food taboos during the famine of the Khmer Rouge regime (1975)(1976)(1977)(1978)(1979). Food scarcity during this period forced many women to eat any available food to F I G U R E 1 Categorization of women's rationales for adhering to food taboos. Circle size represents the overall frequency of each category, whereas colours represent the intended beneficiary of food avoidances. Promoting maternal health was provided as a rationale for 69% of food avoidances, whereas promoting the health of the fetus/infant accounted for 21% of rationales. Other reasons included following traditions (4%), supporting breastmilk production (2%), enhancing the baby's beauty (1%), preventing maternal death (1%) and preventing fetal death (0.5%).
ensure survival (Hoban, 2002  ̶ Causes diarrhoea ̶ Causes stomach aches ̶ Causes sickness ̶ Causes cough ̶ Baby will get a cough ̶ Causes food poisoning a ̶ Causes bloating/gas a ̶ Causes allergies a ̶ Affects breastmilk production a ̶ Advice from elders a ̶ Advice from doctor a ̶ Baby will get a cold ̶ Baby will get diarrhoea ̶ Baby will get sick ̶ Baby will get allergies ̶ Baby will get a fever a ̶ Affects health a ̶ Affects womb a ̶ Causes reproductive health issues a ̶ Causes wound infection a ̶ Affects wounds a Note: The postpartum period includes responses from early (0-2 weeks), mid (2-12 weeks) and late (12-24 weeks) postpartum. All reasons listed refer to the mother unless otherwise noted. Some participants provided more than one reason for avoiding specific foods. a Reason provided by only one woman for that specific food/beverage. b Reason provided by ≥50% of women for that specific food/beverage. T A B L E 4 (Continued) LABONTÉ ET AL. | 7 of 12 women and children. Many of the foods avoided have the potential for adverse health outcomes, including alcohol (Flak et al., 2014), coffee (Gleason et al., 2021;Qian et al., 2020), energy drinks (Qian et al., 2020) and other sugar-sweetened beverages (Jen et al., 2017).
While pregnancy guidelines in Cambodia recommend avoiding alcohol, restricting the intake of foods high in sugar and discouraging concurrent coffee and iron-folic acid supplement consumption (National Nutrition Program, 2009), it does not appear these food taboos were based on medical advice given that only three participants noted avoiding these foods on the recommendation of a medical professional. Although the restriction of nutrient-poor foods such as sugar-sweetened beverages is generally perceived as a healthful practice, the reduction in total caloric intake that may result can present a health risk to the 14% of women in Cambodia, who are Maternal and infant health was the strongest driver of dietary restrictions among our study participants and has been seen as a primary reason for practicing food taboos across the globe (Iradukunda, 2020;Jamaludin, 2014;Köhler et al. 2018Köhler et al. , 2019Meyer-Rochow, 2009;Sein, 2013). Many of the specific health concerns reported by our participants have been described in other studies in Cambodia as experiences of toa, a physical or psychological illness that is believed to afflict postpartum women who act against cultural customs (Turner et al., 2017). For instance, women in our study reported restricting foods to avoid diarrhoea, stomach aches (Turner et al., 2017;White, 2002White, , 2004, weakness, headaches, vomiting (White, 2002(White, , 2004, seizures (Turner et al., 2017), stiff backbone, jaw tightness, dry skin and inadequate breastmilk production (White, 2004), which have all been previously associated in the literature with toa. Some of the health outcomes related to avoidance of specific foods would be considered plausible through the lens of Western medicine. For example, fish, meats and raw vegetables were avoided by some women out of fear of gastrointestinal illnesses. Biological contamination of meats and vegetables is highly prevalent in Cambodia (Thompson et al., 2021) and pregnant women are more susceptible to foodborne pathogens (Smith, 1999); therefore, concerns regarding the consumption of these foods are warranted.
Other food taboos observed in our sample were based on biologically implausible mechanisms. For example, one participant avoided drinking coffee during pregnancy out of fear of getting diabetes, whereas other women eliminated pineapple, chilli, porridge and eggplant from their pregnancy diets to prevent getting 'thick' amniotic fluid. Implausible outcomes such as these have been identified in other regions in Cambodia and throughout Southeast Asia. For example, in Thailand, shellfish and specific relishes are taboo during pregnancy, as they are believed to 'prevent the perineum from drying out properly after giving birth' (Liamputtong et al., 2005;p. 143). In Laos, white buffalo, chicken meat, fermented fish, beef and duck are avoided postpartum as they are believed to cause leprosy (Holmes et al., 2007).
Healthcare providers can play a key role in managing such food avoidances by identifying health concerns women have during antenatal care visits and empowering women with knowledge of the factors that can cause the outcomes they aim to avoid. For instance, the fear of inhibiting breastmilk production through the consumption of specific foods was reported by women in Cambodia, yet empirical evidence shows that dietary intake has little impact on breastmilk volume (Ballard & Morrow, 2013). If these women were equipped with specific knowledge of nutrition and lactation, they may no longer feel the need to restrict their dietary intake to support breastfeeding.
The lack of sociodemographic and health predictors of food taboos is intriguing. Mixed results on this topic have been reported in the literature. In Laos, women with higher socioeconomic status and education, who are older and attend more antenatal care visits were less likely to have a restrictive diet (Smith et al., 2022), while in Malaysia, maternal age, education and household income were not associated with food taboos (Mohamad & Ling, 2016). Our finding that multiparity decreases the practice of food taboos has also been observed in Laos (Barennes et al., 2009;Smith et al., 2022) and Malaysia (Mohamad & Ling, 2016). Multiparous women may be less concerned with the risks of eating taboo foods as they have already had a successful pregnancy, delivery and postpartum experience, which can lead to reduced levels of fear or uncertainty with these periods of the life cycle. In northern Cambodia, Turner et al. (2017) observed that primiparous women used traditional medicines more frequently than during their subsequent pregnancies, as they were more reliant on the advice of elders to guide them through this new experience. Prior pregnancy experiences may therefore enhance self-confidence and help women make independent decisions about peripartum dietary practices. We did not ask women about their sources of information on taboos, yet 24 women described that they followed food taboos due to the advice of others, with elders and family members being the most commonly cited advisors. Women with more intergenerational connections may therefore be at higher risk of following food taboos, but further research is needed to confirm this hypothesis. The current absence of predictors for food taboos makes it more difficult for healthcare providers and public health officials to identify groups of women who would benefit from additional nutritional and health counselling. As such, it is essential for healthcare providers to ensure they discuss food taboos and dietary practices with all women during antenatal care visits.
To our knowledge, this study provides the most in-depth published analysis of maternal food taboos in Cambodia to date. A key strength of this study was its large sample size and the use of open-ended questions to obtain information on foods taboos, which enabled us to capture nuances in behaviours and beliefs. The longitudinal design of this study was another strength: collecting data for four different time points helped highlight clear differences in food taboos throughout the peripartum period. The main limitation of this study was the use of long-term recall methods. Given that this study was a secondary analysis, data collection protocols were optimized for the objectives of the overarching thiamine trial rather than for the current analysis. Participants were asked to recall their dietary behaviours during pregnancy and between study visits. Such an approach could have led to recall bias and as such to missing data, as some women may forget which foods they avoided in their diet.
However, heightened nutrition awareness has been identified during pregnancy (Szwajcer et al., 2012), which could reduce the risk of women forgetting their recent dietary modifications. Finally, while we asked women to explain motivations for their dietary modifications, we did not explicitly ask about ang pleung, a potential limitation of our study given that most restrictions occurred in early postpartum, a time when mother roasting would have been most common.

| CONCLUSION
Peripartum food taboos were commonly practiced in Cambodia, yet were limited in duration and severity. Although 71% of women avoided at least one food or beverage postpartum, most avoidances occurred in the first 2 weeks after childbirth. Food taboos were far less prevalent during pregnancy and during this time, many had the potential to confer positive health impacts.
This study revealed discrepant dietary practices and beliefs among women in Cambodia. First, a wide breadth of foods was avoided in women's diets, with certain foods being considered harmful by some women and beneficial by others. Second, rationales for dietary modifications generally centred on supporting the health of women and their children, yet the health outcomes believed to be associated with specific foods varied between participants. Third, sociodemographic and health characteristics were found to have little influence on food avoidances. These findings suggest that peripartum food taboos are shaped less by a strict belief system within the