Patient–physician communication on herbal medicine use during pregnancy: a systematic review and meta-analysis

Abstract

Introduction

Lack of transparent communication between patients and physicians regarding the use of herbal medicine (HM) presents a major public health challenge, as inappropriate HM use poses health risks. Considering the widespread use of HM and the risk of adverse events, it is crucial for pregnant women to openly discuss their HM use with healthcare providers. Therefore, this systematic review and meta-analysis aims to estimate the pooled prevalence of pregnant women’s HM use and disclosure to healthcare providers and to examine the relationship between HM disclosure and various maternal and child health (MCH) measures.

Methods

A systematic search of five databases was conducted for cross-sectional studies on HM use during pregnancy published from 2000 to 2023. Data extraction followed a standardised approach, and Stata V.16.0 was used for data analysis. Also, Spearman’s correlation coefficient was calculated to examine the association between use and disclosure of HM and various MCH indicators.

Results

This review included 111 studies across 51 countries on the use of HM among pregnant women. Our findings showed that 34.4% of women used HM during pregnancy, driven by the perception that HM is presumably safer and more natural than conventional medical therapies. However, only 27.9% of the HM users disclosed their use to healthcare providers because they considered HM as harmless and were not prompted by the healthcare providers to discuss their self-care practices. Furthermore, a significant correlation was observed between HM disclosure and improved MCH outcomes.

Conclusion

Inadequate communication between pregnant women and physicians on HM use highlights a deficiency in the quality of care that may be associated with unfavourable maternal outcomes. Thus, physician engagement in effective and unbiased communication about HM during antenatal care, along with evidence-based guidance on HM use, can help mitigate the potential risks associated with inappropriate HM use.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• As the use of herbal medicine (HM) increases among pregnant women, promoting open and proactive patient–physician communication is essential to help women make informed decisions about HM use during pregnancy, prevent potential harm and ensure optimal care. However, existing reviews on pregnant women’s HM use provide limited insights into communication between pregnant women and their physicians regarding HM use.

WHAT THIS STUDY ADDS

• Despite the widespread use of HM among pregnant women, less than one-third of women discussed their use with obstetric care providers due to the perceived safety of HM and the lack of inquiry from physicians.

• Furthermore, the non-disclosure of HM use showed a significant correlation with higher maternal and neonatal mortality, lower proportions of births attended by skilled birth attendants and lower antenatal care coverage. This correlation may be influenced by suboptimal patient–physician communication in regions with poorer maternal and child health (MCH) outcomes.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• To prevent potential adverse MCH outcomes resulting from inappropriate HM use, effective patient–physician communication regarding self-care practices should be facilitated.

• Advocating training for healthcare providers on commonly used HM, raising awareness about the safe use, benefits and potential toxicity of HM through community programmes, and integrating inquiries about HM use into standard antenatal case-taking procedures are critical steps to promote open discussions about HM use during antenatal care.

Introduction

Pregnancy is characterised by hormonal changes that often result in significant physical discomfort and mood swings.^1–3 To address these health issues, expectant mothers commonly turn to self-care approaches, as they fear the potential teratogenic effects of conventional pharmaceutical products.^{4 5} Furthermore, during global health crises, such as the outbreak of COVID-19, self-care practices gained increased attention due to the rise in social distancing, self-isolation and strain on the public health system.^{6 7} Self-care approaches have proven beneficial not only because they allow individuals to actively participate in their own healthcare and exercise patient autonomy, but also because they improve access to healthcare for those facing geographical and economic barriers.⁸

However, certain self-care modalities, such as herbal medicine (HM), contain bioactive compounds that can pose risks to both the mother and the fetus if misused.^9–15 While safety and efficacy data exist for some HM, many others lack conclusive evidence and, in some cases, rely on anecdotal evidence and information from mass media and other digital platforms.^{9 16–18} Specifically, the abundance of information available online blurs the line between lay and expert knowledge, making it challenging for individuals to discern accurate information.⁸ Furthermore, the use of HM is often cited as one of the reasons for delay in seeking professional healthcare services and is associated with inadequate antenatal care (ANC) utilisation,^19–24 potentially increasing the risk of unfavourable maternal and child health (MCH) outcomes.^{25 26} Therefore, effective management of pregnant women’s HM use has emerged as a significant public health concern in order to prevent the misuse of HM and mitigate subsequent threats to MCH.

Promoting patient–doctor communication on pregnant women’s health regimen during ANC is a non-pharmacological strategy that enables healthcare providers to support the women in the safe and appropriate use of HM.^27–29 This approach allows healthcare providers to assist pregnant women in making informed decisions about their use of self-care interventions and provides an opportunity for early intervention if potential harm arises. Nevertheless, as many as 93% of complementary medicine (CAM) users do not discuss their use with conventional healthcare providers because the importance of discussing HM use is often overlooked.³⁰ Additionally, the lack of clinical guidelines encouraging healthcare providers to discuss patients’ use of additional healthcare approaches during medical consultations, coupled with limited physician knowledge of HM, creates barriers to effective communication.³¹ Consequently, inadequate patient–physician communication regarding HM use, along with the neglect of patient preferences and choices in healthcare, hinders coordination between complementary health practices and conventional medical care, potentially exposing HM users to health risks.^{27 32–35}

Despite the importance of open and effective communication on self-care practices between pregnant women and their obstetric care providers,^{9 27 29} discussions about HM use are relatively rare and mostly initiated by patients rather than healthcare providers in current clinical practice. Thus, patient disclosure of HM use serves as a critical first step for patient–doctor communication on HM use and becomes an important factor in achieving optimal MCH outcomes. However, previous studies on the use of HM during pregnancy offer limited insights into the discussion between pregnant women and their healthcare providers about HM use and focus primarily on patterns and characteristics of HM use.^36–39 Therefore, this review aims to estimate the pooled prevalence of pregnant women’s HM use and examine their disclosure practices to healthcare providers, while also exploring the potential association between HM disclosure and improved MCH status.

Methods

This review was registered in the International Prospective Register of Systematic Reviews (PROSPERO) database (CRD42022352846), and the findings of this systematic review are presented following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines (online supplemental appendix 1).⁴⁰

Search strategy

An electronic database search was performed on 16 March 2023, using PubMed/MEDLINE, EMBASE (Ovid), CINAHL, Scopus and Web of Science to identify studies that report pregnant women’s HM use and their disclosure of HM use to healthcare providers. The search was conducted using MeSH and related keywords of three major terms: herbal medicine, pregnancy and cross-sectional study (online supplemental appendix 2). The search was limited to cross-sectional studies because such study designs are commonly used to assess participants’ health behaviours and to improve the comparability of the findings.^{9 30} The publication years were restricted to 2000–2023, considering the increasing popularity of HM use over the last two decades.⁴¹ The language was restricted to English-language only. An ancestral search was also conducted to identify any additional relevant articles.

Eligibility criteria

This review includes studies that are: (1) conducted between 2000 and 2023 and (2) report cross-sectional data on the use and disclosure of HM during pregnancy. For our study, the time frame of pregnancy included the period from conception to birth, excluding HM use during labour or postpartum. This review specifically focused on HM use to reduce the heterogeneity among the studies due to varying definitions of complementary and CAM. Moreover, studies focusing on a single HM modality or a specific indication for use were excluded, as they may not provide an accurate representation of the prevalence rates for HM use and could affect the estimation of the general HM use during pregnancy.

Study selection and risk of bias assessment

Three reviewers completed the full-text review of the selected studies to examine various study characteristics (ie, publication year, study design, setting and research subjects) and inclusion of key outcomes (ie, the prevalence of HM use and disclosure, as well as any additional quantitative data related to patient–physician communication regarding HM use). In cases where a study explored multiple CAM modalities, only the prevalence of HM use was extracted. After independently extracting the data, the three reviewers compared their results and resolved any disagreements with the assistance of senior researchers.

The quality of selected studies was evaluated using the critical appraisal tool developed by Hoy et al,⁴² which is designed to evaluate the risk of bias in prevalence studies and consists of 11 items. Each item was scored either 0 or 1, with lower scores indicating a lower risk of bias. Studies that scored 7 or higher on the quality assessment scale were excluded from the final analysis.

Data synthesis and statistical analysis

A meta-analysis was conducted to estimate the pooled prevalence of HM use and disclosure among pregnant women and the corresponding 95% CIs, using the Stata V.16.0. Given the high heterogeneity among the studies (I²=99.00%), a random effect meta-analysis was conducted, and the funnel plot and the Egger test were used to examine publication bias (online supplemental appendices 3 and 4). A subgroup meta-analysis was performed to analyse differences in the HM use and disclosure rates by geographical region (ie, by countries and by continents: Africa, Americas, Eastern Mediterranean, Europe, South-East Asia and Western Pacific) and the study year (ie, studies conducted between 2000–2010 and 2011–2023).

Furthermore, the HM use and disclosure rates were also compared based on the MCH status of the study settings because we hypothesised that pregnant women’s disclosure of HM use is associated with better MCH outcomes. As open communication on HM use between expectant mothers and their healthcare providers could help prevent inappropriate HM use,^{23 27} we considered the disclosure of HM as an intervention for achieving optimal MCH. The MCH status of each study setting was determined by assessing the major MCH indicators, such as maternal mortality rate (MMR), neonatal mortality rate (NMR), skilled birth attendant (SBA) utilisation rate and the coverage of at least four ANC visits at the time of each study.

To perform the subgroup analysis based on MCH status, each study was categorised based on whether the country of the study setting had achieved the WHO’s Sustainable Development Goal (SDG) target for MMR (70/100 000), NMR (12/10 000) and SBA utilisation (90%).^{43 44} Regarding ANC utilisation rate, countries were grouped based on whether the study setting met the global target of at least four ANC coverage (90%).⁴⁵ Country-specific MCH indicator values were obtained from the WHO Data Platform,⁴⁶ UNICEF Data Warehouse⁴⁷ and the World Bank Open Data.⁴⁸ In cases where country-specific data were unavailable, the researchers assumed the MCH situation of the country, classifying it as below or above the global average based on the country’s economic status (ie, gross domestic product per capita).^{49 50}

Lastly, using the SPSS V.26.0., Spearman’s correlation coefficient and scatter plots were computed to explore the potential association between various MCH measures and HM utilisation and disclosure rates. A p<0.05 was considered statistically significant for all analyses.

Patient and public involvement

There was no direct participation or input from patients or the public in the design or conduct of this systematic review and meta-analysis. However, our research findings will contribute to inspiring healthcare providers and policy-makers to develop interventions that improve the quality of patient care and ensure the safety and well-being of pregnant mothers and their children.

Results

Of 23 767 studies retrieved from the database searches, 294 studies met the criteria for full-text screening. Among these, 188 records were excluded from the review for one of the following reasons: non-cross-sectional study design, inappropriate study population, unavailability of full English text and inadequate reporting of pregnant women’s HM use (figure 1). Additionally, 1376 records were identified by reviewing the reference lists of the remaining 106 studies. After removing duplicates, 238 additional studies were included for further review, and 35 studies met the eligibility criteria after the full-text appraisal. Furthermore, 16 studies were excluded due to using data collected before 2000 and identical datasets. Consequently, 121 studies were subjected to a risk of bias assessment, and after excluding 10 studies with a high risk of bias,^51–60 a total of 111 studies were included in the final review (online supplemental appendix 5).

Figure 1.

Flow diagram of search strategy.

Characteristics of studies

Among 111 studies, 102 studies were cross-sectional in design.^{27 61–161} Three studies used mixed-method approaches,^162–164 and six studies had non-cross-sectional designs (ie, longitudinal and case–control),^165–170 but reported cross-sectional data on HM use during pregnancy (online supplemental appendix 6). Regarding the geographical distribution of included studies, a majority of them were conducted in Africa.66 69 77 81 82 89 91 95 98 100 101 105 106 115 116 119 121–123 129 132–135 137–140 143 144 146 150 153 156 158–164 Lastly, 21 out of 111 studies presented findings related to pregnant women’s disclosure of HM use to their healthcare providers.27 70 73 83 85 86 91 109 110 115 117 122 125 126 129 133 142 149 155 160 164

Prevalence of HM use among pregnant women

The pooled prevalence of HM use during pregnancy was 34.4% (95% CI 31.1% to 37.8%), ranging from 2.2% to 93.8% (table 1, figure 2).^{55 136} Among the 51 countries included, HM use was the lowest in Trinidad and Tobago and the highest in Indonesia (figure 3).^{76 101} The prevalence varied significantly by study year (p=0.048) and geographical region (p<0.001).

Table 1.

Pooled prevalence of HM use and HM disclosure by study characteristics

Characteristics	HM utilisation††				HM disclosure††
	Mean*	95% CI	n†	P value	Mean*	95% CI	n†	P value
Overall	34.4%	31.1% to 37.8%	111		27.9%	20.0% to 36.5%	21
Study year
2000–2010	28.8%	23.1% to 34.7%	29	0.048	24.3%	11.2% to 40.3%	3	0.643
2011–2023	36.1%	31.9% to 40.5%	82		28.5%	19.6% to 38.2%	18
Geographical region
Africa	37.6%	31.9% to 43.6%	41	<0.001	9.7%	4.6% to 16.5%	7	<0.001
Americas	16.0%	11.2% to 21.3%	8		38.1%	26.1% to 51.2%	1
Eastern Mediterranean	39.5%	30.9% to 48.5%	28		44.5%	36.2% to 52.9%	7
Europe	27.5%	21.0% to 34.5%	19		30.1%	9.9% to 55.4%	3
South-East Asia	60.5%	30.9% to 86.3%	5		35.4%	18.2% to 54.7%	3
Western Pacific	36.1%	30.9% to 41.5%	12		–	–	–
Maternal mortality rate (SDG target 3.1.1)‡
Less than 70/100 000 live births	30.8%	26.9% to 34.9%	62	0.008	40.6%	29.6% to 52.2%	10	0.003
Greater than 70/100 000 live births	40.4%	34.6% to 46.2%	49		17.8%	9.4% to 28.2%	11
Neonatal mortality rate (SDG target 3.2.2)§
Less than 12/1000 live births	31.3%	27.3% to 35.4%	58	0.033	37.7%	27.4% to 48.5%	9	0.036
Greater than 12/1000 live births	39.0%	33.2% to 44.9%	53		21.1%	11.4% to 32.8%	12
Proportion of births attended by skilled health personnel (SDG target 3.1.2)¶
Less than 90%	39.7%	33.9% to 45.6%	48	0.019	17.8%	9.4% to 28.2%	11	0.003
Greater than 90%	31.3%	27.4% to 35.4%	63		40.6%	29.6% to 52.2%	10
Antenatal care coverage—at least four visits (global target)**
Less than 90%	37.1%	32.0% to 42.4%	65	0.104	23.8%	15.1% to 33.6%	15	0.107
Greater than 90%	31.5%	27.2% to 36.0%	51		38.8%	23.7% to 55.0%	6

*Pooled estimate.

†Number of studies included in each subgroup.

‡SDG target for maternal mortality rate=70/100 000 live births.⁴³

§SDG target for neonatal mortality rate=12/1000 live births.⁴³

¶SDG target for proportion of deliveries attended by skilled birth attendants=90%.⁴⁴

**Global target for proportion of pregnant women with at least four antenatal care uptake=90%.⁴⁵

††Estimated using a random effects model.

HM, herbal medicine; SDG, sustainable development goal.

Figure 2.

Pooled prevalence of pregnant women’s HM use. Forest plot of random effect meta-analysis. A total of 111 studies reporting the prevalence of pregnant women’s HM use were included for the pooled estimation of HM use. The black diamond dots and dashed line passing through represent the effect size and corresponding 95% CI reported in individual studies, and the green diamond on the bottom and the size of its lateral tips denote the pooled effect size and its 95% CI.

Figure 3.

Map of prevalence estimates of herbal medicine use by country. The 111 studies included in this review reported the prevalence of HM use across 51 different countries. Darker colours represent a higher prevalence of HM use.

A comparison of HM use with various MCH indicators revealed that the prevalence was significantly higher in the countries that have not reached the SDG target for MMR (p=0.008), NMR (p=0.033) and SBA utilisation (p=0.019). Spearman’s correlation analysis also revealed a positive correlation between HM use and MMR (p=0.013) and NMR (p=0.011), while a negative correlation was found between HM use and SBA utilisation (p=0.033) and the coverage of at least four ANC visits (p=0.011; table 2 and figure 4).

Table 2.

Correlations between HM utilisation and disclose rate and MCH indicators

Variables	HM utilisation		HM disclosure
	Spearman’s correlation coefficient (rs)	P value	Spearman’s correlation coefficient (rs)	P value
Maternal mortality rate	0.222	0.013	−0.545	0.013
Neonatal mortality rate	0.227	0.011	−0.597	0.004
Proportion of births attended by SBA	−0.190	0.033	0.587	0.005
Antenatal care coverage—at least four visits	−0.243	0.011	0.581	0.007

HM, herbal medicine; MCH, maternal and child health; SBA, skilled birth attendant.

Figure 4.

Correlation between HM utilisation or disclosure rate (%) and MCH indicators. (A) maternal mortality rate (per 100 000 live births), (B) neonatal mortality rate (per 1000 live births), (C) the proportion of births attended by skilled health personnel (%) and (D) antenatal care coverage—at least four visits (%). Scatter plot between the prevalence of HM use or disclosure (x-axis) and MCH indicators (y-axis). The blue dots represent the estimation of various MCH indicators of the study setting at the time each study was conducted. The red line denotes the fitted line of each plot.

Pregnant women’s disclosure of HM use to their healthcare providers

Overall, 27.9% (95% CI 20.0% to 36.5%) of pregnant women disclosed the HM use to their healthcare providers, ranging from 4.8% to 65.8% across different studies (table 1, figure 5).^{85 122} Geographical variations were significant (p<0.001), with the highest disclosure rate observed in the Eastern Mediterranean region (44.5%) and the lowest in Africa (9.7%).

Figure 5.

Pooled prevalence of pregnant women’s disclosure of HM use. Forest plot of random effect meta-analysis. Out of 111 studies on pregnant women’s HM use, 21 studies reported the HM disclose rate and were included for the pooled estimation of HM disclosure. The black diamond dots and dashed line passing through represent the effect size and corresponding 95% CI reported in individual studies, and the green diamond on the bottom and the size of its lateral tips denote the pooled effect size and its 95% CI.

The disclosure rate was significantly higher in the countries that achieved the SDG targets for MMR (p=0.003), NMR (p=0.036) and SBA utilisation (p=0.003). Furthermore, negative correlations were observed between the disclosure of HM use and both MMR (p=0.013) and NMR (p=0.004), whereas positive correlations were identified between HM disclosure and the utilisation of SBA (p=0.005) and achieving at least four ANC coverage (p=0.007; table 2 and figure 4).

Additional findings regarding patient–physician communication on HM use

A significant percentage of women (91.7%) reported that their physicians did not inquire about HM use during ANC consultations (table 3).¹²² Furthermore, the majority of pregnant women perceived the physicians’ response to HM disclosure as indifferent (35.3%), with a higher proportion of women receiving warnings about the harmful effects of HM (34.5%) rather than recommendations on HM use (8.0%).^{91 109 134} Lastly, the most commonly reported reasons for pregnant women to withhold information about their HM use were the perception that HM is harmless (53.6%) and the lack of inquiry from healthcare providers (39.4%).27 83 117 125 126 129 133 149 155 164

Table 3.

Additional findings related to disclosure of HM use

Categories	No of studies	Findings
		Pooled estimate	Reported values
Inquired by healthcare providers regarding HM use
Yes	1122	8.3%	8.3%122
No		91.7%	91.7%122
Physicians’ response regarding HM use
Showed indifference	391 109 134	35.3%	31.0%109, 40.0%91
Advised them about harmful/side effects		34.5%	20.0%91, 49.0%109
Discouraged HM use		28.1%	7.0%109, 20.0%91, 57.4%134
Recommended herbal medicine use		8.0%	8.0%109
Discouraged the use of prescription medicine		5.0%	5.0%109
Level of satisfaction on patient-doctor communication on HM use
Very satisfied	186	55.0%	55.0%86
Fairly satisfied		27.0%	27.0%86
Somewhat satisfied		18.0%	18.0%86
Not satisfied		0.0%	0.0%86
Reason for non-disclosure
HM is not harmful	1027 83 117 125 126 129 133 149 155 164	53.6%	31.0%126, 36.4%129, 93.3%125
Healthcare providers did not ask		39.4%	3.0%126, 29.6%129, 30.7%117, 39.4%27, 51.7%149, 52.2%164, 69.2%83
Forgot to inform the doctor		23.5%	0.9%149, 4.0%126, 17.4%27, 32.2%164, 36.3%155, 51.2%117
It is not important to disclose		21.6%	3.3%164, 7.1%117, 18.0%126, 27.6%83, 29.4%27, 44.8%149
Routine use of HM		16.0%	16.0%83
Without any particular reason		10.0%	1.2%83, 19.0%126
Afraid of doctors’ response		6.6%	1.0%126, 3.6%83, 3.8%133, 9.4%117, 11.0%27, 12.2%164
Did not visit healthcare providers/used HM before visiting the doctor		3.2%	1.8%27, 2.0%126, 2.6%149, 6.8%83
Potential predictors of HM disclosure	127	secondary education level (OR 2.25, 95% CI 1.12 to 4.51), four or more visits to ANC (OR 3.74, 95% CI 1.41 to 9.91)27

ANC, antenatal care; HM, herbal medicine.

Discussion

This review examined the pattern of HM use among pregnant women and highlighted the importance of discussing their use with healthcare providers. Due to potential toxicity concerns or dissatisfaction with conventional healthcare services, many individuals turn to natural remedies to address their unmet needs.^{91 117 133} Pregnant women from various countries rely on HM because they assume that HM is a safer alternative to conventional medicine, posing fewer risks to the developing fetus.^{171 172} This reliance is substantiated by the traditional practices and scientific validation of several herbs known for their efficacy in addressing physical discomfort during pregnancy.^{12 173 174} For example, ginger, peppermint and chamomile have proven effective in relieving nausea and vomiting,^{173 175–177} while cranberry consumption has demonstrated efficacy in preventing urinary tract infections.^{178 179} Furthermore, garlic has shown preventive properties against pre-eclampsia.¹⁷⁴ Based on these perceived benefits, cultural beliefs and available scientific evidence, more than one-third of the women included in this review reported using at least one type of HM during pregnancy.

Nevertheless, despite the perceived safety and expectations associated with HM as a treatment option for pregnancy-related conditions, certain remedies were found to cause adverse effects and poor delivery outcomes when used inappropriately.^{9 62 180} Although ginger and chamomile are commonly used to alleviate morning sickness, inappropriate use of ginger can lead to stomach irritation, heartburn and even cardiac arrhythmias.27 62 117 181 Prolonged use of chamomile has been associated with a higher risk of preterm delivery and low birth weight.^{182 183} Furthermore, there is insufficient evidence to confirm the safety of many commonly used herbal products in pregnancy, requiring continued research.¹⁵ As such, herbs that are often consumed as food can also possess medicinal properties that may harm the health of pregnant women and their unborn children if used inappropriately. The consequences of inappropriate HM use are particularly profound during the first and third trimesters of pregnancy, as these stages are critical for fetal development and final maturation.^{174 184 185}

To prevent inappropriate HM use, scholars have advocated the need for physicians to be aware of patients’ HM use and engage in open communication about self-care practices.^{27 73 186} Previous studies have shown that open communication regarding patients’ self-medication through medication reconciliation helped to reduce the risk of adverse drug interactions and medication errors.^{187 188} Improved patient–doctor communication was also associated with better patient health outcomes.^{189 190} Similarly, promoting transparent and effective communication between pregnant women and their healthcare providers about self-care activities could serve as a non-medical approach to lower the risk of adverse effects and unfavourable MCH outcomes. Furthermore, as culturally competent and patient-centred care is linked to increased utilisation of essential MCH services,¹⁹¹ physicians can improve the uptake of ANC services by recognising and understanding pregnant women’s health beliefs and practices through non-judgmental communication about HM use.

However, less than one-third of the pregnant HM users discussed their use with healthcare providers, highlighting the need for effective patient–doctor communication regarding the use of additional medical treatments. Significantly higher rates of HM disclosure were observed in countries with better MCH outcomes, as characterised by lower MMR and NMR and a higher proportion of births attended by SBA. Furthermore, a positive correlation was identified between the attendance of four or more ANC sessions and an increased disclosure of HM. This suggests a potential correlation between pregnant women’s disclosure of HM use and the availability of quality maternal healthcare services, along with better patient–doctor communication. Previous studies have already established a positive association between the disclosure of CAM use and more physician visits, as well as a higher quality of the patient–doctor relationship.^192–194 Having four or more ANC visits was also identified as a potential predictor of HM disclosure during pregnancy.²⁷ These findings illustrate the importance of ANC coverage, as it is a foundation for ensuring maternal health. ANC visits provide an opportunity to regularly check pregnant women’s health, discuss any health-related concerns and exchange health information.^{23 195 196} Routine ANC visits can also facilitate effective patient–doctor dialogue about HM use and help prevent inappropriate use of HM.^{27 180}

In addition, healthcare providers were found to play a crucial role in bridging communication gaps related to HM use.83 104 197–199 Most women reported the perception of HM being harmless^{126 129} and the lack of inquiry from their physicians regarding the use of additional healthcare modalities. as the reasons for not disclosing HM use.27 83 117 126 129 However, up to 92% of women were willing to discuss HM use if their doctor inquired them about the use, and 70% of women desired HM-related information from their doctors.^{113 197} These findings highlight the importance of healthcare providers initiating the discussions about HM use in an unbiased manner and providing evidence-based feedback to ensure appropriate HM use.^{18 197} Moreover, these conversations should begin early in pregnancy, as HM use during the first trimester can have critical implications for fetal health and development.^{10 184 200} Lastly, further research is needed to explore the relationship between the practice of open dialogue between pregnant women and physicians on HM use and the corresponding pregnancy outcomes to determine whether improved patient–physician communication could potentially predict the reduction in unfavourable pregnancy outcomes.

The following limitations can be found in this review. First, the inclusion criteria of the reviewed studies were limited to English-language, which may have restricted the overall representation of HM use during pregnancy. However, efforts were made to minimise the omission of potentially relevant literature by reviewing the references of included studies. Second, the studies included in this review were primarily conducted in a single health facility using non-random sampling methods, impacting the generalisability of the findings. Furthermore, the variation and limited number of studies representing each geographical region hindered the examination of regional differences in subgroup meta-analysis of HM use and disclosure rate. Additionally, differences in the sample sizes and to whom the women disclosed their HM use existed across the studies, with some studies examining HM disclosure to physicians only and others also including midwives and health workers. Thus, these variations must be considered when interpreting the results of this review. However, despite these limitations, this systematic review and meta-analysis estimated the global prevalence of HM use during pregnancy and provided valuable insights into the patterns of HM disclosure among pregnant populations.

Conclusion

Lack of communication on HM use between pregnant women and obstetric care providers was correlated with unfavourable MCH outcomes, highlighting the need for prevention efforts across various health sectors to promote open communication about self-care practices during ANC and to encourage physicians to provide guidance on the safe and appropriate use of HM. These prevention initiatives could include advocating training for healthcare professionals working with pregnant mothers on the therapeutic effects of commonly used HM, implementing community outreach programmes to raise awareness of HM’s potential toxicity and integrating questions about HM use into routine antenatal case-taking procedures. Finally, additional research is needed to understand the factors influencing open patient–physician communication on HM use during ANC and to develop strategies for overcoming communication barriers to promote culturally responsive and integrative maternity care.

Data availability statement

Data are available on reasonable request. All datasets generated and analysed, including the study protocol, search strategy, list of the included and excluded studies, data extracted, analysis plans, quality assessment, and assessment of the publication bias will be accessible by contacting the corresponding author of this study.

Ethics statements

Patient consent for publication

Not applicable.