Women’s health literacy and the complex decision‐making process to use complementary medicine products in pregnancy and lactation

Abstract Background Little is known about women's decision‐making processes regarding using complementary medicine products (CMPs) during pregnancy or lactation. Objectives To explore the decision‐making processes of women choosing to use CMPs in pregnancy and lactation; and to investigate how women's health literacy influences their decisions. Design, setting and participants In‐depth interviews and focus group discussions were held with twenty‐five pregnant and/or breastfeeding women. Data were analysed using thematic analysis. Results Key to women's decision making was the desire to establish a CMPs safety and to receive information from a trustworthy source, preferably their most trusted health‐care practitioner. Women wanted positive therapeutic relationships with health‐care practitioners and to be highly involved in the decisions they made for the health of themselves and their children. Two overarching components of the decision‐making process were identified: (a) women's information needs and (b) a preference for CMP use. Women collated and assessed information from other health‐care practitioners, other mothers and published research during their decision‐making processes. They showed a strong preference for CMP use to support their pregnancy and breastfeeding health, and that of their unborn and breastfeeding babies. Discussion and Conclusions Complex decision‐making processes to use CMPs in pregnancy and lactation were identified. The participants showed high levels of communicative and critical health literacy skills in their decision‐making processes. These skills supported women's complex decision‐making processes.


| INTRODUC TI ON
Complementary and alternative medicine (CAM) includes multiple CAM practices (therapies) as well as complementary medicine products (CMPs) like vitamin or mineral supplements or herbal medicines. [1][2][3] The World Health Organization refers to CAM as 'a broad set of health care practices that are not part of that country's own tradition or conventional medicine and are not fully integrated into the dominant health-care system'. 4 Similarly, the use of most CMPs is not considered to be part of conventional biomedical practice, 3 although some CMPs have been studied in clinical trials and subsequently have been co-opted or included in biomedical practice (eg some herbal medicines and some probiotic strains). 5,6 CMPs like herbal medicines and nutritional supplements are commonly used in pregnancy and lactation by women around the world. [7][8][9] The practice of herbal medicine is often based on traditional knowledge and use, as passed down by traditional medicine healers in different cultures. 10 Some nutritional supplements (eg iron, folic acid and iodine supplements) are part of evidence-based maternity care practice and are recommended in pregnancy and lactation by both medical practitioners 9,11,12 and complementary medicine practitioners. 9,13,14 High rates of herbal medicine use in pregnancy have been noted. One multinational study found that 28.9% of participants reported use of herbal medicines in pregnancy, with highest rates reported in Russia (69%), Eastern Europe (51.9%) and Australia (43.8%). 7 Herbal medicine use in lactation is also common internationally (eg see refs [15][16][17] and by Australian women. 18,19 Other studies have found high CMP use in Australia too: around 50% of Australian women have been shown to use herbal medicines in pregnancy, and around 90% to use vitamin or mineral supplements. 14,20 Previous research has established that women in high-income economies use complementary medicines during pregnancy and breastfeeding for several reasons. These include the desire for self-determination and choice in health-care decisions, 21,22 including the desire for natural childbirth, 23 to prepare for labour, 24,25 treat common conditions of pregnancy, 9,18,[24][25][26][27] and promote their own and their babies' health and well-being. 8,24,25 During breastfeeding specifically, herbal galactagogues are used to correct perceived or diagnosed breastmilk insufficiency, and other herbs are used to support post-partum health and recovery after birth or to treat common conditions like mastitis or upper respiratory tract infections. 8,9,18,19,28 Complementary or alternative medicine (CAM) use prior to pregnancy has been associated with use during pregnancy. 24,25,29,30 CMP use in pregnancy or breastfeeding is also associated with biomedical or CAM health-care practitioner prescription or recommendation. 9,12 A positive relationship with their CAM practitioners has been linked to pregnant or breastfeeding women's use of CAM. 21,28 Women appreciate CAM practitioners' holistic approaches to health, including consideration of mentalemotional, physical, social and spiritual health. 31,32 They prefer CAM practitioners who facilitate and encourage self-empowerment and autonomy in health care and demonstrate positive patient-provider communication. 21,31,33,34 Self-prescription of complementary medicine products (CMPs) is common 7,13,17,[35][36][37] with some women perceiving CMPs to be safer to use in pregnancy and lactation than pharmaceutical medications. 28,[38][39][40] Women also often take vitamin and mineral supplements due to the belief that supplementation will ensure they meet the additional nutritional requirements of pregnancy and lactation. 41 In affluent economies, CAM use in pregnancy is greater in women with higher incomes, university education and is associated with primiparity. 24,42,43 An important socio-demographic component of health literacy is education, and advanced literacy and education levels have been shown to be strong predictors of positive health status. [44][45][46] Whilst the demographic profile of most pregnant and breastfeeding CAM users in wealthy countries like Australia means they may not initially be considered to be at risk for limited functional health literacy, 47 their actual health literacy levels have not been previously explored. In instances where self-prescribing is common, functional health literacy may be particularly important.
Most pregnant or breastfeeding mothers want to promote their babies' and their own health. Studies from Australia, 18,28,37 and similar overseas economies 22,48,49 confirm that safety of CMPs is very important to mothers. However, little is known about the role of health literacy in pregnant or breastfeeding women's decision-making processes regarding the use of CMPs. A three-tiered hierarchy of health literacy skills proposed by Nutbeam 50 describes the skills consumers need to acquire, understand and use information when making health-care decisions. 50,51 Functional health literacy skills are the first level and involve the reading, writing and numeracy skills required to understand factual health information regarding risks and medication prescriptions. 51 The second level is communicative health literacy and requires more advanced cognitive and communication skills 50,52 to extract health information, apply it to different circumstances and communicate with health-care practitioners (HCPs). 52 Third, the most advanced level is critical health literacy whereby consumers' skills are used to critically analyse and reflect information to support health-care decisions. 50,52 This study aimed to explore the decision-making processes pregnant and breastfeeding women go through when choosing to use CMPs from the perspective of the women themselves. It also aimed to investigate how women's health literacy skills influenced their decisions to use CMPs. Operational definitions used in this research appear in Box 1.

| Participants and recruitment
Purposive followed by snowball sampling approaches were used for recruitment and were directed at pregnant and breastfeeding women who used CMPs. This enabled the study aims to be investigated whilst ensuring that the sample was rich enough to enable participation from women of diverse experiences and backgrounds. 57 The study was advertised on posters and flyers at playgroups, antenatal classes, pregnancy and postnatal yoga classes and support groups, in pharmacies and allied health practices, on free local classified advertising networks, and through [the Institution's] electronic media channels.
Women over the age of 18 who were currently pregnant and/or breastfeeding and who lived in the Northern Rivers region of New South Wales, or in the metropolitan regions of Sydney, Brisbane or the Gold Coast were invited to participate. Women also needed to be currently taking or have taken at least one CMP in the last 12 months and able to speak English well enough to participate in an in-depth interview [IDI] or focus group discussion [FGD].
Women in the Northern Rivers area participated in face-to-face interviews and focus groups, FGDs and women at a distance from the interviewer participated in telephone or Skype interviews. All participants were given a $20 grocery voucher in recognition of their time.
Thematic saturation 57 determined final sample size and was reached at 22 participants. An additional three interviews were held to confirm thematic saturation.

| Study design
Qualitative methods were chosen to elicit in-depth, detailed descriptions of the experiences, beliefs, values 58,59 and views of pregnant and breastfeeding women, and their motivators for using CMPs during pregnancy and breastfeeding. Qualitative methods allowed a compelling picture of the experience of CMP use to be collected and deepened understandings of these phenomena. 58,59 The use of both IDIs and FGDs allowed women to choose which format they would prefer and could participate in. This assisted with recruitment and enabled interviews to go ahead when FGDs were not achievable.

| Data collection
A seven-item semi-structured interview guide was used during FGDs and IDIs (Table 1). Feedback from pre-testing for face, content and construct validity from an interview with one pregnant woman, and a focus group with one pregnant and three breastfeeding women, was used to refine the questions. Pre-testing also helped ensure that women who used CMPs in pregnancy and lactation had a voice in the design of the research.
All participants received an information sheet and had the opportunity to discuss the study before giving consent to participate.
Participation was voluntary, and women could choose to withdraw from the study at any time. The decision to participate in an IDI or FGD was primarily the choice of the participant and depended on how comfortable the participant was in a group or individual setting, whether she wanted to bring her child/children to the IDI or FGD (babies and children were welcome), her work and family commitments, and distance from the interviewer-researcher. The first author conducted all IDIs and FGDs.
Demographic details and data on women's use of CMPs at the time of the interview and in the previous 12 months were also collected.
Women's functional health literacy levels were measured using two validated health literacy screening tools. The first was the standard single question health literacy measure How confident are you filling out medical forms by yourself? 60 with response options: "extremely", "quite a bit", "somewhat", "a little bit" and "not at all". Those that chose "somewhat", "a little bit" or "not at all" were considered to be at risk of inadequate health literacy. 60,61 The second was the Newest Vital Sign, a three-minute direct test of consumer abilities that identifies people at risk of limited functional health literacy by measuring reading ability and interpretation skills, as well as aspects of numeracy necessary to understand nutritional information on food labels. 62,63 Participants who answered four or more of the six questions correctly were considered to have Box 1 Operational definitions • CMPs were defined as herbal medicines in ethanolic extract, tablet, capsule or tea form, 5,53 micronutrient supplements containing vitamins or minerals, and food supplements (eg probiotics or protein powders), 54 topical preparations. CMPs could be purchased over the counter or after consultation with a HCP. 55 • Women's health literacy needs were defined as the information needed and desired to make decisions about using CMPs in pregnancy and lactation, and the factors involved in obtaining and understanding this information. 56 TA B L E 1 Guide for semi-structured interviews and focus group discussions Interviews

| Data analysis
The results from the demographic survey and health literacy assessment tools were analysed using descriptive statistics. All IDIs and FGDs were audio-recorded and transcribed by an independent transcription service, then checked for accuracy against the original recording by LAJB. Transcripts were thematically analysed using the six steps of thematic analysis as described by Braun, Clarke 65 using NVivo10 for data management. All transcripts were read multiple times to ensure a thorough understanding of the themes as they emerged, with themes grouped into major and minor subthemes. Constant comparison of findings was an essential part of the inductive thematic analysis, as potential codes and themes were identified, reviewed, defined, named, and refined, and relationships between themes identified. Participants shared information freely in IDIs and FGDs, including potentially sensitive data like complex health histories. The flexibility of the semi-structured interview guide also facilitated the use of follow-up questions within IDIs and FGDs, and in subsequent FGDs or IDIs to confirm the significance of the information. As no notable differences appeared between data from FGDs and IDIs, and between pregnant versus breastfeeding women, the data from all participants were grouped together for analysis. To increase validity, PA coded several transcripts and LAJB and PA met several times to review, discuss and agree on identified themes and subthemes for the final analysis. LB and KM also participated in reviewing and discussing the thematic analysis in final stages of the writing process. All participants were de-identified and assigned pseudonyms for data reporting.
Additionally, LAJB kept a detailed research journal where ideas and themes from each interview and focus group were documented in an on-going iterative process.

| Demographic information
Participants ranged in age from 23 to 40 years, and the average age was 32 years. Around half were first-time mothers. Fourteen had between one and four older children, ranging in age from 2 to 11 years old. All women with older children reported having breastfed these children for 6-34 months (mean 18 months). All but one woman completed the two health literacy screening tests (

| Complementary medicine use
Women listed the types of complementary medicines they currently took and had previously taken during their most recent or current pregnancy or breastfeeding journey (Appendix 1). A range of CMPs was reported. Pregnancy and breastfeeding multivitamin formulas were the most popular dietary supplements taken regularly across the sample. Probiotics, essential fatty acid supplements and iron supplements were also used widely. Consumption of herbal medicines during pregnancy was reported far less frequently than in lactation. The use of CMPs for breastfeeding issues and support was evident. Breastfeeding women reported using herbal teas and extracts to support breastmilk production and treat mastitis, and dietary supplements like lecithin to treat and prevent blocked milk ducts. A few participants reported using CMPs specifically chosen by their HCPs according to their specific health conditions.

| Information sought in the decisionmaking process
Women sought information from three main areas when deciding whether to take a CMP: HCPs, their own and other's experiences, and published research ( Figure 1). Primarily, they wanted information from their most trusted HCP -usually midwives, naturopaths and integrative GPs, (medical doctors who combine conventional biomedicine and evidence-based CAM in practice 66 ) -but for some CMPs, they sought second opinions from other HCPs (pharmacists, naturopaths in pharmacies, health food stores or herbal dispensaries, and HCPs staffing CAM or hospital medication helplines). Although a few women mentioned having obstetric care, they did not identify their obstetricians as primary sources of CMPs information.
Upon receiving a recommendation to use a specific CMP from their most trusted HCP, and being assured of its safety, some women immediately decided to take the CMP. However, if the recommendation was general (eg to take a pregnancy multivitamin),  Higher income (> AUD $1815 per week) 11 Prefer not to answer 1 Obviously, I make an informed decision and then go on that gut feeling whether it's going to work for me or not.

| Factors influencing the decisionmaking process
Thematic analysis determined two overarching components of the decision-making processes (  All participants expressed the desire to know whether a CMP would harm their unborn or breastfeeding baby and most (21/25) wanted to know that the CMP was also safe for them. Women often double-checked safety with their trusted HCP, even if the HCP had recommended the CMP, and sometimes used multiple sources (eg Internet, asking HCPs when purchasing over-the-counter products, medication helplines) to further assure themselves of safety.
F I G U R E 3 Functional, communicative and critical health literacy as demonstrated by participants within the overarching concept of maternal health literacy, defined as 'the cognitive and social skills that determine the motivation and ability of women to gain access to, understand and use information in ways that promote and maintain their health and that of their children. 56

| Supporting pregnancy and breastfeeding health for mother and baby
Women took specific CMPs to support their babies' health during pregnancy (15/25) and whilst breastfeeding (8/25). Women also reported taking CMPs to help optimize their own health and thus their baby's health. In pregnancy, this usually involved taking folic acid and/or pregnancy multivitamins, but some women used specific CMPs for their individual health conditions. During breastfeeding, some women took CMPs with the aim of providing prophylactic immune support for their breastfeeding children through their milk.
A few women were investigating or using CMPs as galactagogues, due to diagnosed or perceived milk supply issues. They also commonly spoke about using CMPs to treat breastfeeding-specific medical problems like cracked nipples, blocked ducts and mastitis, and appreciated that the use of CMPs helped them to continue to breastfeed successfully.
One duct would keep getting blocked, so I got a herbal tincture from the herbalist, and pokeroot cream to loosen it, break it up, to help move it, so I could keep breastfeeding.

| Past experience with CMP use and desire for holistic health care.
For half the sample, women's preference for CMP use was also re-

| Women's communicative health literacy
Communicative health literacy describes a person's motivations, confidence and abilities to act independently on health knowledge, 50 interpret health information meaningfully and apply it in different circumstances. 51 Participants in this study demonstrated high communicative health literacy in several ways (Figure 3), including collating CMP-related information from multiple sources. Previous research has identified many similar information sources to those used by participants in this study, and the use of plural resources by mothers when seeking information about CMPs. 14,18,25,[67][68][69][70] However, this study identified that women did not rate the information received from family, friends, peers and the Internet as highly as that received from trusted HCPs who had qualifications and experience in CAM modalities. Whenever possible, women preferred to determine a CMPs safety and indications through discussions with their trusted HCPs. Women used multiple sources of information to determine the quality of information obtained. Shared social bonds may be an important influence on self-prescription 9,71 and are evident in other studies on CMP use where pregnant or breastfeeding women share CMPs information with each other and receive CMPs information from their family, friends and HCPs. 13,17,18,28,29,36,37,49,72 In this study, the sharing of information both in person and in online forums with peers was an important consideration in the decisionmaking process, 29 especially when participants described receiving recommendations for CMPs from several non-HCP sources.
The second major demonstration of communicative health literacy in this study was participants' active engagement in discussions with their HCPs to obtain CMPs information and safety profiles. The positive therapeutic relationship identified between study participants and their trusted HCPs was a key factor in women's perceptions of the high quality of information received from these HCPs. This is especially important to note when considering women's primary desire to know that the CMPs they chose to take were safe.

| Women's critical health literacy
Critical health literacy builds on communicative health literacy and describes how well an individual can analyse and consider health information and use it to increase their autonomy in health-care choices and other life events. 50,52 The ways women evaluated the CMPs information they collated to determine whether it was trustworthy, valid and reliable was a key component of the way they demonstrated their critical health literacy skills (Figure 3).
Determining a CMPs safety in pregnancy or breastfeeding was imperative and frequently drove participants' complex informationgathering processes, especially if they received CMPs information from a source other than their trusted HCPs. In order to validate their trusted HCPs CMP recommendation, many participants gathered information from multiple sources and sought information from at least three sources before making their final decisions (Figures 1 and 2). This high level of critical health literacy reflects the women's keen engagement with their own health and the 'active consumer' noted in CAM users previously. 43 Using critical health literacy skills to evaluate CMPs information also required women to assess whether the use of the CMP was applicable to their own or their babies' health and required some complex assessments due to the limitations in empirical ev-

| Strengths and limitations
Using two validated assessments of health literacy levels demonstrated greater reliability of results regarding participants' health literacy levels. Nevertheless, an important limitation was that all but one participant in the sample demonstrated high functional health literacy skills, and the entire sample showed sophisticated communicative and critical health literacy skills, which may explain their extensive information seeking and complex decision-making processes. This study does not represent the full range of health literacy levels and further research on CMP use with lower health literacy samples is needed, especially considering that qualitative research cannot be generalized outside the study sample. However, this limitation can also be considered a strength of the study, as the demographics of the study sample reflect the typical Australian woman who uses CMPs in pregnancy or breastfeeding. 14,18,35,36 Investigating CMP use in a sample of preg-

| CON CLUS IONS
Women's decision-making processes were quite complex and in- informed approach to CMP use in pregnancy and lactation, which ultimately may enhance woman-centred maternity care.

ACK N OWLED G EM ENTS
The authors would like to acknowledge Dr Claire O'Reilly for her input into an initial draft of the research project. Thanks also go to Dr Jennifer Johnston who provided valuable feedback on an early draft of the manuscript. We would also like to acknowledge Blackmores Ltd. for the PhD scholarship provided to Larisa Barnes.

Philanthropical funding from Blackmores Ltd. funded Larisa Barnes'
PhD scholarship during the course of this research. However, Blackmores has had no input into the design, execution or the dissemination of her research. The authors declare that they have no conflicts of interests.

E TH I C A L A PPROVA L
This study received approval from the University of Sydney's Human Research Ethics Committee (approval number 2015/730).

DATA AVA I L A B I L I T Y
The data sets generated and analysed during the current study are not publicly available as participants did not consent to transcripts of interviews and focus group discussions being shared. Additional details relating to other aspects of the data are available on reasonable request from the authors.