Primary healthcare providers’ knowledge, practices and beliefs relating to preventive sexual and reproductive health care for women from refugee and asylum-seeking backgrounds in Australia: a national cross-sectional survey

Design

The research study design was a cross-sectional anonymously completed national online survey.

Setting

The survey was conducted across all states and territories of Australia through an accessible online survey tool (Qualtrics, ver. 22, https://www.qualtrics.com). The survey was open between November 2022 and March 2023.

Participants and recruitment procedure

In this exploratory study, the authors aimed to investigate a phenomenon and gather preliminary data to better understand a topic about which little is known. Approximately 96,000 primary healthcare nurses and 26,600 GPs provide primary care in the community (Australian Government 2024). However, we acknowledge that the exact number of providers offering care to people with refugee and asylum-seeking backgrounds is not known. Despite this, the authors sought a nationally representative sample of primary HCPs and obtained as large a sample size as possible to maximise the chances of capturing meaningful insights and understanding the phenomenon under investigation.

Inclusion criteria were to be either a GP or a registered nurse (RN) (including NPs) practising in a primary care setting in Australia. General practitioners or RNs providing care to the general population and/or to women of reproductive age who are refugees or seeking asylum in Australia as part of their primary healthcare practice/service delivery were included. Representatives of 19 national- and state-based health professional organisations and 25 Primary Health Networks (PHNs) were contacted and asked to disseminate survey information. Australia’s PHNs are regional organisations designed to improve coordination, accessibility and effectiveness of primary health services in specific geographic areas (Australian Government 2023b). Communications representatives at each PHN nationally were contacted to disseminate information about the survey to their members through e-newsletters and social media. A diversity of viewpoints was sought by targeted sampling methods including contacting general practices, refugee health networks, SRH service organisations, family planning and women’s health organisations.

Detailed information about the study purpose, eligibility criteria and what participation involved was provided via an electronic link included in an introductory email sent to contacts in professional organisations and PHNs. One to two follow-up emails and/or phone calls were undertaken by the first author (ND) to encourage organisations to distribute the survey. Follow-up emails were sent to health professional organisation contacts to confirm if and when surveys were distributed.

Potential participants received the introductory email with the survey link from participating organisations. As participants clicked on the link, they were presented with the landing page of the survey. Potential participants were then presented with an explanatory statement outlining what the research involved, why they were invited to participate, procedure for consenting to participate in the project and withdrawing from the research, possible benefits and risks to participants and procedure for maintaining confidentiality.

Eligibility was based on answering ‘Yes’ to the question ‘Do you provide clinical care to women from refugee-like backgrounds in the practice you work in?’. After reading the participant information, those who were willing and eligible to participate could proceed by clicking on a web link or scanning a QR code to complete the survey.

The survey was hosted by Qualtrics (Provo, UT, USA). Respondents were informed that by completing the survey, they provided implied consent to participate. In respectful recognition of their time, participants could enter a draw for one AUD$50 gift card.

Data source

A study-specific survey was developed based on a systematic review of access to preventive SRH care for women who have been displaced, who are refugees or women seeking asylum (Davidson et al. 2022) and other studies of SRH needs of women in these groups. Other studies included those that did not meet the inclusion criteria for the systematic review (Mengesha et al. 2017) or were published after the review was completed (Davidson et al. 2024). The researchers’ refugee health experience also informed the development of the survey. Studies by Davidson et al. (2022), (2024) identified several factors impacting preventive SRH care access for refugee women and women seeking asylum. The barriers to healthcare access identified in these studies enabled the researchers to tailor survey questions to address the specific challenges faced by healthcare providers. Supplementary File S1 outlines how evidence informed the development of survey items.

The first author has a clinical nursing background and experience providing primary health care to refugee women and children in humanitarian aid settings. This first-hand experience offers valuable insights into the specific health needs, challenges and cultural considerations of refugee women. This ensured the survey content was relevant, addressing the unique aspects of providing care for this group of women. She was attuned to cultural nuances, preferences and potential sensitivities of caring for refugee women. This was essential in formulating survey questions and response options appropriate for HCPs. Her clinical background ensured the survey was feasible and practical for implementation in real-world, time-limited primary healthcare settings.

The survey explored primary HCPs’ knowledge, practices and beliefs of preventive SRH care needs of refugee women and women seeking asylum. It also assessed HCPs’ views about barriers to providing preventive SRH care to these groups of women. Data on five topics: contraceptive care, cervical screening, breast awareness, breast screening and HPV vaccination were collected. The outcomes of interest were: (1) who initiates SRH conversations (‘In most cases I do’, ‘In most cases women do’, ‘It’s rarely discussed’, or ‘It’s never discussed’); and (2) whether HCP discuss SRH opportunistically when women present for other health concerns (‘Yes’, ‘Sometimes’, ‘No’ or ‘Don’t know or unsure’). Potential explanatory factors included: (1) demographic characteristics; (2) HCPs’ self-reported knowledge (e.g. ‘Please rate your level of knowledge in relation to: (a) contraceptive care, (b) cervical screening, (c) breast awareness, (d) breast screening, and (e) HPV vaccination’); (3) HCPs’ awareness of women’s understanding of their preventive SRH care needs; (4) HCPs’ perceived need for training; and (5) characteristics of direct professional experience (e.g. frequency of consultation with refugee women and women seeking asylum).

The survey questions, response options and their corresponding dichotomised categories are listed in Supplementary File S2. Demographic characteristics sought are shown in Table 1.

The survey was pilot tested by 10 experienced HCPs to determine if questions were valid, clear and understandable. Seven RNs and three GPs with extensive experience of providing SRH care to refugee women and women seeking asylum participated in the piloting of the survey. Following feedback, some response options were changed. The final survey is available in Supplementary File S3. Pilot data were not included in the results.

Data management and analysis

Data were imported from Qualtrics to SPSS (ver. 25.0, IBM Corp, Armonk, NY, USA) for cleaning, checking and analysis. Not all respondents provided a response to all questions. Missing data were identified by running frequency tables for each variable to check for missing values. For missing data analysis, Little’s test was used to test the assumption that data were missing completely at random (MCAR) (Little 1988). This yielded a P-value of >0.05, indicating that there were no existing patterns in the missing data. No imputation of missing values was undertaken. Cases with >60% of missing values were excluded from analysis. The number of responses included in the analysis varied between questions. Frequencies and distributions of variables were analysed by descriptive statistics. Professional and demographic categorical variables were dichotomised. The two primary outcomes, ‘HCPs initiating conversations on SRH’ or ‘HCPs discussing SRH opportunistically’ were also dichotomised. Categorical variables were divided into two categories, dichotomised, based on a clinically meaningful and practical cut-off point. Each variable had four categories. Those categories with <10 data points in each were grouped together. Supplementary File S2 provides a description of the individual categorisation of each variable. This ensured there were enough data points in each of the resulting dichotomised groups to make certain that statistical analyses could be conducted effectively allowing for comparisons between two groups. Chi-square tests are generally used in data analysis to determine if there is a significant difference between two categorical variables (Sullivan 2012). As data management of primary outcome variables in this study involved grouping observations into categories, Chi-square tests were used to assess associations between the two outcomes of interest and professional and demographic variables.

Provider responses to questions that assessed HCPs knowledge and awareness of women’s perception of their preventive SRH care needs, perceived need for training, self-reported ability to provide culturally safe or competent care, and characteristics of direct professional experience with women are shown in Table 2. Mean scores for responses to questions related to HCPs views of refugee women’s preventive SRH were calculated across the five topics: contraceptive care, cervical screening, breast awareness, breast screening and HPV vaccination. The possible range for the resultant mean score was 0–3, with 0 indicating a low mean score or more negative responses across topics and 3 indicating a high mean score or more positive responses across topics (Table 2). Independent sample t-tests compare the means of two independent groups in order to determine if there is a significant difference between population means (Sullivan 2012). Using mean scores, independent t-tests were conducted to assess the associations between the two primary outcomes and the continuous variables related to HCPs knowledge, practices and beliefs.

Following bivariate analyses, covariates significantly associated with the primary outcomes or identified as relevant from previous research, were entered into two hierarchical logistic regression models (one for each of the two primary outcomes – HCPs initiating conversations on SRH and HCPs discussing SRH opportunistically). Hierarchical logistic regression analysis considers the variability of each potential explanatory factor, and thus allows for the cluster effects at different levels to be analysed within each model (Stryhn and Christensen 2014). Hierarchical logistic regression was used in this study to simultaneously account for potential influence of group effects and the effects of group-level predictors. The regression models included all relevant variables to control for confounding. Variables included professional group, language spoken, country of qualification, geographic location, HCPs’ knowledge and beliefs, HCPs’ perceived need for training, frequency of consultation, years practising in Australia, hours worked per week and being offered a female provider (Tables 3 and 4). The models were used to examine associations between the primary outcomes and covariates to establish separate and combined contributions to the variance. Covariates were categorised into five groups: (1) demographic characteristics; (2) HCP knowledge and awareness; (3) HCPs’ perceived need for training; (4) HCPs provision of culturally safe care; and (5) direct professional experience with refugee women or women seeking asylum. Change in pseudo R Square of each model as each group of covariates was added, was used to explain the contribution of the covariates to the two primary outcomes. The level of statistical significance was set at P < 0.05.

Ethics approval

This study was approved by the Monash University Human Research Ethics Committee (Project ID 2022-32458-83553). In the online survey, participants were informed that by completing the survey, they were providing implied consent to participate.

Professional and demographic characteristics

Respondents’ professional and demographic characteristics are shown in Table 1. Most respondents were females working in community health 10.9% (n = 28), general practice 10.2% (n = 26), SRH 8.6%, (n = 22), refugee health 7.8% (n = 20), regional/remote health service 4.7% (n = 12) and family planning 3.1% (n = 8). Approximately 57.6% (n = 91) were employed as RNs, of which 8.2% (n = 21) were NPs. The majority were born 72.2% (n = 117) and trained 80.7% (n = 130) in Australia. Over half of the respondents were aged <45 years 56.4% (n = 92) and had been practising in Australia for >10 years 58.8% (n = 94). Most reported providing services to refugee women or women seeking asylum more often than every 2 months, 60.2% (n = 112), with 82.1% (n = 110) offering fee-free services to these groups of women. The sample was drawn from all Australian States and Territories, with 60.6% (n = 97) from metropolitan locations and 39.4% (n = 63) from regional, rural and remote locations.

Description of healthcare providers’ practices, knowledge and beliefs

The majority of respondents stated that they offered women a female interpreter 71.2% (n = 116) and choice about seeing a female provider 80.0% (n = 132) (Table 1).

Between 64.8% (n = 78) and 72.7% (n = 98) of respondents reported initiating conversations about preventive SRH (Table 5). In most cases, women rarely or never discussed contraceptive care 72.7% (n = 98), cervical screening 64.8% (n = 78), breast awareness 71.5% (n = 95), breast screening 67.6% (n = 85) or HPV vaccination 69.9% (n = 91). In contrast, in most cases, HCPs initiated conversations on contraceptive care 27.3% (n = 70) of the time, cervical screening 35.2% (n = 90) of the time, on breast awareness 28.5% (n = 73) of the time, breast screening 32.4 (n = 83) of the time and HPV vaccination 30.1% (n = 77) of the time.

Between 26.9% (n = 68) and 40.3% (n = 102) of respondents reported bringing up preventive SRH opportunistically when women presented for other health concerns. Providers brought up preventive SRH opportunistically 40.3% (n = 102) of the time for contraceptive care, 34.6% (n = 88) of the time for cervical screening, 34.3% of the time (n = 86) for breast awareness, 26.9% of the time (n = 68) for breast screening and 34.1% of the time (n = 86) for HPV vaccination (Table 5).

Responses relating to HCPs’ knowledge and beliefs regarding SRH are shown in Table 2. Mean score for self-rated knowledge of preventive SRH suggests most (2.46, s.d. = 0.546) had a ‘moderate’ or ‘very good’ level of knowledge and were ‘moderately’ or ‘very comfortable’ providing culturally safe care (2.28, s.d. = 0.688). Mean scores suggest respondents believed women were ‘somewhat aware’ or ‘moderately aware’ of preventive SRH (1.27, s.d. = 0.846), but only ‘sometimes’ ask for information (1.02, s.d. = 0.703). Mean scores also indicated respondents believed women saw cost as a barrier to care ‘sometimes’ or ‘most of the time’ (1.60, s.d. = 0.906); women’s religious beliefs (1.56, s.d. = 0.860) and being unmarried ‘sometimes’ or ‘often reduced uptake’ of SRH (1.63, s.d. = 0.868) and that women face coercion from their husband or partner ‘sometimes’ or ‘most of the time’ (1.52, s.d. = 0.795).

Approximately two-thirds of respondents wanted further training on preventive SRH for refugee women or women seeking asylum. Respondents reported the need for further training on contraceptive care (71.5%), cervical screening (63.6%), breast awareness (67.3%), breast screening (64.8%) and HPV vaccination (66.1%). The majority of respondents (85.4%) also wanted further cultural awareness training.

Healthcare providers’ perceived barriers to providing preventive SRH to women

Fig. 1 displays the HCPs’ perceived barriers to providing preventive SRH care to refugee women and women seeking asylum. The most commonly cited barriers were women’s cultural beliefs, lack of SRH knowledge and religious beliefs. Over half (54.6%) perceived limited English-language skill as a barrier to care. Fewer believed the presence of a husband or intimate partner (36.2%), unavailability of interpreter services (31.3%), having a consultation with a male provider (27.0%) or interpreters not understanding medical terms (24.5%) were barriers to providing care.

Fig. 1.

Perceptions of barriers to providing preventive SRH care.

Factors associated with initiating SRH conversations and discussing SRH opportunistically

Factors associated with initiating SRH conversations were: seeing refugee women or women seeking asylum more often than every 2 months, not charging patient co-payment for consultations, being a female provider, providers not speaking a language other than English, working in a practice where patients were offered the option of seeing a female provider, working in a metropolitan location, working in a general practice and working part-time (Table 6).

Factors associated with discussion about SRH opportunistically when refugee women and women seeking asylum presented for other health concerns were: not charging a patient a co-payment for consultations, being a female provider, being an RN, having practised for <10 years and working part-time (Table 7).

Hierarchical logistic regression analysis for factors associated with initiating conversations about SRH

The first hierarchical logistical regression analysis is presented in Table 3. Providers not speaking a language other than English was significant in model one, Beta coefficient (Odds Ratio for each predictor variable) 5.01, (95% CI [2.16, 11.6], P < 0.001) and model two 5.27, (95% CI [2.00, 13.86], P < 0.001). Model one includes demographic variables: professional group, language spoken, country of qualification and geographic location. Working in a metropolitan location was also significant in models one 4.21, (95% CI [1.83, 9.69], P < 0.001) and two 3.23, (95% CI [1.32, 7.84], P < 0.01) and three 2.86, (95% CI [1.51, 7.12], P < 0.05). Model two includes HCP knowledge and beliefs; self-rated SRH knowledge, self-rated culturally competency and understanding of women’s SRH awareness. Providers who were ‘moderately comfortable’ or ‘very comfortable’ providing culturally safe care to women were significant in models two 2.58, (95% CI [1.19, 5.56], P < 0.05) and three 2.59, (95% CI [1.14, 5.88], P < 0.05). Providers who viewed women as being ‘somewhat aware’ or ‘moderately aware’ of preventive SRH were significant in model two 0.55, (95% CI [0.32, 0.94], P < 0.05); however, these effects disappeared when factors related to direct professional experience including greater frequency of providing services to refugee women 7.64, (95% CI [2.41, 24.22] P < 0.001), more years practising as a HCP in Australia 2.20, (95% CI [0.71, 6.87], P < 0.001) and <32 h worked per week 3.63, (95% CI [0.70, 18.61], P < 0.001) were entered into model four. Model four includes variables associated with HCPs’ direct professional experience with refugee women or women seeking asylum; frequency of consultation with refugee women or women seeking asylum, years practising in Australia, hours worked per week and offering a female provider. The variables in each model are outlined in Tables 3 and 4. Together, the following three factors explained 59.5% of the variance in initiating conversations about SRH; providing services to refugee women or women seeking asylum more often than every 2 months, working part time, and having practised in Australia for >10 years. Variance provides an indication of the amount of variation in the outcome variable explained by complete hierarchical logistic regression analysis. These results demonstrate that three significant factors together explained 59.5% of the differences observed between those who did and those who did not initiate conversations about SRH. These factors, therefore, play a substantial role in shaping the behaviour of initiating SRH conversations within the context of refugee and asylum seeking healthcare provision.

Hierarchical logistic regression analysis for factors associated with discussing SRH opportunistically when women present for other health concerns

The second hierarchical logistical regression analysis is presented in Table 4. In all four models, significant differences between RNs and GPs were found when discussing SRH opportunistically. Being an RN was significant in models one 3.19, (95% CI [1.52, 6.70], P < 0.01), two 2.91, (95% CI [1.37, 6.22], P < 0.01) and three 2.93, (95% CI [1.37, 6.27], P < 0.01). This effect was maintained but was reduced 2.61, (95% CI [1.12, 6.08], P < 0.01) when variables related to direct professional experience were entered into model four. Factors related to direct professional experience, which were significant in model four, included greater frequency of providing services to refugee women or women seeking asylum 2.82, (95% CI [1.07, 7.40] P < 0.05), more years practising as a HCP in Australia 0.40, (95% CI [1.66, 0.95], P < 0.05) and <32 h worked per week 2.43, (95% CI [1.02, 5.76], P < 0.05). The variables in each model are outlined in Table 4. Together, the following four factors explained 28.5% of the variance in discussing SRH opportunistically; being a RN, providing services to refugee women or women seeking asylum more often than every 2 months and working part time and having practised in Australia for <10 years. These factors, therefore, play a substantial role in shaping the behaviour of opportunistic SRH discussions.

Strengthens and limitations

The study’s strength lies in the broad survey distribution across diverse organisations, professional groups, facilitating potential participation across primary HCPs and it was short and convenient. However, some limitations are acknowledged. As this was an opt-in, anonymous online survey, a response rate could not be calculated. Also, there is no way of knowing how many HCPs see refugee women or women seeking asylum in their practice or what their characteristics are. It is therefore not possible to ascertain the representativeness of those who completed the survey. Furthermore, level of SRH knowledge and provision of culturally safe care were self-reported and may therefore be affected by social desirability bias. Nevertheless, previous studies have established strong concordance between HCPs’ self-reported knowledge and objectively assessed knowledge levels, hence this concern may be mitigated (Snibsoer et al. 2018).

The inclusion only of GPs and RNs in this study might limit the generalisability of the findings to other primary HCPs, such as midwives and other allied health professionals. Because the study focuses solely on GPs and RNs, it may not capture the unique perspectives, practices and challenges faced by other HCPs in the context of primary health care. The number of missing responses was another limitation of the survey. This has the potential for non-response bias and reduced generalisability of findings. Potential for bias may be mitigated given that cases with >60% of missing values were excluded from the analysis. Additionally, dichotomising variables can lead to loss of information and may not fully capture the nuances of the original variable categories (Altman and Royston 2006).

Implications for training, policy and research

The findings in this study have implications for training, policy and research. Training and education for all HCPs about how to provide culturally safe care might improve their capacity to engage positively with people who are refugees or seeking asylum (Migrant and Refugee Women’s Health Partnership 2017). The findings suggest that HCPs who expressed a desire for additional training in preventive SRH for refugee women and women seeking asylum are already comfortable with their current level of knowledge and confidence. Therefore, the training needs that could be better met by these findings include: (1) tailoring training content to focus on advanced or specialised topics within preventive SRH for refugee women and women seeking asylum, building upon the existing foundation of knowledge and confidence among HCPs; (2) designing training programs that emphasise practical skills development and application, allowing HCPs to enhance their competency in delivering culturally sensitive and effective preventive SRH services to refugee women and women seeking asylum; and (3) incorporating opportunities for ongoing professional development and support, such as mentorship programs or access to updated resources and guidelines, to sustain and reinforce the knowledge and skills gained through training. This has clear implications for both pre-service curricula and professional development training. This aligns with recent proposals to integrate refugee health into primary care by expanding specialist services into general primary care practices (Patel et al. 2022), particularly in practices where HCPs provide services to refugee women and women seeking asylum less frequently.

At the policy level, efforts might focus on recognising and addressing cultural barriers hindering discussions on preventive SRH identified in this study. For example, advocating for the full scope of practice for specialised SRH and refugee health nurses might be one approach to address the short consultation times respondents identified as a barrier to initiating preventive SRH conversations. In Australia’s Medicare system, HCPs are required to obtain a provider number to bill Medicare for their services. Currently, in Australia, only NPs and GPs can obtain a Medicare provider number and bill Medicare directly for services they provide.

If specialised SRH and refugee health nurses were granted an expanded scope of practice, including the ability to bill Medicare for their services, this could enable the scheduling of longer appointments when required, hence facilitating increased access to more comprehensive SRH care, interpreter services and information provision for refugee women and women seeking asylum. It might also assist in recognising the role that primary care nurses play in delivering SRH care to these groups of women. Incentivising nurses’ involvement in preventive SRH care through the addition of Medicare provider numbers for RNs and introducing a Medicare item that reflects the time involved for GPs to provide SRH care may make HCPs more likely to initiate SRH conversations and discuss SRH opportunistically. Historically, for example, RNs had access to a Medicare item number to undertake cervical screening, but since 2006, this specific item number is no longer available. In the absence of this, policymakers may consider revising the current Medicare Practice Stream funding (Australian Government 2023c). This funding, which provides financial incentives to cover the costs of engaging RNs and other practitioners, could be increased to support RNs in conducting preventive SRH screening. This adjustment would enable GPs to focus on more complex cases. Additionally, the finding that more frequent professional engagement with refugee women and women seeking asylum leads to improved initiation of SRH conversations suggests a need for continuity of care between HCPs and refugees with dependable services that go beyond one-time interventions.

Future research might examine reasons why two-thirds of HCPs were not routinely initiating preventive SRH conversations or discussing SRH opportunistically, as only about one-third of opportunistic SRH discussions were explained by variables in our analysis. This might involve exploring how the different organisational contexts such as community health, SRH and refugee health influence practice in relation to initiating and discussing preventive SRH care. Future research might explore HCPs’ perspectives and attitudes and organisational factors that might promote or hinder discussions to improve understanding of why preventive SRH is not always routinely addressed in primary care. This would help inform interventions and strategies to promote preventive SRH for refugee women and women seeking asylum in primary health care.

New knowledge to emerge from this study suggests that HCPs with more direct professional experience and frequent interactions with refugee women demonstrate higher knowledge and better attitudes toward preventive SRH; significant barriers to care include cultural and religious beliefs, limited English proficiency, and time constraints; fewer than one-third of HCPs routinely initiate SRH discussions; and RNs may be more proactive than GPs in discussing SRH opportunistically. These findings underscore the need for targeted training and support to improve HCPs’ comfort and competence in delivering culturally sensitive SRH care, highlighting the importance of addressing systemic barriers to enhance preventive SRH services for refugee populations.