Introduction
Approximately two-thirds of adults aged over 50 years have either osteoporosis or osteopenia, contributing to an increased risk of fracture [
1,
2]. Osteoporosis and the associated fractures cause significant morbidity and mortality, with estimated morbidity greater than many other non-communicable diseases, such as ischaemic stroke or chronic obstructive airway disease [
3].
Geographic discrepancy in osteoporosis care is evident, with variation in availability of DXA, use of treatment and access to specialist services both across and within countries [
4‐
8]. A systematic review suggested lower fracture risk in rural compared to urban areas of higher income countries; however, this pattern was not seen in lower income countries and may be country- or region-specific [
9]. Similar disparities exist in rural areas for other chronic conditions globally [
10].
In Australia, despite similar rates of self-reported osteoporosis in rural and urban areas, there are higher rates of hip fractures in rural areas [
5,
11]. The reasons for geographical discrepancy are unclear, and likely multifactorial, with patient-related factors, clinician-related factors and system factors, such as variation in health-related behaviours, osteoporosis knowledge and availability of bone densitometry or bone health specialists [
11,
12].
In 2019, the Australian Government launched a National Strategic Action Plan for Osteoporosis, prioritising actions to improve osteoporosis management [
13]. A National Consumer and Community Forum identified key areas for improvement including increasing community awareness, improving risk factor identification, diagnosis and fracture management.
Approximately one third of Australians live outside major cities [
11]. However, people in rural areas are less likely to be included in medical research [
14,
15]. Stakeholder engagement is a key component of designing and implementing models of care [
16]. Communities and clinicians living outside major cities have distinct experiences and preferences for healthcare, which are important to capture in order to improve care delivery and outcomes [
12,
17].
Our study aimed to explore the unique needs and perspectives of people living in a large regional area of Australia, in regard to osteoporosis. We sought to explore knowledge gaps, risk factors, preferences for care and barriers to care, in order to inform development of osteoporosis models of care.
Methods
This mixed methods study design included a quantitative survey (phase 1), followed by qualitative semi-structured interviews (phase 2), with triangulation of results. The survey provided information on osteoporosis prevalence, risk factors, management and knowledge, while interviews provided further insight into perceptions, preferences and experiences of osteoporosis care. Data from the survey and interviews will be used to inform the development of osteoporosis care programs.
Participants and setting
Adults aged ≥ 18 years living in Gippsland, a large region of Victoria, Australia were invited to participate. This region was chosen because it had the highest rates of hip fractures (a fracture indicative of osteoporosis) in Victoria in 2012–2013 [
5]. We included both people with and without osteoporosis, because osteoporosis is often undiagnosed, and the perspectives of those without osteoporosis are an important aspect to prevention. Exclusion criteria were aged < 18 years, or non-English speaking. Participants were recruited via advertisement at local community groups, social media, local radio stations and local newspapers, as well as word of mouth.
Phase 1: survey
The quantitative survey involved a 38-question anonymous survey, completed online or via mailed hard copy distributed at community meetings (Online resource
1). The survey was developed by the research team, pilot tested with members of the community and refined based on feedback. The survey contained questions regarding demographic details, residential location (based on the Modified Monash Model level of remoteness [
18]) osteoporosis diagnosis, management and osteoporosis risk factors, osteoporosis knowledge (using the Osteoporosis Knowledge Assessment Tool, (OKAT)) [
19] and preferences for managing osteoporosis. Due to the expansion of telemedicine across Australia during the COVID-19 pandemic, specific questions were included regarding experience of telemedicine. An explanatory statement was provided at the start of the survey, and consent was given electronically (online) or implied by return of the survey (hard copy). We initially aimed to recruit at least 100 participants; however, the recruitment was difficult, and we closed the survey after 18 months (available October 2020–June 2022).
Surveys were analysed quantitatively, with categorical data reported as number (%) and continuous data as mean (standard deviation). One question was omitted from the OKAT as it did not align with current Australian guidelines, and thus the potential total score was 19 instead of 20 [
20]. Linear regression was used to assess the relationship of age, educational status, residential location or a diagnosis of osteoporosis, and osteoporosis knowledge. Logistic regression was used to assess the relationship between age, educational status and residential location, and satisfaction with telemedicine. A significance level of
p < 0.05 was used. REDcap electronic data capture tools hosted at Monash University were used for data collection and management, and Stata V15 (StataCorp LLC, Tx, USA) was used to analyse data [
21,
22].
Although the survey was anonymous, participants could record their email address to be contacted to participate in an interview. To encourage participation, by leaving an email address, participants entered a draw to win one of five $75 AUD gift vouchers, which was selected via a random number generator. The email addresses were stored separately, so they could not be linked to the survey data.
Phase 2: semi-structured interviews
Semi-structured telephone interviews were conducted between November 2020 and July 2021. All participants who completed the survey and expressed interest in an interview were contacted via email to confirm interest and determine availability to participate. Of the 23 who expressed interest, 16 responded; one was not eligible as they lived mainly in a metropolitan area, the remaining 15 agreed to an interview and an explanatory statement was provided to these participants. Verbal consent was obtained by telephone prior to commencing the interview. Participants were reimbursed for their time with a $50 AUD gift voucher.
Interviews were conducted by an Endocrinologist with experience in osteoporosis and research (AJ) and lasted 20–25 min. The interview schedule (Online resource
2) was developed by the research team and informed by previous literature [
23‐
25]. Demographic details were collected prior to commencing the interview, and the interview schedule differed based on whether the participant had a diagnosis of osteoporosis or osteopenia or not. Interview topics included osteoporosis knowledge, osteoporosis diagnosis and current management, their ‘ideal’ care program (participants without osteoporosis were asked to imagine that they did have osteoporosis), educational needs and preferences in relation to bone health and use and satisfaction with telemedicine.
Interviews were audio-recorded and transcribed verbatim by an independent transcribing service. Deidentified transcripts were analysed by an Endocrinologist / researcher with experience managing osteoporosis (AJ) and an experienced qualitative researcher (BG). Thematic analysis was used with an inductive approach informed by our research question and review of the literature, to draw out shared meaning and patterns across the data [
26]. Both researchers independently coded three transcripts and met to review codes and reach agreement on interpretation of data and coding. AJ then coded the remaining transcripts, followed by review of these by BG. After assigning initial codes, these were grouped into broader themes, with in-depth discussion between researchers to achieve consensus. Data were informed by a range of participant viewpoints and held sufficient information power to develop new knowledge; data saturation rarely truly occurs in qualitative research of this type [
27,
28]. The Dedoose platform was used to store and organise data.
Discussion
This mixed methods study explores osteoporosis among people living in a large regional area of Australia. The survey demonstrated gaps in osteoporosis screening, management and knowledge and a desire for more information about bone health. The qualitative interview findings were also reflected in the survey, with concerns over osteoporosis care, the desire for tailored, individualised and local care and the preference for in-person appointments, with the option of telemedicine.
Our study highlights known gaps in osteoporosis care, with failure to identify a fragility fracture as osteoporosis, and low treatment rates in the surveys, reflected in the interviews as concerns over care [
29,
30]. These gaps in care are not unique to Australia or to rural areas; however, the solutions need to be customised. Various models of care to improve osteoporosis outcomes exist, including fracture liaison services, education, orthogeriatric services, universal or targeted screening and exercise. A review of models of care in osteoporosis found mixed efficacy for these programs, and minimal information on implementation characteristics were reported, making it difficult to determine whether existing programs will be suitable and efficacious to scale to new settings [
31]. Participants wanted local care if possible, and therefore, approaches such as fracture liaison services, which are mainly located in metropolitan hospitals in Australia currently, will not address this. Virtual fracture liaison services, which have been studied in some countries, may not be desired by the community, as most in our study preferred to see new clinicians in person [
32].
Rural and regional areas have poorer access to DXA than metropolitan cities [
8,
12]. Our interview participants voiced frustration at having to travel for a DXA but also perceived poor awareness of the need for DXA among the community and clinicians. Aligned with this, over one third of surveyed participants in our study with risk factors for osteoporosis had not had a DXA. Government reimbursed DXA is available in Australia for anyone ≥ 70 years, as well as younger people with specific osteoporosis risk factors. However, unlike other screening programs in Australia, such as for breast cancer or colon cancer, DXA requires a referral from a doctor. Initiatives to improve DXA rates in rural areas include mobile bone densitometry vans; however, a study of a service in rural Queensland still reported lower rates of investigation in more remote areas, suggesting access is not the only barrier [
7].
The majority of participants with osteoporosis were managed by their GP; however, GPs have reported knowledge gaps in osteoporosis, with uncertainty around when to start or stop medications, fear of side effects and a desire for more information [
12,
33]. A variety of educational programs targeting GPs have been shown to increase DXA and prescribing rates, including electronic reminders, dissemination of guidelines and academic detailing, although most studies also incorporate a patient-directed component [
31,
34]. Project ECHO (Extension for Community Healthcare Outcomes) telementoring is one way to offer targeted education and improve osteoporosis care in rural areas [
35]. GPs meet with specialists virtually, receive education and training and discuss difficult cases, to upskill them to deliver care for complex patients closer to home. Studies of Bone Health TeleECHO programs in the USA have shown increased confidence in managing osteoporosis among attendees [
36]. Improving health professional access to osteoporosis education and research is a key priority of the Australian National Strategic Action Plan for Osteoporosis, and the first Bone Health TeleECHO in Australia launched in 2021 [
13]. An updated RACGP/Healthy Bones Australia guideline on the ‘Management of osteoporosis in postmenopausal women and men over 50 years of age’ will also be available in the next year [
20].
Interview participants reported that they did not receive enough information about osteoporosis after diagnosis, and this is reflected in the survey osteoporosis knowledge scores, which did not differ between those with or without osteoporosis or fractures. Most people want more information, and there was consensus that information should come from health personnel, so GP practices or community health centres are likely to have a primary role. However, in keeping with previous studies, the desired volume, method, and types of information differed considerably between participants, and so, the approach to education should be individualised [
25,
37,
38]. Educational programs are a key component of disease management for a variety of medical conditions and may improve clinical outcomes and self-efficacy; however, outcomes vary considerably between different programs, and the key factors that improve success are unclear [
39]. Patient preferences for education may affect these differing outcomes [
39,
40]. Clinicians should therefore explore information needs with their patients, and offer tailored information, such as written pamphlets and websites (e.g. Healthy Bones Australia [
41]) that patients can refer to.
Most of the survey and interview participants preferred in-person appointments to telemedicine appointments, with the major concerns of telemedicine being communication and the doctor-patient relationship. It is important to note that most participants had used telephone consults, which are now more limited in Australia. Video appointments may address some of the concerns of patients in regards to communication barriers; however, the technology may be more difficult, and internet speeds may be prohibitive for some people [
42]. A study of patients attending a metropolitan Australian tertiary osteoporosis clinic via telemedicine found roughly one-quarter preferred in-person appointments, one-quarter preferred telemedicine and the remainder had no preference [
43]. Interestingly, confidence with technology was not associated with telemedicine preference. The Endocrine Society have recently published a perspective on the use of telehealth for endocrine conditions [
44]. They consider flexibility of appointment type (in-person, telephone or video) important, and the patient preference in regard to this should be strongly considered, with the opportunity to choose their appointment type unless there is strong reason against this. Whether future generations, who have greater familiarity with technology throughout life, will prefer telemedicine appointments is yet to be established.
The three themes identified in the interviews, concerns regarding adequacy of care, desire for tailored local care and desire for hybrid in-person or telemedicine appointments, are interrelated. They all focus on the need for personalised care. Gaps in care, while widespread, are also individual; local services required are personal; and satisfaction with telemedicine depends on the person and clinician involved. Solutions therefore must be thorough and adaptable to the needs of the person.
A strength of our study is the mixed method design and congruent findings from the survey and interviews. However, our study has several limitations. Our participants came from a region within Victoria, Australia, and so findings may not be applicable to other areas. Most of our participants were Australian-born, and only English-speaking participants were included. However, Australian census data suggests that the vast majority of residents of this region are Australian born (80%), and few (6.8%) speak a language other than English at home [
45]. A greater proportion of survey respondents had tertiary education compared to the broader region, and while educational attainment was not associated with OKAT scores, it may influence other responses and behaviours. Some survey respondents’ provided postcode was metropolitan, despite stating they lived in a rural area. Survey recruitment was difficult, resulting in a smaller sample size than planned. Recruitment issues may have been due to the COVID-19 pandemic, which prohibited travel of the researchers to the region during recruitment, and meant advertising and interviews had to be performed remotely. This may also reflect a lack of interest in the local community, in general, or due to the pressing importance of the COVID-19 pandemic. Survey respondents were likely therefore more invested or interested in bone health, reflected by relatively high screening rates. Thematic analysis is by its nature subjective, and there may be bias related to the researchers’ knowledge and experience. This analysis was performed by two researchers with different backgrounds, to try and bring diversity to the analysis.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.