ABSTRACT
Objectives
To assess what motivates medical imaging utilisation and to understand what referring clinicians, and those who operationalise these referrals, consider important when making decisions to refer to medical imaging in the South Australia Public Health System.
Design, Setting
Two cross‐sectional surveys were purpose‐built for online delivery. One was conducted from March–August 2023 for SA Health referring clinicians, and another was conducted from February–July 2023 for South Australia Medical Imaging (SAMI) clinical staff.
Participants
Two cohorts were surveyed: referring clinicians within SA Health (n = 138) and SAMI clinical staff (n = 251), from both metropolitan and regional locations.
Main Outcome Measures
Two customised survey instruments.
Results
A range of factors was reported to influence how clinicians' decisions to request medical imaging, including patient‐related, staff‐related and system‐related factors. Both clinicians and SAMI staff agreed that ‘improving overall patient health’ and ‘reassurance or confirmation of disease absence’ were key motivators for requesting medical imaging. However, notable differences emerged in how other motivations were rated, as well as factors clinicians consider important when referring patients for medical imaging. While a small number of clinicians acknowledged over‐requesting, a much larger proportion of SAMI staff perceived over‐requesting as a common issue among referring clinicians.
Conclusions
This study is the first to explore the perspectives of South Australian clinicians on the factors that motivate medical imaging requests. It highlights a complex array of drivers influencing referral practices and differing perceptions of overuse between those who refer and those who operationalise those referrals. The findings can inform the development of targeted, evidence‐based strategies to optimise medical imaging referrals, improve resource use and enhance healthcare delivery for the South Australian community.
1. Introduction
Medical imaging has been used for over a century to visualise the human body [1]. Advances in technology have significantly enhanced the detail and variety of medical imaging modalities, improving diagnosis and treatment outcomes [2, 3]. However, the increased use of medical imaging services places strain on the healthcare system and costs.
Australian Medicare began funding diagnostic medical imaging in 1984, and expenditure has grown from $255 million (1984–85) to $2.15 billion (2009–10) [4] accounting for around 15% of Medicare costs [5]. Between 1984–85 and 1993–94 most types of Medicare services increased, especially radiology and pathology services which increased by the highest amounts. There was a 78% increase in radiology services, a 71% increase in pathology services, a 42% increase in specialist services and a 33% increase in general practitioner services per person [6]. There was a 4.3% increase in the population who had a diagnostic imaging service between 2013–14 and 2018–19 [7]. Between 2018–19 and 2022–23 there was a 1.2% increase in the population who had a diagnostic imaging service, with Medicare funding over 27 million diagnostic examinations in 2022–23. Almost 2 in 5 Australians (39%, over 10 million people) had a diagnostic medical imaging service in 2022–23 [8].
Factors influencing demand for medical imaging include increased disease burden of the aging Australian population, healthcare workforce shortages, and pressures on hospital capacity creating an environment where medical imaging is used as a triage tool, to support hospital discharge, or to reduce hospital admissions [9, 10]. Similar factors appear to have led clinicians to shift from approaches such as history taking and clinical examination towards a greater reliance on diagnostic testing [11]. Furthermore, unnecessary medical imaging can result from clinicians' lack of knowledge about available medical imaging modalities, a culture of thoroughness, clinical curiosity, discomfort with uncertainty, external pressures from patients, and legal concerns [12].
The Choosing Wisely campaign, launched in the United States of America in 2012 and in Australia in 2015, aimed to reduce unnecessary medical tests, by promoting dialogue between clinicians and patients [13]. A review of the campaign found that targeting public awareness had limited success [14] and supported the ownership and engagement of the referring clinicians as paramount drivers of change for any intervention [14]. South Australia Medical Imaging (SAMI) is a statewide service provider that performs around 700,000 medical imaging examinations annually and being part of the public health system, patients have no out‐of‐pocket costs.
To advance our understanding of medical imaging referrals in South Australia, this study sought to capture the views of two cohorts—clinicians who make the referrals to SAMI and SAMI clinical staff, about the reasons clinicians refer for medical imaging.
2. Methods
The checklist for reporting results of internet E‐surveys (CHERRIES) was used as a framework [15] (See Appendix 1).
2.1. Study Design
This cross‐sectional survey included two cohorts: referring clinicians and SAMI clinical staff. The referring clinician group included consultants (fully qualified medical specialists), fellows, registrars, medical officers, nurse practitioners, and other clinicians who initiate medical imaging referrals such as physiotherapists and podiatrists. The SAMI clinical staff included radiologists, nuclear medicine specialists, radiology registrars and fellows, diagnostic radiographers, sonographers and nuclear medicine technologists. In this study, the term SAMI clinical staff refers to all clinical staff involved in the delivery of the medical imaging services. The study took place within the public health system of South Australia (SA Health) and included all 10 Local Health Networks (LHNs). These LHNs cover both metropolitan and regional areas, ensuring a broad geographical representation of participants, and responses were received from nine hospitals. For the purposes of confidentiality, hospitals A–F are metropolitan hospitals, hospital G is regional, whilst H is a dental location, and I includes responses from ‘other’.
The study aimed to recruit a minimum sample of 360 referring clinicians (from a total population of 5564 within SA Health) and 250 SAMI clinical staff (from a total population of 710), to estimate population proportions with 95% confidence and a margin of error of approximately ±5%. Sample size calculations were based on a representative sample of this total population, using the Survey Monkey sample size calculator [16]. Participants were informed about the study's objectives and participation was voluntary. To ensure anonymity and confidentiality, no personal identifying information was collected. Consent was stated in the preamble and implied by agreeing to participate in the survey and submitting responses [17]. All data were securely stored on South Australian Government and University of South Australia systems. This study has authorisation from Central Adelaide Local Health Network (CALHN) Human Research Ethics Committee (ID 16802), and the University of South Australia Human Research Ethics Committee (ID 206203).
2.2. Recruitment
Distribution of the referring clinician survey was between March 2023 and August 2023. Participants were invited via organisational email with a link and Quick Response (QR) code for the survey. Distribution for the SAMI staff survey was between February 2023 and July 2023. Recruitment for this cohort was done through the internal staff bulletin and posters placed at each SAMI site with a QR code.
2.3. Questionnaires/Surveys
Separate customised survey instruments were developed for two cohorts—referring clinicians and SAMI staff. Data collection was via electronic survey using Microsoft Forms [18]. SAMI radiation protection (KH, CB) and clinical research staff (MK, MA) developed and administered the survey, with the consultation of two University of South Australia academics with expertise in survey design.
Surveys were piloted among experts to ensure clarity and usability [19]. Multiple‐choice questions gathered demographic information, such as the participants' profession, years of experience, clinical specialty, and practice location. Questions asked participants' views on medical imaging overuse. In addition, Likert scale questions assessed clinicians' motivations for requesting medical imaging, perceptions of colleagues' motivations and behaviours for medical imaging, and the views of the SAMI staff who operationalise the requests, providing insights from three perspectives. Likert scale responses were collected via a 5‐point Likert scale from 1 to 5 (never, rarely, sometimes, often, always), and rating importance of various factors from irrelevant to essential (0–5). Open‐ended questions explored the participants' definitions of over‐requesting and their experiences with medical imaging requests. SAMI staff were asked about their views on referrer habits of over‐requesting they observed in practice [20].
2.4. Data Analysis
Data were exported from Microsoft Forms into SPSS 20 [21] and Microsoft Excel [22] for analysis. A biostatistician provided advice on the statistical analyses. Demographic characteristics and closed‐ended question responses were summarised using descriptive statistics [23]. Open‐ended responses were analysed using thematic analysis [24] which allowed the research team to identify common patterns, themes, and relationships within the data [25]. Responses were coded to obtain recurring themes [26] and an inductive approach allowed the themes to emerge from the data rather than from a prior theory [26].
3. Results
The distribution of respondents' site of practice is shown in Table 1, with the referring clinicians' working across nine sites and SAMI staff over seven sites. Most participants in both groups were from Hospital A.
TABLE 1.
Site of practice of both referring clinicians and SAMI clinical staff survey respondents.
| Referring clinician survey | SAMI clinical staff survey | |
|---|---|---|
| Total respondents | 138 | 251 |
| Number of hospitals | 9 (A to I) | 7 (A to G) |
| Hospital A | 44 (31.9%) | 85 (33.9%) |
| Hospital B | 16 (11.6%) | 21 (8.4%) |
| Hospital C | 20 (14.5%) | 50 (19.9%) |
| Hospital D | 7 (5.1%) | 29 (11.6%) |
| Hospital E | 19 (13.8%) | 28 (11.2%) |
| Hospital F | 1 (0.7%) | 22 (8.8%) |
| Hospital G | 9 (6.5%) | 16 (6.4%) |
| Additional hospitals (H and I) | Hospital H: 9 (6.5%), Hospital I: 13 (9.4%) | Not present |
The referring clinician survey reached a 2.48% response rate. Of the 138 respondents 91% (n = 125) primarily worked for SA Health, while the remaining 9% (n = 13) also worked in either GP practices, large private hospitals, or small community private hospitals. The length of practising was less than 5 years (n = 11), 5–10 years (n = 25), 10–15 years (n = 24), 15–20 years (n = 16), 20–25 years (n = 16), 25–30 years (n = 27), and greater than 35 years (n = 19). The SAMI staff survey achieved a response rate of 35.35%. The length of practising in the profession was less than 5 years (n = 44), 5–10 years (n = 47), 10–15 years (n = 33), 15–20 years (n = 26), 20–25 years (n = 47), 25–30 years (n = 40), and greater than 35 years (n = 14). Both groups took less than half an hour to complete the survey.
The majority of referring clinicians were medical professionals (n = 85), including consultants, registrars, fellows, resident medical officers, surgeons and general practitioners. Allied health staff, including podiatry, physiotherapy and speech pathology were the next most common participants (n = 21). Other professionals included nursing and nurse practitioners (n = 20), dentists, and dental hygienists (n = 12).
The SAMI staff participants included consultants, registrars and fellows (n = 80), and allied health professionals: radiographers (n = 128), nuclear medicine technologists (n = 21) and sonographers (n = 22).
3.1. Motivations for Requesting Medical Imaging
Tables 2, 3, 4 summarise three views: self‐reporting of referring clinicians, referring clinicians' views on their colleagues and views of SAMI staff.
TABLE 2.
Overview of referring clinicians' self‐thoughts, their thoughts on their colleagues' and SAMI staff thoughts on statements on what clinicians consider when referring a patient for imaging?
| Self | Colleagues | SAMI | ||
|---|---|---|---|---|
| Total respondents | 138 | 138 | 251 | |
| Improving overall patient health | Never | 0 (0%) | 12 (9%) | 2 (1%) |
| Rarely | 2 (1%) | 3 (2%) | 19 (8%) | |
| Sometimes | 6 (4%) | 22 (16%) | 83 (33%) | |
| Often | 44 (32%) | 55 (40%) | 123 (49%) | |
| Always | 86 (62%) | 46 (33%) | 24 (10%) | |
| Pressure or expectations from the patient, patient's family or carers | Never | 14 (10%) | 16 (12%) | 3 (1%) |
| Rarely | 63 (46%) | 30 (22%) | 30 (12%) | |
| Sometimes | 49 (36%) | 65 (47%) | 86 (34%) | |
| Often | 9 (7%) | 21 (15%) | 111 (44%) | |
| Always | 3 (2%) | 6 (4%) | 21 (8%) | |
| Reassurance or confirmation of disease absence | Never | 2 (1%) | 12 (9%) | 0 (0%) |
| Rarely | 10 (7%) | 9 (7%) | 8 (3%) | |
| Sometimes | 53 (38%) | 46 (33%) | 36 (11%) | |
| Often | 50 (36%) | 48 (35%) | 160 (64%) | |
| Always | 23 (17%) | 23 (17%) | 55 (22%) | |
| Expectation of other clinicians, e.g., preparation for ward rounds or MDTs | Never | 34 (25%) | 21 (15%) | 2 (1%) |
| Rarely | 33 (24%) | 17 (12%) | 19 (8%) | |
| Sometimes | 38 (28%) | 52 (38%) | 58 (23%) | |
| Often | 27 (20%) | 36 (26%) | 124 (49%) | |
| Always | 6 (4%) | 12 (9%) | 48 (19%) | |
| Streamlining patients through discharge | Never | 37 (27%) | 26 (19%) | 3 (1%) |
| Rarely | 39 (36%) | 20 (14%) | 15 (6%) | |
| Sometimes | 30 (22%) | 48 (35%) | 61 (24%) | |
| Often | 21 (15%) | 33 (24%) | 119 (47%) | |
| Always | 11 (8%) | 11 (8%) | 53 (21%) | |
| Defensive medicine—Fear of missed diagnosis—Being sued, complaints, Coronial inquests etc | Never | 25 (18%) | 22 (16%) | 4 (2%) |
| Rarely | 54 (39%) | 33 (24%) | 14 (6%) | |
| Sometimes | 57 (30%) | 50 (36%) | 29 (12%) | |
| Often | 9 (7%) | 25 (18%) | 100 (40%) | |
| Always | 8 (6%) | 8 (6%) | 104 (41%) | |
| Surrogate for scarce clinical resources, for example overnight | Never | 63 (46%) | 57 (30%) | 9 (4%) |
| Rarely | 40 (37%) | 31 (22%) | 36 (14%) | |
| Sometimes | 26 (19%) | 41 (30%) | 68 (27%) | |
| Often | 3 (2%) | 19 (14%) | 98 (39%) | |
| Always | 6 (4%) | 5 (4%) | 40 (16%) |
TABLE 3.
Overview of the referring clinicians' self‐thoughts, their thoughts on their colleagues' and SAMI staff thoughts on statements of how important clinicians consider clinical factors when referring a patient for imaging?
| Self | Colleagues | SAMI | ||
|---|---|---|---|---|
| Total respondents | 138 | 138 | 251 | |
| Eliminating the need for subsequent diagnostic testing | 0 (irrelevant) | 9 (7%) | 18 (13%) | 12 (5%) |
| 1 | 5 (4%) | 9 (7%) | 24 (10%) | |
| 2 | 15 (11%) | 15 (11%) | 36 (14%) | |
| 3 | 32 (23%) | 25 (18%) | 66 (26%) | |
| 4 | 52 (38%) | 49 (36%) | 76 (30%) | |
| 5 (essential) | 25 (18%) | 22 (16%) | 37 (15%) | |
| Perception of better care | 0 (irrelevant) | 19 (14%) | 17 (12%) | 3 (1%) |
| 1 | 27 (20%) | 13 (10%) | 17 (7%) | |
| 2 | 36 (20%) | 24 (17%) | 30 (12%) | |
| 3 | 30 (22%) | 37 (27%) | 63 (25%) | |
| 4 | 20 (14%) | 29 (21%) | 99 (39%) | |
| 5 (essential) | 14 (10%) | 18 (13%) | 39 (16%) | |
| To improve the health of the patient | 0 (irrelevant) | 0 (0%) | 43 (31%) | 9 (4%) |
| 1 | 1 (1%) | 25 (18%) | 45 (18%) | |
| 2 | 5 (4%) | 31 (22%) | 61 (24%) | |
| 3 | 5 (4%) | 21 (15%) | 84 (33%) | |
| 4 | 43 (31%) | 12 (9%) | 29 (12%) | |
| 5 (essential) | 84 (61%) | 6 (4%) | 23 (9%) | |
| Use of outdated guidelines/protocols/clinical pathways | 0 (irrelevant) | 54 (39%) | 31 (22%) | 9 (4%) |
| 1 | 33 (24%) | 25 (18%) | 14 (6%) | |
| 2 | 24 (17%) | 25 (18%) | 40 (16%) | |
| 3 | 12 (9%) | 29 (21%) | 35 (14%) | |
| 4 | 10 (7%) | 18 (13%) | 82 (33%) | |
| 5 (essential) | 5 (4%) | 10 (7%) | 71 (36%) | |
| Reduced reliance on the clinical diagnostic model (history and examination) | 0 (irrelevant) | 31 (22%) | 13 (10%) | 4 (2%) |
| 1 | 44 (32%) | 4 (3%) | 20 (8%) | |
| 2 | 24 (17%) | 11 (8%) | 37 (15%) | |
| 3 | 18 (13%) | 15 (11%) | 72 (37%) | |
| 4 | 14 (10%) | 44 (32%) | 71 (36%) | |
| 5 (essential) | 7 (5%) | 51 (37%) | 47 (19%) |
TABLE 4.
Overview of the referring clinicians' self‐thoughts, their thoughts on their colleagues' and SAMI staff thoughts statements of how important clinicians consider practical factors when referring a patient for imaging?
| Self | Colleagues | SAMI | ||
|---|---|---|---|---|
| Total respondents | 138 | 138 | 251 | |
| Scientific/Academic curiosity | 0 (irrelevant) | 41 (30%) | 34 (25%) | 26 (10%) |
| 1 | 36 (26%) | 25 (18%) | 55 (22%) | |
| 2 | 23 (17%) | 29 (21%) | 62 (25%) | |
| 3 | 25 (18%) | 31 (22%) | 61 (24%) | |
| 4 | 9 (7%) | 11 (8%) | 38 (15%) | |
| 5 (essential) | 4 (3%) | 8 (6%) | 9 (4%) | |
| Timing/Patient flow pressures | 0 (irrelevant) | 26 (19%) | 25 (18%) | 9 (4%) |
| 1 | 20 (14%) | 19 (14%) | 6 (2%) | |
| 2 | 10 (7%) | 19 (14%) | 24 (10%) | |
| 3 | 34 (25%) | 32 (23%) | 45 (18%) | |
| 4 | 27 (20%) | 36 (20%) | 94 (37%) | |
| 5 (essential) | 21 (15%) | 15 (11%) | 73 (29%) | |
| Lack of opportunity to clinically assess | 0 (irrelevant) | 44 (32%) | 34 (25%) | 8 (3%) |
| 1 | 37 (27%) | 24 (17%) | 24 (10%) | |
| 2 | 23 (17%) | 36 (20%) | 33 (13%) | |
| 3 | 18 (13%) | 24 (17%) | 52 (21%) | |
| 4 | 9 (7%) | 20 (14%) | 79 (31%) | |
| 5 (essential) | 7 (5%) | 8 (6%) | 55 (22%) | |
| Knowledge regarding use of diagnostic imaging tests | 0 (irrelevant) | 8 (6%) | 13 (10%) | 24 (10%) |
| 1 | 10 (7%) | 8 (6%) | 57 (17%) | |
| 2 | 12 (9%) | 21 (15%) | 55 (22%) | |
| 3 | 20 (14%) | 43 (31%) | 66 (26%) | |
| 4 | 49 (36%) | 34 (25%) | 38 (15%) | |
| 5 (essential) | 39 (36%) | 19 (14%) | 26 (10%) | |
| Knowledge of radiation risks for diagnostic imaging tests | 0 (irrelevant) | 2 (1%) | 15 (11%) | 45 (18%) |
| 1 | 15 (11%) | 9 (7%) | 81 (32%) | |
| 2 | 22 (16%) | 32 (23%) | 41 (16%) | |
| 3 | 23 (17%) | 32 (23%) | 35 (14%) | |
| 4 | 41 (30%) | 32 (23%) | 32 (13%) | |
| 5 (essential) | 35 (25%) | 18 (13%) | 17 (7%) | |
| Recent technological advancements | 0 (irrelevant) | 7 (5%) | 18 (13%) | 36 (11%) |
| 1 | 15 (11%) | 7 (5%) | 63 (25%) | |
| 2 | 18 (13%) | 36 (26%) | 66 (26%) | |
| 3 | 36 (26%) | 32 (23%) | 51 (20%) | |
| 4 | 57 (30%) | 32 (23%) | 26 (10%) | |
| 5 (essential) | 20 (14%) | 13 (10%) | 17 (7%) |
Table 2 summarises the reasons for referral to medical imaging. Most clinicians ‘always’ and ‘often’ request medical imaging ‘to improve overall health of patient’. In addition, ‘reassurance or confirmation of disease absence’ consistently scored highly across ‘sometimes’, ‘often’ and ‘always’. Factors clinicians ranked as of low importance for referring were, ‘pressure or expectations from the patient, patient's family or carers’, along with ‘surrogate for scarce clinical resources, for example overnight’. There was considerable overlap between factors referring clinicians consider themselves, and the factors they thought their colleagues considered. There was one exception however—‘preparation for ward rounds or MDTs’, and ‘streamlining patients through discharge’—where referring clinicians thoughts on their colleagues were distributed across the full range of ‘never’ to ‘often’.
However, the views of the SAMI staff on the same listed statements contrasted with those of the referring clinicians. The only similarities in all three perspectives was ‘improving overall patient health’, followed by ‘reassurance or confirmation of disease absence’. SAMI staff predominantly felt that medical imaging referrals are ordered, ranging from ‘sometimes to always’, for reasons such as ‘streamlining patients through discharge’, ‘defensive medicine’, ‘expectation of other clinicians’, ‘pressure or expectations from the patient, patient's family or carers’ and as a ‘surrogate for scarce clinical resources, for example overnight’.
Table 3 summarises various clinical factors that clinicians consider important when requesting medical imaging. ‘Eliminating the need for subsequent diagnostic testing’ was an important factor from all three viewpoints. The majority of clinicians 61% (n = 84) endorsed medical imaging as essential to ‘improve the health of the patient’, while they perceived that only a small proportion of their colleagues (4%, n = 6) considered this factor. Only a few clinicians (5%, n = 7) self‐reported a ‘reduced reliance on the clinical diagnostic model (history and examination)’ whereas they perceived this to be a factor influencing their colleagues' practice (37%, n = 51).
A similar pattern of results was seen in ‘eliminating the need for subsequent diagnostic testing’, ‘perception of better care’, and ‘use of outdated guidelines/protocols/clinical pathways’ for clinicians self‐reporting and of their colleagues. SAMI staff placed greater importance on, and held contrasting views, to clinicians on the reasons medical imaging was ordered, specifically citing ‘perception of better care’, ‘a reduced reliance on the clinical diagnostic model (history and examination)’ and ‘use of outdated guidelines/protocols/clinical pathways’. Just under half of SAMI staff did not support the view that medical imaging was performed ‘to improve the health of the patient’.
Table 4 summarises practical factors that clinicians consider important when referring a patient for medical imaging, The data suggests that referring clinicians consider ‘knowledge about the use of diagnostic imaging tests’ and ‘awareness of radiation risks’ essential for themselves, but not for their colleagues. SAMI staff did not believe that clinicians have a sound ‘knowledge of radiation risks for diagnostic imaging tests’ or ‘knowledge regarding the use of diagnostic imaging tests’. Responses were similar across all three views regarding ‘scientific or academic curiosity’ when referring. Clinicians did not view the ‘lack of opportunity to clinically assess’ as an important practical factor when referring, either for themselves or their colleagues; however SAMI staff, considered this an essential factor when referring. Referring clinicians gave varied responses regarding ‘timing/patient flow pressures’ as reasons for medical imaging, whereas SAMI staff consistently reported that ‘timing/patient flow pressures’ influenced the request process.
Open‐ended text questions asked if any additional factors had not been included in the survey and participants were invited to provide comments. Some clinicians stated that they did not know the motivators of their colleagues and could not answer on behalf of them. Others mentioned that failing to perform the correct medical imaging the first time could occur, possibly due to the availability of equipment or the timing of the referral, such as overnight requests. Other views included the experience of the clinician, the lack of confidence by junior doctors in their clinical diagnostic skills and using medical imaging as a tick box approach to requesting.
People in regional, and remote areas may undergo more medical imaging due to limited clinical opportunities to be seen, and in some cases, access to different types of medical imaging modalities in rural areas was another factor influencing the decision‐making process. SAMI staff also reported factors such as clinicians' poor clinical skills and heavy reliance on diagnostic tests that could influence the referral process. Some staff reported that clinicians lack the time to assess the patient thoroughly, especially in the Emergency Department, with some patients having multiple medical imaging visits.
3.2. Perceived Overuse of Medical Imaging
Response rates to the question ‘What is your definition of over‐requesting?’ were high with 130 of the 138 referring clinicians, and 246 of the 251 SAMI staff providing explanations. Definitions varied widely as can be seen in Data S1.
Key themes for over‐requesting of medical imaging that emerged broadly related to staff, knowledge, and resources. Some examples of these themes included concerns that clinicians may lack professional experience or have limited experience, leading to the use of medical imaging as a triage tool in place of clinical assessment. Participants also noted instances where imaging was required despite not being clinically indicated or unlikely to alter patient management. Other issues raised included ordering multiple medical imaging studies for the same clinical question, repeated scans when the outcome would likely not change, and the fear of missed diagnoses potentially resulting in litigation. Workload pressures were also highlighted, such as insufficient information or clinical history provided on referrals, lack of justification for medical imaging requests, and ordering outside of current guidelines.
Referring clinicians were subsequently asked ‘Have you ever deliberately requested medical imaging that would meet the definition you have given to the previous question?’. Of the 138 respondents 22.5% (n = 31) stated yes, 71% (n = 98) said no, and 6.5% (n = 9) stated prefer not to answer. Referring clinicians were also asked if they believe their colleagues over‐request medical imaging. From the 138 responses, 18.8% (n = 26) stated no, 20.3% (n = 28) stated rarely, 25.4% (n = 35) said occasionally, while 24.6% (n = 34) stated sometimes and 10.9% (n = 15) said often. Of the 251 SAMI staff that were asked if they believe referring clinicians over‐request medical imaging, 89.6% (n = 225) said yes and 10.4% (n = 26) stated no.
4. Discussion
Despite a multitude of clinical guidelines on best practice in medical imaging, there continue to be challenges in how this is implemented in practice. Therefore, the aim of this study was to capture the views of clinicians and the people who operationalise their requests about factors that are related to medical imaging requests. The findings from this study indicate a range of important outcomes. While clinicians report that several factors influence their requests for medical imaging, only a small number acknowledged views related to over‐requesting or admitted to deliberately over‐requesting. This was in stark contrast to SAMI staff, who reported that a much larger proportion of clinicians over‐request medical imaging. The observed disparity between the referring clinicians and SAMI staff regarding perceptions of over‐requesting medical imaging may reflect differing perspectives and areas of expertise. Referring clinicians may have limited awareness of the technical and resource implications associated with imaging requests, whereas SAMI staff may have less insight into the clinical reasoning underlying imaging decisions. Consequently, the true extent of over‐requesting is likely to lie somewhere between these perspectives. This highlights the need for ongoing collaboration between referring clinicians and SAMI staff to optimise the appropriate use of medical imaging resources.
A myriad of factors appears to influence how clinicians decide to request medical imaging, including patient‐related, staff‐related and system‐related factors. This research contributes to addressing an important knowledge gap in Australian literature on medical imaging.
The findings from this research add to the global literature on medical imaging, and inappropriate medical imaging requests. A survey of radiologists in Norway investigated their perceptions on the increasing use of medical imaging in the Norwegian context and highlighted several factors such as over‐investigation and insufficient referral information as the most frequent causes [27]. In a separate Norwegian study, a better understanding of how radiologists and general practitioners view the medical imaging referral process, including issues of justification and unnecessary medical imaging, was sought to gain a better understanding of these perspectives. The study found a high rate of unnecessary medical imaging referrals or referrals lacking appropriate justification, largely driven by pressure from patients or referring clinicians. To reduce unnecessary medical imaging and improve patient safety, clearer guidelines and better communication between doctors and radiologists are needed. Currently no further research has tested any of the interventions to address unnecessary medical imaging [28]. Therefore, in addition to lessons learnt from other jurisdictions and the perspectives of medical imaging professionals on medical imaging practices, our survey findings offer valuable insights that can inform strategies to change or improve practice. Understanding what motivates and drives clinicians can support the development of effective strategies for improving medical imaging practices. An Australian study explored the mapping of drivers of over diagnosis to potential solutions within healthcare services. This work highlighted both the responsibilities of clinicians and the systemic factors that influence decision‐making [12].
An interesting finding from our research was the varying drivers for medical imaging requests. While the referring clinicians' drivers related to logistical challenges, access to results, and the need to improve clinical diagnostic confidence, SAMI staff drivers focus more on improving referral quality, reducing reliance on medical imaging, and accountability. Both cohorts acknowledge the demands and pressures involved in managing and treating patients. The study found both referring clinicians and SAMI staff acknowledged a need for better medical imaging training, improved decision‐support tools, and system‐level changes to manage medical imaging demand.
Despite decades of evidence on medical imaging practices, and the use of evidence‐based recommendations, achieving this in practice remains challenging [14]. Implementation of best practices and interventions indicates that understanding local context and jurisdictional drivers is critical in enhancing medical imaging practice.
4.1. Strengths and Limitations
This research was underpinned by best practice standards in the conduct and reporting of survey research. The participant groups included referring clinicians, and SAMI staff, capturing a range of perspectives. Respondents represented a diverse range of professions, and varying levels of experience, ranging from junior level to seasoned professionals with decades of service. This diversity provided valuable insights from their roles and responsibilities across a wide range of geographical locations in South Australia. Some limitations should be acknowledged, including the potential for self‐reporting bias and a low response rate. The low response rate could be attributed to competing and conflicting SA Health demands, winter furlough, ramping, staff shortages, and survey fatigue. Feedback was received from referrers who did not respond due to the length of the survey, the design and not having the option to skip questions. Researcher bias, particularly in the qualitative data analysis should be acknowledged [29]. However, this was mitigated through the use of reflexivity and strategies to enhance trustworthiness such as peer review and debriefing.
5. Conclusion
Our data suggest that various motivations influence medical imaging referrals. To support improvement, locally tailored strategies are needed to raise awareness of key issues and identify areas of change. While most referring clinicians reported not engaging in over‐requesting, a smaller group acknowledged doing so. In contrast, most SAMI staff perceived that over‐requesting of medical imaging does occur. This discrepancy highlights the need for further investigation, and ongoing research to explore the underlying reasons and identify strategies to reduce overuse.
Ethics Statement
This study has authorisation from Central Adelaide Local Health Network (CALHN) Human Research Ethics Committee (ID 16802), and the University of South Australia Human Research Ethics Committee (ID 206203).
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Appendix S1: Supporting information.
Data S1: Supporting information.
Acknowledgements
The authors would like to thank Professor Maureen Dollard, and in particular, the contributions of Professor Kurt Lushington, both Professors in Psychology at University of South Australia, in the study design. A special note of thanks to Mr John Hillier, SA Medical Imaging for his valuable assistance with Excel.
Data Availability Statement
The data for this study will not be shared, as we do not have permission from the participants or ethics approval to do so.
References
- 1. Bercovich E. and Javitt M. C., “Medical Imaging: From Roentgen to the Digital Revolution, and Beyond,” Rambam Maimonides Medical Journal 9 (2018): 9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Hussain S., Mubeen I., Ullah N., et al., “Modern Diagnostic Imaging Technique Applications and Risk Factors in the Medical Field: A Review,” BioMed Research International 2022 (2022): 5164970. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Waheed S., Tahir M. J., Ullah I., et al., “The Impact of Dependence on Advanced Imaging Techniques on the Current Radiology Practice,” Annals of Medicine and Surgery 78 (2022): 103708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Australia Po , “Review of Funding for Diagnostic Imaging Services: Final Report 2012,” https://www.aph.gov.au/DocumentStore.ashx?id=3af9956d‐9e28‐4855‐a75f‐c7e4f4ac2419&subId=252414.
- 5. Mendelson R. M., “Diagnostic Imaging: Doing the Right Thing,” Journal of Medical Imaging and Radiation Oncology 64 (2020): 353–360. [DOI] [PubMed] [Google Scholar]
- 6. Statistics ABo , “Australian Social Trends,” (1996, accessed 13 October 2025), https://www.abs.gov.au/ausstats/abs@.nsf/2f762f95845417aeca25706c00834efa/867467e96d7f7e47ca2570ec0073d9eb!OpenDocument.
- 7. Australian Institute of Health and Welfare , “Medicare‐Subsidised GP, Allied Health and Specialist Health Care Across Local Areas: 2013–14 to 2018–19,” (accessed 2022 February 15), https://www.aihw.gov.au/reports/primary‐health‐care/medicare‐subsidised‐health‐local‐areas‐2019/contents/diagnostic‐imaging.
- 8. Australian Institute of Health and Welfare , “Medicare‐Subsidised GP, Allied Health and Specialist Health Care Across Local Areas: 2022–23,” (accessed 2025 July 7), https://www.aihw.gov.au/reports/primary‐health‐care/medicare‐subsidised‐care‐2022‐23/contents/diagnostic‐imaging‐services.
- 9. Maskell G., “Why Does Demand for Medical Imaging Keep Rising?,” BMJ (Clinical Research Ed.) 379 (2022): o2614. [DOI] [PubMed] [Google Scholar]
- 10. Parliament of Australia , “Overview of the Health Workforce [Internet],” (2019, accessed 2023 November 9), https://www.aph.gov.au/About_Parliament/Parliamentary_Departments/Parliamentary_Library/pubs/BriefingBook46p/HealthWorkforce#:~:text=While%20the%20overall%20number%20of,healthcare%2C%20and%20new%20health%20technologies.
- 11. Litkowski P. E., Smetana G. W., Zeidel M. L., and Blanchard M. S., “Curbing the Urge to Image,” American Journal of Medicine 129 (2016): 1131–1135. [DOI] [PubMed] [Google Scholar]
- 12. Pathirana T., Clark J., and Moynihan R., “Mapping the Drivers of Overdiagnosis to Potential Solutions,” BMJ (Clinical Research Ed.) 358 (2017): j3879. [DOI] [PubMed] [Google Scholar]
- 13. Livingston C. J., Freeman R. J., Mohammad A., et al., “Choosing Wisely in Preventive Medicine: The American College of Preventive Medicine's Top 5 List of Recommendations,” American Journal of Preventive Medicine 51, no. 1 (2016): 141–149. [DOI] [PubMed] [Google Scholar]
- 14. Tian E. J., Nguyen C., Chung L., Morris C., and Kumar S., “The Effectiveness of Public Awareness Initiatives Aimed at Encouraging the Use of Evidence‐Based Recommendations by Health Professionals: A Systematic Review,” Journal of Patient Safety 20 (2024): 147–163. [DOI] [PubMed] [Google Scholar]
- 15. Eysenbach G., “Improving the Quality of Web Surveys: The Checklist for Reporting Results of Internet e‐Surveys (Cherries),” Journal of Medical Internet Research 6 (2004): e34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. SurveyMonkey , Sample Size Calculator San Mateo (Survey Monkey, 2022). [Google Scholar]
- 17. Gupta U. C., “Informed Consent in Clinical Research: Revisiting Few Concepts and Areas,” Perspectives in Clinical Research 4 (2013): 26–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Burns K. E. A., Duffett M., Kho M. E., et al., “A Guide for the Design and Conduct of Self‐Administered Surveys of Clinicians,” Canadian Medical Association Journal 179 (2008): 245–252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Elangovan N. and Sundaravel E., “Method of Preparing a Document for Survey Instrument Validation by Experts,” MethodsX 8 (2021): 101326. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Yates J. and Leggett T., “Qualitative Research: An Introduction,” Radiologic Technology 88 (2016): 225–231. [PubMed] [Google Scholar]
- 21. IBM , IBM SPSS Statistics for Windows, Version 27.0 (IBM Corp, 2020). [Google Scholar]
- 22. Corporation M , Microsoft Excel. Version 16.0 (Microsoft Corporation, 2021). [Google Scholar]
- 23. Vetter T. R., “Descriptive Statistics: Reporting the Answers to the 5 Basic Questions of Who, What, Why, When, Where, and a Sixth, So What?,” Anesthesia & Analgesia 125 (2017): 1797–1802. [DOI] [PubMed] [Google Scholar]
- 24. Braun V. and Clarke V., “Using Thematic Analysis in Psychology,” Qualitative Research in Psychology 3 (2006): 77–101. [Google Scholar]
- 25. Creswell J. W. and Creswell J. D., Research Design: Qualitative, Quantitative, and Mixed Methods Approaches, 3rd ed. (Sage, 2017). [Google Scholar]
- 26. Naeem M., Ozuem W., Howell K., and Ranfagni S., “A Step‐by‐Step Process of Thematic Analysis to Develop a Conceptual Model in Qualitative Research,” International Journal of Qualitative Methods 22 (2023): 205789. [Google Scholar]
- 27. Lysdahl K. B. and Hofmann B. M., “What Causes Increasing and Unnecessary Use of Radiological Investigations? A Survey of Radiologists' Perceptions,” BMC Health Services Research 9 (2009): 155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Kjelle E., Andersen E. R., Brandsæter I. Ø., and Hofmann B. M., “Norwegian General Practitioners' and Radiologists' Perspectives on the Referral, Justification, and Unnecessary Imaging—A Survey,” Scandinavian Journal of Primary Health Care 42 (2024): 574–581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Braun V. and Clarke V., “Toward Good Practice in Thematic Analysis: Avoiding Common Problems and Be(Com)ing a Knowing Researcher,” International Journal of Transgender Health 24 (2023): 1–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Appendix S1: Supporting information.
Data S1: Supporting information.
Data Availability Statement
The data for this study will not be shared, as we do not have permission from the participants or ethics approval to do so.