Patient initiated radiology requests: proof of wellness through images

Introduction

Traditionally, GPs have been central in patient care (Palanica et al. 2019), acting as the ‘gatekeepers’ to medical interventions by, for example, directing medical services, prescribing medications and facilitating decision-making (Palanica et al. 2019). However, with increased availability of online health information, patients are more informed and active in their health, particularly with some patients having a sense of consumeristic entitlement (Lee et al. 2014). This is not a new phenomenon, as patients are seen as developing their intellectual curiosity and ‘getting what they want’ (Lee et al. 2015). One element of consumer-driven care is seen in patients requesting a referral for a radiological study from their GPs. It is thought that some of these requests are fuelled by the availability of online health information, where individuals are more aware of their health needs and goals (Lee et al. 2015; Hodyl et al. 2020). Platforms, such as social media (Esen et al. 2019), mobile applications (Groß and Schmidt 2018) and community led online groups (Fiksdal et al. 2014), are encouraging individuals to become proactive during patient–doctor consultations. As such, doctors’ gatekeeping roles are diminishing as consumer driven needs increase (Lee et al. 2015).

In Australia, only qualified health practitioners, such as GPs, can write a referral letter for radiological services, including imaging and image guided procedures in computer tomography (CT), magnetic resonance imaging (MRI), X-rays and ultrasounds. Thus, patients who want to have a radiology test performed need to approach their GP for the referral. Moreover, with Medicare subsidised services, most of these requests for studies come at little to no cost to patients. To explore factors contributing to patients’ requests for radiological imaging, we used the Andersen Newman Model (ANM) for healthcare utilisation (Andersen and Newman 1973). This model provides a framework with which to discover aspects that facilitate or impede the use of healthcare resources. Past studies using the model have addressed the phenomenon of doctors fulfilling requests for patient satisfaction (Jerant et al. 2018), explored patients’ spoken and unspoken expectations during a clinical consultation (Fenton et al. 2015) and further explored patients’ perceived needs during clinical visits (Benetoli et al. 2018). Most of these studies have been situated within a general practice context with requests for prescribed medications (Kravitz et al. 2003; Paterniti et al. 2010; Hogue et al. 2012; Fenton et al. 2015), while limited work has focused on patient perceived needs for radiological tests and their experiences around these requests (Alamri et al. 2014; Walderhaug et al. 2022).

The ANM (Andersen and Newman 1973) provides a conceptual framework in recognising antecedents for healthcare use through understanding human behaviour, specifically within the patient–doctor relationship (Andersen 1995; Bradley et al. 2002; Petrovic and Blank 2015). A study by Strain (1990) on GP home visits to the elderly, suggested the need to include psycho-social and socio-cultural elements to ANM (Strain 1990). Petrovic and Blank (2015) described adherence to prescribed medication among the elderly with chronic illnesses using ANM, recognising how the model lends itself to examining elements of trust between patients and doctor. Thus, the original model (Andersen and Newman 1973), the purpose of which was to discover conditions that facilitate or impede the use of health care, has evolved throughout the years, originally focusing on the family, but now applicable to individual patients (Andersen 1995; Bradley et al. 2002; Andersen and Newman 2005). Kravitz et al. (2003) used a modified version of ANM on patients requesting services within a general practice. Their study demonstrated the socio-cultural characteristics or ‘predispositions’ that exist prior to illness, and extended this to include the logistics or ‘enabling’ factors of accessing care, and described the ‘need’ for seeking such care. Using this modified framework by Kravitz et al. (2003), we describe ‘predisposition’, ‘enabling’ and ‘need’ factors that initiate patients’ requests for radiological studies. We hope, by using a two-phase explanatory design, to address the gap in understanding such patient-initiated radiological requests, particularly with consideration of patient–doctor interactions.

Thus, ANM is an appropriate model to examine in our study, as the individual unit of analysis is the patient requesting a radiological service, but it also lends itself to the expansion of the patient–doctor dyad in which its application occurs during clinical negotiations, such as the doctor acquiescing to a patient’s specific request. In this study, we included the survey instrument, the open-ended questions and concurrent interviews to further identify unique views and experiences.

Methods

This mixed method study was approved by the University of Sydney Ethics Review Board. The data was collected from January 2022 to February 2022, starting with a survey that aimed to understand participants’ experiences as a patient requesting a radiological referral from a doctor when attending a primary care facility. The questionnaire provided volunteers with the option of a follow-up interview, to further explore their responses. Sixty participants who had requested a radiological referral from their doctor within the last 5 years responded to the survey, and eight participants were interviewed.

Results

There were 60 respondents who completed the survey of whom 10 agreed to be interviewed, however, thematic saturation was achieved after eight and further interviews were ceased. A true survey response rate was unable to be determined using the external market research panels, as some panel members may have overlooked the invitation, or declined participation based on the incentives offered. Most of the requests were for X-rays and ultrasounds, as indicated in Table 1.

Fig. 1 indicates the steps that drive patients to request radiological referrals and demonstrates variables at play within clinical negotiations between patients and their GPs. These variables form the basis of patients having requests fulfilled.

Fig. 1.

Emerging themes from interviews and surveys with respect to drivers of patient requests. Adapted from a study by Kravitz et al. (2003), p. 1674.

Discussion

This study aimed to understand, through cross-sectional surveys and semi-structured interviews, key findings during doctor–patient clinical negotiations when patients request a radiological referral from their doctors in primary care facilities. The results showed that participants were well-informed about their health and desired visual proof of wellness to alleviate their anxiety over new symptoms. The study found that visual proof, predominantly using X-rays and ultrasounds, played a vital role in satisfying patients’ needs. Most participants sourced information from reputable online sites and felt validated by their doctors, indicating that they were not making unnecessary requests. For example, in our study, visual proof through real-time ultrasound technology helped a mother see her baby’s beating heart and provided reassurance. The participant stated that seeing the images on the screen enabled her to visualise in her mind what was happening within her. Participants espoused the view that care has been received, once images were acquired, so both the GP and the participant could assess the cause for concern. This is in contrast to a study on consumer needs by Lee et al. (2015) who found almost half their population reported at least some difficulty locating desired health information (Lee et al. 2015). However, according to Jutel (2017), using online sourced information has several caveats, one of which is in ‘playing Dr Google’, where patients’ incorrectly self-diagnose and potentially influence the doctor’s better judgement (Gransjøen et al. 2018; Docking et al. 2022).

However, using the ANM to assess clinical requests in office practices, Kravitz et al. (2003) concluded that patients do in fact influence the clinical outcome during consultations. Their results supported the notion that patients’ requests are increasingly pervasive and there is a need to better understand and manage such requests, particularly non-clinically indicated requests. The ANM posits that healthcare use is determined and often weighted by the individuals’ beliefs. For example, the patients’ attitudes about health services, their desire to know about diseases and how they position themselves within the system of healthcare use (Andersen and Newman 2005). Our study indicates that participants are actively seeking resources, notably evidence-based websites, that grant them confidence in requesting certain tests or procedures. Such confidence in a consumeristic patient leads to higher expectations during consultations. Where such expectations involve doctors recognising and validating their requests, and if such acknowledgement is not granted, participants are willing to seek alternate doctors to fulfil their requests.

However, we recognise that granting patient-initiated requests, particularly with non-clinically indicated tests, may be seen as depleting doctors’ valuable time and healthcare resources (Llanwarne et al. 2017). Our study does not highlight the negative impact of patient influence causing unnecessary testing, interventions and possibly delayed diagnosis and poor stewardship of healthcare resources (Le et al. 2018; Walderhaug et al. 2022). Although patients’ expectations for doctors to recognise and validate their requests, particularly for non-clinically indicated tests, seem to be prevalent in our study, employing strategies to discourage unwarranted imaging, such as using a ‘watch and wait’ strategy (Fenton et al. 2021) or suggesting alternatives, may prove beneficial to decrease healthcare costs (Docking et al. 2022) However, doctors’ inclination towards patient satisfaction and practicing defensive medicine may contribute to responding favourably to patient requests (Nilsen and Malterud 2017).

Interestingly, Jerant et al. (2018) undertook a cross-sectional observational study on clinicians denying patient requests on 1319 clinical visits made to 56 doctors by 1141 outpatients. The study found most participants reported their request being fulfilled, n = 1441 (85%), and of the visits within the requested categories, there were 153 (11%) radiological requests. But some participants understood the doctors’ gatekeeping role and tried to remain ‘a patient’ (Kravitz et al. 2003). However, overall, seeing a doctor seems to have become a formality to secure a referral letter, rather than relying on the doctors’ knowledge and guidance (De Silva et al. 2023). This study, albeit small, found that doctors, in competing for patient satisfaction, maintaining relationships and keeping pace with clinical productivity, tend to adopt a default approach of agreeing to patient requests (Sabbatini et al. 2014; Carey et al. 2015).

Thus, in answering our objectives, themes emerging from our study were mapped onto the ANM as ‘predisposing’, ‘enabling’ and ‘need’ elements for patients requesting radiological imaging. The predisposing elements were desire to know, proof in images and health anxiousness. The enabling elements included knowledge source, expert guidance and patient expectations, while the need factors included symptoms and ‘Dr options.’ Our study found that the primary driver for requesting referrals was anxiousness over new undiagnosed illness or experience of symptoms where visual proof was needed mainly through X-rays and ultrasounds. Such visual proof satisfied individuals to pursue goals in restoring their health and wellness. Furthermore, our study supported the idea that participants were satisfied achieving the intended, desired outcomes, and if requests were not met, participants were inclined to choose an alternative doctor who would fulfil their request.

Limitations and strength

This study has several limitations that should be considered when interpreting the results. Firstly, the data was limited to the Australian population and within the context of a health system providing predominantly ‘free’ imaging services to patients. Secondly, it was collected retrospectively relying on participants’ recall, which could introduce bias, affect the accuracy and may tend to allow overinterpretation of findings. Additionally, the study only investigated participants’ views, and it would be valuable to assess doctors’ perspectives to gain a more comprehensive understanding of the issue. Moreover, the recruitment of participants was conducted online using Qualtrics and limited to purposively sampled participants requesting referrals from a GP working in primary care facilities. This could potentially skew the results and limit the generalisability of the findings to the broader Australian population. Our study also did not address whether GP’s performance was reflective of the proportion of tests ordered, which may influence GP’s responses to patients requests.

Our study did employ a mixed method approach, which provided valuable insights into patients’ perspectives. The quantitative component was employed to bring a broad approach to the research question while using the interview responses to gain more depth. Additionally, further research could confirm these results, and investigate full paying patients who are not covered under Australian Government’s Medicare health insurance rebate to determine if frequency of requests would be different to those who absorb little to no costs. Moreover, further investigation could determine if responses from GPs would be similar to those of our study.

Conclusion

This paper identifies several elements that are involved in clinical negotiations between patients and doctors. The ANM was used in the study as a way of explaining the use of health care within the Australian population, in particular, where patients request imaging referrals from their GPs in a primary care setting. Our findings suggest that patients are more informed than ever before and are confident in their knowledge of the types of radiological scans and interventions they require. Patients also found comfort in seeing what was happening inside their bodies, particularly during obstetric imaging, and wanted their doctors to validate their concerns by complying with their requests for imaging referrals. While our study did not highlight the importance of appropriate use of imaging, particularly for non-clinically indicated tests, it recognised the increasing involvement of patients as healthcare participants. Overall, our study provides some new insights into the evolving patient–doctor dyadic relationship in radiological services.

Data availability

Data supporting the results can be found at the University of Sydney research data storage facility.

Conflicts of interest

The authors declare that they have no conflicts of interest.

Declaration of funding

This research did not receive any specific funding.

Author contributions

All authors contributed to previous versions of the manuscript and the final draft written by Lizzie De Silva who conceived and designed the study as part of her dissertation. All authors read and approved the final manuscript.