Optimising access and provision of interventional radiology service to patients by utilising a new referral system

Abstract

Objective:

Implementing a streamlined interventional radiology (IR) service in the UK has been a challenge. This study aims to review a set of changes introduced in IR at a tertiary centre, including a new referral process and the designation of IR clinical nurse specialists.

Methods:

A new process of referring patients to IR using a single generic referral pathway was implemented, replacing an order dropdown-based system. A qualitative survey was designed and distributed as a single-use web link in order to assess the satisfaction and impact of this new process. Responses were based on Likert scale and pertained to perceived qualities of the new referral process. Data analysis was performed to identify specialty and grade-specific trends and possible differences amongst groups.

Results:

Findings from 98 respondents revealed a strong overall satisfaction with the new referral method and support for its continuation. Subgroup analysis by specialty, concluded medical specialties rated the new referral system more favourably than surgical specialties across all aspects: time efficiency, ease of use, periprocedural support and overall user experience. The new system also increased departmental productivity with an increase in the number of patients treated by 11.2%.

Conclusion:

Micropolicy changes within individual IR departments such as the replacement of a request-based referral system to one which puts IR in control of vetting and patient flow is one of many changes that reinforce the transformational phase of this specialty.

Advances in knowledge:

Micropolicy changes within IR departments are key in the progression and widespread recognition of the specialty.

Introduction

Analysing the pace of change in healthcare over the last century, one realises the massive impact rapid turnover of technologies has had on the role of the average physician. In the context of ever-increasing budget constraints coupled with the growing need for cost reduction and quality improvement,¹ a spiral of technological advancements has “forced” medicine into its constituent specialties and subspecialties²; i.e. cohorts of physicians devoting their intellect to narrower areas of interest.

Whether the status quo resembles an overfragmented system or indeed one which needs further fragmentation still, has been a topic of debate.³ Take interventional radiology (IR): a specialty less than 100 years of age, whose technological advancements are growing faster than the specialty itself. IR is growing laterally, “poking the borders” imposed by neighbouring specialties, claiming outdated or unclaimed territory. While “territories” are open for debate, one thing is irrefutable: IR has emerged as a specialty and it’s clearly here to stay.

As IR continues to expand, it demands more from the hospital environment in which it is sited. Of course, the relationship is reciprocal as the hospital and its constituent specialties are demanding more from IR too. And as this reciprocity grows, the need for a more streamlined, organised and integrated specialty becomes more important than ever.⁴ In recognition of the above, one of the largest IR departments in a tertiary centre in the UK has embarked on a series of changes to refine its IR services. The most notable of these are the integration of a new referral system, introduction of an electronic vetting process and the designation of dedicated IR clinical nurse specialists to liaise with referrers. Together, these mark a paradigm shift, away from an outdated order-based system to one which puts IR in control of referral vetting, perioperative optimisation and overall patient flow. These changes also mark the beginning of a true clinical specialty, where the interventionist serves the patient and not solely the referrer. This study aims to assess the perceived effects of these changes on the referrers and the IR department, and discuss their implications in the transformation of IR.

Majority of referrals to UK’s NHS IR departments across the country, are still based on traditional radiology request forms, commonly submitted through radiology information systems (RIS) or paper-based forms. Traditionally, referring clinicians have been able to choose a specific procedure from a long list of IR procedures, deemed, in their eyes, to be most suitable. This oftentimes creates confusion and causes unnecessary delays as the vetting radiologist is technically unable to amend that request, therefore having to reject it. To overcome this challenge, a new referral pathway was implemented and assessed. This incorporates a unidirectional referral pathway to IR by completing a single referral form through RIS (Figure 1). The vetting radiologist is then able to electronically vet the request and choose the appropriate IR code in RIS, while taking into account the specific clinical needs of the patient at hand. The latter is achieved by asking the referrer to provide comprehensive and up-to-date clinical information for the current indication. The radiologist is now asked to use a predefined template that instructs for any pre-procedural investigations that need to be conducted.

Figure 1.

Old and new system of referring to the IR department. Abbreviations: CNS, Clinical nurse specialist; IR, interventional radiologyLA, Local anaesthetic.

Methods and materials

Study objectives

The aim of the study was twofold: firstly, to assess the impact of the new referral system on the overall satisfaction of referrers by using criteria that underpin value-based healthcare¹; and secondly, to investigate the combined effect, if any, of the referral system, electronic vetting and the newly designated role of clinical nurse specialists on activity levels for the department.

Participant recruitment

To assess user satisfaction, a list of eligible participants was compiled through retrospective identification of staff members who referred patients to the IR department within the 12-month period from the day of implementation of the new referral system. Eligible participants were emailed directly using the hospital’s email network and were asked to participate by completing an anonymised single-use web link. A total of 300 emails were distributed with a response rate of 32.6%, yielding a total of 98 participants.

Data collection

Participants completing the survey were asked to report basic demographic details including their grade, specialty and whether their referral was for a paediatric or adult inpatient. The participants were then asked to rate the following aspects of the new IR referral system using a Likert scale from 1 to 5: time commitment, ease of use (including change, if any, in ambiguity and need for information-collection), uncertainty of the need for referral, peri-procedural support and overall user experience (Table 1). A rating of 1 corresponded to “not at all” whereas a rating of 5 corresponded to “significantly improved”. One question was reserved for participants to qualitatively comment on their experience in words.

Table 1.

Rating parameters and relevant questions of questionnaire

Rating parameter	Question
Time	How has the new Referral System altered the time taken to complete IR referrals?
Ease of use	How has the new Referral System changed the ambiguity in the procedures requested?
	How has the new Referral System altered the collection of patient information prior to submitting the request?
Uncertainty of referral	Has the new referral system helped reduce the uncertainty of referring to IR?
Periprocedural support	Has the new referral system altered your ability to liaise with IR department for any requested input before or after the procedure?
User experience	Did you feel more supported through the new referral system?
	How likely is it that you will recommend the continuation of the new referral system?
	What is your overall satisfaction with this change?
	What would be one thing you would like to see improved in the IR referral system

IR, interventional radiology.

Data analysis

A t-stat was calculated for each question using the averages of all 98 responses, across all groups. Normally, t-stat would be calculated if response >1/5, which would identify anything above “not at all”, however, given the nature of the responses which allowed for positive bias we chose to set it at an elevated cut-off of >3.5/5. This would ensure that any degree of positive bias is well-accounted for.

As part of investigating possible differences in responses between grade and specialty, average scores and standard deviations for each question were calculated, giving a score between 1 and 5. A t-test was applied to obtain the significance of each of the questions with a null hypothesis of the mean not exceeding 3.5, and with a confidence interval set at 0.95. To identify possible discrepancies between subgroups, means across different grades and different specialties was assessed.

Results

Impact on referrers

Observed in Table 2 are the means, standard deviations as well as the calculated t-stat for each of the questions, using all 98 responses. As evidenced in Table 2, a statistical significance was observed in five out of the eight questions. The highest t-stats are observed in questions 7 and 8, indicating a very high degree of overall satisfaction with the new system, as well as support for its continuation.

Table 2.

Means, standard deviations and t-stats of questions

		Mean	Standard deviation	t-stat
	Time
(1)	How has the new Referral System altered the time taken to complete IR referrals?	3.8	1.15	2.55
	Easy to use
(2)	How has the new Referral System changed the ambiguity in the procedures requested?	3.94	1.2	3.62
(3)	How has the new Referral System altered the collection of patient information prior to submitting the request?	3.41	1.14	−0.8
	Appropriateness of referral
(4)	Has the new referral system helped reduce the uncertainty of referring to IR?	3.72	1.32	1.68
	Periprocedural support
(5)	Has the new referral system altered your ability to liaise with IR department for any requested input before or after the procedure?	2.87	1.32	−4.74
	Overall experience of the referrer
(6)	Did you feel more supported through the new referral system?	3.46	1.11	−0.36
(7)	How likely is it that you will recommend the continuation of the new referral system?	4.08	0.98	5.87
(8)	What is your overall satisfaction with this change?	3.99	0.92	5.24

IR, interventional radiology.

In question 7, “the likelihood of recommending continuing the new referral system”, 76 out of 98 respondents voted 4 or 5, indicating high preference for continuing the new system. Only six people voted below three in that category. For question 8, looking at overall satisfaction, results were similarly high, with 92 out of 98 responders voting three or higher, indicating most participants were at least moderately satisfied at the new referral system.

There was also a statistically significant improvement shown in questions 1, 2 and 4. These are relating to the time taken to complete an IR form, the ambiguity surrounding the referral and a reduction in the uncertainty of referring to IR, respectively. Areas which fared less well were the alteration of collection of patient information and the support felt by the person referring, with a t-stat slightly below 0, even though the means were still above 3.

Question 5 was the question which stands out as least impressive, sitting at an average of responses of 2.87 and a t-stat of −4.74 and relating to the ability to liaise with the IR department.

Impact on referrers by grade

As indicated in Figure 2, 66.3% of the participants were consultants, while the rest 33.3% comprised of junior doctors and clinical nurse specialists. Foundation doctors rated the new referral system highest across 7 of the 8 parameters (Figure 3). Comparatively, clinical nurse specialists rated the new referral system positively, yet with the lowest scores across 7 out of 8 parameters when compared to other grades. The only question which they did not rate lowest was regarding reduced uncertainty surrounding the new IR referral system, which consultants rated lowest. Ability to liaise with IR was the most lowly rated parameter across all grades, when compared to the other parameters.

Figure 2.

Breakdown of participants’ grades

Figure 3.

Parameters as rated by different grades. IR, interventional radiology.

Impact on referrers by specialty

In total, there were participants from 29 different specialties which, for the purpose of data analysis, were simplified into medical and surgical. Of the 35 surgical referrers, 19 were from vascular surgery, general surgery or urology. Additionally, there was 63 medical referrals of which 43 were comprised of medical oncology, paediatrics, gastroenterology and respiratory. Medical specialties rated the new referral system more favourably than surgical specialties across all aspects including time, ease of use, appropriateness of referral, periprocedural support and overall experience of referral (Figure 4). Despite the apparent discrepancy, both medical and surgical specialties indicated strong support for the continuation of the new IR referral system, with averages of 4.2 and 3.9 out of 5, respectively. Similar preference was observed in the overall satisfaction section, with averages of 4.1 and 3.8, respectively.

Figure 4.

Parameters as rated by surgical vs medical specialties. IR, interventional radiology.

Effects on department

In addition, this study sought to explore the effects, if any, of the new referral system and its facilitation by clinical nurse specialists on the IR department itself. Specifically, data pertaining to activity levels (quantified by absolute number of patients treated monthly, and over a period of 12 months) were analysed to compare the department’s performance before and after the implementation of the referral system. This revealed an increase of 11.2% in the number of patients, and as shown in Figure 5, the level of activity was consistently higher across all months with the exception of March 2020 which coincided with the COVID-19 pandemic.

Figure 5.

Activity levels of IR department in the 12-month interval before and after the implementation of the new referral system. IR, interventional radiology.

Discussion

The repercussions of converting a request-based system to one based on “referrals” have been profound. It represents a paradigm shift from “demanding” to “asking”, and one from merely performing tasks to holistically taking over patients. This shift encapsulates the true trajectory of IR, which is now in a better position to assess, weigh and communicate decisions regarding patient management to referrers in the same way that all other interventionalists or surgeons do.⁵ The impact of this change on the current department has been realised in many ways: increased caseload, higher departmental activity levels with reduced cancellations, better procurement and recognition from the referrers of the needs and considerations before a referral is made. Importantly, it has maximised vetting power and has allowed for better scrutiny and screening of referred patients. The latter has also highlighted the need for the radiologist to be the ultimate consent taker and patient optimiser, allowing enough time to discuss all options with candidate patients.

The impact on the referrers has been equally positive at first glance. Within the first 12 months of the roll-out, staff members from most grades and most specialties rate the change favourably, or at least in some ways better than previously. Perceived improvement in the process was observed with statistical significance in five out of eight questions, rating time, ease of use, appropriateness of referral and overall user experience as improved.

When looking at results by specialty, medics, as opposed to surgeons, seem to vote for the change more favourably. The observed difference can be due to a number of reasons. One possible reason is the fact that surgical specialties deal with patients who have already gone or are planning to go through additional interventions, before or after the proposed IR procedure. This implies that the referral is likely to be very specific and in fact, more likely to be vetted positively. Therefore, in the eyes of a surgeon, it may be that the new referral system increases the time needed to refer an otherwise “straightforward” referral.

Moreover, investigating inter group differences after stratifying responses by grade, highlighted subtle, yet interesting differences amongst juniors and seniors and between doctors and clinical nurse specialists. The majority of respondents being consultants, limits the reliability of observed differences, with in fact none reaching statistical significance. There is a trend, however, that suggests consultants generally reporting a relatively less positive change with regards to their ability to liaise with IR periprocedurally. Conversely, the latter was rated highly by foundation doctors. The apparent discrepancy may be attributed to a pager-based system upon which the IR CNS operates, and which therefore may be more accessible to junior doctors as opposed to consultants who spend a large part of their day in clinics or theatres. As the trust’s communication system is currently being upgraded to overcome such limitations, this seems to be a rather short-lived problem. Moreover, failing to capture a significant improvement in the ability to liaise with IR can also be a marker of satisfactory baseline reachability to begin with, prior to the implementation of the above changes. However, in the absence of pre- and post-implementation data, this is a difficult conclusion to make. While the absence of statistically significant data renders it difficult to conclude the true added benefit of IR CNS in improving liaising with other teams, it is important to emphasise that the new scheme is only at its infancy, of which the true effects have been further downplayed by the fragmentation of the service during the COVID pandemic.

This study is based on a relatively small sample, therefore, it would benefit from further cycles of auditing to ascertain the true and longlasting effect, if any, of the change. Nevertheless, it has highlighted important points about the nature of changes radiologists across the country should pursue in an effort to further ensconce the status of IR in hospital medicine. While this has been a success in a tertiary and financially sound centre, transferability of similar micropolicies across centres (e.g. to district hospitals) and regions remains the real challenge. Dissemination of such changes represents the single most important catalyst for a synchronised progression of IR as a clinical specialty across the world.

Conclusion

In an era where new subspecialties are emerging in part out of need to utilise the ever-expanding technological armamentarium, IR finds itself in the core of this wave. Although oftentimes categorically described as an unnecessarily overspecialised system, some specialties like IR do fulfil the rigorous criteria necessary to identify itself as a true clinical specialty. In shaping this new specialty, micropolicy changes within individual IR departments such as the replacement of outdated request-based systems to one which puts IR in control of vetting and patient flow is one of many changes contributing in entrenching the independence of this specialty.