Clinical and Economic Impact of Implementing HealthPathways at a Musculoskeletal Radiology Department

J Clin Pathways. 2023;9(6):35-43. doi:10.25270/jcp.2023.11.02

The National Health Service (NHS) in the UK provides a comprehensive service, free at the point of delivery, to all individuals living in the UK. The provision of care is based solely on a patient’s clinical need irrespective of their demographics. The service is, however, committed to providing the best value for taxpayers’ money, adhering to the principles of prudent health care, and is accountable to the public, communities, and the patients it serves. Primary care providers (PCPs) are the gatekeepers to health care for nonemergency conditions, and they manage patients’ access to investigations and hospital specialists based on a patient’s clinical condition. To support this role, guidance and clinical pathways are available from national bodies such as the National Institute for Health and Care Excellence and the Medical Royal Colleges, but their interpretation and implementation is varied across the different health boards and regions within the UK. To ensure the application of the principle of prudent health care among PCPs, HealthPathways were introduced into our University Health Board and a review of the current clinical guidance was undertaken.

In 2019, more than 2000 lumbar spine and over 1200 knee MRI scans were being performed within the University Health Board following a scan request from PCPs. This represented over one-quarter (27%) of the entire MRI scanning capacity of the University Health Board. Over 1300 ultrasound scans of the shoulder were also undertaken, placing a significant burden on ultra-sonographer service.

Back then, there was no health board–specific guidance for PCPs on appropriate patient selection for these investigations. Although the radiology department undertook a certain degree of referral vetting, the process was inconsistent due to the lack of locally agreed guidance. As such, large numbers of patients were being referred and many investigations were undertaken for an undefined clinical benefit.

In 2017, the Health Board procured HealthPathways, a web-based health information system that centralizes all locally agreed assessment, management, and referral guidance for a wide range of clinical conditions that PCPs may encounter,¹ which went live in 2019. Initially developed in New Zealand by the Canterbury District Health Board in 2008, it has since been adopted by various health systems in New Zealand, Australia, and the UK.² The pathways are designed to assist with patient care in a primary care setting and are written in a collaborative approach involving general practitioners (GPs), hospital specialists, nurses, and allied health professionals. They are living documents that benefit from end-user feedback and can be rapidly redesigned based on changing clinical practice. The main benefit of using HealthPathways is in proving localized clinical guidance to primary care clinicians to support improvements in the quality of care they provide by reducing harm, waste, and variation.^3-5 This also has the effect of increasing clinical confidence in referral appropriateness and an overall improvement in service integration.^4,5

Since 2019, over 500 clinical HealthPathways have been written by local clinical teams, including pathways pertaining to knee, shoulder, and lumbar spine conditions.⁶ MRI scans are the investigation of choice for many lumbar spine and knee conditions.^7-9 An ultrasound scan of the shoulder is also considered by many to be the investigation of choice for many soft-tissue shoulder conditions.¹⁰ Access to these scans can be vital in making the correct diagnosis of clinical conditions. There are, however, economic constraints to all health systems, with these investigations representing a significant resource requirement.^11-13 Appropriate patient selection based on experienced clinical assessment, and auditing the benefits of such investigations, is required to ensure value-based health economics are maintained.

The focus of this study is to determine the impact of implementing HealthPathways on referral rates for MRI lumbar, spine, and knee scans and ultrasound shoulder scans, including the system’s ability to communicate new, locally determined clinical guidance and the subsequent use of that guidance.

First, we reviewed current referral practices after identifying the lack of local guidance for PCPs on clinical conditions affecting the knee, shoulder, and lumbar spine that require radiological referrals in the community.

To establish the appropriateness of MRI scan requests, we retrospectively reviewed 106 consecutive patients who underwent an MRI scan of the knee and 100 consecutive patients who underwent lumbar MRI scans following a referral from their GP in September 2019 by a consultant musculoskeletal (MSK) radiologist and GP; no formal review of ultrasound requests was undertaken at this phase. We examined the conclusions in the MRI reports, the management of the patients by PCPs, and the outcomes of those patients referred to secondary care services. The radiology department received over 150 requests for scans each month and was unable to cope with this level of demand. Anecdotal evidence suggested that few scans showed significant pathology.

The results of this retrospective review and the service pressures on the radiology department prompted the creation of clinical HealthPathways for knee, shoulder, and lumbar spinal conditions. These pathways altered the threshold of clinical symptoms needed before an MRI or ultrasound scan could be requested. The HealthPathways were written in a collaborative approach between clinicians from primary care, radiology, physiotherapy, orthopedics, and other engaged individuals, including input from the local medical committee. Following their creation, the HealthPathways were visible to all GPs on the Cardiff and Vale Community HealthPathways website, and all practices were informed of their availability via a quarterly update newsletter. GPs were also given the ability to contact the MSK radiologist on call each working day via a new dedicated Consultant Connect phone line to discuss cases and request scans.

Next, a review of the referral rates and number of investigations undertaken for MRI lumbar spine and knee scans and ultrasound shoulder scans were then tracked from January 2018 until September 2022. This data collection and analysis was descriptive in nature. We collected the data and analyzed it using spreadsheet-based software.

By the end of November 2020, new HealthPathways provided advice and guidance to PCPs on acute and chronic knee conditions. This guidance recommended the use of plain radiographs in the first instance for suspected knee pathology and increased the threshold of symptoms required before an MRI scan could be requested.

In October 2021, the “Acute Lumbar Back Pain in Adults” HealthPathway was published, which provided advice on the management of patients who failed primary care–based treatments. The advice was that MRI requests are not usually clinically indicated, although requests could be made following direct discussion with the MSK radiology via Consultant Connect or if a patient had red flag symptoms (defined in the HealthPathway) or “acute and continuous non-traumatic nerve pain that fails to settle with appropriate prescribing of neuropathic modulating agents for over 6 weeks.”¹⁴

By May 2021, a comprehensive set of HealthPathways providing guidance to PCPs on shoulder problems had been published. These encouraged community-based treatments and physiotherapy assessment rather than an ultrasound scan. Finally, referral rates to the associated orthopedic departments and physiotherapy were collected between January 2019 and October 2022 to assess any changes in demand following the introduction of the HealthPathways.

The cost of performing radiological investigations within the UK NHS is variable between health boards within Wales. The National Payment System, within the UK NHS, provides a standardized cost for each investigation and was used to determine the financial savings described.15

Knee

The outcomes of 106 consecutive patients who underwent an MRI scan of the knee following a request from a GP during September 2019 were analyzed. Of the 106 patients, 63% were referred to secondary care mostly at the same time the scan was requested, independent of the MRI results. Of the 37% patients not referred to secondary care, the MRI was normal or essentially normal for the patient’s age group or did not provide any additional information over the plain x-ray result in 85% of cases (Figure 1). A total of 6% of the patients had a clinical abnormality on the MRI scan but were managed within a primary care setting.

We analyzed the MRI request rate and number of scans undertaken from January 2018 until September 2022 (Figure 2 and Figure 3), noting that the introduction of knee HealthPathways occurred in November 2020. From March 2020 through December 2020, there was a marked reduction in scans requested and undertaken due to COVID-19. More relevantly, the data shows how, following the introduction of the HealthPathways, the request rates never returned to prepandemic levels and were reduced by 56%. Figure 3 shows that while 47 scans on average were requested each month since April 2021, only an average of 19 scans were undertaken. The others were returned to the GP for failure to follow the guidance set out in the HealthPathway. Overall, the number of MRI scans undertaken annually has dropped by 82%. The financial impact of these changes is a theoretical annual saving of £121 104 ($150000). The freeing up of MRI capacity, especially in a post-COVID recovery era, has been invaluable.

There has been no increase in the referral rates since the introduction of the HealthPathway to either the knee surgery or physiotherapy departments (Figure 4).

Lumbar Spine

The 100 consecutive lumbar spine MRI requests from GPs from September 2019 were reviewed together with GP referral data, and the outcomes are summarized in Table 1. Of the patients who underwent an MRI scan, 70% were not referred to secondary care. Of that group, 76% of the scans showed no abnormality or showed degenerative changes consistent with age and no neural compromise. Of the remainder, 20% had some degree of neural impingement but were not referred. In two cases, scans were undertaken for other reasons: an acute fall leading to a diagnosis of a stable L1 fracture and a post operative scan that required no referral.

Of the 30% of patients scanned and referred to secondary care, 50% of the scans showed no abnormality or showed degenerative changes consistent with age; despite this, the patients were still referred. Of the 15 patients who were referred with suggestions of nerve root impingement, the majority were referred to physiotherapy (81%) and no patients underwent a surgical procedure.

It appears that MRI scanning did not alter the referral patterns of GPs. The number of patients referred to secondary care with or without a degree of nerve impingement on the MRI was the same. This indicates that the decision to refer to secondary care was undertaken on clinical rather than radiological grounds.

We also analyzed MRI lumbar spine referral rates and the number of scans undertaken from January 2018 until September 2022 (Figure 5 and Figure 6), again noting the introduction of lumbar spinal HealthPathways in October 2021.

Once again there was a rapid reduction in MRI requests from GPs and scans undertaken in secondary care from March 2020 until the end of 2020 due to COVID-19. The results also mirrored the outcomes seen in the MRI knee referral data showing that, following the introduction of the HealthPathways, the referral numbers never returned to prepandemic levels. There was a 56% reduction in scan request rates and of these, 52% were returned to primary care as they failed to follow the HealthPathway. Overall, the number of MRI lumbar spine scans undertaken annually has dropped by 70%.

There has been a very small increase in the referral rates to spinal consultants since the introduction of the HealthPathways (Figure 7), but referral rates to physiotherapy have not been impacted. We also looked at the average referral rates for MRI lumbar spine from the spinal physiotherapy service and orthopedic consultants pre- and post-pathway introduction to ensure that there was not simply a shift in the point of referral. There was a very small increase in referral requests from the spinal physiotherapist (Table 2 and Figure 8) but a marked reduction in consultant referrals, resulting in an overall reduction in MRI referrals of 60%. The financial impact of just the reduction of referrals from primary care is an annual theoretical saving of £196 824 ($250 000).

Shoulders

For shoulder MRIs, we noted a similar, if not more exaggerated, reduction in referral and scanning rates following the introduction of the HealthPathways in April 2021, with reductions of 82% and 92%, respectively (Figure 9 and Figure 10).

The advice published in the HealthPathways has also led to a drop in the number of patients being referred to shoulder surgeons (Figure 11). The exact number of patients referred to shoulder physiotherapy is not available, because the department does not tabulate referrals by this joint specifically, but there has not been an associated marked increase in overall referral rates to MSK physiotherapy. The financial impact of these changes is an annual theoretical saving of £56 496 ($70 000).

The introduction of locally determined and collaboratively written referral criteria ensures that investigations and referrals are undertaken based on best clinical practice and local resources. In the new HealthPathways created for knee, shoulder, and lumbar spinal conditions, scans are now only recommended when the referring clinician has significant concerns and there are positive clinical findings. At present, the use of HealthPathways in the community is not consistent or universal, and PCPs who fail to follow the pathways find their scan referrals being declined. This is not unexpected given the recent and ongoing pressures on primary care and the time for any new innovations to be accepted and embedded into practice.

There has been a substantial reduction in the number of scans undertaken. The financial saving equates to more than £370 000 ($470 000) per year, just from the reduction in referrals from primary care. Reductions in MRI lumbar spine referrals were also reported from the orthopedic consultants, and overall, the more appropriate utilization of scans has reduced waiting times for patients with other medical conditions, as the number of scans undertaken has reduced without a change in overall capacity.

The economic impact we detailed that followed the introduction of HealthPathways has been mirrored in other regions that have introduced them,^16-18 with overall health care savings. Other studies have exalted the benefits of HealthPathways on the quality of referrals,^19,20 patient care,^4,21 and the experience patients have as they journey through a health system.^22,23

However, HealthPathways are a recent innovation, especially in the UK, and so our study faced several limitations. First, their utilization is not consistent or universal within primary care, and the criteria for accepting referrals within radiology does not always mirror the advice given in the HealthPathways. Second, the introduction of these HealthPathways came during the COVID-19 pandemic, which placed significant challenges on health care across the system and has made it more challenging to effectively analyze their impact on referral rates to secondary care departments. There are also limitations in the cost-effectiveness analysis as to whether HealthPathways have produced true financial savings. To achieve a high-quality cost effectiveness analysis would require patient-level data in both primary and secondary care. Such an analysis is currently beyond the abilities of our team and would require appropriate data linkages with administrative willingness, policy changes, and potential changes to privacy laws. Finally, it is possible that some patients may have experienced delayed care due to not receiving a scan or receiving it late; referral patterns, however, suggest this was likely a rare event.

Overall, the introduction of HealthPathways to provide guidance and support to primary care clinicians appears to be an effective way of improving the clinical management of pa-tients with access to radiological investigations based on clinical need, and HealthPathways have not had a negative impact on other secondary care services. The resulting reduced radio-logical demand, with a consistent capacity, has increased the availability of scans for patients with both MSK and other clinical conditions who require an investigation, which hasbeen invaluable in a post-COVID recovery era. There is no evidence that the reduced access for MRI knee, MRI lumbar spine, and ultrasound shoulder scans has been detrimental to patients, although this study focuses on the financial impact of HealthPathways, with HealthPathway advice and Consultant Connect access facilitating ongoing primary care radiology access if a clinician has significant concerns or clinical findings.

Authors: Simon R. Davies, MBBCh, MD; Kathleen Lyons, MBBCh; Kausik Mukherjee, MBBS; Ramakrishna Kishore, MBBS; Khurram Hashmi, MBBCh; Maria Dyban, LLM; Anna Kuczynska, MBBCh

Affiliations: Surgical Clinical Board, Cardiff and Vale University Health Board, Cardiff, UK; Musculoskeletal Radiology Department, Cardiff and Vale University Health Board, Cardiff, UK; HealthPathways writing team,Cardiff and Vale University Health Board, Cardiff, UK; Primary, Community and Intermediate Care Clinical Board, Cardiff and Vale University Health Board, Cardiff, UK.

Address Correspondence to: 

Simon Davies

32 Tair Onen

Cowbridge

CF71 7UA

Wales, UK

Email: simondavies@doctors.org.uk

Acknowledgments: The authors would like to thank Richard Evans, Iqroop Chopra, and Rhys Williams, who were the subject matter experts within the orthopedics department and helped write the HealthPathways; Rob Letchford, JoHutchings, and George Oliver, from the physiotherapy department, and Kevin Thomas, from the Local Medical Committee, for their support and work in developing and socializing these HealthPathways; and the Bevan Commission for their support and advice.

Disclosures: The authors reported no relevant financial or other conflicts of interest.