A supported primary health pathway for mild traumatic brain injury quality improvement report

Introduction

Nearly a third of adults will have experienced at least one mild traumatic brain injury (mTBI, including concussion) before the age of 25.¹ Evidence now shows that without treatment, between a third to a half of people experience persistent symptoms^2–4 that can affect their ability to perform at work/school and function in everyday life.^2,5 One of the biggest challenges experienced by patients and their whānau is that the health system is difficult to navigate.⁶ In New Zealand (NZ), a key barrier to health care access is the assumption that a person and/or their whānau have the knowledge and resources to seek help for their injury if they do not recover naturally.

A challenge for health professionals in primary care is confidence in how to support a person’s recovery.^7,8 This is particularly the case as 40% of GPs see <5 mTBIs per year.⁷ International mTBI guidelines recommend that patients should be followed up routinely by a medical professional to check if the patient has recovered or not.⁹ However, in NZ, Accident Compensation Corporation (ACC) data reveals that only one-third of mTBI claims between 2016 and 2018 received a follow-up by any health professional.¹⁰ If people do not recover naturally, referral to interdisciplinary rehabilitation services (often known as concussion services) is recommended.⁹ Access to interdisciplinary rehabilitation services can improve symptoms, functioning, mood and community participation.¹¹ Studies in both NZ and overseas have shown that delayed presentation to a concussion service is linked to higher symptoms on admission.¹² Additionally, early advice and management (within 7 days of injury) is linked to quicker recovery¹³ and lower health care costs.¹⁴

However, in NZ, there can be long delays (often several months) to access a concussion service for those experiencing a prolonged recovery.¹⁰ This is likely due to patient delays in seeking medical care after injury¹⁴ and uncertainty among general practitioners (GPs) about who, when, where and how to refer.⁷ Consequently, there is a need to identify people at risk of a prolonged recovery and implement a change to clinical practice to facilitate earlier access to health care for those who need it. The aims of this quality improvement study were to increase the proportion of patients receiving a follow-up and to decrease the mean duration in days to a concussion service for those who need it.

Methods

Overview of quality improvement methodology

The Plan, Do, Study, Act (PDSA) as a stand-alone quality improvement method was utilised to test the feasibility and evaluate the implementation of the supported health pathway on health system access after mTBI. To address the identified issues within the current health pathway for mTBI, the supported primary care pathway for mTBI included a range of components outlined in Box 1. Further details of the BIST can be seen in Supplementary Information 1.

Box 1.Components of the supported primary care pathway for mTBI.

Integration of a digital pathway support tool into the MedTech patient management software system (Versions 32 and Evolution), with capability of linking follow-up assessments to track recovery over time. Integration of the Brain Injury Screening Tool (BIST) to facilitate assessment and identification of patients at risk of prolonged recovery.15 Funding for up to three 15–30-min consultations per patient within primary care. Development of two online training modules (30-min duration) to upskill primary care practitioners. Development of an updated patient recovery advice sheet that was made available in electronic and paper form (https://www.acc.co.nz/assets/im-injured/acc8319-concussion-education-sheet.pdf).

A collaborative approach between the Accident Compensation Corporation (ACC), Primary health organisations (PHOs: ProCare, Pegasus, Well South and Nelson Bays), Manage my Health (secure health portal provider) and academics was undertaken. The digital health pathway was developed, tested and refined by Manage my Health in collaboration with the project team. ACC and academic team members contributed to the development of the education and training resources. PHOs were responsible for recruiting practices and clinicians, reviewing all referrals, providing peer support, relaying feedback from clinicians to the project team and administrative help eg application of funding reimbursement codes.

Once a practice had signed up, GPs, nurse practitioners and nurses from that practice could register to be part of the quality improvement project. Participating clinicians had to complete at least one online training module to be given access to the digital supported health care pathway tool for mTBI. Participating practitioners were asked to use the tool to guide their assessment of children (aged ≥8 years) and adults who presented with a suspected mTBI (including concussion). mTBI was defined as being an injury to the brain resulting from an external force resulting in disruption of brain function and alteration in mental state (such as feeling dazed and confused or loss of consciousness).¹⁶ While the pathway could be used for all patients with a suspected mTBI to support clinical assessment and diagnosis, eligibility for follow-up was based on a clinical diagnosis of mTBI or concussion. Delivery of the supported health pathway is outlined in Fig. 1.

Fig. 1.

Overview of the ACC supported primary health care pathway improvement strategy for mTBI.

On completion of the pathway assessment, if there was a high symptom burden and/or presence of another risk factor (eg previous prolonged recovery), the digital pathway generated a recommendation for the clinician to consider a referral to a concussion service. If no risk factors were present, the tool generated a recommendation to arrange primary care follow-up in 7–10 days to determine how they were recovering. Indicators of high-risk patients who may need a CT scan generated a recommendation for the clinician to consider a referral to an emergency department (ED). Referrals to concussion services could still be made at follow-up if the person was not recovering as expected. Clinicians were informed that the tool was only a guide and that triaging decisions ultimately remained their responsibility. An open text box was made available in the pathway tool to enable clinicians to outline their reasons if their action taken was different to the tool recommendation to aid identification of potential tool modifications.

Quantitative data was collected continuously over the course of the project, alongside feedback received from clinicians via the PHOs. This enabled a continuous and responsive test and learn approach. In line with the PDSA method, two PHOs were initially identified and recruited by ACC (ProCare and Pegasus) to ensure inclusion of clinicians working across the north and south islands of NZ. Initial learnings from the first PDSA cycle revealed that it was difficult to initially enrol practices due to high demands placed on primary care at the time. However, once enrolled there was positive feedback from clinicians on usability and utility of the tool. Refinements made to the tool following clinician feedback included the need for flexibility in the timing of appointments (eg on some occasions a 30-min second consultation was needed to complete the initial assessment, particularly in cases of multiple injuries or complex presentations). The wording of the ED referral recommendation was reframed to reflect cases where an ED referral was not appropriate eg person presents several months post-injury. Further clarification for practices on claiming for funding of the appointments was also provided.

Following these changes, the project was extended to include other PHOs servicing other areas of the country, including rural and high tourist areas (WellSouth and Nelson Bays PHOs). By the end of the second PDSA cycle, 10 PHOs were involved, encompassing 56.8% of the NZ population. The PHO areas were broadly representative of the national context as shown in Supplementary material S2, based on age, sex, ethnicity and deprivation extracted from the Ministry of Health.

Following the second PDSA cycle, further refinements were identified including the need to separate out triggers for the ED and concussion service. It became evident that in a few cases, early referral to a concussion service was being recommended unnecessarily eg patients presenting with a history of mental health but who had low acute symptom burden. On review of these cases, it became evident that this was a result of each risk factor being weighted equally. To improve the specificity of referrals, weightings were assigned to individual risk factors following a conjoint analysis process. Concerns were raised about the inability to provide appropriate follow-up appointments after the initial assessment due to service demand constraints. User feedback also favoured access to the pathway tool with as few ‘clicks’ as possible, such as access through a desktop or toolbar icon.

Results

This manuscript presents data extracted as part of the supported primary care pathway for mTBI between 18 May 2022 and 30 June 2023. Reasons for non-participation included: (1) not having the capacity to provide follow-up appointments; (2) using a different patient management software system to MedTech; and/or (3) being too busy to take something new on. Further issues raised by practices within high tourist areas was that the reimbursement payments from ACC did not cover the additional expense they incur in providing services to non-registered patients. Links with HealthPathways (HealthPathways provides clinicians with an online place where locally agreed information on management of a wide range of medical conditions and injuries is stored to support clinicians at the point of care) were needed to clarify referral processes and ensure consistency.

The 31 practices represented 99 practitioners (GPs and nurse practitioners). The supported health pathway was implemented with 220 patients (Table 1). The sample was 51.4% female, with a mean age of 31.5 (s.d. 17.7) years. The most common injury code used was S60 Concussion. There were 17 (7.7%) patients who did not have a concussion code. Median days from accident to GP appointment was 8 days (interquartile range of 15) with a wide range between 0 and 384 days. Sixty-seven (30.4%) patients attended their GP >14 days after their injury.

Discussion

This quality improvement project increased the proportion of people with mTBI receiving a follow-up by any health professional by 20.8% and reduced time to concussion services by 18.2 days compared to usual practice. The findings provide preliminary evidence that a supported health pathway is feasible to implement within primary care and may improve service accessibility post-mTBI for those at risk of prolonged recovery post-mTBI.

This quality improvement project aimed to close mTBI evidence-practice gaps by supporting clinicians to identify those at risk of prolonged recovery, proactively monitoring recovery and facilitating appropriate early referral where needed. As the supported health pathway contained multiple components, it is difficult to determine if any specific components made the biggest contribution to health system efficiency. However, our consultation with stakeholders across the sector and people with personal experience of mTBI to co-create this pathway suggests that without each component, the supported health care pathway would not have addressed need. This is aligned with evidence suggesting that successful change in primary care is often multifactorial in nature.¹⁷ Further work is warranted to evaluate the impact of the supported health care pathway for mTBI in other health care settings such as in emergency departments to facilitate consistency of care. Research studies would also be beneficial to determine if the supported health pathway for mTBI effectively improves patient recovery.

There were 9% of patients who did not have a brain injury code documented in their medical notes. These cases included high risk injuries (eg high speed car crash, traumatic incidents such as assaults and high symptom burden) suggesting that the experience of multiple injuries may confound coding accuracy. This finding supports evidence from the emergency department that mTBIs are often incorrectly coded or are missed.¹⁸ Accuracy of coding is important to prevent delays in access to services post-mTBI. While coding of mTBI was not a specific aim of the project, this was an unanticipated learning that suggests that additional guidance may need to be integrated into the supported health pathway to facilitate accurate coding in the future.

While there was an improvement in follow-up rates and a reduction in time to concussion services, there remained a proportion of people (43%) who did not receive a follow-up. The reasons why follow-up did not occur is not clear. Previous research has shown that about half of people recover naturally in the days to weeks following mTBI.^2–4 It may also be the case that receiving up to date evidence-based medical advice on first medical presentation may have reduced the need for subsequent follow-up appointments.¹⁹ However, it is also likely that current pressures on primary care in NZ may have prevented some patients from being able to book in for a follow-up appointment. Further work is needed to determine if other health professionals such as physiotherapists could assist in providing follow-up appointments to increase accessibility for patients.

More than half of the sample in this quality improvement project was female. Males have been found to have an increased risk of mTBI.²⁰ Evidence has also shown that males are less likely to seek medical help than females.²¹ It is not clear if males are attending other services such as physiotherapy or the emergency department or if they are simply not seeking health care. It was interesting that the median time from injury to first primary care appointment was 8 days, with >30% of people presenting to a primary care practitioner >2 weeks after their injury. This suggests that people may be delaying seeking health care until their recovery has not progressed as they would have expected it to.

A key limitation of this project is that only national pre-implementation data were available. It was not feasible to get pre-implementation data specific to the 31 practices due to gradual expansion of the project over time and need to combine data from different ACC databases. Additionally, not all clinicians within a practice engaged in the project. The only locality specific issue identified was the failure of universal reimbursement to cover the expense incurred by rural practices in providing services to non-registered patients. More widely, many practices were not able to participate in the project as they did not use the MedTech patient management software system. This project is now being extended to determine impact of the refined supported health pathway on clinician confidence, time in the pathway and health care costs, and to test if the risk factor weightings integrated into the BIST tool improve specificity of recommendations.

Conclusions

The findings provide preliminary evidence of improved follow-up rates and reduced time to concussion services following implementation of a supported health pathway for mTBI. Use of a quality improvement method enabled continuous testing and refinement of the supported health pathway. Key learnings from this quality improvement project included the need to take into account time between injury and appointment, separation of ED and concussion service criteria and the need to weight the risk factors of prolonged recovery to improve specificity of recommendations. There is also a need to make the tool accessible to other patient management software systems to facilitate implementation.

Supplementary material

Supplementary material is available online.

Data availability

Deidentified ACC data is accessible for use via data.govt.nz and Stats NZ.

Conflicts of interest

A. T. was part of the BIST development working group. A. M. was employed by Manage my Health during the course of the quality improvement project. The other authors have no competing interests to declare.

Declaration of funding

This quality improvement project was funded by the Accident Compensation Corporation. A. T. was supported by a Rutherford Discovery Fellowship administered by The Royal Society – Te Aparangi.

Author contributions

A. T., A. S., J. C., N. H., P. R., L. S., R. B. contributed to the design of the quality improvement project. A. T. conducted the analysis and drafted the manuscript. All authors contributed to conduct of the study, attending monthly governance group meetings, and revision of the manuscript.

Consent for publication

Not applicable.