Referrers' point of view on the referral process to neurosurgery and opinions on neurosurgeons: a large-scale regional survey in the UK

Historical and future trends in emergency pituitary referrals: a machine learning analysis

PURPOSE

Acute pituitary referrals to neurosurgical services frequently necessitate emergency care. Yet, a detailed characterisation of pituitary emergency referral patterns, including how they may change prospectively is lacking. This study aims to evaluate historical and current pituitary referral patterns and utilise state-of-the-art machine learning tools to predict future service use.

METHODS

A data-driven analysis was performed using all available electronic neurosurgical referrals (2014-2021) to the busiest U.K. pituitary centre. Pituitary referrals were characterised and volumes were predicted using an auto-regressive moving average model with a preceding seasonal and trend decomposition using Loess step (STL-ARIMA), compared against a Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) algorithm, Prophet and two standard baseline forecasting models. Median absolute, and median percentage error scoring metrics with cross-validation were employed to evaluate algorithm performance.

RESULTS

462 of 36,224 emergency referrals were included (referring centres = 48; mean patient age = 56.7 years, female:male = 0.49:0.51). Emergency medicine and endocrinology accounted for the majority of referrals (67%). The most common presentations were headache (47%) and visual field deficits (32%). Lesions mainly comprised tumours or haemorrhage (85%) and involved the pituitary gland or fossa (70%). The STL-ARIMA pipeline outperformed CNN-LSTM, Prophet and baseline algorithms across scoring metrics, with standard accuracy being achieved for yearly predictions. Referral volumes significantly increased from the start of data collection with future projected increases (p < 0.001) and did not significantly reduce during the COVID-19 pandemic.

CONCLUSION

This work is the first to employ large-scale data and machine learning to describe and predict acute pituitary referral volumes, estimate future service demands, explore the impact of system stressors (e.g. COVID pandemic), and highlight areas for service improvement.

Use of an Online Referral Service for Acute Neurosurgical Referrals: An Institutional Experience

OBJECTIVE

We sought to review the current scope of emergency neurosurgical referrals and examine the long-term use of a web-based referral system.

METHODS

This was a single-center retrospective observational study. Referral information was collected retrospectively over a 1-year period after the implementation of a web-based referral system (June 2019-June 2020). Information such as demographics, clinical information, referrer details, and neurosurgical response times and outcomes were collected. Statistical analyses were performed using R Version 6.3.1.

RESULTS

Our unit received 5949 referrals with a median age of 63 years old (range: 0-107 years) (male = 50.3%). We observed an average of 16.3 referrals per day (range: 4-32), with Fridays having the highest average and the weekend days receiving statistically fewer referrals (P < 0.001). More than a third (35.9%) of referrals occurred within hours (8:00-17:00 Monday-Friday), with A + E making up approximately 50% of referrals. Common reasons for referral were traumatic brain injury, intracranial tumors, and degenerative spine. The median time from referral to first response was 32 minutes, occurring within an hour in 72.9% of cases. The median time to definite response was 83 minutes, occurring within 2 hours in 58.2% of cases. Factors found to impact the response time were referral emergency and time of day. Our acceptance rate over this period was 18.5%.

CONCLUSIONS

With an increasing number of referrals, it is feasible to provide traceable advice in a timely manner through an electronic web-based referral system transferable to any specialty. Insights could be used to direct resources and workforce planning according to emergency referral patterns.

A Novel Neurosurgery Referral Course: Feasibility, Validation, and Inferences for Patient Care

OBJECTIVE

The speciality of neurosurgery is under-represented in the majority of medical school curriculums, and those rotating within this specialty tend to be those with career aspirations within this field. Consequently, few emergency medicine trainees are exposed to this specialty. The aim of this educational project was to develop and validate a neurosurgery referral course for the target audience of emergency medics.

DESIGN

Development of a single day neurosurgery referral course, developed with accreditation from the Royal College of Surgeons England. The curriculum covered commonly referred pathologies. Content validity was assessed using a 5-point Likert Scale. Median Likert scores were compared to "indifferent" (3) (indifferent = 3 in this study Likert scale) using the Wilcoxon matched-pairs signed-rank test. Construct validity was assessed using a standardized pre and postcourse 10-single best answer exam and results compared using paired t tests.

SETTING

A pilot "Neurosurgery for Emergency Medics" referral course, hosted at a single UK based neurosurgery unit.

PARTICIPANTS

A cohort of 19 delegates, working in emergency departments various regions within the UK.

RESULTS

The subjective feedback showed significantly higher than the expected median Likert scale satisfaction scores (p = 0.0001). Construct validity was confirmed, with significant improvement in proportion of students getting the answers in the single best answer exam after the days training course (p = 0.017).

CONCLUSIONS

We demonstrate feasibility, content, and construct validity and conclude that this pilot "Neurosurgery for Emergency Medics" course was beneficial. Integration of this 1-day course into local doctor's induction programmes for emergency medicine and neurosurgery may advance both local and national standards for referrals and consults alike, with the ultimate goal of improving patient care.

Predictors and ocular outcomes of rescue treatment in preterm infants with treated retinopathy of prematurity-a retrospective study

AIM

To determine predictors of rescue treatment among infants treated for retinopathy of prematurity and to evaluate their ocular outcomes at 18-24 months of corrected age.

METHODS

This is a single centre retrospective study of infants who received treatment for type 1 ROP, using laser photocoagulation or anti VEGF agents. Multivariable logistic regression was used to generate a prediction model for rescue treatment of ROP. The primary outcome was an abnormal refractive outcome by 24 months of corrected age, among infants primarily treated with laser therapy.

RESULTS

Two hundred and eight infants (including 416 eyes) who received single (n = 151) or rescue (multiple) treatments (n = 57) were included. Ninety three percent of the infants were primarily treated with laser photocoagulation. Lower gestational age, small for gestational age, early packed red blood cell transfusion (within 2 weeks of postnatal age), and presence of Zone 1 retinopathy predicted the need for rescue treatment in treated infants [area under the receiver operating characteristic curve: 0.81 (0.73-0.89)]. The incidence of abnormal refractive outcome, assessed in a total of 174 infants, was found to be significantly higher in the rescue treatment group (67% versus 21%, adjusted odds ratio: 7.56 (3.3-17.2), P < 0.001). Myopia, very high myopia and use of spectacles was significantly higher in the rescue treatment group (P < 0.001 for each).

CONCLUSIONS

Rescue treatment for ROP was associated with an increased incidence of refractive errors and requirement of spectacles by 2 years of age. Larger prospective multicentre studies are required to confirm the findings from our study.

"The Brain Society": The First Two Years of an Undergraduate Neuroscience Society in Northern Ireland

The Queen's University Belfast Brain Society was set up in September 2018 to promote interest in the human brain. There were three main goals: firstly to provide opportunities for medical students to learn from neurologists and neurosurgeons outside their formal curriculum; secondly the Brain Society aimed to organise events that included students from other disciplines and to members of the general public who were interested in learning about aspects of neuroscience; thirdly to tackle neurophobia. In the last two years, there have been 14 events, ranging from formal lectures, to practical sessions and to patient-focused information evenings. We have sold over 1,600 tickets. This article covers how the Brain Society was set up, to inform students in other universities about the Belfast experience.

Development and evaluation of an electronic hospital referral system: a human factors approach

Coordinating care across hospitals has been identified as a patient safety risk as referrals are often paper-based and poorly documented. Electronic referral systems have the potential to improve the situation but can fail to gain uptake. We applied a human factors/ergonomics (HFE) approach to place analysis of local workflow and user engagement central to the development of a new regional electronic referral system. The intervention was evaluated with a before-and-after study. Referral quality improved, referrals containing sufficient clinical information for continuation of care increased from 36.9% to 83.5% and completeness of referral information significantly improved. There was a 35.7% reduction in the number of calls to the on-call specialist, and the mean period between admission and surgery for expedited transfers was reduced. Applying HFE informed design with use-based evidence; the system maintains sustained uptake three years after implementation. Reliable recording of information translates to better patient safety during inter-hospital transitions. Practitioners summary: This study developed, implemented and evaluated a clinical referral system using a human factors approach. Process analysis and usability studies were used to inform the application requirements and design. Region-wide implementation in hospitals resulted in the improved quality and completeness of clinical referral information and efficiencies in the referral process.

Completeness of obstetric referral letters/notes from subdistrict to district level in three rural districts in Greater Accra region of Ghana: an implementation research using mixed methods

OBJECTIVE

To assess the completeness of obstetric referral letters/notes at the district level of healthcare.

DESIGN

An implementation research within three districts in Greater Accra region using mixed methods. During baseline and intervention phases, referral processes for all obstetric referrals from lower level facilities seen at the district hospitals were documented including indications for referrals, availability and completeness of referral notes/forms. An assessment of before and after intervention availability and completeness of referral forms was carried out. Focus group discussions, non-participant observations and in-depth interviews with health workers and pregnant women were conducted for qualitative data.

SETTING

Three (3) districts in the Greater Accra region of Ghana.

PARTICIPANTS

Pregnant women referred from lower levels of care to and seen at the district hospital, health workers within the three districts and pregnant women attending antenatal clinic in the district and their family members or spouses.

INTERVENTION

An enhanced interfacility referral communication system consisting of training, provision of communication tools for facilities, formation of hospital referral teams and strengthening feedback mechanisms.

OUTCOME

Completeness of obstetric referral letters/notes.

RESULTS

Proportion of obstetric referrals with referral notes improved from 27.2% to 44.3% from the baseline to intervention period. Mean completeness (95% CI) of all forms was 71.3% (64.1% to 78.5%) for the study period, improving from 70.7% (60.4% to 80.9%) to 71.9% (61.1% to 82.7%) from baseline to intervention periods. Health workers reported they do not always provide referral notes and that most referral notes are not completely filled due to various reasons.

CONCLUSIONS

Most obstetric referrals did not have referral notes. The few notes provided were not completely filled. Interventions such as training of health workers, regular review of referral processes and use of electronic records can help improve both the provision of and completeness of the referral notes.