Introduction
Surgical cancellations are a cause of inconvenience and patient dissatisfaction. Reference Koh, Phelan, Hopman and Engen1 Patients also report post-operative complications to a greater extent after a cancelled outpatient procedure. Reference Cho, Lee, Lee, Kim and Kim2 While standard of care is often considered unimpacted when rescheduling cancelled procedures, there is mounting evidence to support a negative association between patient’s physical well-being, psychological health and longer duration waiting for surgery. Reference Herrod, Adiamah and Boyd-Carson3,Reference Sommer, Jacobsohn and El-Gabalawy4 Neurosurgical patients are particularly vulnerable given the relatively invasive nature, high complexity, propensity for acute decline and some procedures being performed under awake conditions. This may result in a high degree of pre-operative anticipatory patient anxiety and a resultant reduction in postoperative HRQoL. Reference Herrod, Adiamah and Boyd-Carson3,Reference Sommer, Jacobsohn and El-Gabalawy4
Previously reported neurosurgical cancellation rates range between 4% and 20%. Reference Koh, Phelan, Hopman and Engen1,Reference Cho, Lee, Lee, Kim and Kim2,Reference Wahba, Cromwell, Hutchinson, Mathew and Phillips5 A survey study by Koh et al. indicated that 83.5% of cancellations were reported as due to administrative or facility reasons. Reference Koh, Phelan, Hopman and Engen1 These findings are similar to those reported in the broader surgical literature. Reference Koh, Phelan, Hopman and Engen1,Reference Oteri, Martinelli, Crivellaro and Gigli7–Reference Hussain and Khan16 Other factors relevant to procedural cancellation include pre-anaesthetic workup, comorbidity optimization, type of surgery and surgeon-specific factors. Reference Jiang and Carvalho17 Canada faces challenges with the burden of surgical cancellations due to the nature of a single-payer social health care system, particularly with regards to inefficient use of operating room time and limited opportunity to utilize time with an alternative case.
To date, no prior studies have examined the nature of neurosurgical procedural cancellations across Canadian institutions. As such, here we aimed to address this knowledge gap and identify any factors associated with procedure cancellations. Identifying causative factors may provide an opportunity for targeted quality improvement initiatives aimed at improving efficiency and optimizing the care of neurosurgical patients.
Methods
The Canadian Neurosurgery Research Collaborative (CNRC) Reference Maclean, Touchette and Duda18–Reference Dakson, Tso and Ahmed21 performed a cross sectional, multicentre retrospective cohort study of operative cancellations at five Canadian academic neurosurgical centres. Sites included were Dalhousie University, University of Alberta, Western University, University of Calgary and McMaster University. Research Ethics Board approval was waived by the lead site for being a quality improvement initiative. Health record departments were consulted at each site to compile neurosurgical operative cancellations from January 1, 2014 to December 31, 2018. Deidentified data collected pertaining to each cancelled case included patient age, sex, procedure name, reason for cancellation, admission status (outpatient or inpatient), emergency status (elective or emergency) and American Surgical Association (ASA) score.
Each case was reviewed by two team members (MM, ARP) and coded for neurosurgical subspecialty focus area and stratified into procedural categories: cranial, spine, peripheral nerve, endovascular, miscellaneous, or not provided. Miscellaneous procedures included those that did not specifically fit into one of the aforementioned categories. Finally, reasons for cancellation were reviewed and clustered according to the following groups: (1) patient-related; (2) staffing-related; (3) operational or resource-related factors; and (4) reason not provided or assigned. Patient-related factors included new acute illness, inclement weather precluding transportation to hospital, comorbidity status, expiry, patient not fasting or could not be contact on the day of surgery. Staffing-related factors included reasons for cancellation provided by the surgeon (i.e., surgeon overbooked slate, surgeon called to emergency) or reasons provided by anaesthesia (i.e., anaesthesia unavailable, further pre-operative investigations required, patient did not meet anaesthetic candidacy requirements). Operational and resource-related factors included bed availability constraints, cancellations due to higher acuity cases booked on the emergency slate, rooms running late, or other miscellaneous factors (i.e., patient surgery postponed due to unavailable equipment).
Five-year case cancellation rates were determined for two neurosurgical centres for which total annual case counts were obtainable. A χ 2 test was performed to assess for differences between groups. Historical data pertaining to the proportion of Canadian neurosurgical operative cases completed per anatomic region (i.e., cranial, spine, peripheral nerve, miscellaneous) was determined using data reported recently by the CNRC. Reference Wahba, Cromwell, Hutchinson, Mathew and Phillips5 In order to determine if cases for a given anatomic region were disproportionately cancelled, we compared the actual proportion of operative cancellations to the historical proportion of surgical cases in Canada by anatomic region. Given that case type and reason for case cancellation were reviewed and arbitrated by two independent team members, an interrater Cohen’s kappa statistic was calculated to evaluate degree of agreement.
Statistical analysis was completed using SPSS 21.0 (IBM, Chicago, IL) software. Parametric data were compared by one-way ANOVA. Non-parametric data were compared by Mann–Whitney U test for non-parametric quantitative data and by χ 2 test for qualitative data. Interrater reliability was expressed as Cohen’s kappa statistic.
Results
Across all five sites, 7,734 total neurosurgical cancellations were recorded. Cancellations were categorized by surgical centre, with 1,225 cancellations at Centre 1, 893 cancellations at Centre 2, 187 cancellations at Centre 3, 3,887 cancellations at Centre 4 and 1,542 cancellations at Centre 5. Total completed case counts were only available for centres 1 and 2. The five-year rate of neurosurgical cancellations differed between Centre 1 and 2 (Centre 1: 25.9%; Centre 2: 13.0%, p = 0.008; Table 1). Overall, the combined cancellation rate for the two centres was 18.2%.
Table 1. Cancellations, total cases and cancellation percentage at two Canadian centres. Yearly cancellation rates were compared by Mann–Whitney U test
Demographics, procedural data and reasons for cancellation are presented in Table 2. The mean age was 57.1 ± 17.2 years. Sixty-three per cent of cancellations were observed in patients between the age of 50–79 years. Cancelled patients were 44.9% female. Approximately, two-thirds of cancellations were outpatients (Table 2). Cancelled cases were 74.2% elective and 25.8% emergency. The mean ASA score for all study patients was 2.99 ± 0.79. Regarding type of procedure, spine cases were most often cancelled (47.7%), followed by general neurosurgery (18.7%), oncology (14.4%), functional (8.1%), vascular (6.5%), peripheral nerve (3.6%) and epilepsy (1.1%). Cohen’s kappa statistic for interrater reliability pertaining to arbitration of case type data was 0.998 ± 0.0001 for procedure type and 0.996 ± 0.001 for subspecialty focus. Reason for cancellation was most often due to surgeon-related factors (28.2%), followed by cancellation for a case of higher acuity (23.9%), patient condition (17.2%), other miscellaneous factors (17.0%), resource factors including lack of bed availability (7.0%), cancellation secondary to a prior case running late (6.4%) and anaesthesia factors (0.3%). When clustered, the reason for cancellation was patient-related in 17.2%, staff-related in 28.5% and operational or resource-related in 54.3% of cases (Fig. 1). Cohen’s kappa statistic was 1.00 ± 0.0001 for both reason for cancellation and clustered reason for cancellation.
Figure 1. Graphical representation of (a) surgical cancellations by case type, (b) surgical cancellations by subspecialty, (c) rationale for surgical cancellations and (d) high-level categories of reasons for cancellation.
Table 2. Demographic and procedural data pertaining to neurosurgical operative cancellations across five Canadian academic centres
* Resource availability: Examples include no post-operative inpatient bed (e.g., ward or intensive care unit) available or elective time no longer available.
A comparison of procedural cancellations across centres is presented in Table 3. Differences were identified between the five centres in mean patient age, age distribution stratified by decade, admission status (i.e., inpatient versus outpatient), case acuity (i.e., emergency versus elective), procedure type, subspecialty focus, reason for cancellation and ASA scores. There were more outpatient than inpatient cancellations at Centres 1, 3 and 4. There more spine cancellations than cranial cancellations at Centre 1, 4 and 5. The most common reasons for operative cancellations were related to patient condition at Centres 3 and 5, surgeon-related factors at Centre 4, higher case acuity taking priority at Centre 2 and miscellaneous (other) and Centre 1. Upon clustering, reasons for cancellation revealed operational and resource-related factors were the most common reason for cancellations at all sites except Centre 4, where staff-related factors were most common.
Table 3. Demographics and surgical data by centre. Parametric data are compared by one-way ANOVA, non-parametric data are compared by χ 2 test
* Resources: Examples include no post-operative inpatient bed (e.g., ward or intensive care unit) available or elective time no longer available.
Comparison of the proportion of operative cancellations to the historical proportion of neurosurgical cases completed in Canada per anatomic region, indicated a disproportionately high rate of spine cases (p < 0.00001; Table 4). Information pertaining to the number of times a procedure was cancelled and duration of surgical delay after cancellation was only available at select centres (Supplementary Table 1). At Centre 2, 15.7% of cancelled cases were cancelled at least once prior. The average delay to surgery following cancellation was 19 ± 26.1 days at Centre 3 and the average time from placement on the surgical waitlist to cancellation was 5 ± 7.7 months at Centre 4. Author’s did not receive funding from any sources to support this work.
Table 4. Statistical comparison of proportion of cancelled cases to historical case record. Statistical analysis performed by χ 2 test
Discussion
Here, we present the first multicentre review of neurosurgical operative cancellations in Canada. The current work provides several insights into the nature of neurosurgical cancellations.
The rate of neurosurgical operative cancellations reported in the literature widely ranges 4%–20%. Reference Koh, Phelan, Hopman and Engen1,Reference Cho, Lee, Lee, Kim and Kim2,Reference Wahba, Cromwell, Hutchinson, Mathew and Phillips5 Herein, the combined cancellation rate for Centres 1 and 2 was 18.2%. The only prior study including data on Canadian neurosurgical operative cancellations is from a medium-sized neurosurgical centre in Kingston, Ontario. Reference Koh, Phelan, Hopman and Engen1 They reported 265 procedural cancellations and 1,271 cases successfully completed over a four-year period, yielding a cancellation rate of 16.9%. A comparison of institutional data entry and coding methods would be helpful to assess for differences in the accuracy of recording and coding of operative cancellations. Disparity in data capture, entry and coding may at least partly explain the observed difference in cancellation rates. Centre 1 had a higher cancellation rate for spine and outpatient procedures. These findings are in keeping with the overall trend observed across all five participating Centres. It also aligns with the disproportionately higher than expected rate of spine procedure cancellations based on historical Canadian neurosurgical case frequencies. Reference Tso, Dakson and Ahmed22
For the combined cohort, patients most often cancelled were male, outpatients and on the elective surgical list. Patients in their fifth to eight decades of life comprised the majority of cancelled patients. This trend likely reflects the overall demographic composition of adult neurosurgical patients. Given that the reasons for operative cancellations may be disparate between centres, and in an effort to avoid masking differences in site-specific reasons underlying cancellations, we presented data stratified according by centre (Table 3). We also identified differences in age, admission status, acuity status, ASA grade, procedure type, subspecialty of operation and reason for cancellation between neurosurgical centres. The lack of demographic data on the overall patient populations undergoing electively booked neurosurgical procedures at these centres limits the ability to comprehensively evaluate case cancellation on a demographic basis and produce meaningful statistical comparisons, which should be the subject of future work.
In terms of subspecialty focus, the most commonly cancelled cases were spine (47.7%), followed by general neurosurgery (18.7%) cases. Oncology, vascular, functional, peripheral nerve and epilepsy cases were underrepresented in the cancellation figures. These findings are in keeping with prior literature suggesting that cancer operations are less frequently cancelled. Reference Wong, Harris and Moonesinghe10 This also likely reflects the urgency of oncologic procedures and provincial or health authority wait time benchmarks. We assessed for differences in the proportions of cancelled cases, compared to historical procedure rates (previously published by Tso et. al (2017)), Reference Tso, Dakson and Ahmed22 stratified by anatomical region (i.e., cranial, spine and peripheral nerve). The cancellation rate for spine procedures was disproportionately higher than expected based on historical case frequency. These data imply that patients undergoing spinal operations may be more at risk of cancellation than those undergoing other procedures.
The most frequent reasons for surgical cancellation in the study population were operational and resource-related, which specifically included post-operative bed availability (e.g., ward or high acuity), administrative reasons, more urgent case booked on emergency slate, surgeon’s operative slate running late and other related factors. This falls in line with the earlier work by Koh et al., Reference Koh, Phelan, Hopman and Engen1 who demonstrated that neurosurgical cancellations in their cohort were largely due to operational or resource-related factors (89.4%). Rates of operative cancellations for all surgical specialties in the literature range from 0.037% to 16.7%. Reference Maliha, Bruce and Anstadt8–Reference Jiang and Carvalho17 Specific surgeon practices, Reference Maliha, Bruce and Anstadt8 caregiver preference. Reference Maliha, Bruce and Anstadt8 insufficient pre-operative assessment, Reference Sato, Ida, Naito and Kawaguchi9,Reference Hussain and Khan16 type of surgery, Reference Wong, Harris and Moonesinghe10 wait times, Reference Da’Ar and Al-Mutairi11 patient no-show, Reference Abeeleh, Tareef and Hani12 operating room resources, Reference Abeeleh, Tareef and Hani12 operating room management, Reference Kaye, Mcdowell, Diaz, Buras, Young and Urman13 country of cancellation Reference Leslie, Beiko, van Vlymen and Siemens14 and lack of financial clearance Reference Leslie, Beiko, van Vlymen and Siemens14 have all been identified as barriers precluding planned surgical care.
Regional health care delivery method likely plays a role in operative cancellations. Neurosurgical care often takes place at tertiary or quaternary settings due to specialty-specific resource constraints, such as requirement for high acuity beds after certain procedures, neuroscience-trained nursing staff, multidisciplinary subspecialty support and a variety of other infrastructure factors. This adds a further layer of complexity for patients travelling from rural communities or those distant from neurosurgical centres, as last-minute cancellations affect this population disproportionately. Canada is a single-payer social health care system and financial clearance for surgery does not generally play a role, although as indicated above, travel costs for surgical intervention are covered by patients and their families.
From the perspective of countries serviced under other healthcare insurance climates, rates of cancellation in 2003–2004 at a hospital in the USA (16.5%) and in Norway (14.6%) Reference Leslie, Beiko, van Vlymen and Siemens14 were grossly similar to the current study, so it remains unclear if there is an effect imposed by the insurance system. Hospitals functioning in similar single-payer social healthcare systems, such as the United Kingdom, Italy, New Zealand, may see similar rates of cancellation as compared to countries such as the USA that largely function in a private health insurance ecosystem. Further evaluation of operative cancellations in other countries would assist in answering this question.
While our study spans years in the pre-COVID era, the COVID-19 pandemic has made surgical cancellations, surgical wait times and surgical backlog a pressing global issue. COVID-19 testing and cancellation algorithm vary by region and have had various effects on different types of surgery. 23–Reference Glasbey, Abbott and Ademuyiwa25 Further evaluation of surgical cancellations in the post-COVID era is warranted in order to develop strategies to mitigate cancellations and prevent worsening of surgical waitlists.
Limitations
A primary study limitation was the lack of total completed case counts for three of five participating sites, at least partially attributable to a complex data collection process employed in a national, multicentre, resident-led study. Furthermore, we did not have access to institutional data entry and coding methods for documenting neurosurgical operative cancellations. The decision to broadly categorize cancellations (e.g., cranial, spine, peripheral nerve) was also primarily based on lack of granularity in data coding at the respective participating centres (e.g., surgery for a lumbar spinal fusion for degenerative spondylolisthesis was not differentiated from neurological compromise from a lumbar spinal abscess). While there is a considerable difference in the overall number of cancelled cases at each centre, this largely reflects the size of the participating centres and variability in their annual case volume. Four of five participating sites in this study were small-medium size and therefore these results may not be generalizable to larger academic centres. For this reason, cancellation rate was provided only for two sites with total annual case booking data. Case coding across sites was not standardized as Canadian national standards for coding neurosurgical operative cancellations do not exist.
Conclusion
The current study represents the first Canadian multicentre evaluation of neurosurgical operative cancellations. The overall rate of neurosurgical cancellation was 18.2%. We identified factors related to neurosurgical cancellations; cases more often cancelled included, spine, elective and outpatients, respectively. The most common reasons for cancellation pertain to administrative and operational factors. In addition, points of variability between centres are highlighted that merit further investigation. Future data analysis of interprovincial administrative databases may provide the granularity required to describe and address such a multifactorial and complex challenge in health care delivery.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/cjn.2024.265.
Author contributions
MMM conceived and coordinated the study, performed data acquisition and revised the manuscript. ARP analysed the data, provided an initial draft of the manuscript and revised the manuscript. MAR revised the manuscript. NRP processed the data and helped with the drafting of the manuscript. DS performed literature review. JC, TD, MEE, SH, JJ, MMKS, ML, NS and ET acquired data and provided feedback on the drafted manuscript. SC oversaw the project and provided administrative support.
Funding statement
The authors did not receive or utilize funding to support completion of this work.
Competing interests
None.
Open access
