The number of operations in the United States is increasing1,2 with 13,668,000 hospitalizations involving procedures in 2006.3 Patients have higher comorbidity indices,1 and an increasing number of procedures are performed during terminal admissions4 with nearly one-third of elderly Americans undergoing a surgical procedure during their last year of life.5 Concurrent with these changes, providing patient-centered care is a critical part of the healthcare reform agenda.6 Patient-centered surgical care necessitates not only that the clinical indications for surgery are correct, but also that the expected outcomes are concordant with the patient’s goals and priorities for medical treatment and that the patient can withstand the overall burdens of recovery.
The preoperative clinic often serves as a pause point for comprehensive assessment and medical optimization. A patient-centered assessment must include not only application of risk minimization strategies, but discussion with an informed patient regarding the value of the procedure in the setting of individual treatment goals and priorities; in some cases, the right decision may be to forego surgery.7 It has been well established that cancellation on the day of surgery has a negative impact on costs and efficiency. In addition, patients have mentally and logistically prepared themselves and their families for their procedures. Preoperative clinics have been shown to be associated with decreased delays and cancellations on the day of surgery.8,9 Case cancellations surrounding the day of surgery have been well studied and provide a wealth of insightful information surrounding efficiency, economics, and practice management.10–14 Patients cancelled because of reasons associated with assessment in the clinic may go on to have an operation at a later date; some may end up not having operative intervention at all.
Factors resulting in cancellations in a patient-centered framework are poorly understood. Such an exploration can provide a better understanding of the reasons scheduled cases are cancelled at the time of preoperative assessment as well as a perspective on factors that differ from those cases that proceed without cancellation. We describe our process in the preoperative clinic at Brigham and Women’s Hospital and include quantitative and qualitative factors associated with cancellations of scheduled surgical procedures at the time of preoperative assessment in this anesthesiologist-led preoperative clinic of a tertiary care academic medical center. We describe our overall framework that highlights the clinical and nonclinical issues that impact the decision for a scheduled surgical case to not proceed as scheduled during a specific period in the surgical episode (namely, the time between a preanesthesia clinic assessment and the day before a surgical procedure).
Institutional review board approval was obtained from the review board of the hospital (protocol no. 2011P002233; Partners Human Research Committee, Brigham and Women’s Hospital institutional review board), and the requirement for written informed consent was waived by the institutional review board.
We conducted a mixed-methods observational review15 of elective surgical cases done at a tertiary care center over a 10-month period (July 1, 2011, to April 30, 2012). We included patients having operating room procedures from the following specialties: general surgery, gastrointestinal surgery, surgical oncology, gynecology, neurosurgery, head and neck surgery, orthopedic surgery, plastic surgery, cardiac surgery, thoracic surgery, transplant surgery, urologic surgery, vascular surgery, and anesthesiology (i.e., chronic pain procedures performed in the operating room by an anesthesiologist). All patients scheduled into 1 of the 2 booking categories were included in the analysis: (1) “same-day admission” (patient scheduled to come to the hospital on the day of his or her operation with the plan to be admitted postoperatively); and (2) “day surgery” (patient scheduled for an operative procedure with a plan to be discharged on the same day). By the hospital’s policy, all such patients were required to have a preoperative assessment by the institution’s anesthesiologist-led preoperative evaluation center (Weiner Center for Preoperative Evaluation) within 30 days before the procedure. At the time of this preoperative assessment, the surgical visit and case booking had already been completed.
For purposes of this study and describing how our preoperative clinic functions, a “cancelled case” was defined as a scheduled operative case where the decision was made not to proceed with the operation on the scheduled date after the patient’s preoperative clinic appointment and before the scheduled date of the operation. We did not include cases that were cancelled on the day of the procedure or cases that were cancelled for reasons other than those associated with preoperative evaluation. When a patient had an issue identified in the preoperative anesthesia clinic that potentially precluded the case from proceeding as scheduled, a “hold-sheet” was placed on top of the patient’s paper medical record (“chart”) for the operation and the chart was kept in a specified location for further evaluation. If the issues on a patient’s hold-sheet were not resolved by the date of the operation and a final decision was made not to proceed with the case as scheduled, the case was included in the study. If the patient was seen in the preoperative clinic and had no issues identified during the preoperative assessment, the completed paper chart was sent to a file room holding area for the day of surgery.
The following demographic variables were obtained from the electronic record for the study population: age, sex, race, American Society of Anesthesiologists (ASA) physical status, body mass index (BMI), and the surgical service for which the patient was having the procedure. For all cancelled cases (and in instances in which an electronic anesthesia record was unavailable for a noncancelled case), the information was obtained by physician review of the patient’s medical record.
Continuous variables (age, BMI) for the cancelled versus noncancelled cases were compared using robust generalized estimating equation t tests (i.e., not assuming normality or constant variance) with clustering at the patient level (to adjust for the fact that a given patient may have had >1 operation over the course of the study period).16 Binary and categorical variables (sex, race, ASA physical status, surgical service) were similarly compared using Rao-Scott χ2 tests with clustering at the patient level.17 As a secondary analysis, we also compared the previously noted variables (age, BMI, sex, race, ASA physical status, surgical service) with clustering by month, as has been similarly done in previous studies on case cancellation.18 As a post hoc analysis, we compared cancellation rates by sex (male versus female) and by a binary version of ASA physical status (ASA physical status of I or II versus ASA physical status of III or greater) by applying a Student 2-sample t test to the transformation of the numbers of cases and cancelled cases from each 1-month period, as described in the Methods of the study by Dexter et al.18 (Appendix 1; our study was not powered nor designed to compare each individual category of race or procedure type using this method). The overall cancellation rate and the 95% confidence interval (CI) for this overall rate were also computed using this method. With an α of 0.025 (as a Bonferroni correction for 2 power calculations), we estimated that we were over 80% powered to detect a difference of 4 years (or more) in patient age and a difference of 15% (or more) in the proportion of cases that had an ASA physical status of III or IV when comparing cancelled cases with those that proceeded as scheduled. All P values were 2-sided, and P values <0.05 were considered statistically significant. All statistical analyses were performed using SAS 9.3 or SAS 9.4 (SAS Institute, Cary, NC).
In-Depth Narrative Review of Cancelled Cases
In addition to the previously noted demographic variables, we performed an in-depth medical record review of the narratives surrounding each cancellation, including an assessment of whether the procedure was ultimately performed (or an alternative option pursued) for a 1-year period after the initially scheduled operation. The physician review of cancelled cases consisted of collection of procedure information, reasons for cancellation, and management of the initial diagnosis from the time of the preoperative evaluation to 1 year after the initially scheduled operative date. Procedure information included collection of the overall indication for the procedure (categorized as: “therapeutic,” “diagnostic,” or “palliative”). A given procedure was defined as: therapeutic if intended to cure disease with or without diagnostic components, diagnostic if solely intended to assist with the diagnosis of disease, and palliative if done for symptom relief without intent to cure. A similar classification scheme has been reported elsewhere.19 The cardiovascular risk of procedure was also collected and was grouped according to low and elevated risk per American College of Cardiology/American Heart Association guidelines.20
To better understand the nuances that led to cancellation, a short vignette for each cancelled case was compiled by the physician reviewers. Reasons for cancellation were grouped into 8 primary categories to allow for a descriptive analysis. These categories were (1) uncontrolled or severe medical comorbidities that necessitated further evaluation or management; (2) patient decision to no longer have the procedure after reassessing the clinical risks versus the potential benefits; (3) changes that took place during the preoperative period rendered the procedure no longer indicated or outweighed by potential risks; (4) social issues such as lack of transportation or insurance denial; (5) surgical intervention deemed inappropriate based on the patient’s social habits or medication noncompliance; (6) clerical or scheduling issue in the preoperative clinic or the operating room; (7) a change in the surgical indications during the preoperative period necessitated a different procedure; and (8) the patient did not show up for the preoperative evaluation. Representative vignettes were chosen to illustrate each category of cancellation. Where applicable, the cancellations because of patient comorbidities were grouped according to organ system (e.g., cardiac, pulmonary) and representative narrative vignettes were chosen.
Follow-up management of the cancelled cases for 1 year after the initially scheduled operative date was grouped into the following categories: (1) completion of the original procedure at a later date; (2) completion of a different procedure to treat the operative indication; (3) ongoing medical issues that preclude having the procedure; (4) nonoperative management; (5) no treatment; and (6) palliative or hospice care. The collected primary reasons for cancellation and follow-up management were reviewed by at least 2 physicians with review by a third expert reviewer if there was uncertainty.
The preoperative assessment center evaluated 16,955 cases for operative procedures during the study period with demographic and procedural information listed in Table 1. The mean age of the study population was 55 years. Sixty-four percent of the study participants were females, which reflects the large-volume gynecology service at the institution studied (18% rate of gynecologic procedures for all comers during the study period). The majority of patients had an ASA physical status II (51.4%) or III (38.4%).
One hundred forty-seven of the 16,955 cases were cancelled for reasons associated with their preoperative assessment before their operative date, representing approximately 1% of the study sample (0.87%; corrected cancellation rate and 95% CI, 0.84% [0.84%–0.85%]). For cancelled cases seen in the clinic, the median time between preoperative appointment and planned surgery was 6 days (interquartile range of 3–12 days). Information on comparison of cancelled and noncancelled cases is listed in Table 2. Patients with case cancellation were older than noncancelled cases (P < 0.0001) and had a smaller proportion of females (P = 0.02). Patients whose cases were cancelled had higher ASA physical status (P < 0.0001).
The results comparing the variables noted in Table 2 with clustering by month as well as the results comparing cancellation rates by sex and the binary version of ASA physical status using the method described by Dexter et al.18 are shown in Appendix 1. These results were very similar to the P values reported in Table 2 with no change in statistical significance.
The in-depth physician review of the cancelled cases can be found in Tables 3 and 4. Approximately 70% of cancellations were because of uncontrolled or severe medical comorbidities that received further evaluation or management (69% [95% CI, 61.3%–76.7%]). Further analysis of the organ system responsible for cancellation showed that more than half of these cancellations were for cardiac (50%) and pulmonary (16%) issues.
Nine percent of cases were cancelled because the patient judged the risks of the scheduled procedure to outweigh the clinical benefits (Table 3). Another 4% of cases were cancelled by the clinical team because preoperative clinical changes caused the risks to outweigh the potential benefits. Social, administrative, and patient compliance issues each contributed an additional 4% of cancellations. Of the cancelled cases, 23% (n = 33) were at low risk for cardiovascular events and the remaining 77% (n = 113) were elevated risk (per American College of Cardiology/American Heart Association guidelines).20 The overall indication for the procedure of cancelled cases was 89% (n = 129) for therapeutic reasons, 9% (n = 13) for diagnostic reasons, and 2% (n = 3) for palliative reasons. Representative vignettes of the reasons for cancellation are found in Table 3.
Information on evaluation and management of the cancelled cases can be found in Table 4. At 1 year from the initially scheduled operative date, 69% of these patients had completed further evaluation/treatment and had undergone either the original procedure (67%) or a different procedure (2%) to treat the operative indication.
For the other 31% of patients, a perceived lack of likely benefit by the patient and/or provider(s) precluded completion of their procedure at 1 year. This included approximately one-fourth of patients who were managed nonoperatively (14%), referred for palliative management (4%), or required no further treatment (5%). There were an additional 8% of patients for whom it was unclear at 1 year whether the patient would become appropriate for operative management because of ongoing medical issues.
Approximately 1 in 100 patients scheduled for elective surgery and evaluated in the preoperative clinic were cancelled before the date of the procedure. Seventy percent of these cancellations were because of predominantly cardiac or pulmonary comorbidities that required further evaluation to determine the balance of that patient’s risk–benefit ratio and/or to allow for preoperative risk modification. A perceived unfavorable risk–benefit balance based on existing information accounted for an additional 13% of cancellations with either the patient (9%) or the provider (4%) initiating case cancellation at the time of preoperative assessment for a perceived lack of overall benefit. One-year follow-up found that 69% of patients had undergone surgery for their initially diagnosed condition after further evaluation and optimization. For the other 31% of patients who had not undergone their procedure, clarified priorities rendered medical management, palliative/hospice care, or no treatment more appropriate.
Most cancellations were caused by the need to evaluate or treat poorly controlled medical conditions and optimize the patients’ physiologic fitness for surgery. Patients whose cases were cancelled were older than noncancelled counterparts, and a significantly higher proportion were ASA physical status III or IV. However, even the noncancelled cases were consistent with national trends toward an increasingly aging and medically complex surgical population. The complexity of this population may contribute to the incomplete patient information available at many initial surgical evaluations. In many cases, surgeons will use the period between the decision for surgery and the procedure itself to perform additional testing and obtain information from other providers about comorbidities that can adversely affect surgical outcome, leaving final decisions until after the evaluation in the preoperative testing center. This is consistent with previous research done in our preoperative clinic demonstrating that retrieval of information was required for 84% of issues needing further evaluation of an existing medical condition.21 However, by this time, with the surgery already scheduled, there is little time to modify risk or discuss overall patient goals.7 A recent study found that 13% of patients who had met with their surgeon and signed consent nonetheless had critical or important deficits in their basic understanding of the scheduled procedure such as knowing what procedure they were scheduled to receive or the expected risks and benefits of the procedure.22 The fact that nearly one-third of the cancelled cases did not have operative management at 1 year highlights the need to create systems to identify patients with significant operative risks who might benefit from additional time for joint decision-making.
Preoperative evaluation centers have been shown to be effective in implementing risk factor modification, which additionally may decrease hospital costs through averting expensive complications. Preoperative risks have been found to predict 33% of hospital cost variation, and improved preoperative management of comorbidities has been suggested as a way to decrease hospital costs.23 Approximately 70% of patients in our study who were cancelled for medical reasons may therefore present an avenue for overall cost savings. More than two-thirds of patients with case cancellation in our study were evaluated and/or received medical treatment for their comorbidities and subsequently underwent a surgical procedure to treat their diagnosis within 1 year. Better management of their comorbidities may have helped avert costly complications. In fact, cost savings from risk modification are likely even higher because previous research with this population showed that 12.7% of all comers to our institution’s preoperative evaluation center had additional information requested or management done as a part of the preoperative assessment.21
Our preoperative clinic serves a tertiary care center, which likely has a more medically complex patient population than might be found at community health centers. However, we suspect that the overall types of categories identified are likely to be similar across populations, institutions, and time points.
We observed a difference in sex between cancelled and noncancelled cases. Although small studies have reported differences in cancellation by sex in operations such as cosmetic or elective reconstructive procedures,12 we suspect that the small difference we observed simply reflects a large-volume gynecology service at the institution studied, a pattern that has also been seen in other studies of this population.19 We also observed a difference in the overall distribution of procedure types between cancelled and noncancelled cases, as shown in Table 2. Other small studies on case cancellations have similarly found a difference in cancellation rate by surgical service.11 It is possible that certain issues may be sufficient to cancel, for example, an elective orthopedic procedure but would not suffice to cancel an abdominal procedure scheduled by a surgical oncology service. Nonetheless, lessons learned from examination of these cancellations are likely still applicable to smaller centers with healthier populations, centers that focus on a particular procedure type, or larger centers with a slightly different case mix. We did not review patients taken to the operating room directly from the emergency department or inpatients scheduled for procedures during their admission. Both inpatient and day of surgery cancellations have a larger proportion of cancellations for institution-specific and/or unpredictable issues such as operating room staffing constraints, surgeon availability, or unexpected patient events (e.g., patient noncompliance with nil per os [“npo”] status). With regard to the day of surgery cancellations, previous work from this institution showed that preventable case cancellation on the day of surgery was rare.24 Furthermore, as noted earlier, day of surgery cancellations have been extensively studied.8–14 For these reasons, we narrowed the scope of our project to focus on elucidating the function of the preoperative clinic in managing complex patients, assisting with their risk–benefit assessment, and facilitating alignment of surgical and patient goals.
It is unlikely that hospital rules governing surgeon use of the operating room would have resulted in surgeons scheduling preoperative evaluations for patients who they knew were likely to be cancelled. This would result in a significant waste of preoperative resources and poor patient satisfaction. Metrics on preoperative clinic use by service as well as block time use are presented to the operating room committee for review.
The Perioperative Surgical Home model, proposed by the ASA to improve efficiencies and reduce variations in utilization through increased perioperative care coordination, is one proposal for improving the overall value of surgical care. This has been suggested as a means of leveraging the perioperative expertise of anesthesiologists with the goal of improving individual and population health while decreasing costs.25 Although we reported that 9% of cases did not proceed as scheduled because the patient decided clinical risks outweighed the benefits (Table 3), the patient clearly plays a major role in most if not all the final decisions noted in Table 4, including the decision for palliative/hospice care, a different surgical procedure, or no further treatment. In this era of patient-centered care, a study that explored these roles in further detail would represent an important direction for future work. It is possible that patient-centered care may increase overall cancellation rates both at the time of preoperative assessment and on the day of the procedure.26 It has been suggested that increasing value in health care would involve a shift in the way that we quantify healthcare outcomes and cost to be more inclusive than the scope of our current accounting methods.27,28 As hospitals attempt to grapple with the increasing cost of health care, predicted to exceed one-fourth of the gross domestic product by 2035,29 interventions to tighten inefficiencies and improve patient-centered care will be of increasing importance for hospitals, providers, and patients.
As a result of this work, we are developing protocols to structure the surgical workflow to identify patients at risk of suboptimal surgical outcomes because of their underlying medical fragility, social, or other factors earlier in the workflow. Once identified, we hope to provide the opportunity for providing appropriate comanagement expertise at the preoperative clinic appointment and are piloting these pathways in geriatric, diabetic, and chronic pain patients. In addition, we are training our providers to feel comfortable with conversations about advanced care directives, do-not-resuscitate orders, and patient values and goals to ensure that these issues are addressed appropriately.
Our findings suggest that the preoperative evaluation provides an additional pause point for important reconsideration as to whether the goals of surgery are concordant with the patient’s overall objectives as well as an opportunity for developing and implementing high-value integrated care perioperative pathways.
Appendix 1. Patient Characteristics by Case Cancellation Status (Showing P Values with Clustering by Patient, Clustering by Month, and via the Method Described by Dexter et al.a)
1. Semel ME, Lipsitz SR, Funk LM, Bader AM, Weiser TG, Gawande AA. Rates and patterns of death after surgery in the United States, 1996 and 2006. Surgery. 2012;151:171–82
2. National Center for Health Statistics (US). Health, United States, 2011: With Special Feature on Socioeconomic Status and Health. 2012 May Hyattsville, MD National Center for Health Statistics (US) Available at: http://www.ncbi.nlm.nih.gov/books/NBK98752
. Accessed May 25, 2012
3. Weiser TG, Semel ME, Simon AE, Lipsitz SR, Haynes AB, Funk LM, Berry WR, Gawande AA. In-hospital death following inpatient surgical procedures in the United States, 1996–2006. World J Surg. 2011;35:1950–6
4. Barnato AE, McClellan MB, Kagay CR, Garber AM. Trends in inpatient treatment intensity among Medicare beneficiaries at the end of life. Health Serv Res. 2004;39:363–75
5. Kwok AC, Semel ME, Lipsitz SR, Bader AM, Barnato AE, Gawande AA, Jha AK. The intensity and variation of surgical care at the end of life: a retrospective cohort study. Lancet. 2011;378:1408–13
6. Lee TH. Care redesign—a path forward for providers. N Engl J Med. 2012;367:466–72
7. Mythen M. Fit for surgery? Anesth Analg. 2011;112:1002–4
8. van Klei WA, Moons KG, Rutten CL, Schuurhuis A, Knape JT, Kalkman CJ, Grobbee DE. The effect of outpatient preoperative evaluation of hospital inpatients on cancellation of surgery and length of hospital stay. Anesth Analg. 2002;94:644–9
9. Knox M, Myers E, Hurley M. The impact of pre-operative assessment clinics on elective surgical case cancellations. Surgeon. 2009;7:76–8
10. Dexter F, Shi P, Epstein RH. Descriptive study of case scheduling and cancellations within 1 week of the day of surgery. Anesth Analg. 2012;115:1188–95
11. Schuster M, Neumann C, Neumann K, Braun J, Geldner G, Martin J, Spies C, Bauer MCASCAES Study Group. . The effect of hospital size and surgical service on case cancellation in elective surgery: results from a prospective multicenter study. Anesth Analg. 2011;113:578–85
12. Guyuron B, Zarandy S. Causes for cancellation of aesthetic and reconstructive procedures. Plast Reconstr Surg. 1993;92:662–8
13. Epstein RH, Dexter F. Rescheduling of previously cancelled surgical cases does not increase variability in operating room workload when cases are scheduled based on maximizing efficiency of use of operating room time. Anesth Analg. 2013;117:995–1002
14. Tung A, Dexter F, Jakubczyk S, Glick DB. The limited value of sequencing cases based on their probability of cancellation. Anesth Analg. 2010;111:749–56
15. Wisdom JP, Cavaleri MA, Onwuegbuzie AJ, Green CA. Methodological reporting in qualitative, quantitative, and mixed methods health services research articles. Health Serv Res. 2012;47:721–45
16. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics. 1986;42:121–30
17. Rao JN, Scott AJ. On simple adjustments to chi-square tests with survey data. Ann Stat. 1987;15:385–97
18. Dexter F, Marcon E, Epstein RH, Ledolter J. Validation of statistical methods to compare cancellation rates on the day of surgery. Anesth Analg 2005;101:465–73. Erratum in: Anesth Analg. 2012;114:693
19. Barnet CS, Arriaga AF, Hepner DL, Correll DJ, Gawande AA, Bader AM. Surgery at the end of life: a pilot study comparing decedents and survivors at a tertiary care center. Anesthesiology. 2013;119:796–801
20. Fleisher LA, Fleischmann KE, Auerbach AD, Barnason SA, Beckman JA, Bozkurt B, Davila-Roman VG, Gerhard-Herman MD, Holly TA, Kane GC, Marine JE, Nelson MT, Spencer CC, Thompson A, Ting HH, Uretsky BF, Wijeysundera DNAmerican College of Cardiology; American Heart Association. . 2014 ACC/AHA guideline on perioperative cardiovascular evaluation and management of patients undergoing noncardiac surgery: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines. J Am Coll Cardiol. 2014;64:e77–137
21. Correll DJ, Bader AM, Hull MW, Hsu C, Tsen LC, Hepner DL. Value of preoperative clinic visits in identifying issues with potential impact on operating room efficiency. Anesthesiology. 2006;105:1254–9
22. Ankuda CK, Block SD, Cooper Z, Correll DJ, Hepner DL, Lasic M, Gawande AA, Bader AM. Measuring critical deficits in shared decision making before elective surgery. Patient Educ Couns. 2014;94:328–33
23. Davenport DL, Henderson WG, Kauri SF, Mentzer RM Jr. Preoperative risk factors and surgical complexity are more predictive of costs than postoperative complications: a case study using the National Surgical Quality Improvement Program (NSQIP) database. Ann Surg. 2005;242:463–8
24. Harnett MJ, Correll DJ, Hurwitz S, Bader AM, Hepner DL. Improving efficiency and patient satisfaction in a tertiary teaching hospital preoperative clinic. Anesthesiology. 2010;112:66–72
25. Vetter TR, Boudreaux A, Jones KA, Hunter JM, Pittet JF. The perioperative surgical home: how anesthesiology can collaborately achieve and leverage the triple aim in health care. Anesth Analg. 2014;118:1131–6
26. Hsia IK, Dexter F, Logvinov I, Tankosic N, Ramakrishna H, Brull SJ. Survey of the National Drug Shortage Effect on Anesthesia and Patient Safety: a patient perspective. Anesth Analg. 2015;121:502–6
27. Lee TH. Putting the value framework to work. N Engl J Med. 2010;363:2481–3
28. Porter ME. What is value in health care? N Engl J Med. 2010;363:2477–81
© 2016 International Anesthesia Research Society
29. Bloch MG. Beyond the ‘R word’? Medicine’s new frugality. N Engl J Med. 2012;366:1951–3