The relative importance of different institutional interventions that the largest hospital in Iowa could take to grow the anesthesia department’s intraoperative care was described in 2009.1,2 Most (>50%) patients having elective surgery had not previously had surgery at the hospital.1 Patients’ perioperative experience was unimportant for influencing total anesthesia workload and numbers of patients.1,3 More important was the availability of surgical clinic appointments within several days.1
These results were from 1 large teaching hospital, located hours from any large metropolitan area.1,2 We expected the results to be generalizable, because the teaching hospital performed a large diversity of procedures.4–6 Nevertheless, before anesthesia groups enhance surgeon-focused operating room finance and operations management as compared to patient experience,7–16 consider that the conclusions depend on most (>50%) patients’ time to their next elective surgery exceeding the time for a substantive change in numbers of surgeons and staffed operating rooms (eg, annually).1 The results would not be generalizable if, among all hospitals in a region, the median time from surgery to a patient’s next surgical procedure were <1 years, and might not be generalizable if the median time was <2 years.1
In this article, our primary aim was to evaluate the generalizability of the previous research from the largest hospital in Iowa1 by testing whether the median time from surgery to a patient’s next surgical procedure exceeds 2 years among all hospitals statewide and regardless of the hospital where the patient undergoes the next procedure. We sought to learn what hospital factors, if any, have a large influence on the median times to the next procedure. We performed the study using data for all patients undergoing outpatient surgery at a hospital. (In the United States, this also means surgery with an expected hospital length of stay of <2 midnights, like most laparoscopic cholecystectomy cases.)17,a We considered outpatient surgery for 2 practical reasons. First, outpatient surgery is elective and, in the United States, hospitals code these procedures using the Current Procedural Terminology (CPT) system. We used the American Society of Anesthesiologists’ Crosswalk to obtain relative value guide units from these CPT codes.18 Second, the US Center for Medicare and Medicaid Services reports the results of hospital outpatient departments’ Consumer Assessment of Healthcare Providers and Systems Outpatient and Ambulatory Surgery Survey (OAS-CAHPS).19
The OAS-CAHPS “is designed to measure the experiences of care for patients … for a surgery … The survey is designed to meet the following goals: To produce comparable data on the patient’s perspective that allows objective and meaningful comparisons between” hospital outpatient departments “…on domains that are important to consumers … Public reporting will allow consumers to make more informed choices when choosing” a hospital.20
Our secondary aim of this article was motivated by these described OAS-CAHPS objectives. The objectives’ references to “consumers” seem inconsistent with previous studies’ results.21–25 As summarized in a recent review, patients generally choose their surgeon or group of surgeons.21–26 Growth of the hospital and its anesthesia department’s surgical practice depends on the surgical suite maintaining a productive environment for surgeons (eg, to assure that the many surgeons performing 1 or 2 brief cases27 on days that they operate have access to operating room [OR] time). We therefore compared the median days from when a patient has surgery until the patient undergoes another procedure versus the median days from when the surgeon performs the case to the surgeon’s next case. We made this comparison among all 116 hospitals in Iowa with outpatient surgery, and without regard to the hospital where the next procedure was performed.
The data studied were from the Iowa Hospital Association’s inpatient and outpatient data sets of all encounters at hospitals in the state of Iowa, excluding encounters related to behavioral health and complications from human immunodeficiency virus infection. The patient data were edited, by the Iowa Hospital Association before release, for integrity and for compliance with Health Insurance Portability and Accountability Act regulations. The University of Iowa Institutional Review Board determined that the project did not meet the regulatory definition of human subjects research.
Data From the First Quarter
There were 37,172 surgical cases at hospital outpatient departments of any of the 117 hospitals in Iowa from July 1, 2013, to September 30, 2013.28 During this period, there was at least 1 such case at 116 of the hospitals. Table 1 shows, in detail, the steps that were followed to obtain the data.27 Each of the studied lists of cases included at least one of the 4159 surgical CPT codes with nonzero corresponding intraoperative work relative value units (RVUs) and with nonzero corresponding American Society of Anesthesiologists’ anesthesia base and time units.18,29 We considered outpatient surgery, because those cases’ billing (administrative) codes were CPT; in contrast, inpatient surgical cases were coded using the International Classification of Diseases, Ninth Revision, Clinical Modification (ie, there are not corresponding intraoperative RVUs). The data extracted about these surgical cases from the first quarter are listed in Table 2: case’s procedures’ intraoperative work RVUs, patient’s residence county versus hospital’s county, patient’s age, patient’s principal type of insurance, and hospital’s total outpatient intraoperative work RVUs among all 37,172 cases.
Our primary interest was in results weighted by RVUs because cases are not of equal importance to anesthesia departments, whether considered from the perspective of payment, personnel time, or resources used. Among cases of the first quartile of RVUs (Table 2), the most common CPT was 69436, tympanostomy with insertion of ventilating tube under general anesthesia. In contrast, among cases of the fourth quartile of RVUs, the most common CPT was 47563, laparoscopic cholecystectomy with cholangiography. Analyses of waiting time by RVUs rather than by cases gave 5.8-fold greater weight to a patient undergoing laparoscopic cholecystectomy than 1 undergoing myringotomy tube placement, because our focus was economic. The most common procedures of the studied patients are listed in Table 3.
Matching Data by Patient
The 37,172 cases’ patient linkage identifiers were matched to all outpatient records statewide for the next 2 years meeting the same criteria for RVUs and base units (Table 1). The identifiers allow tracking of patients among outpatient and inpatient encounters at all hospitals, but contain no patient identifying information.
The 37,172 cases’ patient linkage identifiers were also matched to all Iowa Hospital Association inpatient records with a principal surgical procedure code; these codes represent major therapeutic procedures (ie, associated with an operating room charge).30
Matching Data by Surgeon
Among the 37,172 cases from the first quarter, there were 36,914 with the surgeon listed using the National Provider Identifier “of the physician with primary responsibility for performing the principal surgical procedure” (99.3%). The 36,914 cases were performed by 1820 different surgeons. The identifiers of these surgeons were blinded for alignment and statistical analysis.
The 36,914 cases’ surgeon identifiers were matched to the surgeon identifiers of all outpatient records for the next 2 years meeting the criteria for RVUs and base units (Table 1).
The 36,914 cases’ surgeon identifiers also were matched to all Iowa Hospital Association inpatient records with a principal surgical procedure code, same “physician with primary responsibility for performing” that principal surgical procedure, and elective admission type. The criterion of elective surgery was used based on surgeons choosing hospitals for elective (ie, scheduled) surgery, as compared to nonelective surgery (see Discussion). The date used for comparison with the date of the first surgery was the date of the elective admission.
There was a later outpatient case performed by the surgeon for 36,538 of the 36,914 cases from the first quarter studied (99.0%). There was a later inpatient case performed by the surgeon for 27,891 of the original 36,914 cases (75.6%). There was a later case, outpatient or inpatient, performed by the surgeon for 36,653 of the original 36,914 cases (99.3%) and 1719 of the original 1820 surgeons (94.5%). The remaining cases and surgeons were all treated as having time from first surgery to next case as being longer than 2 years; the precise choice for the censoring had no effect on any of our estimates or confidence intervals (CIs; see below).
Ideally, the data by patient (see above) would be summarized using the median days among patients from outpatient surgery to the next surgery, either outpatient or inpatient. However, if more than half the patients do not have surgery again within 2 years, the median cannot be calculated; practically, this would be irrelevant in comparison to surgeons (see Results and Discussion). Therefore, functionally, the primary end point would be the percentage of patients having surgery again within 2 years. Not all cases are equivalent from the perspective of an anesthesia department; the analyses are performed while weighted by the intraoperative component of the work RVUs of the procedure (ie, essentially the typical hours of OR time; see Discussion).1 The percentage of patients having surgery within 2 years was calculated with weighting by treating the RVUs as sampling weights.1 The resulting number of observations used in the calculations of the CIs remained 37,172 cases. Based on Oranje’s31 simulation study of percentages estimated with sampling weights, and with our estimated percentages being far from 0% or 100%, we used Wilson’s method to calculate the 2-sided CIs. Because the sample sizes were large, we used asymptotic standard errors (STATA 15, StataCorp LLC, College Station, TX).32 These calculations were then repeated using different categories of patients (eg, pediatric versus elderly; Table 2). Because we performed multiple such exploratory analyses, we report 99% CIs. All the CIs were then recalculated without weighting by RVUs.
The data by surgeon (see above) were analyzed in 2 ways. First, the days were calculated by surgeon including other cases on the same date, but without duplication. Second, the days were calculated excluding other cases on the same date. For example, suppose that a surgeon performed 3 cases on 1 day and then operated 7 days later. In the first analysis, the median among those 0, 0, and 7 days would be 0 days, and would contribute N = 3 cases to the CI. In the second analysis, the median among those 4 cases would be 7 days, with only N = 1 case contributing to the CI. Using the 36,914 cases with known surgeon, upon treating multiple cases on the same date as functionally 1 case, the sample size for calculating the median and its CI was 18,561 surgeon days. Both CIs for the median were calculated using the Clopper Pearson method without interpolation (ie, conservative).33,34
The statistical power analysis was based on the analysis by patient rather than by surgeon, because it was known a priori that the sample size needed to measure the time for patients to undergo future surgery would be much greater than that needed for surgeons to perform future cases. We previously analyzed the percentage of patients having surgery at the largest teaching hospital in Iowa in 2007 with a previous anesthesia record.1,2 The incidence was 27% with weighting by American Society of Anesthesiologists’ Relative Value Guide Units and 28% without.1 Our objective was to test whether the median time from outpatient surgery to the next case was at least 2 years (ie, whether the percentage of patients with surgery within 2 years was <50%). Performing 2-sided comparison of 28% with 50% using type I and type II error rates of 1.0% would require a sample size of 116 patients. With 3 months of data, our sample size of 37,172 cases was much more than enough, and thus we proceeded. We aimed to use data from all the hospitals in the state to answer the generalizable question.
By patient, the median time to their next surgical case, either outpatient or inpatient, exceeded 2 years, both with weighting by intraoperative RVUs and when unweighted (both P < .0001; Table 2). With weighting, 65.9% (99% CI, 65.2%–66.5%) of the patients had no other surgery within 2 years, at any hospital in the state (Tables 2 and 4). The median exceeded 2 years for multiple categories of patients and similar measures of time to next surgery (Tables 2 and 4; all P < .01). The median was only slightly longer than 2 years among patients in the oldest quartile of ages, but not so when limited to care at the initial hospital (Table 4).
By surgeon, the median time to the next outpatient surgical case was 1 calendar day (99% CI, 0–1 day). The median was 3 days to the next day with at least 1 outpatient case (99% CI, 3–3 days).
Patient experience is an important attribute of quality of care. The surgeons’ experience is important from the vantage point of marketing and growing an anesthesia practice.35 In this article, we quantified these relationships. The median time to the next surgery was >2 years for patients versus 1 day for surgeons. Thus, managers of hospitals and anesthesia departments should focus on surgeons’ experiences when making resource decisions. Financial analyses of surgical suites should be performed stratified by surgeon.7–9 Operations research type optimization of operating room scheduling should be performed using objective functions and models designed to increase physician productivity. Case scheduling should be done when a surgeon is available, provided the case can be performed safely.7,11 In addition, surgeon blocks should be calculated validly for coordination of surgical days among surgeons.12,13 For anesthesia groups contemplating the implications of the generalizability of the previous research,1 we recommend that they be engaged in the operating room management, including when hospital-group agreements are signed, so that these interventions7–13 actually get done.14–16
Our results have implications for patient-centered care and informatics. Operationally, the expectation should be that patients undergoing surgery at a given hospital have no previous experience in navigating the perioperative process at that hospital.1,36 The expectation also should be that there will not be prior perioperative information about the patient (eg, history of difficult intubation).36 Patients’ reliance on the availability of surgeon clinic appointments for choosing where to have surgery, and then on the lag time to surgery when choosing the surgeon, should be expected.21–26 It is in the interest of an anesthesia department seeking to grow its practice that its surgeons have clinic appointments available on whatever date patients would prefer.1 In other words, ensuring that patients have a good perioperative experience is certainly important, but our results match the limited previous data in concluding that experiences unrelated to clinic and surgical appointment availability are unlikely to increase caseloads substantively.1,21,37 Whether increasing caseloads would be considered of value likely will differ markedly among countries, hospitals, etc.
Our study was limited to data from just 1 state, Iowa. However, likely, our result of most patients having a period >2 years between surgeries will be generalizable to other states based on Iowa including 117 different hospitals and our conclusions being fully insensitive to the many hospital characteristics studied. Nevertheless, the only related study from outside Iowa1 that we are aware of is a report of patient data at the preoperative clinic of a hospital in Philadelphia.36 There, 82% of patients having elective surgery had no previous surgical care, over a preceding period averaging 2.7 years. The extent to which there are such limited prior data indicates the value of this study, even if from only 1 US state.
Our study was limited to weighting using intraoperative RVUs and to an unweighted analysis by case. However, anesthesia departments may be concerned about American Society of Anesthesiologists’ Relative Value Guide (ASA RVG) units or their equivalent (eg, Canadian schedule of benefits for anesthesiology billing). Several studies’ results show that results would be indistinguishable. First, the probability distributions of numbers of preoperative visits and time from first preoperative visit until surgery was the same when analyzed by ASA RVG units or cases.1 Second, over an interval of 26 years, the annual number of anesthetics at a hospital was highly correlated with the annual number of ASA RVG units (Pearson r = 0.99).38 Third, the correlations were calculated between each surgeon’s numbers of ASA RVG units and cases, and these pooled among surgeons.39 The mean Pearson r = 0.95.39
Iowa is a geographically large state, 12% bigger than England. Therefore, we doubt that we are substantively overestimating the percentage of patients not having surgery again within 2 years because of data missing data from patients undergoing a later procedure out of the State. However, it is possible because Iowa lacks water barriers along its borders.
Finally, our study was limited to hospital outpatient surgical departments, because the Iowa Hospital Association data do not include freestanding surgery facilities.40 Because surgical cases at freestanding facilities tend to be brief, and, thus, some surgeons will perform several such cases per day, the implication is that our results have systematically underestimated the importance of surgeons’ experiences relative to that of patients’ experiences. The substantive competitive effects of freestanding surgery centers on hospitals’ outpatient departments’ caseloads have also not been included.41–43 The fact that this limitation would seem inconsequential, given that the median time to the next surgery was >2 years for patients versus 1 day for surgeons, indicates the reliability of our conclusion that surgeons’ experiences versus patients’ experiences may be orders of magnitude more influential in growing anesthesia workload.
Name: Franklin Dexter, MD, PhD.
Contribution: This author helped design the study, analyze the data, and write the manuscript.
Name: Craig Jarvie, MMR.
Contribution: This author helped obtain the data and revise the manuscript.
Name: Richard H. Epstein, MD.
Contribution: The author helped design the study and write the manuscript.
This manuscript was handled by: Tong J. Gan, MD.
1. O’Neill L, Dexter F, Wachtel RE. Should anesthesia groups advocate funding of clinics and scheduling systems to increase operating room workload? Anesthesiology. 2009;111:1016–1024.
3. Tiwari V, Queenan C, St. Jacques P. Impact of waiting and provider behavior on surgical outpatients’ perception of care. PCORM. 2017;7:7–11.
4. Dexter F, Wachtel RE, Yue JC. Use of discharge abstract databases to differentiate among pediatric hospitals based on operative procedures: surgery in infants and young children in the state of Iowa. Anesthesiology. 2003;99:480–487.
5. Dexter F, Ledolter J, Hindman BJ. Quantifying the diversity and similarity of surgical procedures among hospitals and anesthesia providers. Anesth Analg. 2016;122:251–263.
6. Dexter F, Ledolter J, Epstein RH, Hindman BJ. Operating room anesthesia subspecialization is not associated with significantly greater quality of supervision of anesthesia residents and nurse anesthetists. Anesth Analg. 2017;124:1253–1260.
7. Dexter F, Blake JT, Penning DH, Lubarsky DA. Calculating a potential increase in hospital margin for elective surgery by changing operating room time allocations or increasing nursing staffing to permit completion of more cases: a case study. Anesth Analg. 2002;94:138–142.
8. Dexter F, Ledolter J, Wachtel RE. Tactical decision making for selective expansion of operating room resources incorporating financial criteria and uncertainty in subspecialties’ future workloads. Anesth Analg. 2005;100:1425–1432.
9. Dexter F, Lubarsky DA, Blake JT. Sampling error can significantly affect measured hospital financial performance of surgeons and resulting operating room time allocations. Anesth Analg. 2002;95:184–188.
10. Wachtel RE, Dexter F. Tactical increases in operating room block time for capacity planning should not be based on utilization. Anesth Analg. 2008;106:215–226.
11. McIntosh C, Dexter F, Epstein RH. The impact of service-specific staffing, case scheduling, turnovers, and first-case starts on anesthesia group and operating room productivity: a tutorial using data from an Australian hospital. Anesth Analg. 2006;103:1499–1516.
12. Dexter F, Macario A, Traub RD, Hopwood M, Lubarsky DA. An operating room scheduling strategy to maximize the use of operating room block time: computer simulation of patient scheduling and survey of patients’ preferences for surgical waiting time. Anesth Analg. 1999;89:7–20.
13. Dexter F, Macario A, Traub RD, Lubarsky DA. Operating room utilization alone is not an accurate metric for the allocation of operating room block time to individual surgeons with low caseloads. Anesthesiology. 2003;98:1243–1249.
14. Masursky D, Dexter F, Nussmeier NA. Operating room nursing directors’ influence on anesthesia group operating room productivity. Anesth Analg. 2008;107:1989–1996.
15. Dexter F, Wachtel RE. Strategies for net cost reductions with the expanded role and expertise of anesthesiologists in the perioperative surgical home. Anesth Analg. 2014;118:1062–1071.
16. Dexter F, Epstein RH. Associated roles of perioperative medical directors and anesthesia: hospital agreements for operating room management. Anesth Analg. 2015;121:1469–1478.
21. Dexter F. Factors substantively influencing numbers of surgical cases performed at a research hospital. J Res Hosp. 2017;2:6.
22. Schwartz LM, Woloshin S, Birkmeyer JD. How do elderly patients decide where to go for major surgery? Telephone interview survey. BMJ. 2005;331:821.
23. Wilson CT, Woloshin S, Schwartz LM. Choosing where to have major surgery: who makes the decision? Arch Surg. 2007;142:242–246.
24. Zwijnenberg NC, Hendriks M, Bloemendal E, et al. Patients’ need for tailored comparative health care information: a qualitative study on choosing a hospital. J Med Internet Res. 2016;18:e297.
25. Gourevitch RA, Mehrotra A, Galvin G, Karp M, Plough A, Shah NT. How do pregnant women use quality measures when choosing their obstetric provider? Birth. 2017;44:120–127.
26. Logvinov II, Dexter F, Dexter EU, Brull SJ. Patient survey of referral from one surgeon to another to reduce maximum waiting time for elective surgery and hours of over-utilized operating room time. Anesth Analg. 2017 July 11 [Epub ahead of print].
27. Dexter F, Jarvie C, Epstein RH. At most hospitals in the state of Iowa, most surgeons’ daily lists of elective cases include only 1 or 2 cases: individual surgeons’ percentage operating room utilization is a consistently unreliable metric. J Clin Anesth. 2017;42:88–92.
31. Oranje A. Confidence intervals for proportion estimates in complex samples. ETS Res Report Series. 2006;2006:i–26.
33. Clopper CJ, Pearson ES. The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika. 1934;26:404–413.
34. Hahn GJ, Meeker WQ. Statistical Intervals. A Guide for Practitioners. 1991:New York, NY: John Wiley & Sons; 82–84,100105.
35. Scurlock C, Dexter F, Reich DL, Galati M. Needs assessment for business strategies of anesthesiology groups’ practices. Anesth Analg. 2011;113:170–174.
36. Dexter F, Witkowski TA, Epstein RH. Forecasting preanesthesia clinic appointment duration from the electronic medical record medication list. Anesth Analg. 2012;114:670–673.
37. Dexter F, Birchansky L, Bernstein JM, Wachtel RE. Case scheduling preferences of one surgeon’s cataract surgery patients. Anesth Analg. 2009;108:579–582.
38. Masursky D, Dexter F, O’Leary CE, Applegeet C, Nussmeier NA. Long-term forecasting of anesthesia workload in operating rooms from changes in a hospital’s local population can be inaccurate. Anesth Analg. 2008;106:1223–1231.
39. Dexter F, Masursky D, Ledolter J, Wachtel RE, Smallman B. Monitoring changes in individual surgeon’s workloads using anesthesia data. Can J Anaesth. 2012;59:571–577.
40. Funk RJ, Owen-Smith J, Landon BE, Birkmeyer JD, Hollingsworth JM. Identifying natural alignments between ambulatory surgery centers and local health systems: building broader communities of surgical care. Med Care. 2017;55:e9–e15.
41. Courtemanche C, Plotzke M. Does competition from ambulatory surgical centers affect hospital surgical output? J Health Econ. 2010;29:765–773.
42. Carey K, Burgess JF Jr, Young GJ. Hospital competition and financial performance: the effects of ambulatory surgery centers. Health Econ. 2011;20:571–581.
43. Al-Amin M, Housman M. Ambulatory surgery center and general hospital competition: entry decisions and strategic choices. Health Care Manage Rev. 2012;37:223–234.