 |
||
 |
Frequent Questions |
|
|---|---|---|
|
|
||
|
|
Frequent Questions |
|
|---|---|---|
|
Frequent
Questions
*****
Answers
to Frequent Questions ***** Section: Consulting Information What relationships does Dr. Dexter have with healthcare consulting firms? A primary focus of Dr. Dexter and his colleagues' work is assisting other organizations develop and use state-of-the-art methods to analyze OR information system, anesthesia information management system, anesthesia billing, hospital financial, and hospital discharge abstract [marketing] data. These methods include optimization based on linear, quadratic, and stochastic programming techniques. Much of the work is performed for healthcare consulting firms. Supplemental services offered include analysis and interpretation for consulting firms. When working with some consulting firms, clients have been unaware of the University of Iowa’s involvement. Likewise the client and origin of the data is unknown to Dr. Dexter and his team. Some consulting firms have chosen, alternatively, to identify Dr. Dexter as an academic sub-contractor providing advanced statistics and robust mathematics. Clearly, the largest part of any process improvement process occurs not in the analysis, but by implementing change after interpreting the data and changing the parent organization. These steps are directed by the consultants, not the University of Iowa.
What are Dr. Dexter's financial relationships? Dr. Dexter receives no funds personally, including honoraria, other than his salary and allowable expense reimbursements from the State of Iowa. He and his family have no financial holdings in any company related to his work, other than indirectly through mutual funds for retirement. He has tenure and does not participate in any incentive programs.
Can you obtain the data that are needed from my information systems? Dr. Dexter has a well established business process for the efficient retrieval of information from hospital OR information systems, anesthesia information management systems, and/or anesthesia group billing information systems. Dr. Dexter’s team is familiar with most commercially available systems and have considerable experience in extracting and cleaning data.
How can I get the software to run the analyses that you have published? Dr. Dexter’s team provides outsourced analytical support. If you send your OR information system, anesthesia information system, and/or financial data to him, they will perform the statistical analyses for you, generally within one week. Many of these analyses are performed using the CalculatOR™ software package. Reports are then discussed by phone and/or web conferencing. Whereas any interested manager or clinician can quickly learn the results of the analyses and how to implement them, many organizations lack in-house staff with the strong background in statistical methods that is required to perform the analyses, test the statistical assumptions, compensate for missing data, and so forth. The alternative to outsourcing data analysis is to train individuals in your organization in the appropriate methodologies. To become facile in performing the calculations, typically several weeks of full-time training are required. Maintaining these skills is challenging, since the methods are typically performed only once or twice a year. If you are not sure, you may want to take one of the courses listed on the Home page.
How can Dr. Dexter help a new OR manager? For the operational and financial aspects of OR management, it is important to learn and apply the science, because it is not intuitive and there are not data that experience improves decision-making. One of the quickest and least expensive ways to learn the science is not to rely on yourself to find precisely the right material or to hope that a conference will cover precisely the right topic in a format that you can apply. Instead, budget a few hours of an expert’s time per month, available by telephone, e-mail, web conference, etc. As needed, describe your problem and have the person you choose send you to the relevant section of the most appropriate and recent paper, critique your recommendations, and/or recommend how to improve the quality of the internal reports that you are being provided. Dr. Dexter’s team has been doing this type of work for several years. Frequently, the OR manager starts by having him perform a quantitative assessment of operational and financial performance of the surgical suite, as described in the Operations Research of the web site. This baseline assessment can help the new OR manager determine rapidly which problems to focus on first to improve a desired goal.
Does Dr. Dexter consult on monitoring anesthesia quality? The consulting performed by Dr. Dexter and his colleagues does not directly involve monitoring anesthesia quality. In Dr. Dexter’s classes, he teaches repeatedly that two people answering the same question should get nearly identical answers if they are basing their responses on well-established statistical methods. This principle holds, for example, when calculating the appropriate number of hours of financial support a hospital should provide to an anesthesia group. However, statistical methods do not currently exist for assessing anesthesia quality. The reason is that there is no one pain or nausea score, rate of compliance, or rate of complications that is optimal. There is no one best rate of unexpected admission to an intensive care unit from an ambulatory surgery center. If a facility strives to reach zero, the consequence can be an increase in the scheduling of certain types of surgical procedures as inpatient, such as pediatric bronchoscopy and laser cases. Consequently, benchmarking is needed to monitor anesthesia quality. Dr. Dexter does perform systematic literature reviews, including those with quality end points. He also develops and applies statistical methods for comparing end points among providers or facilities (i.e., hierarchical analyses). Benchmarking necessitates such appropriate risk adjustment methodologies. If your hospital were considering implementation of systematic quality monitoring and/or credentialing, Dr. Dexter can: (a) review the proposal in the context of current scientific literature and evaluate any inconsistencies, (b) identify issues that need to be addressed regarding the proposed statistical methodology, and (c) recommend additional areas amenable to quality monitoring.
Section: Surgical Services Management Click here to download slides or lectures. That may be the most helpful. I have selected several papers that include long background sections or that stand alone with key results. Click on each of the following links to get the reference and abstract for each paper, or click on [PDF] to download the full article. Impact of service-specific staffing, case scheduling, turnovers, and first-case starts on anesthesia group and operating room productivity: tutorial using data from an Australian hospital [PDF] Making management decisions on the day of surgery based on operating room efficiency and patient waiting times [PDF] Tactical decision making for selective expansion of operating room resources incorporating financial criteria and uncertainty in sub-specialties’ future workloads [PDF] Tactical increases in operating room block time based on financial data and market growth estimates from data envelopment analysis [PDF] Tactical increases in operating room block time for capacity planning should not be based on utilization [PDF] Economic analysis of linking operating room scheduling and hospital material management information systems for just in time inventory control [PDF] How to schedule elective surgical cases into specific operating rooms to maximize the efficiency of use of operating room time [PDF] Use of operating room information system data to predict the impact of reducing turnover times on staffing costs [PDF] Strategies to reduce delays in admission into a postanesthesia care unit from operating rooms Optimizing second shift OR staffing
What management reports do you recommend for operating rooms?
Click
here for a sample report from an OR Staffing and Allocation consult.
This includes the operational reports I recommend.
Click here
for a review article that describes many of those analyses. In addition,
financial performance should be measured.
Click here for a two-page summary of financial performance
measurement, and
click here
for a lecture on the topic. The first half of the lecture provides
multiple reasons why neither adjusted utilization nor raw utilization is
a valid surrogate for the statistic "contribution margin per OR hour".
Click here to read the abstract describing that contribution margin
per OR hour applies to tactical decision making or
click here
to download the full article.
Click here to read the abstract of the follow-up article or
click here to download the full article.
Click here to read the abstract describing that the standard error of the contribution
margin per OR hour can be measured and its important effect incorporated
into the analysis or
click here
to download the full article.
With respect to OR allocations and staffing on workdays, the most important data are the time of the end of the last case of the day in each OR. Sometimes this can be inferred from employee time cards. The incremental value of each additional datum is sometimes small. Consider two scenarios. In scenario A, there are data for two months on the times of the last case ending in each OR. In scenario B, there are data for two weeks on all of the cases. Scenario A would be much more beneficial. In some situations, the incremental value of collecting more data, as opposed to simply using expert opinion, will be negligible. For example, suppose that the workday begins at 7 AM. The managers are polled to estimate when the salaried orthopedic surgeons finish their cases in an OR. The managers guess that the surgeons finish their elective cases between 2:45 PM and 4:15 PM each workday. In that circumstance, there would be little incremental value in collecting more time data. Click here to download the full article. Adjusting staffing would have negligible impact on OR efficiency, as would reducing turnover times. Click here for the abstract or click here to download the full article. With respect to improving decision-making processes, creating scenarios manually can be a quicker way to evaluate processes than via observation. For a description of scenarios, download this PDF. In one full day with one or two experienced on-site day-to-day manager(s), the scenarios will have been created manually. By the end of the day, (a) you'll have a good assessment for how all of the managerial decisions are being made and (b) how far those decision processes are from that which is the optimum based on the ordered priorities. Using detailed data on all cases to create the scenarios automatically is preferred, because such data are unbiased and do not take the managers time. However, use of the scenarios as pre-designed role plays and mock up of decisions is a far faster process than observation. The scenarios may also train the manager(s) in the process of their assistance. Completing the scenarios for longer-term educational value would take another day of work.
This is sometimes precisely the situation where the tactical (strategic) analyses based on financial criteria apply. For example, consider the hypothetical hospital three paragraphs above that finishes all ORs between 7 hr and 9 hr after the start of the workday. This seems to occur often when salaried physicians won't make more money for working longer for elective cases. In that circumstance, reducing turnover times will generally create more under-utilized OR time, not increase OR efficiency because there are no over-utilized hours to reduce. Click here to download a paper reviewing these concepts. Staffing analyses are of the greatest value when some ORs have under-utilized OR time and some have over-utilized OR time each workday. In this scenario, there may be little opportunity for improvement in OR efficiency by adjusting staffing, because there are few under-utilized or over-utilized hours. Click here for corresponding lecture. Often it seems that, at such hospitals, the limiting factor in caring for more patients is the annual budget. Tactical (strategic) decision making based on the large differences in variable costs per OR hour among surgical specialties may be of greater value than consideration of OR efficiency. My impression is that, at such facilities, often when people on-site speak about "efficiency," what they truly mean is providing care for more patients with available resources. That is not OR efficiency as studied scientifically, but a tactical (strategic) decision-making problem. Click here for the related lecture. The incremental reimbursement for each patient may equal zero, or can be represented from a societal perspective as a value per patient treated. The principal issue is to use resources wisely focused on the individuals, departments, and specialties providing the greatest return.
We are implementing “block” OR allocation – any pointers? First, nothing is more important financially when implementing or adjusting blocks than calculating the correct allocations. A good summary of a decade of science is to allocate OR time based on OR efficiency, not based on OR utilization. Allocating OR time based on OR utilization is both logically and computationally flawed, and consequently will often give the wrong answer to the problem. If you allocate too much OR time, then much will be under-utilized, thereby reducing OR efficiency. If you allocate too little, then there will be many over-utilized to finish the cases, resulting in even more expensive over-utilized hours. Click here for a lecture or click here for a review article. If you are focusing on surgeon blocks, click here for the appropriate lecture, click here for the appropriate review article, focus on the following science article, and look at the "Surgeon Blocks" page in our example report that uses this statistical method.
What is the average OR utilization in the United States?
The value is not known, nor can it be known, because operating room
utilization cannot be measured accurately with sufficiently brief duration
data sets as to be practical. There are three reasons. First, the
“surgical service” refers to a group of surgeons who share allocated OR
time. An individual surgeon, a group, a specialty, or a department can
represent a surgical service. There is usually heterogeneity among
services in their adjusted and raw utilization. Thus, the overall average
utilization at a facility is of unclear importance. Second, for services
that have been allocated one OR on some days of the week, the utilization
cannot be measured accurately unless the value is too low or high as to be
of no practical value (click
here for the abstract or
click here to download the full article). Third, for services that
have been allocated two or more ORs on some days of the week, one such OR
can have under-utilized OR time (i.e., adjusted utilization < 100%) while
another such OR has over-utilized OR time. Then, the average utilization
has no relationship to costs, efficiency of use of OR time, or OR
staffing.
With this being said, there are many ORs in the US with fewer than 8 hr
of cases per OR per day. Eleven community anesthesiology groups in the
U.S. had an average of 6.0 hr of anesthesia time per OR per day (click
here). Eight community hospitals in the U.S.
had an average of 5.5 hr of OR time per OR per day in their ORs used for
knee or hip replacement surgery (in press). US hospitals nationwide
averaged 2.1 cases per OR per day (click
here). At a series of academic hospitals,
many ORs had less than 8 hr of cases per OR per day (click
here and
click here).
What is a good OR utilization value for a surgical suite? What’s too low or high? Sixty percent is absurdly low, and 95% is too high. The range is too large to be useful, which is why the answer to the question is that analysis needs to be performed for each surgical suite. Almost always when someone is measuring utilization, this is for tactical decision making as described here and reviewed here.The reason for this is that utilization best applies when one considers ORs as being a fixed resource, not finishing late. From a “macro” perspective, which is quite suitable from a tactical perspective, that is fine. From an operational perspective, and particularly on the day of surgery, this is quite absurd. Allocating OR time based on OR utilization is both logically and computationally flawed, and consequently will often give the wrong answer to the problem. Instead, OR time should be allocated based on OR efficiency (Click here for a lecture or click here for a review article). The latter considers not just under-utilized OR time (i.e., utilization), but also the higher cost of planning too little OR time resulting in more expensive over-utilized OR time. Whereas decision-making based on OR utilization relies first on knowing “what utilization is best,” there is one single answer to best staffing based on OR efficiency and minimizing staffing costs.
Consider a service with total hours of elective cases including turnover times averaging 5 hours every Monday. The service was allocated a single OR for 8 hours. Then, its adjusted utilization is 62%. There are 3 under-utilized hours and 0 over-utilized hours. Because there are no over-utilized hours, allocation based on OR efficiency is identical to allocation based on OR utilization. In contrast, suppose that the same surgical suite has 3 of its 8 ORs as unblocked, open, first-come first-served, other time. The surgical suite staffs in 8 hr, 10 hr, and 13 hr shifts, where 13 hr = 40 hours a week / 3 days per week. Then, those 3 ORs could be allocated as 8/8/8, 8/8/10, 8/10/10, 10/10/10, 8/8/13, 8/13/13/, 13/13/13, 8/10/13, 10/10/13, and 10/13/13. Only by calculations based on OR efficiency, which considers both expected under-utilized and over-utilized hours of OR time, can a good staffing decision be made. Click here for a review article.
Surgical service refers to a group of surgeons who share allocated OR time. An individual surgeon, a group, a specialty, or a department can represent a surgical service. Surgical service simply refers to the unit of OR allocation. For example, suppose that all of the cardiothoracic surgeons practicing at a hospital are allocated OR time. Then, cardiothoracic surgery would be a service. For example, suppose that two otolaryngologists are partners in one of three otolaryngology groups that practice at a hospital. If the two otolaryngologists are together allocated OR time, then they would represent a service. For example, suppose that a busy surgeon is personally allocated 10 hr of OR time every Wednesday. Then, from the perspective of OR time allocation, that surgeon would represent a surgical service. Click here for a review article.
"Staffing" refers to the number of OR teams planned at each time of the day and on each day of the week. During scheduled work hours on weekdays, staffing by specialty is determined from the OR workload of different specialties. Click here for a lecture on the topic. Click here for a review article. During the afternoons and evenings of weekdays, staffing is often less, by specialty, than during weekdays. Staffing can again be determined from the OR workload, but using different statistical methods. Click here for a lecture on the topic (based on anesthesia staffing, but also applicable to OR nurses). Click here for the abstract of the paper in AORN Journal (directed at OR nursing staffing). This graphical method to analyze 3 PM to 11 PM staffing also works well for 11 PM to 7 AM staffing, and in practice is implemented simply by subtracting 8 hours from each of the DateTime fields thereby treating all cases performed between 11 PM and 7 AM as if they were done between 3 PM and 11 PM. During the weekends, staffing in-house and on call from home should be determined from OR workload and the patient acuity. Click here for the abstract in AORN Journal describing the steps. Click here for an article about holiday and weekend staffing. Download Acrobat PDF describing one-time and on-going assessment of all three methods at your surgical suite. Download Acrobat PDF describing afternoon and weekend staffing at your surgical suite. Click here for a sample report.
What is an appropriate subsidy for and productivity of an anesthesiology department? Articles in the bibliography show that typical annual values are 8,200 to 12,100 ASA units per OR. The range is wide, because productivity depends on CRNA:MD concurrency, number of anesthetizing locations covered, typical durations of the OR workday, how OR time is allocated, how cases are scheduled, how anesthesia providers are scheduled, how cases are assigned, and so forth. Appropriate productivity and a subsidy can be calculated using OR information system, anesthesia information system, and/or anesthesia billing data. For details on how Dr. Dexter performs a subsidy analysis, download this file. Click here for the full article describing the methodology on which the subsidy analysis is based.
There are two reasons why this is so. First, when variation by day of the week is addressed by the 1st shift (OR efficiency) analysis (click here for lecture or click here for review article), two factors are modeled: service (i.e., unit of OR allocation) and day of the week. When variation by day of the week is addressed by the 2nd shift (afternoon staffing) analysis, there are three factors: specialty team, day of the week, and time of the day. The 1st shift analysis does not need to consider time of the day as an independent variable, because OR time is allocated and cases are scheduled based on OR efficiency, thereby relating the dependent variable of workload with the time of the day at which cases are performed. For example, if there were 10 hours of cases, the 1st shift (OR efficiency) analysis would assume that the day would end at 5 PM, whereas the 2nd shift (afternoon staffing) analysis would make no assumption about when the workload would end. The result is that there is more uncertainty in an estimate from the 2nd shift (afternoon staffing) analysis than from the 1st shift (OR efficiency) analysis. Generally, results from the 2nd shift (afternoon staffing) analysis are such that uncertainties in estimates of appropriate staffing for combinations of team, day of the week, and time of the day exceed the variations among day of the week in appropriate staffing for combinations of team and time of the day. Second, the 2nd shift (afternoon staffing) analysis is based on team (i.e., skill mix), not service (the unit of OR allocation). A limit to how much work can be done safely on any given workday is the number of ORs with staff having the skills to perform a case. Generally, teams will work every workday. The consequence is that, by design, little variation is expected by day of the week in how many cases are performed by each team possessing special skills.
Where can I get a list of guidelines for scheduling cases, moving cases, etc.? The scheduling guideline is to follow the ordered priorities: (1) safety, (2) surgeon open access to OR time on any future workday for elective cases, (3) minimizing over-utilized OR time on the day of surgery to maximize OR efficiency, and (4) reducing patient and surgeon waiting from scheduled start times on the day of surgery. For additional information click here to download a lecture and click here to get details about how to customize lessons for your facility. My colleagues and I have also published two review articles: Making management decisions on the day of surgery based on operating room efficiency and patient waiting times [PDF] and Impact of service-specific staffing, case scheduling, turnovers, and first-case starts on anesthesia group and operating room productivity: tutorial using data from an Australian hospital [PDF].
For operations researchers planning to simulate a surgical suite, what hints can you recommend? Click here to read one of our review articles. First, for scheduling, use a hierarchical process, where the primary “job” is the surgeon, and the secondary “job” are the surgeon’s list of cases for the day. Thus, you will have 1 or 2 non-preemptive jobs per machine per day. Second, for a tactical (e.g., 1-2 year) perspective, surgical suites can be considered to have fixed hours into which cases are scheduled. Because few ORs run 24 hr a day, unlike hospital wards, intensive care units, and emergency departments, the fixed hours are for staff and specialized equipment. For simulations of operational processes (e.g., 0 to 3 months before the day of surgery), staffing is determined stochastically based on the workload, not vice-versa. Third, the objective of the simulation may be to assist clinicians in understanding why a change in how cases are scheduled would be of value to the facility, clinicians, and/or patients. Instead of using simulation, consider using the actual OR information system data to identify some real examples of what would happen. For example, to show to a surgeon how cases could be sequenced, identify some specific days providing good examples. I describe how I use such vignettes on the second page of this downloadable file.
How should a surgeon’s list of elective cases in the same OR on the same day be sequenced? I recommend that you read the following review article: Making management decisions on the day of surgery based on operating room efficiency and patient waiting times [PDF]. Safety is usually unaffected by the sequence of elective cases. Nonetheless, if it were, safety is the first priority. Next, check if there is conflict over equipment or specialized personnel such that sequencing cases among ORs will reduce over-utilized hours. For the mathematics, see: Bayesian prediction bounds and comparisons of operating room times even for procedures with few or no historical data [PDF]. Typically, the decision will still not have been made. Sequence the cases from the most to the least predictable in duration. Predictability can be quantified as the difference between (i) the upper 90% prediction bound for the duration of the case and (ii) the mean of historical durations for the case. Generally, shorter cases will be more predictable than longer cases. However, there are so many exceptions that predictability should be calculated explicitly for each surgical case.
The validity and usefulness of the methods of scientific OR management are based on findings in the peer-reviewed literature. During the peer-review process, manuscripts are scrutinized by other experts in the field, who carefully consider the statistical validity of the analysis and the general applicability of the study. Limitations or caveats are presented in the discussion section of each paper, which may indicate if the paper is relevant to your situation.
To evaluate a surgical suite’s practice of declaring cases as Urgent versus Elective, we routinely calculate the difference between the dates at which each case was scheduled versus performed. This data can also be used to evaluate a surgical suite’s practice of releasing allocated OR time (e.g., see Impact of service-specific staffing, case scheduling, turnovers, and first-case starts on anesthesia group and operating room productivity: tutorial using data from an Australian hospital). Otherwise, this date information has been challenging to use. The date that a case is scheduled in an information system may not be the date that the patient requested to be scheduled for surgery (e.g., see Monitoring trends in waiting periods in Canada for elective surgery: validation of a method using administrative data). For many OR management decisions, the waiting time of interest is that of each subspecialty at a facility. There are substantial statistical problems in measuring waiting times accurately for small numbers of surgeons, because patients’ waiting times are not independent random samples – one patient’s wait affects another patient’s wait (e.g., see 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). Generally, valid statistical estimation of mean days waiting cannot be done by measuring days and taking the average, but by measuring days between patients’ requests to be scheduled for surgery and case durations, and then inferring the average waiting time (e.g., Click here to download the full article). If the management objective is to determine an appropriate OR capacity for patients to undergo surgery in a reasonable number of weeks, then the current waiting time need not be measured (e.g., see Changing allocations of operating room time from a system based on historical utilization to one where the aim is to schedule as many surgical cases as possible). Finally, the management objective may be to monitor patient waiting times daily to focus surgical clinic scheduling toward surgeons who have open OR time. This practice can cause oscillations in OR workload (see Enterprise-wide patient scheduling information systems to coordinate surgical clinic and operating room scheduling can impair operating room efficiency).
Our software attributes the case to the surgeon with the longer incision to closure time, as an arbitrary but systematic approach. Nevertheless, this is a minor issue when OR time is allocated correctly, as the process should have a negligible, if any, impact on appropriate OR allocations. When allocating OR time operationally (i.e., adjusting staffing to match existing workload), the problem is best avoided by applying good statistical analysis whose results are insensitive to how the case is attributed. The allocation of OR time to individual surgeons is highly sensitive to changes in data, making forecasts unreliable statistically (click here for the abstract or click here for the full article). In contrast, allocations based on the efficiency of use of OR time are insensitive to changes in the data (click here for the abstract or click here for the full article). Downloads available include a review article, lecture, consultation information, and examples of analyses. When allocating OR time tactically based on contribution margin per OR hour, results must be insensitive to each patient, as there are so many outliers in financial data. Thus, even if patients with multiple surgeons of different subspecialties were excluded, decisions should be the same. An appropriate tactical financial analysis (click here for the abstract or click here for the full article) includes calculation of confidence intervals (click here) and their use to exclude the influence of outlier patients on results (click here for the abstract or click here for the full article). Downloads available include a lecture and consultation information.
How do I apply quadratic programming to analyze OR financial data? Quadratic programming in OR management was described originally in the paper: Managing risk and expected financial return from selective expansion of operating room capacity: mean-variance analysis of a hospital's portfolio of surgeons [PDF]. Also use the follow-up papers: Tactical decision making for selective expansion of operating room resources incorporating financial criteria and uncertainty in sub-specialties’ future workloads [PDF] and Tactical increases in operating room block time based on financial data and market growth estimates from data envelopment analysis [PDF]. Individuals knowledgeable in operations research usually ask about applying Fieller’s method to calculate the variance for the assessment of contribution margin per OR hour by surgeon. The equation used is that given in reference 6 below equation (6). For individuals without experience, I recommend reference (5) to learn about quadratic programming. Click here for information about the University of Iowa performing the analysis with you. Click here for a lecture on the topic.
How can I calculate the operating room cost from cancelled cases at my facility? The opportunity cost of the revenue from the case may be included, but usually not. Rarely would a case that is cancelled on one day be performed instead at a different facility. However, if that happens, consider not the revenue, but the contribution margin. The contribution margin is revenue minus variable costs (e.g., implants). Contribution margins are typically around $1700 per OR hour. Click here, click here [PDF], and click here [PDF] for financial data from three hospitals. The indirect/ intangible cost to the patient and family can be included. Click here for some references – however be cautious as this excellent, early paper was performed before managerial cost accounting for ORs was refined. Click here for some survey results. Also, include the incremental cost of all patients waiting longer on the day of surgery. The appropriate managerial response to a high cancellation rate is to have other patients arrive earlier in the day for their cases (click here and click here [PDF]). The majority of the cost of cancellations is from the creation of a (very) prolonged turnover. Click here for the methodology of measuring the cancellation rate for each specialty (service) in a statistically sound manner [PDF]. Click here for the methodology of measuring prolonged turnover times [PDF]. Calculate the baseline staffing cost, with staffing planned based on having as low a cost as possible. Then, repeat the calculation after having turnover times for each day that would not have been present had the case not been cancelled (click here for an article). Do not over-estimate costs by taking a cost per minute of OR time and multiplying that by the scheduled duration of the cancelled case. To understand why, consider an outpatient facility with an 8-hr workday, all the surgeons’ cases scheduled each workday, and rarely finishing late. For this very common scenario, the incremental cost of a cancellation is only the indirect/intangible cost to the patient. There is no cost of the empty OR time per se, because there is no lost revenue and the labor costs are fixed. Do not under-estimate costs by claiming that most potential holes in the schedule from cancelled cases are usually filled by add-on cases. As the cancellations were not planned, staffing had to have been planned months in advance for whatever add-on cases typically are scheduled. This is why the OR analysis needs to consider all services (specialties) simultaneously, not only those for which cancellations are being studied.
How can we determine the appropriate number of operating rooms for our cases? Operating room capacity cannot be determined accurately using a ratio of cases per operating room or hours of cases per operating room. The reason is that surgical suite capacity is exquisitely sensitive to how block time is planned, staffing is adjusted to match the workload, and how the cases are scheduled. Click here for a review article. The planning of block time is the allocation of operating room time tactically (e.g., once a year). This choice of the number of first case of the day starts affects the ability of surgical practices to grow. The choice appropriately affects the surgeons’ flexibility in growing their practices. How to allocate operating room tactically (i.e., how many operating rooms to plan) is well understood. Click here to download a lecture. Click here for the abstract of a paper or click here for the full article. Click here for the abstract of a paper combining with estimation of market growth or click here for the full paper. The planning of the number of staffed hours for each operating room of each service (specialty) is the allocation of operating room time operationally. This decision also influences the appropriate number of operating rooms. If staffing can be planned for some operating rooms for 12 hr or 13 hr, then fewer operating rooms are needed than if staffing is planned for 8 hr. Click here to download a lecture. Click here for the latest review article. The appropriate number of operating rooms cannot be chosen reasonably by taking the total hours of cases and dividing by 8 hr, or something like that, because the day to day variability in operating room workload can be large, as can the absolute differences between scheduled and actual operating room times. Click here for the abstract of an article or click here to download its full text.
What data are available on the role of anesthesiologists in OR management? Anesthesiologists’ day to day OR activities can result in decreases and increases in over-utilized OR time. The latter is the economically rational end-point for OR efficiency on the day of surgery. Click here to download a review article. Anesthesiologists can be instrumental in choosing the durations of allocated OR time for each service on each day of the week. Observational studies have shown need for improvement. Among US facilities studied and permitting publication, ten of eleven could have achieved significantly lower labor costs if OR time had been allocated based on OR efficiency (for abstracts, click here and click here; for an article, click here). The same was true of a studied German hospital (click here) and an Australian hospital (click here).
Why is the OR workload not routinely predicted or monitored on a short-term basis? Short-term prediction or monitoring of the total hours of cases
including turnovers (i.e., “OR workload”) seems of little value, based on
the decisions that can be made with the information.
Section: Group Management Strategies What is the single best way to reduce anesthetic drug costs? The single easiest way to reduce anesthetic drug costs is to choose the therapeutically equivalent drug giving the lowest purchase cost. The second easiest way is to reduce drug wastage. For additional solutions, click here.
How should anesthesia group institutional support agreements be calculated?
Finding industry accepted standards regarding anesthesia group productivity is challenging,
because such information is often not helpful. Anesthesia group productivity is determined
predominantly by how OR time is allocated and cases are scheduled.
To download a copy as an Acrobat (PDF) file, click here. The paper describing the instrument is: Development of a measure of patient satisfaction with monitored anesthesia care: the Iowa Satisfaction with Anesthesia Scale. The instrument should not be used for general anesthesia or regional anesthesia. For another study of the instrument for cataract surgery patients, click here. For two papers reviewing methods of measuring patient satisfaction with anesthesia care, click here and click here.For patients undergoing general anesthesia, instruments measure patient satisfaction with anesthesia and surgery (not the anesthetic per se), click here and click here, and click here. For applications on providing information to patients, click here. If your goal is to measure patients’ recovery of functional ability after surgery, click here.
I am not aware of a valid instrument when patients have received a regional anesthetic
(click here).
Section: PACU Staffing Why does our surgical suite have delays in PACU admission? Delays in PACU admission are often caused by delay in discharge. However, the most effective solution is often not to focus on the root cause. Instead, make small adjustments to PACU nursing to match the arrival of patients. For additional information click here and download this file.
Suppose that after laparoscopic surgery patients remain in the phase I PACU for an average of 1 hour. Adding 1 or 2 such patients a week would result in negligible increased costs, much less than the cost per hour estimated as (total PACU nursing budget) / (total PACU patient hours). The reason is that PACU nurses typically and appropriately have substantial idle time, such that they can care for a few extra patients if scheduled appropriately. To understand why, suppose that PACU staffing was sufficient to only have ORs backup and cause delays in admission once every two weeks. This would represent staffing to the 90th percentile of workload. That may seem a high percentile, until the consequences to OR nurses, anesthesia providers, delayed patients, and frustrated surgeons are considered. Typically, administrators staff PACU to the 95th percentiles of workloads to prevent such unpopular incidents. Yet, surgical suites tend to have large day-to-day variability in workload, particularly late in afternoons. As a result, PACU tend to have large day-to-day variability in workload. The result is that staffing PACU to the 95th percentiles of workload result in much larger staffing than needed for the average day. Typically, the difference in staffing is double. For more information, see the preceding question.
What is the most accurate way to perform cost accounting for post-anesthesia care unit time? When I perform a PACU analysis consult for a hospital, one analysis that I always like to perform is calculation of the incremental reduction in total PACU time attributable to delays in discharge. Since most patients are not delayed in discharge, this result often does not match the difference in length of stay between patients with and without delays. Click here for the original paper, and click here for the review article summarizing the method. For a large PACU with many nurses, this analysis may be sufficient. For a small PACU, the costs may be fixed to small changes in the length of stay, as described in the preceding question. Click here to download a lecture including consideration of costing OR and PACU time.
How do we best determine how many phase I PACU beds to construct and staff? If you have not yet browsed the relevant section of the bibliography, I recommend that you start by clicking here. If you will be performing construction that will result in large changes to patient flow, you should perform a discrete-event simulation. Click here for an example. For most hospitals, the modeling is straightforward, because it is not necessary to specify in detail how cases will be scheduled. Click here for details. If you have an existing facility and you want to right-size your staffing, click here for a review on preventing delays in admission into the PACU. If your phase I PACU has patients with a wide range of acuities (e.g., overflow from intensive care units), then click here for the corresponding methodology. Click here for information on services provided by the Department of Anesthesia Operations Research and click here for an example of a staffing analysis.
Section: Economics and Mathematics of Decreasing Anesthesia, Turnover, or Surgical Times How can we reduce delays in when surgical cases start? The key step in applying science to reducing delays is not to just estimate case durations using the average value. Instead, incorporate into the decision-making the estimated uncertainties in the estimates. Just as one surgeon in an OR cannot start until another has finished, painters cannot start until the plumbing is completed. Good project management considers the uncertainty in the decisions. For more details, download this file.
How can the cost savings from reducing OR time and/or turnover times be calculated? The impact of interventions on labor costs can be forecasted using each facility’s own data. Corresponding confidence intervals can also be calculated. For example, turnover times can be reduced between each case. Click here for a review article. Surgical times can be reduced to national average values for each procedure, as reported publicly for Medicare patients in the US. Click here for that article. The impact of new surgical procedures or anesthetic techniques can be modeled. For example, Oncura funded Dr. Dexter's development of an Excel program to assist hospitals in evaluating cryosurgical ablation products. Click here for another example. Regardless of the intervention, the analysis proceeds as follows. First, the labor cost is calculated assuming that OR time is allocated and cases are scheduled based on OR efficiency. Click here for a lecture describing precisely what this means. Second, the intervention is performed, thereby reducing OR workload for each of the services (i.e., surgeons, groups, or specialties) affected. Third, using the revised workload values, OR time is reallocated based on OR efficiency, and the new estimates for labor costs calculated. Fourth, the differences are taken. The result is consideration of two sources of variability for an OR schedule. One source of variability is differences in the durations of cases of the same procedure. Another source of variability is variation in the workload scheduled each day. Click here for a comprehensive bibliography of OR management articles studying the economics of time reduction.
How can the revenue enhancement from reducing OR time and/or turnover times be calculated? Revenue enhancement from reducing OR time and/or turnover times applies only if lack of OR time is truly the bottleneck to care. Click here to download a lecture on the topic. Click here for the abstract of the most recent paper on the topic or click here for the full text. Although the formal topic of the paper is tactical decision making for selective expansion of OR time incorporating financial criteria and uncertainty in sub-specialties’ future workloads, the results are identical for reducing OR time while maintaining the same revenue.
Can I validly monitor turnover times or first case of the day starts by anesthesiologist? The validity of comparing average turnover times among anesthesiologists is limited. Click here [PDF] for details. First, 4 turnovers in a day are not N = 4 independent random samples of turnovers. Thus, correlations among turnovers need to be modeled statistically. Second, if an anesthesia group has 40 anesthesiologists, resulting in 40 confidence intervals to be compared to one another, the family-wise error rate needs to be controlled. Third, turnover times need to be adjusted for the preceding and following procedures and patients. Not only is something simplistic like “general surgery” absurdly insufficient because of heterogeneity in setup and cleanup times among procedures within a specialty, but turnovers often involve cleaning up from one specialty and setting up for a procedure of a different specialty. The usefulness of monitoring turnover times by individual anesthesiologist is also limited. Click here [PDF] for details. Suppose that each anesthesiologist were responsible for one OR. Then, decisions that reduce turnover times would always serve to increase OR efficiency or have no effect (i.e., would never be disadvantageous). However, this is not true when anesthesiologists are responsible for more than one OR (i.e., is not true for facilities making precisely the appropriate interventions to increase productivity – click here and click here). Thus, to encourage anesthesiologists to make decisions to increase OR efficiency, over-utilized OR time (i.e., OR efficiency) should be monitored, not turnover times. Better yet, individuals should be provided with real-time recommendations to focus attention on those turnovers that, if reduced, would serve to increase OR efficiency and/or to reduce surgeon waiting. The same principles apply to monitoring first case of the day starts by anesthesiologist. Click here [PDF] and click here [PDF] for details. That is, such procedures are of limited validity and usefulness.
Should we monitor first case of the day start times? The objective of improving on time starts may be to reduce over-utilized OR time. Click here [PDF] for an article describing a screening tool for the economic cost associated with late first case of the day starts. Click here [PDF] (and go to Table 7) for process of monitoring the expected impact of reducing the lateness of first case of the day starts on the efficiency of use of OR time. This should be done by combination of service and day of the week. There is consistently a > 500% difference among combinations of service and day of the week in the value of each 1 minute reduction in the average lateness of first cases of the day. This is why policies designed to achieve overall improvements in on time first case starts are generally counter productive. Surely do not monitor the percentage of late on time starts, use mean lateness (click here [PDF]). Click here [PDF] and click here [PDF] for articles describing that if these findings seem counter-intuitive, that is because of psychological bias. The objective of improving on time starts may be to reduce the tardiness of starts of subsequent cases. Click here [PDF] for an article on the psychological biases. Click here [PDF] for observational studies. Click here [PDF] for intervention.
Section: Miscellaneous Topics What are typical numbers of nursing staff per case ("skill mix") in the United States? Generally, in the US, there is one registered nurse and one surgical technologist or, in some cases, a second RN, per operating room. The circulator (RN) is busy drawing up medications, obtaining supplies and equipment, and so forth. The scrub nurse (surgical technologist or RN) is passing instruments, suture, and other supplies to the surgeons. When a laser is being used, during a liver transplant or burn case, when there is a second sterile table, and so forth, often then there will be another registered nurse in the theatre. For certain cases, there may be a nurse or physician assistant who is scrubbed and assists the surgeon. However, this is not my area of expertise. For more information on your question, click here to be forwarded to OR Manager. I serve on the advisory board of this monthly publication. They have some articles on staffing listed under "Favorite Articles." In their April 2001 issue, they report results of a survey evaluating skill mix for 19 types of procedures. You can subscribe to the publication on their website and receive an electronic copy monthly.
© 2003-20 10 Franklin Dexter. All Rights Reserved. Website feedback? |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
