Chapter One
INTRODUCTION
Statistics is a subject that for many people is pure tedium-a little bit like eating hay. For others, it is more likely to be anathema. The last thing they want to do in their life is have to take a course in statistics. Of course, there are those strange souls who find statistics interesting, even stimulating. But they are usually in the minority in any group.
This book is posited on the recognition that in the health field, as indeed among people in any discipline, there are at least these three different views of statistics, and that any statistics class is likely to be made up more of the former two groups than the latter. It is the goal of this book to provide an introduction to statistics in health policy and administration that will be relevant and useful, and perhaps finally interesting, to people in the first two groups, while still being challenging and informative to the people in the latter.
Section 1.1 How This Book Differs from Other Statistics Texts
The primary difference between this statistics text and most others is that this text uses Microsoft Excel as the tool for carrying out statistical operations and understanding statistical concepts as these relate to health policy and health administration issues. This is not to say that there are no other texts in statistics that use Excel. Levin, Stephan, Krehbiel, and Berenson (1999) have produced a very useable text entitled Statistics for Managers Using Microsoft Excel. But that book focuses almost exclusively on non-health-related topics. In many years of teaching statistics, especially to mid-career professionals, it is clear that the closer the applications of statistics are to the real-life interests and experiences, the more effective students will be in understanding and using statistics. Consequently, this book focuses its examples entirely on subjects that should be immediately recognizable to people in the health sciences.
Microsoft Excel, which most people will know as a spreadsheet program for creating budgets, comparing budgeted and expended amounts, and generally fulfilling accounting needs, is also a very powerful statistical tool. Chapter Two is devoted specifically to the ways in which Excel can be used as a statistical tool. Books that do not use Excel for teaching statistics (and, as has been said, this is most other books) generally leave the question of how to carry out the actual statistical operations in the hands of the student or the instructor. It is often assumed that relatively simple calculations, such as means, standard deviations, and t tests, will be carried out by hand or with a hand calculator. For more complicated calculations, the assumption is usually that a dedicated statistical package such as SAS, SPSS, STATA, or SYSTAT will be used. There are at least two problems with this approach that the current book hopes to overcome. First, hand calculations, or even the use of a hand calculator, can make the simple statistical operations overly tedious and prone to errors in arithmetic. Second, because dedicated statistical packages are designed for use rather than for teaching, they often obscure the actual process of calculating the statistical results, which comes between the student and an understanding of both how the statistic is calculated and what the statistic means.
In general, this is not true with Microsoft Excel. It is true that a certain amount of time in using this book must be devoted to the understanding of how to use Excel as a statistical tool. But once that has been done, Excel makes the process of carrying out the statistical procedures under consideration relatively clear and transparent. It is hoped that the student will end up with a better understanding of what the statistic means, through an understanding of how it is calculated, and not simply with the ability to get a result by entering a few commands into a statistical package. This is not to say that Excel cannot be used to shortcut many of the steps needed to get particular statistical results. As discussed in Chapter Two, a number of statistical tests and procedures are available as add-ins to Excel. However, using Excel as a relatively infallible, powerful, but transparent calculator can lead to a much clearer understanding of what the statistic means than that which can be obtained by other methods.
Section 1.2 Examples of Statistical Applications in Health Policy and Health Administration
In many iterations of teaching statistics to health policy and health administration students, the same question arises. Every semester sees students who say something like, "All these statistics are fine, but how do they apply to anything I am concerned with?" The question is not only a reasonable one, but it also points directly to one of the most important and difficult challenges for a statistics teacher, a statistics class, or a statistics text. How can it be demonstrated that these statistics have any real relevance to anything that the average person working in the health field ever needs to know or do? Happily, it has seemed that by the time a student has finished one of the courses that is the inspiration for this book, he or she usually sees how the knowledge of statistics can be useful. But it would be nice to be able to provide this kind of insight at the very beginning of a book, or course, as a way of getting rid of at least one stumbling block in the process of learning statistics.
To work toward a better understanding of why and when the knowledge of statistics may be useful to someone working in health policy or health administration, six examples have been selected of situations in which statistical applications can play a role. All six of these examples were inspired by real problems faced by students in classes in statistics, and they represent real statistical challenges that students have faced and hoped to solve. In virtually every case, the person who presented this problem recognized it as one that could probably be dealt with using some statistical tool. But also in every case, the solution to the problem was not obvious in the absence of some understanding of statistics. Although these case examples are not likely to resonate with every reader, perhaps they will give many readers a little better insight into why knowledge of statistics might be useful.
Documentation of Medicare Reimbursement Claims
The Pentad Home Health Agency provides home health services in five counties of an eastern state. The agency must be certain that its Medicare reimbursement claims are appropriately and correctly documented in order to ensure that Medicare will reimburse these claims in a timely manner. Appropriate documentation requires that all physician orders, including medications, home visits for physical therapy, home visits of skilled nursing, and any other orders for service be correctly documented on a form 485. Inappropriate documentation can lead to rejection or delay in processing of the claim for reimbursement by the Medicare administration.
The Pentad Agency serves about eight hundred clients in the five-county region. In order to assure themselves that all records are appropriately documented, the administration runs a chart audit of one in ten charts each quarter. The audit seeks to determine (1) whether all orders indicated in the chart have been carried out and (2) if they have been correctly documented in the form 485. Orders that have not been carried out, or orders incorrectly documented, lead to follow-up training and intervention appropriate to ensure that the orders and documentation are carried out correctly in the future.
Historically, the chart audit has been done by selecting each tenth chart, beginning at the beginning or at the end of the chart list. Typically, the chart audit determines that the majority of charts are correctly documented, usually 85 to 95 percent. But there are occasionally areas, such as skilled nursing care, where correct documentation may fall below that level. When this happens, the administration initiates an appropriate intervention.
One of the questions of the audit has been the selection of the sample. Because the list of clients changes relatively slowly, the selection of every tenth chart often results in the same charts being selected for audit from one quarter to the next; therefore, a different strategy for chart selection is desirable. It has been suggested that a strictly random sample of the charts might be a better way to select them for quarterly review, as this selection would have a lesser likelihood of resulting in a review of the same charts from quarter to quarter. But how does one go about drawing a strictly random sample from any population? Or, for that matter, what does strictly random actually mean and why is it important beyond the likelihood that the same files may not be picked from quarter to quarter? These are questions that are addressed by statistics, specifically the statistics associated with sample selection and data collection. They are the subjects of Chapter Three.
Another question related to the audit is the question of when to initiate an intervention. Suppose a sample of one in ten records is drawn (for eight hundred clients, that would be eighty records) and it is discovered that twenty of the records have been incorrectly documented. Twenty of eighty records incorrectly documented would mean that only 75 percent of the records were correctly documented. This would suggest that an intervention should be initiated to correct the documentation problem. But it was a sample of the eight hundred records that was examined, not the entire eight hundred. Suppose that the twenty incorrectly documented records were, by the luck of the draw, so to speak, the only incorrectly documented records in the entire eight hundred records. That would mean that only 2.5 percent of the cases were incorrectly documented.
If the intervention to correct the problem were expensive-a five-day workshop on correct documentation, for example,-the agency would not want to initiate that intervention when 97.5 percent of all cases are correctly documented. But how would the agency know from a sample what proportion of the total eight hundred cases might be incorrectly documented, and how would they know the likelihood that fewer than, say, 85 percent of all cases was correctly documented if 75 percent of a sample was correctly documented? This, again, is a subject of statistics and is discussed particularly in Chapter Five, which deals with probability.
Emergency Trauma Color Code
The emergency department (ED) of a university hospital was the site of difficulties arising from poor response time to serious trauma. Guidelines indicate that a trauma surgeon must attend for a certain level of trauma severity within twenty minutes and that other trauma, still severe, but less so, should be attended by a trauma nurse within a comparable time. Less serious trauma did not require such immediate response. In general, it had been found that the response time for the ED in the university hospital was more or less the same for all levels of severity of trauma-too long for severe cases and often quicker than necessary, given competing priorities for less severe cases.
The ED director knew that when a trauma case was en route to the hospital, a call was put in from the ambulance to the ED to indicate that the emergency was on the way. Part of the problem as perceived by the director of the ED was that the call-in did not differentiate the trauma according to severity. The ED director decided to institute a system whereby the ambulance attendants would assign a code red to the most severe trauma cases, a code yellow to less severe trauma cases, and no color code to the least severe trauma cases. The color code of the trauma would be made known to the ED as the patient was being transported to the facility. The object of this coding was to ensure that the most severe traumas were attended within the twenty-minute guidelines. This in turn was expected to reduce the overall time from admission to the ED to discharge of the patient to the appropriate hospital department (all trauma cases at the red or yellow level of severity are transferred from the ED to a hospital department).
A major concern of the director of the ED was whether the new system actually reduced the overall time between admission to the ED, treatment of the patient in the ED, and discharge to the appropriate hospital department. The director of the ED has considerable information about each ED admission going back a period of several months before the implementation of the new color coding system and six months of experience with the system after it was implemented. This information includes the precise time that each trauma patient was admitted to the ED and the time that the patient was discharged to the appropriate hospital department.
It also includes the severity of the trauma at admission to the ED on a scale of 0 to 75, gender, age, and, of course, whether the admission occurred before or after the color coding system was implemented. The ED director also has information about the color code assigned after the system was initiated that can generally be equated to the severity score assigned at admission to the ED. Trauma scoring 20 or more on the scale would be assigned code red, below 20, code yellow, and those not on the scale would not be assigned a color.
The question the ED director wishes to address is how she can use her data to determine whether the color-coding system has reduced the time spent by trauma victims in the ED before discharge to the appropriate hospital department. At the simplest level, this is a question that can be addressed by using a statistic called the t test for comparing two different groups. The t test is discussed in Chapter Nine. At a more complex level, the ED director can address the question of whether any difference in waiting time in the ED can be seen as related somehow to changes in severity levels of patients before or after the color coding scheme was introduced. She can also examine whether other changes in the nature of the people who arrived as trauma victims before and after the introduction of the color-coding scheme might be the cause of possible differences in waiting time, if these are found. These questions can be addressed by using regression analysis, which is presented in Chapters Eleven through Thirteen.
It might be useful at this point to mention two caveats to the use of statistics that apply directly to this example. The first of these caveats is that no statistical analysis may be needed at all if the difference in waiting time after the initiation of the color-coding scheme is clearly shorter than the waiting time before.
Continues...
Excerpted from Statistics for Health Policy and Administration Using Microsoft Excel by James E. Veney Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
