Employing Self-Directed Computer and Printed Resources to Reform a Medical School Epidemiology and Biostatistics Curriculum

This project identified self-directed learning resources, evaluated their utility, and suggested implementation strategies that can revise significant parts of a first-year medical Epidemiology and Biostatistics course. Computer resources and printed materials were researched for their usefulness in redesigning the curriculum, incorporating self-directed learning, and employing the use of computers. Some classroom contact can be replaced with or complemented by these new, self-directed materials. This can allow students to make more efficient use of their study time. However, no present product or combination of products was found to be appropriate for the majority of medical school needs. We expect that newer products will change this predicament in the near future.

Web-based search engines, such as Yahoo and Infoseek, were used to find web-related materials during summer semester, 1997. In addition, a question was posted on the medical education (Med Ed) listserver to inquire if any members knew of web-based learning tools for medical epidemiology or biostatistics. A visit to The Learning Center at the Cognitive Science Branch of the National Library of Medicine, a department that specializes in compiling web-based medical education tools, was the final step in locating programs and websites.

Most of the computer programs evaluated for this project were downloaded from websites or found through medical library searches, including the Health Sciences Consortium based in Chapel Hill, NC. Other programs were already designed for epidemiology or biostatistics courses at either West Virginia University School of Medicine or other medical schools.

Detailed searches of printed materials (both books and periodicals) for epidemiology and biostatistics were completed at the National Library of Medicine in Washington, D.C. and the library at the West Virginia University School of Medicine. Instructional materials from the Centers for Disease Control in Atlanta, GA were obtained through the Public Health Training Network (PHTN) and the Environmental Intelligence Service (EIS).

Dr. James Harner of the Department of Statistics and Computer Science at West Virginia University has created an excellent computer-based introductory statistics course that is easily adaptable for a medical epidemiology and biostatistics course. The program is available though a common website, and students can enter the specific module they are interested in. The modules contain text, Java-based dynamic examples, examples, and exercises that can be completed on computer and e-mailed to the professor. With additional development, this program could be the best answer for a medical school class. At this time, some of the modules that are completed or under construction are the following:

1. variable types
2. estimates of central tendency and confidence intervals
3. probability
4. discrete and normal distributions
5. central limit theorem
6. hypothesis testing

A program of this sort could also be used as a distance learning tool. A similar project is being completed by Andy Lentz, division administrator at the Johns Hopkins University School of Hygiene and Public Health Distance Education Division.

The home page for the Global Health Network includes information and links to relevant sites. It also includes an excellent on-line course called SUPERCOURSE: Epidemiology, the Internet and Global Health. The course includes over 40 interconnected lectures on the epidemiology or specific diseases complete, such as AIDS.

A non-interactive, comprehensive report on distance-based education.

Investigating an Outbreak: Pharyngitis in Louisiana, 3050-G. Centers for Disease Control, 1994 An excellent program that uses funny graphics and simple situations to teach complex concepts and real-life epidemiology. Through a fictionalized account of pharyngitis in a public school, the student learns the investigative steps for an outbreak, creating case definitions and questionnaires, how to interpret data, and how to apply attack rate and relative risk. The student may use EpiInfo (included) to compute the statistics or choose to have the program interpret the data.

The Draft Outbreak Report that students fill out at the end is a means of grading. The entire program takes a few hours, so it may take too long to be used in a class.This program can be downloaded free from http://www.cdc.gov/phtn.

Drug Study Evaluation for Pharmacy 255. West Virginia University Health

Sciences Center. This program was designed to complement an existing course at West Virginia University with a computer-based lab. It consists of a computerized textbook with clickable words, definitions, and concepts; there are also dynamic visuals and interactive questions and quizzes. The sections on how to read a study and what statistical tests to use are quite useful. Although most of the topics don’t go into enough detail for a first-year medical school course, this program is extremely well done and educational and could be added to or used as an introduction. Information on this program can be acquired through the West Virginia University School of Medicine.

Risk 97. Santos, J.L., Vio, F., and Albala, C. Available through Medical Education Online, 1996. This interesting program uses both static and interactive pages to illustrate biostatistical concepts. For example, the user can input their risk factors and see how their relative risk for heart disease rises. There is also an interactive two-by-two table that shows how sensitivity and specificity change as the number of cases and controls change. This program would be most useful if used by the professor as a demonstration. This program can be downloaded free from http://www.utmb.edu/meo.

EpiInfo. Centers for Disease Control. DoEpi. Centers for Disease Control. EpiInfo is a program designed by the CDC to be used by epidemiologists in the field to create databases, organize questionnaires, and compute statistical tests. Although DoEpi is an excellent tutorial for EpiInfo with many interesting and relevant step-by-step exercises, the program is too in-depth for use in a first-year medical school class. These programs can be downloaded free from http://www.cdc.gov/phtn.

Principles of Epidemiology. Howard Strasberg. Health Sciences Consortium, 1995., Understanding Biostatistics. Ronald P. Cody and Steven B. Just. Health Sciences Consortium, 1990. Neither of these commercial programs were evaluated for this project but may be of use in a medical school class. These courses can be ordered from HSC, 201 Silver Cedar Ct, Chapel Hill, NC 27514-1517 (919)942-8731 FAX (919)942-3689

The following materials are useful self-directed learning tools that could be incorporated into self-study or classroom discussion:

Centers for Disease Control. EIS Summer Courses. This is an excellent series of step-by-step exercises used by the CDC to teach Environmental Intelligence Officers about relevant topics. It is very well-designed, with clear objectives, probing questions, and modern examples. The exercises cover such topics as outbreaks, study design, confounders, biases, odds ratios, relative risk, attributable risk, causation, incidence, prevalence, two-by-two tables, surveillance, and vaccinations. These materials may be difficult to acquire through the CDC or EIS. Centers for Disease Control.

Principles of Epidemiology, 3030-G. A printed continuing medical education course distributed by the Public Health Training Network (1-800-41TRAIN). It does not cover the topics with enough depth for the course, especially biostatistics, but certain sections may be useful for the class. It is divided into six sections:

1. Agent, Host, Environment
2. Measures of Central Tendency
3. Statistical Measures Used in Epidemiology
4. Methods for Organizing Epidemiologic Data
5. Disease Surveillance
6. Investigation of Disease Outbreaks

Riegelman, Richard K. and Hirsch, Robert P. Studying a Study and Testing a Test: How to Read the Health Science Literature, 3rd ed. Boston: Little, Brown and Company, 1996. Although most of the book is not concise or complete enough for the class, the "Flaw-Catching Exercises" are excellent self-directed or classroom learning tools. In an easy-to-understand format, they include analyses of study design, interpretation of statistical tests, and application of rates.

The following books should be considered when choosing the text for the class:

Abrahamson, J.H. Making Sense of Data: A Self-instruction Manual on the Interpretation of Epidemiologic Data, 2nd ed. New York: Oxford University Press, 1994. A fascinating and brilliant text that is meant to be a complete self-directed course. It is unfortunately too in-depth for a first-year medical school class. The Alice in Wonderland quotes add a light-hearted flair.

Bland, Martin. An Introduction to Medical Statistics, 2nd ed. New York: Oxford University Press, 1995.Very complete but too in-depth for first-year medical students. The examples at the end of each chapter are excellent and would make very good classroom exercises.

Essex-Sorlie, Diane. Examination and Board Review: Medical Biostatistics and Epidemiology. Norwalk, CT: Appleton and Lange, 1995. The simplest board review book that covers every topic.

Fletcher, Robert H., Fletcher, Suzanne W., and Wagner, Edward H. Clinical Epidemiology: The Essentials, 3rd ed. Baltimore: Williams & Wilkins, 1996. The best overall text from a medical student perspective. Logical order of presentation, and every topic is related to clinical experience.

Gordis, Leon. Epidemiology. Philadelphia: W.B. Saunders, 1996. A well-written and readable text. The epidemiology section covers unusual but relevant topics such as disease, intervention, cause, and policy.

Ingelfinger, Joseph A. Biostatistics in Clinical Medicine, 3rd ed. New York: McGraw-Hill, 1994. A fascinating approach for presenting the material: Each concept is illustrated with a recent, striking example from the literature that is expanded through each chapter; however, it may not be comprehensive enough as a sole text. It covers most of biostatistics, but it only covers epidemiology necessary for use in clinical medicine (e.g. clinical trials and meta-analysis.)

Lilienfeld, David E. and Stolley, Paul D. Foundations of Epidemiology, 3rd ed. New York: Oxford University Press, 1994. This is the text that has been used in the past for the course at West Virginia University School of Medicine. Although it has some good sections, it does not cover much biostatistics and is rather dry.

Morton, Richard F., Hebel, J. Richard, and McCarter, Robert J. A Study Guide to Epidemiology and Biostatistics, 4th Ed. Gaithersburg, MD: Aspen Publishers, 1996. The best quick and comprehensive review of medical biostatistics and epidemiology. It is extensive, concise, and well-designed, but has no textbook-like explanations. Each topic includes sample test questions.

All of the competencies listed in the curriculum can be met by using the resources discussed above, but no single resource and certainly no self-directed or web-based resource was presently adequate to do the job alone in 1997. The World Wide Web is a voluminous encyclopedia of information, and students in the course can use it to find facts and download programs. However, most useful materials are not posted on the web in their entirety--if at all--because the authors do not profit from it. Programs that are not based on the web but which are obtainable in printed form (often for free) are usually more applicable to the course.

The web may also be used to connect students in the class, such as in the program under development for statistics by Dr. Harner at West Virginia University. Through a common website, a privately-designed program can be disseminated easily to the students as well as allowing student and professor to communicate, complete homework sets, and grade the student’s work through e-mail.

Many of the non-web-based computer programs incorporate the useful features of a computer--interactivity, connected links--with the text of a printed book. Currently, this seems to be the most useful type of program for a medical student. These same materials could be made available to students through the World Wide Web, either as downloadable programs or as a website. Many of the non-web-based programs are designed with programs such as Hypercard or Authorware and can be easily written and amended.

Some excellent self-directed learning tools exist as printed materials, and they should be used as background texts and explanatory exercises. However, similar materials have been developed for the computer that take advantage of its unique interactivity with the user. As soon as feasible, these computer-based tools should be the basis of the course.

All references are included in the body of the text.

This project was completed by Adam Cohen, MSIII with the supervision of Alan Ducatman, MD, MS in the Department of Occupational and Environmental Medicine. Other faculty and staff that were of great help were James Shumway, Ph.D.; Gerry Hobbs, Ph.D.; James Harner, Ph.D.; Sandra Magnetti, Ph.D.; Charles Muntaner, MD; Bill Shockor, MD; and R. John Pearson, MD, MPH, all of West Virginia University School of Medicine. I would also like to thank Rita Washko, MD of the National Institutes for Occupational Safety and Health for the EIS training materials, Cathy Slemp of the West Virginia Division of Surveillance and Disease Control, and Suzanne Stensaas, Ph.D. of Cornell University Medical College.