Simulation Modeling and Health-care Decision Making

Banks J. Verifying and validating complex simulation models by analogy. Simulation. Jan 1990;54(1):33-6. Discussion of a procedure for developing approaches for verification and validation of large simulation models.

Christy DP , Watson HJ The application of simulation: a survey of industry practice. Interfaces. 1983;13(5):47-52. A look at some of the ways simulation is actually used in industry.

Dittus RS , Roberts SD , Wilson JR Quantifying uncertainty in medical decisions. J Am Coll Cardiol. Sept 1989;14(3 Suppl A):23A-28A. Overview of approaches to quantifying uncertainty. Compares simulation with other techniques.

Keller L. , Harrell C. , Leavy J. The three reasons why simulation fails. Industrial Engineering . April 1991;23(4):27-31. Discussion of techniques for developing and communicating simulation models and their results.

Law AM , McComas MG Pitfalls to avoid in the simulation of manufacturing systems. Industrial Engineering. May 1989;21(5):28 - 31 +. Eleven common simulation mistakes are described. Most would apply to health care as well as manufacturing.

Law AM , McComas MG Secrets of successful simulation studies. Industrial Engineering . May 1990;22(5):47-53 +. An outline of the steps required to develop a successful simulation model.

Mahachek AR An introduction to patient flow simulation for health-care managers. Journal of the Society for Health Systems. 1992;3(3):73-81. A basic description of when a manager should consider simulation followed by a discussion of simulation terminology, the resources needed to conduct a simulation, and the possible results that can be anticipated.

Roberts SD , Klein RW Simulation of medical decisions: applications of SLN. Simulation . Nov 1984;43(5):234-41. Description of medical decision-making simulation applications and issues surrounding them.

Rossi JA An informatics approach to complex research problems. Dynamic process modeling . Computers in Nursing. Jan-Feb 1991;9(1):7-14. Provides basic introduction to simulation terminology and random variates. Discusses validation and gives examples of how simulation can be useful to nurse researchers.

Weinstein MC Methodologic issues in policy modeling for cardiovascular disease. J Am Coll Cardiol. Sept 1989;14(3 Suppl A):38A-43A. Contrasts economic-evaluation models and population-simulation models. Describes key methodologic issues and problems related to both.

Estrine E. 1991 Directory of Simulation Software. La Jolla, CA: Society for Computer Simulation, 1991. ISBN:0-911801-94-4. An annual compilation of software packages for simulation applications. Contains brief descriptions and approximate costs obtained from the vendors. All types of simulation software other than simulators are included. Over 115 listings, cross-referenced by keywords.

Fries BE Applications of Operations Research to Health Care Delivery Systems: A Complete Review of Periodical Literature. New York: Springer-Verlag, 1981. A review of published applications of operations research, including simulation, to health-care delivery. Organized by application area.

Hooper JW Strategy-related characteristics of discrete-event languages and models. Simulation. April 1986;46(4):153-9. A comparison of common simulation languages. Discussion of different world views with respect to "next-event" scheduling strategies.

Pollacia LF A survey of discrete event simulation and state-of-the-art discrete event languages. Simulation Digest. Fall 1989;20(3):8-25. A history and classification of discrete-event simulation software through 1988. Emphasis on new trends such as special-purpose features, graphics capabilities, and artificial intelligence concepts.

Roberts SD , England WL Survey of the Application of Simulation to Health Care. La Jolla, CA: Society for Computer Simulation, Simulation Series 10(1), 1981. A comprehensive bibliography (over 400 references) of simulation in health-care publications through 1980. Good discussion of the many ways in which simulation can be used in health care.

Swain J. World of choices: simulation software survey. OR/MS Today . Oct 1991;18(5):81-102. Brief summaries of 56 discrete-event simulation products for microcomputers or workstations.

Valinsky D. Simulation. In: Shuman LJ , Speas RD , Young JP , eds. Operations Research in Health Care. Baltimore : Johns Hopkins University Press, 1975. ISBN:0-8018-1642-4. A good introduction to simulation and health care, included primarily for its references to pre-1975 applications.

Banks J. , Carson JS Discrete-event System Simulation. Englewood Cliffs, NJ : Prentice-Hall, 1984. ISBN:0-13-215582-6. Primarily an undergraduate industrial engineering text. Examples using FORTRAN, Simscript, GPSS, and SLAM focus on queueing and inventory problems. Besides overview of discrete-event simulation (especially event scheduling), there are sections on mathematical models that are useful in simulation, random numbers, and input and output analysis, including validation.

Boldy D. Operational Research Applied to Health Services. New York : St. Martin's Press, 1981. ISBN:0-312-58682-5. An anthology describing international applications of operations research to strategic and tactical planning. Provides much information and references about the Health Care System Model component of the world model of the International Institute for Applied Systems Analysis. Also has applications of simulation to radiology staffing and ambulance location.

Bratley P. , Fox BL , Schrage LE A Guide to Simulation, second edition. New York: Springer-Verlag, 1987. ISBN:0-38-790820-X. Strictly a graduate engineering or computer science text. Very comprehensive treatment of variance reduction and output analysis.

Bulgren WG Discrete System Simulation. Englewood Cliffs, NJ: Prentice-Hall, 1982. Overview of discrete-event simulation. Comparison of programming in GPSS and Simscript.

Carroll JM Simulation Using Personal Computers. Englewood Cliffs, NJ: Prentice-Hall, 1987. ISBN:0-8359-6924-X. A book for self-teaching simple uses of computer simulation. Includes many examples of simulations written in Basic and tips for programming such models. One chapter describes using GPSS.

Davies RM , O'Keefe RM Simulation Modeling with Pascal. New York: Prentice-Hall, 1989. ISBN:0-13-811571-0. After five overview chapters, a hospital simulation is used to demonstrate more advanced concepts. Suitable for a one-semester course if a Pascal-based language will be used.

Delaney W. , Vaccari E. Dynamic Models and Discrete Event Simulation. New York : Marcel Dekker,1989. ISBN:0-8247-7654-2. Intended to be an undergraduate text for engineers or computer scientists. Covers deterministic and discrete-event simulation. Unique chapters on software engineering and artificial intelligence as they relate to simulation. Contains FORTRAN code and description of a hospital simulation example. More what and why than how to—authors are physicists.

Hoover SV , Perry RF Simulation: A Problem-solving Approach. Reading, MA: Addison-Wesley, 1989. ISBN:0-201-16880-4. Suitable for a comprehensive first course in simulation model building in engineering or computer science. Gives examples and language primers in three languages (GPSS, SIMAN and Simscript). Has a problem-solving orientation.

Ingels DM What Every Engineer Should Know about Computer Modeling and Simulation. New York: Marcel Dekker, 1985. ISBN:0-82-477444-2. Contains many lists, such as steps for verifying and validating and uses for simulation. Material is suitable for presentation at an introduction to simulation workshop.

Kreutzer W. System Simulation Programming Styles and Languages. Sydney , UK: Addison-Wesley, 1986. ISBN:0-201-12914-0. Promotes structured programming approach to simulation using primarily Pascal and SIMULA. Discusses Monte Carlo, continuous, queueing, and combined systems. Contains excellent annotated bibliography of 30 older simulation texts.

Law AM , Kelton WD Simulation Modeling and Analysis, second edition. New York : McGraw-Hill, 1991. ISBN:0-07-036698-5. A comprehensive language-independent introduction to simulation. Suitable for an upper-level undergraduate engineering or management course.

Lewis Paw , Orav EJ Simulation Methodology for Statisticians, Operations Analysts, and Engineers . Pacific Grove, CA: Wadsworth and Brooks, 1989. ISBN:0-534-09450-3. After four introductory chapters, the focus is on statistical techniques for data analysis at a graduate statistics level. Much emphasis on problems in mathematical statistics. Simulation is treated as a controlled statistical sampling technique. Very mathematical treatment of variance reduction. A second volume for only the mathematically sophisticated.

McHaney RW Computer Simulation: A Practical Perspective. Harbor Springs, MI: Academic Press, 1991. ISBN:0-12-484140-6. A practical look at what simulation can do today. Includes new topics such as concept modeling, logic transfer, and animation. Contains many examples.

Mitrani I. Simulation Techniques for Discrete Event Systems. Cambridge , UK: Cambridge University Press, 1982. ISBN:0-521-28282-9. A short text with more attention to programming issues, such as basic approaches to programming discrete-event simulations. Compares early versions of GPSS, Simscript, and SIMULA. Includes a SIMULA primer as an appendix.

Morgan Bjt. Elements of Simulation. London, UK: Chap-man and Hall, 1984. ISBN:0-412-24590-6. Simulation viewed as an operations research tool and a mathematical method. The focus is on the uses of simulation in statistics. The only programming is Basic. Primarily Monte Carlo simulation.

Neelamkavil F. Computer Simulation and Modelling. Chichester, UK: John Wiley & Sons, 1987. ISBN:0-47-191129-1. Introductory overview text. Good balance among continuous, discrete-event, and Monte Carlo methods. Examples in three languages.

Pidd M. Computer Simulation in Management Science. Chichester, UK: John Wiley & Sons, 1984, ISBN:0-471-90281-0. Directed toward management science students, but also useful to anyone with some knowledge of programming but no experience building simulations. Provides an overview of simulation, an introduction to tools, techniques, and terminology for building discrete-event simulations, and a unique discussion of the principles of modeling feedback systems and the systems dynamics approach.

Ripley BD Stochastic Simulation. New York: John Wiley & Sons, 1987. ISBN:0-47-181884-4. Statistical perspective to simulation at beginning graduate student level. Excellent look at random-number and random-variate generation. Adequate treatment of output analysis.

Smalley HE Hospital Management Engineering: A Guide to the Improvement of Hospital Management Systems. Englewood Cliffs, NJ: Prentice-Hall, 1982. ISBN:0-13-394775-0. The definitive book on hospital industrial engineering. Contains case studies of simulations of clinic queues, forecasting product requirements, staffing, and scheduling.

Smith JM Mathematical Modeling and Digital Simulation for Engineers and Scientists . New York: John Wiley & Sons , 1987. ISBN:0-47-108599-5. Emphasis on fast numerical methods, continuous simulation, chaos. Directed toward graduate electrical engineering and computer science students.

Szymankiewicz J. , McDonald J. , Turner K. Solving Business Problems by Simulation, second edition. London, UK: McGraw-Hill, 1988. ISBN:0-07-084946-3. A "teach-yourself text" presenting a practical modeling approach. Covers both discrete and continuous processes. A lot of emphasis on visual presentation of results. Uses HOCUS.

Thesen A. , Travis LE Simulation for Decision Making. St. Paul, MN: West Publishing Company, 1992. ISBN:0-314-83549-0. Integrates hands-on computer skills and statistical issues. Comes with a disk containing student versions of GPSS/PC, Sandie (for data analysis), and TBS-II (a template-based modeling system). The text is at an introductory level for undergraduate business or engineering. Statistics are used throughout without theorems or proofs.

Warner DM , Holloway DC Decision Making and Control for Health Administration. Ann Arbor, MI: Health Administration Press, 1978. ISBN:0-91-490424-8. Most of the chapter on complex stochastic analysis is devoted to simulation. It contains detailed examples of simulation models to find appropriate numbers of beds and admissions scheduling.

Widman LE , Loparo KA , Nielsen NR Artificial Intelligence, Simulation, and Modeling. New York: John Wiley & Sons, 1989 . ISBN:0-471-60599-9. A compilation of 20 articles relating simulation and artificial intelligence. These include an introductory survey, eight basic conceptual articles, three using simulation to enhance artificial intelligence, and eight applying artificial intelligence to enrich simulation. The article on the nature of modeling and several on qualitative or semiquantitative systems provide unique insight.

Butler TW , Reeves GR , Karwan KR , Sweigart JR Assessing the impact of patient care policies using simulation analysis. J Soc Health Systems. 1992;3(3):38-53. Describes results of a Simscript IL5 model of inpatient room placement. A variety of scenarios are examined as to their effects on numbers of patient room transfers and room type misplacements.

Cox TF , Birchall JP , Wong H. Optimising the queueing system for an ear, nose and throat clinic. J Appl Statistics. 1985;12(2):113-26. Clinic scheduling model in FORTRAN. Describes data collection and assumptions. Provides many output graphs.

Davies H. , Davies R. A simulation model for planning services for renal patients in Europe. J Oper Res Soc. Aug 1987;38(8):693-700. More details about the discrete-event renal model introduced in the other Davies articles. Description of enhancements to PascaLSIM that allows their model to interact in real time with the European Dialysis and Transplant Association Registry database. Discussion of other features desired for a complete decision-support system.

Davies R. An assessment of models of a health system. J Oper Res Soc . Aug 1985;36(8):679-87. Expounds the virtues of discrete-event simulation for comparing policy options for patient treatment. Characterizes a good model. Presents a model for treatment of patients with kidney failure in a renal unit. Critiques all previous such models with respect to the stipulated characteristics of a good model.

Davies RM An interactive simulation in the health service. J Oper Res Soc. July 1985;36(7):597-606. A discrete-event model of renal failure treatments and their effects on bed, dialysis unit, and staffing needs. Written in Pascal for Apple computers. The model structure, input and output data, validation, and the advantages of interactive microcomputer simulation are discussed.

Dumas MB Hospital bed utilization: an implemented simulation approach to allocating and maintaining appropriate levels. Health Serv Res. April 1985;20(1):43-61. Many results are presented. Good companion to other Dumas article in Simulation.48

Dumas MB Simulation modeling for hospital bed planning. Simulation . Aug 1984;43(2):69-78. A SIMSCRIPT model for bed allocation among hospital services. Good discussion of the model structure and the outcome measures.

Hancock WM , Walter PF The use of admissions simulation to stabilize ancillary workloads. Simulation. Aug 1984;43(2):88-94. An example of simulation of admission scheduling for reducing workload variability. Uses existing data to drive a PL/I network flow model.

Hannan EL , Gimbrone CJ Predicting the impact of instituting a priority readmission policy in nursing homes. J Comput Oper Res. 1987;14(6):493-505. A FORTRAN model of the macro view of patient flow between hospitals and nursing homes. Includes model description, validation, and results.

Harris RA Hospital bed requirements planning. Eur J Oper Res. April 1986;25(1):121-6. Discrete event Basic model that examines the relationships between surgery schedules, lengths of stay, and bed allocations. Its applications for planning are discussed.

Ishimoto K. , Ishimitsu T. , Koshiro A. , Hirose S. Computer simulation of optimum personnel assignment in hospital pharmacy using a work-sampling method. Med Informatics . Oct-Dec 1990;15(4):343-54. An excellent example of data collection and analysis. The application is perfect for discrete-event simulation, but a fixed time advance was used instead. Other unnecessary assumptions limit the stochastic nature of the simulation and thus make the results suspect.

Kenvin JC , Sommerfeld JT Discrete-event simulation of large-scale poliomyelitis vaccine production . Process Biochemistry. June 1987;22(3):74-7. A GPSS batch-production model. Application for production scheduling, equipment and labor utilization.

Lambo E. An optimization-simulation model of a rural health center in Nigeria. Interfaces. June 1983;13(3):29-35. Discusses advantages and disadvantages of linear programming optimization models and simulation models for staffing and personnel allocation problems. Uses a model designed in ECSL producing FORTRAN. Model details and numerical results not included, but has a good list of operational variations that may improve efficiency.

Levy JL , Watford BA , Owen VT Simulation analysis of an outpatient services facility. J Soc Health Systems. Nov 1989;1(2):35-49. Uses a SIMAN model to design part of a clinic facility. Good example of focusing on the results of interest.

Liu SL , Lee JT A simulation of a hospital emergency call system using SLAM II. Simulation. Dec 1988;51(6):216-21. Location and speed of ambulances and numbers of hospital beds needed in Taiwan are obtained from a SLAM II model with FORTRAN extensions. Extends the work of Uyeno and Seeberg.65

Lowery JC , Martin JB Design and validation of a critical care simulation model. J Soc Health Systems. 1992;3(3):15-36. Describes development and excellent validation of a GPSS/H model of critical care bed usage. Contains extensive review of bed-sizing literature.

MahachekAR , Knabe TL Computer simulation of patient flow in obstetrical/gynecology clinics. Simulation. Aug 1984;43(2):95-101. A GPSS model finds serious flaws in both space and staffing requirements for a proposed operational change. Describes the education value of simply gathering and organizing the data needed to build a simulation model.

Mukherjee AK A simulation model for management of operations in the pharmacy of a hospital. Simulation. Feb 1991;56(2):91-103. A good application using SIMAN to investigate pharmacist and technician utilization. Could have been improved with better insight into input modeling, rather than assuming uniform distributions and using goodness of fit for weak support of the assumption.

Pallin A. , Kittell RP Mercy Hospital: simulation techniques for ER processes. Industrial Engineering. Feb 1992;24(2):35-7. A good discussion of the steps involved in creating a model that is a permanent resource for a hospital rather than a quick tool to answer a few questions. The process modeled and the objectives are described in detail, but description of the actual GPSS/H model is scant.

Romanin-Jacur G. , Facchin P. Optimal planning of a pediatric semi-intensive care unit via simulation. Eur J Oper Res. May 1987;29(2):192-8. Uses a combination of FORTRAN and GPSS to model changes in a pediatrics ward. Both the number of beds needed and the number and schedules of nurses are examined.

Shuman LJ , Wolfe H. , Gunter MJ RURALSIM: the design and implementation of a rural emergency medicine service simulator. J Soc Health Systems. 1992;3(3):54-71. An overview of more than ten years of applications of a general emergency service simulator developed in SIMULA. Conclusions from the modeling effort and limitations of the model are clearly described.

Taket AR Equity and access: exploring the effects of hospital location on the population served—a case study in strategic planning. J Oper Res Soc. Nov 1989;40(11):1001-10. An attraction-constrained spatial-interaction model (continuous) for facility location.

Torn A. The simulation net approach to modelling and simulation. Simulation. Sept 1991;57(3):196-8. A technical note that uses a hospital patient treatment system as the example to demonstrate a modeling technique.

Uyeno DH , Seeburg C. A practical methodology for ambulance location. Simulation . Aug 1984;43(2):79-87. Using simulation to fine-tune (as third step) ambulance location. Discussion of the many "what-if" questions a validated model can address.

Vassilacopoulos G. Allocating doctors to shifts in an accident and emergency department. J Oper Res Soc. Jun 1985;36(6):517-23. FORTRAN simulation used to verify a dynamic programming solution to allocating staff to an emergency room. Determines that results are extremely sensitive to patient arrival process— simulation uses piecewise constant time-varying Poisson process.

Vassilacopoulos G. A simulation model for bed allocation to hospital inpatient departments. Simulation. Nov 1985 ;45(5):233-41. A FORTRAN model for bed allocation. Extensive discussion of the data collection and model parameterization. Possible extensions to the model and illustrations of its use are included.

Vemuri S. Simulated analysis of patient waiting time in an outpatient pharmacy. Am J Hosp Pharm. June 1984;41(6):1127-30. GPSS model of waiting time at a hospital outpatient pharmacy. Discussion of data collection and statistical design of simulation experiments to evaluate cost per waiting time saved for various staffing levels.

Wears RL , Winton CN Load and go versus stay and play: analysis of prehospital i.v. fluid therapy by computer simulation. Ann Emerg Med. Feb 1990;19(2):163-8. Uses the INSIGHT simulation language to put variability into a deterministic model of prehospital intravenous fluid therapy. Has an excellent discussion of distributions used to parameterize the model.

Wilt A. , Goddin D. Health care case study: simulating staffing needs and work flow in an outpatient diagnostic center. Industrial Engineering. May 1989;21(5):22-6. Case study of facility design before construction. Uses the Micro SAINT language.

Wood R. A simulation study of the Westgard multi-rule quality-control system for clinical laboratories. Clin Chem. Mar 1990;36(3):462-5. An APL Monte Carlo model. Demonstrates the effects of realistic skewness on quality control model decisions.

Wright MB The application of a surgical bed simulation model. Eur J Oper Res. Oct 1987;32(1):26-32. A Pascal discrete-event model of surgical bed requirements under a variety of different schedules. Good discussion of input required and assumptions used.

Dasbach EJ , Fryback DG , Newcomb PA , Klein R. , Klein BE Cost-effectiveness of strategies for detecting diabetic retinopathy. Med Care. Jan 1991;29(1):20-39. Adds strategies of care to embedded Markov model of disease progression. Evaluates screening strategies using cost per sight-year.

Goldman L. , Weinstein MC , Williams LW Relative impact of targeted versus populationwide cholesterol interventions on the incidence of coronary heart disease. Projections of the Coronary Heart Disease Policy Model. Circulation. Aug 1989;80(2):254-60. Describes the use of the demographic-epidemiologic component of the state-transition-simulation heart disease policy model to investigate cholesterol interventions. Results indicate that population-wide programs may offer advantages over targeted treatment. The paper is balanced between methods and results; complete model details may be found in Weinstein et al.81

Habbema Jdf , van Oortmarssen GJ , Lubbe Jtn , van der Maas PJ The MISCAN simulation program for the evaluation of screening for disease. Comput Meth Progr Biomed. 1984;20:79-93. Introductory discussion of the MISCAN Monte Carlo simulation program. Describes the two parts (Disease and Screening), input needed, and output available. Short discussion of breast cancer and cervical cancer screening examples included.

Habbema J. , Lubbe J. , van Oortmarssen GJ , van der Maas PJ A simulation approach to cost-effectiveness and cost benefit calculations of screening for early detection of disease. Eur J Oper Res. 1987;29:159-66. Focus is on the extensions to the MISCAN simulation program that enable full cost-benefit and cost-effectiveness calculation. Uses cervical cancer screening as an example. Allows for and discusses the implications of discounting.

Milsum JH On the optimization of cholesterol screening. Meth Inform Med. 1991;30(1):36-43. Basically a stochastic decision tree in a spreadsheet. Simplified model for cholesterol screening. Good discussion of necessary input data.

Screenivas V. , Prabhakar AK , Ravi R. , Luthra UK A simulation approach for estimating the loss of woman-years due to cervical cancer and probability of developing cervical cancer. Neoplasma. 1989;36(5):623-7. Demonstrates how much information can be obtained from a simplified "quick & dirty" model. Uses Basic language and random-number generator.

Tsevat J. , Weinstein MC , Williams LW , Tosteson AN , Goldman L. Expected gains in life expectancy from various coronary heart disease risk factor modifications. Circulation. Apr 1991 ;83(4):1194-201. Describes use of a state-transition model to examine long-term population-wide effects of various interventions, such as cholesterol levels. Balanced between methods and results. Uses the demographic-epidemiologic submodel of the Coronary Heart Disease Policy Model described by Weinstein et al.81

van Oortmarssen GJ , Habbema Jdf , van der Maas PJ , de Koning HJ , Collette Hja , Verbeek Alm , Geerts AT , Lubbe Ktn. A model for breast cancer screening. Cancer. Oct 1990;66(7):1601-12. Gives results from MISCAN, a state transition model. Displays many validation plots and describes parameterization in an appendix.

Weinstein MC , Coxson PG , Williams LW , Pass TM , Stasson WB , Goldman L. Forecasting coronary heart disease incidence, mortality, and cost: the coronary heart disease policy model. Am J Public Health. 1987;77:1417-26. Detailed descriptions of the three components of the coronary heart disease policy model. Good discussions of simplifying assumptions and their consequences. Specific data assumptions and references are in appendices.

Weissfeld JL , Weissfeld LA , Holloway JJ , Bernard AM A mathematical representation of the expert panel's guidelines for high blood cholesterol case-finding and treatment. Med Decis Making. Apr-Jun 1990;10(2):135-46. A computer simulation of the National Cholesterol Education Program protocol for high cholesterol case-finding and treatment. The model deals with the calculation of probabilities at chance nodes that are dependent on sequential measurements. Modeling the uncertainty related to the variability of the effects of diet and drugs on HDL C is also discussed in detail.

Ahlgren DJ , Stein AC Dynamic models of the AIDS epidemic. Simulation. Jan 1990;54(1):7-20. Epidemiologic systems dynamics with STELLA. Provides good introduction to STELLA templates and language. Includes extensive bibliography of AIDS-related epidemiology and systems-dynamics literature.

Bloom BS Medical management and managing medical care: the dilemma of evaluating new technology. Am Heart J. Mar 1990;119(3 Pt 2):754-61. Simulation from a population perspective. Results show what a model can do, but few details of the specific model are included.

Bongaarts J. A model of the spread of HIV infection and the demographic impact of AIDS. Statistics Med. Jan 1989;8(1):103-20. Very detailed mathematical examination of a complicated Markov model of HIV spread. Good discussion of assumptions.

Finkelstein SN , Homer JB , Sondik EJ Modelling the dynamics of decision-making for emerging medical technologies. R & D Management. July 1984;14(3):175-91. Systems dynamics using the Dynamo language to model generic emerging medical technologies. Contains a detailed case study of PCTA as an emerging technology. Provides extensive references to systems-dynamics literature.

Gebski V. , McNeil P. , Coats A. , Forbes J. Monitoring distributional assumptions and early stopping. Statistics Med. Sep 1987;6(6):667-78. Monte Carlo model to "complete" censored data from partially completed clinical trials. Allows assessment of assumptions made in sample-size calculations, thus improving stopping rules for patient accrual.

Gheorghe FC , Stefanescu S. Simulation models for the study of the active population evolution and of the health state of entire population. Econ Computation Econ Cybernet Stud Res. 1984;19(4):37-46. A theoretical look at modeling active versus invalid populations using absorbing Markov chains. The algorithms for simulation programs (with some stochastic components) implementing the theory are presented.

Homer JB A diffusion model with application to evolving medical technologies. Technol Forecasting Social Change. May 1987;31(3):197-218. A systems-dynamics model that addresses both the adoption and the changing extent of uses of evolving medical technology. It shows influence diagrams and case-study results for pacemakers and the antibiotic clindamycin.

Jezek Z. , Grab B. , Dixon H. Stochastic model for inter-human spread of monkeypox. Am J Epidemiol. Dec 1987;126(6):1082-92. Monte Carlo epidemiology model in FORTRAN. Good discussion of validation and appropriate application of the model.

Leslie WD , Brunham RC The dynamics of HIV spread: a computer simulation model. Comput Biomed Res. Aug 1990;23(4):380-401. Uses a Simscript model. Details the differences in assumptions between this model and deterministic math models. Demonstrates how variability makes models predictive only for large populations.

McKusick KB , Schach SR , Koeslag JH Social mechanisms in the population genetics of Tay-Sachs and other lethal autosomal recessive diseases: a computer simulation model. Am J Med Genet. June 1990;36(2):178-82. C-language simulation of population genetics. Examines effects of marital customs and reproductive compensation on genetics of lethal recessive genes.

Stolwijk JA , Canny PF Simulation model of lung cancer incidence related to smoking and radon daughter exposure. Int J Epidemiol. 1990;19(Sup 1):S73-8. This is a SAS model that randomly creates a population, uses risk equations to generate disease groups and calculates the relative risk for a population. The model is described in detail in an appendix.