The amount of activity in open source medical software has stepped up considerably over the last months. Several projects, among them FreePM, GNUmed, Freemed and TK_familypractice are moving forward at a good pace. Many are at the alpha stage, with planned beta’s in 2001. What is all this activity for, one may ask, particularly if new to open source medical software. Where is it all going? What is the problem to be solved? A vision of what is beyond the sunset for open source medical software is in order.

30 years is a long time in computing, yet that is how long efforts to computerize medicine have been in existence. A local survey of hospitals in Texas reveals that those efforts have largely failed. Few use clinical computing software beyond obtaining lab results. A ubiquitous, standard computing platform for all aspects of medicine does not currently exist. The usual commercial model of locking customers into a proprietary system that has failed for so many years to produce widespread, robust clinical software, continues to fail to produce a comprehensive, beneficial and economically feasible system.

Fine, but what are we missing by not having widespread open source medical software? Let’s look at what it could look like, beginning in medical school.

Guadalupe Cornelius Chen would get her white coat, stethoscope, fingerprint scan and wireless computer on her first day at school. After the initial speeches by the dean and faculty, she would report to her first class which would consist of a bare room with only a comfortable chair and a light. The lights dim and the walls become alight with images so that the student finds herself inside a convincing illusion of a hospital hallway. After fumbling with controls on her wireless computer she makes herself ‘walk’ through the simulated hallways of the hospital. She is greeted as ‘Dr. Chen’ and is shown to her first patient: a diabetic.

She is speechless at first. After time has elapsed, a kindly attending walks into the room and instructs her as to the proper questions to ask. The student then finds appropriate practice guidelines on her wireless computer based on real world, constantly gathered outcome data sponsored by the National Institute of health, which is easily accessible through an open source client that has been customized and extended for the simulator. The data has made it possible to tailor the treatment to the patient profile that Ms. Chen has gathered and are most likely to achieve success for this patient. The computer even suggests a few questions to ask that will greatly improve the chances of success for both diagnosis and treatment. This particular module was built by a consortium of medical schools and would have been very difficult or very limited if the clinical programs at its base had not been adaptable for this non-commercial purpose.

Dr. Chen slowly keys in the required documentation for the patient and moves on to the next patient. She ends the first patient encounter by ordering the proper medications electronically in a way that is virtually guaranteed to be error free and automatically chooses the most cost-efficient medication. The drug ordering module, open sourced years before, is constantly improved by government, insurance agencies and drug companies since it has been shown to drastically reduce errors and the price of delivering care.

Six months later, the student begins to see a few real patients, but she has ‘treated’ approximately 3000 simulated patients. Faculty have followed her progress through the simulator and have verified her progress with the real patients. She has ‘killed’ 50 and has been ‘sued’ 3 times. She is now achieving a sustained rate of treating 40 general practice patients a day with an error rate lower than any clinician today. Her ‘attending’ must intervene only a few times a day now. In six more months, she will be treating 50 simulated patients a day and will not be confined to a single specialty. This high rate of patient processing can be done because time wasters such as an illegible chart, hunting for the chart and associated paper work as well as redundant drug data ordering is gone. The work is still difficult, but her fine brain is working on problems worthy of it, diagnosis and treatment, not mechanics.

The patients Dr. Chen is seeing are cases that have been automatically built based upon actual local cases in the teaching hospital or clinic which the computer. The ability to extract these cases is possible because the University computer system is all open source and modular so that a data stream can be easily built from the University system and diverted to the simulator. It will be sanitized (by an open source module originally intended for researchers) for confidential data but still retaining locally important issues.

Dr. Chen only feels it necessary to call her ‘attending’ once a week now with the simulator giving her constant feedback on her performance. No useless lectures, no esoteric research topics, the entire time has been spent efficiently learning how to take care of patients using standard open source technology. She has even been able to build a few interesting cases herself that were accepted for inclusion in a national database.

During her past medical school year, medical knowledge, forms, practice parameters, managed care plans and government rules have been updated several thousand times. The extra effort to deal with these changes has been a light burden indeed for the student because her wireless computer has updated itself subtly so that she did not have to worry or care about these details. Those details were left to over a thousand system designers from various open source foundations, patient advocacy groups, insurance companies and government agencies that have updated the open software base in response to changing conditions. Dr. Chen is left to focus on the patient and establish a relationship with each one. Fraud can only be accomplished with great and deliberate effort since mere clerical errors are quite rare, having been designed out of the system.

In the next three months, the simulator trains her in managing her own appointments, billing and staff with well tested open source software sponsored by Franklin-Covey. Initially, 3 of her simulated workers quit in disgust at her incompetent, overbearing manner. The remaining staff coaches her in how to be a team player and she eventually acquires and trains more staff who express high satisfaction with her as a boss. Her final three months prior to entering a real hospital is spent negotiating contracts, making executive decisions about the day to day operations of a hospital and clinic. No exam is required, as her examinations were always inherent in the system.

Her actual contact with real patients has accelerated so that at the end of her second year she is functioning as a physician and only has to have her work given a cursory check by her attending since software safeguards for treatment error are always in place. She could also spend time in the simulator and get refreshed on subjects she had difficulty with. Her wireless computer has been a great help and she now has a sizable database of real and simulated cases that she can continue to measure her skills against and refer to for the rest of her life. The computer has noted her weak areas and she was able to reduce her error rate and improve the outcome of her patients quickly once she had seen a sufficient number of patients.

Many of her real attendings lamented the days in which they had an entourage of students staring at their shoes, but they also noted that the volume of patients the hospital was able to see had expanded tenfold as well as achieving a tremendous reduction of errors.

Dr. Chen graduated with honors, but the difference between the valedictorian and the bottom of the class was slight. Even the poorest physician in the group was more than adequate. Gone was the hit or miss teaching methods and critical people that made a medical teaching program collapse like a house of cards if key faculty left. Medical training for Dr. Chen is considered too important to be left to such fragile methods. Below standard physicians usually leave the profession since high error rates in a well trained physician could only be achieved by impairment, distraction or sabotage, quickly noticed by valid statistics securely and confidentially accessible by State boards.

Even with physicians impaired by circumstances such as residency, the system is a great help since it makes errors difficult to make. Particularly errors based on lack of information.

Dr. Chen moved onto private practice where she found the same system in place that she was trained on in medical school. Her faithful wireless computer and its now large database of treated cases went with her so that she was fully productive in a few days. She achieved in a few weeks a level of effectiveness in treating her patients that formerly clinicians achieved after years. In addition, all the power tools required to run her practice: appointment scheduling, billing, pharmacy ordering use the same interface she learned in medical school. A fact that enables her to take on other physicians patients with ease since they use the same standard system. The only difference is which drug company logo adorns her generally free system…

The above vignette of Dr. Chen’s life is the goal of open source medical software. Most of the above systems are either currently unavailable, or poorly intergrated and without intraoperability since there is a disincentive to do so. Furthermore, many of the above systems will not be possible with closed source because necessary additions or changes to base software in order to run derivative software (the patient simulator) could not be accomplished. Dr. Chen’s future life is the goal of open source medical software.