Problems with medical records

George Dunea, MD

(Summer 2015)

When Lawrence Weed first unveiled his vision for reforming medical record documentation, he unleashed a revolution that captivated the imagination of the medical public but may also have brought about unintended consequences from which we suffer even today.

Dr. Weed first published his new method in 1967. 1 A few years later, in 1972, he chaired a two-day seminar on Problem Oriented Records in Atlanta, Georgia, with Dr. William Hurst as co-chair. There he contended that doctors, unlike other professions, had to deal simultaneously with multiple patient problems and accordingly needed a disciplined system to allow them to address each problem separately. He even contemplated a time when medical practice would more closely resemble the airline industry, with maps, road-signs, pilots, mechanics, air traffic controllers, and computerized navigation systems. Well-trained physician extenders or assistants would obtain a complete history (thus saving expensive physician time) and then key it into a database coupled to an extensive knowledge databank. Detailed computer-based analysis would then apprise the physician of the most likely possibilities instead of having him rely only on his own knowledge, judgment, and intuition. The system was rational and practical, designed for the men in the trenches, not for the rare individual combining the genius of Sigmund Freud and William Osler.

Dr. Weed further explained how this system would work in a busy teaching hospital. An intern or medical student would most likely take the initial history, often in the middle of the night. He would then present his findings to the assembled medical team in the morning, starting by outlining the patient’s “problems”. Each finding would then have to be checked and confirmed. Thus if the intern found an enlarged spleen, the senior members of the firm would in turn palpate the abdomen to confirm or disprove his finding. A consensus would be reached and the finding recorded as such. There would be only a single history on the chart – complete, reliable, and definitive.

Next the record would have a section summarizing how the various findings were related to one another. A diagram would be helpful, for example rheumatic fever leading to carditis – then aortic valve disease – endocarditis – heart failure – and systemic embolization. This section, potentially of great value, was the first casualty as soon as the system reached the public. It failed to catch the attention of the instructors and administrators, who however immediately fell in love with problem lists and especially with SOAP progress notes. These were to be written daily, using the subjective, objective, assessment, and plan format. Each listed problem was to be identified by a number and addressed separately until it was resolved or the patient was discharged. Thus if the patient with aortic incompetence (problem number one) also had a drug rash (two), depression (three), and dysuria (four), every problem would be fully addressed every day with its own SOAP note until resolved or the patient discharged.

By the end of the Atlanta meeting the audience was enchanted, mesmerized. But those visiting the local teaching hospital the next day were disappointed. Most patients had lengthy problem lists, sometimes as many as twenty, a simple urinary tract infection being listed as white cells in the urine, bacteruria, frequency, dysuria, and fever. Universally lacking was the section and the neat diagram by which findings were to be confirmed and interrelated.

Nevertheless the new religion spread like wildfire. It was adopted by edict by the Veterans Administration and by most teaching hospitals. Coming at a time of societal turmoil and questioning of traditional values, it readily became dogma that patients had problems, not diagnoses. But the system was never fully understood, never fully implemented, and in time only the famous SOAP notes survived, in high favor with students and residents, always partial to using abbreviations and acronyms.

Two generations later, perhaps still inspired by the Weed system, the government has virtually mandated electronic medical records, using a stick and carrot approach, threatening to cut reimbursement for nonusers and offering significant moneys to physicians who can demonstrate “meaningful use”. But in the trenches computers tend to be slow, clinically unhelpful, apparently more designed for billing, often “down” at busy clinic times, and slowed down by complex passwords and sign-in requirements. Many doctors type slowly, some not at all, and everybody says that even simple tasks have become inordinately time-consuming. “Documentation” takes precedence over everything else; doctors have to keep their eye on potential malpractice issues, government regulations, and avoid being charged with fraudulent billing; the patients complain that their doctors look at the computer and not at them; and the hospitals send out to the referring doctors huge computerized patient notes full of legal disclaimers and irrelevant details, but often with scant information on what actually was the matter with the patient, what was the diagnosis,  and what was done for him.

It is all a long cry from what was originally proposed – almost half a century ago. In an interview given in 2009, Dr. Weed expressed his disappointment at the lack of progress, nor was he optimistic that changes would soon be forthcoming2. Yet to fairly judge the Weed system, well intentioned but complex and impractical as it was, one might pause and consider, like Edward Gibbon did in a different context, not only by whom but also to whom this revelation was given. But as America always does the right thing (after it has tried everything else) there are grounds for optimism about the future.


  1. Weed L. Medical Records that Guide and Teach. New England Journal of Medicine 1968; 278:593 – 600.
  2. Jacobs  L L.  Interview with Lawrence Weed. Permanente Journal, 2009; 13:84 – 89.



George Dunea MD, Editor-in-Chief