Data Processing

The division of quality control activities into simple tasks is essential to the development of an effective computerized analysis system. The main stumbling block to the effective use of computers in hospitals has been the temptation to computerize all of the hospital's information flow. The have in face been several well-funded attempts to computerize a complete hospital data base (including medical records), but each of these efforts has failed because of the overwhelming amount of data that must be stored and manipulated. In contrast to these attempts. the hospital administrative functions that have been successfully computerized are billing and electronic mail. Both of these functions are simple, requiring no special data handling techniques. They are cost-effective to automate because they are repetitive and provide easily appreciated benefits over manual methods. These are examples of problem-oriented data processing. In each case, the problem to be solved was well-defined and limited in scope.

The use of data processing techniques is crucial to the long-term effectiveness of medical quality assurance programs. The amount of information that must reviewed to meet current JCAH standards is very large. As these standards tighten, requiring larger amounts of information to be analyzed, it will become impossible to comply with them by manual review techniques. The standards are not onerous, but the growing complexity of health care delivery makes even a cursory analysis of quality control data a massive task.

The problem of managing complex information is not limited to quality assurance activities. Health care providers face this problem in all areas of health care delivery. Just as most hospitals have been forced to automate their billing procedures, they will be forced to automate their quality control efforts, their pharmacy functions, their linen service activities, and all other functions that demand the monitoring of large amounts of data. This automation will be necessary both for economic reasons and to satisfy the hospital's legal duties.

Automated data handling will be legally necessary to document effectively that the hospital is diligent in carrying out its surveillance duties. The hospital is not legally liable for all negligent injuries suffered by patients, only those it could have prevented and those caused by its employees. Thus, to defend itself properly, the hospital must be able to demonstrate that it did in fact monitor the activities of the personnel involved and that this monitoring did not indicate anything untoward that would have alerted the hospital to the necessity for intervention. This monitoring must be more detailed than just reviewing incident reports. It must include a review of all oversight committee function (tissue committee reports, pharmacy committee reports, infection control reports, and so on) to determine if there is a pattern of substandard care. The monitoring must also be sufficiently well-documented to convince a jury that it was conscientiously performed. This demands detailed written records of all surveillance activities.