Structured programming is an area of computer science which is now undergoing rapid development. Although .it has many potential .-advantages to medical computing, there have been few applications in this subject area. Such an application is included in this issue.
Medical computing has a need to solve similar computing problems in many places by means which differ in detail among the users. The underlying subject matter, human biology, is much the same everywhere, and clinical practice does not vary greatly from place to place, although it is continually changing with time. However, medical computing must satisfy a large number of diverse legal, administrative, and record-keeping requirements, which vary from place to place. In the United States we must meet the diverse needs of the Federal government, the 50 states, all the hundreds of Blue Cross-Blue Shield plans and other third-party payers, and several thousand hospitals. Furthermore, these needs are continually changing through legislative, judicial, and administrative decisions. Often little if any thought is given to the implications for computing in making these decisions. As a result a computer program to solve a given medical problem at a given time and place may not then be applicable without modification at other times and places.
The use of structured programming permits the more economical production of programs which are then easier to maintain and modify, although at some price in efficiency of operation. For some time, programmer's salaries have been increasing with time, whereas the cost of computer memory and central processors has been decreasing even faster. Under these conditions it becomes more efficient to modify a suitable structured program to meet local conditions, as well as to maintain and update it, than for local users with diverse requirements to separately implement different solutions to the same basic problem, even if these solutions make more efficient use of computer hardware. Growth and evolution of such programs will be facilitated by structuring: they will be easier to modify, change the design of, and add to in a piecemeal fashion. This might particularly apply to medical information handling and record keeping, where the administrative component is particularly large, but it also applies to other technical areas within medical computing: the example reported in this issue is concerned with biological simulation.
It is not the purpose of this editorial to suggest any one means of structuring programs as best for medical applications, but rather that the time is ripe for the medical computing community to examine this opportunity and decide how best to handle it.