Finally, it may be possible to examine various medical outcomes, such as injuries, illnesses, or hospitalizations during the first term that could be related to initial medical or physical conditions, some of which might be subject to disqualification waivers. This type of analysis would involve merging files of the U.S. Military Entrance Processing Command (MEPCOM) with medical record files, which may require special approvals but should be technically possible.
The final step, assuming that a physical or medical condition is reliably correlated with an acceptable outcome, is to determine a specific standard for that condition. There are at least two scenarios. On one hand, if the physical or medical condition is a simple binary “present” or “absent,” then, as mentioned earlier, the standard may simply be a decision to disqualify a potential recruit if the condition is present (which assumes that a person with that condition cannot perform military duties). If, on the other hand, the condition has a continuous underlying metric, such as BMI, a score on a fitness test, or blood pressure, then a specific cutoff value is required.
The process of determining a specific cutoff value ranges from simple to complex, depending on the way in which utility is assessed. In the simplest case, military experts may be able to use a qualitative procedure based on experience that determines a value above or below which a recruit is highly unlikely to meet the requirement in question.
In the more complex case, particularly when there is no expert opinion about acceptable levels of the condition (or even sometimes when there is an expert opinion), a cost-performance trade-off model can be applied. For example, if BMI or some type of substance abuse is correlated with attrition or some performance outcome, one can compare the cost of disqualifying persons above certain values (based on supply information about the cost of recruits) with the cost of various performance or attrition levels (e.g., training costs associated with different rates of attrition).
In this section, we discuss an example of the type of cost-performance trade-off model that has been applied to aptitudes and education. The full model is a computer-based optimization model that evaluates enlistment standards along two dimensions: high school graduation status and aptitudes. We present a much simplified version of this model and illustrate it with high school graduation status, a characteristic not unlike such medical conditions as asthma or drug use. We use this model later for those