home email us! sindicaci;ón

Which Outcomes Should Be Attributed to Disease Management?

by Scott MacStravic

There has been some recent discussion on the subject of which changes in healthcare costs found among people who are participating in disease management (DM) programs should be credited to the program.  After all, change is natural, even inevitable, so it would be cavalier to assume that every change noted in healthcare utilization and expenditure is due entirely to DM interventions.  And errors in attribution would tend to lead to persistence in investments that are really not causing the returns claimed for them.

One of the topics in this discussion has been labeled “therapeutic specificity”.  It reflects the argument that when a person has a particular disease, such as diabetes, congestive heart failure, depression, or any other diagnosed condition known to lead to healthcare expenditures, as well as absences, impaired performance while at work (“presenteeism”), disability costs, etc., and that disease is managed, only changes in the direct consequences of the disease should be counted as due to its management.

On the face, this may seem to make sense.  It would make no sense to count as costs due to the disease in the year prior to participation in a DM effort the results of an automobile accident, for example.  So clearly the total costs for a given person in the year prior to participation should not include the care needed because of an accident, unless the disease somehow caused it.  And if the DM program focuses specifically on a single disease, then only the costs of care caused by that disease should be counted in before vs. after comparisons.

But there is a strong argument against this position – one directly related to the label applied to it, namely “therapeutic specificity”.  It is not the disease that should limit which changes are attributed to the DM intervention, but the treatment involved.  With diabetes, for example, it is usual to find high levels of blood pressure and cholesterol, not merely blood sugar.  Moreover, depression is a common co-morbidity of diabetes, as it is for other diseases such as heart disease and cancer.

Moreover, when either face visits with a physician or phone coaching by nurses is part of the DM intervention, it is almost inevitable, and a matter of professional pride, that any concerns the patient expresses during visits or coaching sessions be addressed by the provider.  In such cases, the “treatment”, though focused on managing the specific disease in question, would include recognizing and addressing other conditions that patients mention, and others that providers worry about, for that matter.

If the DM intervention includes coaching, urging of lifestyle changes, monitoring of related conditions, and similar matters that are not direct consequences of the disease being managed, then it is this intervention that should limit the therapeutic specificity, not the disease.  If coaching on blood pressure and cholesterol succeeds in lowering the patients’ levels of these risk factors, and the consequences of this include lower healthcare costs, absences, and performance impairment unrelated to diabetes per se, then these therapeutic consequences should surely be counted as attributable to the DM intervention, even if they have no demonstrable link to the disease.

It is usually the case that DM evaluations tend toward over-counting of the consequences of interventions.  But if only changes that can be directly linked to the disease were counted, where the DM intervention included efforts and information that addressed other issues, this would lead to under-counting.  Since the aim should be counting accurate, precise, and attributable results to the intervention in question, the scope and limits of the evaluation should extend to what can be logically credited to the intervention, not merely to the disease.


No comments yet »

Your comment

HTML-Tags:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>