Disease management is so yesterday, unless it is “digital”. But actually, disease management programs are still highly prevalent in commercial health insurance and Medicaid. One reason enthusiasm for the tactic waned is that research isn’t entirely supportive of benefits from these efforts. Research published in the Journal of Health Economics looks at one particular program used in Georgia’s Medicaid system. (JHE Article) As with most populations, the Medicaid one sees spending concentrated among a relative few members who typically have significant chronic diseases. Disease management seems like a natural way to try to improve treatment of those diseases and reduce spending. Most beneficiaries with high and moderate risk participated in a comprehensive and intensive disease management effort and results for that group were compared with those for other similar beneficiaries who only received the usual low-intensity intervention. The disease management programs were pretty standard, covering the usual list of expensive chronic diseases, with the most intensive effort directed at the highest risk beneficiaries and including home visits as well as ongoing phone contact, education and coaching. A data processing error inadvertently led to some high and moderate risk members being put in the low-risk category, so a natural experiment opportunity was created.
Medical expenditures for the high-intensity intervention declined by about $89 per person per month relative to the low-intensity group, which is not insignificant. Most of the savings occurred in regard to treatment for the chronic diseases targeted, with COPD, asthma and coronary artery disease treatment showing greater expense reductions. By treatment type, there were reductions in ER and inpatient use and spending, particularly for those with asthma. There was an increase in medication spending among moderate risk patients and a decrease for high-risk ones. The program, however, cost $126 a month, so it is pretty obvious it wasn’t cost-effective, looking solely at spending and not quality. Even within the intervention population, however, there was wide variation in pre-intervention spending, and for the highest cost tier of this group, the intervention was cost-effective, saving $200 per person per month. The lessons, as usual, are to limit hype about savings from disease management, to target the interventions to clear higher-risk patients so that there is some chance of a return on investment, and most importantly, to consider improvements in patient outcomes and patient health separately, because that is a value separate from any economic analysis.