Scott MacStravic, PhD
As hospitals become significantly at risk for re-admissions due to the same problem as the original one within 30 days of discharge, they might be forced to invest in disease management (CM) for at least the patients and diseases for which it is having a significant number of such re-admissions. At a minimum, this means creating a discharge planning and follow-up process that optimizes discharged patients’ ability to manage their recovery and chronic condition in general in order to prevent the crises, complications and worsening that cause re-admissions.
The payoff to hospitals will be not having Medicare refuse to pay for what it decides are “avoidable/preventable” re-admissions, as is already happening for “never” events, which are far more rare and therefore produce less damage when not paid for overall. A recent article reported that there were about 5 million such re-admissions a year in U.S. hospitals according to the Institute for Healthcare Improvement, and as many as 46% could be prevented. [L. Landro “Keeping Patients from Landing Back in Hospital” Wall Street Journal Online Feb 12, 2007 (online.wsj.com)]
Of course, the traditional reasons for not following up with disease management of patients discharged from inpatient treatment for some crisis, complication or worsening of their condition(s) still applies. There is no currently built-in basis for charging for Disease Management (DM) unless the hospital is part of a CMS demonstration project (where they can’t seem to consistently deliver cost savings) or has contracts with insurers or employers for such services. The same is true for physicians, who may also prefer that hospitals stay out of the market if DM ever becomes profitable for physicians.
DM has a seriously flawed history among many of its early reports of success, due to reporting as cost savings what was likely to be “regression to the mean” natural declines in expenditures for DM patients who were identified and selected for DM interventions based on outlier levels of expenditures in a baseline year. Other reports have suffered from “self-selection bias”, reflecting the fact that when patients can choose whether to participate in DM or not, those who enlist in DM are likely to be more engaged in self-management and complying with medications than are those who do not, so their results are likely to overstate what is true for non-participant comparison groups.
Both flaws can be overcome fairly simply, however, and hospitals can avoid overstating the gains they deliver to others, or those they enjoy themselves, when using DM internally for their own employees. Whether for internal or external use, the keys to successful DM are likely to at least include defining DM broadly, and measuring is effects broadly. The definition should include all investments and efforts aimed at reducing the incidence and prevalence of chronic diseases, and of “health conditions” that impair workers’ productivity and performance on the job. And the measured effects should include measures of the full economic benefit of reducing both direct expenditures – for sickness care, disability and workers compensation costs – and indirect gains from improved productivity and performance, across all the measurable or credibly estimable impacts that improved employee health are known to have.
Chronic diseases tend not to have that much impact on workforce productivity and performance, which tends to reflect more as much as double the overall economic costs and benefits of DM, but affects what is usually only a minority of the workforce. Other chronic conditions, including both disease risk and productivity/performance impairment factors, usually have greater prevalence rates, higher impairment impact, or both. And these often have costs that are as much as five times the costs of chronic diseases when applied to the workforce as a whole.
For example, when the Dow Chemical Company calculated the costs of ten chronic conditions, including both healthcare expenditures and lost productivity, it found that lost productivity amounted to almost exactly five times as much as did healthcare expense, in terms of total impact on the workforce. The top four cost conditions were allergies, arthritis, back/neck pain and emotional disorders, rather than diseases in the usual DM domain. [J. Collins, et al. “The Assessment of Chronic Health Conditions on Work Performance, Absence and Total Economic Impact for Employers” JOEM (Journal of Occupational and Environmental Medicine June 2005 547-557]
But even more than the costs of chronic diseases, risks and impairment factors, there is the measured benefits and gains demonstrated by DM providers. Interventions that prevent chronic diseases in the first place tend to save all the negative effects of each disease, where managing such diseases only recovers some of the losses. And interventions that prevent or minimize risk and impairment factors can have significantly greater impact over a far larger portion of the workforce.
For example, HealthMedia, Inc. in Ann Arbor, Michigan, has reported the success it has achieved after six months post participation in its chronic disease and risk condition programs. Its Diabetes DM program yielded a 2.00% increase in productivity, for example, which would be at least 2% of $50,000 = $1000 per participant per year. But because only roughly 5% of the employees in its database had diabetes, this would only amounted to an average of $1000 x 5% = $50 per employee across the workforce if every diabetes patient participated in DM.
By comparison, its general chronic disease/risk condition DM program achieved a productivity reduction of 1.88% per participant, slightly less per participant than was true for diabetes. But because there were roughly 40% of the population that had one of these other conditions, the effect would equal $1.88% x $50,000 x 44% = $376 gained per FTE in the workforce. The difference in the economic benefit per employee between diabetes DM and a general risk/disease DM intervention is important whenever the DM program charges are based on the total number of employees in the workforce, as is true with HealthMedia. The gains of $50 vs. $376 per employee would make charges of say $50 per employee look quite different for the two different interventions. [HealthMedia’s charges vary with the size of the workforce, and may be more in small workforces, or far less than $50 per employee per program in large workforces.]
Moreover, even identifying the size of the problem and potential gains adds costs per worker, since health risk assessments that include productivity/performance impairment questions are usually charged per worker, as well. So the more that costs per worker can be reduced and productivity gains per worker achieved, the better the investment is likely to work.
