Scott MacStravic, PhD
Whether HCOs choose to invest in employee health management (EHM) as part of their internal cost control strategy, as a revenue-generating new service line, or both, there are the same key numbers involved in all three. These are:
- The costs per EHM participant for the intervention
- The extent of reduction in costs achieved thereby
- The relationship between the two
In the simplest case, the cost reduction relates to lower sickness care utilization and expense. For HCOs with more effective measurement methods, cost reductions in disability and workers compensation may also be achieved. For the most effective HCOs, measures of reduced absenteeism, presenteeism and turnover, and perhaps even increased revenue due to direct (pay-for-performance bonus revenue) and indirect (resulting from happier customers, increased market share, etc.) can be achieved and measured.
For example, a Stanford study of disease/case management for heart disease patients focused on reducing participants’ risk of heart disease “events” and expenditures. An estimate was made that the cost savings resulting from avoiding such an event could be $40,000. The program costs, including an average of 14 visits with nurse coaches or dietitians, averaged $1250 per participant. The results of the intervention, as reported for the 17 months of the study, included lower blood pressure readings and other risk indicators that amounted to a reduced risk of coronary heart disease events equal to 1.6%. [“Stanford Study Highlights Cost-Effective Method of Lowering Heart Disease Risks,” Business Wire.com Aug 20, 2007]
This was found to be statistically significant, given the 341 patients participating, with half receiving the HM intervention, and half getting usual primary physician care alone. The question arises, however, if it was also financially significant. What was the estimated ROI from this investment? Based on just the figures given, the ROI would be 1.6% times the number of participants involved, say 170 as half of the total number, times the cost per case of $40,000.
The math is simple enough – 1.6% of 170 participants is 2.72 cases presumably avoided over the participation period. Since each case was assigned a cost of $40,000 in medical/hospital expense, the 2.72 cases avoided means a savings of 2.72 x $40,000 = $108,800. On the other hand, with 170 participants, each costing $1250 apiece for the HM intervention, the total program costs were 170 x $1250 = $212,500. From a financial perspective, the program was not a success, since it cost almost twice as much as it saved. The ROI ratio would be $108,800/$212,500 = 0.512:1, or a loss of 49 cents on every dollar invested.
Since these figures require knowing up front how many participants there will be, and what reduction in the number of cases there will be, they are more useful for evaluation than for planning. At the point of considering whether to invest in HM or not, and planning an intervention of any given type, there is a simpler way to approach the same challenge. It is called “numbers needed to succeed” (NNS) analysis. It depends on a combination of decisions and predictions made up front, rather than waiting to evaluate after the fact. [A. Linden & T. Biuso “In Search of Financial Savings from Disease Management: Applying the Number Needed to Decrease (NND) Analysis to a Diabetic Population” Oregon Health & Science University, Portland, OR (undated working paper)]
In the above example, if it is decided that an HM intervention costing $1250 is the one to be considered, and the predicted effect of this intervention is the reduction of 1.6% in the frequency of particular expenses, regardless of what these expenses come from, these are the only two numbers needed for NNS analysis. If the costs are $1250 per participant, and 1.6% of participants will yield the expected cost savings, then the amount of the cost savings needed for each such participant has to be $1250 divided by 1.6% = $78,125.
If the HCO is considering only one HM intervention, it can therefore look to see what the chances are that there will emerge from this investment as much as $78,125 in savings per case. If healthcare costs are only $40,000 for example, other costs, such as days absent and the lost productivity that means, short and long term disability costs, impaired productivity at work, for example. One study of over 200,000 employees, for example, found that employees at work who were recovering from a heart attack, had a 14% productivity impairment. [“Population Impairment Productivity Dashboard” (www.HealthMedia.com) May 6, 2007] This would amount to a loss of at least $8400 in output for a worker earning the average hospital employee wage of $60,000 per year, though it would depend on the hospital and labor market involved.
Moreover, there is known to be a “multiplier effect” for absent or impaired workers, in terms of their impact on team performance. This has been estimated at 1.4 for hospital nurses. [S. Nicholson, et al. “How to Present The Business Case for Health Care Quality to Employers” Applied Health Economics and Health Policy 4:4 2005 209-218] With such a multiplier, the economic loss from an employee recovering from a heart attack could be 1.4 x $8400 = $1176. When this amount is added to the $40,000 of avoided costs per case in hospital/medical costs, it is still only $51,176. If disability costs, days absent from work, impairment, and other costs ended up adding up to more than the $78,125 needed for breakeven, the program might be seriously considered.
On the other hand, there may be another HM intervention that has lower costs, say $500 per participant for illustration. If that program were predicted to reduce the frequency of adverse events by 1%, the amount needed in avoided costs to justify an investment in this alternative intervention would be only $500/.01 = $50,000. With a $40,000 start in terms of sickness care costs, finding the other $10,000 may not be as hard as finding the $38,125 extra needed for the original intervention.
By the same token, if there is a $200 intervention that would yield a .05% reduction in adverse cases, then the amount needed to breakeven on the program would be only $200/.005 = $40,000, the amount already specified as sickness care cost reductions expected. All other savings in disability, absence and presenteeism, in avoided turnover, plus any predicted increase in revenue that can be foreseen, would result in pushing the ROI above the breakeven point. In any case, the NNS analysis provides a useful tool for considering EHM, or any other investment whose costs and effects can be predicted on a per participant basis.
