Paying and charging per participant, rather than per member of the population, or per eligible member, is the most easily translated into a cost/benefit prediction.  Since only HM participants figure to actually yield any cost savings or revenue enhancement benefits to the HCO for an internal HM program, or to its clients for an external one, this method offers and apples to apples comparison of costs and benefits for different HM programs and competing providers.

Flat/per population fees have been popular primarily among insurance plans, since they fit perfectly with the per-member per-month (PMPM) cost and revenue accounting systems insurers employ.  On the other hand, costs per eligible make more sense when considering a particular program, since there is no truly reliable way to estimate from a given population, prior to analysis, how many will likely yield any benefit.  But cost per participant is often preferred by payors, including Medicare in its disease management demonstration projects, since costs for DM tend to be high per participant, and fit with its accounting of savings per participant.

An added effect of charges per participant is that it will motivate HM providers to push participation as high as possible, since the more participants the more revenue they gain, though each one adds costs to sponsors.  By contrast, both flat fee/per population, and per-eligible fees motivate the client or HCO sponsor of HM investments to push participation, since once the per-member or per-eligible fee is charged, adding participants does not add to costs.  However, as mentioned in preceding discussions, it may take incentives for participation to increase beyond the level of a minority who are pre-disposed to participate.

For this reason, clients of per-participant charging HM providers may choose to put the onus on HM providers to pay incentives, since providers are the ones who most clearly and directly benefit if the number of participants increases.  Of course, as long as clients or the sponsor in an internal HM program continue to benefit as well, they may also choose to contribute to overall incentives.  In any case, the per-participant fee approach fits perfectly with any incentive program, since incentives are only paid to people who participate, even though they are offered to all eligibles.

Before HCOs or their prospects/clients agree to a per-participant fee arrangement, they should probably conduct an HRA survey that includes both risk and potential benefit information.  Productivity and performance impairment may be calculated by clients or HCOs that have well-established performance measurement systems, often part of pay-for-performance systems.  Or they may be estimated based on employee self-reporting of impairment, once validated by comparison to actual performance measures elsewhere.

There are probably half a dozen self-reporting productivity impairment surveys that have been validated by comparing to objective measures of output.  Fewer have been validated for broader performance or employee value measures, of course, since these are less frequently measured in any objective way.  It may logically be estimated that improvements in productivity will be accompanied by similar improvements in performance, of course, since the same dynamics should apply to both, at least until validations have been made based on objective performance measures as well.

One challenge associated with productivity/performance self-reporting, of course, is that validation is not the same as mathematical equality.  If employees report themselves to be 10% impaired, for example, that does not mean their objective impairment will be exactly the same 10%.  In one study, for example, call center staff were asked to report their impairment due to migraine headaches.  The average self-reported impairment was roughly 20%, but the measured decline in output for these same workers was only 8%. [G. Pransky, et al. “Performance Decrements Resulting from Illness in the Workplace” JOEM 47:1 Jan 2005 34-40] If such a relationship applied generally, than a simple translation could easily be made in self-reported impairment numbers, to estimate actual levels.

To compare flat fee or per eligible charges into per participant costs is roughly the same as translating flat fees into per eligible costs.  Flat fees are simply divided by the percent of the total population participating in order to determine “equivalent” or “effective costs per participant, which can then be compared to the total positive economic impact to determine ROI.  Of course, this can equally well be done by comparing total charges to total economic impact, without even worrying about how many participants were needed to gain it.

But since the numbers and potential value of participants is an element of the HM process that purchasers can best control, and when they pay per participant will want to control, it is useful to be able to apply a calculated cost per participant even before knowing how many eligibles will participate.  This can be done simply, by dividing the predicted value per participant by the sum of per participant fees and incentive costs per participant.  Then the result of this calculation can be used to choose the best mix of incentive levels and participation levels in terms of costs vs. net savings.

The challenge in per-participant charging situations is to avoid relying on ROI ratios as the best way to determine how many participants to pay for.  Even if additional participants will deliver a lower ROI ratio than the highest potential eligibles, such as those with chronic diseases, for example, in considering HM for saving sickness care costs alone, there are likely to be plenty more who promise at least positive ROI.  And as long as each added participant yields a positive ROI ratio higher than could be obtained in some other available investment, the addition of such participants should be warranted.

For example, if DM potential participants are known to yield a 2:1 ROI ratio on average, while other HM participants will yield only 1.5:1, the tendency might be to invest only in DM.  But if DM can only attract 10% of the population, while HM can attract 50%, the amount of total savings will be far better when both kinds of participants are recruited.  10% of 1000 people with 2:1 ROI on a $100 per participant investment will yield 100 x $100 = $10,000 in net gain.  But 50% of 1000 with a 1.5:1 ROI will yield another 500 x $50 = $25,000 more.