Spending Money to Save Money? Navigating Population Health “Hard-Dollar” Investments vs. Downstream “Unicorns and Rainbows” Medical Savings.

Most medical groups considering risk contracts usually understand that they’ll need to make significant investments in population health infrastructure for capitation to work out financially.  After all, capitation can only be profitable if the risk-bearing group has the ability to manage the use of expensive downstream healthcare services and minimize unneeded care.

The problem is that the relationships between investments in population health and total medical expense (TME) savings are pretty complex. You’re going to need to spend some money to save even more money.  The problem is that few people can tell you where, how much or when.

As I’ve discussed before, the risk bearing financial model is significantly more complicated than fee-for-service.  In risk, revenue is a function of covered lives times a negotiated rate (and most risk bearing health systems have some fraction of fee-for-service volume too).

Screenshot 2015-04-21 16.21.11

Here is the model: note that on the expense side, there are primary expenses: direct, which include things such as the cost of the facilities, staff, medical supplies and the like, including your population health infrastructure (e.g. case managers, chronic disease coordinators, geriatric managers, behavioral health….)

The second, and bigger cost is medical expense, which includes both internal expenses (the cost of health care services provided within the organization), and external medical expenses (which includes things such as hospitalizations, visits to nonaffiliated physicians, ambulance services, pharmacy, and the like).  In risk-bearing groups investments in population health can (and should) decrease medical spend.  Thus, direct expenses and TME are in opposition– though not proportionately.

In risk, you are trying to arbitrage by getting outsized TME savings per unit of money invested in population health programs.  This is the art of capitation.

Thus, paying clinical pharmacists means that your pharmacy costs will fall and you’ll “make” money (or rather “save” money) by making the investment.  Paying care managers and chronic disease managers and geriatric home visit nurses to check on sick, frail old people will have a direct impact on your hospital, ED and drug costs, and so on.

I’ve learned this:

The debate that should occupy every business meeting at every risk-bearing group should be about how to properly spend money on direct expenses to save even more money on healthcare expenses.  

This is a bit counterintuitive for managers from a FFS background.  After all, in a typical fee for service business, managers will look for efficiencies by controlling as many expenses as possible. Most organizations look to reduce costs by strictly analyzing overhead, leases, staff, and the like.  The more you can sit expenses, the more profit to be made.  Running lean via smaller overhead is the path to success.

Except in risk.

In risk, the real money is in the pass-through TME dollars, which flow through the company and are used to pay for hospitals, outside doctors, pharmaceuticals and the like. Managing this expense is the margin opportunity.  Organizations that take on capitated contracts will quickly find that they will sink the ship if they skimp on the backend infrastructure needed to manage populations.  Unmanaged, over 50c of every dollar of revenue coming in will flow right back out in the form of payments to outside providers.  The only way to manage this is through robust IT infrastructure, case managers, data analytics operation, risk management, etc. You simply can’t be successful in risk without these essential services.

At first pass, these investment decisions aren’t rocket science:  you’d need to do a typical ROI calculation looking at the immediate direct costs and anticipated downstream TME savings, while factoring in the present/ future value of money.

But, the devil is in the details of each ROI assumption.  I’ve listed these as “The 6 Laws of Investment in TME Reduction”

1) Law One: Total Medical Expense (TME) savings will always be undervalued, and direct costs overvalued, especially during budget season.

When it comes to budgeting, your director of finance will tell you that the direct costs of a pop health program are immediate and quantifiable, whereas the TME savings are hypothetical, unproven and down the road.  I’ve heard TME savings referred to as “unicorns and rainbows”, feel-good, mythical things that you look for but never find….

This isn’t true, or course, since the ratio of payroll to TME passthrough in many risk-bearing groups is easily 1:10 and the TME impact (cost/outcome) can be strikingly high. But, your finance guys will always overvalue the cost of the hard-dollar investment and undervalue the potential upside. They are wired to do this mainly because they are loathe to put too much weight on performance that’s out of their hands and in the realm of the clinicians. They just don’t trust that the savings will reliably come through when it comes time to close the books.

2) Law Two: you’ll never be able to prove causation between investment and outcomes.

In other words: you’ll see savings in your TME spend, but you’re unlikely to be able to tie it back to specific investments in population health.  This is because you’re likely to make multiple pop health investments at the same time, and these work synergistically: case managers working with psychologists and pharmacists and so on.  Your total pharmacy spend is going to go down, probably a lot.  But, I’ll promise that you’re going to be unable to isolate this to one program or another, and the results from the multiple initiatives will meld together. So, unless you’re prepared to do a rigorous randomized trial, which is functionally impossible, you’ll need to be satisfied with watching correlation.

3) Law Three: Depending on your internal funds flow process, investment and reward won’t sit together.  Don’t confuse “expense” departments with “money-losing” departments. 

From an accounting perspective, you’re not going to make investments in, say, pharmacy and see pharmaceutical medical expense savings pile up in the same pharmacy business unit. In our company, pharmacy savings actually accrue in the primary care business unit, in the form of reduced medical spend.

This is one of the things that makes measuring program effectiveness hard, and one of the reasons why you have to become comfortable running operational “losses” in “expense” departments.  They may not generate revenue and look like expenses, but in aggregate they generate huge savings for the company.  Underinvest in clinical pharmacy, and watch primary care take the hit…

4) Law Four: it takes a while to see the TME benefit, but direct costs accrue immediately.

Changing behavior takes a while.  You are going to need a “float” period where you are paying direct expenses but before you’ve appreciated the efforts of the work. It can take a while to see TME savings, particularly as they wind their way through the payer networks, etc.

5) Law Five: You’re going to struggle with your investment’s hurdle rate.

The hurdle rate of an investment should reflect what you would make on your money over time if invested for other purposes.  These days, a payback hurdle (MARR or minimal acceptable rate of return) of ~ 12% seems to be fashionable, based on the historic performance of the equities markets. Given the uncertain returns associated with pop health investments, you could argue that it should be higher than this.

But, given that few risk-bearing systems are making anywhere near 12%, makes this argument a tough one. Why should the payback of an operational improvement be expected to pay more than the organization’s margin as a while?  What should the hurdle be?

6) Law 6: There is a law of diminishing returns in population health investments. 

I’ll make this final argument:  the goal of population health is to keep people healthy and mitigate the use of wasted and unnecessary healthcare services, or those provided needlessly at high-cost locations. There is a logical end-point to this activity: it’s when people receive perfect preventive care, and rigorous evidence-based care at high performance/ high-value providers. That’s the green dashed line in this graph comparing direct costs to TME savings.

Screenshot 2015-04-21 16.05.18

I’ll make the observation that the relationship between direct investments in population health, and TME reductions looks something like this:  The easy stuff is done first, and cheaply.  Then, achieving TME reduction becomes harder and harder, you break even — one dollar or pop health equals one dollar of TME savings.  Then, it’s into negative territory, where it takes two dollars to achieve one dollar of savings…


I’m often asked how much a risk bearing organization should spend on direct costs to reduce its TME…  I’d argue that when you’re getting close to a 1:1 ratio of expense to savings (as the slope on the red line above begins to flatten) then you’re where you need to be. The investments generate benefits beyond the numbers in terms of keeping people healthy and happy- but anything more than a 1:1 is something that most companies can’t afford.



Photo: Tanakawho via Flikr, Creative Commons license

%d bloggers like this: