The Lowdown on E-Health ROI

One of the arguments for electronic health records is that they will create huge savings by reducing medical errors, improving medical decisions (which lead to better outcomes and fewer procedures) and creating efficiencies by reducing paperwork and

administrative overhead. The House and Senate versions of the economic stimulus bills include about $20 billion to $25 billion in incentives to encourage hospitals, clinics and doctors offices to adopt the digital records.

So, how long before the government's investment in e-health records -- mostly spent in the form of "incentives" provided through Medicare and Medicaid -- starts to pay off? Longer than you may think, if you look at the Congressional Budget Office's analysis of the House stimulus bill. (See page 16, subhead "Health Information Technology.")

Here's a simplified breakdown:

Cost of incentives to encourage e-health record adoption from 2009-2019:

$32.8 billion

Savings from using e-health records from 2011-2019:

$12.5 billion

Increased tax revenues (assuming health insurance premiums go down and companies pass those savings on to employees as taxable wages):

$ 3.2 billion

Total return on investing in e-health records by 2019:

- $17 billion

Because most of the savings from e-health records comes in the latter part of that 10-year period, subsequent years would continue to bring positive returns, eating into that negative $17 billion return until eventually the investment in e-health records begins to move into the black.

One of the shortcomings of the analysis -- and it's not in CBO's purview to analyze this -- is the inevitable costs and delays (which create more costs) such enormous IT projects inevitably run into. The increased price tag for an IT project like this (and they could be sizable) most assuredly will eat into the savings and could possibly push the period for the return on investment many more years out.

That isn't to say pursuing e-health records is not a good idea. The argument that it can save lives is compelling enough. But costs will increase, unforeseen problems will occur, and delays will happen. It may be a good idea for someone to try to quantify that and include it in the equation.