Skip to content Skip to navigation

Measuring IT Value: What Intel Got Right - and Wrong

May 12, 2008
by E2556BEF60524A5689D02EEBDAEFEDB6
| Reprints

I recently attended Intel's webinar on Measuring the Value of Mobility at the Point of Care. In this session, Intel representatives teamed with their clients to present several case studies of the combined value of mobile devices, applications and wireless infrastructure. In general, they did a nice job, and Intel's "Business Value Model", as applied to healthcare, is a reasonable framework for understanding IT value. The things I liked about their approach included:

1. They used a systematic and logical approach in their "Value Model Workshop" to identify and reach agreement on potential IT benefits

2. They proposed specific metrics that could be used to measure the potential benefits

3. They completed what seemed to be a fairly detailed workflow analysis to identify the specific mechanisms of IT benefit - or HOW the IT "made the benefits happen"

4. They gathered quantitative baseline and post-implementation data, using their proposed metrics

5. They identified lessons learned about how value is realized

These strengths of the Intel approach fit nicely with the lessons learned from our experience with over 100 early adopter hospitals, which we describe in "EHR Value: Don't Expect What You Don't Plan For" (Healthc Fin Mgt, December, 2006) and "A Review of Methods to Estimate the Benefits of Electronic Medical Records in Hospitals and the Need for a National Benefits Database" (JHIM, Winter, 2007). Basic as they may seem, these strategies are NOT typically used by most hospitals to measure the value of their IT investments.

However, I question one of the methods Intel described for "monetizing" benefits, namely, imputing cost savings based on the calculated value of an "hour saved" instead of actual salary cost reductions. In one example Intel claims $11,000 in annual cost savings based on a three hour daily reduction in time spent on "corrections" by health information center staff. But these savings are based on the hourly pay rates of the affected staff times the three daily hours "saved", with no evidence that any salary reductions were achieved. This appears to be a purely theoretical "savings" with no practical value. What did these staff DO with the time they saved? Could the value of any additional activities they performed or results they produced be measured? In another example, Respiratory Therapists were found to have saved 5 minutes in documentation time per treatment, with an imputed value of $2.50 "saved" per treatment ($30,000 annually). It's unlikely that Respiratory Therapy staffing was reduced based on these time savings (nor was this claimed). But, did the number of treatments performed each day go up? And if so, did that result in any more net revenue to the hospital? If not, was there a real, "monetized" benefit to this change? If there was no real financial benefit, was there a demonstrated improvement in therapist or patient satisfaction, or a reduction in errors, or an improvement in the quality of the treatments provided?

This method of imputing financial benefits that don't really exist is quite common, not only in vendor benefit studies but in the published literature, as described in our upcoming article "Using Published Data to Estimate Hospital EMR Benefits" (Healthc Fin Mgt, in press). Intel would have done better to omit these savings or quantify them in another manner. The use of this method, and other common benefit forecasting methods described in our upcoming article, invites the skepticism of hospital executives who have to manage to real budgets.