Don’t just do something – stand there!

By Shawn Vincent

I had the good fortune to shadow a physician client a few weeks ago, to learn more about workflow in his clinic. It was a profoundly educational experience, and I wholeheartedly recommend it to anybody working in this industry who is not a physician themselves.

During the debrief at the end of the day, he shared a piece of healthcare wisdom: “Don’t just do something – stand there!” Tests and treatments often have undesirable side effects. Taking action can easily make things worse. It’s in our human nature to want to act: to make things better. However, acting isn’t enough: you need to take the right action.

I’ve heard this physicians’ adage before, but it’s been churning around in my mind ever since my clinic visit a few weeks ago.

What strikes me is the applicability of such a sentiment to healthcare IT. Far too often, substantial investments are made in healthcare IT that don’t improve patient care, and don’t save enough money to provide meaningful return on the initial investment. How do you ensure that the investment you make does both?

We need to move away from an industry culture that says taking action is better than acting carefully.

I’d like to encourage a different way of thinking about IT investments in healthcare: evidence-based healthcare IT. If it’s a good idea in medicine, why aren’t we applying the same principles to the technology that enables it?

A succinct definition of evidence-based medicine is the Sicily statement on evidence-based practice (http://www.biomedcentral.com/1472-6920/5/1). It lays out a five-step process that I think healthcare IT would do well to systematically follow:

1. Translation of uncertainty to an answerable question

2. Systematic retrieval of best evidence available

3. Critical appraisal of evidence for validity, clinical relevance and applicability

4. Application of results in practice

5. Evaluation of performance

Translating these steps into terms directly applicable to the IT space, I’d propose the following:

Step 1: Have specific and modest goals. Rather than a goal like “Reduce costs,” which is both vague and ambitious, set a goal like “Reduce waiting times by 10%.” The more modest and measurable the goal, the better.

Step 2: Do your research. Look at projects with similar goals (search broadly – in other institutions, jurisdictions and countries). Look at the results that they have had. Reach out and talk to the people involved. Learn as much as you can about the space you’re trying to change with your project. Also, do a deep dive into the problem you’re trying to solve in your own organization. Interview people involved day to day. Do some studies to measure the depth of the problem before you change anything, to establish a baseline.

Step 3: Critically appraise others’ results. Just because something worked for somebody else doesn’t guarantee success for your project. Look carefully at the reasons for success. Was the technology the only change?

Step 4: Plan and execute your project mindfully. Take your project seriously and make sure that its success is the responsibility of a specific person who has the power and influence to get things done. Plan your change, and execute efficiently. Odds are your project will run over time and budget. If your project does start to go off the rails, make a conscious decision about whether to proceed. Don’t fall prey to the sunk cost fallacy (http://youarenotsosmart.com/2011/03/25/the-sunk-cost-fallacy/). Better to fail soon with a small investment and have funds to try something else than sink all your money in a project that cannot succeed. Deciding to pull the plug is harder than it sounds.

Step 5: Objectively evaluate your success. Even if you do everything right, you still may not achieve the benefits that you wanted. It’s important, particularly for large projects with big budgets, to objectively measure your success. Use the results of your pre-studies (you did do pre-studies in Step 2, right?) and see whether you exceeded your expectations, met your expectations, or did not meet your expectations. Regardless, look at things that were done well and poorly, and learn from the inevitable mistakes that you made. Ideally, freely publish your results so that other people who are at Step 2 have more data to work with.

There are many projects in this industry that follow, more or less, this strategy. These are the projects that provide real value (even if it’s just learning what not to do again!). Unfortunately, there are far too many projects that provide no value at all: these divert resources away from truly improving health care. The opportunities are huge. Let’s not squander our resources!

Health care dollars are precious, and the stakes are high. Don’t just do something – stand there!

Shawn Vincent is Vice President, R&D, at MD Practice Software LP.

Posted January 17, 2013