On the Road

John D. Halamka, MDOver the past few months, I’ve been in England, China, Denmark, New Zealand, and Canada. Each of them is rethinking their healthcare IT strategy and is not entirely satisfied with past progress. I’m often asked by senior government officials to help harmonize IT strategy at the country level. That I can do. I’m also asked to discuss the US Presidential campaign, but that defies rational explanation. I frequently say that healthcare IT issues are the same all over the world. Here’s a few common observations

1.  Top down never works

In every country I’ve visited (there are 195 in the world right now and I’ve been to about half), I’ve never found a healthcare IT program that succeeds by disenfranchising stakeholders and imposing a solution from above. Asking users what the want/need, then working collaboratively to deliver a workflow solution that enables them to practice at the top of their license tends to overcome any resistance to technology implementations.

2.  A single EHR for a state, province our country never works

The VA, Kaiser, and Department of Defense are completely vertically integrated which means that payers and providers in all sites of care (inpatient, outpatient, emergency, urgent care, long term care) are part of the same organization and management structure. A single EHR platform works in those circumstances. However, when a country has private payers, private providers, or a mixture of a public payer with private providers, there is not a single command and control structure. There will be heterogeneity in requirements and care processes. A single EHR vendor cannot support all use cases. Similarly having 50 different EHRs is unlikely to provide the data integration and care coordination needed by a regional group of healthcare organizations. The right answer is a parsimonious approach - the fewest number of EHRs and technology tools to meet the needs of the region - not 1 and not 50. In Eastern Massachusetts we use about 6.

3.  Interoperability needs a business case, a workflow and good policies

I was recently asked to define interoperability. I suggested that interoperability is having access to the data you need to coordinate care when you need it without a lot of effort or cost. If clinicians are paid more for repeating tests, they will repeat tests. If sharing records requires a convoluted workflow using some application outside of the EHR, clinicians will rarely take the time to exchange data. If privacy policies do not clearly define consent and allowable uses of data, clinicians will be too intimidated by compliance issues to embrace healthcare information exchange. Make data sharing part of the job/pay program, make it integrated into the EHR, and standardize the process for making data available to all stakeholders who need it, then data will flow.

4.  There may not be a measurable return on a healthcare IT investment

One international hospital I recently visited said their hospital information system was a failure because they did not see a return on investment one month after implementation.  Another I visited said they would reduce costs by shifting all the work to the clinicians, saving on administrative costs. Both are completely unrealistic expectations. In my 20 years of traveling the world, I’ve seen healthcare IT projects that improve quality in measurable ways - reducing readmissions, enhancing medication compliance, and improving processes. I’ve seen safety enhancements that markedly reduce errors. I’ve seen automation such as complex order sets that improve efficiency. However, Ive never experienced an IT project that reduces costs when all expenses of implementation and operation are accounted for.

5.  The experience of past patients can inform the care of future patients.

Different countries have different terms for this idea - big data, precision medicine, a learning healthcare system, population health, care management. If a society pools clinical data, financial data, social determinants of health data, government services data (food stamps, criminal justice, family services), and patient generated data, it is possible to ask questions about care already delivered to refine care yet to be delivered. Many societies are thinking about the benefits of population level data aggregations that may be anonymized to protect the privacy of individual patients.

Over the next few months I’ll visit Israel, Scotland, and Japan. You’ll see the details about their healthcare systems and the healthcare IT I discover in my “Dispatch from…”  series. I’ll even go as far to suggest the best model I’ve seen that aligns incentives, embraces innovation, and balances burden/benefit. At the moment, I think the Nordic countries have done a good job and New Zealand has a chance to eclipse them all. But there are miles to travel before I decide on the winner for 2017.