Tapping into health care data to save lives and reduce costs

March 15, 2012 5 Comments »
Tapping into health care data to save lives and reduce costs
Can the power of data improve health care? (Flickr photo: The Javorac)

In British Columbia we are overlooking a vault of public treasure. If we allowed people we trust to open it, and analyze the treasure within, we could save thousands of lives and, potentially, billions of dollars in B.C. alone.

This treasure is in the form of electronic data collected over many decades, under the stewardship of successive provincial governments, on everything from hospital admissions to doctor visits and prescriptions filled by every pharmacist in this province. It’s one of the best collections of health data anywhere in the world.

It’s time to use it to help researchers save lives, to make better use of taxpayer dollars and to discover cost-effective ways to get better health outcomes.

As soon as anyone raises the subject of accessing health data, the immediate reaction by many is, “You can’t do that – it would violate an individual’s privacy rights!” Well, we have laws and systems already in place that make sure privacy is protected by removing personal identifiers. Researchers have no way of tracing the data they receive back to individual citizens.

The upside of opening British Columbia’s unique trove of data sets is huge.

For example, how many people suffer (or even die) from avoidable adverse drug interactions? Drug manufacturers do clinical trials before a new drug can be declared safe. These trials are in very controlled settings and they do try to ensure that new products are not only effective but also safe and that any side effects are identified up front.

But the trials are often short, and involve only certain population groups. What happens when the drug is used in the real world, often over long periods of time? Sometimes two harmless drugs, when taken at the same time, can trigger effects that the clinical trials did not identify.

Sometimes side effects don’t show up until a drug has been taken for longer than a trial might have run. Sometimes drugs intended for one use or one group during a trial end up being used for other purposes and on other people once they have been approved. Adverse drug reactions are often identified by doctors or pharmacists who observe the effects on the front lines and then trigger an investigation.

By tapping into health care data that we already have, more of those adverse reactions could be identified earlier. Lives could be saved.

Making use of our health care data also has the potential for huge savings in delivering health care.

We often throw around the term “evidence-based medicine.” In theory, every procedure and every medication has been scientifically tested to prove that what common medical practice says should be done will, in fact, lead to the desired improvement in health outcomes. In reality, much of what is done in medical practice is done because that is the way it has always been done. We have the data to determine whether some of the things we do really do help, whether other things we do offer no benefit whatsoever, and whether there are ways of achieving better health results at less cost.

A study done last year by the McKinsey Global Institute concluded the annual value of using data effectively could be more than US$300 billion, “two-thirds of which would be in the form of reducing national health care expenditures by about eight per cent.” And that is in the United States alone. In Europe, the McKinsey report also states, “we estimate that government administration could save more than 100 billion euros (C$149 billion) in operational efficiencies alone.”

Just think of what that could mean in this province. Our health care budget is $18 billion a year. Eight per cent of that is $1.44 billion a year. This could make a huge difference in our ability to deliver effective health care in the future in B.C., as well as Canada.

The McKinsey study predicts it would take a decade to achieve those cost savings in the United States. Why? Because the Americans don’t have the data. They would have to start today to systematically build the data sets needed. In B.C., we already have those data sets waiting to be harnessed, to help us take the lead in improving health care.

British Columbia is already known globally for leading-edge health research. The research community adds hundreds of millions of dollars to our economy. But we limit researchers’ access to this data, because of concerns about protecting privacy that provincial regulations now make moot.

It’s time we proactively opened this B.C. advantage to health researchers and research funders locally, nationally and internationally. Not only would a more open-door policy bring in millions of new research dollars and human talent, it would lead to discoveries that will save more lives, improve quality of life and cement British Columbia as a centre of excellence for bringing efficient/effective health care solutions to Canada and the world.

This is one of those rare opportunities where there are no downsides, only remarkable potential benefits. The protection of individuals’ privacy must be paramount. And it is. B.C. has the safeguards to guarantee privacy is protected. We now owe it to ourselves, and to our fellow citizens, to unlock the treasures hiding in our data and use them to save lives.

- post by Colin Hansen



5 Comments

  1. Steven Forth March 15, 2012 at 7:10 pm -

    Where is the Like button? Evidence-based medicine is necessary if we are going to have an effective and affordable healthcare system in the 21st C and evidence-based medicine requires evidence. We need to make this happen as soon as possible.

  2. Paul LeBlanc March 15, 2012 at 9:59 pm -

    This is an excellent idea.  

  3. Steven Forth March 16, 2012 at 6:54 am -

    This would also be a way to stimulate innovation. Making the data available and letting new companies apply new approaches to pattern recognition, complex event processing, social sharing and so on could lead to a surge in activity in BC.
    I would have some concern that the government would favour large multinationals rather than local growth and innovation, this has happened in the past, but the solution to this is a larger cultural shift in which we believe in and invest in our local companies.

  4. Bobh March 16, 2012 at 8:40 am -

    Hard not to be in favor of this, but my understanding is that the information on drug prescribing does not have the medical problem being addressed, or indeed the full list of problems that the patient has. If this understanding is correct, then how can researchers determine best practices? Physicians would have to have coded each patient visit with an ICD9 code for a reasonably accurate description of the problem. That is not done.

  5. John March 27, 2012 at 8:58 pm -

    Very much agree!
    The application of privacy in BC healthcare hinders our ability to share data and deliver optimal services.  We bureaucrats spend an inordinate amount of time completing Privacy Impact Assessments, information sharing agreements, reviews and meeting about the perceived issues – lots of talk and no coherant and consistent application of privacy across the Health Authorities.   Everyone is well-intentioned; however, the outcome is leaving our patients' care compromised.  As one doctor states, "no one went to their grave happy that their privacy was in tact!"
    Releasing the anonymized datasets to the public following similar models (see: http://www.google.com/publicdata/directory) is a great plan.  The potential beneftis far outweigh the risks.  I can only imagine how creative our BC students/researchers could be with access to this information. What a great way to spur practial innovation. 
    One of the biggest challenges is the quality of the data.  Our health information systems have evolved in silos.  There are multiple "buckets" of data that is of variable quality.  The data is difficult to link across systems, and structures are inconsistent.  Terminology standards have not been followed across the Province and coding/abstracting practices are variable.   On the otherhand, we do have a repository of medication histories, 20+ yrs of cancer data and the Ministry of Health must have loads of MSP billing data.   We have a master patient index key'd from our PHN.  Lastly, we are only 4.5 million people – it's really not a lot of data. 
    In short, the data sources aren't great, but they are there and there is hope.
    Now only if we could find the time to do a "public data" project in addition to keeping our legacy systems running, and spending the budgets reinforcing the silos.  Oh – and then there's another meetng on privacy/audit to attend.