Useful Clinical Data: The Biggest Casualty in HIT Non-Adoption

March 16, 2011

While many industries are wringing even more efficiencies out of their IT systems investments, many in healthcare circles are still quarreling over whether healthcare IT is even a good idea.

In most industries, the migration to automated and largely paperless systems happened many years ago.

Industries such as shipping, retail, hospitality, financial services, and travel are on their fourth- or fifth-generation systems, and routinely “data mine” that information to spot important customer trends and to improve internal efficiencies. In other words, they use it to reduce costs and improve the customer experience.

By contrast, many healthcare facilities and practitioners are on their first- or second-generation systems, and there are many influential industry pundits who argue that electronic systems do not even belong in healthcare. Indeed, while many industries are wringing even more efficiencies out of their IT systems investments, many in healthcare circles are still quarrelling over whether healthcare IT is even a good idea, arguing that healthcare is too complex to automate, that healthcare IT systems are not yet mature enough, that it would cost more to implement than it is worth. Some influential thought leaders even claim it would harm patient care and reduce quality.

These negative opinions exist in spite the enthusiastic support of many practitioners who have made the leap of faith and gone electronic, the quantum improvements in HIT software feature/functionality, the tremendous price drops on ever-more-powerful hardware systems, and the billions of dollars promised by ARRA/HITECH to stimulate IT adoption in healthcare.

People can poke fun at the old-fashioned clipboard and mountains of paper charts. They can scoff at the number of file folders stacked high throughout a typical medical practice, the lost charts and the lack of privacy - with health records frequently in full view of patients and visitors. They can complain about how hard it is to get information electronically from your provider or clinic, and the annoyance of a lost chart or lab result. The cost of searching for and retrieving a patient’s medical file - and the cost of replacing a lost file - should be obvious to most people.

These inefficiencies, however, mask an even bigger casualty - quantitative clinical information. Because clinical information is still stored largely on paper, or in isolated silos of digital storage, it means that at least one trillion clinical data elements are essentially lost or functionally unavailable in the United States today.

Here is the math: The population of the United States is approximately 300 million. For every one of us, it is not unreasonable to assume that there are 20 different “healthcare records” floating around, when you consider all the many doctors’ office visits, hospitalizations, annual physicals, checkups, labs, imaging tests, specialty clinics, and treatment center visits over the years. Let’s assume each paper record consists of 20 pages (a very conservative estimate), and each page contains on average 10 clinical data elements, such as blood pressure, weight, diagnosis, treatment, etc., and of course a date.

If you multiply these numbers together, that means there are more than 1.2 trillion clinical data elements on the U.S. populace! This is vital information that cannot be readily compiled, tracked, quantified, sorted, searched, measured, or otherwise made into some useful form to support improved patient care.

This is indeed tragic. We can spot trends in the financial market on thousands of investment instruments down to the minute, 24x7x365, and even apply literally dozens of complex analytical tools to track trends. We can follow the votes on “American Idol” in real time. We can get instant updates on the over/under on any major sporting event in the world. You can track a package as it winds its way from a web retailer to your business. You can even watch the progress of your pizza being made and delivered to your door.

But if you - or even your physician - want to just produce a simple graph of your own blood pressure or weight over time, which are two parameters that have been shown to be linked to the leading causes of death, you cannot do so without extreme effort and lots of manual processing – if the data are available at all. At best, your provider can only look at your last few exams - if he/she has the paper chart handy - and determine if your numbers are “up” or “down,” without a real sense of the time element and numeric scales involved.

The lack of automation means that none of this data can be easily tracked and utilized in any meaningful way. It cannot be used by providers to track the trend of an individual patient in the practice. It can’t be used to easily link things like obesity or drug use or any one of a number of environmental and behavioral factors on patient health. It can’t be used by public health leaders to monitor the epidemiology of a patient population.

It can’t be used by health informaticists to track and implement quality measures. No, this information can only be gleaned through expensive data-harvesting studies, where pre-selected paper records are accessed manually and those data elements that have been predetermined as being potentially important are entered into pre-set templates. (As a side note, this manual data entry leads to errors of its own.)

These studies are typically large and costly, so they are commonly funded by the government or educational institutions, but also frequently by pharma companies or device manufacturers (who may have a vested interest in showing certain outcomes). But these endeavors are so exceptional that the resulting conclusions - if any - are considered incredibly rare and valuable.

That’s because they are indeed valuable, but they shouldn’t be rare.

In short, for something that is vital to every person in the United States, and which represents fully one-sixth of the U.S. economy, it is almost a tragedy that we have over a trillion potentially useful healthcare data elements stuffed into file cabinets, sitting in stacks on billers’ and coders’ desks, languishing in providers’ duffle bags or stored in cardboard boxes in cavernous warehouses, that could otherwise be used to lower healthcare costs, improve patient care and make our lives better - and we can’t get to it.