It was only about 10 minutes into the game when I fell on the soccer pitch this summer and tore a ligament in my knee. My subsequent trip to the hospital garnered me a specific diagnostic code that went to my insurance company. My insurer was then able to see why I sought care and billed accordingly. Despite significant upgrades in medical knowledge and care, the same thing would have happened a decade ago. Those codes have remained largely unchanged for more than 30 years. But soon a big change to that collection of numbers and letters may be making a splash—and should bode well for consumers.
Come October 1 a code update will go into effect that will take the current 19,000 diagnostic and procedure codes and catapult that number to 142,000. The transition promises to offer greater granularity to why we seek care. It gets wonky, but with the change researchers that deal in health data might gain more insight into what types of care get good results. That know-how could then trickle down into better care for you. Soon, instead of a code that simply indicates “torn anterior cruciate ligament” there will be separate codes that directly correspond to whether I tore the ACL in my right knee versus my left. Was it my first visit for care for this injury? The new coding system will cover that, too. Under the new system one code will indicate I tore my left ACL and this was my first visit for care. That larger compendium of choices will provide greater specificity for my doctor’s future reference and also for insurers trying to suss out whether my care was necessary. Yet one of the most significant aspects of this change continues to go largely ignored by medical workers bracing for rejected insurance claims and frustrations next month: More detailed medical billing codes could eventually improve your health care. Those new codes could provide a clearer picture of why individuals seek care and which health problems are growing or contracting in communities —helping inform what health issues should be researched and improved. At least, that’s the hope.
At the same time, some clinicians anticipate serious headaches as insurance companies and medical providers adjust to the new system—called the International Classification of Diseases (ICD-10). It is dizzyingly complex. Compared with the 15,000 diagnostic codes in the current system there will now be 70,000. The number of codes for inpatient hospital procedures—now totaling in at 4,000—will spike to 72,000. Many of the codes will not be needed on a regular basis (like V97.33CD, which indicates you were sucked into a jet engine, and this is your subsequent visit to a doctor). “The average internist probably won’t need more than 40 to 50 ICD codes for diagnosis,” says William Rogers, the ICD-10 Ombudsman for the Centers for Medicare & Medicaid Services (CMS) and a practicing emergency physician at Georgetown University Hospital. But officials overseeing the transition at hospitals and the doctor’s office are expecting a significant learning curve. “Other countries have said coders have become very confident in their coding probably within six weeks to six months,” says Lynne Thomas Gordon, CEO of the American Health Information Management Association.
In anticipation of these difficulties, CMS announced that during the first year of this new policy they will not reject valid insurance claims as long as health claims were in the right ballpark. That means if you coded for heart failure but did not click the most specific code for “heart failure” the physician will still get paid (or the insured patient will still be reimbursed). “The policy says you didn’t get the exact right one, but you got the right category,” says Pat Brooks, a senior technical advisor for CMS.
Next month, complications at rollout are expected to match the scale of concerns about Y2K, according to Lisa Iezzoni, director of the Mongan Institute for Health Policy at Massachusetts General Hospital. “I think it will be akin to the concerns with the millennium because we are talking about new alphanumeric systems,” she says. For ICD-10, computer systems needed to be altered to accept billing codes that might begin with any letter (as opposed to mostly numbers)—and it’s a big change, she adds. In the end, however, “the millennium happened without a whimper and I think that will happen here as well.”
The codes actually represent a U.S.-specific tweak of the ICD, a set of World Health Organization categories used internationally to record causes of death. The current U.S. codes classify types of sickness or procedures and largely adhere to the state of medical knowledge circa 1975—which is another reason it is time for a change. Every other industrialized country has already made the transition, Gordon says, including Iceland and Australia. “We didn’t have any code to monitor Ebola. I think it was embarrassing for our country but now we will catch up with everybody,” she says. Over the years, “hundreds” of small changes occurred on an ad hoc basis when physicians requested them, says Nelly Leon-Chisen, the director of coding and classification at the American Hospital Association, a group that helps educate hospitals on how to use the codes. New codes distinguishing the types of skin cancer were added, for example. But for diagnoses the codes only had so many numerical options before running up against another category of disease so there was an artificial cap on what could be added.
For researchers, the new system will offer the difference between “knowing there are apples in the supermarket and if there are Granny Smith apples versus McIntosh apples,” Gordon says. Some medical examples: the new codes will specify what trimester of pregnancy a patient is in when she seeks care. When it comes to orthopedics there is also more detail about which particular bones or tendons are affected.
The promise of gleaning data from these codes is greater than with electronic medical records. Those patient records usually differ across health care settings so there is often no easy way to combine them and extract health data. These medical billing codes, by contrast, will be uniform across the nation. “It’s true you can look at a paper record or electronic medical record, but it takes a whole lot of time,” Brooks says. “If you have nationally reported codes, you are talking about a simple program with trend analysis with thousands of patients.” Right now, even with the less precise codes, the codes help drive research on the quality, cost, accessibility and outcomes of health services. They also help identify trends in care. Will better codes eventually lead to better health? That’s what researchers are counting on. After all, the changes are what the doctor ordered.