You only have to go back a few years to hear proponents of ICD-10 promising that more detailed codes would result in more timely and accurate claims and reimbursements. Was anyone really surprised, though, that adding more codes led to more coding errors?
ICD-10 coding has increased miscodes and lowered office productivity and morale due to the sheer number of codes (an eightfold increase!), their specificity, and the fact that the ultimate responsibility for codes reflecting what actually happened in the exam room resides squarely on the shoulders of the provider, whose time is the most valuable to the practice. Unfortunately, coding isn’t a skill providers are taught in medical school.
A very recent survey by ICD-10 Monitor reports improvements in some areas of coding accuracy between Q1 and Q2 of 2016, but miscodes for ambulatory services are at 17.3% overall, with some categories, such as symptoms, as high as 27.8%.
Missing, incomplete or inaccurate codes result in reimbursements being reduced, delayed or denied. Revising and appealing these claims cost the practice time and money, in some cases up to a third of the value of the claim. In other cases, claims may be approved, but they may be downcoded, bringing lower reimbursement to the practice. On the flip side, sometimes miscodes lead to inflated reimbursements, which if not rectified, can lead to expensive legal action against the practice. All these costs of miscoding affect medical practices by lowering revenue at a time when operating costs are escalating and margins are shrinking. It’s not just monthly cash flow that is at stake, but the financial viability of the practice.
Everyone involved with coding and billing needs to be adept at quickly identifying the correct billing code in every circumstance. This expertise depends on extensive training on ICD-10 codes for both providers and administrative staff—an expense of time and money that may be a hardship for many small practices.
In addition, billing staff needs the training and experience to exercise sound judgment in deciphering which medications and diagnoses are actually current amid an avalanche of data saved in lengthy and complex electronic health records. Otherwise, outdated information can impact submissions to, and decisions made, by the insurance provider.
Identifying the correct ICD-10 codes often involves time-consuming back-and-forth communications between clinical and administrative staff within the practice, staff at insurance carriers and government agencies, and oftentimes the patient, too. This is tedious even on those rare occasions when all the parties are available at the same time. The more typical scenario involves waiting on hold, waiting for information to be found, or waiting for the right individual to be available. This is compounded by the fact that staff members at insurance providers are also still coming up to speed on the new codes.
Spending more time on coding detracts from time spent caring for patients and time spent with family and friends outside the office. But, it’s not only office staff that is affected. Miscodes can be costly to patients, too, resulting in longer wait times, misdiagnoses, higher co-pays, denial of tests and treatments, and in extreme cases, loss of insurance coverage. According to a SERMO poll, 86% of 200 surveyed physicians report that ICD-10 implementation is negatively impacting patient care.
As with any new technology-based initiative, early adopters tend to be more technically proficient. As a result, it’s likely that miscodes will continue to increase as more providers transition to ICD-10. Physicians who are griping today about spending 20+ hours a week completing their patient notes, may soon think of these as the good old days.
One way to simplify the selection of ICD-10 codes is for the provider to be more descriptive when entering diagnoses. This can be a hardship for physicians typing in their notes. It’s much easier to type “pulmonary” than “pulmonary arterial hypertension.” Using voice entry for patient notes makes it much easier to enter a more descriptive diagnosis, which helps to narrow the field of code options. For example, “pulmonary” results in 100+ code choices, while “pulmonary arterial hypertension” results in only 15 choices.
Providers who rely solely on free-text notes for their patient records are at a much higher risk of miscoding because free text doesn’t generate a code and isn’t supported by billing software in the manner that structured data is. It’s likely that someone other than the physician will later use their judgment to decide upon and manually enter a code. In order to minimize the likelihood of billing errors, providers should utilize every possible opportunity to enter structured data into the EHR. With the EHR alone, or just the EHR and speech recognition software, structured data is entered via 100 or more annoying mouse clicks per note. Entering structured data by voice is faster and easier, so providers are more inclined to enter it, resulting in a patient note that supports billing and reporting.
NoteSwift helps providers to document in a manner that supports accurate coding. When NoteSwift is added to the EHR/speech recognition combination, it enables providers to use their voice to navigate through the note, enter more descriptive diagnoses and related treatment plans as structured data, and still enter their narrative text. With NoteSwift, patient notes are completed in fewer than 5 mouse clicks, ICD-10 coding is streamlined through easier entry of descriptive diagnoses, and accurate structured billing information is captured for timely, maximized reimbursements.
Art Nicholas is VP of Sales and Business Development at NoteSwift, Inc.