Analyzing ICD-10’s influence on billing operations a year after implementation
Last year, U.S. providers feared falling productivity, increased denials and reduced revenue as they prepared for the inevitable shift from ICD-9 to ICD-10. The new coding system increased diagnostic codes from approximately 14,000 to more than 68,000 and procedure codes from 4,000 to 87,000 with promises to improve quality reporting and outcomes measurement, and streamline reimbursement processes.
The United States was one of the last developed countries to transition to ICD-10, trailing years behind Australia, Canada and many European countries. This delay allowed the United States to learn from the mistakes of others, like Canada, whose healthcare system saw a 50% drop in coder productivity after ICD-10 was implemented in 2001. According to Canadian authorities, this drop could be linked to an underestimation of the amount of work required to implement ICD-10 codes.
Taking note from Canada’s loss, the United States concentrated on training and preparation of not only coders, but also health information management (HIM) staff and physicians. After a yearlong postponement, the new coding system went into effect on Oct. 1, 2015, requiring all providers covered by the Health Insurance Portability and Accountability Act (HIPAA) to be compliant with ICD-10 to avoid reimbursement issues.
While ICD-10 had the potential to have an extreme negative impact on the U.S. healthcare industry, especially on revenue cycle management, it was a successful transition due to the country’s extensive preparation. Today, nearly a year after implementation, we can analyze ICD-10’s impact on our billing operations and extract important lessons that can contribute to continued success.
Clinical Documentation Improvement and Case Mix Index
ICD-10 was not just a coding issue and providers who thrived in the year following implementation did not focus exclusively on coder education; they also dedicated resources to improving clinical documentation to support the transition. Doing so ensured providers met the level of coding specificity needed to obtain proper reimbursement, comply with regulatory requirements and accurately reflect patient care.
This attention to Clinical Documentation Improvement (CDI) programs and coding accuracy contributed to there being no significant changes in Case Mix Index (CMI) levels across the industry. For example, leading up to the transition, Quorum Health Resources (QHR) client hospital Clarion Hospital in Clarion, Pa., implemented a CDI program focused on the training and communications between physicians and coders to ensure documentation accurately reflected the high quality of care provided by the hospital. The hospital’s more accurate and specific documentation resulted in increased CMI and inpatient reimbursement.
Overall, better documentation leads to a better bottom line. This success relies on the ongoing assessment of infrastructure, processes and training of the clinical documentation team. To prepare for Oct. 1 updates, providers must review CC/MCC lists to confirm what codes have been removed and added.
Coder Productivity and DNFB Increases
U.S. providers were warned they should expect a 50% productivity loss comparable to what Canada experienced when they transitioned to ICD-10. However, extensive preparation in the United States., and coder training in the years leading up to implementation, allowed us to avoid a similar deficit. Providers that saw a successful transition to ICD-10 dedicated countless hours to education and training.
As a whole, the United States initially saw a decrease of only about 25% in charts coded per hour. This slight decrease in productivity caused minimal increases in Discharged Not Final Billed (DNFB) numbers due to increases in clinical documentation queries and medical necessity edits.
To put it in perspective, the American Health Information Management Association (AHIMA) reported that, on average, coders recorded 24 inpatient charts per day in ICD-9. With a 25% decrease in productivity, coders complete an average of 18 charts per day in ICD-10. This increases the DNFB because providers are holding claims longer, unless they grow their coding staff. These numbers vary from hospital to hospital depending on the hospital type and CMI. The higher the CMI, the lower the average number of inpatient charts coded per day will be.
Nearly a year later, productivity has stabilized and providers now see only a 10-15% decrease. This does not mean providers can relax, however. The Centers for Disease Control and Prevention and Centers for Medicare and Medicaid Services (CMS) recently announced the addition of 1,900 diagnostic codes, including 260 new diabetes combination codes formulated for reporting manifestations, and 3,651 inpatient procedure codes for fiscal year 2017. Providers, therefore, must continuously prepare and train for ongoing changes, and routinely evaluate coder and documentation performance to sustain productivity levels.
Payer Denial Behavior
Payers deny claims for a variety of reasons – procedures are improperly coded, care is provided outside of patient’s network or services are deemed as medically unnecessary. It appeared that payers took a conservative approach on creating additional claim review criteria during the ICD-10 implementation. For the most part, providers saw minimal payer behavior changes after implementation. If providers had problems with medical necessity in ICD-9, for example, they continue to face the same issues in ICD-10. Providers did see a slight increase in requests for additional information due to the specificity of the new codes.
Moving forward, providers must continue to monitor payer denials – especially in the coming months. Payers will soon have a year’s worth of ICD-10 data to understand critical patterns of care under ICD-10 and will likely adapt processes accordingly. Providers must be prepared for these potential changes to mitigate financial risk.
Providers that focus on improving clinical documentation, providing ongoing training of coders and continuously monitoring payer denial behaviors will see financial benefits in the second year of ICD-10.