June 20, 2011
Regulations Cramp Coder Productivity
By Cheryl Servais, MPH, RHIA, and Cindy Doyon, RHIA
For The Record
Vol. 23 No. 12 P. 8
The fact that there is a coder shortage is nothing new, but the situation is getting worse, according to industry reports. Increased reporting requirements and other government initiatives may be part of the cause. The coding shortage creates a backlog of uncoded records that delays claims submission to payers and increases the number of days in accounts receivable.
While a coder shortage presents a challenge, it also offers an opportunity to monitor coder productivity and develop plans to augment staffing or improve coder efficiency.
Productivity Study Results
A recent Precyse study measured the productivity of coders who completed a large volume of inpatient records. Specifically, it tabulated the difference in the average number of charts coded per hour in 2009 vs. 2010. In 2009, the coders completed a total of 64,107 records, an average of 2.38 inpatient charts per hour. In 2010, the average dropped to 2.1 inpatient charts per hour, a decrease of nearly 12%.
Put another way, in 2009 a coder could complete an average of 17.85 inpatient records per day (7.5 hours) compared with 15.75 charts per day in 2010, a decline of 2.1 records per day. To code the same number of records in 2010 as were coded in 2009, a coder would need to work at least 20 additional hours per month.
If those lost coding hours are not replaced, the unbilled claims backlog could grow quickly. For example, if a facility has just one inpatient coder, the backlog after the first month would be approximately 40 records. Assuming an average reimbursement of $6,000 per record, the backlog represents nearly $250,000 in revenue that cannot be billed.
Why is there a productivity loss? There are several reasons.
Present on Admission
One reason for the decrease in productivity is the additional time required to enter the present-on-admission (POA) status of each diagnosis code. In 2007, the Centers for Medicare & Medicaid Services (CMS) launched a new initiative to classify codes to indicate whether the condition represented by the code was present at the time of admission or whether it developed during the hospital stay.
The classification system relies on the assignment of a POA code to most ICD-9-CM diagnosis codes. (External cause of injury codes, or E codes, are exempt from reporting.) The CMS uses POA to prevent a condition that developed during hospitalization from being counted as a major complication/comorbidity (MCC) or a complication/comorbidity (CC) that could increase Medicare severity diagnosis-related group (MS-DRG) reimbursement.
The CMS issued a list of codes that describes hospital-acquired conditions that could have been prevented if the hospital and the treating physicians had followed proper protocols.
With the implementation of the 5010 claim form (effective January 1, 2012), providers have the option to submit more diagnosis and procedure codes. Claim forms will allow up to 24 diagnosis codes (rather than the current nine) and up to 24 procedure codes (instead of six).
While some providers have already been coding up to 24 diagnosis and procedure codes, many have not. For these providers, there is an opportunity to increase the information submitted to payers and others about medical necessity and services provided.
It is unclear how much, if any, extra time will be required for coders to provide additional codes. In facilities where coders have been limited to nine diagnoses and six procedure codes, the change may require staff to spend more time reviewing records and determining which codes to assign. The coders will also need to determine the correct POA code for all additional diagnosis codes.
There are several reasons a provider may choose to increase the number of codes submitted. First, with an increased number of diagnosis codes, it will be possible to record additional MCC or CC codes. While it takes only one such code to influence MS-DRG assignment, the presence of additional MCCs or CCs makes it less likely that reimbursement will change if an outside reviewer feels there may not be sufficient documentation to support an MCC or a CC or if a reviewer determines that a particular code has been incorrectly assigned.
Also, with the government and private groups increasing their focus on quality of care, the addition of secondary diagnosis codes provides a more complete picture of the severity of illness. Coders limited to six codes often had to choose to either omit codes for ancillary procedures or codes for some procedures in complex trauma surgeries.
Entities such as recovery audit contractors, Medicare administrative contractors, and Medicaid integrity contractors are monitoring coding accuracy. In addition, codes are now the basis for reporting core quality measures—a key to value-based purchasing, creating physician report cards, and monitoring quality indicators.
With this increased scrutiny of code assignments comes greater attention to medical record documentation to support the most appropriate code assignment. If coders have questions about documentation, they must take time away from coding to send a query letter to the physician to obtain clarification, resulting in a loss of productivity.
How to Improve Coder Productivity
Facilities must have a process in place to monitor coder productivity by coder and record type. When recording productivity by record type, the following groups should be included (at minimum): inpatient, observation, emergency department, outpatient surgery, clinic, diagnostic services, rehabilitation, psychiatric, and skilled nursing. Many facilities have further subcategories to distinguish inpatient records by service type and/or length of stay.
Outpatient services may also be categorized by service type (eg, gastroenterology, cardiology, radiology), whether facility and/or professional fee evaluation and management codes are required, whether coding includes physician services, or charge-capture review.
When a new regulation, service line, or coding/abstracting system is implemented, loss in productivity is often attributed to a “learning curve.” It is assumed productivity will return to previous levels within a few weeks or months. While productivity will improve after a major change, it generally does not return to the preimplementation levels. It is critical to have detailed baseline production rates so productivity trends can be mapped over months and years. Any decreases can be measured and an action plan developed to accommodate the loss of record completion by coders.
Also, examine the workflow of the coding process. Is all critical documentation required to support diagnosis and procedure codes present in the record in a timely fashion? Is there clerical support to locate missing records, documents, etc? Do coders have access to all aspects of the hybrid record system, including the ancillary, ordering, and nursing systems? Is the follow-up query process effective and efficient?
The HIM director or an outside consultant can provide productivity statistics and workflow analysis. Once this information is gathered, steps can be taken to streamline workflow and improve coder productivity.
Another option is enlisting computer assistance to improve coder efficiency (document access, workflow, etc). If coders must access multiple systems to view all medical record documentation, consider software that will create a single record view. This software provides coders with a seamless view into multiple systems through a single sign-on process.
Coder workflow software allows a manager to assign records or documents into work queues. Operating similar to transcription workflow software, it ensures that the oldest, highest-dollar, and most complex records are routed to the appropriate coder and tracked for completion. Computer-assisted coding can review text documentation and provide coders with a list of suggested codes for review. These types of software have been shown to increase coder productivity and improve accuracy.
Facilities looking to boost productivity should also develop a plan for supplemental coding staff. A March article in HFM Magazine addressed staffing for variable workloads. While the article focused on nursing services, a similar approach can be applied to any function that fluctuates with patient volumes. In applying the article’s advice to coding, a facility would staff to its average needs and augment high-volume periods with overtime, on-call staff, or vendor staffing services.
A facility may also consider outsourcing the entire coding function or the coding of a specific patient type or service.
For numerous reasons, coder productivity is diminishing. However, with careful analysis and an action plan to improve coder efficiency and workloads, a facility does not have to suffer from insurmountable backlogs.
— Cheryl Servais, MPH, RHIA, is corporate vice president of compliance and privacy officer and Cindy Doyon, RHIA, is vice president of coding services and West outsourced accounts at Precyse.