T his is the second installment in a series of articles that began in the May 2004 issue about a radiology group’s experiences moving from “average to peak performance” within a designated time frame. Radiology Consultants of North Dallas (RCND) agreed that the first emphasis should be upon achieving measurable improvement and breaking out of average performance when compared to industry business indicatorsthen setting new business performance levels.

The goals for Radiology Consultants of North Dallas are ambitious and rely upon the achievement of quantum change. At the heart of the program is the goal of moving business performance standards from average to peak performance within 12 to 18 months. This includes the achievement of new productivity levels, reduced administrative costs, greater profitability, and implementation of formal quality improvement processes to ensure continued refinement.

While it is possible to noticeably improve business indicators and productivity with modifications to existing processes, we recognized the achievement of peak performance goals would require a level of change possible only with the use of advanced technology to leverage staff capabilities and efforts. This meant seeking a “vendor plus” relationship, where leading edge software enhancements were matched by a willingness to work with the radiology practice “change team” to adapt and refine processes.

We were confident with the concept, our experience, and the vendor commitment. However, as usual in grand plans, the process of quantum change slams into the wall of reality. Any practice manager who has ever installed a new time clock knows change management is a challengeand we were out to change the fundamental operation and structure of the billing operation as a first step to becoming the radiology office of the future.

THE FIRST HURDLE

In the process of preparing for the practice management software conversion, we faced the first indication things would move more slowly than expected. The administrative office had previously installed several new computer workstations, staff was using the Internet for claims follow-up, and demographic information was downloaded from the hospital via modem. The framework appeared to be in placeright up until the technical conversion team began the task of networking computers.

We knew current workstations were not networked and would have to be in order to maximize the benefits of the new software. The existing “network” had linked workstations via commonality of the old practice management software, but not through the operating system. Different versions of Windows operating systems were found, including several desktops with Windows 98, which presented problems achieving sufficient security levels. That was the least of the problems.

The current configuration could be described as one that “just grew” as the number of staff members expanded over the yearsand Internet access was not accompanied by appropriate business-level controls. There had been no policies and procedures regarding Internet and e-mail usage, and firewalls were inadequate. As a result, a few workstations were clogged with virus infections. The technology team faced inconsistency, compounded by cleaning workstations and installing a level of security we were confident would meet the impending Health Insurance Portability and Accountability Act (HIPAA) Security Standards. After installing and checking system intrusion detection software, the technology team advised we were fortunatewhile there had been intrusion attempts, there were no successful invasions.

Network problems imposed delays of more than 2 weeks in the conversion process as problems continued to emerge during the task of standardizing and securing the network. However, final security audit and validation processes confirmed that firewalls, access limitations, and intrusion detection measures were successfully in place to protect from outside-in attempts. In addition, new Internet blocking mechanisms were implemented to restrict the downloading of potentially harmful attachments, as well as unauthorized use of Internet capabilities. The team also installed tracking mechanisms that could monitor employee use of Internet sites.

Additional support involved the development of appropriate policies and procedures regarding the use of e-mail, attachments, and the Internet. Employee training regarding security, policies, and procedures (and penalties) was then scheduled as part of the HIPAA security compliance plan development.

Before

After

Downloaded information reviewed account by account and compared to hospital database

New data definition files eliminate the need for manually comparing to second hospital database

Fields with missing or incomplete information drop to exceptions report for review and correction

Radiology reports manually matched with hospital orders for verification of “count” and coding

Radiology reports scanned into system and attached to patient accounts; coding info on order sheets picked up on new download format

Reports coded manually and other information added for person completing charge entry process

Codes pulled from download and verified and/or corrected by coder; radiology report and supporting documentation available on screen

Coded reports passed to charge entry

Charge entry occurs with verification of codes as elements from radiology report and order form are matched by software

Charges filed through clearing house

Charges filed directly to key payors and through clearinghouse for non-contracted and/or commercial claims

Table 1. Several manual front-end processes are replaced and/or simplified by the practice management system. Examples of several modifications are included in this table.

STAFF UNCERTAINTY AND UNREST

The initial staffing challenge was not unusual. Since we had elected not to move data from the old system to the new, it was necessary to ensure both databases continued to be workedto ensure not only a coordinated transition but also consistent cash flow as the new system ramped up. Theoretically, the new practice management software needed fewer people on key functions. Cumbersome manual front-end processes would be replaced by electronic ones, with examples as follows:

Since numerous tasks formerly requiring an 8-hour commitment of time would now be completed in minutes (or a minimal number of hours), we began to collapse job descriptions, which meant employees on the “new” team needed a broader range of skills. In addition, several people (payment posters, for example) would be required to shift back and forth between the two software versions and would therefore need to be comfortable with juggling different processes and software demands. The systems and processes were so dissimilar we knew staff members working the “crossover” functions would need to be among the more adaptable and technologically adept of the employees. While they appeared to be excited about learning new skills, it was evident they could also face greater frustration dealing with the vast differences in systems.

Other employees faced the threat of having their positions eliminated entirely (Table 2). Three of them resigned rather than be assigned to other job functions, and we elected to absorb their positions rather than hire, a decision that placed a greater level of stress on remaining employees. But they were willing to assume additional levels of responsibility while trying to learn new processes. While it was our intent to eventually reduce staff size, we did not anticipate doing so this early in the process.

At the same time, more rigorous, standardized processes initiated for insurance follow-up on the old system, combined with inherent software inefficiencies, added to the workloads of staff members assigned to back-end functions. This meant a disproportionate number of staff members would remain working accounts receivable on the old system, since its limitations had to be addressed with less efficient manual processes.

Ambiguity often represents the greatest stressor for staff members and those assigned to insurance follow-up would live in the land of “it depends” the longest. How quickly would they move to the new system? How would it impact their workloads and at what point would the remaining A/R be moved to the new system? Their positions currently involved reviewing monthly reports and working through them based on established criteriaalthough there was literally no way to monitor compliance with standardized procedures and priorities. In addition, they dealt with an increasing level of insurance correspondence, as the old software and clearinghouse functions experienced failed and refiled transmissions in recent months. We increased support functions, which included a person to pull and copy explanation of benefits (EOB) forms, radiology reports, and other necessary documentation, including prior appeals that had been filed.

Insurance follow-up was also destined to undergo significant changes, with “before and after” changes in processes and job descriptions. All information, including EOBs, correspondence, a copy of the HCFA form (even though filed electronically, a form could be created for viewing), and other pertinent information, was scanned into the system and stored by patient account. That meant, instead of looking for documentation through numerous file folders (especially if it was not filed correctly), virtually everything needed to correct and/or appeal a claim could be viewed, resubmitted, or printed from the employee’s workstation. In addition, rather than waiting for a monthly A/R report to be printed and/or correspondence to be distributed, employees would be working from an electronic distributed worklist based on criteria established by management. Rather than develop standardized procedures and hope for compliance, the computer system (and administrator) could assign priorities for follow-up on a real-time basis.

DENIALS MANAGEMENT PROGRAM

After hearing that up to 30% of claims are denied specialty-wide, the radiologists of RCND eagerly embraced implementation of a denials management programeven though it meant they might have to make changes too. Since our ability to pull solid information from the old software package was limiteddue to both the inflexibility of custom reporting and poor denials/adjustments classificationswe began with manually auditing radiology reports.

Table 2. The yellow boxes represent electronic processes and the green boxes represent tasks completed by a staff member. Approximate times and full time equivalent (FTE) staffing levels are noted for various steps in the process. The most notable point is that fewer people and less time are required in the ‘new’ processes, largely due to the elimination of data cleaning and manual matching of reports and demographics. Preliminary net gain in time and staffing demands from receipt of information to claims submission is logged, with further improvements anticipated.

While our initial sample size was not ideal, we pulled 25 hospital reports per radiologist randomly (a total of 525 reports) and compared them to the American College of Radiology communications guidelines. This “first pass” enabled us to identify whether there were problem patterns in dictation when compared to the ACR guidelines as an objective set of criteriaand we were primarily looking for problems that would make the reports difficult to code. Compared to audits completed for other radiology groups, RCND was very clean in terms of matching the ACR guideline criteria; however, there was still evident room for improvement that could translate into increased collections as better coding documentation was available.

Each radiologist received a binder containing the ACR communications guidelines, a summary of his or her results, samples of blinded reports from the group that had been denied for payment due to coding problems, and other resource materials. We discussed the significance of detailed indications for the examination so the coders could assign more specific codesand the radiologists offered suggestions regarding how they believed they could improve this information in their dictation.

The group was unusual in its level of cooperation and initiative, and requested additional resources to better educate its members regarding ICD-9 codes; we set dates to follow up on progress. During the interim period, we are receiving feedback from the coders that they are seeing improvement in patient history/indications information.

OTHER INITIATIVES AND CHANGES

With the staff heavily immersed in conversion changes, we knew there were several months ahead of getting through the mire of cleaning the old system A/R and fighting its limitations. The system allowed us to begin filing claims directly to key payors again in order to minimize the impact of problems encountered by the clearinghouseand we changed clearinghouses as well.

As firm believers in the precept “you can’t manage what you can’t measure,” other members of the change team focused on the development of meaningful management reports that would also help the radiologists better understand their business. Highlights included the introduction of the following, although the list included other areas of productivity as well:

  • The Radiology Business Management Association (RBMA) chart of accounts, not only for internal revenue/expense monitoring, but so we could begin to compare performance with industry averages and participate in surveys
  • An account resolution tracking system to measure the percentage of accounts resolved within 60 days (as developed and promoted by Claudia Dwyer, RBMA past president)
  • A trend report that enabled tracking of charges, adjustments, payments, refunds, days in A/R, and bad debt percentages, all on one page

In addition, the new computer system provided “real time” reporting, making it possible to monitor productivity by user as well as overall charge entry and payment posting progress and potential backlogs. Since most of this type of management reporting information is usually available with month-end close reports, this represented culture shock in that, for example, we could look at reports for 1:47:34 pm and see how many unbilled charges were in the system. We had to remind ourselves backlogs in various areas had always existedwe just did not have a good way to measure them, and therefore, it was important to react, but not overreact. This information allowed us to quickly mobilize staff to address backlogs or check on problems causing delays.

The agonizing process of securing the network and server room had met most of the HIPAA “technical mechanism” and “physical safeguard” requirements, but there was still much to be done with disaster and business continuity planningas well as policy and procedure development and training. We determined if we were to illustrate “best practices” performance, we definitely had to meet the compliance deadline for the Security Rule.

WHAT’S NEXT?

In our next article, we should be able to begin reporting measurable improvement although we know we will still be tightening down processes and adjusting to those many factors out of our controlsuch as a hospital download failure that continued for nearly 2 weeks. Performance improvement is, however, a way of life rather than a projectand requires constant adjustments. How well do we think we will be doing? It will all depend on our success clearing “old system” accounts receivable, refining new processes (including denials management), and ensuring our employees develop the skills we believe they must have to support the radiology office of the future. And then, of course, there is the challenge of meeting HIPAA compliance deadlines. We are behind schedule at the momentdue to the usual factors beyond the control of any radiology practice managerbut remain focused on our goals.

Patricia Kroken, FACMPE, is a principal of Healthcare Resource Providers, a consulting firm based in Albuquerque, NM. [email protected].

Phillip W. Carmody, MD, is a principal of Healthcare Resource Providers, a consulting firm based in Albuquerque, NM.