Improving Emergency Department Cash Flow Via Lean Six Sigma

A not-for-profit hospital system noticed that many of its patients were routinely leaving the E.D. (emergency department) without providing complete insurance verification information or making appropriate financial arrangements. This information was critical to ensure a timely and accurate billing process and cash flow.

In addition, the hospital system was only collecting 30% of the charges billed for E.D. patients that were treated and released and less than $300 per day in co-pays.

The hospital’s goal was to increase the collection ratio to 40% within six months by securing each patient’s financial responsibility information before they were discharged. It sought a clearly defined and consistently applied process to capture this information for every patient, every time.

In addition, the hospital wanted to increase the amount of cash collected from patients at the time of service from $300 per day to $1,200 per day.

The project scope was unsecured patient accounts. The beginning boundary for the project was the time a patient arrived in the E.D. The ending boundary was when the patient was discharged from the E.D. and financial responsibility was secured.

The team selected the DMAIC process as its improvement method. A Six Sigma DMAIC project is defined as a project that eliminates a chronic problem that is causing patient dissatisfaction, defects, costs of poor quality, or other deficiencies in performance.

Using the previous year’s data, the hospital outlined its current process capabilities. There were total charges of roughly $71 million and 65,756 hospital visits. The median billed charge per visit was $1,080. As previously calculated in its eight-week analysis of E.D. gross revenue charges, there were $356,000 in unbilled accounts. This translated to 41 defects
per week ($356,000/$1080/8 weeks).

The estimated current process capability was:

• Sigma Level: 3.4 (short term)
• % defects: 3%

The target sigma level was 4.0 or greater, meaning there would be less than one incomplete account per day.

Using a cause-and-effect diagram, the hospital team identified some of the major factors that were contributing to unsecured accounts. They included:

1. People – Probable root causes included each person’s level of training and experience, possible bad habits, and which registrars were available.
2. Patients – Probable root causes included the severity of the patient’s physical state, as well as their lucidity. Each patient’s primary language was also a factor, along with whether or not he or she had their insurance card on hand.
3. Facility – Probable root causes included the location of the registration station(s), as well as the number of unsecured exits from the E.D. In addition, there was no defined checkout process or location within the E.D.
4. Regulatory – Probable root causes included meeting EMTALA, HIPAA, and JCAHO/CMS requirements.
5. Tools – Probable root causes included the proximity of computers, access to wireless Internet, access to the online eligibility system, and access to Medipac.
6. Process – Probable root causes included inconsistency in registration and the checkout process.

Left Without Being Seen (LWBS) was the primary reason for unsecured accounts. The majority of accounts were not secured because of patients who left from the lobby without being seen. Over a third of the accounts were not secured because the patient left before being checked out.

1. Co-pays were being collected on just more than 3% of the patients in the E.D.
2. The majority of E.D. patients did not have co-pays due. The top three reasons co-pays were not collected were: 1. Self-pay patients were not asked for a deposit. 2. Patients requested to be billed 3. The hospital was simply not asking.
3. There was a lack of a defined, consistent checkout process; patients were “escaping” from the E.D. before account information had been collected or verified.
4. For patients who were seen, the providers or nurses were discharging them before the registrar could return to the room or simply losing track of them after they had seen a provider.
5. Communication was poor between clinical staff and registrars.
6. Registrars needed help in knowing when the patient was available to interview and verify the information.
7. There were problems with patient flow; patients were leaving from several exits.

Once detail-level designs were created and approved, the project team developed an implementation and control plan that included:

1. Bedside registration of emergency patients using wireless data entry.
2. Central intake/registration of non-emergent patients (level 3-5).
3. A hardwired checkout process where all patients are to be escorted to the checkout desk.
4. Integrated clinical care process and registration.
5. Visual alerts to improve communication between providers and registrars when patients were available to complete registration.
6. Process transparency and operational metric reporting.
7. Training for registrars and clinical staff of the process changes and new data collection tools. ▪ Adding a back-up workstation to the checkout desk for peak periods of patient flow.
8. Collecting data on key metrics at the end of every shift. ▪ Instructing hospital registrars to always ask for co-pays and inform patients of their financial obligation.