Automation + scheduling = an enterprise win: Duke implements automated and self-scheduling for more than 3,000 employees.

Scheduling the right personnel, at the right place, at the right time--and at the right price--is an ongoing and monumental challenge for any healthcare organization. Duke University Health System, however, has adopted an automated solution that streamlines scheduling activities by eliminating inefficient manual processes and facilitating each manager's ability to make the best possible use of available staff resources.

Identifying Our Needs

Duke University Health System, a top-ranked healthcare network in Durham, N.C., is comprised of Duke Hospital (995 beds), Durham Regional Hospital (369 beds) and Duke Health Raleigh Hospital (186 beds). Systemwide, we schedule more than 3,000 in-house and agency employees.

Traditionally, each of our facilities relied upon time-consuming manual systems to manage staffing and scheduling. The limitations of a paper-based process were compounded by the complexity of our work environment. A centralized float pool assigns staff for three facilities, and each site displays a different "culture"--adopting diverse temporary staffing policies, disparate activity codes and individualized training programs. As a result, traditional methods proved ineffective in ensuring fair scheduling across the system.

In 2003, we committed ourselves to finding an electronic solution that would:

* automate staffing and scheduling functions;

* optimize staff allocation by identifying staffing trends and assigning qualified internal employees wherever possible;

* increase cost efficiency by forecasting staffing levels accurately to minimize reliance on agency staff; and

* reduce scheduling errors.

Beyond these benefits, we wanted software that would help managers better understand staffing issues through robust reporting and analysis functionality that would allow us to compare anticipated needs to what actually took place. We also wanted to ensure that the software could integrate with the existing time and attendance system, and we wanted advance efforts to achieve consistent scheduling policies across the enterprise.

Finally, we recognized that our situation was also an opportunity to increase morale by expanding the use of self-scheduling, previously offered on a limited basis. Self-scheduling provides nurses with greater input to their own schedules and significantly increases job satisfaction. With today's shortage of nurses, we believed this feature would attract larger numbers of nurse candidates and help retain those already on staff.

Critical Planning Stages

We first developed a business plan to outline strategies for maximizing staff resources. We decided that the new application should not only address the challenges of scheduling nurses, but also exhibit the capability to be deployed in other departments like pharmacy, case management, radiology and special diagnostics.

At this stage, we assembled a 30-member technology search team that included staff nurses, nursing managers, clinical directors, administrators, information systems, hospital education, payroll and human resources. Our team conducted preliminary evaluation of 12 products and invited seven vendors to submit formal proposals. Besides evaluating the functionality of the software, we also assessed the level of customer service each company offered. Ultimately, four vendors provided on-site demonstrations.

At the end of the nine-month search, we selected the ActiveStaffer staffing/scheduling system from api Software Inc. (API) because of its extensive functionality and proven track record. Duke had deployed API's Time and Attendance system previously, and we knew the two applications would integrate with one another.

As we began planning for implementation, we selected one individual from each department to serve as a "functional administrator," charged with assisting in the development of policies and procedures, coordinating relevant rollout schedules, serving as the primary contact for functional and triaging technical questions. In addition, we assembled two teams:

The pilot, training and support team was composed of a functional administrator from each facility and the technical team leader, as well as representatives from hospital education and nine specific roll-out areas. This team was responsible for building department and employee information tables, creating schedules and testing schedule rules, as well as coordinating and executing training for all employees.

The health implementation team was comprised of a project manager, a functional administrator and a technical team leader, along with representatives from HR, payroll, hospital education and clinical operations. This group planned implementation activities, supervised rollout and provided support during the transition.

Focus on Training for Implementation

With the preliminary work accomplished, the pilot, training and support team devised a comprehensive training program to maximize success. Because each facility was transitioning from a manual system, we viewed training as an integral part of the implementation process and scheduled specific training classes prior to rolling out the system. At the core of the effort were 11 classes of "Scheduling 101." Each covered the theory and mechanics of developing a good schedule and included topics like:

* facilitywide scheduling policies;

* mathematical formulas to determine FTEs required to maintain a predetermined staffing level;

* formulas to determine how many weekend-option and off-shift staff are required;

* formulas to calculate the amount of paid time off managers could grant per shift, per scheduling period;

* unit-specific policies to ensure regular schedules are fairly distributed; and

* methods to determine whether 8- 10- or 12-hour shifts would be most advantageous to specific units.

The health implementation team began implementation in April 2004, initially rolling the system out among functional administrators. From there, we implemented the system incrementally, with about three to six departments in each rollout group.

It took about six weeks for groups to be brought fully up-to-speed. During that period, each group was introduced to the application through general information meetings, followed by Level I (setting up specific departmental schedules) and Level II (creating schedules and running reports) training sessions. Classes were offered to charge nurses addressing daily maintenance issues. We also offered Web-based training, which could be accomplished on an individual basis or as a group during staff meetings.

Goal Attainment

After more than two years of use, ActiveStaffer has enabled us to:

Monitor and control staff overtime. Not only does this help reduce costs, it helps ensure patient safety.

Eliminate scheduling of unqualified individuals. The automated system relies upon staffing rules and adheres to parameters that designate licensure, credentials and experience required for specific shifts on specific units.

Improve communication among departments and facilities. This allows us to allocate human resources more efficiently and prevents agency staff from inadvertently being scheduled in two places at once.

Improve communication with all employees. Staffers can view schedules on the Web and find out more easily if their self-schedule requests have been accepted.

Track competencies more readily. The system allows users to define specific fields to monitor licensure and educational class history.

Evaluate and improve business processes. Information about new employees is entered into the system centrally, so staffing rosters are up-to-date. In addition, the assignment of temporary staff is handled more judiciously, and related billing is processed in a more accurate and timely manner.

Audit scheduling activities. Comprehensive reporting allows us to identify error patterns and, subsequently, to better train users regarding how to achieve maximum benefits from the application.

Expand self-scheduling. We also have been able to expand the use of self-scheduling from 30 percent of our employees to 80 percent. Now we provide nurses access to the scheduling system from their homes via a secure Web portal for added convenience. While no quantitative analysis has been conducted, nurses tell us that this functionality has increased their job satisfaction and morale.

By replacing the paper-based process, we have eliminated a wide range of staffing-related problems, such as scheduling errors and ineffectual use of costly agency workers. Plus, we have freed up managers to pay more attention to "big picture" issues throughout the Duke system, like identifying trends and predicting staffing requirements.

For more information about ActiveStaffer and other staffing and scheduling solutions from api Software Inc., www.rsleads.com/701ht-203

By Sylvia Alston, M.B.A., M.S.N., R.N.

Sylvia Alston is administrative director at Duke University Health System, Durham, N.C. Contact her at [e-mail].