82 ◾ David Z. Cowan and Joyce T. Siegele
the system is only a tool to use in identifying opportunities for improvement, as baseline data
for more detailed stang studies, and for an illustration in mentoring and coaching managers.
Productivity monitoring systems are almost always built into accounting systems, but because the
information is buried and lacks graphic representation, standalone systems are almost always pre-
ferred. Productivity systems report on a monthly or biweekly schedule. e denition of outputs,
the development of targets, and the analysis of the reports for opportunities should all be included
in the role of the management engineer.
Data mining: is is an excellent source for developing productivity standards because it
enables the management engineer to have access to several data points without having to gather-
ing them personally. Information systems that are in place today allow us to mine the data avail-
able to develop productivity standards much more easily than before. An example of this would
be radiology. Within the radiology system, one can obtain tens of thousands of records within
a short timeframe to determine the amount of time to perform dierent radiology procedures.
e radiology information system can report the type of the exam or procedure, modality (e.g.,
computerized tomography [CT], magnetic resonance imaging [MRI], ultrasound, etc.), start time
with the technologist, end time with the technologist, and patient identier. Calculations can be
made to determine the length of time required by exam/procedure or by modality. Depending on
the level of detail desired, productivity standards could then be developed for the department as a
whole with procedure mix based on the point in time that the data was pulled. Another data-rich
department is surgery. Surgical information systems can report the provider, procedure, patient
identier, time entered in the operating room (OR), start time for the surgeon (cut time), stop time
for the surgeon (close time), and time exited OR. Productivity standards could be developed or a
weighted average determination of a standard can be applied to a case minute.
Stang studies: Detailed analysis is often required when a department confronts signicant
changes. e work of a healthcare department is often complex and dynamic. e representation
of productivity by a simple ratio of hours per unit volume rarely represents the broad range of
tasks and services or the roles of varied skilled workers. e management engineers will need to
immerse themselves in the department to understand the work—what is done, when it needs to be
done, how it is done, and so on. is often involves many other tools described in this book such
as process improvement, quality management, and Lean/Six Sigma. But the end result needs to
recommend a new department organization, stang budget, and sta schedule to reect the new
demands of the department. Almost always, the request comes from the department manager with
an expectation of increasing stang levels and ends up with a redesign of work and budget-neutral
sta reorganization.
BUILDING BUYIN
e hospital’s pharmacy department expenses were growing beyond the budget because
new regulatory requirements added a number of new steps to each process; many tasks
were taking twice as much time to complete as before. e management engineers, working
alongside the pharmacists and techs, observed the new procedures. e regulations certainly
increased the workload and new productivity standards were developed to reect these
changes. e management engineer and the pharmacy director searched for any opportu-
nity to nd new labor eciencies in other areas of the department’s function to limit the
request of additional budgeted sta and expense. Together they reported back to the COO
with the stang budget request and the revised productivity targets.
Workforce Management ◾ 83
Sta scheduling: Scheduling sta in a 24/7 operation can be quite complicated. Most depart-
ment managers will build schedules manually, but there are many automated scheduling systems
in use, particularly to support the stang of inpatient nursing units. Stang algorithms that drive
the automated systems can also be used to support manual scheduling. ese stang matrices
(described in a following section of this chapter) simplify the determination of how many sta
are needed on each shift based on the predicted workload. A scheduling process can be developed
to help managers set up the schedule to avoid the need for overtime, the use of work stretches
(consecutive days of work) that are too short or too long, to fairly allocate unpopular weekend and
night shifts, and to assure that the right mix of skills and leadership are available on each shift.
Managers without a computerized system can spend many hours preparing biweekly or monthly
sta schedules. But an engineer can help a manager establish a methodology to more quickly build
a schedule with the needed coverage and the best t to employee expectations.
Stang matrix: One tool that is helpful for managers as well as frontline supervisors on
inpatient nursing units is a stang matrix. is is a simple table relating the unit census to the
shift-by-shift stang needs. is is based on an average acuity level for each patient. In a previ-
ous section of this chapter, a discussion of a method of classifying patients based on their clinical
acuity and care needs is another level of renement of this stang matrix. In order to develop a
stang matrix, one must start with the productivity standard. A typical unit of measure for inpa-
tient nursing units are hours per patient day (HPPD). ere are several dierent methodologies to
develop HPPD productivity standards (e.g., nurse stang ratios, acuity-based standards, etc.). e
HPPD standard is the variable component of the standard for the inpatient nursing unit. ere
is also usually a xed component of the productivity standard relating to leadership and clerical
support as well as other activities that are needed to support the unit overall. Both the variable and
xed components are utilized in developing the stang matrix.
To develop a stang matrix, one typically starts with the census levels that apply to the inpa-
tient nursing unit. Most matrices start at a census of 1 and stop at the total number of beds physi-
cally available on the unit. Another methodology centers around the census levels that are most
common on the particular inpatient nursing unit. One then does the mathematical calculations
at each census level with both the variable and xed components to develop the number of hours
available to sta at that census level. A management engineer can develop a rst pass of the appro-
priate stang at each census level. However, it is usually best to have nursing leadership involved
as well, especially regarding appropriate skill mix.
If, for example, a unit has a nurse-to-patient ratio of 1:5 and a tech-to-patient ratio of 1:10, one
would want to use that as the basis for developing the stang level at each census. Certainly, the
census levels that are in between the whole number dividers make it challenging to determine the
appropriate skill mix and stang levels. is is when it is helpful to involve nursing leadership.
Once the stang matrix has been developed and agreed upon, it is a benecial guide to frontline
supervisors in determining the number of sta needed at each census level. Critical thinking and
clinical judgment are also factors in determining the number of sta, especially if the census is
right at the point where an additional sta member may be needed. Frontline supervisors appreci-
ate having a tool such as this in which to base their day-to-day and shift-to-shift stang decisions.
Managing premium pay: One of the challenges of building stang exibility is the reliance
on nurse stang agencies and overtime. Nurse stang agencies have a pool of nurse employ-
ees that are dispatched to area hospitals on short notice. ese nurses typically work for higher
wages without benets. Hospitals pay the agencies 1.5–2 times typical wages to use these nurses.
Hospitals with “holes” in their sta schedules nd it convenient to just call the agency to ll the
gaps. In most healthcare organizations, the use of these agency nurses might grow over time to be
84 ◾ David Z. Cowan and Joyce T. Siegele
very signicant and the extra cost (premium pay) of stang with these nurses over routine sta
becomes a “budget buster.” Hospitals can rein in these costs with better sta scheduling, but to
maintain the stang exibility, the hospital can build their own temporary stang pools. ese
pools can be of PRN (as needed) sta. ese are part-time sta that are willing to be considered
for scheduling at the need of the organization. ere is a need for a fairly large pool of PRNs, and
scheduling coordinators may need to call many of these sta before lling all the “holes.” Another
interesting approach used by many hospitals and aided by web-based applications is shift bid-
ding. After a sta schedule is developed, the unlled slots are oered to the PRN sta to indicate
which shifts they are willing to ll. ose slots that are dicult to ll may be incentivized with
a premium pay (usually much less than an agency would cost). Overtime is also a key to sta-
ing exibility. Casual overtime is an issue of management and process improvement eorts and
should be made to eliminate it. But many times, overtime is the most economical decision to meet
workload requirements. It is almost always more desirable than overstang and agency stang.
e management engineer can help develop rules for appropriate overtime use. In most acute care
organizations, overtime in nursing units of 2–3% and in the emergency department and surgical
suite of 3–5% is usually appropriate and represents good sta management (if the other indicators
show that the total hours per patient is at or below industry averages). Addressing premium pay,
building optimization models, and providing managers with eective stang decision support
tools will almost always result in cost savings and improved stang levels.
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.