Nurse Staffing and Patient Outcomes
Study Workgroup – December 9, 2013
Meeting Summary (DRAFT)
Meeting Details
Date: Monday, December 9, 2013
Start/End Time: 2:00 – 3:30 p.m.
Location: Stassen Office Building, St. Paul
Project champion: Diane Rydrych, Minnesota Department of Health (MDH)
MDH project staff: Stefan Gildemeister and Nate Hierlmaier
Facilitator: Kris Van Amber, Management Analysis & Development (MAD), Minnesota Management and
Budget
Workgroup members present: Connie Delaney, University of Minnesota School of Nursing; Marie
Dotseth, Minnesota Alliance for Patient Safety; Linda Hamilton, Minnesota Nurses Association
(MNA)/ Minneapolis Children's Hospital; Sandra “Mac” McCarthy, Essentia Health (by phone);
Steven Mulder, Hutchinson Health; Maribeth Olson, Mercy Hospital - Allina Health; Robert
Pandiscio, Minnesota Nurses Association; Sandy Potthoff, University of Minnesota School of Public
Health; Eric Tronnes, Abbott Northwestern/MNA; Vonda Vaden Bates, patient representative.
Others present (based on sign-in sheet): Wendy Burt, Minnesota Hospital Association (MHA); Julia
Donnelly, MNA; Tara Erickson, MLPNA; Sarah Ford, MHA; Shawntera Hardy, HealthPartners; Eric
Haugee, AFSCME Council 5; Andrea Ledger, MNA; Kristin Loncorich, MHA; Tara Mulloy; Bob Ryan, USW
District 11; Mark Sonneborn, MHA; Diane Twedell, Mayo; Nikki Vilendrer, Mayo; Anna Youngerman,
Children’s Hospitals and Clinics.
Workgroup members unable to attend: Shirley Brekken, Minnesota Board of Nursing; Betsy Jeppesen,
Stratis Health; Christine Milbrath, Metropolitan State University
Welcome and introductions
Stefan Gildemeister welcomed the group to the meeting. The goal for today’s meeting is to hear two
perspectives about data that may be relevant to MDH’s study of the correlation between nurse staffing
and patient outcomes. MDH hopes for a robust study, so workgroup feedback about these two
perspectives and how the data can be used in the study will be important. Today’s presentations will
also help the group develop a common understanding about what information may be available to MDH
to conduct the study.
Kris Van Amber asked workgroup members to introduce themselves. She asked workgroup members
and observers to avoid side conversations that might disrupt the group’s discussion or make it difficult
for members on the phone to hear what’s going on.
1
Conceptual framework debrief
Kris asked workgroup members to share their thoughts about the small group discussions at the last
meeting and about feedback shared via email after the last meeting. Member discussion included:













The small group discussion process was effective. Everyone had chance to talk about this
difficult and complex issue. We got a lot of work done in a short time.
We were able to see the gap between the ideal and the doable. For example, from a staffing
standpoint, the level of detail that would be ideal to develop a study with valid and actionable
results is difficult to come by and it varies across organizations. This is a big challenge.
There are many common benchmark data points that facilities collect, but these may not
account for all relevant activities.
We have to consider feasibility, and we have to realize that we are disregarding important
factors (hospitalists, for example). We may need to rank order the things we are forced to
disregard for feasibility reasons.
These were productive conversations about what things are most impactful and how to quantify
them.
It was important to talk about the many factors that affect patient outcomes and that affect
nurse staffing—it will be challenging to isolate these for study.
We share a common commitment to quality and patient outcomes.
We should focus on the research questions at hand and not go too far down other paths.
Even if hospitals don’t have consistent data on some measures, there may be other sources of
consistent information (like mortality rates) from JACHO or bargaining units.
Surgical Care Improvement Project (SCIP) data may be of use.
Best practices associations (such as pediatrics) might be collecting data on staffing and
outcomes.
Core measures data are available from hospitals. The correlation to staffing is the challenge.
Within those core measures, we find inconsistency—it’s not 100%. Not all hospitals have the
processes and systems in place to get the information in the right box. Staffing usually isn’t the
cause of variation.
Patient acuity systems presentation
Diane Twedell (DNP, RN, CENP), Chief Nursing Officer at Mayo Clinic Health System (MCHS) - Southeast
Minnesota Region gave a presentation on patient acuity systems. The text from her presentation slides
is at Appendix 1 on page 5.
Additional workgroup member and presenter discussion included:



The primary focus of the patient acuity system is to manage staffing volume and workload—to
be able to provide the critical care for an individual patient.
These systems can also be used as productivity tools—they may be used in conjunction or linked
with accounting systems.
The systems can have different factors for different areas, such as inpatient psych units.
2









Factors and formulas are built into the system—these may be different depending on the type
of hospital.
Data from the system can be used to assess staffing retroactively according to outcomes—it can
be used to correlate with patient falls on a night shift, for example. However, it is more of a tool
for scheduling and forecasting.
Data from EMR or ADT systems may be fed automatically into these systems, but additional
checking would be required to confirm how and whether this occurs for particular systems.
For these systems to function, they have to be maintained and monitored appropriately.
These systems look at all direct patient care staffing—LPNs, RNs, patient care assistants. It
doesn’t factor in other services like wound care.
The vendor doesn’t determine the factors and assumptions behind the system—nursing does.
Different hospitals use different systems: QuadraMed and API are two examples.
There are inconsistencies in these systems—it’s not comparing apples to apples. Many do not
capture important work like surveillance or psychosocial needs. There are lots of issues with
classification tools.
The weighting system behind these can be changed—this affects validity of comparisons (and
affects staffing assignments).
Kris recommended that workgroup members make note of any additional comments or concerns about
patient acuity systems and provide them to MDH.
Minnesota Hospital Association nurse staffing plan presentation
Mark Sonnenborn (MS, FACHE) VP Information Services, Minnesota Hospital Association (MHA) gave a
presentation on the MHA’s plan to collect and publish information on nurse staffing plans and data. The
text from his presentation slides is at Appendix 2 on page 8.
Additional workgroup member and presenter discussion included:





The legislation that requires the MDH study also required MHA to collect and publish
information about nurse staffing and nurse staffing plans.
The law has some specific directions about data, such as that the staffing data include
nonmanagerial, direct care staff over a 24 hour period. Staff spending less than 50% of their
time on the unit (such as float staff) would not be counted.
Some seemingly clear legislative directions were not entirely clear, such as inpatient care unit. A
lot of non-inpatient care is provided on inpatient units. MHA’s data will include any unit that
provides inpatient care, even if much of their work is non-inpatient care, such as blood
transfusions.
There was a lot of discussion about whether swing beds should be included; they will be
included in the MHA reports.
The staffing plans are due next month, and MHA has already begun collecting the data. They will
be published on the website by April 1; by July 1, there will be actual staffing data. The website
will display the budgeted data and the actual data for comparison.
3









MHA is reporting unit type based on the Labor Management Institute’s list, with an additional
type of “Critical Access Hospital-Mixed” because that captures an important distinction.
MHA is going beyond what is required by the law: reports will also a list of other staff that are
part of the care team, such as pharmacists, dieticians, hospitalists. This will be a list, not
information about amount of work.
The data will include staffing information for all nurses, not nurses aids or nurses assistants.
There is no differentiation between RNs and LPNs because MHA doesn’t interpret that as being
required by the statute.
The reports will be purely hours worked per patient averages; there will be no adjustments for
transfers in and out or for acuity.
In July (with the first report on MHA’s site), the general public will be able to compare staffing
information with publicly reported outcome information.
The data will be published quarterly, with average staffing data per 24 hour period presented—
this means one ratio (staff to patient) per unit per quarter, not 120 daily averages.
Outcome data is also aggregated, but not by unit.
MDH was charged with studying the correlation between nurse staffing and patient outcomes—
it’s not clear how the MHA data will help that study.
The MHA data is one source of information—the workgroup should consider how and whether
that will help the study, and what other sources may be available.
Closing comments
Kris and Stefan asked the group to consider how these two data sources might help the study, and what
other information is needed to inform the group. A follow-up email with a specific request for
information will be sent to the group.
Kris, Diane, and Stefan closed the meeting by thanking the workgroup for their discussion today.
Future meetings
Due to some internal scheduling conflicts and competing legislative priorities, the January 13 and
February 10 workgroup meetings are canceled. MDH continues to consider options for how to most
effectively synthesize and reflect back to the work group what was learned so far, the approach to the
study, and which areas are needed to focus input next. MDH would like to make sure that the next
meetings uses the talents and time of workgroup members as efficiently as possible, and will be better
positioned to do so with additional time for internal planning . MDH will send information as soon as
possible about a March workgroup meeting.
4
Attachment 1: Content from presentation slides – Diane Twedell
Inpatient Workload Patient Classification
Overview of Content




Patient classification function and types
Staff RN role in classification of patients
Data accuracy and monitoring
Coordination of patient classification program
Patient Classification Major Functions


Calculate inpatient acuity and workload.
Determine number and kind of personnel required for patient care.
Patient classification systems do not measure:


A patient's severity of illness.
Nursing acuity cannot be used as a severity of illness index.
Two types of patient classification systems


Prototype - broad narrative descriptions of typical patients.
Factor system - classifies patients by one of two methods.
a. According to their need for specific nursing care.
b. Time standards associated with tasks.
Acuity:


Specific to nursing care requirements not linked directly to severity of illness.
Can correlate as ICU level patients have higher nursing care needs but there may be other
reasons why patients need high levels of nursing care.
Patient Classification Tool



Measures the workload of the direct care unit based staff.
Is based on the needs of the patient.
Credible and reliable: data collected, analyzed and tested.
Patient Classification Tool




Methodology has specific indicators that can be selected based on patient need.
Inpatient
Mental health
Perinatal
Weighted indicators



Place the patients into appropriate category for overall nursing care hours.
Are used in combination with one another, workload is not driven by one indicator alone.
Take into account the overall patient need.
5
Workload


Each patient category has an associated acuity value.
Workload is calculated utilizing:
a. Acuity value
b. Number of patients - census
c. Other workload - activities/procedures, ADT activity, care for patients off of the unit.
Staffing



Calculated utilizing the measured workload and hour of care per measured workload.
Prospective staffing - forecast for shift by shift and next day.
Retrospective staffing analysis - day of week by shift patterns, seasonal patterns, hour of day
patterns.
Direct Care RN inputs workload




Expectations clearly communicated.
Integrated into their work on a daily basis.
RN providing care classifies the patient.
Charge RN checks to ensure classifications are completed.
RN staff engagement







Orientation and education of staff.
Staff are supported to staff to patient care need.
Staff see the data being utilized.
Staff input for workload measurement tool enhancement - some serve as monitors for data
collection.
Ongoing education on indicators and importance.
Using reports effectively.
Recommended versus actual staffing - applying critical thinking and nursing knowledge to
recommendations.
Data Accuracy: Monitoring Classifications




Ongoing monitoring every classifying shift, every pay period.
Transparent reporting
Presence on unit.
Opportunity for staff to ask questions.
Classification Monitoring Program


Staff RNs.
Set units to monitor - not their home unit.
Data Accuracy in addition to monitoring

Internal and external benchmarking
6

Analyzing workload trends - practice changes, population changes.
Patient Classification Program Coordination







Education.
Monitoring program and oversee data collection.
Maintain system parameters.
Data analysis.
Consultation.
Liaison to vendor
Keeping data alive - building and continuing the culture.
Creating a Successful Program



Consistent messages.
Awareness of current practice and changes.
Equality, same set of rules throughout.
Final thoughts:


Data use and analysis can be powerful.
Be open to what the data may uncover.
And the Bottom Line

The right staffing to successfully care for patients each and every day.
7
Attachment 2: Content from presentation slides – Mark Sonneborn
MHA’s Hospital Unit Staffing Website
Subd. 1 - Definitions
(b) "Core staffing plan" means the projected number of full-time equivalent nonmanagerial care staff
that will be assigned in a 24-hour period to an inpatient care unit.
(c) "Nonmanagerial care staff" means registered nurses, licensed practical nurses, and other health care
workers, which may include but is not limited to nursing assistants, nursing aides, patient care
technicians, and patient care assistants, who perform nonmanagerial direct patient care functions for
more than 50 percent of their scheduled hours on a given patient care unit.
(d) "Inpatient care unit" means a designated inpatient area for assigning patients and staff for which a
distinct staffing plan exists and that operates 24 hours per day, seven days per week in a hospital
setting. Inpatient care unit does not include any hospital-based clinic, long-term care facility, or
outpatient hospital department.
(e) "Staffing hours per patient day" means the number of full-time equivalent nonmanagerial care staff
who will ordinarily be assigned to provide direct patient care divided by the expected average number
of patients upon which such assignments are based.
Subd. 2. Hospital staffing report.
(a) The chief nursing executive or nursing designee of every reporting hospital in Minnesota under
Minnesota Statutes, section 144.50, will develop a core staffing plan for each patient care unit.
(b) Core staffing plans shall specify the full-time equivalent for each patient care unit for each 24-hour
period.
(c) Prior to submitting the core staffing plan, as required in subdivision 3, hospitals shall consult with
representatives of the hospital medical staff, managerial and nonmanagerial care staff, and other
relevant hospital personnel about the core staffing plan and the expected average number of patients
upon which the staffing plan is based.
Subd. 3. Standard electronic reporting developed.
(a) Hospitals must submit the core staffing plans to the Minnesota Hospital Association by January 1,
2014. The Minnesota Hospital Association shall include each reporting hospital's core staffing plan on
the Minnesota Hospital Association's Minnesota Hospital Quality Report Web site by April 1, 2014. Any
substantial changes to the core staffing plan shall be updated within 30 days.
(b) The Minnesota Hospital Association shall include on its Web site for each reporting hospital on a
quarterly basis the actual direct patient care hours per patient and per unit. Hospitals must submit the
direct patient care report to the Minnesota Hospital Association by July 1, 2014, and quarterly
thereafter.
8
MHA Website – Descriptive Info



Name of unit (e.g “3 West”)
Type of unit (e.g. “Medical” or “ICU”)
o Using list of types from Labor Management Institute
 Added a “Critical Access Hospital-Mixed” unit
 Allowing “write-ins” if unit type is not on list
List of other care team members
o Just a list, no other data
o May be available to multiple units
MHA Website – Staffing Info



FTE/hours worked per 24-hour period, split by:
o Nurses
o Other Assistive Personnel
Average number of patients per 24-hour period
Staffed Hours Per Patient Day (calculated)
Outcome measures


Publicly available:
o SQRMS
o CMS
o Adverse Health Events
Measures generally available at hospital-level
o I.E., not unit-level
Summary prepared by: Beth Bibus, Management Analysis & Development, Minnesota Management &
Budget
9