Transcript ORGANIZATIONAL LEARNING FROM FRONT

Operational Failures and Interruptions
in Hospital Nursing
Anita L. Tucker, Assistant Professor, Wharton
and Steven J. Spear, IHI
Cincinnati Innovations in Healthcare Delivery
September 22, 2006
1
Motivations for studying nurses’ work environment

Nurses’ work has a direct impact on patient outcomes


Typical Policy suggestion: Increase Nursing Staff, but is
challenging




More nursing care = better outcomes (Aiken et al, 2002; Kovner et
al., 2002; Needleman et al., 2002)
Nursing shortage (Buerhaus et al., 2000)
Expensive
Increasing documentation takes time away from patient care
(Beaudoin and Edgar, 2003)
Need to investigate and improve nurses’ work environment
(JCAHO 2002, Page 2004)


Increase nurses’ job satisfaction and retention
Reduce work requirements that take time away from patient care
2
Alternate Recommendation

Improving work systems by reducing Operational
failures (Tucker 2004; Beaudoin & Edgar 2003)


Disruptions in employees’ abilities to effectively
complete tasks due to problems or errors in supply of
information or materials
Reducing operational failures can lead to


Fewer interruptions (may reduce medical errors)
More time to care for patients
3
Examples of Operational Failures on Nursing Units
Type
Problems
Errors
Information - Unsure what supplies
surgeon wants
- Change of shift doesn’t mention
patient was nauseous
- Patient not being observed by
any nurse
Materials
-
Missing linen
--Missing container
- Sputum sample lost on way to
lab
Equipment
- Can’t find thermometer
- Nurse mistakenly left sleeping
infant on ICU bed after transfer
to regular floor
Medications - Missing medications
- Nurse forgot to give patient his
medications all shift
4
Research Questions


To what extent do operational failures interfere
with effective nursing work?
How do operational failures shape the nursing
work environment?
5
Methods: 3 Sources of Data

Direct Observation




Interviews



11 nurses at 6 hospitals, each for a complete shift
Mean observation time: 9 hours 51 minutes
Recorded minute-by-minute information about their work
activities
Structured interviews with 6 of the observed nurses
Nurse’s perceptions of how operational failures affect
productivity and patient care
Survey Data (520 nurses in 48 units at 21 hospitals)

# of times experienced operational failures during last shift


I started to prepare a patient’s medication, but it was missing or incorrect
Also surveyed Managers from those units: # of operational
failures (i.e. medication, orders, equipment, supplies) nurses
encountered per shift
6
Findings: Frequency of Operational Failures
Average number of operational failures experienced by nurse by shift
Nurse Survey
Manager Survey
Observer
Mean (st dev)
Mean (st dev)
Mean (st dev)
1) Medication
1.2 (0.97)
1.5 (2.3)
1.0 (1.1)
2) Supply items
(including food)
1.2 (1.01)
1.4 (1.7)
0.9 (1.1)
3) Medical Orders
.54 (.55)
1.2 (1.1)
1.2 (1.1)
4) Equipment
0.98 (.87)
0.80 (1.0)
0.80 (1.1)
5) Insufficient staffing
0.59 (0.50)
0.80 (1.1)
0.60 (1.1)
Total in an average 8hour shift
4.51
5.7
4.5
Category of failure
7
Findings: Nature of Nursing Work

Workload pressure


Staccato Pace of Work





Administering medications within certain time periods
Monitoring vital signs during and after blood transfusions
Preparing patients for procedures (i.e. surgery)
Wide breadth of responsibilities



Average task time: 3.1 minutes
Nurses switched among patients every 11 minutes
Time-specific procedures


Worked 45 minutes of unscheduled overtime at end of shift to “catch up”
84 different types of activities
Cognitive components, such as administering medications that were
contingent upon the patient’s laboratory test results or vital signs
Frequent Interruptions

Interrupted mid-task 8 times per 8-hr shift
8
Lois* experienced frequent interruptions
3 sets of interrupted tasks: Preparing TPN, IV, Discharge patient
Even her interruptions were interrupted!
Minute (4:14 P.M. start)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
Prepare TPN
Interruption #1 - Ms. Park IV
Interruption #2 - Mr. Almay to bed
Interruption #3 - Mr. Holmes leaving
Interruption #4 - paged & Talk to Mrs. Almay
Interruption #5 - Find double pump for TPN
Interruption #6 - Lifting help
Interruption #7 - Mr. Muccini's sister here
Interruption #8 - No soap in Ms. Park's dispenser
8 interruptions in total to “Target Task”
* All names disguised to protect confidentiality
Findings: Three Tactics for Managing Workload
Planning Work
1. Partition


Spread care for each patient over shift (rather than one block)
Often due to medical necessity (assess vital signs every 2 hrs)
2. Interweave

Switch back and forth between different patients’ care
Managing disruptions to the plan
3. Reprioritization


Adapt work plans by adding, subtracting, and reordering tasks
as patients’ conditions changed
Newly admitted patients often caused reprioritization as nurses
had to fit the new patient into their work loads
10
Nurse Kendra Brown*’s Day
Mrs. P
Partition
Mrs. Q
Mr. Q
Mr. N
Interweave
Mr. B
14:49
14:35
14:21
14:07
13:53
13:39
13:25
13:11
12:57
12:43
12:29
12:15
12:01
11:47
11:33
11:19
11:05
10:51
10:37
10:23
10:09
9:55
9:41
9:27
9:13
8:59
8:45
8:31
8:17
8:03
7:49
7:35
7:21
Kendra's time spent on each patient
other
Reprioritize
While working on discharging Mr.
Q, Mr. B. complained of chest pain
11
Implications of the three tactics
Tactics may be associated with error because they introduce
 Delays


Interruptions


Delays in completing a task can cause a person to forget to return to
the task (Reason 1990)
Interruptions can cause a person to pick up a task at the wrong point,
repeating or omitting steps, or doing the right task, but on the wrong
object/person (Rudolph and Repenning, 2002; Flynn et al., 1999)
Distractions

Medical errors have occurred when healthcare professionals receive
wrong or incomplete information or materials (i.e. Chassin and
Becher, 2002)
12
Quote about recalling interrupted tasks

“I am in [a patient room] trying to do a medication, priming
the IV tubing. The phone rings, the clerk comes and gets
me. I stop what I am doing, dial the flow clamp off, hang it
over the IV pole, walk out and deal with the phone call.
Then someone else comes and asks me, ‘Can you come and
help with whatever.’ And I totally forget I have this IV
that I really haven't hung and haven't given to the patient
yet. Until I walk back in the room again-usually pretty
quickly-and see the IV hanging there and say ‘shoot I really
need to get that going.’” - Norma Garvin, Shock/Trauma H9
13
Partitioning and human error

Requires nurses to recall what has already been done and
what remains to be done




“Mental bookkeeping” (Cook and Woods 1994)
Interdependencies among tasks (i.e. patient awaiting surgery
needs lab tests done, but no food or water)
If stock of undone tasks increases faster than they are
completed, it causes stress, decreasing cognitive processing
(Rudolph and Repenning 2003) and conflicting priorities (i.e.
productivity versus quality)
Operational failures can increase stock of undone tasks
(i.e. administering medication)
14
Interweaving Care and human error



Switching back and forth requires “recovery
time” to bring details of patient’s case to the
forefront (Speier et al, 1999), especially with
complex work.
Can lead to accidentally mixing up patient details
Operational failures can lead to more switches
15
Reprioritization and human error

Additional cognitive load because




Attention process: determine whether interruption should be
addressed
Strategic process: determine which goals get priority (Cook
and Woods 1994)
Increases stress when have to abandon task that was
originally planned
Operational failures can cause reprioritization

25 minutes delay in getting Mr. Bartlett his medication because
Dr. did not realize Ms. Rollins amputation was that day and he
needed to come and sign consent form in the hour before
surgery
16
Findings: operational failures’ impact
Operational
Failure caused
18 (35%)
Patient care
caused
34 (65%)
Total
2. Interweaving
3 reprioritization
15 (2%)
8 (9.4%)
714 (98%)
77 (90.6%)
729
85
4. Interruptions
Total
4 (4.5%)
45 (4.7%)
85 (95.5%)
910 (95%)
89
955
Type
1. Partition
52
17
DISCUSSION


Conditions inherent to meeting patient needs make 95%
of the interweaving and reprioritization unavoidable
Design nursing processes to minimize negative impact
of interruption



Mistake-proofing and human factors engineering: Design
physical space to make it more difficult to commit errors, even if
interrupted (Grout 2003)
Visual Signals to Reduce Interruptions (e.g. a hat or apron with
the words, “Please don’t interrupt- preparing medications”) that
alerts other nurses and patients’ families that the nurse should not
be interrupted (IOM 2004)
Filter messages through a secretary or by provide nurses with
enough information to triage their messages (IOM 2004)
18
DISCUSSION


5% stem from operational failures and are
avoidable
Improve work systems by removing known
problems

Use failure occurrence to trigger removal of underlying
causes, rather than the common approach of relying on
people to work around failures (Spear, and Schmidhofer
2005; Tucker, Edmondson, and Spear 2002)
19
Conclusion: Two Avenues for Improvement

Operational Failures



What: Problems or errors in
supply of materials,
information, and equipment
to health care professionals.
Avoidable
Response: Work around
failure to provide patient
care
Strategy: Improve work
systems to reduce future
occurrences

Nature of health care



What: New information
about a patient’s health
status becomes evident
Unavoidable
Response: Need to update
the patient’s plan of care to
reflect new knowledge
Strategy: Design health care
work to be robust to
interruption
20