15th October 2010, Volume 123 Number 1324

Mary E Seddon, David Hay

The quality of hospital care provided after hours has been a concern for a number of years.1–3Factors identified include: lower staffing levels, key services not offered after hours or offered only by a skeleton staff on call (e.g. clinical pharmacists), fatigue of shift workers, overall clinical workload, and fragmented communication between staff. Various solutions have been put forward, including the UK Hospital at Night approach. In this paper we report on a tool designed to improve after-hours communication in a large (800 bed) metropolitan hospital.
Communication between staff has been identified as a problem generally4,5 but is particularly acute after hours. There are two specific problems: the quality of the communication, particularly between ward nurses and the junior doctors, and the distraction caused by our main communication tool—the pager (or beeper). Studies of after-hours communication have shown that pagers ringing frequently6 (for an individual doctor this may be as often as once every 7 minutes),7 and often interrupt clinical care,8 with one study showing up to 65% of the time pagers interrupted direct patient contact.9
The concern is that pagers may inhibit junior doctors from performing effectively and safely.10They also contribute to after hours stress on junior medical staff (related to the perceived lack of control one has over one’s workload).11 With the frequent pagers, junior doctors can feel overwhelmed, they cannot see a view of their overall workload, are unable to prioritise their tasks, and are often unable to complete their work.
"The beeper interrupts; pagers come more quickly than they can be answered."12
Furthermore there is evidence that pagers are not a good form of communication, with major differences between junior doctors and nurses in their perception of the appropriateness of calls and their urgency.13,8,9 Earlier work at this institution had identified that the paging process was not standardised and there were significant differences between doctors and nurses when it came to their use. Numeric paging, where a nurse adds the prefix 93 to the pager number and then hangs up and waits for the junior doctors to call back, was of little value to the junior doctor, as they did not know who had paged them, or what the problem was. Often on calling back, they were unable to directly communicate with whoever had called them as the nurse had moved away from the phone. There was a widespread belief amongst the doctors that numeric paging could be responded to as a lower priority than text pages.
Nurses, on the other hand assumed that doctors would see numeric pages as more urgent than text pages. Text paging was seen by the nurses as a lower priority message. Examination of the actual text paging data showed that it often consisted of updating patient investigation status (e.g. “patient is back from X-ray”), as opposed to worsening patient status. So there was a mismatch between nurses’ and doctors’ expectations of the paging system and considerable time wasted responding to pages without any effective communication taking place. Doctors reported getting so many pages during a shift that they were constantly being interrupted and distracted causing ‘page rage.’
We therefore introduced Task Manager, designed to capture after-hours ward tasks - from 1600 hours to 0800 hours on weekdays and 0800 to 0800 hours for the entire weekend. The aim was to improve communication after hours, by prioritising each task based on urgency, by making it visible to nurses that the junior doctors had seen the message and would be attending to the task, and to give staff an overview of how busy each shift was. It was also hoped that Task Manager would significantly reduce interruptions from pagers and decrease ‘page rage.’

Method

The idea for Task Manager came from a demonstration of a system at Hutt Hospital (Hutt Valley District Health Board). They had already successfully introduced SPADE (Simply Prioritize and Distribute Electronically), and they were willing to share their results. SPADE was developed using Java Server Pages, and it was not designed to integrate with the Hutt Valley Patient Management System. It was unclear whether it could be used for a much larger hospital, who would support the application (which was an unfamiliar application to CMDHB) and how much it might cost if Hutt Valley sold it to a vendor.
We decided to design our own system, extend it, and to link it with our electronic Patient Management System (PMS). Task Manager is accessed via a log on to our PMS. This provides security and is also the place where staff will be checking electronic laboratory and radiology results so Task Manager fitted with the usual workflow.
Development time for the initial version was six developer weeks at a cost of approximately $30,000. This was only possible as the Hutt Hospital system was used as the model for development, so the usual requirements and design phases were not required.
Task Manager displays after hours tasks, colour codes them for urgency, and filters tasks by patient location or speciality (see Figure 1). It also displays the Early Warning Score—known in our institution as the Physiologically Unstable Patient (PUP) score—if appropriate. Once a task is accepted, the accepting person is displayed. When the task is completed, it disappears from the active screen and is captured in the ‘completed in 24 hours’ tab.
Figure 1. Task Manager as displayed in the Patient Management System
sed1
It was decided that Task Manager would use 'drop-down' boxes to streamline the process for creating and accepting tasks. The tasks came from discussion with the services, and the responders to tasks were also identified. As Middlemore is a large hospital, there were a large number of responders including junior doctors (6 surgical house officers, 7 medical house officers and 9 registrars), nurses, Patient At Risk (PAR) teams and phlebotomists.
A test version was established for training purposes and a small team undertook training of the staff members. Most staff used the computer frequently and required very little training.
Task Manager was introduced into the general and sub-speciality Medical wards (9 wards) in late December 2009 and extended to the Surgical wards (7 wards) in February 2010. For the first week in each area, there was support for the users of Task Manager with a team of roving 'experts' on hand. A handout was produced taking people through the various steps:
  • how to create a task
  • how to accept a task
  • how to complete a task
Posters of these steps were put near each of the ward computer areas.
Task Manager Newsletters were circulated with updates and fixes to problems.
A convenience sample of junior doctors and nursing staff were surveyed using Survey Monkey™.
The impact of Task Manager on the number of pages was assessed by accessing the pager records for each after-hours page in November 2009 and again in March 2010.

Results

In the first 6 months since Task Manager’s introduction, there have been over 21,000 tasks completed. On average there are just over 100 tasks completed each evening/night shift (1600 to 0800 hours) and 190 completed on the weekend days (0800 to 0800 hours). The most common tasks were blood tests and cannulations, which made up 47.8% of all tasks (see Table 1).
The vast majority of tasks were entered as routine (59.9%), 33.1% were coded semi-urgent and 6.96% were coded as urgent The average time from creation of a task to completion was 65 minutes, 94 minutes and 111 minutes for urgent, semi-urgent and routine tasks respectively.
A third of all tasks were completed by phlebotomists (on duty from 1500 hours to 2300 hours), approximately one-third by medical house surgeons and the final third by surgical house surgeons. Very few nurses completed tasks (1.1%) and there were no tasks completed by registrars (see Table 2)
Table 1. Type of tasks completed and frequency of task
Task
n
%
Electronic Discharge Summary Required
Family request meeting
New Admission /Transferred Patient
Patient Deceased
Urinary catheter needed
Low haemoglobin
Consent Patient
Admit Patient
Patient At Risk Nurse Review
Chart Anti-emetic
Patient for Discharge
Arterial Blood Gas (ABG)
Review medications
Fluid Review
Multiple tasks—see comment
Chart/re-chart pain relief
Review Labs
Review Patient raised PUP score
Chart Warfarin
Review Radiology
Blood Test / Cannula
Review ECG
Chart/re-chart IV fluids
Other Task—see comment
Chart/re-chart meds
Review patient - see comment
Blood Test
Cannula
2
12
13
19
33
35
40
50
92
112
133
232
225
233
280
352
400
431
443
469
597
817
1313
1530
1697
2349
3489
6015
0.01%
0.06%
0.06%
0.09%
0.15%
0.16%
0.19%
0.23%
0.43%
0.52%
0.62%
1.08%
1.05%
1.09%
1.31%
1.64%
1.87%
2.01%
2.07%
2.19%
2.79%
3.82%
6.13%
7.15%
7.93%
10.97%
16.29%
28.09%
Total
21413
100.00%
Table 2. Health Professionals completing tasks
Task completed by
n
%
Phlebotomist
Medical On Call House Surgeon (OCHS)
Medical B-call House Surgeon (HS)
Med S-call HS
Med C-call HS
General Surgery On Call House Surgeon (OCHS)
Orthopaedics OCHS
Orthopaedic Admitting HS
Plastics OCHS
Plastic Hands OCHS
Nurses
Patient At Risk Nurses
5430
4242
1258
941
148
3918
1245
54
448
100
201
105
30.0%
23.4%
7.0%
5.2%
0.8%
21.7%
6.9%
0.3%
2.5%
0.6%
1.1%
0.6%
The busiest time for tasks to be created coincided with the afternoon change of shift, peaking at 1600 hours and then remaining busy until 2400 hours (see Figure 2).
Figure 2. Time of day that tasks were created
sed2
Since the introduction of Task Manager there has been a 30% reduction in pager calls to junior doctors (see table 3) and a 40–50% reduction in pager calls to the phlebotomists.
Table 3. Pager numbers by role pre and post introduction of Task Manager
Role
Pre Task Manager
(1/11/09–30/11/09)
Post task Manager
(1/3/10–30/3/10)
% change
Medical On Call House Surgeon (OCHS)
Medical B call
General Surgery OCHS
Orthopaedics OCHS
Plastics OCHS
Medical Phlebotomist
Surgical Phlebotomist
719
141
257
142
405
120
86
499
102
175
100
267
60
49
- 31%
- 28%
- 32%
- 30%
- 31%
- 50%
- 43%
Forty-nine junior doctors (out of a possible 174) responded to the survey, but it was clear that Task Manager was predominantly used by house officers (96% used it regularly compared to 9% of registrars). On the basis of this, the responses were from 27 house officers only. Most (83%) found it easy to use and 71% found that it was easier to manage their workload with Task Manager (see Table 4).
“Able to prioritise some clerking, also can get on and do tasks without being interrupted for simple requests that are non-urgent”
“Not constantly being interrupted. Task Manager - nurses are required to state what the job/concern is and then you can prioritise; paging - they often just leave an extension number to call back and then when you call the number, you often wait for them to find the person who paged and found what the job is - lot of time wasted.”
All felt that it had led to a reduction in pages, with most house officers rating the reduction as more than 50% and a quarter of respondents thought that pages had decreased by 80%. The overall satisfaction rating was good with 29.6% satisfied compared to 7.4% unsatisfied.
Table 4. Survey responses from house surgeons
Question
Yes
No
Do you find TM easy to use?
Has it made it easier to manage after hours workload?
Has TM increased your after hours workload?
Has TM reduced how often you are paged after hours
83.3%
71%
57%
100%
17%
29%
43%
0%
If it has decreased pager calls, please estimate by how much
Estimated reduction in pager calls
10%
20%
50%
80%
100%
0%
17%
58%
25%
0%
Overall satisfaction with TM
1
Very unsatisfied
2
Unsatisfied
3
4
Satisfied
5
Very Satisfied
3.7%
7.4%
59.3%
29.6
0.0%
Years since graduation
0
1
2
3
>3
33.3%
7.4%
44.4%
11.1%
3.7%
213 nurses responded to the survey and of these 89% found it easy to use, 68% thought that it had improved communication with junior doctors and 70% felt that it resulted in tasks being completed in a more timely fashion.
We are now sure that our tasks will be acknowledged and actioned
Don't have to sit near the phone and wait for the doctors or phlebotomists to respond

Discussion

Task Manager was borne out of our desire to improve after-hours care and communication, decrease interruption by pagers, and allow junior doctors to prioritise their workload. It has succeeded in decreasing pager interruption, and has improved the display and prioritisation of after-hours tasks. It has high user acceptability and it has proven to be a very inexpensive, innovative solution. Furthermore, we were surprised by how little formal training was required as Task Manager uses intuitive drop down boxes and simple instructions.
‘After-hours’ actually constitute 75% of the hours in a week, and the care provided in this time has been a concern recently. The overall shortage of junior doctors, and the requirements to conform to working time directives, means that we need to be more innovative in how we organise after-hours care.
Task Manager is one of the tools which can help prioritise and rationalise after-hours work. Other initiatives, such as Hospital At Night, advocate a central coordinator role (usually a senior nurse) to prioritise work and a physical ‘control centre’ that allows a single point of call for all after-hours calls.1 We have not gone down this route as it was estimated that we would require 11 FTEs to cover this role and Task Manager allows all staff to see where the tasks are in real time.
Another initiative13 aimed at decreasing interruptions from pagers, was to have whiteboards on each ward for nurses to write up tasks. Again Task Manager effectively does this, but has the added advantage of allowing a hospital-wide or service specific overview of all the tasks from any computer.
We had investigated a number of options to improve after hours communication. The first used mobile phones to send messages to junior doctors after hours. This was rejected as it was very costly, there were concerns about delays with the phone network, and feedback from junior doctors who had worked with this system was overwhelmingly negative.
We also looked at a product that used text pagers that automatically escalated tasks if there was no response. Again this was rejected on the basis of cost, that it did not necessarily deliver better communication, and it did not provide a hospital-wide view of the pending tasks.
Several studies have tried to estimate the workload after hours, actually shadowing junior doctors as they work. Christchurch Hospital identified that there was wide variability in the activity levels of junior doctors after hours and that most tasks were generic in nature.6 This is supported by our data. Task Manager has the in-built ability to rapidly generate reports on activity and this has demonstrated for the first time the number and type of tasks undertaken after hours.
The majority (nearly 50%) of after-hours tasks are phlebotomy-related (blood tests and cannulations). We intend to review this in more depth to see whether all the phlebotomy tasks are indicated - especially after hours cannulation.
It is likely that Task Manager does not capture total workload as junior doctors on the wards will be asked to do tasks verbally and these will not be captured in Task Manager. So the workload in Task Manager should be seen as a minimum indication of the total workload. It has however, highlighted some inequalities in workload (such as high rate of phlebotomy tasks), which may be used to better organise the after hours workforce. As the tasks after hours are generic in nature (i.e. not requiring sub-speciality expertise) there may be an opportunity to review the traditional roster system for junior doctors.
One of the central tenets of the Hospital At Night work is that routine work is not carried over into the out-of-hours period when there are fewer staff to carry out such work. The Task Manager reports show that such tasks are spilling over into the after-hours. For instance, Warfarin is traditionally charted during the day once the blood monitoring is reviewed, but is given at 1800 hours.
In at least 443 cases, the daytime workers had gone home without charting Warfarin and this was picked up by the after-hours staff. It could be argued that at least some of the IV fluid tasks could have been anticipated by the day staff. Interestingly of note, there are some tasks which are rarely requested such as an ‘Electronic Discharge Summary’ (0.01% of tasks), and ‘Consent Required’ (0.19%). We will be revisiting the list and may delete some of the infrequently requested tasks.
As far back as 1990, it was suggested that "if use of bleepers was restricted to emergencies with the institution of a non-urgent messaging system, the number of interruptions to work and rest, and thus workload, could be reduced considerably".14 We believe that Task Manager is such a system and since its introduction there has been a significant reduction in pagers and ‘page-rage.’ Pager calls for junior doctors have decreased by 30% but the perception is that the reduction is far more (closer to 50%) with junior doctors noticing significantly fewer interruptions.
The reduction for phlebotomy staff is even greater with 40–50% reduction in actual pager calls. They had previously had a fairly rigid work process, starting with a 'ward round' doing four phlebotomy tasks on each ward. With the introduction of Task Manager they were able to see where the most urgent tasks were and to prioritise their work accordingly. At the same time nurses can be reassured when they see that the phlebotomy task has been accepted and are therefore not paging the phlebotomists to check up.
Previous studies have assessed the appropriateness of pager calls to junior doctors. Inappropriate calls have ranged from 17% to 26%.8,9 Task Manager has largely replaced pagers and ward lists for routine tasks but there has been some education of nurses needed for use of 'urgent' priority in Task Manager.
To avoid a really urgent task (e.g. asked to assess a rapidly deteriorating patient) sitting on Task Manager unseen, a fix advocated by junior doctors was instituted. If a nurse assigns an urgent priority to a task, before they can finalise the task, a pop-up box comes up to remind them to page the intended recipient for an urgent task. Urgent tasks make a up a small proportion of tasks (just under 7%), but as the risk is there, we will soon be looking at a forcing function - whereby an automatic page is sent to the on- call house surgeon if an urgent task is entered, and a similar system will work if the nurse requests a Patient At Risk (PAR) team review.
Limitations—There are several limitations to our study. We had attempted to get accurate data on junior doctors’ tasks prior to the introduction of Task Manager, but despite hiring people to shadow junior doctors, this provided only very basic data, and was difficult to collect. We have used the number of pagers per junior doctor as a proxy for workload pre and post Task Manager. However, it is not a perfect proxy as we were unable to capture on-the-ward referrals and the use of junior doctors’ mobile phones. Our survey numbers are also very small with only 49 junior doctors and 213 nurses responding, however, we have backed this up with monthly forums for feedback and ongoing discussions about improvements.
Future directions—Task Manager has shown its utility in the adult medical/surgical wards and there is now interest from other disciplines (Obstetrics and Paediatrics). However, what has become apparent is that there are multiple 'tasks' that are ordered in a multitude of ways in our hospital and many could be served by Task Manager. We are working on a non-clinical Task Manager for our orderly services. Ward nurses will be able to book an orderly task on Task Manager, the orderly service can then prioritise jobs, direct their staff and keep an overview of activity. Similarly, booking transit nurses could be done in the same way.
Conclusion—Task Manager is a simple yet powerful tool for prioritising routine tasks after hours. It allows staff to quickly create tasks, and communicate effectively with other members of the team. It has reduced the number of times junior doctors are paged and they can continue with their work with fewer interruptions. It also provides accurate information on the type of tasks undertaken, the busy shifts and the approximate workload of staff. Whilst it was introduced to improve effective communication after hours it has become apparent that it will also work for other services that respond to tasks 24/7, such as the orderly and transit nurse services.

Summary

In the evenings and overnight the hospital is staffed with fewer doctors and nurses than it is during the day. It is important that these healthcare professionals are able to communicate quickly and clearly so that patients receive the care that they need. At Middlemore Hospital we have introduced a tool (Task Manager), that staff are able to access on any computer. This tool shows all the tasks that need to be done, it colour codes them for urgency and it displays who has accepted the task. This has dramatically reduced the number of pagers that each doctor receives (allowing them to do their work without distraction) and has shown for the first time, what tasks are actually being done, by whom, and when. This should inform how we staff hospitals at night.

Abstract

Aim

To improve communication between doctors and nurses after hours, by developing a tool to display ward tasks, allowing staff to prioritise their work, without constant interruption from pagers (beepers). Middlemore Hospital, a large metropolitan 800-bed hospital in Auckland, New Zealand.

Method

Introduction of computerised system (Task Manager) to identify, allocate and complete after-hours tasks.

Results

In the first 6 months 21,000 tasks have been completed in Task Manager. Paging of junior doctors has decreased by over 30% and there is broad acceptance of the tool by both nursing and medical staff. Task Manager has collected real-time data on the type of after hours tasks (nearly 50% are phlebotomy-related tasks), busy times of the day (1600 hours to 2400 hours) and who is performing most of the tasks.

Conclusion

Task Manager is a simple yet powerful tool for prioritising routine tasks after hours. It allows staff to quickly create tasks, and communicate effectively with other members of the team. It has reduced the frequency of junior doctors paging so that they can continue their work with fewer interruptions. Whilst it was introduced to improve effective communication after hours, it has become apparent that there are multiple 'tasks' that are ordered in a multitude of ways in our hospital and many could be served by Task Manager.

Author Information

Mary E Seddon, Clinical Director Quality Improvement Unit, Counties-Manukau DHB, Auckland; David Hay, Enterprise Architect (Clinical Systems), HealthAlliance, Auckland

Acknowledgements

Hutt Valley DHB for original idea and for sharing their experiences.

Correspondence

Dr M Seddon, Clinical Director Quality Improvement Unit, Counties-Manukau DHB, Private Bag 93311, Auckland, New Zealand. Fax: +64 (0)9 2593865

Correspondence Email

MZSeddon@middlemore.co.nz

Competing Interests

None.

References

  1. Beckett DJ, Gordon CF, Paterson R, et al. Improvement in out-of hours outcomes following the implementation of Hospital at Night. Q J Med. 2009;102:539–46.
  2. Bendavid E, Kagnova Y, Needleman J, et al. Complication rates on weekends and weekdays in US hospitals. Am J Med. 2007;120:422–8.
  3. Hendey GW, Barth BE, Soliz T. Overnight and post call errors in medication orders. Acad Emerg Med. 2005;12:629–34.
  4. Gawande AA, et al. Analysis of errors reported by surgeons at three teaching hospitals. Surgery. 2003;133(Jun):614–21.
  5. Dunn EJ, Mills PD, Neily J, et al. Medical team training: applying crew resource management in the Veterans Health Administration. The Joint Commission Journal on Quality and Patient Safety. 2007;33(6):317–25.
  6. Morton J, Williams Y, Philpott M. New Zealand's Christchurch Hospital at night: an audit of medical activity from 2230 to 0800 hours. N Z Med J. 2006;119(1231). http://www.nzmj.com/journal/119-1231/1916/content.pdf
  7. Chiu T, Old A, Naden G, Child S. Frequency of calls to "on-call" house officer pagers at Auckland City Hospital, New Zealand. N Z Med J. 2006;119(1231). http://www.nzmj.com/journal/119-1231/1913/content.pdf
  8. Patel R, Reilly K, Old A, Naden G, Child S. Appropriate use of pagers in a New Zealand tertiary hospital. N Z Med J. 2006;119(1231). http://www.nzmj.com/journal/119-1231/1912/content.pdf
  9. Katz MH, Schroeder SA. The Sounds of the Hospital. Paging Patterns in Three Teaching Hospitals. New Engl J Med. 1988;319(24):1585–9.
  10. Harvey R, Jarrett PG, Peltekian KM. Patterns of paging medical interns during the night calls at two teaching hospitals. Can Med Assoc J. 1994;151(3):307–11.
  11. Kuper H, Marmott M. Job strain, job demands, decision latitude, and risk of coronary heart disease within the Whitehall II study. Journal of Epidemiology & Community Health. 2004;57(2):147–53.
  12. McCall TB. The impact of long working hours on resident physicians. N Engl J Med. 1988;318:775–8.
  13. Lear JT, Kelly MJ, Worley C. Disruption caused by the house offices bleep: a simple solution. Annals of the Royal College of Surgeons of England. 1993;75(Suppl):9–41.
  14. Turnbull NB, Miles NA, Gallen IW. Junior doctor's on call activities: differences in workload and work patterns among grades. BMJ. 1990;301(Nov 24):1191–2.