Implementing an automated pharmacy system

1
Implementing an automated pharmacy system

• Cheryl Way
• Principal Pharmacist
• Llandough Hospital

2
Aims of Presentation

• To explain the background to the Automated
  Pharmacy System Project
• To describe the implementation of automation at
  Llandough Hospital
• To outline future plans for the project

3
The Automated Pharmacy System (APS) Project

• Business case submitted to Wales Assembly
  Government (WAG) in 2002 by Welsh Chief
  Pharmacists (WCP) Committee
• Three phase project to install APS in Welsh
  Hospital Pharmacies
• Funding for phase one announced Jan 03
• 500,000 p.a. for five years

4
The APS Project

• PRINCE project board appointed
• Three pilot sites chosen
• West Wales General Hospital
• Llandough Hospital
• Ysbyty Glan Clwyd
• Contract Awarded in April 03
• All installed by Dec 03

5
Tendering process

• OJEC advert placed by WHS Jan 03
• 3 suppliers tendered
• ARX
• Baxter
• Swisslog
• Visits to live sites April 03
• Suppliers visits to hospitals
• Adjudication of tenders 28/4/03

6
Award of tender to ARX

• All three sites
• Duplex machines for W.Wales, Llandough
• Triplex for Glan Clwyd
• W. Wales delivery July, live September
• Llandough delivery Sept., live November
• Glan Clwyd delivery Oct., live December

7
Llandough Project Group

• Set up to manage implementation at site
• First meeting Feb 03, then monthly
• Worked closely with other pilot sites
• Reported to project board
• Building work started May 03
• Went live November 03

8
Building and refurbishing work

• Department redesigned around robot
• Cleaners room relocated and wall removed
• Dispensary, stores and box assembly gutted
• Holes in walls and dormer for conveyors
• New dispensary ceiling
• New flooring, paint and fittings
• Emergency cupboard fittings moved
• Air-conditioning installed

9
Service Changes

• July 28th
• New ward pharmacy service started
• No returns accepted for processing
• September 8th
• Dispensary moved to dressing store
• FP10(HP)s for out-patients
• Additional portering rounds
• Saturday visits to all wards

10
Installation of robot

• Delivered 18/9/03 three hours to unload
• Took up half box assembly area and more
• Assembly started 22/9/03 for 2 weeks
• German engineers work long days
• Conveyors installed over next 2 weeks
• Fridges installed end October
• Went live 3/11/03

11
IT issues

• Bar-code for every robot item input on EDS
• Bi-directional interface between EDS and robot
• Dial in to pharmacy system and robot via lap-top
  from home for on-call

12
Future plans

• Automatic goods receiving on robot
• Controlled drugs in robot
• Stock reconciliation
• Over-labelling of bar-coded pre-packs

13
Phase 2 of APS project

• Three new sites
• Royal Glamorgan Hospital
• Princess of Wales Hospital
• Royal Gwent Hospital
• Additional funding of 1.2 million
• Tenders received
• Purchase, not lease
• Contract awarded to Westfalia

14
The Welsh automated pharmacy system
projectIntroduction

• Dr Ceri Phillips

15
Efficiency
'In a system with limited resources, health
professionals have a duty to establish not only
that they are doing good, but that they are doing
more good than anything else that could be done
with the same resources' (Williams, 1993).
16
Inefficiency 'costs NHS 6bn each year'?
THE NHS may be wasting up to 6 billion a year as
a black hole of rising inefficiency consumes as
much as 9 per cent of the extra cash being pumped
into the service, an official study showed
yesterday. The measure of productivity in the
health service, gauging how much extra it
delivers to patients for every pound spent, shows
that it has suffered a steady decline from 1995
to 2003. The analysis by the Office for National
Statistics suggests that the slide in efficiency
means that while the real resources pumped into
the NHS over this eight-year period have risen by
32 to 39 per cent, after allowing for inflation
in its pay and costs, its output of services to
patients has increased by just 28 per cent. The
gap between funding and results is accounted for
by tumbling productivity. The ONS estimates this
has dropped by between 3 and 8 per cent since
1995. An 8 per cent plunge in productivity means
that if the health service were still as
efficient as it was in 1995, it could achieve the
same results for patients this year with a budget
6 billion lower than the 69.4 billion it will
actually receive. If account is also taken of
the impact of rising costs and wage bills, then
the ONS estimates imply that for every extra 100
spent on the NHS, between 56 and 62 goes in
inflation, and between 3 and 9 on reduced
productivity, leaving only a 35 real rise in
output of services.
THE TIMES 19/10/2004
17
Adverse events

• Adverse drug events
• US study
• 341 potential and adverse drug events identified
  from 4031 adult admissions
• Small number of preventable errors which caused
  adverse drug events were due to dispensing (4)
• (Bates et al.,1995)

18
Adverse drug events

• Systematic analysis of events on all medical and
  surgical units at 2 US tertiary care hospitals
• Prescribing 39
• Transcription 12
• Dispensing 11
• Administration 38
• Approx 33 of dispensing errors intercepted
  primarily by nurses
• (Leape et al., 1995)

19
Adverse events in the UK

• Adverse events detected by retrospective record
  review in 2 acute hospitals
• 11 of the 1014 patients experienced an adverse
  event of which 50 were avoidable
• (Vincent et al., 2001)

20
Results from a systematic review

• Overall impact of adverse drug events 380
  million per year
• 4 of available bed-days and equivalent to 13.6
  400-bed hospitals
• (Wiffen et al., 2001)
• http//www.jr2.ox.ac.uk/bandolier/Extraforbando/AD
  RPM.pdf

21
Cost of medical negligence in NHS

• All negligence claims arising from treatment in
  Oxfordshire Health Authority (1974-98)
• Litigation volume doubled between 1990 and 1998
• In 1998 1 claim closed for each 1200 finished
  consultant episodes
• Overall expenditure 84 million, inc legal costs
• (Fenn et al., 2000)

22
Medication error

• NPSA
• is any preventable event that may cause or
  lead to inappropriate medication use or patient
  harm while the medication is in the control of
  health professional, patient or consumer

23
Medication use process

• Prescribing
• Transcription
• Dispensing
• Preparation
• Administration

24
Human Error

• Person approach
• System approach
• building defences to avoid error
• evaluating conditions
• fair blame culture

25
Safety?

• Arthur Dent If I asked you where the hell we
  were, would I regret it?
• Ford Prefect Were safe.
• Arthur Dent Oh good.
• Ford Prefect Were in a small galley cabin in
  one of the space ships of the Vogon
  Constructor Fleet.
• Arthur Dent Ah, this is obviously some strange
  usage of the word safe that I wasnt aware
  of.

Douglas Adams, The Hitch Hikers Guide to the
Galaxy, 1979
26
Dispensing process

• Prescription interpreted
• Label generated patient, drug dose, frequency,
  precautions
• Label attached to pack with correct drug strength
  and form
• Correctly labelled medicine given to correct
  patient with advice

27
UK Dispensing Error Analysis Scheme

• Overall error rate 18.1 per 100,000 items
  dispensing (Spencer Smith, 1993)
• Most common errors (23) wrong drug dispensed
  and wrong strength of right drug (Roberts et al.,
  2002)
• Latest data most common error (23) wrong
  strength of right drug (Roberts et al., 2004)

28
Background

• Audit Commission Report, 2001
• A spoonful of Sugar medicines management in
  NHS hospitals
• Van den Bemt et al, 2000
• Drug related problems in hospitalised
  patients

29
Background

• Bates, 2000
• Using information technology to reduce rates
  of medication errors in hospitals
• Slee et al., 2002
• Implementing automated dispensing system
  drop in error rate by 50 in first four months

30
Background

• DOH, 2004
• Building a safer NHS for patients improving
  medication safety
• Safer medication through improved information
  management and technology
• Greater use of ..robotic dispensingpotential
  to reduce further the risk of medication errors
  p11

31
Evidence

• To date, however, there are only limited
  published data to demonstrate improved accuracy
  over traditional manual dispensing systems
• DOH, 2004 p 127

32
The Welsh automated pharmacy system
projectResults

• Dr Cate Whittlesea

33
Aims of the study

• Multi-centre study to assess the costs and
  benefits of APS based at the three pilot
  hospitals, with a view to rolling out this
  evaluation to other hospitals in Wales
• Development of a tool kit to evaluate the APS
  and associated processes

34
Timescale

• Tendering process complete
• West Wales General
• Llandough Hospital
• Ysbyty Glan Clwyd
• Report to National Assembly April

April 2003
September 2003
November 2003
December 2003
March 2004
35
Preliminary work

• Negotiation of access to data via Chief
  Pharmacists Committee (Wales)
• Determination of what data evaluated the impact
  of the robot rather than the pharmacy computer
  system
• Evaluation of data currently being reported at
  sites and comparability of data
• Information from other sites with robot
  dispensing systems e.g. Dudley, Arrowe Park and
  Barts the London related to type of dispensing
  errors observed
• Information obtained from ARX

36
Preliminary work

• Site visits and discussion with staff related to
  hospital processes to be evaluated
• Determination of survey and data available from
  other sources
• Pharmacy bench marking 2001 NHS national
  performance advisory group July 2001
• Bench marking the current dispensing rate of
  Welsh Hospital Pharmacies 2002
• Survey and protocol development
• Sue Ashwell Kettering General Hospital

37
Existing data

• Pharmacy issue statistics particularly internal
  orders, stock adjustments, inter-location
  transfers
• Turn around times for prescriptions
• Dispensing errors
• Out of hours supply
• Stock taking
• Ward distribution stock ordering incidents

38
Existing methods

• Dispensing rate skill mix
• (Hiom et al., 2003)

39
New methods

• Outpatient satisfaction
• Ward staff satisfaction
• Attitudes of pharmacy support staff
• Dispensing incidents
• Work load data (ward box assembly)
• Whittlesea et al. (2004) Hospital Pharmacist,
  11(7) 283-285

40
Benefits of APS Turn around time - TTHs

• Reduction in the median time
• 96.8 to 61.1 minutes at West Wales General
• 66.5 to 60.3 minutes at Llandough

41
Benefits of APS Turn around time - Outpatients

• Reduction in the median time
• 8.1 to 6.6 minutes, West Wales General
• 20.0 to 16.4 minutes, Llandough

42
Dispensing rate in Wales

• Average rate achieved during the benchmarking
  study (2002)
• 9.9 items/person/hour (CI 95 0.9)
• 17 hospitals
• (Hiom et al., 2003)

43
Benefits of APS Dispensing rate

• Increase in dispensary rate at Llandough
• Pre automation 11.6 items/person/hour
• Post automation 15.7 items/person/hour

44
Benefits of APS Dispensary skill mix

• Change in dispensary skills mix of staff at WWG -
  rise in technician time decrease in pharmacist
  time

45
Benefits of APS Outpatient satisfaction

• Patients at Llandough post automation,
  considering their visit to better than the
  previous visit
• Respondents considered the staff
• Less busy
• Less stressed
• Less harassed
• Less interested
• Reduction in the amount of time patients waited

46
Benefits of APS Dispensing errors reduced

• Ysbyty Glan Clwyd
• April - June compared to Dec - February
• 2.9 per 10,000
• July Oct compared to Dec - February
• 2.6 per 10,000
• Llandough
• July Sept compared to Nov - February
• 2.0 per 10,000

47
Dispensing errors Ysbyty Glan Clwyd
48
Benefits of APS Dispensing incidents reduced

• West Wales General
• July - September - compared to October December
• 3.6 per 10,000

49
Dispensing incidents West Wales General
50
Benefits of APS are

• Decrease in staff time processing stock orders
  and requisitions at West Wales Llandough
• Increase in the amount of time spent on ward top
  up activities at Llandough Glan Clwyd due to
  expansion of these services

51
Benefits of APS are

• Decrease in staff time spent tidying the ward
  distribution and stock holding areas (West Wales
  Llandough)
• Decrease in inter location transfers at all
  hospitals
• Decrease in internal orders (West Wales Glan
  Clwyd)

52
Distribution incidents

• The incident rate overall per 1,000
• 12.9 pre APS and 7.2 post APS
• Difference between the incident rates pre and
  post APS of 5.7 per 1,000
• Whittlesea et al. 2004 International Journal of
  Pharmacy Practice, 12 (suppl) R70

53
Benefits of APS Ward staff views

• Increased ward staff satisfaction with some
  aspects of the Glan Clwyd pharmacy services
  (pharmacy opening times)
• Decrease in the time required to undertaken the
  annual Llandough stock audit
• Out of hours supply from Llandough dispensary
  successfully achieve from a remote location

54
Overall attitudes positive BUT
Benefits of APS Technician views

• I am worried that the robot will continuously
  break down (59)
• A robot will reduce the boredom associated with
  some of my job (20)

55
Technician views differences

• My job will be threatened by introducing a robot
• Working with a robot does / will make my job more
  difficult
• The use of the robot has / will reduce the number
  of dispensing errors
• A robot has/will allow me more time to check
  patients medication
• Operation of a robot has/will cause hassle for
  staff
• New job opportunities will occur when robots are
  introduced
• In press Pharmacy World and Science

56
Cost of automated dispensing system(supplied by
PRINCE Project manager)

• Capital cost of APS systems (ex VAT)
• West Wales General
• Basic equipment 190,000
• Software and additional costs 120,650
• Llandough
• Basic equipment 190,000
• Software and additional costs 120,650
• Glan Clwyd
• Basic equipment 256,500
• Software and additional costs 168,150
• Quarterly leasing bill
• West Wales General Llandough 16,321 per
  site
• Glan Clwyd 22,485
• VAT 9,647
• Projected maintenance per annum (all
  sites) 98,647
• TOTAL leasing maintenance (2004/5) 357,416

57
Pharmacy Stock Cost
Annual Stock valuation
58
Limitations

• VERY short timescale pre and post automation
• Extensive building work at 2 sites
• Questionnaires anonymous
• Power of studies
• Pharmacy staff knowledge of research
• Availability of staff for QA procedures
• Support for data analysis

59
Conclusion

• First multi-centre study of APS in UK
• Addresses in part the identified lack of evidence
  of APS versus traditional dispensing and
  distribution
• Improvements in the pharmacy dispensing and
  distribution observed post APS
• Further work required to quantify the benefits of
  APS on redeploying pharmacy staff to wards,
  discharge times and value for money

60
Acknowledgements

• Study
• Dave Roberts
• Robin Burfield
• Sarah Hiom
• Darrell Baker
• Cheryl Way staff at Llandough Hospital
• Jeremy Savage staff West Wales General Hospital
  Carmarthen
• Ann Slee staff at Ysbyty Glan Clwyd, Rhyl
• Gavin Rose - MPharm 4, Welsh School of Pharmacy
• Staff from Royal Glamorgan Hospital, Llantristant
  and University Hospital of Wales, Cardiff who
  participated in the pilot evaluations
• Staff at Wrexham Maelor Hospital, Prince Charles
  Hospital, Merthyr Tydfil and Ysbyty Gwynedd,
  Bangor who participated the support staff survey
• Presentation
• Dave Lewis