Control Charts Robin Henderson Royal Infirmary of Edinburgh Margrethe van Dijke Western General Hospital, Edinburgh National Stroke Audit Coordinators
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Control Charts
Robin Henderson Royal Infirmary of EdinburghMargrethe van Dijke Western General Hospital, Edinburgh
National Stroke Audit Coordinators
Charts have been used in healthcare for a very long time! Carl Wunderlich 1861
Control Charts
Florence Nightingale – the “passionate statistician” - is believed to have invented a type of chart
A patient’s temperature chart from 1963 – a run chart.
The horizontal line at 98.2 °F indicates “normal” temperature
The red horizontal lines indicate the normal range 97.5 to 98.8 °F
The run chart has now been converted to a control chart – Walter Shewhart 1924
Henderson, D. A. et al. JAMA 1999;281:2127-2137.
Typical Temperature Chart of Patient With Smallpox Infection
The control chart helps us to distinguish between “common cause” variation and “special cause” variation
Month
Pro
port
ion w
ith r
eco
rded s
creen
Sep-06J ul-06May-06Mar-06J an-06Nov-05Sep-05J ul-05May-05Mar-05J an-05
100
90
80
70
60
50
40
30
Swallow Screening at RIE
A run chart of the monthly proportion of patients with recorded swallow screens at RIE
To turn this run chart into a control chart we have to calculate a centre line and limits. The calculations are quite easy to do but most users get a computer to perform the calculations and plot the chart. An Excel tool may be downloaded from: -
http://www.indicators.scot.nhs.uk/SPC/SPC.html
Month
Pro
port
ion w
ith r
eco
rded s
creen
100
90
80
70
60
50
40
30
_Average = 66
Upper Limit =100
Lower Limit = 30
Swallow screening at RIE
All points fall between the tram lines and there are no “unusual” patterns of variation so there is no evidence of any special cause variation
Month
Pro
port
ion w
ith r
eco
rded s
creen
Apr-07Jan-07Oct-06Jul-06Apr-06Jan-06Oct-05Jul-05Apr-05Jan-05
100
90
80
70
60
50
40
30
_Average = 66
Upper limit = 100
Lower limit = 30
2
Swallow screening at RIE
We have evidence of improvement! If the average had remained at 66 then nine in a row above 66 would be most unlikely
It appears that a campaign by SALT staff has been effective and increased the average monthly proportion of patients screened from 66% to 82%.
Month
Pro
port
ion w
ith r
eco
rded s
creen
Apr-07Jan-07Oct-06Jul-06Apr-06Jan-06Oct-05Jul-05Apr-05Jan-05
100
90
80
70
60
50
40
30
_Average = 82
UB=100
Lower limit = 51
1 2
Swallow screening at RIE
Case Study – NV Clinic at WGH
Days from referral to examination for all patients
0
2
4
6
8
10
12
14
16
18
20
Months
Da
ys WGH Mean
WGH Median
Days from receipt of referral to examination <= 14 days for DEFCVD patients and <= 7 days,
80% Meets NHS QIS Standard for both
0102030405060708090
100
Months
Pe
rce
nta
ge
<= 7 days
<= 14 days
Target
18
16
14
12
10
8
6
4
2
0
Date
Days
_X=3.55
UCL=6.10
LCL=1.00
1 21 21 2
Mean days from referral to examination for DEFCVD patients
X=11.31_
The mean waiting time from referrals being received to patients being seen in the NV Clinic has decreased from 11.3 to 3.5.
100
80
60
40
20
0
Date
Perc
enta
ge
_X=91.5UB=100
LCL=72.9
1 21 21 21 21 21 2
I Chart of Percentage of NV Clinic patients seen within 7 days
X=28.8_
The mean percentage of DEFCVD patients seen in the NV Clinic within seven days since referral has being received has increased from 28.8 to 91.5.
How do you know that your stroke care is improving?
Control charts are a very useful tool for addressing this question.
“Display is an obligation!”
John Tukey 1986
Have a go!
Related Documents
