Does prompting reduce non- attendance rate? Ranga Rattehalli – ST6 Mahesh Jayaram - Consultant Ihsan Kader - Consultant
Jun 24, 2015
Does prompting reduce non-attendance rate?
Ranga Rattehalli – ST6Mahesh Jayaram - ConsultantIhsan Kader - Consultant
My role in the project
Inspired by the Leeds PROMPTS trial
I took the lead in formulating and developing the idea, data collection, data analysis, writing up and publication
Research was conducted on my special interest day during my ST4 post
Background
Out-patient clinics- an important contactFailed attendance adds to the cost of careAdverse impact – missing medicationsDelay in identifying relapsesIncreases waiting time for othersIt costed NHS 360 million pounds in 1997? > 0.5 billion per year now
Causes of non-attendance?
Conscious decision (benefits vs risk)
Often have a treatable morbidity
Commonest – forgetting the appointment
Forgetting – medication non-adherence
Unrelated to severity (other specialities)
Severe mental illness – more likely
Implications of non-attendance
Delay in identifying EWS of relapse Disengagement from services Trust performance indicator- CQC reportBanded on a scale of 1 to 5 (high=better)Rate < 11% considered acceptableFinancial penalties for high DNA rates?
Evidence for prompting
Leeds PROMPTS trial (N 764, RR 0.76, CI 0.59 to 0.98,NNT 16, 95% CI 10 to 187)
- Psychological Medicine 2008
Systematic review (serious mental illness)5RCTs, N 1184, RR 0.72, CI 0.59 to
0.89, NNT 6, CI 4 to 14 - Cochrane 2001 + PROMPTS Trial
Aim
To implement this evidence based intervention of sending prompt letters in our out-patient clinics in the pragmatic real world non-randomised setting & measure the non-attendance rate.
(Prompting works in trial settings but does it work in the real world?)
Methods
Prompting letter to all patients attending outpatient clinic in South Leeds (2 CMHT)
Letter sent one week before appointment
Royal mail First Class Post
Ongoing practice since June 2007
DNA- failed to attend & no message
Dear [Patient’s name]
Re: Your appointment at Bridge House
This is a short reminder of your appointment at Bridge House on the [Date at Time]. Your appointment will be with Dr. XX and will last for xx minutes. This interview will be private and confidential. It is often helpful if you bring a friend or family member and medications along. Our clinic has a reception and once the receptionist knows you have arrived, she will inform the doctor.
Bridge House is located on Balm Road and a map with directions is enclosed with this letter.
If you have forgotten about the appointment or made other plans, do not worry. Please let me know at the above telephone number and we will rearrange your appointment at a time which is convenient for you.
[Name of Secretary]
Data analysis
Compared DNA rates before and after this intervention in 2007
Compared with DNA rates in the same months in 2006 (seasonal variation)
Trend across the whole year
Compared with DNA rates in West Leeds where no prompting was used
Results
June – Nov 07 June-Nov 06 Jan-May 07
17% 27% 26%
A total of 1433 letters were sent out in the study period.
Average non-attendance rates across different periods
0
5
10
15
20
25
30
35
40Ju
n-0
6
Jul-
06
Au
g-06
Sep
-06
Oct
-06
Nov
-06
Dec
-06
Jan
-07
Feb
-07
Mar
-07
Ap
r-07
May
-07
Jun
-07
Jul-
07
Au
g-07
Sep
-07
Oct
-07
Nov
-07
DNA Rate
Prompting letters introduced
0.5 0.7 1 1.5 2 Favours Prompts Favours Controls
Study Prompt Group Control Group RR Weight RR (95% CI) n/N n/N %
243/1433 280/1074
Total N=2507Total events: 243 (prompts group), 280 (control group) Test for heterogeneity: not applicableTest for overall effect Z=5.53 (P<0.00001)
100.0 0.65 (0.56, 0.76)
NNT 11, 95% CI 8 to 17
A Forest plot showing the effect size
Comparison with West Leeds
June-Nov 2006 June-Nov 2007 P value
22% 23% 0.45 (NS)
So a reduction in the DNA rate in our study is unlikely to bedue to a general reduction across the city!
Average non-attendance rates
Were samples comparable?
1433 appointments575 patients (2007)
90% patients sameas in 2006
10% patients new to 2007
So a reduction in the DNA rate in our study is unlikely to bedue to a having a different set of patients in 2007!
Were the clinicians different?
Clinicians in 2007 may have acted differently toward patients compared to 06
Yes!
- one Consultant changed in Mar 2007
- SHOs, SpRs & Staff Grade changed
Drawback of pragmatic studies in NHS!
Patients getting better over time?
New for 2007
Remained constant
RR 95% CI
15% 17% 1.17 0.79 to 1.74
Average non-attendance rates
So a reduction in the DNA rate in our study is unlikely to bedue to an improvement in their mental state!
Conclusion
Prompting letters do reduce non-attendance rates even under non-trial conditions
Useful and easy to apply intervention with no implication on resources
£80 saved for every £3 spent (NNT 11)!!
(reference cost of each appointment - £80)