1 Who are smoking cessation services in Scotland reaching? A secondary analysis of routine data Jan Kerssens, Cheryl Denny, Linsey Galbraith (ISD) Sally Haw, Linda Bauld (Stirling University ) Rosemary Hiscock (University of Bath)
Jan 02, 2016
1
Who are smoking cessation services in Scotland reaching? A secondary
analysis of routine data
Jan Kerssens, Cheryl Denny, Linsey Galbraith (ISD)Sally Haw, Linda Bauld (Stirling University )
Rosemary Hiscock (University of Bath)
2
Outline• Scotland• Smoking Cessation services• Data set• Interventions• Client groups• Standardised Rates• Funnel plots• Conclusion• Questions
3
Scotland• Northern part of UK
mainland• Just over 5 million
inhabitants, concentrated in the central belt
• Edinburgh capital – Glasgow biggest city – 32 Local authorities (councils)
• Islands in the west and the north
• Health inequalities• 24% smokers (top 6 in
Europe)
4
Smoking prevalence
5
Smoking cessation services
• 14 geographical NHS Health Boards• 32 Local authorities (councils)• 40 Community health partnerships• 4 in Highlands; 3 in Fife; 5 in Glasgow• Rest coterminous with Local authorities• Pharmacy, General practice, Specialist setting• Nicotine Replacement Therapy (NRT),
Varenicline, Group support
The national database
Server at ISD
ISD analysis(no access to person-identifying data)
NHS board staff
NHSNet connection
Data from GG & C
Client information• Age• Sex• Pregnancy• Ethnic
background• Free
prescriptions• Employment
status• Postcode
Tobacco use and Interventions
• Quit date• Pharmaceutical
usage• Behavioural
support• Intervention
setting• Free prescriptions• Follow-up
Outcomes• Success/failure• 3-month follow-
up• 12-month follow-
up
Pharmacy
0
10
20
30
40
50
60
one month cessation three month cessation
Perc
enta
ge (%
)
Pharmacy services All Scotland
• Over 130,000 quit attempts made in 2010-2011
• Broadly, the more deprived the area a client is from, the more chance the client will using a pharmacy service
• Most popular in 18-34 yr age group and least popular with those aged 60+
• Unemployed most likely to use pharmacy service while sick or disabled least likely to
• Variation of Nicotine Replacement Therapy use across NHS boards from 60% to over 90%
NRT
Varenicline
• Variation of Varenicline use across NHS boards from less than 10% to 34% of all quit attempts made
LegendQuit attempt using Varenicline (%)
Other features of different
pharmacotherapy users• Those in paid employment, full-time parents and permanently
sick/disabled use Varenicline more than any other employment groups
• Students and the unemployed use Varenicline the least• Proportionally, more 45-59 year olds used Varenicline than any
other age group (and used NRT the least) while the lowest use of Varenicline was in the younger age groups (e.g. 18-24).
• No difference in prescribing by sex (approx. 75% NRT and 10-11% Varenicline)
• By deprivation: broadly similar percentages of each deprivation group being prescribed Varenicline (around 19%). Lowest in least deprived (14%)
Group Support• Around 30,000 quit attempts
made in 2010-2011 with group support
• The more deprived the area a client is from, the less chance the client will use group intervention. 30% - 5%
• Most popular in <18s and 45-59 age groups (23%) and lowest in those aged 18-34 (14-17%).
• Highest uptake in homemakers, the employed and sick/disabled (22-25%). Lowest in unemployed (11%).
0
10
20
30
40
50
60
one month cessation three month cessation
Perc
enta
ge (%
)
Group Intervention All Scotland
15
2- years reach
16
Client sub-groups
Pharmacy Multiple product
NRT
Varenicline Groupsupport
Total 70% SR 28% SR 10% SR 15% SR
Male 70% 1.00 27% 0.99 10% 1.00 15% 0.99
Female 70% 1.00 28% 1.01 10% 1.00 15% 1.00
Young (15-24) 77% 1.10 27% 0.98 5% 0.50 10% 0.67
Not young 70% 0.99 28% 1.00 11% 1.06 15% 1.04
Most deprived 76% 1.08 27% 0.97 8% 0.79 13% 0.89
Rest of Scotland 67% 0.95 28% 1.02 11% 1.12 16% 1.06
Pregnant 40% 0.62 11% 0.64
Not pregnant 67% 1.03 17% 1.03
Percentage and Standardised Rates
N=192,194
17
Standardised Rates (SR)
• Indirect standardised rates for each service• For example, when 88% of men undertook a
quit attempt in pharmacy (numerator)• And 80% of all the quit attempt in were in
pharmacy (denimonator)• The SR is then 88/80 = 1.10 for men in
pharmacy in that service• Over-representation of 10%• Rates suppressed when numerator < 5 and/or
denomiator < 30• Difference with Scotland’s SR (+/- 3 SD)
18
A funnel plot is a graph with a performance indicator on the vertical-axis
and the number of observations on the horizontal-axis
19
A data point is added for each service…
Scotland’s SR is added as a horizontal line...
20
Control limits are added, assuming that the services differ by chance
only…
21
Young age and group support
22
23
Results/Conclusion• The reach for some services showed gaps for
young people in – Providing multiple product NRT (13 out of 40)– Providing varenicline (13 out of 37)– Group support (19 out of 34)
• The reach for some services showed gaps for people in most deprived areas– Quit attempts in pharmacy (14 out of 40)
• Very few gaps were found for men and women• Pregnancy too many small numbers• Using Standardised Rates for smoking
cessation interventions seems to offer a useful frame work for analysing service reach
24
Questions …
Reach and Outcome
• Absolute and relative numbers of successful quit attempts
• Outcome varies from 27% to 55% (Scotland 38%)
• Reach varies from 6% to 29% (Scotland 18%)• Success in terms of
– High success rate / low volume– Low success rate / high volume
• Success per 100 smokers25
26