Alcohol screening and brief intervention: new evidence Colin - - PowerPoint PPT Presentation

Alcohol screening and brief intervention: new evidence

Colin Drummond

Evidence base for SBI

 Freemantle 1993 - 6 trials in primary care

– 24% drop in consumption (95% CI 18 to 31%)

 Moyer 2002 – 56 trials, 34 relevant to PHC

– Consistent positive effect, NNT 8-12 (smoking=20) – Cost savings found at 4 years in the USA

 Kaner 2007 – 29 trials in PHC & A&E

– Consistent positive effects ~7 drinks less/week – Evidence strongest for men, less work on women – No significant benefit of longer versus shorter BI

 National Institute for Healthcare and Clinical Excellence 2010

• Public health guidelines recommend implementation

SBI primary care 6 months (n=14)

Study or Subgroup 1.1.1 Europe Altisent et al., 1997 Diez Manrique et al.,2002 Fernandez et al., 1997 Heather et al., 1987 Huas et al., 2002 Kaner et al., 2013 Lock et al., 2006 Romelsjo et al., 1989 Subtotal (95% CI) Heterogeneity: Tau² = 0.01; Chi² = 10.32, df = 7 (P = 0.17); I² = 32% Test for overall effect: Z = 2.00 (P = 0.05) 1.1.2 Rest of the World Chang et al., 1997 Fleming et al., 1997 Fleming et al., 1999 Maisto et al., 2001 Ockene et al., 1999 Richmond et al., 1995 Senft et al., 1997 Subtotal (95% CI) Heterogeneity: Tau² = 0.02; Chi² = 15.27, df = 6 (P = 0.02); I² = 61% Test for overall effect: Z = 1.96 (P = 0.05) Total (95% CI) Heterogeneity: Tau² = 0.02; Chi² = 26.83, df = 14 (P = 0.02); I² = 48% Test for overall effect: Z = 2.98 (P = 0.003) Test for subgroup differences: Chi² = 0.04, df = 1 (P = 0.84), I² = 0% Mean 168 293.74

• 107.4

252.55

• 109

140 126.4

• 34.86

130.7 135.03 120.6 138.97 161.28 310 140.95 SD 167.2 186.21 370.4 156.37 164.73 81.73 98.48 209.04 191 127.68 89.88 143.62 181.76 241 177.31 Total 34 206 38 29 270 213 39 35 864 11 337 76 74 233 70 196 997 1861 Mean 280 302.74

• 64.6

317.76

• 92

134.4 199.68 42.64 109.71 174.83 193.08 147.05 170.24 276 160.7 SD 174.4 163.01 278.3 246.68 190.35 105.25 320.8 202.26 191 129.78 152.52 164.69 162.56 267 177.31 Total 30 186 50 62 149 209 32 36 754 8 366 68 85 248 66 215 1056 1810 Weight 3.2% 9.9% 4.2% 3.9% 9.8% 10.2% 3.5% 3.6% 48.3% 1.1% 11.9% 5.8% 6.3% 10.7% 5.7% 10.1% 51.7% 100.0% IV, Random, 95% CI

• 0.65 [-1.15, -0.14]
• 0.05 [-0.25, 0.15]
• 0.13 [-0.55, 0.29]
• 0.29 [-0.73, 0.15]
• 0.10 [-0.30, 0.10]

0.06 [-0.13, 0.25]

• 0.32 [-0.79, 0.15]
• 0.37 [-0.84, 0.10]
• 0.13 [-0.27, -0.00]

0.10 [-0.81, 1.02]

• 0.31 [-0.46, -0.16]
• 0.58 [-0.92, -0.25]
• 0.05 [-0.36, 0.26]
• 0.05 [-0.23, 0.13]

0.13 [-0.20, 0.47]

• 0.11 [-0.30, 0.08]
• 0.16 [-0.31, -0.00]
• 0.15 [-0.25, -0.05]

Brief Intervention Control

• Std. Mean Difference
• Std. Mean Difference

IV, Random, 95% CI

• 1
• 0.5

0.5 1 Favours BI Favours Control

SBI primary care 12 months (n=14)

Study or Subgroup 1.2.1 Europe Aalto et al., 2000 Beich et al., 2007 Cordoba et al., 1998 Huas et al., 2002 Kaner et al., 2013 Lock et al., 2006 Rubio et al., 2010 Wallace et al., 1988 Subtotal (95% CI) Heterogeneity: Tau² = 0.03; Chi² = 24.23, df = 7 (P = 0.001); I² = 71% Test for overall effect: Z = 2.67 (P = 0.008) 1.2.2 Rest of the World Fleming et al., 1997 Fleming et al., 1999 Fleming et al., 2004 Maisto et al., 2001 Reiff-Hekking et al, 2005 Richmond et al., 1995 Subtotal (95% CI) Heterogeneity: Tau² = 0.04; Chi² = 17.69, df = 5 (P = 0.003); I² = 72% Test for overall effect: Z = 1.78 (P = 0.08) Total (95% CI) Heterogeneity: Tau² = 0.03; Chi² = 41.97, df = 13 (P < 0.0001); I² = 69% Test for overall effect: Z = 3.36 (P = 0.0008) Test for subgroup differences: Chi² = 0.01, df = 1 (P = 0.94), I² = 0% Mean 278.3 168 202.4

• 109

134.4 128.64 245.76 304.34 137.76 119.04 57.64 133.98 161.28 326 SD 280.69 152.4 183.27 164.73 121.15 293.28 116.48 184.94 135.72 83.64 106.39 147.52 190.72 211 Total 82 224 104 270 208 36 371 363 1658 353 78 81 74 235 70 891 2549 Mean 262.79 168 295.2

• 92

140 156.8 284.67 386.15 185.52 195.24 65.99 147.33 170.24 290 SD 299.4 156 215.22 190.35 70.2 293.28 116.61 230.97 155.16 146.04 74.34 147.72 167.68 208 Total 73 288 125 149 194 42 381 385 1637 370 67 70 85 210 61 863 2500 Weight 5.7% 8.7% 6.7% 8.1% 8.2% 3.8% 9.4% 9.4% 60.2% 9.4% 5.4% 5.6% 5.8% 8.5% 5.2% 39.8% 100.0% IV, Random, 95% CI 0.05 [-0.26, 0.37] 0.00 [-0.17, 0.17]

• 0.46 [-0.72, -0.20]
• 0.10 [-0.30, 0.10]
• 0.06 [-0.25, 0.14]
• 0.10 [-0.54, 0.35]
• 0.33 [-0.48, -0.19]
• 0.39 [-0.53, -0.24]
• 0.19 [-0.32, -0.05]
• 0.33 [-0.47, -0.18]
• 0.65 [-0.98, -0.31]
• 0.09 [-0.41, 0.23]
• 0.09 [-0.40, 0.22]
• 0.05 [-0.24, 0.14]

0.17 [-0.17, 0.51]

• 0.18 [-0.37, 0.02]
• 0.18 [-0.29, -0.08]

Brief Intervention Control

• Std. Mean Difference
• Std. Mean Difference

IV, Random, 95% CI

• 1
• 0.5

0.5 1 Favours experimental Favours control

Alcohol Screening and Brief Intervention Research Programme

SIPS

A&E St. Mary’s 'Scientia Vincit Timorem'

Programme design

• Funded by Department of Health for 3 years
• Jointly led by IOP & Newcastle University
• 3 cluster randomised clinical trials of alcohol

screening and brief intervention (PHC, AED, CJS) to assess:

– What are the barriers/facilitators to implementation in a “typical setting”? – What is the most effective screening method? – What is the most effective and cost effective intervention approach?

• Total target of 2,403 subjects, completed 2,485

July 2009

• 6 and 12 months follow up, 74% @ 6 months

(mainly phone) 69% @ 12 months

Screening tools and methods

• Universal versus targeted screening –

PHC and ED

• Single Alcohol Screening Question (M-

SASQ)

• Fast Alcohol Screening Questionnaire

(FAST)

• Paddington Alcohol Test (SIPS-PAT) –ED
• nly
• (AUDIT as part of baseline assessment)

Interventions

• Patient Information Leaflet (PIL) – DH How

much is too much?

• Brief Advice (BA) – 5 min of simple

structured advice based on WHO Drinkless Programme modified for SIPS

• Brief Lifestyle Counselling – 20 min

motivational intervention based on WHO Drinkless programme modified for SIPS delivered by a trained Alcohol Health Worker (or Practice Nurse in PHC)

Website

www.sips.kcl.ac.uk

9

Training and intervention tools

10

SIPS in numbers

Staff trained (n) Approached (n) Eligible (n) % Positive (n) % Consent (n) % ED 278 5,992 3,737 62 1,491 40 1,204 81 PHC 195 3,562 2,991 84 900 30 756 84 CJS 227 967 860 88 574 67 525 91 Totals 700 10,521 7,588 72 2,965 39 2,485 84

A&E St. Mary’s 'Scientia Vincit Timorem'

SIPS results

Primary Health Care

PHC survey results

Mean ratings and standard deviations for key conditions Key conditions N Mean SD Mental health problems 86 1.60 1.08 Gastrointestinal problems 84 1.94 1.08 Hypertension 82 2.17 0.95 Injuries 85 2.38 1.01

• 68% preferred targeted to universal
• 14% preferred universal to targeted
• 17% felt both appropriate
• New patient registrations and smoking cessation

also felt to be important contexts for screening

Proportion screened and positive by targeted condition

Targeted

• Condition

Screened n (%) Positive n (%) Prevalence (%)

• Gastrointestinal

Hypertension Injuries Mental Health

• New

Registrations

• Universal

Condition

• 124

(9.8) 623 (49.4) 75 (5.9) 167 (13.2) 273 (21.6)

• 1632
• 44

(9.7) 208 (45.9) 30 (6.6) 86 (19.0) 85 (18.8)

• 421
• 35.5

33.4 40.0 51.5 31.1

• 25.8
• Targeted screening results in significantly more screen positives

(overall 36.2% versus 25.8%)

Relative Risk of being screen positive for targeted conditions in the targeted group versus the universal screen population

ç Relative Risk (95% CI)

• Significant

Mental Health

• Injuries

Gastrointestinal Hypertension

• 1.9963

(1.6865; 2.3630) 1.5506 (1.1613; 2.0705) 1.3755 (1.0700; 1.7683) 1.2942 (1.2773; 1.4859)

• Not

significant New Registrations

• 1.2079

(0.9935; 1.4663)

Proportion screened and positive by targeted condition in the universal arm – presenting conditions in universal arm patients were coded into wider range of conditions and contexts

Targeted

• Condition

Screened n (%) Positive n (%) Prevalence %

• Abnormal

blood results Ante-natal Chronic disease work Convulsions Gastrointestinal Hypertension Injuries Medical Certificates Mental Health

• New

Registration Obesity Sexual Health Work Smoking Cessation Substance Misuse

• No

Target

• Overall
• 25

(1.5) 16 (1.0) 69 (4.2) 4 (0.2) 73 (4.5) 142 (8.7) 23 (1.4) 17 (1.0) 95 (5.8) 20 (1.2) 14 (0.9) 60 (3.7) 12 (0.7) 14 (0.9)

• 1048

(64.2)

• 1632
• 8

(1.9) 2 (0.5) 13 (3.1) 1 (0.2) 20 (4.8) 32 (7.6) 4 (1.0) 8 (1.9) 23 (5.5) 7 (1.7) 1 (0.2) 14 (3.3) 7 (1.7) 4 (1.0)

• 277

(65.8)

• 421
• 32.0

12.5 18.8 25.0 27.4 22.5 17.4 47.0 24.2 35.0 7.1 23.3 58.3 28.6

• 26.4
• 25.8
• Broader targeting misses 66% of alcohol cases
• SIPS targeting misses 80% of alcohol cases

What does it all mean?

Targeting versus universal….

• In PHC targeting yields more screen positives than

universal screening.

• But 66% of PHC presentations in the universal arm do

not meet the (broader) targeting criteria.

• So while targeting is more efficient and more popular

with PHC staff, universal identifies a higher proportion

• f the at risk population – implications for public health

impact of SBI

PHC - Effectiveness

Changes in the proportion of AUDIT positives overall and by intervention at 6 and 12 months

Cost effectiveness plane for BA over and above PIL

Cost effectiveness acceptability curve for PIL and BLC

PHC main points

• FAST more efficient than SASQ in PHC
• Targeted more efficient than universal screening but

identifies only small proportion of cases

• Overall, risk drinking significantly fell between baseline and 6

and 12 months

• No significant differences between BI conditions
• PIL more cost effective than BLC or BA
• “All patients received simple feedback on their screening
• utcome. Beyond this input, however, evidence that brief

advice or brief lifestyle counselling provided important additional benefit in reducing hazardous or harmful drinking compared with the patient information leaflet was lacking.”

• Therefore BLC should be reserved for non-responders to

simple BI

What this study adds

• “Brief advice and brief lifestyle counselling

did not provide a statistically significant benefit in reducing hazardous or harmful drinking compared with a patient information leaflet”

• “Screening followed by simple feedback

and written information may be the most appropriate strategy to reduce hazardous and harmful drinking in primary care”

• Targeted screening more efficient but

universal screening best for PH impact

ED - Effectiveness

Changes in the proportion of AUDIT positives overall and by intervention at 6 and 12 months

Preventing harmful drinking (PH24)

• All NHS professionals and non-NHS
• Routine alcohol screening

– Universal – Targeted “if not feasible”

• Validated screening tool (AUDIT, FAST etc)
• Don’t use biological markers
• Structured brief advice- all hazardous/harmful
• Extended brief- non-responders
• Referral of moderate/severe alcohol

dependence/non-responders to brief interventions

NICE care pathway – case identification diagnosis

Screen (FAST, SASQ, AUDIT-C) indicates possible alcohol use disorder

Administer: AUDIT AUDIT < 8 AUDIT 16–19 Harmful drinking AUDIT 8–15 Hazardous drinking AUDIT 20+ Probable alcohol dependence Referral to specialist assessment Referral to specialist assessment where no improve maintained Extended brief intervention(s) Review of progress Consider Tier 2 interventions Structured alcohol interventions Assisted withdrawal assessment Brief intervention

Overall conclusions

• Universal screening reaches wider target population than

targeted in PHC

• Screening tools:

– FAST best screening tool in PHC and CJS – SASQ best screening tool in ED

• More intensive BI adds no significant clinical benefit to

simple clinical feedback plus alcohol information leaflet

• More intensive intervention should be reserved for more

severe cases and non-responders to less intensive BI

• PHC: feasible to implement but needs financial

incentives

• ED and CJS less feasible unless specialist alcohol staff

deployed

• Need full spectrum of services across the care pathway