Skip to main content

Table 1 Educational Interventions to decrease benzodiazepine use

From: Improving the use of benzodiazepines-Is it possible? A non-systematic review of interventions tried in the last 20 years

Study

Location

Intervention design:

Educational Letter Interventions Targeting Consumers

Study Design and Size

Result

Follow up

Bashir et al, 1994 [57]

London

Letters sent long term users (> 1 year) with a request to visit their GP and allocated to:

1 = informed about the correct use of BZ + self help book. 2 = control (no advice)

RCT

11 GP practices 109 patients.

Sig increase in the number of patients' that reduced BZ prescriptions in intervention group (18%) compared to control (5%).

6 months post intervention

Towle and Adams, 2006 [25]

Scotland

Letter sent to repeat BZ users by pharmacist to tell them that a step down program had been initiated and invited them to see their GP for a medication review + repeat BZ prescriptions were inactivated. Posters displayed in GP practice.

Convenience Sampling

1 practice,

206 patients.

73% collected their step down program. Decreased number of tablets prescribed by 64%. Only 23% of 369 patients stayed on a repeat script. (No statistical analysis).

3 years (end of the study)

Heather et al, 2004

[26]

UK

Long term BZ users (> 6 months) were allocated to:

1 = letter from GP asking them to consider cutting down their BZs. 2 = letter inviting participants to a consultation + leaflet and self help sleep book. 3 = control.

RCT

7 Practices, 299 patients.

Sig reduction in letter group (24%) vs. control (16%) and sig reduction in consult (22%) vs. control (16%). No sig difference between consultation and letter group for BZ decrease.

6 months post intervention

Cormack et al, 1989

[58]

UK

Long term BZ users (> 1 year) were allocated to: 1 = letter from GP advising patients to cut down BZ use. 2 = invitation to see the GP about cutting down BZ use. 3 = control.

CT

5 practices, 75 patients.

Sig reduction in BZ use in both the letter group and the consultation group compared to no change in control group (data not reported).

6 months to 1 year post intervention

Cormack et al, 1994

[23]

England

Long term BZ users (> 6 months) allocated to:

1 = letter from GP to recommending stopping BZs 2 = letter from GP + monthly advice sheets.

3 = control

CT

3 practices 209 patients.

30% reduction in BZ use by intervention groups (sig difference from control) No sig difference between interventions.

6 months post intervention

Gorgels et al, 2005

[59]

Netherlands

1 = discontinuation letter to long term users then approximately 3 months after the letter, an invitation for GP evaluation of BZ use. 2 = control.

RCT

30 practices

1707 patients

(intervention) and 1821 (control).

Sig use reduction of BZ prescriptions in intervention group (24%) vs. control (5%). Sig reduction in BZ use in the intervention group for those who attended the GP evaluation (35%) vs. those who did not go (24%).

21 months post intervention

Morrison, 1990

[60]

UK

Long term BZ users (> 6 months) were informed that they should stop BZs by their GP. If agreed an individual plan (how to decrease and number of follow-up visits) was established.

Convenience Sampling

1 practice, 27/72 eligible patients agreed.

No new long term users started during study. 37.5% quit 33.3% reduced dose by >50% and 15% remained same dose or <50% reduction. No-one increased their dose during the study period.

1 year (end of study)

Onyett and Turpin, 1988 [28]

UK

Recruitment of long term users by notices. Asked to self-reduce BZs then all participants received a pamphlet and allocated to: 1 = group session. 2 = individual GP appointment.

Prospective cohort

18 patients.

59% reduction in dose (group) and 69% reduction in dose (GP meeting) but no sig difference between groups.

15 week post intervention

Brymer and Rusnell, 2000 [27]

Canada

Home assessment by nurse to determine if patients were substance dependent. Saw Geriatrician for medical review, who recommended a treatment plan (also sent to their GP). Were encouraged to join support/educational groups.

Observational study

55 elderly patients.

Significant reduction of BZ use (59%) between pre and post intervention.

6 months post intervention

Study

Location

Intervention Design:

Educational Interventions Targeting GPs

Study Design and Size

Result

Follow up

Zwar et al, 2000

[61]

Australia

1 = face to face 20 min educational visit by a GP focusing on the management of long term users of BZs + guidelines + leaflets on relaxation techniques for patients. 2 = control.

RCT

157 GP registrars

Sig decrease in overall BZs use in both groups (0.6 per 100 encounters for both group), however no difference between control and intervention groups.

3 practice surveys - 6 monthly

Smith et al, 1998

[30]

Washington, USA

1 = mailed intervention package (guidelines, letter, prescriber-specific profile, patient profile) for prescribers of over users (1 tab per day >1 year)

2 = control.

RCT

GPs or pharmacist of 222 over users

Sig decrease in BZ prescribing/dose for the intervention group (27.6%) compared to control (8.5%).

3 months post intervention

Holm, 1990

[32]

Aarhus, Denmark

1 = invite to a meeting on correct use of hypnotics/sedatives + educational material given at the meeting. 2 = mailed information on correct use and feedback on their prescribing rate compared with others. 3 = control.

RCT

356 GPs (245 practices)

Sig decrease (-53) in DDD/1000 pat/week between pre and post intervention. No difference between groups 1 and 2 but sig difference between groups 2 and 3 (2 prescribed more).

1-2 months post intervention

De Burgh et al, 1995 [29]

NSW, Australia

1 = 20 min educational visit to GPs. Educational material left (management guidelines, review cards for long term users). Offered access to sleep aids. Recommended to review 5 patients and received a follow-up phone call. 2 = control.

RCT

286 GPs

An overall decrease in BZ prescribing (23.7%). No sig difference between intervention and control for reduction rate of BZ. Sig decrease between intervention (72% decrease) and control (13% decrease) for rate of new BZ scripts for insomnia.

Approx 6 months post intervention

Midlov et al, 2006 [62]

Skane, Sweden

1 = two educational visits from pharmacist and GP focusing on effects of medium and long acting BZs in the elderly. 2 = control (received intervention after the study).

RCT

15 GPs (8 in intervention, 7 in control)

Sig decreases in all BZ prescribing (26.63%) and 25.8% decrease of long and medium acting BZs vs. control.

1 year post intervention

Berings et al, 1994 [31]

East and West Flanders, Belgium

1 = educational advertisement like mailings (with slogan) + educational visit. 2 = mailings only 3 = control

RCT

128 physicians

Sig decrease for whole sample pre to post intervention. Sig decrease in BZ prescriptions between intervention 1 (24%) and control (3%) and intervention 2 (14%) and control and intervention 1 and 2.

4 weeks post intervention

Study

Location

Intervention Design: Educational Interventions Targeting LTC

Study Design and Size

Result

Follow up

Hagen et al, 2005 [63]

Alberta

1 = Algorithm on non-pharmacological approaches for agitation. Education based on algorithm to nurses, pharmacists, or family members. 30 min education session for GP. 2 = Control

CT

24 LTC facilities, 12 (intervention) and 12 (control)

BZ use in both the control and the intervention increased post intervention (Sig increase for control only). At 6 months post intervention total BZ use sig higher in control vs. intervention.

Every 2 months until 6 months post intervention

Avorn et al, 1992 [64]

Massachusetts

1 = physicians received 3 × drug advertisement like summaries of literature about geriatric medicine, psychopharmacology + 3 face to face visits to each doc by pharmacist. 4 training sessions for nurses on alternatives to psychoactive drugs + ADRs. 2 = Control

MPR

12 LTC facilities 6 (intervention) and 6 (control) both groups were matched

Sig difference in the % change to more appropriate BZs in intervention (64%) vs. control (4%). E.g. long acting to short acting.

30 days post intervention

Schmidt et al, 1998 [65]

Sweden

1 = Monthly meeting for 12 months led by pharmacist and included physician, and nurses. Each patient's medications were reviewed. 2 = Control

RCT

33 nursing homes (15 intervention, 18 control)

Sig increase (from baseline) in prescribing of appropriate hypnotics (+70%) and anxiolytics (+ 50%) in intervention group. No sig difference in control group

1 month post intervention

Gilbert et al, 1993 [36]

Adelaide, Australia

1 = letter to residents inviting participation in relaxation groups (8 × 40 min) audio tape of relaxation for practice and information about sleep & anxiety medications. Nurses received a seminar on dealing with BZ withdrawal. Doctors received letter of progress. 2 = Control

Prospective cohort

2 LTC facilities

There was a sig decrease in the % of BZs users' from baseline (70%) to post intervention (35%). No change in control

12 weeks after baseline

  1. BZ = benzodiazepines, GP = General Practitioner, LTC = long term care, Sig = statistically significant (p < 0.05), tab = tablet, ADRs = Adverse Drug Reactions, DDD/1000 pats/week = the number of defined daily doses dispensed per 1000 patients per week, RCT = Randomized controlled trial, CT = Controlled trial, MPR = Matched pair randomization