| pA2 online © Copyright 2004 The British Pharmacological Society |
210P
GKT, University of London Winter Meeting December 2003 |
|
Variation
in the prescribing of antidepressants, anxiolytics, hypnotics and
sedatives in general practice |
|
Print abstractThere is a widespread concern about the use and prescribing patterns of drugs affecting the central nervous system in part due to their potential for abuse and dependence.
We estimated the extent of variation between general practitioners (GPs) in the prescribing of antidepressants (AD), anxiolytics, hypnotics and sedatives (hypn/sed; primarily benzodiazepines (BZD) and BZD related drugs) within the General Medical Services (GMS) and whether there was any association between the prescribing of these drug groups.
The GMS scheme provides free medical services to over 350,000 persons in the Eastern region of Ireland. It is means tested for those aged under 70years; however all persons aged over 70 are included. Prescription data relating to individual general practitioners during the period January-December 2002 was extracted from the national prescription claims database and analysed. GPs with a panel size greater than 100 patients (83%; n=512) were included in the study. Prescribing data for the selected drug groups were compared between GPs using the Standardised Prescribing Ratio (SPR) (Johnson et al 1997) in order to adjust for the influence of practice demography (age and gender). The drug groups were defined using the WHO Anatomical Therapeutic Classification (ATC). The ratio between the 90th and 10th percentile was used to determine the variability of the prescribing rates between GPs. Regression analysis was used to examine if there was association between the prescribing of these drug groups. All analysis was carried out using SAS version 8 (SAS Institute).
The results are summarised
below;
Table 1: Median SPR and variation indices for selected drugs
| Drug class(ATC code) |
Median
SPR(interquartile range)
|
Variation
index
|
| AD (N06A) |
108.6
(80.6-140.4)
|
3.3
|
| Anxiolytic (N05B) |
102.1
(80.0-135.2)
|
3.1
|
| Hypn/sed (N05C) |
107.8
(81.6-143.1)
|
3.1
|
Table 2: Correlation coefficient (r) between the SPR's of the different drug classes (p<0.001)
| DRUG CLASS |
Anxiolytic
|
Hypn/sed
|
| AD |
0.67
|
0.77
|
| Anxiolytic |
-
|
0.87
|
The variation index indicates a 3-fold variation in prescribing of these drug classes (Table 1) and a highly significant positive correlation (Table 2) between GPs who are high prescribers of one group of drug also being high prescribers of other groups.
The study shows that there is a wide variation between prescribers even when practice demography is controlled for and immeasurable factors such as medical training, patients' demands and expectations could be contributory. Regional guidelines and policies regarding the prescribing of these drugs need to be established to reduce the potential of abuse and dependence associated with their use.
Johnson Z, et al (1997) Pharmacoepidemiology & Drug Safety 6: 337-45.