Abstract
Background
The trends in sedative use have varied in recent years. Benzodiazepines and z-drugs are indicated for anxiety and/or sleep disorders but should be limited to short-term use. The aim of this study is to examine trends and patterns in sedative prescribing in Ireland between 2014 and 2022, as well as comparing trends between Ireland and England within the same period.
Methods
Monthly data on medicines prescribed and dispensed in primary care on the means-tested General Medical Services (GMS) scheme in Ireland were used. Volumes of prescribed benzodiazepine and z-drug use and patterns of prescribing, including initiations, discontinuations, chronic use, and high-risk prescribing were summarized per year. Other sedating agents (sedating antihistamines, antidepressants, and antipsychotics) were also analysed. Volume of use outcomes were compared with NHS data from England for the same period.
Results
The rate of benzodiazepine and z-drug dispensings per 1000 GMS population decreased by 5%, from 1531 in 2014 to 1474 in 2022. By comparison in England, there was a steeper decrease of 27% in the dispensing rate and the level of use was substantially lower, falling from 288 dispensings per 1000 population in 2014 to 210 in 2022. In Ireland, dispensing rates were highest amongst women and older age groups. High-risk dispensings of benzodiazepines and z-drugs decreased over the study period.
Discussion
Despite decreases in benzodiazepine and z-drug dispensings, rates remain high in Ireland and may suggest a need for enhanced availability of non-pharmacological interventions, and improved education and deprescribing support for healthcare professionals.
Keywords: prescribing, primary care, pharmacology/drug reactions, population health, public health
Key messages.
Findings suggest that sedative dispensing is decreasing in Ireland
High-risk dispensings and dispensings to older adults are also decreasing
In comparison to England, dispensing rates remain high
Prevalence of discontinuations have decreased over time
There is a need for education and deprescribing support for health professionals
Background
Benzodiazepines and z-drugs are indicated for acute anxiety and/or sleep disorders, with certain benzodiazepines also used for other primary health conditions, including epilepsy and muscle spasm, and in the context of alcohol withdrawal and end-of-life care. The use of benzodiazepines and z-drugs is associated with significant side effects and a high risk of dependency, with cessation often accompanied by withdrawal symptoms, in particular for chronic users of short-acting variants [1]. For these reasons, many guidelines indicate that prescriptions should be limited to short-term use, with the British National Formulary (BNF) recommending that uninterrupted usage does not exceed 4 weeks [2].
A recent analysis of global trends in benzodiazepine and z-drug use found distinct differences in consumption and trends across countries in regions, with the consumption in high-income countries much higher than in middle-income countries [3]. In Ireland, previous research has shown that there was a reduced trend in primary care benzodiazepine prescribing between 2005 and 2015, while prescribing of z-drugs increased [4]; however, a more recent analysis found that both classes increased 2017–2019 [5]. In the UK, prescribing has reduced in recent years, with the number of people prescribed a benzodiazepine reducing from 1.4 million to 1.1 million in 2017–2021 [6]. Efforts to reduce potentially inappropriate prescribing of benzodiazepines and z-drugs may have the unintended consequence of increasing use of alternative agents with their own risks. This includes drugs from other drug classes used off-licence for sedative purposes, including antidepressants (doxepin, trazodone, and mirtazapine) all of which have seen increased dispensings in recent years in the UK [7]. Similarly, dispensings of atypical antipsychotics (risperidone, quetiapine, and olanzapine), often prescribed at low dosages for sleep difficulties [8], have increased [9]. A recent Canadian study found that while prescribing of benzodiazepines and z-drugs have decreased in recent years, low-dose trazodone and quetiapine have increased [10].
Describing and quantifying patterns and quality of medication use is important to monitor medication safety at the population level, however a range of indicators are needed to capture differences. The Organisation for Economic Cooperation and Development (OECD) has outlined 11 quality indicators for primary care prescribing, including indicators of long-term use of benzodiazepines amongst older adults [11]. Ireland has had the second highest rate of chronic benzodiazepine use among older adults among OECD countries since 2011, and unlike other countries, saw little decline in this indicator between 2011 and 2021 [12]. While this is an important indicator of appropriate prescribing of these medications across different national contexts [13], it provides a limited assessment of utilization, not accounting for rates of initiation or discontinuation, or other forms of high-risk prescribing such as co-prescribing with other sedative drugs or utilization of high doses. Therefore, there is a need to examine broader indicators of use and high-risk use in primary care.
The aim of this study was to examine trends in volume and patterns, including high-risk patterns, of benzodiazepines, z-drugs, and other sedative dispensings in Ireland between 2014 and 2022. A secondary aim was to assess trends in volume of sedatives in Ireland versus England in a cross-country comparison.
Methods
This is an observational administrative database study of medicines dispensing in primary care in Ireland between 2014 and 2022. The protocol for the project this study relates to has been previously published [14]. The study was approved by the Royal College of Surgeons in Ireland (RCSI) University of Medicine and Health Sciences Research Ethics Committee (REC202201015) and the Health Service Executive (HSE) Reference Research Ethics Committee B (RRECB1022FM).
Population and data sources
This study includes individuals of all ages eligible for the General Medical Services (GMS) scheme in Ireland. Eligibility to this scheme is based on age and income, with income thresholds increasing with age, and therefore eligible persons tend to be older and more socioeconomically deprived than the general population [15]. The scheme covers ∼32% of the population, increasing to 78% of individuals aged 75 and older. Approximately 29% of the GMS population was aged ≥65 in 2021, compared to 15% of the general population of approximately 5 million [16].
Data were obtained from the HSE Primary Care Reimbursement Service (PCRS), which administers community drug schemes in Ireland. Pharmacies transmit claims for medications dispensed under these schemes to the PCRS at the end of each month for reimbursement. For GMS eligible individuals, the data provide complete information on dispensed medications. The data are hierarchical, with dispensed medications nested within prescriptions, prescriptions nested within individuals, and individuals nested within prescribers [15].
Data were requested from the PCRS for this study through an information request and provided to the research team under a data transfer agreement. Data were at the level of individual medications dispensed to a patient. For each item, information included anonymised identifiers for individuals, prescribers, and pharmacies, Local Health Office (LHO), date of dispensing, sex and age group of the individual, drug information [World Health Organisation (WHO) Anatomical Therapeutic Chemical (ATC) code, product name and strength, and defined daily dosage (DDD)], quantity dispensed, and cost. For the cross-country comparison of volume sedative use, aggregate data relating to prescribing in all GP practices in England were used. See Supplementary Box S1. Cross-national comparisons play a role in identifying variation in medication use between countries, to enhance understanding of the implications of varying prescribing practices and to develop strategies to optimize medication use. England was selected as a comparator, due to the availability of open data and cultural, geographic, and other similarities with Ireland.
Outcomes
Sedatives, including benzodiazepines, z-drugs, antihistamines, and selected antidepressants and atypical antipsychotics dispensed as low-dose products, were identified using WHO ATC codes in data for Ireland and mapped to the equivalent BNF codes for England (see Supplementary Table S1). Outcomes were derived for all drug classes and individual benzodiazepines, and for long-acting benzodiazepines [17]. Volume outcomes included number of dispensings, costs, and standard doses, expressed as rates per 1000 patients. For Ireland, denominators were obtained from PCRS annual reports which report number of eligible individuals at the end of each calendar year [18]. For England, denominators were obtained from the NHS Digital portal [19]. Standard doses capture medications consumption on a common scale across different medications, drugs, or drug classes, and WHO DDDs [20], and diazepam dose equivalents (DDE) were used [21] (see Supplementary Table S2).
For Ireland, pattern of use outcomes were derived, including volume of use stratified by age group and sex, prevalence of use, and prevalence of initiation, discontinuation, chronic use, and high-risk use, as defined by the Scottish Polypharmacy Guidance [22]. An overview of included outcomes is provided in Supplementary Table S3. High-risk use included high-dose use due to dependency risk, and co-prescription with other sedative or anticholinergic drugs due to fall risk. A list of anticholinergic drugs, as defined by the Anticholinergic Burden Scale [23] is available in Supplementary Table S4.
Analysis
The outcomes above were calculated annually from 2014 to 2022. For volume outcomes, absolute and relative changes from 2014 to 2022 were calculated, as well as the rate ratio (rate in Ireland divided by rate in England). The Jonckheere–Terpstra test for trend across years was run for dispensing rate for each drug for Ireland and England. For the outcomes of prevalence of use, initiations, discontinuations, and chronic use in Ireland, the Cochran-Armitage test for trend was run across years. Considering data from Ireland relates to the GMS scheme, which over represents those with lower socioeconomic status, a subgroup analysis was conducted using data for England from higher deprivation clinical commissioning groups (CCGs) only (specifically those in the top 33 centiles of deprivation, based on the Index of Multiple of Deprivation 2019 [24]. Analyses were largely descriptive without inferential statistics, as study participants represent the full population. All analyses were conducted in Stata v18 (College Station, TX, USA: StataCorp LLC) and statistical significance was defined as P < 0.05.
Results
The number of GMS eligible patients in Ireland decreased from 1 768 700 in 2014 to 1 568 379 in 2022; the NHS population in England rose from 56 545 892 in 2014 to 61 768 942 in 2022 (see Supplementary Table S5).
Volume of utilization
The rate of benzodiazepine and z-drug dispensings per 1000 GMS population increased in the initial study period from 1521 per 1000 population in 2014, peaking at 1569 in 2019, and then decreasing to 1445 in 2022: a 5% decrease from 2014 to 2022 (see Table 1). The rate of long-acting benzodiazepine dispensings decreased from 437 to 399, a 9% decrease. The trends for individual drugs varied, with moderate increases seen for clonazepam (17%, 44 to 51 dispensings per 1000), lorazepam (12%, 26 to 29), zolpidem (7%, 270 to 289), and zopiclone (6%, 341 to 361). The largest relative decreases were seen lormetazepam (100%, 31 to 0), nitrazepam (97%, 13.6 to 0.4), and prazepam (58%, 10.8 to 4.5). Zopiclone and zolpidem had the highest rates of dispensing, followed by diazepam and alprazolam (see Fig. 1). The rate of dispensings of sedating antidepressants (low-dose doxepin, trazodone, and mirtazapine) increased by 112% (92 to 195 dispensings per 1000), while sedating antihistamines (promethazine, cyclizine, and ketotifen) increased by 570% (11 to 71) and sedating antipsychotics (low-dose olanzapine, quetiapine, and risperidone) by 95% (118 to 230). Combining these with benzodiazepines/z-drugs, the overall dispensing rate of sedative drugs rose from 1741 per 1000 in 2014 to 1940 in 2022, an 11% relative increase.
Table 1.
Rate of dispensings per 1000 GMS population in Ireland.
| 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | Absolute change | Relative change | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Benzodiazepines and z-drugs | 1521.0 | 1507.2 | 1549.0 | 1562.0 | 1568.9 | 1565.6 | 1506.4 | 1513.4 | 1444.9 | −76.1 | −0.05 |
| Benzodiazepines | 910.2 | 894.5 | 909.2 | 907.1 | 903.3 | 887.7 | 837.3 | 835.8 | 794.9 | −115.3d | −0.13 |
| Long-acting benzodiazepines | 436.7 | 429.7 | 438.3 | 441.4 | 441.6 | 440.1 | 421.1 | 423.2 | 399.1 | −37.6 | −0.09 |
| Clonazepam | 43.8 | 44.9 | 48.5 | 52.1 | 54.5 | 55.7 | 55.7 | 58.1 | 51.4 | 7.6d | 0.17 |
| Diazepam | 258.1 | 258.1 | 265.6 | 269.0 | 270.8 | 272.0 | 262.9 | 269.7 | 264.6 | 6.4 | 0.02 |
| Chloridiazepoxide | 18.5 | 17.8 | 18.2 | 18.0 | 17.5 | 18.0 | 16.5 | 16.8 | 15.1 | −3.3d | −0.18 |
| Potassium Clorazepate | 0.1 | 0.1 | 0.1 | 0.1 | 0.1 | 0.1 | 0.1 | 0.1 | 0.1 | 0.0 | −0.13 |
| Lorazepam | 26.0 | 25.4 | 26.0 | 26.3 | 27.4 | 28.1 | 28.2 | 29.2 | 29.0 | 3.0d | 0.12 |
| Bromazepam | 56.1 | 54.8 | 55.3 | 55.1 | 54.5 | 53.1 | 50.2 | 48.7 | 45.8 | −10.3d | −0.18 |
| Clobazam | 12.0 | 10.6 | 10.2 | 9.9 | 9.6 | 9.1 | 8.7 | 8.9 | 9.2 | −2.8d | −0.23 |
| Prazepam | 10.8 | 10.3 | 10.0 | 9.7 | 9.6 | 9.2 | 7.8 | 7.6 | 4.5 | −6.2d | −0.58 |
| Alprazolam | 238.4 | 239.2 | 248.5 | 250.8 | 251.8 | 253.3 | 242.8 | 243.4 | 236.8 | −1.5 | −0.01 |
| Flurazepam | 79.9 | 75.7 | 73.8 | 71.6 | 69.5 | 66.3 | 60.5 | 58.7 | 53.8 | −26.1d | −0.33 |
| Nitrazepam | 13.6 | 12.2 | 11.7 | 10.9 | 10.1 | 9.7 | 8.9 | 3.3 | 0.4 | −13.2d | −0.97 |
| Triazolam | 48.2 | 48.4 | 48.8 | 48.7 | 48.5 | 47.9 | 46.2 | 45.5 | 42.8 | −5.4d | −0.11 |
| Lormetazepam | 31.0 | 28.8 | 28.4 | 27.0 | 25.5 | 14.0 | 1.1 | 0.1 | 0.0 | −31.0d | −1.00 |
| Temazepam | 69.5 | 63.3 | 60.9 | 57.5 | 53.9 | 51.2 | 47.7 | 45.8 | 41.3 | −28.2d | −0.41 |
| Z-drugs | 610.8 | 612.7 | 639.8 | 655.0 | 665.6 | 677.9 | 669.1 | 677.6 | 650.0 | 39.2 | 0.06 |
| Zopiclone | 341.1 | 343.7 | 358.5 | 366.6 | 371.1 | 377.6 | 370.3 | 375.6 | 360.8 | 19.7d | 0.06 |
| Zolpidem | 269.7 | 269.1 | 281.4 | 288.4 | 294.5 | 300.2 | 298.8 | 301.9 | 289.2 | 19.5d | 0.07 |
| Antidepressantsa | 91.7 | 104.6 | 118.4 | 131.7 | 146.3 | 159.0 | 172.4 | 192.9 | 194.6 | 102.9d | 1.12 |
| Antihistaminesb | 10.5 | 13.2 | 17.8 | 24.1 | 29.5 | 39.3 | 50.6 | 64.8 | 70.5 | 60.0d | 5.70 |
| Antipsychoticsc | 118.1 | 131.4 | 146.2 | 162.2 | 179.1 | 198.2 | 211.6 | 230.3 | 229.9 | 111.8d | 0.95 |
| All sedatives | 1741.4 | 1756.4 | 1831.5 | 1880.0 | 1923.8 | 1962.1 | 1941.1 | 2001.3 | 1940.0 | 198.7d | 0.11 |
aDoxepin 10 mg, trazodone 50 and 100 mg, mirtazapine 15 mg.
bPromethazine, cyclizine, ketotifen.
cOlanzapine 2.5 mg, quetiapine 25 mg, risperidone 0.5 mg.
dTest for trend significant at 0.05 based on the Jonckheere–Terpstra test.
Figure 1.
Rate of dispensing per 1000 GMS population for individual benzodiazepines and z-drugs.
In England, the rate of benzodiazepine and z-drug dispensings decreased by 27% from 288 in 2014 to 210 per 1000 in 2022. Multiple drugs dispensed in Ireland were not dispensed in England during the study period, including potassium clorazepate, bromazepam, prazepam, alprazolam, flurazepam, and triazolam. All sedative dispensing rate remained steady throughout the study period (392 per 1000 in 2014 and 391 in 2022, see Supplementary Table S6). Figure 2 reports rate ratios for dispensings in Ireland versus England for 2014 and 2022. There were similar changes in the cost rates per 1000 GMS/NHS population, as shown in Supplementary Table S7.
Figure 2.
Rate ratio for dispensings in Ireland versus England for 2014 and 2022*. *Rates above 1.0 indicate higher dispensings in Ireland.
The rate of diazepam dose equivalent (DDEs) dispensed per 1000 GMS population per day for all benzodiazepines decreased from 834 in 2014 to 764 in 2022, a 9% decrease, with most individual drugs decreasing (see Supplementary Table S7). Rate ratio plots for DDEs and DDDs in Ireland versus England for 2014 and 2022 are available in Supplementary Figure S1.
For the subgroup analysis of CCGs in England in the top 33 centiles for deprivation, dispensing rates of benzodiazepines were lower than the general population in 2014 (238 vs 288), and slightly higher in 2022 (216 vs 210). For all sedatives the rate of dispensing increased 31% in the most deprived CCGs, versus a stable trend in the general population (see Supplementary Table S8).
Pattern of utilization
In Ireland, the rate of benzodiazepine and z-drug dispensings varied substantially between age groups. The 75+ years age group persistently had the highest rate of dispensings, however it also saw the largest relative decrease between 2014 and 2022, decreasing from 3736 to 2826 per 1000 population. All other age groups also saw rates decrease between 2014 and 2022, however for the youngest age groups (<24) the differences were small, and for younger adults (25–34, 35–44) there was no significant linear change in trend. The rate of dispensings for benzodiazepines and z-drugs was persistently higher for females compared to males, decreasing slightly from 1865 in 2014 to 1730 in 2022 for females and 1133 in 2014 to 1100 in 2022 for males (see Fig. 3 and Supplementary Data in the Data availability statement).
Figure 3.
Rate of benzodiazepine and z-drug dispensings per 1000 GMS population by age group and sex. Note: rates for age groups below 16 years are all <25 dispensings per 1000.
Like the rate dispensings, the rate of DDEs per 1000 GMS population was highest in the 75+ age group, followed by the 55–64, 65–69, and 70–74 age groups, and higher for females compared to males (see Supplementary Data in the Data availability statement).
The prevalence of benzodiazepines and z-drugs use in the GMS population decreased slightly during the study period, albeit not significantly based on a linear test for trend, with 18.2% of eligible individuals receiving at least one dispensing in 2014, compared to 16.4% in 2022. The prevalence of benzodiazepine and z-drug initiations, based on a medication-free interval of 90 or 180 days, was stable at 7.7–9.0% and 6.3–7.2%, respectively, while discontinuations, based on a medication-free interval of 90 and 180 days, ranged between 4.9–5.6% and 3.2–3.6%, respectively. Similarly, chronic use of benzodiazepines and z-drugs for more than 30 days was stable at 10.6–11.7%, while 8.0–9.6% received dispensings for more than 90 days. Prevalence of initiations, discontinuations, and chronic use was not calculated for 2014, as results would be skewed due to the lack of access to 2013 data. No statistically significant trends were identified (see Table 2 and Supplementary Table S9). Pattern of use by age group and sex generally followed age/sex volumes of use presented above (see Supplementary Figure S2). Full results for prevalence, initiations, discontinuations, chronic use, and high-risk dispensings by age group and sex are available from Zenodo 10.5281/zenodo.15747675.
Table 2.
Prevalence (%) of any dispensings, initiations (90/180 days), discontinuations (90/180 days), and chronic use (30/90 days) of benzodiazepines and z-drugs in the GMS population.
| 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | |
|---|---|---|---|---|---|---|---|---|---|
| Prevalence of any dispensings | 18.2 | 17.8 | 18.3 | 18.4 | 18.3 | 18.1 | 16.5 | 16.9 | 16.4a |
| Prevalence of initiations after 90 days | — | 8.2 | 8.9 | 9.0 | 8.9 | 8.8 | 7.7 | 7.8 | 7.7a |
| Prevalence of initiations after 180 days | — | 6.3 | 7.1 | 7.2 | 7.1 | 7.1 | 6.3 | 6.4 | 6.4a |
| Prevalence of discontinuations after 90 days | — | 5.2 | 5.5 | 5.6 | 5.5 | 5.6 | 4.9 | 5.0 | 5.0a |
| Prevalence of discontinuations after 180 days | — | 3.2 | 3.5 | 3.5 | 3.5 | 3.6 | 3.3 | 3.3 | 3.3 |
| Prevalence of chronic use >30 days | — | 11.2 | 11.5 | 11.6 | 11.7 | 11.6 | 11.1 | 11.2 | 10.6a |
| Prevalence of chronic use >90 days | — | 9.1 | 9.3 | 9.5 | 9.6 | 9.6 | 9.2 | 9.3 | 8.9a |
aTest for trend significant at 0.05 based on the Cochran-Armitage test.
The prevalence of high-risk dispensings decreased during the study period (see Table 3). The prevalence of benzodiazepine or z-drug dispensings with two or more other sedating or anticholinergic drugs in people aged ≥65 years decreased from 18.5% in 2014 to 13.1% in 2022. The prevalence of benzodiazepine or z-drug dispensing at a dose equivalent to >40 mg diazepam per day remained steady at 0.9–1.0% during the study period (see Table 3).
Table 3.
Prevalence (%) of high-risk dispensings in the GMS population.
| 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | |
|---|---|---|---|---|---|---|---|---|---|
| Prevalence of benzodiazepine or z-drug dispensings with two or more other sedating or anticholinergic drugs in people aged ≥65 years | |||||||||
| Benzodiazepines and z-drugs | 18.5 | 17.5 | 17.1 | 16.1 | 15.6 | 15.0 | 14.1 | 13.6 | 13.1 |
| Long-acting benzodiazepines | 5.4 | 5.1 | 5.0 | 4.7 | 4.5 | 4.3 | 4.1 | 3.9 | 3.7 |
| Proportion of benzodiazepine or z-drug dispensings with two or more other sedating or anticholinergic drugs out of all benzodiazepine or z-drug dispensings in people aged ≥65 years | |||||||||
| Benzodiazepines and z-drugs | 53.8 | 54.1 | 53.1 | 52.2 | 51.4 | 50.8 | 50.7 | 49.7 | 49.0 |
| Long-acting benzodiazepines | 54.3 | 54.0 | 53.1 | 52.1 | 50.9 | 50.7 | 51.5 | 50.1 | 49.2 |
| Prevalence of benzodiazepine or z-drug dispensing at a dose equivalent to >40 mg diazepam per day | |||||||||
| Benzodiazepines and z-drugs | 1.04 | 1.02 | 1.03 | 1.04 | 1.02 | 1.01 | 0.94 | 0.92 | 0.88 |
| Long-acting benzodiazepines | 0.66 | 0.65 | 0.66 | 0.66 | 0.66 | 0.65 | 0.60 | 0.59 | 0.57 |
| Proportion of benzodiazepines or z-drugs dispensed at a dose equivalent to >40 mg diazepam per day out of all benzodiazepine or z-drug dispensings | |||||||||
| Benzodiazepines and z-drugs | 5.7 | 5.7 | 5.6 | 5.6 | 5.6 | 5.6 | 5.7 | 5.4 | 5.4 |
| Long-acting benzodiazepines | 9.2 | 9.2 | 9.1 | 9.2 | 9.1 | 9.1 | 9.3 | 8.9 | 9.0 |
Discussion
Since 2014, there was an overall rise in sedative dispensings in primary care in Ireland, but a decrease in benzodiazepine and z-drug dispensings. Our research indicates that benzodiazepine use increased up to 2019, but has fallen below 2015 levels in recent years, while z-drugs have continued to increase with the exception of a small drop in 2022. This is in line with literature from the UK, where the trend in benzodiazepine and z-drug use similarly has decreased in recent years, consistent with our England analysis [6]. In contrast to our findings, recent research from the USA has shown that the onset of the COVID-19 pandemic was associated with increased rates in z-drug prescribing, along with an increase in benzodiazepine prescriptions in women [25].
In Ireland the first formal guidance on appropriate prescribing of these medications was issued in 2018 [26], aligning with long-standing guidance in other jurisdictions, such as the UK [27]. It recommended restriction of benzodiazepine prescribing to the shortest possible duration and for a maximum of 2–4 weeks for anxiety treatment, and avoidance of long-acting benzodiazepines for insomnia. Selective serotonin reuptake inhibitors (SSRIs) were recommended as first-line pharmacological treatment for anxiety, and they also recommend avoiding initiation of sedating antidepressants, systemic antihistamines, or atypical antipsychotics when discontinuing long-term benzodiazepine treatment. Despite the decrease in benzodiazepine and z-drug dispensings, overall sedative dispensings increased, driven by an increase in antidepressants, first-generation antihistamines, and atypical antipsychotics with sedating properties. Previous research from the UK shows an increase in prescribing of these medications [7, 9], in line with our findings. Without information on the clinical characteristics of patients it is unclear what the implications for clinical practice are, and if this is a positive change. While reducing benzodiazepine and Z-drug use may be a positive step, it is important to remain vigilant regarding the adverse effects of alternative sedatives, e.g. hang-over effects and metabolic complications [28–30]. There is no clear evidence or guidance on the comparative effectiveness and safety of benzodiazepines and z-drugs versus these alternative sedatives, and further research should examine this.
Our analysis of benzodiazepine and z-drug dispensing rates by age and sex show higher rates for women and older people, a well-established pattern internationally [31, 32]. One contributor to the high prevalence in older age groups could be use in end-of-life care. Rates of dispensings decreased in all older age groups, with the largest decrease during the study in the ≥75 years group. OECD health indicators for prescribing in primary care relating long-acting and long-term benzodiazepine use among those aged ≥65 have both decreased in most OECD countries between 2009 and 2019 [33]. The prevalence of benzodiazepine and z-drug initiations, discontinuations, and chronic use remained steady over our study, while there was a slight decrease in the percentage of individuals with any dispensings. In the context of the decreasing rate of dispensings, this may indicate that a reduction in the frequency of dispensings, rather than fewer people on these medications, is the main driver. The larger relative reduction in DDEs compared to dispensings suggest an additional trend towards prescribing lower quantity, strength, and/or potency of benzodiazepines and z-drugs with the implications that patients may be less at risk of dependence/tolerance.
England had a substantially lower rate of dispensings for all drugs examined. One explanation may be the difference between the two populations, as the English data included all individuals registered with a GP practice, compared to the Irish data which was restricted to a means-tested population, aligning with evidence of higher benzodiazepine and z-drug prescribing rates in GP practices in deprived areas [34]. Our subgroup analysis to account for differences in deprivation (albeit not in age) between Ireland and England showed that dispensing rates were not higher for most drugs in this subgroup during the study. Despite similar guidance and controlled drug status for benzodiazepine and z-drug in both countries, Ireland has for a long time had higher rates of prescribing. As highlighted above, Ireland has one of the highest rates of chronic benzodiazepine and z-drug use in people aged ≥65 among OECD countries [35], and previous studies have found higher rates of potentially inappropriate prescribing in the GMS population compared to the NHS population of Northern Ireland [36]. Hospitalization and unintended long-term continuation of hospital-initiated benzodiazepines and z-drugs also likely contribute in Ireland, potentially driven by a lack of shared care record to support transitions of care [37]. This more restrictive stance on prescribing of these drugs in the UK is also mirrored by a much more restricted set of drugs available to be prescribed in the NHS, with several drugs not dispensed at all during the study period. There are also differences in access to recommended non-pharmacological therapies, including cognitive behavioural therapy (CBT) and applied relaxation and mindfulness [38], and in waiting time for mental health specialists between Ireland and UK, both which may impact on prescribing [39].
The main strength of our study is the data used, as its primary use for pharmacy reimbursement ensures full coverage for the GMS population and high accuracy and completeness. Dispensing-level data allowed analysis of individual-level changes (e.g. initiations/discontinuations). An important aspect of drug utilization research is the access to data sources. Restrictions to data sources, including charges and lengthy application processes, may impede timely analysis and should be addressed by data governance policies, with the availability of open data when appropriate being crucial.
Our study is limited in only including data on the means-tested GMS eligible population, not directly comparable to the NHS population in England, and although our subgroup analysis addressed deprivation, we could not address the potential age disparity between the populations. Secondly, dispensings do not include indications for the prescription, nor do we know the medical history of individuals. Our analysis of other sedative agents should therefore be interpreted with caution as a proportion of these dispensings will be for indications unrelated to sedation. Similarly, we lack information on whether prescribing was specialist or GP-initiated, as all GMS patients have their secondary care prescriptions reissued by their GP, and interventions aiming to reduce sedative prescribing need to account for this. We also do not have data on hospitalization, and so cannot exclude person-time where individuals are not at risk of being dispensed medication in primary care. Due to lack of available public data, we were unable to identify drug shortages during the study period which may have impacted trends, however analysis on an annual basis may have mitigated the influence of short-term shortages. Finally, some GMS eligible individuals included in the study may have lost eligibility due to changes in circumstances and/or income or died during the study period. As we do not have access to eligibility start or end dates, or mortality data, we are not able to account for these factors when analysing patterns of use (e.g. initiations, discontinuations).
The findings of this study suggest that benzodiazepine and z-drug dispensing is decreasing in Ireland, including high-risk dispensings and dispensings to older age groups. However, in comparison to England, dispensing rates remain high. Qualitative work may be beneficial to identify and contextualize sedative use in primary and secondary specialist care, and the impact for patients of access to specialized mental health care and non-pharmacological interventions. Overall, this study indicates a need for enhanced availability of non-pharmacological services and interventions to address the two main indications of benzodiazepine and z-drug prescribing: anxiety and insomnia, as well as improved education and deprescribing support for healthcare professionals.
Supplementary Material
Contributor Information
Molly Mattsson, School of Pharmacy and Biomolecular Sciences, RCSI University of Medicine and Health Sciences, 111 St Stephen's Green, Dublin 2, D02 VN51, Ireland.
Ahmed Hassan Ali, School of Pharmacy and Biomolecular Sciences, RCSI University of Medicine and Health Sciences, 111 St Stephen's Green, Dublin 2, D02 VN51, Ireland.
Fiona Boland, Data Science Centre, RCSI University of Medicine and Health Sciences, 123 St Stephen's Green, Dublin 2, D02 YN77, Ireland.
Michelle Flood, School of Pharmacy and Biomolecular Sciences, RCSI University of Medicine and Health Sciences, 111 St Stephen's Green, Dublin 2, D02 VN51, Ireland.
Ciara Kirke, National Medication Safety Programme, HSE National Quality and Patient Safety Directorate, Dr Steevens' Hospital, Steevens' Lane, Dublin 8, D08 W2A8, Ireland.
Emma Wallace, Department of General Practice, Western Gateway Building, University College Cork, Cork, T12 XF62, Ireland.
Derek Corrigan, FutureNeuro, RCSI University of Medicine and Health Sciences, 123 St Stephen's Green, Dublin 2, D02 VN77, Ireland.
Mary E Walsh, School of Pharmacy and Biomolecular Sciences, RCSI University of Medicine and Health Sciences, 111 St Stephen's Green, Dublin 2, D02 VN51, Ireland; School of Public Health, Physiotherapy and Sports Science, Woodview House, University College Dublin, Belfield, Dublin 4, D04 C1P1, Ireland.
Tom Fahey, Department of General Practice, RCSI University of Medicine and Health Sciences, 123 St Stephen's Green, Dublin 2, D02 VN77, Ireland.
Frank Moriarty, School of Pharmacy and Biomolecular Sciences, RCSI University of Medicine and Health Sciences, 111 St Stephen's Green, Dublin 2, D02 VN51, Ireland.
Supplementary data
Supplementary data is available at Family Practice online.
Conflict of interest statement
The authors have no relevant interests to disclose.
Funding
This study is funded by the Health Research Board in Ireland (HRB) through the Secondary Data Analysis Projects scheme (CDRx project, grant number SDAP-2019-023). The funder had no role in in study design; in the collection, analysis, and interpretation of data; in the writing of this article; or in the decision to submit this article for publication. E.W. is funded by an HRB Emerging Clinician Scientist Award (grant number: ECSA/2020/002). M.E.W. is funded by an HRB Applying Research into Policy and Practice Award (ARPP/2020/004).
Data availability
The datasets analysed during the current study are not publicly available as this was not covered by the Data Exchange Agreement between RCSI and HSE-PCRS. Data can be requested from HSE-PCRS via an information request as detailed at https://www.hse.ie/eng/staff/pcrs/pcrs-publications/. Full results for dispensing rate, prevalence, initiations, discontinuations, chronic use, and high-risk dispensings by age group and sex are available from Zenodo (doi:10.5281/zenodo.15747675).
References
- 1. Edinoff AN, Nix CA, Hollier J et al. Benzodiazepines: uses, dangers, and clinical considerations. Neurol Int 2021;13:594–607. 10.3390/neurolint13040059 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. British National Formulary . Hypnotics and Anxiolytics. Benzodiazepine Guidance. https://benzo.org.uk/BNF.htm (2013, July 2024, date last accessed).
- 3. Ma T-T, Wang Z, Qin X et al. Global trends in the consumption of benzodiazepines and Z-drugs in 67 countries and regions from 2008 to 2018: a sales data analysis. Sleep 2023;46:zsad124. 10.1093/sleep/zsad124 [DOI] [PubMed] [Google Scholar]
- 4. Cadogan CA, Ryan C, Cahir C et al. Benzodiazepine and Z-drug prescribing in Ireland: analysis of national prescribing trends from 2005 to 2015. Br J Clin Pharmacol 2018;84:1354–63. 10.1111/bcp.13570 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Cadogan CA, Bradley CP, Bennett K. Impact of changes in controlled drugs legislation on benzodiazepine receptor agonist prescribing in Ireland: a repeated cross-sectional study. Eur J Clin Pharmacol 2021;77:903–12. 10.1007/s00228-020-03063-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Optimising Personalised Care for Adults Prescribed Medicines Associated With Dependence or Withdrawal Symptoms: Framework for Action for Integrated Care Boards (ICBs) and Primary Care. NHS England, 2023.
- 7. Bogowicz P, Curtis HJ, Walker AJ et al. Trends and variation in antidepressant prescribing in English primary care: a retrospective longitudinal study. BJGP Open 2021;5:BJGPO.2021.0020. 10.3399/BJGPO.2021.0020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Sateia MJ, Buysse DJ, Krystal AD et al. Clinical practice guideline for the pharmacologic treatment of chronic insomnia in adults: an American academy of sleep medicine clinical practice guideline. J Clin Sleep Med 2017;13:307–49. 10.5664/jcsm.6470 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Roberts R, Neasham A, Lambrinudi C et al. A quantitative analysis of antipsychotic prescribing trends for the treatment of schizophrenia in England and Wales. JRSM Open 2018;9:2054270418758570. 10.1177/2054270418758570 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Thompson W, Kim JD, Carney G et al. Changes in sedative use in British Columbia, Canada from 2012 to 2021: a drug utilization study. J Gen Intern Med 2024;40:980–3. 10.1007/s11606-024-09065-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Organisation for Economic Co-operation and Development (OECD) . Health Care Quality Indicators (Edition 2017). https://www.oecd-ilibrary.org/content/data/e3667ffb-en (July 2024, date last accessed).
- 12. OECD. Health at a Glance 2023: OECD Indicators. Paris: OECD Publishing, 2023. [Google Scholar]
- 13. Vitiello B. An international perspective on pediatric psychopharmacology. Int Rev Psychiatry 2008;20:121–6. 10.1080/09540260801887710 [DOI] [PubMed] [Google Scholar]
- 14. Mattsson M, Boland F, Kirke C et al. Evaluation of policies and practices to support safe and appropriate analgesic and sedative prescribing: the CDRx (controlled drug prescribing) protocol. Res Social Adm Pharm 2022;18:3588–95. 10.1016/j.sapharm.2022.03.004 [DOI] [PubMed] [Google Scholar]
- 15. Sinnott S-J, Bennett K, Cahir C. Pharmacoepidemiology resources in Ireland-an introduction to pharmacy claims data. Eur J Clin Pharmacol 2017;73:1449–55. 10.1007/s00228-017-2310-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Mattsson M, Flood M, Wallace E et al. Eligibility rates and representativeness of the general medical services scheme population in Ireland 2016–2021: a methodological report [version 2; peer review: 2 approved]. HRB Open Res 2022;5:67. 10.12688/hrbopenres.13622.2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Ashton H. Benzodiazepines: How They Work and How to Withdraw (The Ashton Manual). Newcastle upon Tyne: Institute of Neuroscience, Newcastle University, Revised August 2002. https://www.benzo.org.uk/manual/
- 18. Primary Care Reimbursement Service (PCRS) . Eligibility Reports. HSE. https://www.sspcrs.ie/portal/annual-reporting/report/eligibility (July 2024, date last accessed).
- 19. NHS Digital . Patients Registered at a GP Practice. NHS. https://digital.nhs.uk/data-and-information/publications/statistical/patients-registered-at-a-gp-practice (July 2024, date last accessed).
- 20. WHO Collaborating Centre for Drug Statistics Methodology . DDD—Definition and General Considerations. https://www.whocc.no/ddd/definition_and_general_considera/ (July 2024, date last accessed).
- 21. Benzo.org.uk . Benzodiazepine Equivalency Table. https://www.benzo.org.uk/bzequiv.htm (July 2024, date last accessed).
- 22. Scottish Government Polypharmacy Model of Care Group . Polypharmacy Guidance, Realistic Prescribing. 3rd edn. Scottish Government, 2018.
- 23. Boustani M, Campbell N, Munger S et al. Impact of anticholinergics on the aging brain: a review and practical application. Aging Health 2008;4:311–20. 10.2217/1745509X.4.3.311 [DOI] [Google Scholar]
- 24. UK Goverment . English Indices of Deprivation 2019. https://www.gov.uk/government/statistics/english-indices-of-deprivation-2019 (July 2024, date last accessed).
- 25. Milani SA, Raji MA, Chen L et al. Trends in the use of benzodiazepines, Z-hypnotics, and serotonergic drugs among US women and men before and during the COVID-19 pandemic. JAMA Netw Open 2021;4:e2131012. 10.1001/jamanetworkopen.2021.31012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Medicines Management Programme . Guidance on Appropriate Prescribing of Benzodiazepines and z-Drugs (BZRA) for the Treatment of Anxiety and Insomnia. Dublin: HSE, 2021.
- 27. British National Formulary . Hypnotics and Anxiolytics—Treatment Summaries. https://bnf.nice.org.uk/treatment-summaries/hypnotics-and-anxiolytics/ (26 September 2024, date last accessed).
- 28. NHS . Antihistamines. https://www.nhs.uk/medicines/antihistamines/ (July 2024, date last accessed).
- 29. NHS . Side Effects—Antidepressants. https://www.nhs.uk/mental-health/talking-therapies-medicine-treatments/medicines-and-psychiatry/antidepressants/side-effects/ (July 2024, date last accessed).
- 30. HSE . Treatment—Psychosis. https://www2.hse.ie/conditions/psychosis/treatment// (July 2024, date last accessed).
- 31. Agarwal SD, Landon BE. Patterns in outpatient benzodiazepine prescribing in the United States. JAMA Netw Open 2019;2:e187399. 10.1001/jamanetworkopen.2018.7399 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Olfson M, King M, Schoenbaum M. Benzodiazepine use in the United States. JAMA Psychiatry 2015;72:136–42. 10.1001/jamapsychiatry.2014.1763 [DOI] [PubMed] [Google Scholar]
- 33. Organisation for Economic Co-operation and Development (OECD) . Safe long-term care. In: Health at a Glance 2021: OECD Indicators. Paris: OECD Publishing, 2021, 258–9. [Google Scholar]
- 34. Soyombo S, Stanbrook R, Aujla H et al. Socioeconomic status and benzodiazepine and Z-drug prescribing: a cross-sectional study of practice-level data in England. Fam Pract 2020;37:194–9. 10.1093/fampra/cmz054 [DOI] [PubMed] [Google Scholar]
- 35. Organisation for Economic Co-operationd and Development (OECD) . Safe long-term care. In: Health at a Glance 2023: OECD Indicators. Paris: OECD Publishing, 2023, 218–19. [Google Scholar]
- 36. Cooper JA, Moriarty F, Ryan C et al. Potentially inappropriate prescribing in two populations with differing socio-economic profiles: a cross-sectional database study using the PROMPT criteria. Eur J Clin Pharmacol 2016;72:583–91. 10.1007/s00228-015-2003-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Coll S, Walsh ME, Fahey T et al. Hospital initiation of benzodiazepines and Z-drugs in older adults and discontinuation in primary care. Res Social Adm Pharm 2022;18:2670–4. 10.1016/j.sapharm.2021.06.001 [DOI] [PubMed] [Google Scholar]
- 38. National Institute for Health and Care Excellence (NICE) . Generalised Anxiety Disorder and Panic Disorder in Adults: Management. London: National Institute for Health and Care Excellence (NICE), 2019. [PubMed] [Google Scholar]
- 39. European Union . Mental Health. https://europa.eu/eurobarometer/surveys/detail/3032 (July 2024, date last accessed).
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets analysed during the current study are not publicly available as this was not covered by the Data Exchange Agreement between RCSI and HSE-PCRS. Data can be requested from HSE-PCRS via an information request as detailed at https://www.hse.ie/eng/staff/pcrs/pcrs-publications/. Full results for dispensing rate, prevalence, initiations, discontinuations, chronic use, and high-risk dispensings by age group and sex are available from Zenodo (doi:10.5281/zenodo.15747675).