Discussion
National purchasing data for clinics offering specialist mental health services and interdisciplinary specialised SUD treatment show a significant increase in the use of anxiolytics. There has been a decline in the use of antidepressants and mood stabilisers. The antipsychotics group has increased somewhat, while the use of depot injections has grown by 79 %.
Diazepam is the most used of the anxiolytics and has shown a marked increase in recent years. The reduction in the number of beds in mental health services for adults and interdisciplinary specialised SUD treatment, combined with more outpatient clinic consultations and an increase in the number of people involuntarily admitted, may have led to a change in the composition of patients. This could mean that those currently receiving treatment as inpatients are more seriously ill than previously and may, therefore, have a greater need for sedatives (13, 20). The increase in diazepam may also be due to changes in the guidelines for opioid maintenance treatment (21), as benzodiazepines used in tapering or maintenance treatment implemented by the hospital are now paid for by the health trust. Use of prescription benzodiazepines in Norway and the Nordic region fell during the reference period (3, 7). Consequently, despite an increase in use in specialist mental health services and interdisciplinary specialised SUD treatment, it is unlikely that more patients will be discharged from the hospitals with a prescription for benzodiazepines.
A reduction has been observed in the antidepressant group while use has remained more stable in the primary health service during the same period, albeit with an increase during the pandemic 2020–22 (3, 10, 22). A decrease in the use of escitalopram, which is mostly used on prescription, and a rise in sertraline has been observed in hospitals and in the primary health service. In relation to mirtazapine, venlafaxine and fluoxetine, there has been an increase in prescriptions but a reduction in hospital use (22). A possible change in the composition of hospital patients may explain the differing consumption patterns. It cannot be ruled out that therapists in hospitals may have a more critical attitude towards the efficacy and side effects of selective serotonin reuptake inhibitors (23, 24).
The reduction in the use of mood stabilisers may be due to several factors. One possible explanation could be the decrease in admissions of patients with bipolar disorder in recent years (25, 26) or the increased use of antipsychotics as part of their treatment. Of the mood stabilisers, only lithium is used exclusively for the treatment of affective disorders. The proportion of patients using lithium on prescription per 1000 people per annum decreased from 1.6 in 2012 to 1.4 in 2021, and the age distribution changed, with a larger percentage of older people among users (3, 22). During the period of this study, lithium use decreased by 16 %. Reduced use of lithium for bipolar disorders may represent a deviation from the guidelines. A decline in lithium use has been registered in many countries and is a cause for concern as lithium is considered the best medication for bipolar disorder with regard to treating mania and preventing illness and suicide (27–30).
The proportion of antipsychotics administered in the form of depot injections increased in the period 2012–23. This may reflect the increased focus on medication adherence to prevent readmission (31), but also changes to, and the interpretation of, the legislation on the use of coercion from 2017 (32). From 2015 to 2022, there was an increase in involuntary admissions, in the proportion of patients medicated without consent, and in the number of outpatients involuntarily receiving mental health care (33). The treatment of patients in the latter group with depot injections of antipsychotics is paid for by the health trust, which may also have contributed to the observed increase.
The figures show a decline in the use of peroral antipsychotics, while antipsychotics on prescription increased in the years 2017–21 (3). The national clinical guidelines for pharmacological treatment of psychotic disorders were published recently. They recommend that second-generation antipsychotics with a low metabolic impact should be used in relation to the first episode of psychosis (34). The Danish guidelines specify aripiprazole as a first-line medication (35), while there is no specified preferred medication in the Norwegian guidelines. In a survey conducted in the Early Intervention in Psychosis Advisory Unit at Oslo University Hospital in 2022, 43 % of the patients used olanzapine and 29 % used aripiprazole (36). Our study shows that olanzapine and quetiapine are used more than aripiprazole, but that the use of depot injections of aripiprazole is increasing. A decline in clozapine use (9 %) was observed during the reference period. A study based on dispensed prescription data identified major regional variations in the prescription of clozapine in Norway. The study suggested that differences in the degree of implementation of the national guidelines were a possible cause (11). This may also be a contributory factor in the change observed in our figures. There has been a focus on the off-label use of quetiapine as a hypnotic and sedative (37, 38). Use of this medication on a 'white' prescription (usually not reimbursed) has increased in recent years (39), and our figures reveal the increased use of low-dose tablets.
Large-scale events, such as the COVID-19 pandemic, may have impacted medication use. Prescriptions of antidepressants and low-dose quetiapine on prescription increased in Norway during the pandemic (22), as well as anxiolytics and hypnotics in Scandinavia (10). On the basis of this study alone, we cannot determine whether the pandemic affected the use of psychotropic medication in hospitals.
The strength of this study is the complete set of data for purchases of anxiolytics, antidepressants, mood stabilisers and antipsychotics in hospitals offering mental health services and interdisciplinary specialised SUD treatment over a 12-year period. Other available data, such as the figures from the Norwegian Drug Wholesale Statistics and the Norwegian Prescribed Drug Registry (3), reveal little about medication use in hospitals.
A limitation of the study is that the available data for medication use in Norwegian hospitals are based on purchasing figures and not, therefore, actual use by patients. Drugs purchased for a department can be administered to patients, disposed of or stored. The difference between purchase and use is primarily linked to periodisation. If longer time periods are used, from six months up to a year, purchased DDD will provide estimates that correspond to use (40). Only three European countries (Belgium, Italy and Portugal) have national databases of medication use that record what has actually been administered to patients in hospitals (41).
A range of factors, including activities, patient composition, changes to treatment guidelines, opinion leaders, legislation and purchasing agreements will have a potential impact on medication use in hospitals. An increase in the number of DDD indicates that either more patients have received the medication or that the doses have increased. We have used DDD/100 bed-days, as recommended by WHO, as an indicator to correct for activity levels in hospitals (42). Other indicators for medication use, which use bed-days, patient composition and number of inpatients, may provide a more accurate measure (43).
This study does not provide answers regarding causal relations or the effects of the observed changes; however, with a time series spanning many years, changes in the figures can be assessed in relation to concurrent events that could potentially have impacted medication use in hospitals. The study serves as a basis for hypotheses for further research. Future studies should examine the correlation between medication use and patient data, as well as assess practice in relation to guidelines. It will be relevant to compare medication use between different units within the mental health service and interdisciplinary specialised SUD treatment with the Helseatlas [Health Atlas] study, which reports undesirable variations in the treatment provision (44). Regular reporting of psychotropic medication use on a national, regional and local level will be able to provide an overview that helps to optimise medication use in hospitals.