This pharmacoepidemiology retrospective study reviewed prescribing information and population characteristics from 2015 to 2021. The study utilised a stratified random sampling approach, selecting data from 10 major hospitals in Zhejiang, Jiangsu, and Shanghai. This method ensured a representative sample across different types of hospitals, including specialised psychiatric institutions and general hospitals, thereby providing a comprehensive view of benzodiazepine and Z-drug prescription trends.

This study focused on the use of BZDs and non-BZDs in major psychiatric hospitals in economically developed areas of eastern China from 2015 to 2021, as well as prescriptions for BZDs and non-BZDs in outpatient and emergency departments from 2017 to 2021.

The data of BZDs and non-BZDs in the hospital information system (HIS) refer to actual clinical usage and do not include data that are not actually used. Prescriptions for BZDs and non-BZDs used in outpatient and emergency prescriptions, with complete information and excluding unqualified returned prescriptions, were eligible for use in this study.

Defined daily dose (DDD) value: This refers to the average daily dose of medication. The DDD value is the total mean daily dose of the drug. The DDD values of each drug were queried and registered according to the DDD values published on the WHO’s website.

Prescription: This refers to an outpatient or emergency prescription issued by a registered physician.

Second-class mental prescriptions: In China, psychotropic drugs are divided into Class I and Class II psychotropic drugs. Class I psychotropic drugs are prescribed as special prescriptions for narcotic drugs, whereas Class II psychotropic drugs are prescribed with special prescriptions for Class II psychotropic drugs.

Drugs: These refer to BZDs and non-BZDs (class Z). The drugs include primarily BZDs, such as lorazepam tablets, oxazepam, estazolam tablets, alprazolam, clonazepam, diazepam needles, diazepam and nitrazepam tablets, while non-BZDs (class Z) include zopiclone tablets.

Diagnosis (name of disease): The names of mental diseases are classified according to the broad categories of related mental diseases denoted in the Treatment Guide: Psychosis Volume [7th edition] compiled by Australian Treatment Guidelines Limited, 2015. They include insomnia, sleep and anxiety disorders, and depression. Non-mental illness names are classified as ‘other diseases’.

The DDD values of each drug were queried and registered according to the DDD values published on the WHO’s website.

Annual data were calculated and tabulated according to the drug name, specification, quantity, total measurement, and DDD number. Drug names were classified individually according to the prevailing chemical name of the drug.

Prescriptions: These were categorised by year, gender, age group, and drug name.

Drug quantity: The quantity data of BZDs and Z-drugs that were used in relevant hospitals from 2015 to 2021 were collected, including the year, drug name, specification, unit price, and quantity. The data for this study were collected as part of a comprehensive research project on the prescription patterns of BZDs and Z-drugs in China. The study involved a total of 10 hospitals located in three provinces: Zhejiang, Jiangsu, and Shanghai. This included specialised psychiatric institutions and general hospitals with significant outpatient and emergency services.

Prescribed: The information system was based on the individual HIS of each hospital for drugs. Data were collected from 2017 to 2021 about the prescription and use of BZD drugs. The prescriptions contained no privacy information, as the names were displayed as ‘surname * *’. Patient data included their gender, age, diagnosis of diseases (name), drug name, quantity, usage, dosage, and prescription date. The collected data were checked for privacy and compliance.

The quantity of drugs refers to the actual consumption of said specific drug throughout the hospital as a whole.

Number of prescriptions: This was 8%–10% of the total sample collected for all 10 hospitals.

To ensure rigorous analysis and interpretation of the data collected, we employed a comprehensive statistical approach using the SPSS Statistics 21 software (IBM, Armonk, NY, USA). This section outlines the detailed statistical methods used in our study.

Data Preparation and Cleaning: Initially, data underwent preprocessing to ensure accuracy and consistency. This involved checking for outliers, handling missing values through imputation where appropriate, and verifying the data integrity against entry errors.

Descriptive Statistics: We computed descriptive statistics to summarise the demographic and prescription data. This included calculating the means, standard deviations and ranges for continuous variables, as well as frequencies and percentages for categorical variables.

Study population: Male (38.2%) and female (61.8%) patients aged 14–80 years old who required the use of BZDs and non-BZDs.

Age group: The groups were $\le$17, 18–34, 35–44, 45–64, 65–75, and $>$75 years of age.

There was a total hospital consumption of 8 different BZD types and 1 type of Z-drug in the 6 years from 2015 to 2021.

A total of 33,569 valid BZD and Z-drug outpatient and emergency prescriptions were collected from 2017 to 2021.

Inclusion Criteria:

Patients who received benzodiazepine or Z-drug prescriptions between 2015 and 2021.

Data from both specialised psychiatric institutions and general hospitals in Zhejiang, Jiangsu, and Shanghai were collected.

Exclusion Criteria:

Incomplete or missing prescription data.

Patients who were not prescribed BZDs or Z-drugs.

Inpatient prescriptions were excluded.

The total DDD values of the drugs that were used were calculated according to the general name of each drug, based on the DDD value of said drug, published on the WHO website. The annual DDDs of the total drugs showed an upward trend. Zolpiclone is the primary Z-drug available in the regions where our study was conducted. Eszopiclone and Zolpidem are less commonly prescribed due to regional pharmaceutical regulations and availability. According to the annual DDD value of individual drugs, lorazepam and zopiclone tablets showed a rapid rising trend; oxazepam and estazolam tablets showed a slow rising trend; and alprazolam, clonazepam, and diazepam showed a slight rise, followed by a steady decline. Diazepam and nitrazepam tablets tended to be reduced to non-use. See Table 1 and Fig. 1.

Fig. 1.

Trends in benzodiazepines and non-benzodiazepines drug use data (DDDs), 2015–2021.

There is growing concern about the potential for misuse/abuse, dependence, and addiction to BZD drugs, which are increasingly being prescribed in developed countries. Benzodiazepines were used in 16.6 million of 133.3 million outpatient visits, and in 1.9 million of 18.1 million emergency department visits from 2001 to 2010 in the United States [13]. In 2018, more than 47 million BZD prescriptions, and more than 14 million Z-drugs prescriptions were written in the United States alone [14]. Long-term BZD prescriptions in Australia increased from 4.4% in 2011 to 5.8% in 2015, with a relatively stable annual growth rate of +2.5% between 2015 and 2018 [15].

There were more than 50 million health care consultations between 2011 and 2018 [16]. The prescriptions of BZDs or Z-drugs in Ireland increased from 11.9% in 2005 to 15.3% in 2015 [17]. From January 2017 to December 2020, 1.442 billion BZDs were dispensed across Canada [18].

This study found that the annual DDDs of BZDs and Z-drugs showed an upward trend. Furthermore, the proportion of second-class psychiatric prescriptions in outpatient and emergency prescriptions rose slowly from 29.78% in 2017 to 33.10% in 2021, and the proportion of second-class psychiatric prescriptions and the number of second-class psychiatric prescriptions showed a slowly increasing trend year by year.

Expanding on these initial observations, BZDs were used by 16% of elderly patients in Australia’s 65-year-old population [19]. Older patients were six times more likely than younger patients to be on long-term BZD prescriptions. American adults aged 50 years and older were more likely than younger adults to use BZDs more frequently than prescribed [13], and 13.75% of elderly patients in the general hospitals of large cities in Germany were dependent on BZDs and Z-drugs [14]. A study on the correlation between sleep status and frailty in Chinese adults aged 30–79 years found that insomnia disorder was associated with an increased proportion of frailty [9]. This study showed that the prescriptions in different age groups also showed an increasing trend year by year, and the increasing trend was larger in 50–64 and 65–75-year-olds than in other age groups.

A total of 38% of women aged $>$80 years of age in France used BZDs [20]. The 10-year trends in the use of sleeping pills in Switzerland found that, overall, women, the elderly, and people with a history of anxiety and depression were more likely to take sleeping pills. Long-term BZD prescriptions are more common and last longer in older women in Australia [15]. Women in the United States are about twice as likely as men to be prescribed BZDs, according to a press release from the National Institutes of Health on 17 December 2014. The percentage of women using BZDs and Z-drugs in this study was 61.8%. Among them, 63.3% were women between 50 and 75 years old. The number of BZD and non-BZD prescription drug users is much higher in women than in men, based on gender, economic factors, and social and family roles.

Considering the above demographic trends, BZDs are the main therapeutic drugs for patients with anxiety, insomnia, and epilepsy, and also serve as important adjunctive drugs for patients with specific physical and severe mental diseases to improve insomnia, anxiety, agitation, and other symptoms. In this study, the main disease factors for using BZDs and Z-drugs were anxiety disorder (24.3%), insomnia (41.30%), and sleep disorder (17.2%). Statistics showed that the main disease factors of BZDs and Z-drugs were also different in different age groups. Insomnia and sleep disorders accounted for 71.4%, and anxiety disorders accounted for 64.1% of people over 50 years old. Among the 35–49 age group, 19.8% had insomnia or sleep disorders, 23.0% had anxiety disorders, and among the 18–34 age group, 8.2% had insomnia or sleep disorders and 18.9% had anxiety disorders.

With these clinical implications in mind, various adverse events caused by or related to the use of BZDs and Z-drugs have been widely reported [21]. Furthermore, there is an increased likelihood of falls and delirium associated with the long-term use of BZDs in older adults [22], and an increased risk of cognitive decline, fractures, and episodes of drug dependence in older patients [23]. Benzodiazepine users aged 85 and older fall more frequently [13].

These drugs are controlled under various international drug control treaties, according to the Global Drug Report 2018 of the United Nations Office on Drugs and Crime [5]. Benzodiazepines are the most commonly abused type of prescription drug for non-medical use and have become a growing problem. The non-medical use of BZDs is the most prevalent in the UK among illegal stimulants [24]. Non-medical use, including BZDs, is also growing in the Asia–Pacific region, including China [25]. Appropriate interventions are needed to address prescription drug abuse, including BZDs.

This study includes data before and up to approximately two years following the onset of the COVID-19 pandemic. The pandemic may have impacted the study, especially since these medications treat many mental health conditions that were exacerbated by the psychosocial effects of the pandemic. The authors analysed the data, and the histogram from the prescription showed no differences in 2020 compared with 2019 and a small difference in 2021 compared with 2020.

In synthesising the insights provided by the current research results, this study provides critical insights into the prescribing trends of BZDs and Z-drugs in China, revealing significant increases in usage that bear implications for both clinical practice and public health. Our findings highlight a substantial reliance on these medications, particularly among elderly populations, underscoring an urgent need for policy interventions. By documenting demographic and drug-specific trends, this research contributes to a better understanding of the potential risks associated with long-term use, including dependency and adverse effects. The data serves as a foundation for developing targeted guidelines that can enhance patient safety and optimise prescribing practices. This is crucial in settings where ageing populations are particularly vulnerable, and where there is a need to balance therapeutic benefits against the risks of harm.

Our study aimed to evaluate the prescription patterns of benzodiazepines (BZDs) and their alignment with the Zhejiang Provincial Hospital’s guidelines. The findings indicate that prescribed dosages, such as 5 mg/day for diazepam and 1.5 mg/day for lorazepam, were within recommended ranges. Diagnoses primarily included anxiety disorders and insomnia, consistent with the hospital’s criteria. Additionally, 80% of BZD prescriptions adhered to the short-term use guideline of 4–6 weeks, with exceptions clearly documented. These results demonstrate a high degree of conformity with the hospital’s guidelines, reflecting cautious and appropriate prescribing practices. Overall, our data underscore the effectiveness of these guidelines in ensuring the safe and appropriate use of BZDs, with dosages, diagnoses, and treatment durations generally meeting the recommended standards.