Trends and patterns of benzodiazepines prescription from 2014 to 2016: A retrospective cross-sectional study in an outpatient setting of tertiary hospital in Malaysia.

Abstract


Introduction
Benzodiazepines are commonly prescribed for psychiatric disorder.Mother′ sLi ≤ Helpers′ and Sleepers' are among many nick names bestowed upon this class of drugs, to describe benzodiazepines.As it possess sedative, hypnotics, anxiolytics and muscle relaxants properties; it is often medically used to treat insomnia, panic disorder, relieve stress and seizures among other [1].BZD were widely noted as breakthrough drugs, which were believed to surpass its predecessor's effectiveness -the barbiturates.
Thus when it was marketed in the 1960, it became the most commercially successful drug at that time.They were thought to be a solution by many prescribers because they acted effectively on anxiety, mixed anxiety depression, and other nonspeci c physical complaints while at the same time being comparatively safe [2,3].BZDs are also often used in conjunction with other psychotropic drugs.For example, the combination of BZD and antidepressant led to greater response and to less drop-out than antidepressant alone in the acute phase treatment of depression [4].
However, the joy of prescribing it in abundance has taken its toll.Albeit benzodiazepine has legitimate indication, it also has its risk and harm that should not be overseen.After society discovered BZDs are taxed with addictiveness, it remains unclear how addictive they were when compared to other related list of psychoactive drugs [2].While short-term uses of these medications are generally safe and effective, the long term use of benzodiazepines is debatable, chie y because of the potential of dependence, withdrawal and tolerance [5].When use for extended period of time, they may lead to problems associated with discontinuation and withdrawal symptoms and also abuse [6,7].Benzodiazepine use is also associated with adverse drug events such as dementia and psychomotor impairments, which lead to premature death [8].Known risk factors for benzodiazepine-related adverse events include lung disease, substance use, and vulnerability to fracture [9].Although causality has not been de nitively determined, strong associations between benzodiazepine prescribing and mortality have been described in younger population [8].Benzodiazepines are drugs that should be used for very short time, at most for a few days or weeks in selected patients, carefully monitored, and stopped as soon as possible, as recommended in their summary of product characteristics and in all international recommendation [11].The risk-bene t balance of using BZD in the long term is accordingly to be negative.BZD is commonly prescribed for short-term duration (1-4 weeks) to reduce the risk of dependence and withdrawal [11].A WHO guideline for traumatic stress in non-specialized settings suggests that, when psychotherapy is not practical, short-term treatment with BZD may be considered [12].Other guidelines also make similar favourable mentions [13,14].
The Committee on the Review of Medicines in the United Kingdom, for example, has issued a warning regarding the regular long-term use of benzodiazepine and expresses their concern on the high number of patients with repeat prescriptions for an unusually long period [15].The primary basis of the worry seems to be about the risk of normal and excessive usage of benzodiazepine, the tolerance level, dependence potential and the abuse liability of benzodiazepines.Today, the dispute encircling the use of benzodiazepines is still open for debate.Benzodiazepine prescribing trends may vary in each country, but their long-term usage is usually seen and treated as the norm in many.
Quantifying benzodiazepines usage in populations helps to rea rm the current prescribing practices and pinpoint the existence of problematic consumption that could be prophetical for primary adverse outcomes.While several developed countries have established a nationwide register of patient's data to enable them to conduct a study not only on benzodiazepines but also other medications related to current pharmacoepidemiology issues, scant information is available about the nature of benzodiazepines prescribing in Malaysia compared to other countries.Due to this limitation, the data on the trends and patterns of benzodiazepines prescriptions and its use in Malaysia are still limited.This study aimed to investigate the trends and patterns of benzodiazepines prescribing at the out-patient setting of Malaysian public hospital.

Ethics Approval
Approval to conduct this study was granted by the Medical Research Ethical Committee (MREC), Ministry of Health Malaysia (Date: 11th October 2016 / Registration no: NMRR-16-1742-28839).

Study design and data source
This retrospective, cross-sectional study was conducted at the outpatient pharmacy of a public tertiary hospital in Malaysia.This study has been granted ethical approval from the Medical Research Ethical Committee (MREC), Ministry of Health Malaysia.It was carried out with the approved of the Medical Research Ethics Committee (MREC), Ministry of Health Malaysia (NMRR-16-1742-28839).The data on patient information were de-identi ed and only aggregate results were reported.There was no direct patients' involvement in this study that did not require informed consent from patients.
Prescriptions for seven benzodiazepines (alprazolam, bromazepam, clobazam, clonazepam, diazepam, lorazepam and midazolam) issued from January 2014 until December 2016 was identi ed together with these inclusion criteria; having at least one type of benzodiazepine prescribed in a prescription, the prescribed benzodiazepines must be in oral formulation only for patient's aged 18 years old and above.Information extracted from the prescriptions includes patient age and gender, name of medication, dosage form, dose, strength, frequency, duration, the amount supplied, diagnosis, and issuing department.
Calculation for the de ned daily dose (DDD) The number of DDDs consumed is calculated according to the following formula by Truter, Wiseman, & Kotze, (1996).The calculation of DDD was done for each patient starting from the rst day they received their prescription until the last day of our study frame.

Benzodiazepine prescribing patterns
Fundamentally, the patterns of benzodiazepines prescribed in the study are acquired from the information of the prescription slips.The applied patterns analysis is not corresponding to the number of patients or the number of prescriptions since one patient, or one prescription may contain more than one benzodiazepines.Moreover, each prescribing patterns are consistently following the number of dispensation, not the number of patients or prescriptions.

Data analysis
The patients and prescriptions were measured monthly and annually.Prescribing trends and patterns (ATC classi cation, duration of action and onset of action) for each type of benzodiazepine were also analysed and presented as number, percentage and DDD/1000/day.Descriptive statistics and association between patient characteristics and prescribing patterns were performed using STATA® SE v.13.1 (Stata Corp LP, Texas, USA).

Benzodiazepine prescribing patterns
The study discovers that from 5,728 prescriptions, there were 8,929 benzodiazepines issued throughout the three years.The analysis of prescribing patterns is described in Table 2 in regards to the number of benzodiazepines dispensation; n = 8929.In regards to the duration of supply for benzodiazepines, the study recognised there are 79 types of duration available in the data.
The benzodiazepines day supply ranges as one day to a maximum of 240 days (mean 34.43 days).Those durations were then grouped into ve types (1-7 days, 8-14 days, 15-30 days, 31-60 days and > 60 days), as listed in Table 2. Throughout the analysis of data, it was found that 16 types of dosing schedule involved in the study however, the dosing schedule is further categorised into two types; xed schedule and as needed schedule (Table 2).The Association between Gender, Age Group and Type of Benzodiazepines The Association between Gender, Group and Duration of action of Benzodiazepines Short-acting benzodiazepines are the most commonly prescribed benzodiazepines among male patients (n = 1,942, 21.75%) and among patient aged 45-54 years old (n = 1,158, 12.97%) (Table 5).Pearson chi-square analysis shows a signi cant relationship (P < 0.001) between gender, age group and the duration of action of benzodiazepines. .In summary, our study showed similar decreasing trends on the number of patients receiving benzodiazepines, the number of benzodiazepines prescriptions and also the DDD of benzodiazepines in comparison to the MSOM 2015.However, the prescribing trends of this study are comparatively low in contrast to other research outside Malaysia that was done during the same study period.Prior research in Australia also showed decreasing trends in benzodiazepines utilisation, though the overall benzodiazepines consumption remained relatively high.There were a total of 27.7, and 20.8 DDD/1000 patients/day of benzodiazepines dispensed in 1992 and 2011, while 14.2 DDD/1000 people/day were dispensed between 2013 to 2016, respectively [19,20].Furthermore, the yearly study in Finland also documented a declining trend from 15.7, 14.69 and 13.6 DDD/1000 people/day in 2014, 2015 and 2016, respectively [21].
The overall reduction trends in the number of prescriptions and the benzodiazepines de ned daily dose is likely due to strict governmental enforcement and establishment of levels of authority in relation to prescriber's category that prevented the widespread usage and over-prescribing of benzodiazepines in government healthcare facilities.The law and regulations covering the benzodiazepines have also prevented 'doctor shopping' and self-prescribes by patients via community pharmacies or illegal online stores since a prescription is compulsory for the purchasing of benzodiazepines.Moreover, there are also public educations of benzodiazepines potential adverse effects in this country.The reduction of trends can also be attributed to the changes in the price of benzodiazepines in the market throughout the year of study.However, since the study did not include the estimation of sales and procurement data, we cannot con dently associate to this factor.
Consequently, the prescribing trends of benzodiazepines can be seen as a result of multimodal in uence including from patient, prescriber and local prescribing practice [22][23][24][25].Verily, the use of benzodiazepines in each region varies greatly, depending on the nation, legislature and local practice.Benzodiazepine was not uncommon in Japan and was found in up to 11.9% of annual prescriptions in a tertiary care hospital [26].In Hong Kong and Beijing, benzodiazepine prescription was found in up to 29.9% of prescriptions to 505 outpatients [25,27].In North India, half of the psychotic and bipolar patients and two-third of depressive patients of the study group received benzodiazepines [28].reported that the elderly (aged 65 and above) to be the majority in using BZDs [29][30][31].Benzodiazepines are more dominant in female, and its use is increased steadily with age, which indicates the elderly are usually the highest receiver [18,30,[32][33][34][35][36][37].
Nevertheless, there is a signi cant similarity in view to the age group of benzodiazepines patient of this study compared with other established studies.Most of the patient that received benzodiazepines falls under the age of 45 years old and above.The lower rates of benzodiazepines prescribing to the elderly patients (65 years old and above) receiving benzodiazepines compared to other age groups in this study can be contributed to the practice guidelines which strongly recommended initiation of non-pharmacologic approaches and then the use of antidepressant over benzodiazepines as rst-line treatment for insomnia and anxiety among the elderly [38].Conversely, this nding provides evidence of the prescriber's adherence to the guidelines.
Alprazolam remained as the most frequent benzodiazepines prescribed to the study cohort throughout the study period.In Malaysia, alprazolam is available in three strength of tablet (0.25mg, 0.5mg and 1mg) while other benzodiazepines only have one strength of tablet.This could be one of the reasons that appointed them as the golden choice for prescribers.Variable options in tablet strength ensure convenience in dose tapering and high suitability for patient compliance.Alprazolam was observed in our study as the most prescribed benzodiazepines in which it is consistent with the ndings in the previous works of literature [37,[39][40].Though the choice of prescribing alprazolam seems to have its legitimacy, it is essential to note that, while alprazolam has a short duration of action, it has higher toxicity in overdose, and has been reported to be associated with fatality in relative to other benzodiazepines [40].
The choice of benzodiazepines frequently prescribed in the current study differs from other studies, which have been published previously.For instance, Sundaran et al., (2019), Grover et al. (2012) Tor et (2011) found the most prescribed benzodiazepines lorazepam and [23, 28, A 20 (1992-2011) study on benzodiazepines trends in Australia found that surprisingly, temazepam is the most dispensed benzodiazepines in the country, accounting to 35% of the whole prescriptions studied.Meanwhile, previous studies had reported diazepam as the most frequently prescribed [20,[41][42][43].Although clonazepam and diazepam both are listed as prescriber category B (ie. a prescription item) by the Ministry of Health Malaysia, they are not as frequently prescribed compared to other benzodiazepines.This gives rise to a question of whether the prescriber who prescribed alprazolam correctly ful ll the authorisation level provided.Therefore, we found that the differences in the choice of benzodiazepines in each country are assumed to be due to availability factors and differences in clinical practice.
Short-acting benzodiazepines such as alprazolam and midazolam are an excellent choice for acute relief of anxiety, panic attack, and stabilising mood.This is because they are rapidly absorbed, thus takes effect more quickly, and their therapeutic effect did not last too long to affect daily life routine.A recent study mentioned that short-acting benzodiazepines are recommended over long-acting benzodiazepines to alleviate the risk of dependency [44].Furthermore, previous studies also recorded the long-acting formulations of benzodiazepines such as diazepam and lorazepam are the dominant choices among the prescribers [35][36]42].Both the usage of short and long-acting benzodiazepines is recommended in many guidelines for different medical conditions.Due to the insu cient data to support the reason behind the selection of short or long-acting formulation, the study could not justify such preference.
In the present study, ndings showed that the male patient is more likely to be prescribed with anxiolytics, short-acting benzodiazepines and alprazolam.Additionally, looking into the duration of supply patterns, male patients are more likely to be dispensed with duration of 15 to 30 days, and the benzodiazepines were ordered with a xed dosing schedule.The signi cantly higher odds of male prescribed with benzodiazepines can be seen in another study as well [45].There is a statistically signi cant association seen between the tested variables and majority of patients aged 45-54 years old are very likely to receive anxiolytics, short-acting benzodiazepines, the duration of supply and the dosing schedule.Referring to Jessell et al., (2020) and Bushnell et al., (2017), they had described that patient aged 40 years and above and diagnosed with psychiatric disorders are more likely to be dispensed with benzodiazepines, especially if the prescribers are from the psychiatric department, compared with if other departments manage the patient [44][45].The high occurrence of benzodiazepines prescriptions in older patients may re ect an association between age, the prevalence of psychiatric illness and longer benzodiazepine prescriptions.The combination of the above may contribute to adverse clinical outcomes that frequently associated with benzodiazepines in the elderly, such as the risk of fall, dose-dependency and mortality.

Study Limitations
Our study has several limitations.We focused only on the prescription data, and this necessarily does not re ect the exact clinical situation.Moreover, our study was limited to one tertiary care, which may or may not re ect the national trends.
Furthermore, the prescription data may not include the indication for initiating treatments since the principal diagnosis written on the prescription does not necessarily refer to the primary indication of each benzodiazepine.Due to compliance problems, some of the prescribed drugs may not have been consumed according to what has been ordered in the prescription.Hence, we can only get the total number of benzodiazepine prescribed to the patient, but not the total number of benzodiazepine consumed.
Besides, there is insu cient data to support the status of the patient; whether they are naïve or chronic benzodiazepine user.
The study is unable to determine that the rst appearance of the patient in the data collected during the study period is the rst dispensation of benzodiazepine to the patient or the rst time the patient takes the medication.The patient may have defaulted follow up for a few weeks or months before the start of the study.A comprehensive comparison with a longer duration of retrospective years was limited due to lack of data availability.

Figure 1 showed approximately 5 .
Figure 1 showed approximately 5.5% increased of prescriptions from 2014 (n = 1,956) to 2015 (n = 2,063), but later a decreasing trend of 34.3% from 2015 to 2016 (n = 1,709).A line chart with a trendline was constructed to see the trends over months of the study duration, which witnessed an increase in trend in 2014 and decreasing trends in both 2015 and 2016.In general, the study observed a drop of 12.6% number of prescriptions from 2014 to 2016.
outpatient pharmacy throughout the study duration.Based on the ATC classi cation, benzodiazepines that fall under anxiolytics are the highest prescribed by the prescribers since 2014.The result also revealed that short-acting benzodiazepines are the leading choice (n = 3591, 40.22%) in terms of frequency (n) and percentages (%) for three years; compared to longacting (n = 1,860, 20.83%) and intermediate-acting benzodiazepines (n = 3,478, 38.95%).However, the pattern of prescribing changed from short-acting benzodiazepines in 2014, towards intermediate-acting in recent years (2015 and 2016).The analysis also indicates that in view on the onset of action, prescribers frequently prescribed intermediate-action benzodiazepines (57.67%) rather than rapid-action (42.33%) benzodiazepines.This pattern continues from 2014 till 2016.Throughout the study period, there were no changes in the patterns of dosing schedule, duration of supply, concurrent medications, diagnosis and issuing department.All three years saw xed dosing as the most frequent dosing schedule, 15-30 days was the most common duration of supply, antidepressant as the most common type of medication concurrently prescribed with benzodiazepines, a psychiatric disorder was the highest diagnosis written for benzodiazepines patient and psychiatry department mainly prescribed BZD in the study.Prescribing Trends of Benzodiazepines Based on De ned Daily Dose (DDD) On the other hand, Kaufmann et al., (2016), Tu, Mamdani, Hux, & Tu, (2001) and Cunningham, Hanley, & Morgan (2010)

Figures Figure 1 The
Figures

Figure 2 The
Figure 2

Table 3
presents the trends of benzodiazepines in miligrams and DDD for year 2014, 2015 and 2016.In 2014, 117,942.5mg of

Table 5
Association Between Gender, Age Group and Benzodiazepines Duration of Action and Morbidity Survey (NHMS) 2015 reported that there is an increasing trend in the prevalence of mental health problems among adults.The prevalence of approximately escalates from in 1996 29.2% in 2015 Naturally, in with the cases, we would a similar in the prescribing trends of psychotropic drugs, especially benzodiazepines.However, the Malaysian Statistics on Medicines (MSOM) 2015 reported a drop of 22% in the overall usage of benzodiazepines; from 1.313 DDD/1000 patients/day in 2011 to 1.024 DDD/1000 patients/day in 2014[18]