Prescribing pattern of benzodiazepines in outpatients without a diagnosis of mental disorders-Retrospective study

1Medicinski fakultet, Katedra porodične medicine, Univerzitet u Banjaluci, 2Medicinski fakultet, Katedra porodične medicine, Univerzitet u Sarajevu, 3Medicinski fakultet, Foča, Katedra porodične medicine, Univerzitet u Istočnom Sarajevu, 4Dom zdravlja, Banjaluka, Bosna i Hercegovina Obrazac propisivanja benzodiazepina ambulantnim pacijentima koji nemaju dijagnozu mentalnih bolesti Retrospektivna studija

U našoj zemlji je sproveden manji broj studija na temu pretežne upotrebe BZD. Dostupni podaci 9 do kojih smo došli Introduction Benzodiazepines (BZDs) are efficient drugs, mostly prescribed to reduce anxiety, treat insomnia, cause myorelaxation, and prevent seizures 1 . Clinical guidelines generally recommend short-term use of these drugs for up to 6 weeks 2 . The recommendations for the short-term use limit are based on drug tolerance and addiction 3 .
However, regardless of all recommendations, the prescribing pattern and use of BZDs worldwide is increasing. In the USA, BZDs prescription (primary for anxiety disorders and insomnia) has increased by 320% from 1996-2013. In the same period, the percentage of death outcomes related to BZDs overdose has increased by 500% 4,5 . According to the report of The National Institute on Drug Abuse (NIDA) from 2004, 48 million people older than 12, use benzodiazepines in the USA, which is approximately 20% of the total population 6 .
A wide-range study about the use of drugs conducted from 1996.-2002, at the University of pharmacology, British Columbia (BC), Vancouver, Canada, reported that benzodiazepines are among the top prescribed drugs. Only during 2002, 84 million pills of benzodiazepines were prescribed 7 . The same study showed 9.7% of the BC population (400.000 people) got at least one prescription for benzodiazepines during 2002.
The study conducted in Germany from 2009-2014 recruited 31 family physicians and showed that approximately 5% of patients who visited a family physician got at least one prescription for BZD. The results of the study showed that BZDs were routinely prescribed in family medicine 8 . The study about benzodiazepine prescriptions in the Republic of Srpska (Bosnia and Herzegovina) from 2011-2018 revealed that 2.98% of all prescriptions were prescriptions for benzodiazepines. The study showed these drugs were prescribed to 10.63% of patients, and 23.81% of BZDs were prescribed for non-psychiatric diagnoses 9 .
Many physicians often prescribe benzodiazepines for patients with depression, anxious reactions, cardiovascular and gastrointestinal diseases, tension headache, chest pain, or low back pain. It appears that BZDs not only diminish symptoms of anxiety but also diminish somatic difficulties of cardiovascular and gastrointestinal diseases, although the exact pathophysiological mechanism is not yet well understood 10 .
The results of numerous studies revealed benzodiazepines are more often used by female patients than male patients, older patients (> 65 years), patients with chronic diseases, and patients living in urban areas [11][12] .
In our country, a small number of studies on the topic of prevalence and use of benzodiazepines were conducted. The available data 9 , we came across, reveal a high prevalence of benzodiazepine use in the patients in the Republic of Srpska. It is to be expected that patients with mental disorders use these drugs at a higher percentage. However, it is estimated Obrazac propisivanja benzodiazepina ambulantnim pacijentima koji nemaju dijagnozu mentalnih bolesti -Retrospektivna studija Opšta medicina, 2021;27(1-2):1-10 Prescribing pattern of benzodiazepines in outpatients without a diagnosis of mental disorders -Retrospective study General Practice 2021;27(1-2):1-10 that patients without the diagnosis of mental disorders use these drugs in a considerable number.

Objective
Our objective was to investigate the prescribing pattern of family physicians for BZDs in adults, without the diagnosis of mental disorders, and the influence of socio-demographic characteristics on BZD users (gender, age, level of education, marital status, residence -urban or rural).

Method
The research represents a retrospective study of benzodiazepines use in a ten-year period, (2009-2019). The study was conducted in five family medicine teams, from three primary health care centers in the Republic of Srpska: 3 family medicine teams from primary health care center Banjaluka, with the mainly urban population, and one team from primary care center Foca and Celinac, respectively, with the mainly rural population.
The study included all the adults over the age of 18, who were the patients of the five aforementioned family medicine teams. The study included a total of 8.560 patients. Out of this number 259 patients with mental disorders were excluded, so the remaining 8.301 patients were included in the study. For the final analysis, we used the data of the patients that had at least one prescription for benzodiazepines in a ten-year period. The data about prescribing patterns of family physicians and socio-demographic characteristics of patients were obtained from electronic health records (EHRs).
The data collection was performed by reviewing EHRs and a questionnaire was created for the purpose of this study to enter the data from the medical records. Regarding prescribing pattern of family physicians, the following data were analyzed: is the drug recommended by a family physician or prescribed following the recommendation of another specialist; the presence of chronic diseases (cardiovascular diseases, chronic pulmonary diseases, diabetes, musculoskeletal diseases, cancer); the class of used benzodiazepine (bromazepam, diazepam, alprazolam, other); the duration of therapy (<6 weeks, ≥6 weeks to 1 year, 2-5 years, 6-9 years, ≥10 years), and course of benzodiazepine therapy (continuous or "as required").
Socio-demographic data used for the purpose of analysis were: age, gender, level of education, marital status, and residence.
For the purpose of the statistical analysis, in relation to age, patients were divided into four age groups: 18-30 years, 31-45 years, 46-64 years and ≥65 years. In relation to the educational level, patients were divided into four groups: patients without formal education, patients who graduated from primary and secondary school, and patients who graduated from a university. In relation to marital status, patients
were divided into four groups: married, single, divorced, and widowed. In relation to residence, patients were divided into two groups: urban and rural. The patients from the city of Banjaluka were distributed in the group with urban residence, as the three family medicine teams mainly treat the population that lives in an urban area. The patients from Foča and Čelinac were distributed in the group with rural residence, as these two family medicine teams mainly treat the population that lives in a rural area.
This research was conducted according to the Helsinki Declaration about medical researches and the principles of good clinical practice. The approval for conducting the study was obtained from the Ethics Committee of the Primary Health Care Center Banjaluka.

Statistical analysis
The collected data were presented for at least 95% of the included patients. The categorical variables were presented as numbers and percentages and compared using the chi-square test. The continuous variables were expressed as the mean with standard deviation or median with interquartile range and compared with Studentʼs t-test or Mann-Whitney U test according to the data distribution and the number of groups. Kolmogorov-Smirnov test and visual assessment were used to estimate the normal distribution of continuous variables. Univariate, as well as multivariate binary logistic regression analysis, was used to identify independent predictors of `as needed` treatment. All statistical analyses were performed using IBM SPSS Statistics, version 25.0 (New York, USA).

Results
Out of a total of 8301 analyzed medical records, in a 10year period, 1044 patients (12.58%) used benzodiazepines.
Patients using benzodiazepine were most frequently females (71.07%), persons over 65 (44.54%), patients who lived in urban areas (75.96%), and patients with chronic diseases (88.60%). Married patients were the most frequent users, and in respect to education, those with secondary school (60.44%) used benzodiazepines the most ( Table 1).
The largest number of benzodiazepine prescriptions was not recommended by family physicians, but other specialists (65.42%). The majority of patients were recommended to use the drugs, not as a continuous course of therapy, but `as needed', 732 (70.11%). In relation to the duration of benzodiazepine therapy, the majority of patients used these drugs for 2-5 years, and a small number of patients, 169 (16.19%), used these drugs according to recommendations (<6 weeks). The most used benzodiazepines in our patients were bromazepam 837 (80.17%), diazepam 116 (11.11%), and alprazolam 61 (5.84%), while other benzodiazepines were used in a smaller percentage (Table 1).
Obrazac propisivanja benzodiazepina ambulantnim pacijentima koji nemaju dijagnozu mentalnih bolesti -Retrospektivna studija Opšta medicina, 2021;27(1-2):1-10 In relation to residence, all patients were divided into a group that lives mainly in an urban area (4.981 patients) and those who live mainly in a rural area (3.579 patients). The use of benzodiazepines was analyzed in relation to residence, as well as the influence of socio-demographic characteristics on the use of BZDs in these two groups of patients. Patients living in urban areas used BZDs in a considerably higher percentage when compared to those living in rural areas. Therefore, out of a total of 4981 analyzed EHRs from urban area residents, 793 (15.92%) were prescribed benzodiazepines, at least once, in the follow-up period. In the rural area, out of a total of 3579 analyzed EHRs, 251 (7.01%) were prescribed BDZs, at least once, in the follow-up period. (Graph. 1).
Females in urban area used BZDs more often than those in lmural areas and it was statistically significan (p=0.048). The patients who graduated from secondary school and patients over 65, in urban areas, used benzodiazepines more often than those in rural areas, which was (p<0.01) statistically significant. The patients in urban areas used benzodiazepines for longer periods when compared to those from rural areas and the difference was statistically significant (p<0.01). The family physicians in urban areas independently prescribed benzodiazepines more often when compared to their colleagues working in rural areas and the statistical difference was significant (p<0.01). When considering the course of therapy (continuous or`as needed` marital status, and the presence of chronic diseases, the statistically significant difference between the use of benzodiazepines in patients from urban and rural areas was not found, ( Table 2). U odnosu na mesto stanovanja, svi pacijenti su podeljeni u grupu koja uglavnom živi u urbanoj sredini (4.981 pacijenata) i one koji uglavnom žive u ruralnim područjima (3.579 pacijenata). Upotreba BZD je analizirana u odnosu na mesto stanovanja, a ispitivan je i uticaj sociodemografskih karakteristika na upotrebu BZD kod ove dve grupe pacijenata. Pacijenti koji žive u urbanim sredinama koristili su BZD u znatno većem procentu kada smo ih uporedili sa pacijentima koji žive u ruralnim područjima. Od ukupnog broja analiziranih EZK, kod 4.981 pacijenta iz urbanih područja, 793 (15.92%) je imalo propisan BZD najmanje jednom u periodu praćenja. U ruralnim područjima, od ukupnog broja 3.579 pregledanih EZK, 251 (7,01%) pacijent je imao propisan BZD najmanje jednom u periodu praćenja (Grafikon 1).

Discussion
The results of our study showed a high prevalence (12.58%) of benzodiazepine use in our patients. Female patients, patients over 65 years, and patients with chronic diseases used these drugs more often when compared to other groups of patients.
The results of other studies showed similar results considering the prescription and use of benzodiazepines. The study conducted in 2010 among the patients of a family medicine team in Primary Health care Center Banja Luka showed that even 20.41% of patients over the age of 18 got at least one prescription for benzodiazepines in a one-year period 13 . The results of this study showed females and persons over 65 used benzodiazepines more often.
The results similar to ours were obtained in the study from Sweden. This study showed the factors related to the more frequent use of benzodiazepines were older age, female sex, living in an urban area, use of many other drugs, and simultaneous use of psychotropic drugs, especially antidepressants 14 .
The study conducted in Pakistan aimed at investigating the prevalence of benzodiazepine use in suburban and urban population in the city of Karachi. The total prevalence of benzodiazepine use was 14%, which relates to our results. However, unlike our patients, the patients from the suburban areas, from this study, used benzodiazepines statistically sig- Rezultati iz iste studije su pokazali da je najčešće korišćen BZD bio bromazepam, zatim diazepam, što je slučaj i sa našim pacijentima.
Koliko je nama poznato, slično ispitivanje na temu propisivanja BZD pacijentima bez mentalnih bolesti, u našoj zemlji nije rađeno. Ovom studijom smo želeli da ukažemo na preterano propisivanje BZD i da ohrabrimo lekare porodične medicine, ali i druge lekare, da isprave svoju propisivačku praksu i prilagode je preporukama. nificantly more when compared to the subjects from the urban areas 15 . The results from the same study revealed that the most used benzodiazepines were bromazepam, followed by diazepam, which is the case in our patients, as well.
The study conducted in the Republic of Srpska, which also included patients with mental disorders, showed similar results as our study. Females, the elderly and patients living in urban areas used BZDs for longer periods 9 . The same study also showed that people who live alone use BZDs more often, unlike the results of our study where married persons used BZDs more frequently.
The study conducted in the USA from 2003-2015 showed that the number of visits to a family doctor, due to benzodiazepine prescribing, has increased from 3.8% in 2003 to 7.4% in 2015. Family doctors prescribed approximately half of these drugs 16 . Our study showed the opposite results, that is to say, the majority of benzodiazepines were prescribed following the recommendations of other specialists, and approximately one-third of these drugs were prescribed by family doctors independently.
Tension relieving, insomnia or worries due to one's family financial problems, everyday life, and existential problems are mentioned as the most frequent reasons for chronic use of benzodiazepines 3 .
Numerous researches showed a high prevalence of chronic use of benzodiazepines, which calls for the need to educate family physicians on methods that would decrease the overuse of these drugs 17,18 . Family physicians play an important role in decreasing the long-term use of benzodiazepines. Patients should be prescribed a short-term course of these drugs, preferably for up to 4 weeks, and at the lowest effective dose for the treatment of severe anxiety, panic disorder, and acute crisis reactions. Instead of prescribing benzodiazepines for the treatment of mild anxiety and insomnia, counseling, cognitive-behavioral therapy, sleep hygiene, and strategies for self-help should be recommended. Family physicians should introduce patients with side effects of these drugs, including influence on driving, operating machinery, and addiction, before benzodiazepines were being prescribed. Indications for chronic use are exceedingly rare. In these cases, it is best to ask for an opinion of a psychiatrist 19,20 .
The results of our study showed that the patients in urban areas used benzodiazepines more than those in the mainly rural areas and this difference was statistically significant. We think that different lifestyles in these areas may be the reason why. It is an assumption that patients in mainly rural areas live with less stress, and they are more physically active and therefore have less need for the use of benzodiazepines.
Similar research on the topic of prescribing BZDs for patients without the diagnosis of mental disorders was not conducted in our country, as far as we know. With this study, we wanted to make a point on BZD overprescribing, and to encourage family doctors, as well as other doctors to correct their prescribing pattern and adjust them to the recommendations.

Conclusions
The results of our study showed that the use of benzodiazepines in patients without the diagnosis of mental disorder is high and considerably higher than it`s recommended. These drugs are used more by females, the elderly, and persons with chronic diseases, which is similar to the results of other studies. The patients in urban areas use benzodiazepines more than those in rural areas. The strategies to reduce the benzodiazepines prescription, which would include the education of physicians and patients, are necessary to reduce the chronic use of these drugs.