Prescription is written medico-legal document by authorized person for the treatment of patient where each prescription should contain all the four elements, superscription, subscription, inscription and signature with registration number issued by medical council (Sampling to Study Drug Use, WHO, 2000). In superscription, information of doctor (qualification, address) and patient (name, age, sex address) as well as date of prescription are included (Uddin et al., 2014). The subscription of a prescription includes to the information of prescribed drugs such as dosage form, drug name, dose, number etc. And in inscription directions for the use of drug are mentioned. Finally, signature of the registered medical practitioner with their registration number at medical council should be written as last element of prescription. All prescription orders should be clearly written by prescriber so that it can easily communicate between pharmacist and patient. Prescribing errors can be observed if sufficient and necessary information are not mentioned in prescription (Lesar et al., 1997)

Prescribing errors causes the irrational utilize of drugs and reduce patient compliance (Fijn et al., 2002). Study by Kuo et al. described that 70% prescribing errors was noticed among USA family physicians while study managed at West England reported that 43.8% prescriptions accommodate one or more prescribing errors (Kuo et al., 2008; and Seden et al., 2013). Study carried out at Nasik, India described that out of 300 prescriptions 279 prescriptions (93%) were irrational based on the utilize of essential drug WHO and standard treatment guidelines (Chaudhari et  al., 2011). Prescribing is a complex task requiring diagnostic skills, knowledge of medicines, an understanding of the principles of clinical pharmacology, communication skills, appreciation of risk and uncertainty (Khan and Ara, 2011). Irrational use of medicines is a global problem and common finding, particularly in developing country like Bangladesh. Poly-pharmacy, unnecessary use of antimicrobials, over use of injections and vitamins are commonly evident (Islam et al., 2020). On the other hand, aggressive drug marketing, lack of information drug use and inadequate drug supply has been suggested to be the main causes behind the irrational prescribing (Begum et al., 2012). 

The high cost of prescribed drug causes problems in developing countries like Bangladesh. This problem can be minimized by prescribing drugs by generic name and selection of drugs from essential medicine list (Sharif et al., 2019). Generic drugs are substitute of branded drug without any patient protections with similar efficacy but 40 to 60% cheaper than branded drugs (Siddarth et al., 2014). Irrational drug use leads to reduction in the quality of drug therapy, wastage of resources, increased treatment costs, increased risk for adverse drug reactions and emergence of drug resistance (Rahman et al., 2019). According to WHO prescribing indicators includes number of drugs prescribing per prescription, percentage of drugs prescribed by generic name, percentage of encounter by injection and antibiotics prescription and percentage of drugs prescribed from EML. 

Samples of prescriptions (by private practitioners) were collected from patients randomly. Prescriptions were collected either by photocopying or scanning by Smartphone after taking informed written consent from respective patients. A total of 300 prescriptions were the study sample. We had carried out Z-test. Variables were described as numbers and percentages. P=0.05 was considered as statistically significant. The Statistical Package for Social Sciences (SPSS) version 20.0 was used to analyze data.

Prescriptions were not categorized according to patients details like age, sex or disease profile. Afterwards all prescriptions (300) were analyzed using the WHO indicators. Following parameters were analyzed- a) Average numbers of drugs per prescription, b) Percentage of drugs prescribed by generic name, c) Percentage of prescriptions with an antibiotic prescribed, d) Percentage of prescriptions with an injection prescribed, and e) Percentage of drugs prescribed from essential drug list. 

Patients descriptions such as age and gender were not found in 6% prescriptions. It was evidenced that dose, direction of drug and duration of treatment was not completely mentioned in 60%, 72% and 52% of prescriptions respectively. Abbreviations were used in 100% prescriptions. Doctors medical registration number was mentioned in 13% prescriptions. Total 1042 drugs were prescribed in 300 prescriptions. Average number of drug prescribed was 3.38±1.79 (Mean±SD). 

Brand name was used in 93.33% prescriptions and generic name in 6.7% prescriptions. Use of antibiotics and injections was 64% and 8% respectively. 

Table 1: Analysis of prescription according to WHO prescribing indicators (n=300)

300 prescriptions were analyzed by using some of the WHO indicators. According to that analysis our study revealed that there was not fully complete prescription was found. Patients names, address, date of prescription were mention in 100% prescriptions. Similar to report showed by Siddarth et al., where it was 99.3%. Gender, age and address of patient were missing in 6% the prescriptions similar to study conducted at Maharashtra14 while Siddarth et al., reported that in 33% and 0.7% prescriptions address and gender were missing. Study conducted at Ethiopia (Desta and Abdulwhab, 1996) and Dubai (Sharif et al., 2008) reported that age, gender were not mentioned in 36.6%, 18.6% and 9.7%, 12% respectively which was higher as compared to our study. In this study, on an average, 3.38 drugs were prescribed per prescription, which was 3.81 in a study conducted in 2009 and 3.24 in another study conducted in 2011 (Rahman et al., 2009 and Paul et al., 2011). 

In an Indian study, the average number of drugs in a prescription was 3.128. In the current study, only 0.20% drugs were prescribed under generic names but Paul et al.; and Alam et al. reported it to be 0.13% and 1.33% respectively (Paul et al., 2011; and Alam et al., 2011). In an Indian study it was 4.24%, which was 44% in Nepal study (Kafle, 1992). In our study 64% prescriptions were prescribed with antimicrobials. This finding is quite similar with the study Rahman et 

al., 2009, where the result was 72.50%. In the study, about 64% of the drugs were prescribed from the Essential Drug List. Paul et al., 2011 and Baqui and Choudhury have reported that respectively 48.35% and 49% of prescribed medicines were from EDL (Paul et al., 2011). About 8% prescriptions contained an injection which was almost similar with the study done by Saurabh et al. About 60% prescriptions were provided with proper instructions regarding drug dosing, formulation and duration, which was 70% in the study of Rahman et al. 

On the other hand only 6% prescriptions contained proper instructions about patient medication information and advice like side effects of the prescribed drugs, other relevant advice and follow up of the patients (Shahen et al., 2019). These patient medication information parameters were 17.5% in the study Paul et al., 2011 and 35.68% prescriptions follow the proper instructions about prescription format. Irrational prescribing is a habit that is difficult to cure. However, prevention is possible. Intervention is needed to improve prescribing behaviors of doctors such as short problem based training course in pharmacotherapy and rational use focused workshops can improve prescription behavior and skills (Vries et al., 1995; and Thomas et al., 1997). Clear and comprehensives rules should be formulated and implemented by the government to ensure rational prescribing. 

Bulk of the prescriptions we studied, failed to meet WHO standard and antibiotics were surprisingly the commonest prescribed medication compared to other group of drugs. Although it is not a large scale study, our study reflects these two prescribing issues that need to be addressed. We advocate proper monitoring of good prescribing practice and strict regulation of antibiotic use policy in private chambers of medical practitioners throughout the country to enhance optimum patient care.

We are grateful and acknowledging the support from the medical practitioners whose prescriptions was collected for evaluation.   

The authors declare that they have no conflict of interests to publish the present research work.