Pattern of Prescription Practices in Acute Respiratory Infection and Diarrhoea Cases in Out Patient Department of Tertiary Care Hospitals in West Bengal, India; A Cross Sectional Analysis through Rational Use of Medicine Consensus Approach

Background: Evaluation of prescription patterns would determine the drug utilization with main emphasis on rational use of medicine. The problem of irrational use of drugs is rampant particularly in developing nations. The present study was undertaken for evaluating the prevailed prescription patterns in tertiary hospitals with diarrhoea and/or Acute Respiratory Infection (ARI) to address specic areas of deciencies and deviation from the available guidelines. Method: We conducted this observational cross-sectional study from August 2019 to December 2020 in Medicine & Paediatrics outpatient departments and Urban Health Training Centre in two Government teaching hospitals in West Bengal, India. We included 630 prescriptions (511 – ARI, 119- diarrhoea) and evaluated in terms of disease and medicine prescribed including antibiotic related indicators. We compared prescription patterns across different age groups, different strata of prescribers and compared against WHO standards. A Rational Use of Medicine Consensus (RUMC) committee was formed and the prescriptions were assessed for appropriateness independently by a pharmacologist and clinician. Deviations, if any, were ascertained from the available guidelines and the acceptability of the deviations were determined by consensus Result: Age and sex were mentioned in all prescriptions however signs & symptoms, provisional diagnosis and follow up visit were mentioned in 90.3%, 4.9% and 67.9% prescriptions respectively. Body weight was mentioned in 88.5% of prescription of children (< 18 years). Higher rates of Fixed Dose Combination (51%), lower proportion of generic drug (23.3%) and adherence to hospital formulatory (36.5%) were some the major concerns identied. Antibiotics prescription rate (APR) and multiple antibiotic prescription rate (MPR) were respectively 57% and 10%; both found signicantly higher for diarrhoea than ARI. Deviations from Standard Treatment Guidelines were found in 98.9% prescriptions and 90.4% of which were unacceptable. Agreement between clinician and pharmacologist was observed in 90% prescriptions (Kappa -0.114). Deviations were most commonly observed with prescriptions by interns and house-staff (99.6%), whereas acceptable deviations were more frequent among the residents (15%). Conclusion: We conclude that in light of identied irrational prescription patterns,


Background
Evaluation of prescription patterns comprises of studies mentioning drug utilization with main emphasis on rational use of medicine. The drugs are said to have been used rationally when medicines are prescribed to the patients aptly according to their clinical requirement, in doses appropriate to their clinical needs, for the requisite duration with minimum expenditure for both the patients and the community (1,2.) The problem of irrational use of drugs especially antibiotics is prevailing worldwide (3,4). Absence of access to essential medicines have been found in one-third people whereas more than half of the medicines are either incorrectly prescribed or inappropriately dispensed or sold. Also absence of drug adherence was found in 50% patients. (5). The problem is further complex in developing countries (6). In developing countries, respiratory illnesses and diarrhoeal diseases remain the main causes of morbidity and mortality particularly in children, accounting for one in ve deaths and resulting in 1.5 million annual fatalities (9). Majority of diarrhoea cases in children, especially under the age of ve years are caused by viruses whereas both bacterial and viral diarrhoeas are found in adults. Motility disorders, lactose intolerance, irritable bowel syndrome, bile salt enteritis, in ammatory bowel disease are some of the non-infective causes leading to diarrhoea which are also more frequent among the adult population. Though it has been estimated that antibiotics are required in 5% of diarrhoea cases, use of antibiotics in practice are rampant (10). According to WHO, in developing countries half of all viral respiratory tract infections and viral diarrhoea were treated with antibiotics. Also, antibiotics were not found to be prescribed in 70% of pneumonia cases where the use was an absolute necessity (11).
However, Overuse or underuse of Medicines ,prescribing wrong or ineffective medicines, polypharmacy practices, use of expensive xed dose combination products and misuse of antibiotics are the common forms of irrational prescription. (7,8). Irrational prescription continues in practice despite presence of treatment guidelines. Since both Diarrhoea and Acute Respiratory Infections are major public health problem particularly in children, speci c management guidelines are available from WHO, Ministry of Health Family Welfare, Govt of India and in different State Government Health Departments as well. (9,12,13) However, the major challenge in implementing these guideline is lack of optimization between general practitioners and specialists. Guidelines prepared by WHO, Ministry of Health etc are actually tailormade for public health practitioners practising at low resource peripheral health settings. However, implementation of such guideline at tertiary care teaching hospital gets in uenced by scienti c knowledge and experience of different prescribers. Major concerns to prescribers remained: encroachment on professional autonomy and applying the same standards at all levels. (14). Selection of one guideline to follow in presence of many is also a matter of choice. Even in England, there is no predominance of adhering to NICE guideline over those formulated by Guidelines and Audit Implementation Network (GAIN) or the Royal Colleges. (15). The World Health Organization (WHO) has advocated for developing and practising Standard Treatment Guidelines (STGs) and National List of Essential Medicines Lists customised to the country. (16). Adaptation and customization of international & national guidelines to the local guideline, is a pragmatic approach as variation is likely to be expected between disease risk & population pro le and health system determinants. A locally or regionally oriented guideline also build local ownership and acceptance. (17)(18)(19)(20)(21)(22)(23)(24) The Rational Use of Medicine Consensus (RUMC) is a national task force study being conducted by Indian Council of Medical Research (ICMR) in India with the aim to formulate a national standard guideline and training modules for different level of practitioners (intern, private practitioners, practitioners at public health settings etc) based on context speci c evidence evolved from the study. Hence under the ambit of RUMC, the present study was undertaken for evaluating the prescription patterns for acute onset diarrhoea and/or Acute Respiratory Infection mapping the speci c areas of de ciencies and deviation from the available guidelines. Methods a) Aim, design and setting: The primary objectives of this multicentric study were to assess prescribing patterns using the WHO criteria for prescription evaluation and to determine appropriateness of the prescription and acceptability of the deviations from standard guidelines through a consensus committee approach.
We used speci c indicators expressed in terms of proportion (%) in disease related and medicine related domains. Within medicine related domains separate indicators are also used in case of Antibiotics as follows: 2)Multiple antibiotics prescribing rate (MPR %) = number of prescriptions that included at least two antibiotics of different groups /total number of prescriptions that included at least one antibiotic × 100%.
3)Parenteral antibiotics prescribing rate (PAPR%) = number of prescriptions that included at one injectable antibiotic /total number of prescriptions that included at least one antibiotic × 100%.

Study design & setting:
This cross-sectional study was conducted from August 2019 to December 2020 in the OPDs of Medicine and Pediatrics and the Urban Health Training Centre (UHTC) clinic of two Government teaching hospitals of West Bengal, an eastern state of India. Both are tertiary care hospitals. Former is located in the city whereas the later is at the peripheral district, however both these hospitals usually serves large segments of population.

b) Characteristic of participants
Inclusion & Exclusion criteria: The study participants included prescription of patients ful lling the following inclusion and exclusion criteria.

Inclusion criteria:
Prescriptions of patients mentioning the following details: 1. Signs & symptoms and/or diagnosis (provisional or nal) 2. Prescriptions for the disease (any medication) 3. Patients /legally authorized person of the patient have to provide verbal consent/written consent to capture/ copy the prescription for review Exclusion criteria: 1. If critically ill, or consent not taken/cannot be given 2. Prescription without a single medication 3. Prescription not available from the patient.

Sample size calculation
For prescription evaluation studies, according to WHO a minimum of 600 prescriptions (25) are required to be studied. We assessed 630 prescriptions during the study period. Prescriptions were assessed after the physician have examined the respective patients and prescribed the medications. c) Clear Description of Processes: Data collection: Patients coming to the OPD was approached after consultation while exiting from the OPD with the prescription or at hospital dispensary, through consecutive sampling. Prescriptions were screened for eligibility following consenting and those ful lling the inclusion criteria were included in the study with capturing a photograph of the prescription. From the photographed prescription, demographic details, relevant clinical information, diagnosis and medication details, were abstracted in the Case Record Forms.

Ethical Statement:
After obtaining approval from the Institutional Ethics Committees of ICMR-NICED and participating institutes, written informed consent was taken from the patients returning from OPDs.

Data Management and analysis plan:
The data management system comprised of data entry, cleaning, back-up, and generation of regular reports. Built-in quality control mechanisms will be developed to ensure data quality and con dentiality. Prescriptions were analysed using the mentioned WHO indicators.
All above proportions were compared across age group (Pediatric and adults) and different categories of prescribers viz. Interns and House staffs, Post Graduate Residents and Faculties (RMO/Demonstrator/Clinical Tutor, Assistant Professor, Associate Professor and Professor.)

Assessment of prescription through Consensus Committee approach:
A Rational Use of Medicine Consensus Committee was formed with the objective to develop the assessment framework for diarrhoea and ARI prescription towards determining appropriateness and acceptability of the deviations. Both clinician and pharmacologist evaluated the prescriptions independently for appropriateness and identi cation of deviation from standard treatment guidelines.
Following treatment guidelines were consulted as standard: (9,12,13,26) Since all the guidelines are mostly targeted to general practitioners in public health settings, they do not cover many additional drugs such as probiotics, antihistamines, leukotriene receptor antagonist, bronchodialators, mucolytics etc which are commonly prescribed in medical colleges setting. In such cases both pharmacologist and clinician performed the assessment based on the scienti c rationale and their clinical expertise.
Further the pharmacologist judged the prescriptions as appropriate/inappropriate on the basis of signs and symptoms prescribed, adverse effects of drugs, route of administration, dose (appropriate as per age/ body weight, individualization, maximum dose/day mentioned for acute drug), duration being correct, as per documented indication, possibility of drug interaction, prescription of generic names. The clinician judged the prescriptions independently as appropriate/inappropriate according to above criteria as well as their clinical judgement particularly optimizing symptom remission and tolerable adverse effects. Acceptability of Deviation was determined using the following matrix:  Figure 1] Majority of the unacceptable deviations were due to antibiotics, followed by bronchodilators , antihistaminics, Proton Pump Inhibitor/H 2 receptor blocker/Antacids, Probiotics. [ Table 2] Some of the frequently observed unacceptable deviations were as follows: 1 Appropriateness of prescriptions as per clinician and pharmacologist revealed that only 1.1% of total prescriptions were appropriate according to both clinician and pharmacologist. There were 9.6 % prescriptions which were inappropriate as per pharmacologist but appropriate as per clinician. Agreement between pharmacologist and clinician were observed in 90.4% prescriptions (Kappa -0.14) [ Table -3].
These disagreements were resolved in RUMC meeting on case to case basis with following clari cation: 1. Prescription of antihistaminic in ARI in children though identi ed as inappropriate by pharmacologist however a 2nd generation antihistaminic (cetirizine) may be considered as acceptable deviation but 1st generation (Chlorpheniramine) is unacceptable due to excessive sedation.
2. Prescription of Azithromycin in URTI identi ed as inappropriate by pharmacologist, as Azithromycin is not a 1st line antibiotic, but considered acceptable deviation by the clinician as standard practice.
3. Drugs prescribed by brand names are considered inappropriate by the pharmacologist but it was considered as an acceptable deviation.
4. Prescription of albendazole in children or Vitamin D in infants less than 6 months of age though considered inappropriate by pharmacologist when there is no indication, it is considered as acceptable deviation by consensus it adheres to national programme guideline 5. ORS prescribed without speci c indication is also an acceptable deviation as it cause no apparent harm.

Completeness of prescriptions as per different criteria in various age groups
Body weight was mentioned in 88.5% of prescriptions of patients below 18 years age group. Of them, body weight was found in 93.4% patients upto 5 years of age, 88.9% of patients aged 5 -< 12 years, and 14.3% patients of 12-<18 years. None of the adult prescriptions had body weight mentioned in them. Signs and symptoms, provisional diagnosis were more commonly mentioned in adult patients whereas follow up visit was more commonly mentioned in <18 years age group. Higher proportion of prescriptions with generic name (27% vs 17%), and from hospital schedule list (41% vs 29%) and lower FDC (44% vs 63%) were observed in prescriptions of < 18 years as compared to adults. APR was also lower for children than adults (48 % vs 71 %).
Prescriptions with deviations were slightly lower in children (98.5% vs 99.5%). However, proportion of acceptable deviations were more in <18 years age group (12% vs 6%). [Data not presented]

Completeness of prescriptions as per different criteria across types ofprescriber
Body weight, signs and symptoms, follow up visit was mentioned most commonly by residents while provisional diagnosis was commonly mentioned by faculties. Prescriptions of all drugs with generic names and from hospital schedule list was mostly prescribed by residents while xed dose combinations and antibiotics were mostly prescribed by faculties. Deviations were most commonly observed in the prescriptions of interns and housestaffs (99.6%), whereas acceptable deviations were more common among the residents (15%).Out of the unacceptable deviations, chances of ADR was most common among interns and housestaffs whereas prescriptions with inconsistent/irrational indication was most common among the faculties.[ Table 4]

Discussion
In this study, polypharmacy emerged as a major concern as the average number of drugs prescribed per patient was 4.2 ± 1.9 which is much higher than WHO standard of ≤ 2 (27). However, few studies have also mentioned higher average number of drug prescription (28, 29) per patient, whereas, much lower estimates (1.5) were also observed among under ve children with acute diarrhoea in Bangladesh (30). Several others studies also reported an average ranging 2.8-3.2 drugs per patient (31)(32)(33)(34)(35). Mostly Prescribers are unaware of availability of Standard Treatment Guidelines and further lack of point of care rapid diagnostic facilities contributes to higher number of drugs prescription per patient. The higher number of drugs may enhance the chance of adverse drug reactions, antimicrobial resistance, healthcare expenditure and also interfere with prescription adherence. The high number of drugs per prescription may also account for the higher number of (51.0%) xed dose combinations prescribed in this study.
Only 23.3% drugs were prescribed by generic names in this study. This is much lower than the standard cut off of 100% (27). Higher proportions of generic names were found in studies by Viswanath et. Al (36) (62.3%) and Shankar PR et al (29) (58.1% ). Also in various other studies proportion of generic names in the prescriptions were found to be 46.2%-100% (31,32,37). Use of generic names is recommended by Government to reduce the healthcare costs. However, prescription patterns by different categories of prescribers may be in uenced by promotions of particular brands of drugs accounting for such discrepancies. It was observed that 53.6% residents prescribed generic names as compared to only 18% for interns and faculties. However, drugs with Fixed Dose Combinations (FDC) were prescribed in little more than half of the prescriptions and similar ndings had been reported by others also (37). Overall 57.0% prescriptions have at least one antibiotic prescribed. Considering the higher magnitude of infectious diseases in the developing countries, WHO has limited the use of antibiotics in < 30% prescriptions in case of all infectious diseases (27,39). Thus we observed a very high APR and signi cantly higher in case of diarrhoea and adult patients compared to their counterparts in the study (p < 0.05). In India, irrational antibiotic prescription is a serious concern as re ected by rates (20-72.8%) reported by different studies. (31,32,33,34,37). Antibiotic prescriptions without a provisional diagnosis in a rst time patients supports the notion of physician to cover for immediate medical catastrophes than to consider back up antibiotics for future implication in the era of rapidly emerging antimicrobial resistance.
We reported a MPR of 10% which is much lower than studies by Ashraf et.al (40) and Panchal et al (28) however, much lower usage of antibiotics of 1 per prescription i.e. 0% MPR was also reported by Bordoloi et al. (41) Disease wise variation shows that any antibiotic has been prescribed in 55.2% of ARI and 65.5% prescriptions with diarrhoea. In most of the prescriptions antibiotics have been prescribed as an empirical therapy without mentioning any provisional diagnosis. A study by Hekster et al also reported similar ndings where diagnosis was not the deciding factor for prescribing antibiotics in half of the prescriptions (42.) Most episodes of watery diarrhoea in children and sometimes in adults supposed to be of viral aetiology where use of antibiotics is inappropriate, even, Acute Respiratory Infection may also sometimes be of viral origin having no indication for antibiotic prescription. This will ultimately contribute to irrational use of drugs and ultimately antimicrobial resistance (43,44,45,46). Also according to ICMR guidelines, antibiotics should not be used for viral respiratory infections and watery diarrhoea and their use should be limited to Streptococcal pharyngitis, bacterial sinusitis and diarrhoea caused due to cholera, amoebiosis, Giarrdiasis, Shigellosis and those caused by Campylobacter or Aeromonas (26). The Guidelines issued by State of West Bengal in 2011 also inhibits inadvertent empirical use of antibiotics (13). However, our observations are not in adherence with those guidelines.
The most commonly used antibiotic for respiratory infection was a combination of Amoxycillin and Clavulinic Acid which was corroborated by other studies (47,48). The most commonly used antibiotics for diarrhoea was Metronidazole alone or with Cipro oxacin. However, studies by Panchal et al (28) and Maniar M et al (49) reported 3rd generation cephalosporin to be the most common antibiotic used in diarrhoea whereas Sharma S et al (22) reported uroquinolones such as Nor oxacin alone or in combination with a nitroimidiazole to be the most frequent therapeutic choice. The easy availability of metronidazole and cipro oxacin combined with prescriber`s inclination towards a broad spectrum to eliminate the possibility of mixed infection may drive such type of prescriptions. Azithromycin has been recommended by ICMR for patients with penicillin allergy whereas cephalosporins have been used as alternatives to penicillin (26). The state guideline also recommends use of Cotrimoxazole in paediatric population (13).
Deviations from the available treatment guidelines were found in 98.9% prescriptions, with 90.3% being unacceptable deviations. The unacceptable deviations were in the form of preventable ADR, documentation error or drugs prescribed for which rationality could not be explained. Some of these deviations are known to cause serious side effects, while the others, though not cause any signi cant harm, do not follow the standard prescription guidelines as a result of which the treatment may be ineffective or irrational. A study done at outpatient clinics of Saudi Arabia reported omissions of various components of the treatment regimen, with some even up to 91% incompleteness (50). Higher adherence to guidelines will actually lead to treatment regimen completion possibly because of the institutional culture of emphasizing on the treatment regimen prescription writing. (51) In conclusion, the pattern of prescriptions revealed inappropriate practices in the form of multiple drugs, use of brand names, prescribing xed dose combinations and overuse of antibiotics without any rationale and not adhering to the available guidelines (ICMR, State, WHO). Though the guidelines have overlaps, they are not all the same and also do not include all the possible treatment options creating a scope for physician to depend on his/her experience/expertise. The available guidelines are more suitable to the resource constrain primary care settings where simpler cases are expected to be managed, our study re ected the need of having level of healthcare speci c treatment guidelines for curtailing the subjective approach. Apart from this, adequate training of various categories of physician is also required to ensure rationality of drug usage as well as quality of care. The evidence generated from this study will help to assess the practice of physicians in rational use of drugs and gap identi ed thus may be adequately addressed through structured capacity building, robust documentation and monitoring system and nally strengthening the antimicrobial stewardship programme.

Availability of Data and Materials
The data will be available from the corresponding author on request.

Competing Interests
The authors declare that they have no competing interests.      Acceptability of the deviations in assessed prescriptions through a consensus committee approach