Indian Journal of Public Health

: 2014  |  Volume : 58  |  Issue : 3  |  Page : 180--185

Prescribing behavior of diabetes treating physicians in selected health care facilities of the Diabetic Association of Bangladesh

Bilkis Banu1, Md Shah Jalalur Rahman Shahi2, Kausara Begum3, Tofail Ahmed4, Hasan Ali Choudhury5, Liaquat Ali6,  
1 Lecturer, Department of Health Promotion and Health Education, Bangladesh Institute of Health Sciences (BIHS), Dhaka, Bangladesh
2 Lecturer, Department of Immunology, Bangladesh Institute of Health Sciences (BIHS), Dhaka, Bangladesh
3 Sr Executive Officer, Department of Administration, Bangladesh Institute of Health Sciences (BIHS), Dhaka, Bangladesh
4 Professor, Department of Endocrinology, BIRDEM, Dhaka, Bangladesh
5 Additional Coordinator, Department of Distance Learning Program and EDC, Diabetic Association of Bangladesh, Dhaka, Bangladesh
6 Professor and Head, Department of Biochemistry and Cell Biology, BIHS, Dhaka, Bangladesh

Correspondence Address:
Bilkis Banu
Lecturer, Department of Health Promotion an Health Education, Bangladesh Institute of Health Sciences, 125/1, Darus Salam, Mirpur, Dhaka - 1216


Background: Practicing behavior of the physicians varies from population to population due to diverse socioeconomic, cultural, and professional factors. Evidence on these issues is almost nonexistent in the developing countries. Objective: The prescribing behavior of diabetes treating physicians working in selected hospitals of the Diabetic Association of Bangladesh was studied along with the factors affecting those behaviors. Materials and Methods: This was an observational study on 818 prescriptions given by 49 physicians working in 16 health care facilities, which were photocopied by a portable photocopier. The various components of the prescription were scrutinized for presence and absence, and evaluated independently by two expert Diabetologists for their qualitative aspects. Results: The mean ± standard deviation of the total prescribing score (expressed as percentage) was 60 ± 11. Physicians scoring around or below 60% belonged more to lower age (<40 years), less experienced (<7 years) and mid-position (Senior Medical Officers) groups. Most of them also had public medical college background. Physicians with Certificate Course on Diabetology (CCD) had significantly higher score compared with the Non-CCD group (P < 0.001). Direction and duration of drug use were absent in majority of prescriptions (72.0% and 61.6%), respectively. Symptoms were not written in 78.0% and the family histories were not recorded in 98.5% prescriptions. Diet (49.4%) and exercise (51.0%) related advices were not mentioned in a large number of prescriptions. Appropriate change of drug (78.2%) and proper use of drug (99.1%) and brand (93.8%) were found rational, but still, 22.4% of the prescriptions found illegible. Conclusion: A large proportion of prescriptions in Bangladesh related to diabetes care still lack standardization and acceptable quality. Nondrug related issues (such as history, symptoms, and dietary/exercise-related advices) are the most neglected ones in a prescription.

How to cite this article:
Banu B, Shahi MJ, Begum K, Ahmed T, Choudhury HA, Ali L. Prescribing behavior of diabetes treating physicians in selected health care facilities of the Diabetic Association of Bangladesh.Indian J Public Health 2014;58:180-185

How to cite this URL:
Banu B, Shahi MJ, Begum K, Ahmed T, Choudhury HA, Ali L. Prescribing behavior of diabetes treating physicians in selected health care facilities of the Diabetic Association of Bangladesh. Indian J Public Health [serial online] 2014 [cited 2022 Sep 26 ];58:180-185
Available from:

Full Text


A rational, evidence-based, clear, complete and outcome-focused prescription, with consideration of the socioeconomic reality of the patient, is a prerequisite for good quality health care delivery. It also reflects the technical and moral standard of the physician community in general. The prescribing behaviors of the physicians are influenced by a large number of sociodemographic, educational, economic and commercial factors. Hence, the prescribing patterns and their modifying factors need to be studied in individual populations. So far, such studies are scanty, and even the conducted ones have concentrated mostly on drug issues. Although drugs are central for a prescription there are many other aspects (such as patient's name, age, weight, sex, address, sign and symptoms of disease, family history, and dietary and physical activity related advice) are also important for the management of a patient.

Studies related to prescribing behavior (mostly related to drugs) have already revealed substantial inadequacies and irrationalities in most of the societies. Different aspects of the prescribing patterns have been studied in different countries like India, [1],[2],[3],[4],[5] Pakistan, [6],[7] Nepal, [8] and Sri Lanka. [9],[10] Field studies carried out in these populations have highlighted a significant degree of inappropriate use of drugs in their health care facilities. [11] In a French longitudinal study of 9294 subjects aged >65 years, nearly 40% of the participants used at least one potentially inappropriate medication. [12] In United Kingdom hospitals, prescribers make errors in 1.5% of prescriptions; [13] and in primary care errors occur in up to 11% of prescriptions. [14] Communication of prescribing information between the two sectors is also less than ideal: In one study, around half of the patients were failing to take the right medicine, correctly, a month after discharge. [15] Nonadherence, in part a consequence of poor prescribing, affects 30-50% of patients taking medications for chronic conditions. [16],[17] It is particularly acute in developing countries due to lack of proper regulatory activities.

There are several factors, which can influence and in turn, cause a modification to the prescribing behavior of physicians. Such factors may be the age and sex of the prescriber, under- and postgraduate education and the experience of the physician, various social factors, the number of practitioners in a practice and others. [18],[19],[20],[21] One study done in Nigeria found several factors (like in service education, prequalification training and influence of industry) influencing the prescribing practice. [22]

Bangladesh is one of the rapidly progressing developing countries, and technical values and moral standards are fast changing in such a transition society. Diabetic Association of Bangladesh (BADAS) is the second largest comprehensive (primary, secondary, tertiary) health care provider in the country next to the public sector and runs as a parallel health care provider of government health system. Physicians belonging to this organization are perceived to provide good quality care to the patients. In the present study, the prescribing behavior of a group of BADAS physicians have been studied along with the factors, which may affect their behaviors.

 Materials and Methods

Under an observational design the study was conducted on 818 purposively selected prescriptions generated by all (49) diabetes treating physicians working in 16 health care facilities of the Health Care Development Project of BADAS over a period of 2 years during 2008-2010. Of the 16, 11 facilities (Bangladesh Institute of Health Sciences and Hospital, Bashabo Swasthoseba Kendra, Banani Swasthoseba Kendra, Dhanmondi Swasthoseba Kendra, Jurain Swasthoseba Kendra, Keraniganj Swasthoseba Kendra, Rampura Swasthoseba Kendra, Savar Swasthoseba Kendra, Tongi Swasthoseba Kendra, Uttara Swasthoseba Kendra and Motijheel Swasthoseba Kendra) are at capital, three (Dinajpur Diabetes O Swasthoseba Hospital, Thakurgaon Swasthoseba Hospital, and Pabna Swasthoseba Hospital) at district levels and two (Khalashuddin Chowdhury Swasthoseba Kendra, and Ranishankail Swasthoseba Kendra) are located at the upazilla/sub-district levels. The upazilla facilities have primary and secondary care facilities, all others have primary to tertiary care facilities.

Following an exit survey technique each prescription was photocopied by using a portable photocopier and socioeconomic information of prescribers obtained by collecting their curriculum vitae from the Human Resource Department of the organization. Presence and absence of individual components in each prescription were entered in a customized Excel based data collection sheet. The targeted components included Prescriber's identity, superscription denoted by "Rx", inscription (number, name, formulation such as tablet, suspension, capsule or syrup and dose of drugs used), subscription (directions regarding dosage forms and total amount of drug), transcription or signature identification of patient, date of issuing the prescription, additional components (patient's sign and symptoms, family and disease history, diet, exercise and other advices and date for coming to follow-up). Presence of each component was scored as 1 with absence as 0.

After the descriptive input on the presence and absence of components the quality of each prescription were anonymously reviewed by two independent expert reviewers well acknowledged in Bangladesh. Categorization was based on the averages of scores given by the two independent reviewers. The evaluation was conducted under four headings: Rational change of drug, irrational use of drug, irrational brand change, legibility of prescription.

In each heading a scoring scale of 0-100 was used.

The presence of individual components was analyzed in terms of prescriptions. The qualitative aspects of the prescription, under the four separate headings, were categorized as rational and not rational and were expressed as proportions. The scores on prescribing components were expressed as mean ± standard deviation (M ± SD) and the difference of scores between two groups were determined by independent sample t-test or one-way ANOVA as appropriate. Association between prescriber's background and total score of prescribing components was explored by multiple logistic regressions. All statistical analysis was performed with the Statistical Package for Social Science (SPSS) 11.5 version.


The study sample consisted of 818 valid prescriptions that met inclusion criteria from 49 diabetes treating physicians. The M ± SD of the overall score (expressed as percentage) in total number of prescription was 60 ± 11.

Sociodemographic information of prescribers

Majority of prescribers were <40 years of age (73.5%) and completed MBBS/BDS with Certificate Course on Diabetology (CCD) (81.6%) from Government Medical Colleges (91.8%) in opposition to 40+ years of age (26.5%) with Non-CCD (18.4%) from private medical colleges (8.2%). The M ± SD of age was 39 ± 10. Most of the prescribers were Senior Medical Officer (SMO) (49.0%) and <7 years experienced (63.3%) in against to MO (36.7%) and Consultant (14.3%) with more than 7 years experienced (36.7%). The M ± SD of total service experience was 9 ± 9.

Prescribing pattern

In the study, the proportion of different components presence in the total 818 no of prescriptions. Prescriber's name was mentioned in almost all (99%) prescriptions. Some important components of a prescription are drug, formulation, dose and frequency of drug use which were present in large proportion (94%, 93.6%, 93.6% and 93.8% respectively) of prescriptions. However, other vital components like direction (72.0%) and duration (61.6%) of drug use were absent in majority of prescription. Background of patients which includes name (99.9%), age (72.2%), weight (83.9%), sex (62.1%) and address (65.2%) were mentioned in most number of prescriptions. Prescribing date and recommended follow-up date were also present in large number of prescriptions (98.9% and 88.1%). In 89.9% prescriptions, sign of the disease were recorded but symptoms of the disease were not recorded in 78.0% of prescriptions. Similarly, histories of patient illness were recorded in 64.9% of prescriptions, but in 98.5% prescriptions, patient's family history was missing. Diet (50.6%) and exercise (51.0%) related advices were not mentioned in a large number of prescriptions [Table 1].{Table 1}

Rationality of prescription

Appropriate change of drug (78.2%), proper use of drug (99.1%) and brand change (93.8%) were found rational in most of the prescriptions. A large number of prescriptions (77.6%) were found to be relatively legible among total prescriptions in this study [Table 2].{Table 2}

Factors influences prescribing pattern

The study shows that the total prescribing score was found to be higher amongst the prescribers who were <40 years of age (score, M ± SD, 61 ± 11) than those who were more aged (58 ± 10). Total score of all components of prescription found also to be higher amongst the physicians who were graduated from Government Medical College (60 ± 11) than those who were graduated from Private Medical College (59 ± 14). However, the differences were not statistically significant. Physicians with CCD (6 month Refresher's Course by Distance Learning Technique) had significantly higher (P < 0.001) score (62 ± 9) compared with the Non-CCD group (53 ± 14) [Table 3].{Table 3}

The study reveals that junior physicians (MO) had significantly higher score (M ± SD, 65 ± 7) than SMO (60 ± 10) and consultant (50 ± 15) respectively.

Significant association was found in overall score of prescribing components in the prescription with designation of the prescriber after adjustment of other variables in this study [Table 4].{Table 4}


Prescriptions are mirrors of the professional competency of physicians, and it also reflects their moral standard as well as attitude to the society. Although, the present study is limited only for diabetes-related prescriptions by a particular group of physicians under a specific project of BADAS, it generates valuable indications regarding the trends on this important issue.

In this study, a host of important components are missing in a large proportion of prescriptions. Those include superscription, age, sex, weight, and address of the patient; symptoms; relevant history (family, illness, drug, investigation, obstetric, operational); diagnosis; investigation; direction (including direction of drug use); dietary advice, and advice for exercise. Due to absence of literature on diabetes related prescriptions the data cannot be compared with other studies.

In a study on general (for all diseases) prescriptions by Indian physicians (conducted in 2001), the proportion of missing components were as follows-absence of prescriber's name (27%), age (73%), address (95%), weight (88%), Rx (37%), directions regarding total amount of drug to be dispensed (50%), and duration of drug therapy (39%). [23] On the same components, the corresponding proportions in the present study are as follows: Absence of prescriber's name (1.3%), patient's age (27.8%), sex (62.1%), address (34.8%), weight (16.1%), Rx (56.5%), directions regarding total amount of drug to be dispensed (72%), duration of drug therapy (61.6%). Thus, overall, the present prescriptions seem to have better mentioning of components, but it is worse in some crucial components like direction and duration of drug use, symptoms of disease, family history, dietary and exercise related advices. The overall higher quality should not lead to any complacency as, ideally, all prescriptions should be free of errors as the life of a single patient is completely for herself/himself. Comparing to the standard of a developed country where prescribing error has been shown to be very low (4%), [24] the present prescriptions have an unacceptable level of prescribing error.

Analysis of the factors influencing the prescribing behavior could not show any significant difference in score between the groups except the fact that those who has a CCD (a 6-month distance learning based Continuing Medical Education [CME] Program) have better score compared to the Non-CCD Group (total score, M ± SD, 62 ± 9 vs. 53 ± 14, P < 0.023). On the other hand, MOs and SMOs scored better than the consultants as a group, a finding possibly related to the impact of the CME program, as most of the CCD-group physicians belong to MO and SMO categories. The observation seems paradoxical as the consultants have various postgraduate degrees and diplomas. Simultaneously, it highlights the need of CMEs in updating all groups of physicians in the management of diseases. A study in Nigeria has also led to similar suggestions. [19]

The present prescriptions have been collected from fairly reputed nonprofit private facilities. It may be apprehended that both the components and quality aspects of the prescriptions may be even worse in public sector facilities and profit-motivated private sector hospitals/clinics. The major challenges of this study were to take permission from the hospital authority and to convince patient for photocopy their prescription. Large scale studies are required to generate evidence and design intervention programs in this field.

 Conclusions and Recommendations

The findings in this study lead to the following conclusions: A large proportion of prescriptions in Bangladesh related to diabetes care still lack standardization and acceptable quality; nondrug related issues (like history, symptoms, and dietary/exercise-related advices) are the most neglected ones in a prescription; major factors responsible for prescribing practices include education and designation of prescribers.

Large scale studies targeted to specific diseases should be undertaken covering wider strata of physicians. A task force for regular monitoring of prescriptions may be formed by the BADAS. Regulatory Bodies from the government should undertake organized program for prescription auditing. Targeted CME and training programs should be undertaken to improve the technical and moral standard of physicians.


The study is supported by the ORET Grant of the Government of Netherlands through the Health Care Development Project of the BADAS. We gratefully acknowledge the advice and suggestions of Dr. Jurrien Toonen and Dr. Erik Post from the Royal Tropical Institute of the Netherlands in designing the study.


1Srishyla MV, Krishnamurthy M, Naga Rani MA. Prescription audit in an Indian hospital setting using the DDD (defined daily dose) concept. Indian J Pharmacol 1994;26:23-8.
2Mirza NY, Sagun D, Barna G. Prescribing pattern in a pediatric out-patient department in Gujarat. Bangladesh J Pharmacol 2009;4:39-42.
3Biswas NR, Uppal R, Sharma PL. Perinatal prescribing to indoor patients in Nehru Hospital, PGIMER. Chandigarh J Obset Gynaecol India 1993;43:907-10.
4Gupta N, Sharma D, Garg SK. Auditing of prescriptions to study utilization of antimicrobials in a tertiary hospital. Indian J Pharmacol 1997;29:411-5.
5Biswas NR, Biswas RS, Pal PS, Jain SK, Malhotra SP, Gupta A, et al. Patterns of prescriptions and drug use in two tertiary hospitals in Delhi. Indian J Physiol Pharmacol 2000;44:109-12.
6Hafeez A, Kiani AG, ud Din S, Muhammad W, Butt K, Shah Z, et al. Prescription and dispensing practices in public sector health facilities in Pakistan: Survey report. J Pak Med Assoc 2004;54:187-91.
7Najmi MH, Hafiz RA, Khan I, Fazli FR. Prescribing practices: An overview of three teaching hospitals in Pakistan. J Pak Med Assoc 1998;48:73-7.
8Sarkar C, Das B, Sripathi H. Antimicrobial drug use in dermatology in a teaching hospital in western Nepal. Int J Clin Pract 2002;56:258-60.
9Angunawela II, Tomson GB. Drug prescribing patterns: A study of four institutions in Sri Lanka. Int J Clin Pharmacol Ther Toxicol 1988;26:69-74.
10Ruwan CP, Ranasinghe BD. Pattern of private sector drug prescriptions in Galle: A descriptive cross sectional study Galle. Med J 2005;10:7-9.
11Hogerzeil HV, Bimo, Ross-Degnan D, Laing RO, Ofori-Adjei D, Santoso B, et al. Field tests for rational drug use in twelve developing countries. Lancet 1993;342:1408-10.
12Lechevallier-Michel N, Gautier-Bertrand M, Alpérovitch A, Berr C, Belmin J, Legrain S, et al. Frequency and risk factors of potentially inappropriate medication use in a community-dwelling elderly population: Results from the 3C Study. Eur J Clin Pharmacol 2005;60:813-9.
13Dean B, Schachter M, Vincent C, Barber N. Prescribing errors in hospital inpatients: Their incidence and clinical significance. Qual Saf Health Care 2002;11:340-4.
14Sandars J, Esmail A. The frequency and nature of medical error in primary care: Understanding the diversity across studies. Fam Pract 2003;20:231-6.
15Omori DM, Potyk RP, Kroenke K. The adverse effects of hospitalization on drug regimens. Arch Intern Med 1991;151:1562-4.
16Meichenbaum D, Turk DC. Facilitating Treatment Adherence: A Practitioner's Handbook. New York: Plenum Press; 1987.
17Sackett DL, Snow JC. The magnitude of compliance and non-compliance. In: Haynes RB, Taylor WD, Sackett DL, editors. Compliance in Health Care. Baltimore, London: The John Hopkins University Press; 1979. p. 11-22.
18Coleman JS, Katz E, Menzel H. Medical Innovation: A Diffusion Study. Indianapolis: The Bobbs-Merril Company Inc.; 2000.
19Ryan M, Yule B, Bond C, Taylor RJ. Scottish general practitioners' attitudes and knowledge in respect of prescribing costs. BMJ 1990;300:1316-8.
20Bradley CP. Factors which influence the decision whether or not to prescribe: The dilemma facing general practitioners. Br J Gen Pract 1992;42:454-8.
21Muijrers PE, Grol RP, Sijbrandij J, Janknegt R, Knottnerus JA. Differences in prescribing between GPs: Impact of the cooperation with pharmacists and impact of visits from pharmaceutical industry representatives. Fam Pract 2005;22:624-30.
22Erah PO, Olumide GO. Prescribing practices in two health care facilities in Warri, Southern Nigeria: A comparative study. Trop J Pharm Res 2003;2:175-82.
23Sharma P, Kapoor B, Study of prescribing pattern for rational drug therapy. JK Sci 2003;5:107-9.
24Avery T, Barber N, Ghaleb M, Dean-Franklin B, Armstrong S, Crowe S, et al. Investigating the prevalence and causes of prescribing errors in general practice: The practice study. A report for the GMC. The University of Nottingham. May 2012.