By Author
  By Title
  By Keywords

January 1997, Volume 47, Issue 1

Prescribing Practices

Paediatric Prescribing in Karachi

S.Q. Nizami  ( Department of Paediatrics, The Aga Khan University, Karachi. )
I.A. Khan  ( Department of Paediatrics, The Aga Khan University, Karachi. )
Z.A. Bhutta  ( Department of Paediatrics, The Aga Khan University, Karachi. )


To assess amount of drug overuse we studied drug prescribing for common childhood problems by 65 general practitioners (GPs) and 29 paediatricians. A total of 2433 encounters between GPs or paediatricians and children under five years of age were observed. The presenting complaints were fever in 18%, cough in 9%, both fever and cough in.21%, vomiting in 20% and diarrhoea in 41% of encounters. Antibacterials were prescribed in 49% of encounters, analgesics and antipyretics in 29%, antiemetics in 8% and inject­ables in 15%. Antidiarrhoeals were prescribed in 41% encounters with children reported to have diarrhoea. Ampicillin and cotrhnoxazole were the two common antibacterials prescribed by both GPs and paedia­tricians. Antibacterials were prescribed in significantly larger number of encounters with GPs than in those with paediatricians. Mean encounter time of patients with GPs was 3.4±2.7 minutes and with paediatricians 9.7±4.1 minutes. (JPMA 47:29,1997).


Irrational and excessive use of drugs is a world-wide problem both in developed and developing countries1-7. Worldproductions of drugs were estirnatedto beoverUS$ 100 billion8 About 79% of these drugs are used in developed world whose population is only 26% of total world population, whereas, only 21% of these drugs are used in developing countries with 74% of total world population8. In Pakistan, total annual expenditure on drugs was US$ 547 million in 1990, of which US$ 120 million were spent on antibiotics alone9.
Although it is extremely difficult to assess irrational and L excessive prescribing of drugs by practitioners, but evaluation of their prescribing practices usingINRUD criteria10 can give an estimate of irrational and excessive prescribing of drugs. In most studies, drug prescribing at public health facilities has been investigated11 ,but rarely at private practitioners level. In Pakistan, public health system including primary care centres to tertiary care hospitals are providing free treatment but private sector is also very active in providing medical care to population both through private hospitals and individual practices. This private sector is not under any legislative or other administrative control to prescribe or dispense drugs. Although it is possible to monitor and control excessive use of drugs through administrative and legislative measures and peer review of prescribing practices at government controlled health facilities and hospitals12 but may not be so at level of private practitioners. In Pakistan. Akhtar (unpublished data) reported high prescribing rate of antibiotics and injections by GPs from Peshawar, but we could not find any other study in literature describing and comparing drug prescribing trends of GPs and pediatricians for children in Pakistan.
Since prescription of unnecessary medications may contribute to childhood morbidity and mortality, we conducted a study to investigate prescribing practices of GPs and Paediatricians in Karachi. Specifically, we investigated the common childhood problems encountered and frequency ofvarious drugs prescribed by them. We also looked at, if there are differences between GPs and paediatricians in pmblems encountered by them and in their drug prescribing behaviour.


The study was done in Karachi during April, 1992 to December, 1992. In absence of any available data describing differences between GPs and paediatricians and using ourbest judgement. we presumed. a difference of 30% in their drug prescribing behaviour. Using computer progra.mrne EPI.Info (version 5.0 CDC/WHO, April 1990), we needed thirty pediatricians and seventy- five GPs in order to show this difference of 30% in their prescribing behaviour at a confidence level of 0.05 and power of 0.8. Considering refusal and/or dropouts, we selected forty one pacdiatricians and ninety-four GPs. Using random number table paediatricians were selected from a list obtained from Pakistan Paediatric Association and GPs were selected from a list of 2000 practitioners obtained from a drug company. This was done because no organization or association maintains any register of practicing GPs and theiraddresses of practices. Ninety-four GPs and 41 paediatricians were selected for observation of their practice, 15 GPs and 11 pediatricians refused to participate in study. Another 13 GPs and one paediatrician refused to allow their practice to be observed alter their initial consent to participate in the study. Thus, practice of only 65 GPs and 29 paediatncians could be observed. Practice of selected practitioners was observed daily for 3-4 hours during peak hours of their practice for 5-6 days over a week. The observers were graduate in sociology from Karachi University and trained by us specially forobservingprescribing practices and recording relevant information for the study. Four observers were trained during one month of pilot study by observing practices of a group of practitioners (not included in this study) and evaluating their daily peiformance and correcting their mistakes. These observers were then allowed to sit in the offices or clinics of practitioners selected for study after obtaining their consent by research medical officer. To reduce the effect of presence of an observer on prescribing practice. these practitioners were told that the study is aimed towards collecting data about magnitude of paediatric problems faced and solved by them. Data collection forms designed for study were seen and accepted by them and data was recorded by observers in their presence without any interference. The information was recorded about the qualification and duration of practice of the observed practitioners, age and sex of child, presence or absence of fever, cough, diarrhoea etc, their duration, other presenting complaints, encounter time and drugs dispensed orprescribed. No questions were asked from practitioner about diagnosis or reason to prescribe dmgs.
The data so obtained was coded, entered into computer and analysed using computer programme EPI-Info version 5.01 (CDC/WHO). Categorical data are presented as frequency distributions and compared by chi-square test and continuous data are presented as means with standard deviations compared using student t-test.


Study population is shown in Table I.

A total of 1639 encounters with 65 GPs and 794 encounters with 29 paediatricians by children under five years of age were observed. Mean number of encounters with GPs was 25±15 and with paediatricians was 27±18. Out of sixty-five GPs. six (9%) had obtained MCPS in nîedicine or other subjects but were practicing as general practitioners, 61% of all practitioners were practicing for niott than 10 years. Mean encounter time with GPs was three minutes and with paediatricians 10 minutes.
Problems presented to these practitioners are shown in Table II.

It is seen that fever with or without any other major complaint was the eomnionest presenting symptom in 59% of all encounters followed by diarrhoea in 41%. Fever alone without cough or diarrhoea was the presenting complaint only in 18% of encounters. The other major complaints were cough in 30% and vomiting in 20% of all encounters. Most children had more than one complaint. Complaints other than quoted above were grouped according to systems involved. Complaints that could not be grouped together, were included in miscellaneous group.
Regarding encounter with febrile children, a significant difference was seen between GPs and paediatricians. Although GPs encountered slightly higher proportion of children with feveralone orboth fever and cough, but duration of illness was significantly higher in encounters with paediatricians (x2=18.9, p<0.001). Frequency of various drugs prescribed is given in Table III.

It is seen that antibacterials were the commonest group of drugs prescribed in 49% encounters followed b paracetamol prescribed in 25% encounters. Arnoxicillin group (including few prescriptions of penicillin and ampicillin) and co-trimoxazole were the commonest antibacterials prescribed both by GPs and paediatricians. Antibacterials were prescribed in significantly less number of encounters and paracetamol, cough and cold synips and iron and vitamins were prescribed in higher number of encounters with paediatricians.


Prescribing of antibacterials in significantly smaller number and prescribing of paracetaniol, other analgesics, antihistamines and cough syrups in higher number of encounters by paediatricians as compared to GPs. indicate that paediatricians are more selective in prescribing antibacterials and prefer to give symptomatic treatment Despite this fact. prescribing of antibacterials in 42% of encounters by paediatricians cannot be described as satisfactory.
Prescribing of analgesics. antipyretics and cough syrups in higher number of encounters with paediatricians is perhaps due to absence of dispensing of drugs from their own clinics. As GPs dispense medicine in the form of ‘mixtures”, they include analgesics and cough remedies into their “mixtures. Though composition of these mixtures was not revealed by most GPs but some GPs informed the observers about main < ingredients of the mixtures. Besides that observers witnessed presence of drugs like paracetamol syrup, cough syrups, aspirin tablets and various other tablets and svmps in large size packings present in the dispensing corner of most GPs. Dispensing of “mixture” prepared in own clinics by mixing crushed tablets and other ingredients by GPs is due to the nature of their practice and perhaps necessary to get adequate remuneration. Unless majority of GPs decide not to dispense medicines and rely upon issuing prescriptions only, the trend of patients’ expectation to get medicine is unlikely to change. Encountering smaller number of patients with fever and cough. but with prolonged duration by paediatricians show parental anxiety. Patients who can afford financially or not getting desired response within a couple of days of treatment by GPs. prefer to consult pediatricians directly. Therefore, to get best medical care for their children within their financial resources, these parents might have preferred to consult paediatricians rather than GPs.
Higher encounter time (10 minutes) withpaticnts canbe another factor for comparatively lower antibacterial prescribing by paediatricians. With a short encounter time (2-3 minutes with GPs), it is not possible to take an adequate history, examine the patient to makeup his/her mind about diagnosis and selection of drugs or to counsel the patient. Hence, it can be obviously speculated that GPs preferred to prescribe and/or dispense drugs rather than to spend time on diagnosing the case and counselling the patient. Higher use of injectables by GPs reflects both financial incentive and a desire to cure the illness as early as possible, as injectable medication has a higher cost.
Since prescribing of drugs is related to diagnosis, it was not possible to evaluate rationality of drugs prescribed foreach encounter. The diagnosis was not available in majority of encounters in our study. For collective evaluation of irrational or excessive use of drugs in a given setting, it is not only unnecessary but also extremely difficult to analyse critically each prescription or each doctor’s practice individually. To overcome this problem, INRUD10 described several indicators for assessing rational use of drugs by practitioners in various health facilities. Some of these are:
1. Percentage of cases receiving antibiotics.
2. Numberof drugs prescribed percase.
3. Percentage of cases receiving injections.
4. Percentage of patients for whom no drugs are prescribed.
5. Percentage of children under five with diarrhoea receiving ORS.
6. Percentage of children under five years receiving antidiar­rhoeal products.
7. Percentage of drugs prescribed in generic form.
8. Average consultation time with a prescriber.
Using these INRUD indicators, studies from various parts of developing world have shown high prescribing rate of antibiotics and injectables and a low encounter time11. But most of these studies have not looked into prescribing for children particularly in private practice by different types of practitioners. In Pakistan. private practice is the back bone of health care system and majority of those patients who can afford financially, prefer to use services of private practitioners. That is why we looked at their prescribing behaviour for childhood diseases encountered by them according to INRUD indicators but could not use all the criteria quoted above. Using these INRUD indicators, prescribing behaviour of both GPs and paediatricians cannot be termed satisfactory due to higher prescribing rate of antibacterials. antidiarrhoeals and injectables, low encounter time and lack of use of generic names. In Pakistan drugs are registered and sold under brand names and not under generic name. It was also interesting that there was no patient who was not prescribed or not given any drug.
Although our study shows a higher trend for prescribing antibacterials and other drugs but we have not looked into various factors responsible for this prescribing behaviour. It can be speculated that lack of knowledge, competition with fellow practitioners, financial impact on practitioners, detailing by medical representatives of drug companies etc are responsible for irrational prescribing. Though various studies14,15 done in other countries have found one or more of the above factors responsible for irrational prescribing but further studies are needed to determine their role in irratiohal and excessive drug prescribing in Pakistan. This is also necessary for planning any intervention strategy to improve and promote rational prescribing of drugs.


Financial support for this research was provided by the Applied Diarrhoeal Disease Research Project at Harvard University through a cooperative agreement with the U.S. Agency for Internadonal Development. The authors gratefully acknowledge the contributions ofDr. James Trostle for assisting in the development of research proposal. We also acknowledge the help and able guidance of Mr. Jonathon Simon of ADDR at all stages of the study. without whose help and assistance the study would not have been possible.
We are thankful to all those who directly or indirectly helped us in analysing the data and writing this manuscript.
Finally we are thankful to Dr Syed Mairajuddin Shah and ourfield workers (Ms. Tayyaba Hussaini. Mr. Moharmnad Asif Aslam, Mr. Haseeb Fakih and Mr. Fahim Ahmed) for their long hours of field work and Mt Ismail Rehmani for his secretarial assistance.


1. Thamiikitkul. Antibiotic dispensing by drug store personnel in Bangkok, Thailand. J. Antimicrobial Chemother., 1988:21:125-131.
2. Tomson, 0. and Sterky, G. Sail prescribing by way ofpharmaciea inthree Asian developing countrica. Lancet, l986;ii:620-22.
3. Wolf-Gould, C.S., Taylor, N.,Horwitz, S.M. etal. Misinfonnationaboutmedication in rural Ghana (improper use of druga). Soc. Sci. Med.. 1991;33:83-89.
4. Hoaaain, MM.,Glass, MR. and Khan, MR. Antibiotic uae in a rural community in Bangladesh. list. 2. Epidemiol., 1982; 11:402-405.
5. Kunin, CM., Lipton, H.L., Tupasi, T. et al. Social, behavioral and practical factors affecting antibiotics use worldwide: Report of task force 4. Rev. Infect. Dia., 1987;9:S270-284.
6. Sacha, L. and Tomaon, 0. Medicine and culture a double perspective on drug utilization in a developing country. Soc. Sci, Med., 1992;34:307-31-5.
7. Fabricant, S.J. and Hirachhon, N. Deranged distribution, perverse preacription, unprotected use: The irrationality of pharmaceuticals in the developing world. Health Policy Plan., 1987;2:294-213-3.
8. WHO: National drug policy and rational drug use: A model curriculum, July 1985. DAP/85.6 WHO, 1985, Module ‘pp. 2-3.
9. Richardson, R. Use and abuses of antibiotics. The Networks Newsletter, 1992;1 3-5.
10. Special section: Methods for INR UD indicators. Revised list of proposed indicators. INRUD News, 1991 ;2:9.
11. Hogerzeil, H. V., Bimo, Ross-Degnan, D., Laing, R.O. Field test for rational drug uae-in twelve developing countries.Lancet, 1993342:1408-10.
12. Soumersi, S., Quick. J , Avorn, 2. et al. Changing the unchangeable: Principlet and experiences in improving prescribing accuracy. World Psediatrics and Child Care, 1984;62:447-74.
13. Hemminki, E. Review oflitersture on the factors affectingdrug prescribing. Soc. Sci.Med., 1975;9:111-115.
14. Avom, J., Chen, M. and Hartley, R. Scientific versus Commercial sources of influence on the prescribing behaviour ofphysieians. Am. J,Med., 1982;73:4-8.
15. Schwartz; R.K., Soumersi, SB. and Avom, 3. Physician motivatioot for non-scientific drug prescribing. Soc. Sci. Med., 1989;28:577-582.

Journal of the Pakistan Medical Association has agreed to receive and publish manuscripts in accordance with the principles of the following committees: