- Email: [email protected]
Evaluation of drug utilization and prescribing errors in infants: A primary care prescription-based study
Health Policy 81 (2007) 350–357
Evaluation of drug utilization and prescribing errors in infants: A primary care prescription-based study Khalid A.J. Al Khaja a,∗ , Thuraya M. Al Ansari b , Awatif H.H. Damanhori b , Reginald P. Sequeira a a
Department of Pharmacology & Therapeutics, Arabian Gulf University, P.O. Box 22979, Bahrain b Primary Care, Ministry of Health, Bahrain
Abstract The purpose of this study is to evaluate the drug utilization trends and to describe the prevalence and type of medicationrelated prescribing errors in infants treated at primary care health centers in Bahrain. Prescriptions issued for infants were collected over a 2-week period in May 2004 from 20 health centers. Prescribing errors were classified as omission (minor and major), commission (incorrect information) and integration errors. Medications were classified according to the British National Formulary. In infants with a mean age of 6.5 months (±3.1) drugs per prescription were 2.52 (±1.1). Paracetamol and sodium chloride nasal drops were the topmost prescribed systemic and topical drugs, respectively. In 2282 prescriptions, 2066 (90.5%) were with omission (major), commission, and integration errors. In 54.1% of prescriptions with omission errors, length of therapy was not specified in 27.7%, and in 12.8% the dosage form was not stated. In 43.5% of prescriptions with errors of commission, dosing frequency (20.8%) and dose/strength (17.7%)—related errors were most common. Errors of integration such as potential drug–drug interaction comprised 2.4% of all prescribing errors. The proportion of drugs prescribed irrationally were: contraindicated medications, notably chlorpheniramine, promethazine, and corticosteroids (16.1%); medications prescribed on a p.r.n. basis (13.3%); missed information regarding strength of medications (2.8%); medications prescribed over extended periods (2.7%); low dosing frequency (2.6%); supratherapeutic doses (2.3%); excessive dosing frequency (0.8%). Irrational drug therapy in infants, with prescribing errors were apparent in primary care practice, which may be related to a lack of drug information, pharmacovigilance programme, and nonadherence to basic principles of prescribing. Establishing a national drug policy and pharmacovigilance programme for promoting rational drug use are to be considered. There is also a need to evaluate the effectiveness of interventions by measuring the outcomes. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Drug utilization audit; Prescribing errors; Infants; Primary care; Bahrain
1. Introduction ∗ Corresponding author. Tel.: +973 39644642; fax: +973 17271090. E-mail address: [email protected] (K.A.J. Al Khaja).
Infants and children are at a high risk for medication errors [1,2]. This may be attributed to several factors including the need for weight-based dosing, dilution
0168-8510/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.healthpol.2006.07.001
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357
of stock medicine and the vulnerability of infants, particularly those with immature renal and hepatic functions [1]. Although medication errors are a major cause of morbidity and mortality among hospitalized children [3], few studies have addressed the problem of medication errors and adverse drug events in the pediatric population [1,2]. Medication errors can happen at all stages of therapeutic process—prescribing, dispensing, dosage calculation, medication administration and patient compliance [4,5]. Most data concerning prescribing errors in pediatric practice were hospital based, although pediatric patients in such setting represent only a small proportion of population at risk for adverse drug events [2]. Studies on the prevalence and type of medication errors in pediatric ambulatory care are rare, even though the bulk of pediatric services including well-child care, and acute and chronic illness care are in ambulatory setting [2]. A third of all patients admitted to a general pediatric medical or surgical wards were reported to have received one or more unlicensed or off-label drugs during their hospital stay [6]. Emphasis on rational use of medications in the pediatric age group (infants and children) could attenuate the risk and enhance the quality of care [2]. We have conducted a nationwide, primary care-based study in infants to: (a) determine the trends in drug utilization including off-label drug prescribing and (b) to describe the prevalence, and the type of medication-related prescribing errors.
2. Methods 2.1. Setting The Kingdom of Bahrain, located in the Arabian Gulf region, has an approximate population of 700,000; of these, 11.4% are children ≤4 years of age [7]. A network of 20 primary health care centers across the country provide free health care, including essential drugs. The number of primary care physicians in each health centers varied between 4 and 11. Patients requiring special investigations and specialist consultations or admission are referred to Salmaniya Medical Complex, a hospital which provides secondary/tertiary care. Drug-prescribing pattern and drug-related prescribing errors in infants (1–12 months old) were analyzed.
351
2.2. Data source and variability Prescriptions dispensed for infants were collected on a daily basis by the pharmacists at 20 health centers between 9 May and 23 May, 2004, in response to a request from the chief pharmacist for primary care. For the purpose of the present study, absence of prescription components such as date of prescription, any parameter of patient’s personal identifiers, physician’s stamp, and/or direction for use are deemed as minor errors of omission [8,9]. Absence, vague, incomplete and/or illegibility of any component of body of the prescription is considered as major errors of omission [8,9]. Incorrectly written component(s) of body of the prescription is considered as an error of commission [10,11]. Errors of integration [11] or knowledge-based errors [12] in prescribing include potential drug-drug interactions or drug allergies which may reflect a failure of the prescriber to integrate information about the patient or drug history. Skill-based errors of prescribing [12] such as illegible handwriting and/or prescriptions with non-official or unconventional abbreviations, were excluded. Prescriptions with non-drug items (syringes, needles, cotton, bandages, eye pads, surgical spirit, elastoplast, etc.) were also excluded. Based on the above definitions of errors and exclusion criteria, prescriptions with errors were identified. Medications were categorized as per the British National Formulary [13]. Before data entry and analysis, prescriptions were audited to confirm and classify errors using the criteria and classification described earlier [8–12]. The percentage of omission (both minor and major), commission, and integration-related errors were calculated. Data were analyzed using Statistical Package for the Social Science (SPSS/PC+, version 9.0). Descriptive statistics as percentage, mean and standard deviation were used to describe study variables.
3. Results A total of 2282 prescriptions were dispensed to infants from 20 health centers during the study period. Paracetamol and sodium chloride 0.9% nasal drops were the most frequently prescribed systemic and topical drugs, respectively. Chlorpheniramine, an antihistamine, ranked the second most frequently prescribed
352
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357
Table 1 The 12 most commonly prescribed systemic and topical medications in infants Systemic medications
Topical medications
Medications
Number
%
Medications
Number
%
Paracetamola Chlorpheniramine Bromhexine Amoxycillin Allospasmin®b Oral rehydration solution Amydramine®c Ferrous sulfate Multi-vitamins Cephalexin Promethazine Erythromycin
1324 683 356 347 232 155 151 149 101 96 85 73
58.0 29.9 15.6 15.2 10.2 6.8 6.6 6.5 4.4 4.2 3.7 3.2
Sodium chloride 0.9% nasal drops Oxytetracycline 1% eye ointment Hydrocortisone 1% cream Calamine lotion & cream Nystatin oral suspension Chloramphenicol eye & ear preparations Daktacort®d cream Zinc oxide cream Otrivin®e & Nasivin®f nasal drops Travocort®g Betnovate®h lotion, cream and ointment Sulphacetamide eye drops
737 120 107 89 72 61 58 54 48 43 37 37
32.3 5.3 4.7 3.9 3.2 2.7 2.5 2.4 2.1 1.9 1.6 1.6
a b c d e f g h
Both by oral (n = 1042) and rectal (n = 282) route. Homatropine methylbromide plus phenobarbital. Diphenhydramine + sodium citrate + menthol. Hydrocortisone plus miconazole. Xylometazoline. Oxymetazoline. Diflucortolone valerate + isoconazole. Betamethasone valerate.
systemic medication, comprising 29.9% of all prescriptions. Topical corticosteroids such as hydrocortisone 1% cream, potent corticosteroid betamethasone valerate (Betnovate® ) cream, and combinations with antiinfectives like Daktacort® (hydrocortisone 1% plus miconazole) and Travacort® (diflucortolone valerate plus isoconazole) cream were prescribed to 10.7% of infants. Table 1 presents the twelve frequently prescribed systemic and topical medications. Of 2282 prescriptions, 2066 (90.5%) were identified to contain major errors of omission, commission, and errors of integration. In 2282 dispensed prescriptions a total of 5745 medications were included. Of these, 4282 (74.5%) contained drug-related errors. The mean number of drugs per prescription was 2.52 ± 1.1 (range 1–7). The mean age of patients was 6.5 ± 3.1 months. The common minor errors of omission in prescriptions were either missing stamp of the prescriber in 175 (7.7%), or not stating patient’s address in 53 (2.3%). The demographic characteristics of patients, and prescription statistics are shown (Table 2). Different types of prescribing errors are summarized in Table 3. Errors of omission accounted for 54.1%; length of therapy/quantity (27.7%) and dosage form (12.8%) were the common omissions. In contrast,
Table 2 Demographic characteristics of infants and prescription statistics Total number of prescriptions Number of prescription containing drug-related errorsa (%) Total number of prescribed drugs Total number of prescribed drugs with errorsa (%) Number of drugs/prescription (mean ± S.D.) (Range)
2282 2066 (90.5) 5745 4282 (74.5) 2.52 ± 1.1 (1–7)
Minor errors of omission Patient’s name Age Sex Date Address (%) Physician’s stamp (%) Direction for useb
– – – – 53 (2.3) 175 (7.7) NA
Patients’ gender (%) Male Female Age of patients in months (mean ± S.D.)
1173 (51.4) 1109 (48.6) 6.5 ± 3.1
NA, not applicable. a Major errors of omission and commission. b The prescription order format does not include a component known as direction for use; however the latter is inferred by pharmacists from the frequency of dosing and the length of therapy/quantity which are classified as major errors of omission.
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357
errors of commission accounted for 43.5%; in 20.8% of prescriptions the dosing frequency was incorrectly written and in 17.7% the dose/strength of medications was incorrectly stated. In 2.4% of all prescribing errors adverse drug–drug interactions (integration errors) were plausible (Table 3). Irrational drug use (i.e., prescribing of an off-label drug/not recommended drug/contraindicated drug) was observed in 16.1% (925 cases) of overall dispensed drugs. Some examples for such a practice include chlorpheniramine 11.9% (683 cases), promethazine 1.4% (85 cases), moderate-to-potent corticosTable 3 Major prescribing errors of omission, commission and integration Prescription components
Frequency
Percent
Total number of prescribed drug items Drugs without errors Drugs with errors
5745
100
1463 4282
25.5 74.5
Errors of omission Dose/strength Dosage form Frequency of dosing Length of therapy/quantity Subtotal
424 979 621 2122 4146
5.5 12.8 8.1 27.7 54.1
Errors of commission Dose/strength Dosage form Frequency of dosing Length of therapy/quantity Subtotal
1354 4 1594 386 3338
17.7 0.05 20.8 5.0 43.5
120 37 8 8
1.6 0.5 0.1 0.1
Errors of integration Antihistaminea + bromhexine Antihistamineb + Allospasmin®c Antihistamineb + antihistaminebs Antihistaminea,b + bromhexine + Allospasmin®d Allospasmin®e + bromhexine Actifed®f + sympathomimetic nasal decongestantsg Subtotal Total a
6 4 183 7667
0.07 0.05 2.4 100
Antimuscarinic property of antihistamines. Sedative property of antihistamines. c Sedative property due to phenobarbital. d Antimuscarinic and sedative property of Allospasmin® . e A fixed dose combination of homatropine methylbromide + phenobarbital. f Triprolidine + pseudoephedrine. g Xylometazoline, oxymetazoline and naphazoline nasal drops. b
353
teroids 2.5% (143 cases), borax–glycerin 0.14% (8 cases) and Antistine–Privine® 0.1% (5 cases). Medications prescribed on pro re nata (p.r.n) basis such as paracetamol (both oral and rectal), antispasmodic Allospasmin® (homatropine methylbromide plus phenobarbital) and others accounted for 13.3% of overall prescribed medications. Medications such as topical nasal decongestants Otrivin® (xylometazoline) and Nasivin® (oxymetazoline); Amydramine® cough and cold preparation, paracetamol suppository, glycerin suppository and Albucid® (sulfacetamide) eye drops were prescribed without specifying the strength, although these products were available in variable strengths and/or sizes (Table 4).
4. Discussion The prescribing of drugs in infants observed in this study raises three pertinent issues: (a) an overzealous prescribing of paracetamol (orally (44.4%), rectally (11.0%) and by dual routes (2.6%)); (b) an occasional tendency to prescribe drugs contraindicated for infants (e.g., chlorpheniramine), and for children less than 2 years of age (e.g., promethazine) and (c) an overuse/misuse of antispasmodic Allospasmin® (Table 1). Paracetamol is generally considered to be safe and pediatric dosage varies between 10 and 15 mg/kg every 4 h as a safe maximal dose [14]. Although toxicity is the likely outcome if more than 150 mg/kg of paracetamol is ingested [15], a dose as low as 20 mg/kg/day over several days can result in liver failure [14,16]. Even at doses as low as 13 mg/kg every 4 h paracetamol may get accumulated after repeated doses over 2–3 days [17]. In addition to overt prescribing of paracetamol, we found that in approximately half of the instances, paracetamol was prescribed on a p.r.n basis without specifying the recommended fourth hourly frequency of dosing and the appropriate length of therapy. Paradoxically, in 2.6% of infants, paracetamol was prescribed on a p.r.n. basis by both oral and rectal routes. The expression ‘as needed’ may be misinterpreted by parents and can result in medication error. In order to avoid toxicity due to paracetamol, direction for use must specify dosage, frequency and length of treatment. Paracetamol prescribing on a p.r.n. basis is considered unacceptable and should be avoided [18].
354
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357
Table 4 Types of irrational prescriptions, and omission and commission errorsa Types of errors
Example
Frequency
Percent
Contraindicated or not recommended medications
Chlorpheniramine (n = 683), promethazine (n = 85), moderate to potent corticosteroidsb (n = 142), boraxglycerin (n = 8), Antistine–Privine® (n = 5) Paracetamol (n = 653), Allospasmin® (n = 103), Others (n = 8) Otrivin® (n = 17), Nasivin® (n = 7), Amydramine® (n = 84), glycerin suppositories (n = 13), Albucid® (n = 5), paracetamol suppositories (n = 32) Bromhexine (n = 102), Allospasmin® (n = 38), corticosteroids (n = 13), crotamiton (n = 4) Cephalexin (n = 49), nystatin oral suspension (n = 46), Erythromycin (n = 42), bromhexine (n = 7), amoxycillin (n = 4), Allospasmin® (n = 4) Bromhexine (n = 38), ferrous sulfate (n = 34), multivitamins (n = 30), Allospasmin® (n = 30) Otrivin® (n = 29), Nasivin® (n = 8), multivitamin syrup (n = 10)
925
16.1
764 158
13.3 2.8
157
2.7
152
2.6
132
2.3
47
0.8
Medications prescribed on ‘prn’ basis Unspecified strength where medications available in different strength and/or size Inappropriate (prolonged) length of therapy. Inappropriate (low) dosing frequency
Inappropriate (high) doses Inappropriate (high) dosing frequency a
Total of 5745 drugs prescribed. Betamethasone valerate 0.1%; dexamethasone 0.05%; diflucortolone valerate 0.1%; flumethasone pivalate 0.02%; fluocinolone acetonide 0.025%; prednisolone acetate 0.25%. b
Chlorpheniramine, a first-generation sedativeantihistamine, is more likely to produce paradoxical stimulation in young children than other antihistamines. Notwithstanding the indications, sedative effect and paradoxical excitation limit its use in infants [19]. Promethazine, a phenothiazine antihistamine, is widely used for its antiemetic and sedative properties. The use of promethazine is contraindicated in children less than 2-year-old [20], and its use has been associated with agitation, hallucinations, seizures, dystonias [21], sudden sleep apnea and infant death syndrome [22,23], which are often common in infants [24]. Despite the recommendations to the contrary, chlorpheniramine and promethazine were prescribed to 29.9% and 3.7% of infants, respectively. Although colic is common in the first few months of infancy, antimuscarinic–antispasmodics are no longer considered appropriate because of adverse effects. The widely used dicyclomine, for instance, despite its effectiveness, is contraindicated for treating infantile colic because its use is associated with apnea in infants less than 6-month-old [25]. Nonetheless, we noticed that Allospasmin® was not only excessively prescribed but was also misused, because (a) in almost 50% of prescriptions with Allospasmin® , frequency of dosing and length of therapy were not specified, and it was prescribed on p.r.n. basis; (b) in 16.4% instances
Allospasmin® was prescribed over a prolonged period ranging 7–15 days, although colic in infants is a shortlived symptom and (c) in 12.9% instances it was prescribed at supra-therapeutic dose, which was several fold greater than the recommended dosage (results not shown). As regards topical medications, intranasal sodium chloride 0.9% solution was the foremost used preparation. It may relieve nasal congestion by liquefying the mucus. Sympathomimetic nasal decongestant such as Otrivin® and Nasivin® , were prescribed infrequently; none of these were prescribed to infants less than 3month-old, a finding which conforms with the guidelines [13,26]. Decongestants are of limited value, since tachyphylaxis occurs after regular use, and rebound nasal congestion can occur upon discontinuation [13]. Topical ophthalmic anti-infectives such as chloramphenicol, oxytetracycline and sulfacetamide accounted for 9.6% of prescriptions in our study. Sulfacetamide (10% eye drops) has been approved for use in infants ≥2 months [27,28], whereas the safety and effectiveness of ophthalmic chloramphenicol have not been established [27]. In view of its bone marrow toxicity potential, the use of chloramphenicol in infants is controversial [29]. Subjecting infants to the risks of chloramphenicol cannot be justified while safer alternative anti-infectives are available to treat eye infections [30].
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357
Although oxytetracycline (1% eye ointment) was most extensively prescribed, it has never been listed among topical ophthalmic drugs for pediatric use [27,28]. Other medications prescribed to infants were: gentamicin in 34 cases (1.5%) (not approved); prednisolone acetate in 13 cases (0.6%) (approval not established); Sofradex® in 12 (0.5%) cases (status uncertain) and Neosporin® in 8 cases (0.4%)(not approved) [27]. The extent of absorption of ophthalmic medications into systemic circulation is variable and may result in systemic adverse effects [13]. Transconjunctival drug absorption has been reported to be comparable to intravenous administration [31]. Potential teeth discolouration and bone deformity resulting from ophthalmic tetracycline and sulfonamide-induced hemolysis in glucose-6-phosphate dehydrogenase deficient infants/children, a genetic disease with high prevalence in Bahrain [32], are issues to be considered before prescribing these ophthalmic preparations. Topical corticosteroids are needed to treat atopic and other skin disorders in infants and children. The risk of systemic absorption should always be considered while corticosteroids are applied topically because the occlusive effect of nappies and water-proof paints may increase their percutaneous absorption and adverse effects even with a mild corticosteroid, hydrocortisone 1% cream. Prescribing topical corticosteroids to approximately 11% of infants in primary care, 3.5% of whom were exposed to potent preparations (Travacort® & Betnovate® ) without specifying dosing frequency and duration of therapy in several instances is unjustified, and is an irrational therapeutic practice (Tables 1 and 4). Our study revealed that despite vulnerability of infants to adverse drug events, the prescribing errors, notably errors of commission, exceeded considerably (Tables 3 and 4). Out of 7667 drug-related errors, 54.1% were major errors of omission, whereas 43.5% were errors of commission which are undesirable (Table 3). Among the commission errors, dose/strength of drugs, and frequency of dosing were most conspicuous. The rate of commission errors is substantially higher than that has been reported earlier for all age group [33]. Potential drug–drug interactions (errors of integration) merit further attention (Table 3). Bromhexine, a mucolytic, decreases the viscosity of bronchial secretions, with resultant increase in sputum volume [26]. The antimuscarinic (atropine-like) action
355
of first-generation antihistamines (sedative antihistamines) often produces a dry sensation in the throat and nasal passages and thickening of bronchial secretion [26]. In theory, the effect of bromhexine may be diminished by concomitant administration of antihistamines, particularly those with a marked antimuscarinic property. Such potential interaction, albeit speculative, comprised 1.6% of all prescribing errors. A similar interaction may also result from the concurrent use of bromhexine with homatropine methylbromide, an ingredient of Allospasmin® . Depression of the central nervous system (CNS) is a common adverse effect of first-generation antihistamines. Because of their sedative effect, drugs like chlorpheniramine, diphenhydramine, triprolidine, and promethazine can potentiate: (a) the sedative effect of CNS depressant phenobarbital, a constituent of Allospasmin® and (b) sedation due to other antihistamines used concurrently (Table 3). Sedatives, especially in overdose, are associated with the risk of apnea [34]. As a result of their marked antimuscarinic properties, first generation antihistamines given concurrently can result in: (a) potentiation of the antimuscarinic adverse effects of Allospasmin® , and those of other antihistamines and (b) decrease in the effectiveness of bromhexine resulting from thickening of bronchial secretion (Table 3). Such combinations, with a potential for causing drug interactions were occasionally prescribed. In few instances, oral Actifed® that contains triprolidine (sedative antihistamine) plus pseudoephedrine (a systemic sympathomimetic nasal decongestant) with another topical sympathomimetic nasal decongestants or two topical nasal decongestants were prescribed together. Such practice is not justified since CNS-related [35] and cardiovascular-related [13] adverse effects of such combinations are additive. These are examples of errors of integration in prescribing. These prescribing errors perhaps could be attributed to excessive workload, busy clinics and frequent interruptions during consults and polypharmacy practice [4]. To what extent suboptimal knowledge and skill of drug therapy and a lack of adherence to the basic prescribing tenets (attitude) could have contributed to the magnitude of the problem, is uncertain. Perhaps future studies using a qualitative approach and triangulation techniques should explore in-depth the reasons for prescribing errors in primary care. Once the reasons
356
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357
are better understood, prioritizing the training needs of the prescribers, organizational restructuring of primary care, and other interventions can be planned accordingly. Ultimately the evaluation of the outcomes of intervention undertaken should also be planned. Thus the health policy decisions should take into consideration the multidimensional issues including indicators of drug use, in order to improve the quality of care for the benefit of the community.
5. Conclusion In several instances, irrational therapeutic practice and prescribing errors were apparent in infants treated at primary care. These prescribing errors may be related to the lack of drug information and pharmacovigilance in primary care. A lack of adherence to the basic principles of prescribing could also have contributed to observed errors. We recommended establishing a national drug policy and pharmacovigilance programme for promoting rational drug prescribing in Bahrain. There is also a need to evaluate the effectiveness of interventions by measuring the outcomes.
Acknowledgments We acknowledge the help and assistance given to us by the Ministry of Health, Directorate of Health Centers/Primary Health Care: Dr. Awatif S. Sharaf (Chief of Medical Services for Primary Health Care); Dr. Bahyia Al Assoomi (Deputy Chief of Medical Services); the pharmacists and the pharmacy technicians of the health centers involved. Mr. Moh’d Ghali Rashid (Reference Librarian) for information retrieval, and Mrs. Radha Raghavan for her help in preparing and typing of the manuscript.
References [1] Kaushal R, Jaggi T, Walsh K, Fortescue EB, Bates DW. Pediatric medication errors: what do we know? What gaps remain? Ambulatory Pediatrics 2004;4:73–81. [2] American Academy of Pediatrics. Additional statement. National summit on medical errors and patient safety research. September 2000. Accessed on 12/5/2005. Available from: http://www.quic.gov/summit/aaap.htm.
[3] Koren G. Trends of medication errors in hospitalized children. Journal of Clinical Pharmacology 2002;42:707–10. [4] American Society of Hospital Pharmacy (ASPH). Guidelines on preventing medication errors in hospitals. American Journal of Hospital Pharmacy 1993;50:305–14. [5] National Coordinating Council for Medication Error Reporting and Prevention. About medication errors. Available from: www.nccmerp.org. [6] Turner S, Longworth A, Nunn AJ, Choonara I. Unlicensed drug use on paediatric wards. British Medical Journal 1998;316:343–5. [7] Population Census, Kingdom of Bahrain. Manama: Arabian Printing Press; 2003. [8] Howell RR, Jones KW. Prescription writing errors and markers: the value of knowing the diagnosis. Family Medicine 1993;35:104–6. [9] Shaughnessy AF, Nickel RO. Prescription writing patterns and errors in a family medicine residency program. The Journal of Family Practice 1989;29:290–5. [10] Claesson CB, Burman K, Nilsson LG, Vinge E. Prescription errors detected by Swedish pharmacists. International Journal of Pharmacy Practice 1995;3:151–6. [11] Rupp MT, Schondelmeyer SW, Wilson GT, Krause GT. Documenting prescription errors and pharmacist interventions in community pharmacy practice. American Pharmacy (Nova Scotia) 1988;28:30–7. [12] Ferner RE, Aronson JK. Errors in prescribing, preparing and giving medicines: definitions, classification and prevention. In: Aronson JK, editor. Side effects of drug annual-22. Amsterdam: Elsevier; 1999. p. xxiii–xvi. [13] British National Formulary (BNF) (# 50). London: Pharmaceutical Press; 2005. [14] Miles FK, Kamath R, Dorney SFA, Gaskin KJ, O’Loughlin EV. Accidental paracetamol overdosing and fulminant hepatic failure in children. The Medical Journal of Australia 1999;171:472–5. [15] Kwok KL, Fu YM, Ng DKK. Hepatotoxicity and persistent renal insufficiency after repeated supratherapuetic paracetamol ingestion in a Chinese boy. Hong Kong Medical Journal 2004;16:616–24. [16] Heubi JE, Barbacci MB, Zimmerman HJ. Therapeutic misadventures with acetaminophen: hepatotoxicity after multiple doses in children. Journal of Pediatrics 1998;132: 22–7. [17] Nahata MC, Powell DA, Durrell DE, Miller MA. Acetaminophen accumulation in pediatric patients after repeated therapeutic doses. European Journal of Clinical Pharmacology 1984;27:57–9. [18] Penna AC, Dawson KP, Penna CM. Is prescribing paracetamol “pro re nata” acceptable? Journal of Paediatrics Child Health 1993;29:104–6. [19] WHO Model Formulary 2004. Available from: http://www. gujhealth.gov.in/medi servi/mfindex.htm. [20] FDA Med Watch: Phenergan (promethazine) contraindicated in children <2 years due to fatal respiratory depression. Available from: http://www.fda.gov/medwatch/SAFETY/ 2005/safety05.htm#phenergan.
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357 [21] Hickson GB, Altermeier WA, Clayton EW. Should promethazine in liquid form be available without prescription? Pediatrics 1990;86:221–5. [22] Stanton AN. Sudden infant death syndrome and phenothiazine. Pediatrics 1983;71:986–8. [23] Kahn A, Blum D. Possible role of phenothazine in sudden infant death. Lancet 1979;ii:364–5. [24] Hendeles I. Efficacy and safety of antihistamines and expectorants in non-prescription cough and cold preparations. Pharmacotherapy 1993;13:154–8. [25] Lucassen PL, Assendelft WJ, Gubbles JW, van Eijk JT, van Geldrop WJ, Neven AK. Effectiveness of treatments for infantile colic: systematic review. British Medical Journal 1998;316:1563–9. [26] World Health Organization. Cough and cold remedies for the treatment of acute respiratory infections in young children. Department of Child and Adolescent Health and Development; 2001. WHO/FCH/CAH/01.02. [27] Chung I, Buhr V. Topical ophthalmic drugs and the pediatric patients. Optometry 2000;71:511–8. [28] Coutler RA. Pediatric use of topical ophthalmic drugs. Optometry 2004;75:419–29.
357
[29] Al Khaja KAJ, Sequeira RP, Mathur VS. Trends in ophthalmic antimicrobial utilization pattern in Bahrain between 1993 and 2000: a resurgence of chloramphenicol. International Journal of Clinical Pharmacology Therapy 2003;41: 36–41. [30] Fraunfelder FT, Morgan RL, Yunis AA. Blood dyscrasias and topical ophthalmic chloramphenicol. American Journal of Ophthalmology 1993;115:812–3. [31] Shell JW. Pharmacokinetics of topically applied ophthalmic drugs. Survey of Ophthalmology 1982;26:207–18. [32] Al-Arrayed S, Hafadh N, Amin S, Al-Mukharaq H, Sanad H. Student screening for inherited blood disorders in Bahrain. East Mediterranean Health Journal 2003;9:344–52. [33] Al Khaja KAJ, Al-Ansari TM, Sequeira RP. An evaluation of prescribing errors in primary care in Bahrain. International Journal of Clinical Pharmacology Therapy 2005;43:294– 301. [34] Cockfield, Phenergan, Theralene, Algotrophyl—drugs responsible for the death of new born babies. Official Journal of European Communities 1986; 29:C130/25-6. [35] Soderman P, Sahlberg D, Wiholm BE. CNS reactions to nose drops in small children. Lancet 1984;1:573.
Evaluation of drug utilization and prescribing errors in infants: A primary care prescription-based study Khalid A.J. Al Khaja a,∗ , Thuraya M. Al Ansari b , Awatif H.H. Damanhori b , Reginald P. Sequeira a a
Department of Pharmacology & Therapeutics, Arabian Gulf University, P.O. Box 22979, Bahrain b Primary Care, Ministry of Health, Bahrain
Abstract The purpose of this study is to evaluate the drug utilization trends and to describe the prevalence and type of medicationrelated prescribing errors in infants treated at primary care health centers in Bahrain. Prescriptions issued for infants were collected over a 2-week period in May 2004 from 20 health centers. Prescribing errors were classified as omission (minor and major), commission (incorrect information) and integration errors. Medications were classified according to the British National Formulary. In infants with a mean age of 6.5 months (±3.1) drugs per prescription were 2.52 (±1.1). Paracetamol and sodium chloride nasal drops were the topmost prescribed systemic and topical drugs, respectively. In 2282 prescriptions, 2066 (90.5%) were with omission (major), commission, and integration errors. In 54.1% of prescriptions with omission errors, length of therapy was not specified in 27.7%, and in 12.8% the dosage form was not stated. In 43.5% of prescriptions with errors of commission, dosing frequency (20.8%) and dose/strength (17.7%)—related errors were most common. Errors of integration such as potential drug–drug interaction comprised 2.4% of all prescribing errors. The proportion of drugs prescribed irrationally were: contraindicated medications, notably chlorpheniramine, promethazine, and corticosteroids (16.1%); medications prescribed on a p.r.n. basis (13.3%); missed information regarding strength of medications (2.8%); medications prescribed over extended periods (2.7%); low dosing frequency (2.6%); supratherapeutic doses (2.3%); excessive dosing frequency (0.8%). Irrational drug therapy in infants, with prescribing errors were apparent in primary care practice, which may be related to a lack of drug information, pharmacovigilance programme, and nonadherence to basic principles of prescribing. Establishing a national drug policy and pharmacovigilance programme for promoting rational drug use are to be considered. There is also a need to evaluate the effectiveness of interventions by measuring the outcomes. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Drug utilization audit; Prescribing errors; Infants; Primary care; Bahrain
1. Introduction ∗ Corresponding author. Tel.: +973 39644642; fax: +973 17271090. E-mail address: [email protected] (K.A.J. Al Khaja).
Infants and children are at a high risk for medication errors [1,2]. This may be attributed to several factors including the need for weight-based dosing, dilution
0168-8510/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.healthpol.2006.07.001
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357
of stock medicine and the vulnerability of infants, particularly those with immature renal and hepatic functions [1]. Although medication errors are a major cause of morbidity and mortality among hospitalized children [3], few studies have addressed the problem of medication errors and adverse drug events in the pediatric population [1,2]. Medication errors can happen at all stages of therapeutic process—prescribing, dispensing, dosage calculation, medication administration and patient compliance [4,5]. Most data concerning prescribing errors in pediatric practice were hospital based, although pediatric patients in such setting represent only a small proportion of population at risk for adverse drug events [2]. Studies on the prevalence and type of medication errors in pediatric ambulatory care are rare, even though the bulk of pediatric services including well-child care, and acute and chronic illness care are in ambulatory setting [2]. A third of all patients admitted to a general pediatric medical or surgical wards were reported to have received one or more unlicensed or off-label drugs during their hospital stay [6]. Emphasis on rational use of medications in the pediatric age group (infants and children) could attenuate the risk and enhance the quality of care [2]. We have conducted a nationwide, primary care-based study in infants to: (a) determine the trends in drug utilization including off-label drug prescribing and (b) to describe the prevalence, and the type of medication-related prescribing errors.
2. Methods 2.1. Setting The Kingdom of Bahrain, located in the Arabian Gulf region, has an approximate population of 700,000; of these, 11.4% are children ≤4 years of age [7]. A network of 20 primary health care centers across the country provide free health care, including essential drugs. The number of primary care physicians in each health centers varied between 4 and 11. Patients requiring special investigations and specialist consultations or admission are referred to Salmaniya Medical Complex, a hospital which provides secondary/tertiary care. Drug-prescribing pattern and drug-related prescribing errors in infants (1–12 months old) were analyzed.
351
2.2. Data source and variability Prescriptions dispensed for infants were collected on a daily basis by the pharmacists at 20 health centers between 9 May and 23 May, 2004, in response to a request from the chief pharmacist for primary care. For the purpose of the present study, absence of prescription components such as date of prescription, any parameter of patient’s personal identifiers, physician’s stamp, and/or direction for use are deemed as minor errors of omission [8,9]. Absence, vague, incomplete and/or illegibility of any component of body of the prescription is considered as major errors of omission [8,9]. Incorrectly written component(s) of body of the prescription is considered as an error of commission [10,11]. Errors of integration [11] or knowledge-based errors [12] in prescribing include potential drug-drug interactions or drug allergies which may reflect a failure of the prescriber to integrate information about the patient or drug history. Skill-based errors of prescribing [12] such as illegible handwriting and/or prescriptions with non-official or unconventional abbreviations, were excluded. Prescriptions with non-drug items (syringes, needles, cotton, bandages, eye pads, surgical spirit, elastoplast, etc.) were also excluded. Based on the above definitions of errors and exclusion criteria, prescriptions with errors were identified. Medications were categorized as per the British National Formulary [13]. Before data entry and analysis, prescriptions were audited to confirm and classify errors using the criteria and classification described earlier [8–12]. The percentage of omission (both minor and major), commission, and integration-related errors were calculated. Data were analyzed using Statistical Package for the Social Science (SPSS/PC+, version 9.0). Descriptive statistics as percentage, mean and standard deviation were used to describe study variables.
3. Results A total of 2282 prescriptions were dispensed to infants from 20 health centers during the study period. Paracetamol and sodium chloride 0.9% nasal drops were the most frequently prescribed systemic and topical drugs, respectively. Chlorpheniramine, an antihistamine, ranked the second most frequently prescribed
352
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357
Table 1 The 12 most commonly prescribed systemic and topical medications in infants Systemic medications
Topical medications
Medications
Number
%
Medications
Number
%
Paracetamola Chlorpheniramine Bromhexine Amoxycillin Allospasmin®b Oral rehydration solution Amydramine®c Ferrous sulfate Multi-vitamins Cephalexin Promethazine Erythromycin
1324 683 356 347 232 155 151 149 101 96 85 73
58.0 29.9 15.6 15.2 10.2 6.8 6.6 6.5 4.4 4.2 3.7 3.2
Sodium chloride 0.9% nasal drops Oxytetracycline 1% eye ointment Hydrocortisone 1% cream Calamine lotion & cream Nystatin oral suspension Chloramphenicol eye & ear preparations Daktacort®d cream Zinc oxide cream Otrivin®e & Nasivin®f nasal drops Travocort®g Betnovate®h lotion, cream and ointment Sulphacetamide eye drops
737 120 107 89 72 61 58 54 48 43 37 37
32.3 5.3 4.7 3.9 3.2 2.7 2.5 2.4 2.1 1.9 1.6 1.6
a b c d e f g h
Both by oral (n = 1042) and rectal (n = 282) route. Homatropine methylbromide plus phenobarbital. Diphenhydramine + sodium citrate + menthol. Hydrocortisone plus miconazole. Xylometazoline. Oxymetazoline. Diflucortolone valerate + isoconazole. Betamethasone valerate.
systemic medication, comprising 29.9% of all prescriptions. Topical corticosteroids such as hydrocortisone 1% cream, potent corticosteroid betamethasone valerate (Betnovate® ) cream, and combinations with antiinfectives like Daktacort® (hydrocortisone 1% plus miconazole) and Travacort® (diflucortolone valerate plus isoconazole) cream were prescribed to 10.7% of infants. Table 1 presents the twelve frequently prescribed systemic and topical medications. Of 2282 prescriptions, 2066 (90.5%) were identified to contain major errors of omission, commission, and errors of integration. In 2282 dispensed prescriptions a total of 5745 medications were included. Of these, 4282 (74.5%) contained drug-related errors. The mean number of drugs per prescription was 2.52 ± 1.1 (range 1–7). The mean age of patients was 6.5 ± 3.1 months. The common minor errors of omission in prescriptions were either missing stamp of the prescriber in 175 (7.7%), or not stating patient’s address in 53 (2.3%). The demographic characteristics of patients, and prescription statistics are shown (Table 2). Different types of prescribing errors are summarized in Table 3. Errors of omission accounted for 54.1%; length of therapy/quantity (27.7%) and dosage form (12.8%) were the common omissions. In contrast,
Table 2 Demographic characteristics of infants and prescription statistics Total number of prescriptions Number of prescription containing drug-related errorsa (%) Total number of prescribed drugs Total number of prescribed drugs with errorsa (%) Number of drugs/prescription (mean ± S.D.) (Range)
2282 2066 (90.5) 5745 4282 (74.5) 2.52 ± 1.1 (1–7)
Minor errors of omission Patient’s name Age Sex Date Address (%) Physician’s stamp (%) Direction for useb
– – – – 53 (2.3) 175 (7.7) NA
Patients’ gender (%) Male Female Age of patients in months (mean ± S.D.)
1173 (51.4) 1109 (48.6) 6.5 ± 3.1
NA, not applicable. a Major errors of omission and commission. b The prescription order format does not include a component known as direction for use; however the latter is inferred by pharmacists from the frequency of dosing and the length of therapy/quantity which are classified as major errors of omission.
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357
errors of commission accounted for 43.5%; in 20.8% of prescriptions the dosing frequency was incorrectly written and in 17.7% the dose/strength of medications was incorrectly stated. In 2.4% of all prescribing errors adverse drug–drug interactions (integration errors) were plausible (Table 3). Irrational drug use (i.e., prescribing of an off-label drug/not recommended drug/contraindicated drug) was observed in 16.1% (925 cases) of overall dispensed drugs. Some examples for such a practice include chlorpheniramine 11.9% (683 cases), promethazine 1.4% (85 cases), moderate-to-potent corticosTable 3 Major prescribing errors of omission, commission and integration Prescription components
Frequency
Percent
Total number of prescribed drug items Drugs without errors Drugs with errors
5745
100
1463 4282
25.5 74.5
Errors of omission Dose/strength Dosage form Frequency of dosing Length of therapy/quantity Subtotal
424 979 621 2122 4146
5.5 12.8 8.1 27.7 54.1
Errors of commission Dose/strength Dosage form Frequency of dosing Length of therapy/quantity Subtotal
1354 4 1594 386 3338
17.7 0.05 20.8 5.0 43.5
120 37 8 8
1.6 0.5 0.1 0.1
Errors of integration Antihistaminea + bromhexine Antihistamineb + Allospasmin®c Antihistamineb + antihistaminebs Antihistaminea,b + bromhexine + Allospasmin®d Allospasmin®e + bromhexine Actifed®f + sympathomimetic nasal decongestantsg Subtotal Total a
6 4 183 7667
0.07 0.05 2.4 100
Antimuscarinic property of antihistamines. Sedative property of antihistamines. c Sedative property due to phenobarbital. d Antimuscarinic and sedative property of Allospasmin® . e A fixed dose combination of homatropine methylbromide + phenobarbital. f Triprolidine + pseudoephedrine. g Xylometazoline, oxymetazoline and naphazoline nasal drops. b
353
teroids 2.5% (143 cases), borax–glycerin 0.14% (8 cases) and Antistine–Privine® 0.1% (5 cases). Medications prescribed on pro re nata (p.r.n) basis such as paracetamol (both oral and rectal), antispasmodic Allospasmin® (homatropine methylbromide plus phenobarbital) and others accounted for 13.3% of overall prescribed medications. Medications such as topical nasal decongestants Otrivin® (xylometazoline) and Nasivin® (oxymetazoline); Amydramine® cough and cold preparation, paracetamol suppository, glycerin suppository and Albucid® (sulfacetamide) eye drops were prescribed without specifying the strength, although these products were available in variable strengths and/or sizes (Table 4).
4. Discussion The prescribing of drugs in infants observed in this study raises three pertinent issues: (a) an overzealous prescribing of paracetamol (orally (44.4%), rectally (11.0%) and by dual routes (2.6%)); (b) an occasional tendency to prescribe drugs contraindicated for infants (e.g., chlorpheniramine), and for children less than 2 years of age (e.g., promethazine) and (c) an overuse/misuse of antispasmodic Allospasmin® (Table 1). Paracetamol is generally considered to be safe and pediatric dosage varies between 10 and 15 mg/kg every 4 h as a safe maximal dose [14]. Although toxicity is the likely outcome if more than 150 mg/kg of paracetamol is ingested [15], a dose as low as 20 mg/kg/day over several days can result in liver failure [14,16]. Even at doses as low as 13 mg/kg every 4 h paracetamol may get accumulated after repeated doses over 2–3 days [17]. In addition to overt prescribing of paracetamol, we found that in approximately half of the instances, paracetamol was prescribed on a p.r.n basis without specifying the recommended fourth hourly frequency of dosing and the appropriate length of therapy. Paradoxically, in 2.6% of infants, paracetamol was prescribed on a p.r.n. basis by both oral and rectal routes. The expression ‘as needed’ may be misinterpreted by parents and can result in medication error. In order to avoid toxicity due to paracetamol, direction for use must specify dosage, frequency and length of treatment. Paracetamol prescribing on a p.r.n. basis is considered unacceptable and should be avoided [18].
354
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357
Table 4 Types of irrational prescriptions, and omission and commission errorsa Types of errors
Example
Frequency
Percent
Contraindicated or not recommended medications
Chlorpheniramine (n = 683), promethazine (n = 85), moderate to potent corticosteroidsb (n = 142), boraxglycerin (n = 8), Antistine–Privine® (n = 5) Paracetamol (n = 653), Allospasmin® (n = 103), Others (n = 8) Otrivin® (n = 17), Nasivin® (n = 7), Amydramine® (n = 84), glycerin suppositories (n = 13), Albucid® (n = 5), paracetamol suppositories (n = 32) Bromhexine (n = 102), Allospasmin® (n = 38), corticosteroids (n = 13), crotamiton (n = 4) Cephalexin (n = 49), nystatin oral suspension (n = 46), Erythromycin (n = 42), bromhexine (n = 7), amoxycillin (n = 4), Allospasmin® (n = 4) Bromhexine (n = 38), ferrous sulfate (n = 34), multivitamins (n = 30), Allospasmin® (n = 30) Otrivin® (n = 29), Nasivin® (n = 8), multivitamin syrup (n = 10)
925
16.1
764 158
13.3 2.8
157
2.7
152
2.6
132
2.3
47
0.8
Medications prescribed on ‘prn’ basis Unspecified strength where medications available in different strength and/or size Inappropriate (prolonged) length of therapy. Inappropriate (low) dosing frequency
Inappropriate (high) doses Inappropriate (high) dosing frequency a
Total of 5745 drugs prescribed. Betamethasone valerate 0.1%; dexamethasone 0.05%; diflucortolone valerate 0.1%; flumethasone pivalate 0.02%; fluocinolone acetonide 0.025%; prednisolone acetate 0.25%. b
Chlorpheniramine, a first-generation sedativeantihistamine, is more likely to produce paradoxical stimulation in young children than other antihistamines. Notwithstanding the indications, sedative effect and paradoxical excitation limit its use in infants [19]. Promethazine, a phenothiazine antihistamine, is widely used for its antiemetic and sedative properties. The use of promethazine is contraindicated in children less than 2-year-old [20], and its use has been associated with agitation, hallucinations, seizures, dystonias [21], sudden sleep apnea and infant death syndrome [22,23], which are often common in infants [24]. Despite the recommendations to the contrary, chlorpheniramine and promethazine were prescribed to 29.9% and 3.7% of infants, respectively. Although colic is common in the first few months of infancy, antimuscarinic–antispasmodics are no longer considered appropriate because of adverse effects. The widely used dicyclomine, for instance, despite its effectiveness, is contraindicated for treating infantile colic because its use is associated with apnea in infants less than 6-month-old [25]. Nonetheless, we noticed that Allospasmin® was not only excessively prescribed but was also misused, because (a) in almost 50% of prescriptions with Allospasmin® , frequency of dosing and length of therapy were not specified, and it was prescribed on p.r.n. basis; (b) in 16.4% instances
Allospasmin® was prescribed over a prolonged period ranging 7–15 days, although colic in infants is a shortlived symptom and (c) in 12.9% instances it was prescribed at supra-therapeutic dose, which was several fold greater than the recommended dosage (results not shown). As regards topical medications, intranasal sodium chloride 0.9% solution was the foremost used preparation. It may relieve nasal congestion by liquefying the mucus. Sympathomimetic nasal decongestant such as Otrivin® and Nasivin® , were prescribed infrequently; none of these were prescribed to infants less than 3month-old, a finding which conforms with the guidelines [13,26]. Decongestants are of limited value, since tachyphylaxis occurs after regular use, and rebound nasal congestion can occur upon discontinuation [13]. Topical ophthalmic anti-infectives such as chloramphenicol, oxytetracycline and sulfacetamide accounted for 9.6% of prescriptions in our study. Sulfacetamide (10% eye drops) has been approved for use in infants ≥2 months [27,28], whereas the safety and effectiveness of ophthalmic chloramphenicol have not been established [27]. In view of its bone marrow toxicity potential, the use of chloramphenicol in infants is controversial [29]. Subjecting infants to the risks of chloramphenicol cannot be justified while safer alternative anti-infectives are available to treat eye infections [30].
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357
Although oxytetracycline (1% eye ointment) was most extensively prescribed, it has never been listed among topical ophthalmic drugs for pediatric use [27,28]. Other medications prescribed to infants were: gentamicin in 34 cases (1.5%) (not approved); prednisolone acetate in 13 cases (0.6%) (approval not established); Sofradex® in 12 (0.5%) cases (status uncertain) and Neosporin® in 8 cases (0.4%)(not approved) [27]. The extent of absorption of ophthalmic medications into systemic circulation is variable and may result in systemic adverse effects [13]. Transconjunctival drug absorption has been reported to be comparable to intravenous administration [31]. Potential teeth discolouration and bone deformity resulting from ophthalmic tetracycline and sulfonamide-induced hemolysis in glucose-6-phosphate dehydrogenase deficient infants/children, a genetic disease with high prevalence in Bahrain [32], are issues to be considered before prescribing these ophthalmic preparations. Topical corticosteroids are needed to treat atopic and other skin disorders in infants and children. The risk of systemic absorption should always be considered while corticosteroids are applied topically because the occlusive effect of nappies and water-proof paints may increase their percutaneous absorption and adverse effects even with a mild corticosteroid, hydrocortisone 1% cream. Prescribing topical corticosteroids to approximately 11% of infants in primary care, 3.5% of whom were exposed to potent preparations (Travacort® & Betnovate® ) without specifying dosing frequency and duration of therapy in several instances is unjustified, and is an irrational therapeutic practice (Tables 1 and 4). Our study revealed that despite vulnerability of infants to adverse drug events, the prescribing errors, notably errors of commission, exceeded considerably (Tables 3 and 4). Out of 7667 drug-related errors, 54.1% were major errors of omission, whereas 43.5% were errors of commission which are undesirable (Table 3). Among the commission errors, dose/strength of drugs, and frequency of dosing were most conspicuous. The rate of commission errors is substantially higher than that has been reported earlier for all age group [33]. Potential drug–drug interactions (errors of integration) merit further attention (Table 3). Bromhexine, a mucolytic, decreases the viscosity of bronchial secretions, with resultant increase in sputum volume [26]. The antimuscarinic (atropine-like) action
355
of first-generation antihistamines (sedative antihistamines) often produces a dry sensation in the throat and nasal passages and thickening of bronchial secretion [26]. In theory, the effect of bromhexine may be diminished by concomitant administration of antihistamines, particularly those with a marked antimuscarinic property. Such potential interaction, albeit speculative, comprised 1.6% of all prescribing errors. A similar interaction may also result from the concurrent use of bromhexine with homatropine methylbromide, an ingredient of Allospasmin® . Depression of the central nervous system (CNS) is a common adverse effect of first-generation antihistamines. Because of their sedative effect, drugs like chlorpheniramine, diphenhydramine, triprolidine, and promethazine can potentiate: (a) the sedative effect of CNS depressant phenobarbital, a constituent of Allospasmin® and (b) sedation due to other antihistamines used concurrently (Table 3). Sedatives, especially in overdose, are associated with the risk of apnea [34]. As a result of their marked antimuscarinic properties, first generation antihistamines given concurrently can result in: (a) potentiation of the antimuscarinic adverse effects of Allospasmin® , and those of other antihistamines and (b) decrease in the effectiveness of bromhexine resulting from thickening of bronchial secretion (Table 3). Such combinations, with a potential for causing drug interactions were occasionally prescribed. In few instances, oral Actifed® that contains triprolidine (sedative antihistamine) plus pseudoephedrine (a systemic sympathomimetic nasal decongestant) with another topical sympathomimetic nasal decongestants or two topical nasal decongestants were prescribed together. Such practice is not justified since CNS-related [35] and cardiovascular-related [13] adverse effects of such combinations are additive. These are examples of errors of integration in prescribing. These prescribing errors perhaps could be attributed to excessive workload, busy clinics and frequent interruptions during consults and polypharmacy practice [4]. To what extent suboptimal knowledge and skill of drug therapy and a lack of adherence to the basic prescribing tenets (attitude) could have contributed to the magnitude of the problem, is uncertain. Perhaps future studies using a qualitative approach and triangulation techniques should explore in-depth the reasons for prescribing errors in primary care. Once the reasons
356
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357
are better understood, prioritizing the training needs of the prescribers, organizational restructuring of primary care, and other interventions can be planned accordingly. Ultimately the evaluation of the outcomes of intervention undertaken should also be planned. Thus the health policy decisions should take into consideration the multidimensional issues including indicators of drug use, in order to improve the quality of care for the benefit of the community.
5. Conclusion In several instances, irrational therapeutic practice and prescribing errors were apparent in infants treated at primary care. These prescribing errors may be related to the lack of drug information and pharmacovigilance in primary care. A lack of adherence to the basic principles of prescribing could also have contributed to observed errors. We recommended establishing a national drug policy and pharmacovigilance programme for promoting rational drug prescribing in Bahrain. There is also a need to evaluate the effectiveness of interventions by measuring the outcomes.
Acknowledgments We acknowledge the help and assistance given to us by the Ministry of Health, Directorate of Health Centers/Primary Health Care: Dr. Awatif S. Sharaf (Chief of Medical Services for Primary Health Care); Dr. Bahyia Al Assoomi (Deputy Chief of Medical Services); the pharmacists and the pharmacy technicians of the health centers involved. Mr. Moh’d Ghali Rashid (Reference Librarian) for information retrieval, and Mrs. Radha Raghavan for her help in preparing and typing of the manuscript.
References [1] Kaushal R, Jaggi T, Walsh K, Fortescue EB, Bates DW. Pediatric medication errors: what do we know? What gaps remain? Ambulatory Pediatrics 2004;4:73–81. [2] American Academy of Pediatrics. Additional statement. National summit on medical errors and patient safety research. September 2000. Accessed on 12/5/2005. Available from: http://www.quic.gov/summit/aaap.htm.
[3] Koren G. Trends of medication errors in hospitalized children. Journal of Clinical Pharmacology 2002;42:707–10. [4] American Society of Hospital Pharmacy (ASPH). Guidelines on preventing medication errors in hospitals. American Journal of Hospital Pharmacy 1993;50:305–14. [5] National Coordinating Council for Medication Error Reporting and Prevention. About medication errors. Available from: www.nccmerp.org. [6] Turner S, Longworth A, Nunn AJ, Choonara I. Unlicensed drug use on paediatric wards. British Medical Journal 1998;316:343–5. [7] Population Census, Kingdom of Bahrain. Manama: Arabian Printing Press; 2003. [8] Howell RR, Jones KW. Prescription writing errors and markers: the value of knowing the diagnosis. Family Medicine 1993;35:104–6. [9] Shaughnessy AF, Nickel RO. Prescription writing patterns and errors in a family medicine residency program. The Journal of Family Practice 1989;29:290–5. [10] Claesson CB, Burman K, Nilsson LG, Vinge E. Prescription errors detected by Swedish pharmacists. International Journal of Pharmacy Practice 1995;3:151–6. [11] Rupp MT, Schondelmeyer SW, Wilson GT, Krause GT. Documenting prescription errors and pharmacist interventions in community pharmacy practice. American Pharmacy (Nova Scotia) 1988;28:30–7. [12] Ferner RE, Aronson JK. Errors in prescribing, preparing and giving medicines: definitions, classification and prevention. In: Aronson JK, editor. Side effects of drug annual-22. Amsterdam: Elsevier; 1999. p. xxiii–xvi. [13] British National Formulary (BNF) (# 50). London: Pharmaceutical Press; 2005. [14] Miles FK, Kamath R, Dorney SFA, Gaskin KJ, O’Loughlin EV. Accidental paracetamol overdosing and fulminant hepatic failure in children. The Medical Journal of Australia 1999;171:472–5. [15] Kwok KL, Fu YM, Ng DKK. Hepatotoxicity and persistent renal insufficiency after repeated supratherapuetic paracetamol ingestion in a Chinese boy. Hong Kong Medical Journal 2004;16:616–24. [16] Heubi JE, Barbacci MB, Zimmerman HJ. Therapeutic misadventures with acetaminophen: hepatotoxicity after multiple doses in children. Journal of Pediatrics 1998;132: 22–7. [17] Nahata MC, Powell DA, Durrell DE, Miller MA. Acetaminophen accumulation in pediatric patients after repeated therapeutic doses. European Journal of Clinical Pharmacology 1984;27:57–9. [18] Penna AC, Dawson KP, Penna CM. Is prescribing paracetamol “pro re nata” acceptable? Journal of Paediatrics Child Health 1993;29:104–6. [19] WHO Model Formulary 2004. Available from: http://www. gujhealth.gov.in/medi servi/mfindex.htm. [20] FDA Med Watch: Phenergan (promethazine) contraindicated in children <2 years due to fatal respiratory depression. Available from: http://www.fda.gov/medwatch/SAFETY/ 2005/safety05.htm#phenergan.
K.A.J. Al Khaja et al. / Health Policy 81 (2007) 350–357 [21] Hickson GB, Altermeier WA, Clayton EW. Should promethazine in liquid form be available without prescription? Pediatrics 1990;86:221–5. [22] Stanton AN. Sudden infant death syndrome and phenothiazine. Pediatrics 1983;71:986–8. [23] Kahn A, Blum D. Possible role of phenothazine in sudden infant death. Lancet 1979;ii:364–5. [24] Hendeles I. Efficacy and safety of antihistamines and expectorants in non-prescription cough and cold preparations. Pharmacotherapy 1993;13:154–8. [25] Lucassen PL, Assendelft WJ, Gubbles JW, van Eijk JT, van Geldrop WJ, Neven AK. Effectiveness of treatments for infantile colic: systematic review. British Medical Journal 1998;316:1563–9. [26] World Health Organization. Cough and cold remedies for the treatment of acute respiratory infections in young children. Department of Child and Adolescent Health and Development; 2001. WHO/FCH/CAH/01.02. [27] Chung I, Buhr V. Topical ophthalmic drugs and the pediatric patients. Optometry 2000;71:511–8. [28] Coutler RA. Pediatric use of topical ophthalmic drugs. Optometry 2004;75:419–29.
357
[29] Al Khaja KAJ, Sequeira RP, Mathur VS. Trends in ophthalmic antimicrobial utilization pattern in Bahrain between 1993 and 2000: a resurgence of chloramphenicol. International Journal of Clinical Pharmacology Therapy 2003;41: 36–41. [30] Fraunfelder FT, Morgan RL, Yunis AA. Blood dyscrasias and topical ophthalmic chloramphenicol. American Journal of Ophthalmology 1993;115:812–3. [31] Shell JW. Pharmacokinetics of topically applied ophthalmic drugs. Survey of Ophthalmology 1982;26:207–18. [32] Al-Arrayed S, Hafadh N, Amin S, Al-Mukharaq H, Sanad H. Student screening for inherited blood disorders in Bahrain. East Mediterranean Health Journal 2003;9:344–52. [33] Al Khaja KAJ, Al-Ansari TM, Sequeira RP. An evaluation of prescribing errors in primary care in Bahrain. International Journal of Clinical Pharmacology Therapy 2005;43:294– 301. [34] Cockfield, Phenergan, Theralene, Algotrophyl—drugs responsible for the death of new born babies. Official Journal of European Communities 1986; 29:C130/25-6. [35] Soderman P, Sahlberg D, Wiholm BE. CNS reactions to nose drops in small children. Lancet 1984;1:573.