1. IntroductionAntibiotic resistance is increasingly becoming a global health threat, with the irrational use of antibiotics being the major contributor to the development of antibiotic resistance [1,2]. According to the World Health Organization (WHO), the rational use of antibiotics refers to the use of an antibiotic that is appropriate for a patients’ clinical condition, with doses that meet their individual needs for an adequate period of time [3]. Antibiotic resistance is associated with increased morbidity, mortality and healthcare costs. It is estimated that antibiotic resistance contributes to the death of 23,000 patients and an additional cost of more than 20$ billion per year in the United States [4]. Antibiotics are commonly prescribed for hospitalized pediatric patients. This is because children are more prone than adults to contracting infections that are the leading cause of hospital admission in this population [5,6]. It was reported that at least half of the hospitalized children in developing countries received antibiotic treatment [7,8,9,10]. This high percentage of antibiotic prescribing is almost always associated with different types of inappropriate antibiotic prescribing (IAP), which includes unnecessary antibiotic use as well as incorrect antibiotic selection, dose or duration of treatment [9,11,12]. IAP may result in ineffective treatment, adverse drug reactions (ADRs), prolongation of hospitalization and unnecessary additions to cost [8,13]. To avoid IAP, the WHO recommends implementing hospital stewardship programs that include using local guidelines for diagnosis and treatment [1]. Consequently, adhering to local/national antibiotic guidelines is one of the key practices to prevent IAP. Studying the pattern of antibiotic prescribing can help in assessing the prescribers’ adherence to antibiotic guidelines. It was found that respiratory tract infections are the most commonly encountered infections in pediatric wards globally [10,11,14]. Subsequently, penicillins and cephalosporins are the top prescribed classes of antibiotics for children [10,11,14]. However, more than one third of these antibiotics were not appropriately prescribed [11,12]. In Malaysia, respiratory tract infections were also reported as the primary driver for caregivers to bring their children to the emergency department [15] and primary care clinics, where more than 35% of the patients received at least one antibiotic [16]. Limited data are available about antibiotic use in pediatric wards in Malaysia. A one-month study that reported the results from a tertiary hospital in Malaysia also found that respiratory tract infections were the leading cause of antibiotic use. However, the appropriateness of the prescribed antibiotics was reported to be 99.8% [17]. It is worth noting here that a review of the Malaysian ADR reporting system revealed that systemic anti-infective agents were the most common therapeutic group reported for ADRs in pediatric patients [18]. To the best of the authors’ knowledge, there are no studies evaluating the appropriateness of antibiotic prescribing among hospitalized pediatric patients in Malaysia. The objective of this study is to assess the indication, prescribing pattern and appropriateness of antibiotics that are prescribed for hospitalized pediatric patients. 4. DiscussionThe literature is lacking in studies evaluating the use of antibiotics in pediatric inpatients. This study assessed the pattern and appropriateness of antibiotics prescribed for hospitalized children in a Malaysian teaching hospital. It was found that 292 out of 702 (41.6%) hospitalized pediatric patients were prescribed with at least one systemic antibiotic. The antibiotic prescribing rate of the current study is relatively lower than that reported from other developing countries such as Turkey (54.6%) [11], Gambia (54.1%) [10], Nigeria (63%) [8], Costa Rica (65%) [24] and India (66%) [25]. No similar study was reported from Malaysia. However, the prescribing rate was comparable (43.5%) to what was reported in children brought to the emergency department of a tertiary Malaysian hospital [15]. The majority of the patients (71%) received one antibiotic, which resembles what has been reported from India (71%) [26] and Nigeria (63%) [8]. The median duration of treatment was seven days, which is similar to the findings elsewhere [10,26]. About two thirds of the patients (68.1%) were diagnosed with respiratory infections, with LRTIs being more prevalent than URTIs. Pneumonia was the most common type of LRTI, while tonsillitis was the most reported URTI. Acute respiratory infections are the leading cause of children’s hospitalization worldwide, including Malaysia [10,11,14,17,27,28]. This is because children have an immature immune system and are usually surrounded by peers who could carry infections [12]. The main approach to treat pneumonia was a monotherapy of co-amoxiclav, cefuroxime or azithromycin or a combination of co-amoxiclav/cefuroxime and azithromycin. Additionally, co-amoxiclav monotherapy was the mainstay treatment used for tonsillitis/pharyngitis. Because of this, azithromycin, co-amoxiclav and cefuroxime were the top three prescribed antibiotics in this study. Penicillins and cephalosporines were repeatedly reported in the literature as the most common prescribed antibiotic classes for hospitalized children [11,12,14,17,26,28]. However, penicillins without a β-lactamase inhibitor and third-generation cephalosporines, namely ceftriaxone, were the predominately prescribed agents [11,12,14,17,26,28]. The difference in the results between our study and the other mentioned studies can be explained by the preference of the prescribers, where they preferred cefuroxime over ceftriaxone for pneumonia and azithromycin over amoxicillin for tonsillitis/pharyngitis, although these agents are considered as the second-line treatment according to the Malaysian NAG [20]. Third-generation cephalosporines were also commonly used inappropriately in those studies. In addition, co-amoxiclav, which was frequently prescribed in our study for patients with tonsillitis/pharyngitis, was used inappropriately, as will be discussed below. Notably, the use of macrolides in our study was far higher than what is reported elsewhere [11,12,14,26,28]. This can be attributed to the recommendations of the NAG, where azithromycin is recommended as a monotherapy for pneumonia suspected to be caused by atypical bacteria, and it is also the alternative therapy for tonsilitis/pharyngitis. Aminoglycosides (AMGs) use was prevalent in Turkey (16.6%) [11], Nigeria (25.4%) [8] and India (20%) [26]. Nevertheless, they were prescribed for only eight patients (2.73%) with sepsis or a urinary tract infection (UTI) in our study. Again, the difference can be explained partly by the NAG’s recommendations, as AMGs are not recommended for respiratory infections. This claim is supported by a similar result reported from another Malaysian study, where AMGs were used in 2.9% of the cases [17]. Additionally, sepsis and UTIs were more prevalent in those studies than ours [11]. Inappropriate antibiotic prescribing was found in 57% of the patients. This incidence rate of IAP is higher than what was found in Turkey (47%) [11], Pakistan (41%) [12] and Ethiopia (28%) [29]. This could be attributed to the lack of established hospital antibiotic guidelines in our teaching hospital and the difference in the types of infections encountered. On the other hand, non-adherence to NAG recommendations was found to be higher among pediatric outpatients in Malaysia (71.8%) [16]. Improper antibiotic selection represents 42% of all types of IAP identified. This was mainly attributed to the prescribing of co-amoxiclav for URTIs despite not being recommended by the NAG. Co-amoxiclav was also inappropriately selected as a perioperative prophylaxis. Unnecessary antibiotic use (antibiotic without indication) was identified in 19 prescriptions, where antibiotics were prescribed for acute gastroenteritis (AGE) and bronchiolitis. However, both conditions are believed to be viral infections that do not need empirical antibiotic use [20,30]. The use of antibiotics for viral infections and AGE was also reported as a common IAP in other countries [9,29]. It is important to note here that no laboratory confirmatory test for diagnosis was reported in the patients’ electronic charts. Thus, unnecessary antibiotic use could be underestimated in our study. This may explain the differences in the results of this study and studies from other countries where IAP was much higher [12]. However, most of the studies did not differentiate between the wrong selection of antibiotic and unnecessary antibiotic prescribing and considered them to be the same [11,28]. One third of the identified IAP in the current study was related to wrong daily dosage. In most of these cases, low doses were used. The use of lower-than-recommended doses were mainly associated with the use of cefuroxime for pneumonia. This finding is in consistent with a previous study conducted in a general pediatric ward in Malaysia, where incorrect dosage represented 31% of the overall identified medication errors [31]. In fact, the scenario was worse among Malaysian pediatric outpatients where wrong antibiotic dosing was identified in 64% of the prescriptions [16]. One possible reason for inappropriate dosing is the inadequate dosing information for pediatric patients that is available in the most commonly used drug reference in Malaysia: the Monthly Index of Medical Specialties (MIMS) [32]. Likewise, inappropriate dosage was also reported in the literature as one of the top encountered types of IAP in pediatric patients [12,28,29], with underdose being the predominant dosing error [12,29]. The duration of treatment was inappropriate in 26 (9%) patients. In 19 out of those 26 patients, the physicians prescribed antibiotics for a shorter-than-recommended duration, which theoretically might lead to a failure of treatment and/or an increase in bacterial resistance. However, growing data show no evidence of these claimed negative outcomes [33,34]. Studies from Pakistan and Ethiopia showed a higher rate of prescribing antibiotics with the wrong duration of treatment [12,29].Incidence of IAP was not associated with the patients’ demographic characteristics or the length of hospital stay. These findings are not consistent with other studies, where advanced age of children was reported by Yehualaw et al. as an independent factor for IAP [29]. However, that study involved children ≤ 18 years old. Iftikhar et al. found that long length of hospital stay was correlated with antibiotic errors in hospitalized children [12]. Nevertheless, that study evaluated antibiotics prescribed for respiratory infections only. We found that the number of antibiotics and the type of infection were significantly associated with the incidence of IAP. The applied logistic regression revealed that prescribing two antibiotics and treatment of URTIs were independent factors for IAP. It is logical that prescribing multiple medications increases the possibility of medication errors compared with monotherapy, and this was supported by findings from pediatric wards [35]. On the other hand, the most common types of IAP in this study were related to antibiotics prescribed for URTIs, which clearly explains why treating URTIs was an independent risk factor for IAP.