A utilization study was performed in a 2200-bed tertiary care teaching hospital. Data mining was performed on all nasogastric medication prescriptions for patients hospitalized in 2011. Nurses were interviewed by questionnaire. A PDCA (Plan-Do-Check-Act) cycle was used for continuous quality improvement. The proportion of patients with nasogastric tubes (NGT) was 3.2%. A large number of medical orders (n = 6261) involved nasogastric medications with a package insert particularly noting that they should not be crushed or opened (group 1) or medications without a specific formulation recommendation in the package insert but having evidence discouraging NGT dosing (group 2). Of the nasogastrically administered sustained-release or controlled-release formulations, a sustained-release sodium valproate tablet formulation was the most prescribed drug and a sustained-release 2.5 mg felodipine tablet was prescribed with the highest proportion of NGT dosing [NGT/(NGT + oral) = 12.3%]. Among the nasogastrically administered enteric-coated formulations, a myrtol-standardized enteric-coated capsule formulation was the most prescribed drug and a pantoprazole tablet formulation was prescribed with the highest proportion of NGT dosing [NGT/(NGT + oral) = 19.3%]. Proportions of NGT dosing for amiodarone and carbamazepine (group 2) were 4.8% and 6.3%, respectively. The percentage of nurses with adequate knowledge about pharmaceutical dosage formulations was 60%. The rate of answering correctly as to whether medications in group 1 could be crushed or opened was only 30%. Awareness of evidence discouraging NGT dosing of medications in group 2 was zero. Most nurses (90%) left physicians and pharmacists with the entire responsibility for knowledge and decision-making concerning route of drug administration. After a 3-month preliminary intervention, irrational medical orders involving nasogastric administration of medications in group 1 were successfully abolished. The rate of answering correctly as to whether medications in group 1 could be crushed or opened increased to 100%. This utilization study indicates poor awareness concerning nasogastric administration of medication on the part of physicians and nurses, and preliminary intervention measures were efficient in improving knowledge through team cooperation and effort.