Selection of Opportunities for Improvement

Oral medication is administered via enteral feeding tube (EFT) in patients with dysphagia, but some are non-crushable or require special handling. From verification study, knowledge on drug administration via EFT was inadequate and percentage of error was 57% (n=138).

Key Measures of Improvement 

The indicator used was total number of medication errors divided by total potential errors multiplied by 100. The standard was set as 0% error.

Process of Gathering Information

A cross-sectional study was conducted using universal sampling to determine the percentage of medication error in four medical and four critical care wards at Hospital Tuanku Ja’afar Seremban (HTJS). Verification study was conducted in October 2017 (n=138), cycle 1 in April 2018 (n=76), and cycle 2 in January 2019 (n=36). Data collection on prescribing and administration errors was done using data collection form to patients who met the inclusion criteria. Contributing factors were identified using questionnaires distributed to healthcare professionals in the selected wards (n=50).

Analysis and Interpretation

Verification study showed 57% medication error in the selected wards, with 23% errors from critical care wards and 34% errors from medical wards. Main contributing factors identified using Pareto chart were insufficient knowledge on drug administration via EFT (100%), in-patient pharmacist not informed on route of administration (68%) and no local protocol (60%).

Strategy for Change

A protocol on administration of oral medication via EFT was developed, consisting of medication that cannot be crushed, medication requiring special handling or administration technique, and medication with no data for administration via EFT. Continuous education was conducted to all pharmacists in HTJS and doctors and nurses in critical care and medical wards. The protocol was updated, officially launched by the hospital director, and disseminated to all medical and critical care wards in HTJS in cycle 2.

Effects of Change

We successfully reduced both prescribing and administration errors. Prescribing errors reduced from 33% (verification) to 18% (cycle1) and 11% (cycle2) while administration errors reduced from 81% (verification) to 29% (cycle1) and 17% (cycle2). The Achievable Benefit Not Achieved (ABNA) improved from 57% (verification) to 24% (cycle1) and finally 14% (cycle2).

The next step

We would like to use color-coded unit-of-dose labels to clearly identify oral medication which are unsuitable to be crushed. In-patient medication chart can be modified to include options to identify route via EFT. This study can also be expanded to all wards in HTJS.