An explanatory sequential mixed methods study for assessing medication administration safety in medical wards of a state tertiary hospital in the Republic of Cyprus

Abstract

Background: Medication errors in hospitals are a leading cause of injury and avoidable harm that negatively affect the quality of the care provided. One-third of all medication errors causing harm to hospitalized patients occur in the administration stage of the medication process. For developing targeted interventions to reduce the risk of placing hospitalized patients at risk, it is crucial to first quantify the magnitude and dimensions of the problem and understand risk-related factors, such as working environment conditions, suboptimal problematic procedures drug-related factors and individual-related factors. Objective: To record the prevalence and types of medication administration errors, with an emphasis on errors of omission. Also, to investigate error associated factors such as systematic and person-related factors, during administration of medicines to inpatients and explore nurses’ perceptions of the medication error-related factors. Methods: An explanatory sequential mixed methods design was followed. In particular, the study consisted of two phases: a descriptive observational phase and a focus group phase. In the first phase, the medication process in two medical wards of a tertiary hospital was observed by two observers using a structured observation form, to record the frequency and types of errors. Chi Square, and logistic regression analysis were used to explore associations between errors and potential factors. Subsequently, nurses’ perceptions of medication administration error-related factors were explored in two focus group discussions, to explain and enrich the findings of the quantitative phase. Thematic analysis was employed for analyzing the data collected from the focus groups. Results: A total of 665 drug administrations were observed involving 128 patients and administered by 24 nurses. From these administrations, 2371 errors were detected from which 81.2% were omissions and 18.8% were errors of commission. Omissions in the infection prevention guidelines (46.6%) and in the five rights of medication safety principles (35.8%) were more prevalent. In particular, omitting to hand wash before administering a drug (98.4%), omitting to disinfect the site of injection (37.7%), and omitting to confirm patient’s identity (74.4%) were the three most frequently observed omissions. Documentation errors (13.1%) and handling errors (4.5%) were also detected with lower frequency. Regression analysis showed that the therapeutic class of the drug administered (OR=4.11, 95% CI, 2.65-6.38, p<0.001) and the number of medicines taken per patient (OR=1.57, 95% CI, 1.08-2.27, p=0.04) were the two factors which statistically significantly increased the risk of a higher number of errors being detected. Particularly when a cardiovascular drug was administered, or when twelve or more drugs were prescribed for a patient, the risk of five or more errors being made per administration was increased by approximately 4 and 1.6 times respectively. Furthermore, regression analysis revealed that when the administration was carried out by a nurse with more than twelve years of working experience, the risk of five or more errors being made per administration was increased by approximately 48% than when a nurse with less than 12 years of experience was administering the drug (OR=1.48, 95% CI, 1.02-2.15, p=0.05). Four themes were identified from the analysis of the data collected in focus group discussions: (a) professional practice environment and related factors, (b) person related factors, (c) drug related factors, and (d) processes and procedures. Professional practice environment and related factors was the dominant theme. According to nurses’ perceptions, factors like staffing, interruptions and/or distractions, communication lapses, processes and systems failures, management and leadership issues are associated with medication errors. Moreover, nurse being physically or mentally fatigued, the patient’s condition and patients with polypharmacy or in a severely poor health condition were also perceived by nurses to be medication administration errors associated factors. Conclusions: Medication administration errors is a multifactorial and multidimensional problem that requires collective effort to be minimized, thereby improving patient safety. Taking into account nurses’ perceptions of medication errors can help enlightening the underlying conditions contributing to errors. Errors during drug administration are common in clinical practice, with omissions being the most common type of error. The risk of a higher number of errors being made is increased when a cardiovascular drug is administered or when the number of medicines administered per patient is increased. Nurses’ years of work experience is also related to drug administration errors. Staff’s perceptions of the causes of medication errors, when supplemented with evidence derived from observational studies, can provide a comprehensive picture of the factors that contribute to errors and thus inform and shape targeted interventions for preventing medication errors in hospitals wards.