In the present study, 72% of nurses reported that they experienced MAEs. This result is consistent with self-reported study carried out in Jimma 71% (32, 43), AA 67%(43), Iran (70%)(44), Turk 66.7% (45) and Korea 69.6%(46). On the other hand, this result is lower than another study carried out in Ethiopia Gondar 98.1% (30), Jimma 89.9% (47) and 92.6% in Korean nurses during 3 months 17, and higher than the study in Ethiopia in ICU of JUSH was 621 (51.8%)(48), Tigray 62.7%(32), in Nigeria (65%)(40). The difference might be due to variation in the definitions of MAEs or the components counted as MAEs: For example some studies did not consider time, documentation error as MAE while the other study did consider. This is confirmed by a systematic literature review of 45 studies have confirmed the inconsistency in defining medication errors as a result, the prevalence is varied with the range of 2% to 75% (49). The other possible reason for the difference may be due to variation in the study settings (the previous study was conducted in the pediatric ward only)(32, 48) as compared to ours is in overall unit. This may be explained by those nurses working in paediatrics unit were more experienced compared to those working in adult medical-surgical unit. Moreover, the assessment method may also contributed for the variation that is whether the assessment method was observational, self-reported and patient chart review. For example a previous study in Jimma revealed the prevalence of MAE was 71% compared to observational method (97%) (33). Study from Korea also support this (46). This may suggest the need of both methods to understand the difference between perceived and actual experience of MAE. Though the proportion of MAE was vary for each type of error, majority 179 (63.5%) of the MAE in this study was found to be documentation error. This results is consistent with another studies (30, 33). For example, in Egypt, out of the overall MAE the most frequent errors was wrong documentation (90.96%)(50). Another study also revealed documentation error was 85.4% as compared to right time 58.5% and wrong routes 40% (33).
Regarding the associated factors, sex (AOR = 2.75, CI: 1.131, 6.672), age ?45 (AOR=5, CI: 8.131,39.868) nurse (AOR = 4.03, CI: 1.646, 9.863), system (AOR = 4.49, CI: 1.664, 12.130), pharmacy (AOR = 0.05, CI: 0.011, 0.268) (46, 51) and documentation (AOR = 11.18, CI: 2.470, 50.596 ) were factors associated with MAE. As to the statistical association, female participants were around three time (AOR = 2.75, CI: 1.131, 6.672) more likely perceived MAE as compared to male. This is consistent with study carried out in Saudi Arabia (52). Those participants with age ?45 were five times (AOR=5, CI: 8.131,39.868) more likely perceived MAE than those participants with age 20-24. This may be due to the fact that as the number of years increase, the participants perceptions of making an error is increased, although practically they provide safe medication administration. In other words individual who perceived high error, were may give more attention on the actual work in order to prevent the perceived error. Study in Korea showed a higher rate of MAEs experienced compared to the rate of perceived MAEs (46)
Regarding the perceived causes of MAE, the most common reason for MAEs is system reasons (3.3014 ± 1.26491), which was agreed by majority of participants 164 (58.2%). Physician reasons (3.2291 ± 1.07928), nurses’ reason (3.0449 ± .93537), medication packaging reasons (2.8806 ± 1.29250), pharmacy reasons (2.6463 ± 1.14268) and documentation reasons (2.4121 ± 1.06580).This perceptions of MAEs were also supported with statistical association for system reasons (AOR = 4.49, CI: 1.664, 12.130), nurse reasons (AOR = 4.03, CI: 1.646, 9.863), pharmacy reasons (AOR = 0.05, CI: 0.011, 0.268) and documentation reasons (AOR = 11.18, CI: 2.470, 50.596). Though physician reason and packing reasons were not associated with MAE, which need further study. This finding is consistent with another studies (46, 51). This results suggest the needs of all stakeholders’ attention including work environments/system in the preventions of MAE. This is supported by evidence that showed the associations of MAE with professional practice, health care products, procedures, and systems, including prescribing, order communication, product labeling, packaging, and nomenclature, compounding, dispensing, distribution, administration, education, monitoring, and use