Health Care Professionals' Knowledge, Attitude, and Practice towards Adverse Drug Reaction Reporting and Associated Factors at Selected Public Hospitals in Northeast Ethiopia: A Cross-Sectional Study

Background The role of health care professionals among other stakeholders in early detection, assessment, documentation, and reporting as well as preventing suspected adverse reactions is very crucial to mitigate drug-related problems in health facilities. Previous reports from literatures have indicated that adverse drug reaction reporting is highly linked to the knowledge and attitude of the health care professionals. Objective To assess knowledge, attitude, and practice of health care professionals about adverse drug reactions and the associated factors at selected public hospitals in Northeast Ethiopia. Methods A hospital-based quantitative cross-sectional study design was employed. A structured self-administered questionnaire was used to collect data on KAP of selected health care providers by the convenience sampling method. Data were entered into Epi info version 3.5.3 and analyzed using SPSS Version 20. Association between dependent and independent variables was found by using bivariate and multivariate logistic regression analysis where p < 0.05 was considered to be statistically significant. Results Out of 120 questionnaires distributed, 114 respondents filled and returned, giving a 95% response rate. From total, 49 (43%) were nurses, 26 (22.8%) physicians, 17 (14.9%) pharmacy professionals, 12 (10.5%) health officers, and 10 (8.8%) midwives. About 86 (75.44%) study participants had an inadequate knowledge towards ADR reporting, and half of participants failed to report the adverse drug reactions they encountered. But the majority of participants (84, 73.68%) had a favorable attitude towards ADR reporting. Nurses [AOR = 0.069, 95% CI (0.018–0.275)], health officers [AOR = 0.10, 95% CI (0.015–0.647)], and physicians [AOR = 0.14, 95% CI (0.03–0.64)] were found to be less likely to have adequate knowledge on ADR reporting compared to pharmacy professionals. Conclusion Even though the majority of health care professionals had a positive attitude, they had inadequate knowledge and poor practice towards ADR reporting.


Background
Globally, several numbers of drugs are brought into the market everyday; however, the safety of medicines remains to be a major concern for various population groups due to inadequate knowledge. According to the World Health Organization (WHO), an adverse drug reaction (ADR) is any deleterious, inadvertent and unwanted reaction to drugs when used for prevention, diagnosis, or treatment purposes at therapeutic doses or for modification of physiological malfunction which precludes accidental or deliberate overdosage or drug maladministration [1]. ADRs can be predictable and dose related, unpredictable and nondose related, both dose and time related, time related (delayed reactions), withdrawal reactions, and unexpected reactions due to treatment failure [2,3].
Modern medicines have substantially changed the diseases treatment schemes that improve the treatment outcomes in many medical conditions. However, adverse reactions to medicines are a common cause of morbidity, hospital admissions, longer hospital stay, disability, and even mortality [4]. Not only they have a major impact on public health but they also reduce patients' quality of life and impose a significant financial cost in the health care delivery system [5,6]. e history of ADR monitoring has dated back as much as forty years since the milestone of thalidomide disaster that caused phocomelia in thousands of children in many countries [5,7]. Pharmacovigilance is the science and activity related to the detection, assessment, understanding, and prevention of adverse effects or any other possible drug-related problems [3].
Findings from various studies revealed that ADR reporting is linked to the knowledge and attitude of the health care professionals (HCPs). All HCPs in the country need to alert the Ethiopian Food and Drug Administration (EFDA) about suspected adverse drug reactions in order to facilitate immediate and appropriate actions to be taken to prevent or minimize medicine-related injuries for other patients in the future [2,8].
All ADRs ranging from minor to severe reactions should be reported with particular concern to ADRs to new medicines, serious adverse drug reactions, unexpected reactions, and drug interactions which are potentially serious or clinically significant. In addition, uncertainty of the causal relationship between the drug and ADR should not be a reason for not reporting [2,9].
Underreporting of ADRs by HCPs to the concerned body has long been a big challenge in Ethiopia. Search of literature with regard to knowledge, attitude, and practice (KAP) of HCPs towards ADR reporting in the study areas yielded no results; hence, it is found important to conduct this study to investigate the gaps on KAP of HCPs about ADR reporting and identify factors attributable for inadequate knowledge.

Study Area, Design, and Period.
A hospital-based quantitative cross-sectional study design was used among HCPs (nurses, medical doctors, pharmacy professionals, midwives, and health officers) working at Kemisse General Hospital (KGH) and Ataye Primary Hospital (APH), Northeast Ethiopia. KGH is located in Northeast Ethiopia, Oromia Special Zone, Amhara National Regional State, 331 km away from north of Addis Ababa. KGH is the only general hospital in a zone that serves 1.2 million populations. APH is located in the North Shewa zone, Amhara National Regional State, 300 km away from north of Addis Ababa and 31 km away from south of Kemisse. e study was conducted from May 1 to May 30, 2019.

Study Population.
e study population included all physicians, pharmacy professionals, health officers and nurses, midwives who were working in selected hospitals during the study period.

Data Collection Tool and Data Collection Process.
A structured self-administered questionnaire was used. e questionnaire was developed by modifying the tools from other similar studies and the EFDA guidelines [2,6,[10][11][12][13][14]. Information about the sociodemographic characteristics of respondents and detailed questions used to assess knowledge, attitude, and practices of respondents was included. Prior to commencing the study, the questionnaire with attached written consent form was distributed to HCPs. Participation of the respondents was entirely voluntary.

Data Entry, Management, and Statistical Analysis.
Data were entered into Epi info version 3.5.3, cleaned, and then exported to Statistical Package for the Social Sciences (SPSS) version 20 statistical software for analysis. Both descriptive and analytical statistical tests were done. Bivariate and multivariate binary logistic regression analysis was performed to find association between different independent variables and participants' overall knowledge and attitude towards ADR reporting. A p value <0.05 was considered as statistically significant association.
In this survey, knowledge of HCPs about ADR reporting was assessed with twelve questions. Each correct response had a score of 1, and each wrong response had a score of 0. us, the total score varies from 0 to 12 points (Table 1). e overall level of knowledge was categorized by using the mean score. Participants who score more than or equal to the mean score were classified as having adequate knowledge, and scores below the mean were classified as having an inadequate knowledge. e participants' attitude was evaluated by using thirteen questions rated on a threepoint Likert scale such as agree, neutral, and disagree. A response of "agree" was given a score of 3, "neutral" a score of 2, and "disagree" a score of 1, and a score more than or equal to 75% was indicative of a favorable attitude, and a score below 75% would be indicative of unfavorable attitude towards ADR reporting. HCPs' practice was assessed by identifying whether they documented and reported ADRs or not.

Data Quality Control Measures.
After the data collection format was prepared, pretest of the tool was done on 10 HCPs working at Dessie Referral Hospital for any necessary amendments to check its suitability for the actual data collection. And, the data were cleared and checked everyday for completeness and consistency before data processing and analysis.
Adequate knowledge on ADR reporting: those who had a score of more than or equal to a mean score out of 12 questions used to assess knowledge of the respondents Favorable attitude towards ADR reporting: those who had a score of more than or equal to 75% response (score ≥ 29.25 on 3-point Likert scale) of the total 13 questions used to assess attitudes of the respondents Good practice: those who documented and reported the encountered ADRs Inadequate knowledge: those who had less than 6 correct response of the total 12 questions Poor practice: those who did not document and report the encountered ADRs Unfavorable attitude: those who had less than 75% response (score < 29.25) of the total 13 questions

Knowledge and General Awareness of HCPs on ADR
Reporting. Twelve different questions were used to assess knowledge of HCPs on ADR reporting. 100 (87.7%) respondents knew that all drugs available in the market are not safe, and 76 (66.7%) were able to differentiate ADR from overdose toxicities. Only 23 (20.2%) respondents knew the term pharmacovigilance and understood its function. Likewise, 24 (21.1%) and 26 (22.8%) respondents knew the availability of national reporting system and ADR reporting form in Ethiopia, respectively. 35 (30.7%) respondents knew the responsible body that monitor ADRs in Ethiopia. Less than half of respondents (55 (48.2%)) knew that ADR reporting is a professional obligation. 25 (21.9%) respondents replied that the possibility of an ADR should be the first differential diagnosis at all times in patients taking medicines. Moreover, significant proportion of the respondents (93 (81.6%) and 52 (45.6%)) replied that ADRs should be reported only when they are serious and life-threatening and severe and cause disability, respectively (Table 1).
When the general awareness of respondents was assessed, the study has found that more than half of the respondents (64 (56.1%)) were using National Drug Formulary and Standard Treatment Guideline (STG) followed by standard text books (53 (46.5%)) as the main sources of information about ADR. 37 (32.5%) and 32 (28.1%) of the respondents responded that ADRs should be reported to EFDA and Drug and erapeutic Committee (DTC) of the respective health facility, respectively. Prescribing error, dispensing error, life style of the patient, overdose, and nonadherence are the possible factors that predispose a patient to ADR in 92 (80.7%), 87 (76.5%), 86 (75.4%), 79 (69.3%), and 55 (48.2%) respondents, respectively (Table 3).

Attitude of HCPs towards ADR Reporting.
Regarding the attitude of HCPs towards ADR reporting, 100 (87.7%) respondents agreed that ADR reporting should be part of their duty and 87 (76.3%) supported that ADR reporting should be compulsory and 84 (73.7%) agreed that one report of ADR makes a difference. Besides, most respondents 108 (94.7%) and 101 (88.6%) agreed that reporting ADR is important for the public and improves quality of patient care, respectively. 88 (77.2%) health professionals agreed that ADRs should be reported spontaneously at a regular base with 87 (76.3%) emphasizing that there should be certainty for ADRs related to the drug before reporting. Approximately half of respondents (60 (52.6%)) worried about legal problems while ADR reporting (Table 4).

Practice of HCPs regarding Identification, Recording, and Reporting of ADRs.
e present study has found that only a small number of respondents (34 (29.82%)) encountered at least one patient with ADR in the past 12 months of their clinical practice, out of which 24 (70.59%) and 17 (50%) respondents recorded and reported ADRs, respectively. From those who have reported ADRs, 8 (47.06%) respondents reported to hospital and pharmacy department and 5 (29.41%) respondents to EFDA. Although about half of the respondents (59 (51.8%)) preferred yellow card for ADR reporting, most of them did not use it regularly due to unavailability of the form. In addition, 44 (38.60%) HCPs have not been usually given proper advice to their patients on possible adverse effects of drugs they prescribed, dispensed, or administered (Table 5).

HCPs' Reasons for Not Reporting ADRs and Suggested
Strategies to Improve ADR Reporting. On a 5-point Likert scale evaluation of the reasons for not reporting ADRs, respondents agreed that lack of feedback (67 (58.8%)), reporting forms are not available when needed (53 (46.4%)), not knowing where to report (53 (46.4%)), not knowing how to fill and report the report form (47 (41.2%)), other colleagues are not reporting ADR cases (43 (37.7%)), and uncertain that causal association between the drug and ADR (41 (35.9%)) were the leading discouraging factors contributing to underreporting in hospitals (Table 6). e most common strategic approaches suggested by the respondents to foster ADR reporting were availability of ADR information sheets at OPD (92 (80.7%)), encouraging all health professionals to report (86 (75.4%)), training to report ADR (83 (72.8%)), encouraging patients to report (76 (66.7%)), drug information center assistance (76 (66.7%)), and easy accessibility to ADR forms (68 (59.6%)) ( Figure 1).

Factors Associated with HCPs' Knowledge of ADR
Reporting. Bivariate logistic regression analysis was run to identify any association between different independent variables and knowledge of HCPs on ADR reporting. Accordingly, attending training on ADR reporting and type of profession were found to have statistically significant association with knowledge.  to look into associations controlling any potential confounders. Accordingly, only the type of profession was found to have statistically significant association with knowledge. Nurses, health officers, and physicians were 93.1% [AOR � 0.069, 95% CI � 0.018-0.275], 90% [AOR � 0.10, 95% CI � 0.015-0.647], and 86% [AOR � 0.14, 95% CI � 0.03-0.64] times less likely to have adequate knowledge on ADR reporting compared with pharmacy professionals,

Discussion
Since ADRs are an important cause of morbidity and mortality and increased health care costs, all HCPs should be alert and keen towards any unexpected or suspected reactions occurring in patients taking medicines, assessing, managing, and reporting the encountered adverse events, which are an integral part of the pharmaceutical care process [2,8,15].
In the current study, 66.67% respondents said ADR is different from overdose toxicities or side effects. is figure is higher than a study conducted in Gondar town (64.7%) [10] and Addis Ababa (53. 5%) [14]. However, this result is lower than a result reported by pharmacy professionals in Addis Ababa (85.2%) [21]. But the WHO recommended that the term side effect refers to minor effects related to the pharmacological properties of the drug [7,22]. In addition, 20.18% of the respondents knew the term pharmacovigilance and its activities which are comparable with a study done in the Jimma zone (19.5%) [16]. Moreover, the present study indicated that a small number of respondents knew the availability of national reporting system (22.81%) and ADR reporting form (21.05%), which is lower than 40.4% in a study done in Lagos, Nigeria [23,24], 49.02% in Gondar [10], and 63.4% at TASH, Addis Ababa [14].
Besides, 58.77% of respondents were aware that all HCPs are responsible and an obligation to report ADRs to the  concerned body in contrast to 81.1% in a study conducted in the Amhara region [12]. Moreover, 21.93% of the HCPs said that the possibility of ADR is the first differential diagnosis at all times in patients taking medicines opposed to a report by EFDA, the regulatory agency of Ethiopia [2]. One possible explanation for insufficient knowledge of respondents in this study was that only 19 (16.67%) respondents had taken training on ADRs and their reporting. Hence, creating awareness, providing training, and upgrade knowledge of HCPs about how ADRs are the priority causes of morbidity and mortality and the essence of ADR reporting to the public is essential. Significant proportion of the respondents (81.58%) replied that ADRs should be reported only when they are serious and life-threatening, while a few respondents (21.05%) were aware that mild to moderate unexpected, certain, and suspected reactions need to be reported. In contrast, a study done in one of the tertiary centers in Northern Nigeria showed that a majority (>70%) of the respondents were aware that suspected, serious, and certain reactions should be reported [25] and 48.6% in a study conducted in Amhara [12]. ADRs are the most ignored but yet constitute the major drug therapy problems and the most important ones to cause major impact on public health, reducing patients' quality of life and a considerable financial burden, and hence, their reporting deemed to be very essential.
is study showed that 87.7% of  respondents agreed that ADR reporting should be part of their duty, which is in line with the findings from similar studies in Addis Ababa, 84% [14], Gondar, 84.3% [10], but higher than a study in Jimma, 57.31% [16]. In addition, 76.3% respondents agreed that affirmation of ADRs related to the drug is necessary before reporting in contrast to 83.3% respondents in a study conducted at Gondar [10]. is perceived obligation to establish causal relation and feeling of legal problems might be one reason for the low reporting rates. Considerable number of respondents (73.7%) agreed that ADR reporting is time-consuming activity and about half (52.6%) of HCPs believed that ADR reporting creates additional workload.
is perception of the respondents may substantially affect their motivation of ADR reporting in their clinical practice.
Looking at the practice of HCPs about ADR reporting, the present study revealed that 29.82% (Figure 4) in contrast with 38% in TASH, Addis Ababa [14], 55.9% in Gondar [10], 65% in Turkish [29], and 81% in Northern Nigeria [25] encountered patients with ADR in their clinical practice in the past 12 months. Among those HCPs who encountered ADRs, 50% claimed that they have reported the ADR despite only 29.41% reported to appropriate body EFDA. Underreporting of ADRs by HCPs remains to be a challenge in Ethiopia. Not only the health professionals working in hospitals but also the medicine regulatory agencies have to work jointly to monitor drug safety issues and maintain a balance between medicines' benefits and risks. e major reasons for not reporting of ADRs were lack of feedback; reporting forms are not available when needed, not knowing where to report, not knowing how to fill and report ADR, uncertain of the causal association between the drug and ADR, and insufficient clinical knowledge, among others. Regular sensitization programs such as continued education and workshops and trainings, timely availability of ADR forms and ADR information sheets, feedback, and communication with the concerned bodies (e.g., EFDA) and encouraging all health professionals and patients to report suspected ADRs were some of the strategies forwarded by the respondents to foster ADR reporting (Figure 1). ese training programs and continuous feedback clear all misconceptions associated with the ADR reporting as the previous reports have shown significant improvement in knowledge, attitude, and practice of health care workers about ADR after intervention [30][31][32].
Conduction of this study was not without limitation. One of the limitations was that we did not include all HCPs working in a hospital, and only those having frequent exposure to drugs and patients were selected. e sample size was small. However, the inclusion of different HCPs does make it valuable. As the study used self-administered questionnaire, recall and personal bias may have affected the data obtained. Above all, it is a simple assessment of KAP in specific hospitals that does not assess multisectors and sort out problems in the area of pharmacovigilance, but without denying that the findings from this study are inputs for further researchers.
From total 114 HCPs, 28 (24.56%) had adequate knowledge, among which pharmacy personnel accounted the majority 42.86%. On the contrary, 86 (75.44%) had inadequate knowledge, of which half of them were nurses followed by physicians (24.42%).

Conclusion
e present study identified that HCPs working in KGH and APH had inadequate knowledge on ADR reporting and poor documentation and ADR reporting practice which contributed to underreporting in hospitals to EFDA despite the majority HCPs had favorable attitude towards ADR reporting. Hence, taking into account HCPs' suggestions and the study findings, the present study strongly underscore that creating awareness and improving the knowledge of all HCPs through regular sensitization programs, trainings, and timely feedback is a very crucial strategy to enhance spontaneous ADRs reporting in health facilities to the concerned body, which ultimately impacts the provision of quality patient care.

Data Availability
All data generated or analyzed during this study are included within the manuscript and supplementary file (STROBE checklist for observational studies). Moreover, the datasets used and/or analyzed (SPSS data) are available from the corresponding author on reasonable request.

Ethical Approval
First, an official cooperation letter was obtained from Wollo University, College of Medicine and Health Sciences, and submitted to the clinical director of both hospitals. en, data collection was commenced after all the relevant communication was made with hospitals officials, and objective of the study was disclosed and understood by HCPs.

Consent
Written informed consent was obtained from all the participants.

Conflicts of Interest
e authors declare that they have no conflicts of interest.