key: cord-1003532-v145shj8 authors: Cheoun, Mee-Lang; Heo, Jongho; Kim, Woong-Han title: Antimicrobial Resistance: KAP of Healthcare Professionals at a Tertiary-Level Hospital in Nepal date: 2021-09-24 journal: Int J Environ Res Public Health DOI: 10.3390/ijerph181910062 sha: 5fd70b490a118989ae475ec05fe5f344cd38c522 doc_id: 1003532 cord_uid: v145shj8 Although increasing antimicrobial resistance (AMR) is a substantial threat worldwide, low- and middle-income countries, including Nepal, are especially vulnerable. It is also known that healthcare providers (HCPs) are the major determinants of antimicrobial misuse. A cross-sectional, self-administered survey was conducted among 160 HCPs to assess the knowledge, attitudes, and practices (KAP) of Nepali HCPs regarding AMR and its use. Descriptive statistics and nonparametric tests were performed to evaluate KAP dimensions and investigate subgroup differences. HCPs scored higher on theoretical than practical knowledge. Regarding practical knowledge, men scored higher than women (p < 0.01), and physicians scored higher than nurses (p < 0.001). Participants aged < 25 years scored lower on practical knowledge than older participants (p < 0.001), while those with <3 years work experience scored lower than those with >6 years (p < 0.05). Participants from the medical department scored higher on practical knowledge than those from the surgical department (p < 0.01). AMR control was more accepted in the medical than in the surgical department (p < 0.001). Regarding practices, women and nurses scored higher than men (p < 0.001) and physicians (p < 0.01), respectively. An educational intervention that is tailored to the sociodemographic and professional characteristics of HCPs is necessary to reduce the gap between theoretical and practical knowledge and improve their attitudes and practices. Antimicrobial resistance (AMR) is a global concern, affecting both countries with limited resources and developed countries. Since an article was first published in 1996 urging the strengthening of AMR control through antimicrobial stewardship in health systems [1] , continuous efforts have been made [1, 2] . Most recently, in 2019, AMR was proposed for the first time as a specific indicator of Good Health and Wellbeing, which is Goal 3 of the United Nations' 2030 Sustainable Development Goals [3] . However, lowmiddle-income countries (LMICs) with poor infection control and prevention systems are more vulnerable to the increasing threat posed by AMR [4] . The World Bank Report predicted that increasing AMR is estimated to drive an additional 28 million people to extreme poverty by 2050, mainly in LMICs [5] . AMR in Nepal has significantly increased, with an increasing trend in the proportion of multidrug-resistant organisms over the past 20 years [6, 7] . In a recent study, more than 50% of Escherichia coli, Klebsiella pneumoniae, and Streptococcus pneumoniae isolates and more than 30% of some Shigella sp. and Vibrio cholerae isolates were resistant to first-line antibiotics, and many other bacterial pathogens were resistant to most first-line and some second-line antibiotics [8] . The lack of AMR control in the Nepali health systems is attributable to the rarity of bacterial confirmation and susceptibility tests, lack of well-equipped facilities [6] , and the low ratio of physicians to patients (0.17 per 1000 population) [9] . These factors may contribute to irrational antibiotic provision, which, in turn, may lead to more self-medication [6] and a high degree of dependency on informal drug dispensers [8] , exacerbating the current AMR situation in Nepal. Inappropriate prescribing patterns of HCPs have been identified as a major determinant of antimicrobial misuse in Nepal [6] ; these actions include prescribing broad-spectrum antibiotics at incorrect doses, using antibiotics as a routine course of supportive care, and providing inadequate medication counselling [10] . Given that knowledge, attitudes, and skills are the main factors affecting adequate AMR control [11] , HCPs need to increase and update their knowledge and practices on an ongoing basis to keep pace with the constantly changing multisectoral factors surrounding the AMR threat. To the best of our knowledge, KAP studies on AMR targeting HCPs have rarely been conducted in Nepal. Only two KAP studies on AMR were recently conducted among university students [12] and community members [13] in Nepal. Thus, this study aimed to assess the KAP of HCPs toward AMR and the rational use of antibiotics at a tertiary level hospital in Nepal. Additionally, this study aimed to examine the differences in the KAP levels across subgroups categorized by sociodemographic and professional variables. We conducted a cross-sectional survey among HCPs at a tertiary, nongovernment hospital in south-eastern Kathmandu, Nepal, in January 2020. The university hospital has more than 400 beds, covering the population from the Kavrepalanchok district and other surrounding districts. From August 2020, this institute has been recognized as a COVID-19 care hospital by the government of Nepal, which implies its substantial role in the community. HCPs working at the institute during the study period who were willing to participate in the survey were recruited. HCPs from the Departments of Psychiatry, Ophthalmology, Hair Transplant, Radiology, and Forensics were excluded because they are relatively less involved in the antibiotic process. A minimum sample size of 160 was calculated considering the expected response rate of 50% when a 95% confidence level and a 5% margin of error were applied. A self-administered questionnaire was distributed for data collection by co-investigators from both the JW LEE Centre for Global Medicine and the institute during official work hours (08:00-16:00). A total of 200 questionnaires were distributed and only voluntary participants were asked to respond to the questionnaire on site. The study investigators fully explained the purpose of the survey to the participants. Written informed consent was obtained before the survey and 164 (response rate: 82%, 164/200) responses were received. Since four of them missing the demographic information were excluded, the final 160 questionnaires were analysed for this study. The questionnaire was developed based on several related studies [12, [14] [15] [16] and adapted to the Nepali setting (Supplement Document S1). The questionnaire was evaluated by six experts in the field using the content validity index (CVI). For the calculation, we adopted the scale level CVI (SCVI)/Ave, which is defined as the average of the item-level CVI for all items on the scale [17] . A minimum I-CVI of 0.78 and SCVI/Ave of 0.90 indicated excellent content validity [18] . The first domain comprised theoretical knowledge and practical knowledge. This distinction was based on Rolfe's typology [19] , which defines theoretical knowledge as information discovered from books, journals, or lectures, while practical knowledge is defined as information obtained from one's experience performing relevant tasks in specific situations. With regard to the AMR rate question in the section on practical knowledge, Klebsiella spp. was selected not only because it is the second leading bacterial aetiology of healthcare-associated infections (HCAIs) in Nepal [20] , but also because it is included in the WHO Global Antimicrobial Resistance Surveillance System (GLASS) list. The true AMR rate was obtained from the GLASS 2017-2018 report [21] , and other previous studies were also considered [20] . For the theoretical knowledge domain, HCPs were asked to reply with "True" or "False" to each question. Scores of 1 or 0 were given for the correct or incorrect answers. The practical knowledge domain consisted of two different themes which are 'Current AMR in the community' and 'AMR term familiarity'. For the theme of 'Current AMR in the community', five different choices including "0-25%", "25-50%", "50-75%", "75-100%", and "don't know" were offered. Scores of 1 and 0 were given for the correct and incorrect answers. Lastly, for the theme of 'AMR term familiarity', a score of 1 was given to the positive responses: "I've heard the term and I can explain what it is", "I've used the term before", and "I use the term in everyday practice". A score of 0 to the negative responses: "I've never heard of it" and "I've heard the term but I'm not sure what it is". The second domain included questions on attitudes towards the severity of AMR and the acceptability of AMR interventions that were scored on a 5-point Likert scale. Acceptability was related to education and regulations. The average severity score and the average acceptability score were calculated and ranged from 1 to 5. Additionally, preferences regarding future intervention methods and the sectors perceived as responsible were also ascertained with multiple-choice questions (Supplementary Figure S1 ), which were not included in the average acceptability score. The practice domain incorporated questions asking their management of antibiotics and factors influencing their decisions regarding antibiotic prescriptions and were answered on a 5-point scale with options ranging from "strongly disagree (1)" to "strongly agree (5)". Based on Bloom's original cut-off scale, which has been widely used in KAP studies [22] [23] [24] , ≥80% was considered good, ≥60% was considered moderate, and <60% was considered poor when assessing the mean score in each domain. Descriptive statistics were used to summarize the data. Because the data were not normally distributed by the Shapiro-Wilk test, Mann-Whitney (U) test and Kruskal-Wallis test (χ 2 ) were used to assess the differences between subgroups stratified by sex, age, work experience, position, and department in each domain (K, A and P). Dunn's pairwise comparisons were also used for post hoc tests with the Bonferroni adjustment. All analyses were performed using Stata 16.0 (Stata Corp, College Station, TX, USA). Table 1 shows the sociodemographic characteristics of the respondents. Overall, nearly twice more women (70.6%) participated as men (29.4%), and half of all the respondents (50.0%) were aged between 25 and 29 years. More than half of the respondents (55.6%) had practiced for less than three years, while 47 (29.4%) had practiced for 3-5 years, and 24 (15.0%) had practiced for more than six years. The respondents included 75 (46.9%) physicians, 76 (47.5%) nurses and 9 (5.6%) pharmacists and pathologists. In total, 151 physicians and nurses were from the medical department, surgical department, anaesthesia department, emergency department, and special care unit. The medical department included internal medicine, paediatrics, and gastrointestinal endoscopy groups, while the surgical department included ear, nose, and throat, obstetrics and gynaecology, orthopaedics, surgery, and post-surgery groups. Nine pathologists and pharmacists were also included in the category of "others" with other physicians and nurses from the anaesthesia department, emergency department, and special care unit. Overall, the respondents showed a good level of theoretical knowledge (3.51 ± 0.59), while the practical knowledge level was poor (1.35 ± 1.41) ( Table 2 ). Regarding the theoretical knowledge questions, more than 95% of the respondents provided the correct answers to all questions except regarding the antibiotic combinations. In response to the practical knowledge questions concerning the current local AMR of Klebsiella pneumonia, only 8 (5%) and 48 (30%) respondents provided the correct answers for ciprofloxacin and meropenem, respectively (Table 3) . Regarding ciprofloxacin, almost half (48.8%) of the respondents underestimated, while the other half (46.3%) answered "I do not know", as shown in Figure 1 . The proportion of correct answers was higher for meropenem than for ciprofloxacin, although the proportion of respondents who answered, "I don't know" (52.5%), was still more than half. Furthermore, most of the participants answered that they had never heard of or were not sure of the practical terms related to AMR (ASP (84.4%) and antibiogram (83.2%)). The attitudes toward the severity of AMR and acceptability of AMR control were at a moderate level (Table 2) . Detailed results regarding the answers to each question are shown in Figure 2 . Of all respondents, 83.7% (134/160) considered AMR to be a serious public health issue at the national level, while 61.9% (99/160) considered it to be a serious issue at the facility level. Furthermore, only 30.5% (49/160) of the respondents agreed (combined "Strongly Agree" and "Agree") that antibiotics were overused in their facility, in contrast to the result that most strongly agreed or agreed with the need to establish education programs regarding rational antibiotic use and antibiotic policies in their facility (79.4% and 81.9%, respectively). 18.8% (30/160) of the respondents believed that limiting the use of antibiotics could impair patient care, and approximately 50.1% of the participants disagreed. shown in Figure 2 . Of all respondents, 83.7% (134/160) considered AMR to be a serious 175 public health issue at the national level, while 61.9% (99/160) considered it to be a serious 176 issue at the facility level. Furthermore, only 30.5% (49/160) of the respondents agreed 177 (combined "Strongly Agree" and "Agree") that antibiotics were overused in their facility, 178 in contrast to the result that most strongly agreed or agreed with the need to establish 179 education programs regarding rational antibiotic use and antibiotic policies in their facil-180 ity (79.4% and 81.9%, respectively). Nevertheless, 18.8% (30/160) of the respondents be-181 lieved that limiting the use of antibiotics could impair patient care, and only approxi-182 mately 50.1% of the participants disagreed. On average, the respondents had a moderate level of accurate practices regarding 187 antibiotics ( Table 2 ). The detailed results regarding the answers to each question are 188 shown in Figure 2 . Regarding the related factors, bacterial confirmation and sensitivity 189 reports were considered important prerequisites for antibiotic prescriptions by 71.3% 190 (114/160) of the respondents. Similarly, almost 70% of the respondents agreed that the cost 191 of an antibiotic should be considered before it was prescribed. Nevertheless, 13.3% (10/75) 192 and 17.3% (13/75) of the physicians strongly agreed that their prescription practices are 193 influenced by patient demands and that they are more influenced by the availability of 194 the antibiotics than by the diseases being treated, respectively. 195 The Kruskal-Wallis test revealed significant between-sex differences in two domains: 196 practical knowledge (p<0.01) and practices (p<0.001) ( Table 4 ). It also revealed significant 197 differences between age groups (p<0.001) and work experience groups (p<0.05) in practi-198 cal knowledge. Dunn's multiple comparisons test showed that respondents younger than 199 25 years (a) had the lowest level of practical knowledge (a