Knowledge, perceptions and practices on antibiotic use among Sri Lankan doctors.

INTRODUCTION: Prescribers have a major role in preventing antimicrobial resistance (AMR) through appropriate prescribing. However, in countries like Sri Lanka, where continuous professional development is not mandatory for license renewal and antimicrobial stewardship is not implemented, prescribing practices go largely unchecked.
OBJECTIVES: To identify the knowledge on antibiotic use and practices related to antibiotic prescribing among Sri Lankan doctors.
METHODS: This cross-sectional study was conducted in 2020. We used a validated, pretested Google-form based questionnaire with multiple choices, single best answer questions, polar questions (Yes/No) and five-point Likert scale questions. The Google-sheet generated was used for data analysis. Knowledge and practice scores were calculated.
RESULTS: Of the 262 respondents, 40.1% were males. Majority (61.8%) were aged 25-35-years and in medical practice for 0-5 years (48.9%) while 46.2% had or were engaged in post graduate studies. Knowledge scores ranged from 98.31% to 46.55% [mean:71.27% (SD±10.83); median:71.18% (IQR 64.4-79.7)]. Most (98.09%) obtained ≥50 marks while 45.8% scored more than the mean. The practice scores ranged from 100% to 0% [mean:65.33% (SD±18.16), median:66.67% (IQR53.3-80)]. The majority (81.3%) scored ≥50 in the practice score while 52.3% achieved more than the mean practice score. The knowledge score and the practice score differed significantly (p<0.001, related sample Wilcoxon Signed Rank Test) but the knowledge and practice scores were significantly correlated [Spearman correlation, p<0.001, r = 0.343 (Bias corrected 95% CI 0.237-0.448)]. Knowledge scores and the practice scores were significantly higher in those with or undergoing postgraduate training.
CONCLUSIONS: While the knowledge and practice scores were high, and knowledge and practice scores were correlated, the practices score was lower than that of knowledge indicating the need to encourage correct practices through means other than solely promoting knowledge.

Introduction

Antimicrobial resistance (AMR) has been identified as one of the ten threats to global health in 2019 by the World Health Organization (WHO) [1]. The rapid emergence of resistant microorganisms is a consequence of inappropriate use of antibiotics in many sectors including health sector [2] and has a negative impact on patient outcomes [3].

Inappropriate use of antibiotics in healthcare includes incorrect dosing, inadequacy of duration, inappropriate frequency, lack of compliance, unsupervised self-medication and over-the-counter (OTC) sale of antibiotics without prescriptions. These global issues are escalating at an alarming rate [4-7].

Although inappropriate use of antibiotics is prevalent globally, developing countries are most affected due to higher infection rates and limited resources [7]. In some countries, one of the most important factors affecting antimicrobial resistance is the lack of proper training of prescribers in proper antimicrobial use and infectious disease diagnosis. Demand and pressure from the patients have been identified as another cause driving the irrational use of antibiotics in low and middle-income countries [8]. Many studies conducted in developing countries have revealed that antibiotics are commonly used in day-to-day practice for being fever and respiratory symptoms, irrespective of etiology [8, 9]. Sri Lanka being a lower-middle-income economy is no different from other similar countries. High rates of antibiotic use at the first contact is a well-recognized issue in Sri Lanka [10]. High patient volume and fear of bacterial superinfections have been cited as reasons for antibiotic overuse in Sri Lanka [11].

Doctors are the main prescriber in many countries including Sri Lanka. In addition to the high workload and lack of facilities limiting the opportunities to deliver evidence-based care, [2] a prescriber’s decision may also be influenced by factors such as updated knowledge [12], fear of losing patients’ and lack of information on rational antibiotic use [13].

Community-acquired antibiotic resistance is on the rise [14] and up-to-date knowledge and correct practices regarding rational use of antibiotics among prescribers are now more important than ever. Prescribers should also consider AMR as an important factor in prescribing [15, 16]. The WHO World Health Assembly adopted an action plan to combat AMR and preserve the efficacy and availability of antimicrobials in 2015. Two of its objectives are improving awareness and understanding on AMR and optimizing the use of antimicrobials [17].

Hence this study was conducted to identify the knowledge and practices related to antibiotic use among Sri Lankan doctors across different specialties and levels of experience, in the context of AMR.

Materials and methods

This was a cross-sectional study conducted recruiting 262 participants from various fields of Medicine working in Sri Lanka, through convenience sampling. The ethical clearance was obtained from the Ethics Review Committee, Faculty of Medicine, University of Peradeniya, Sri Lanka (2020/EC/36) and participants indicated consent for participation through pressing the next button as informed through instructions. The questionnaire was Google form-based. It was validated by a Specialist in Medical Education and two Consultant Microbiologists with experience in similar studies and relevant clinical practice. The tool was piloted to a group of 10 medical officers and modified as needed. The finalized tool was disseminated to doctors through hospital based social media groups. Any doctor practicing in Sri Lanka, registered with the Sri Lanka Medical Council were eligible to answer the questions. One attempt was allowed per-email address; therefore, a single response was taken per-participant. The questionnaire was open to accept answers from 15/10/2021 to 12/01/2021.

The questionnaire collected socio-demographic details of the participants including their duration of practice, work unit, their current role, and their post-graduate qualification status. The questionnaire consisted multiple choice questions, single best answer questions, polar questions (Yes/No), five-point Likert scale questions, as well as open-ended questions.

The questionnaire was used to assess the knowledge on antibiotics and antimicrobial resistance using a score. For this purpose, eight defined (Yes/No/ Don’t know) questions and six multiple choices were allocated. For the former group, 1 mark was awarded for the correct answer, and none were awarded to any answer labeled as incorrect/ don’t know/ blank. In the multiple-choice questions, 0.25 marks was given for each correct choice in identifying beta-lactam antibiotics and causes for AMR. For the same questions, 0.25 was deducted for each incorrect choice. Negative marks were not carried forwards from question to question. The answers provided for the multiple-choice questions that assessed the knowledge on selecting the most appropriate antibiotic for resistant phenotypes [Methicillin resistant Staphylococcus aureus (MRSA), Carbapenem resistant Enterobacteriaceae (CRE), Extended Spectrum Beta Lactamase producing Enterobacteriaceae (ESBLE) and vancomycin resistant enterococci (VRE)] were categorized as the “correct choice”, “correct choice + others”, and “incorrect choice”. Only the correct choice was awarded 1 one and incorrect choices were not given negative marks.

The answers provided for the questions which assessed the knowledge on abbreviations of resistant bacterial phenotypes, were initially categorized as know the meaning, don’t know the meaning, and unanswered. Those that categorized as “don’t know the meaning”, were subcategorized as “don’t know the meaning, but have heard the term” and “don’t know the meaning and not heard the term” (S1 Table).

In the questions that assessed prescribing related personal experiences and perspectives, the categories “agree” and “strongly agree” were merged to a common category as “agree” and the categories “disagree” and “strongly disagree” were merged as “disagree”.

Ten questions were allotted to assess the practices regarding antibiotic use (S2 Table). They were five-point Likert scale (almost always/ often/ sometimes/ rarely/ never) based or polar questions (Yes/ No). For assigning marks, “almost always” and “often” and, “rarely” and “never” were grouped together. Depending on the nature of the question, the most appropriate category was given 2 marks whereas the least appropriate was awarded 0. If the participant had stated his/her answer as “sometimes” 1 mark was given. Any question that was left unanswered was not given marks. The polar questions were awarded 1 and 0 for the correct and incorrect answers respectively.

For the question on how the dose of antibiotics was decided, 2 marks were given for “by referring to BNF or another formulary”, 1 mark for any other choice except for “by remembering”, and 0 was given for, “by remembering”.

The answers provided for the open-ended questions were thoroughly assessed and categorized under common themes. Those that were unable to do so, were located under the theme “others”.

Any questions left blank were labeled as either “blank” or “unanswered”.

The questionnaire had a voluntary option to indicate if they wanted to have a feedback on the answers and the answers were emailed to those who indicated so.

The maximum mark achievable for knowledge score was 14.75 while it differed for the practice score depending on if participants were engaged in hospital practice and/or private practice. All total marks were converted to %, so the maximum achievable mark was 100% while the minimum was 0%.

Knowledge and practice scores were checked for normalcy. T test and ANOVA were used to compare the knowledge scores across groups as the knowledge score was normally distributed while the Mann-Whitney U test and Kruskal Wallis test was used to compare the practice scores as this was not normally distributed. Spearman correlation co-efficient was calculated between the knowledge and practice scores. The knowledge and practice scores for individuals were compared using the related sample Wilcoxon Signed Rank Test. Data analysis was done with SPSS (IBM statistics, version 23).

Results

Out of 262 participants, 105 (40.1%) were male. A majority (n = 162, 61.8%) were from 25–35-year group and have been in medical practice between 0–5 years (n = 128, 48.9%), while a larger proportion (n = 141, 53.8%) were not postgraduate trained (Table 1).

Table 1

Socio-demographic details of the participants.

		n	%
Age	25–35 years	162	61.8
	36–45 years	76	29
	46–55 years	22	8.4
	56–65 years	2	0.8
Sex	Male	105	40.1
	Female	157	59.9
Years in practice	0–5	128	48.9
	6–10	63	24
	>10	70	26.7
	Unanswered	1	0.4
Post graduate qualifications	Yes, currently enrolled	69	26.3
	Yes, completed	52	19.8
	No	141	53.8
What post graduate studies	M.D.	83	68.6
	M.Sc.	15	12.4
	*P.G. Diploma	18	14.9
	Other	4	3.3
	Unanswered	1	0.8
Current role	Academic	2	0.8
	Consultant	23	8.8
	Senior Registrar	18	6.9
	Registrar	40	15.3
	Senior House Officer	19	7.3
	Medical Officer	132	50.4
	Relief House Officer	6	2.3
	House Officer	14	5.3
	*PG trainee	3	1.1
	Medical Officer of Health	5	1.9
Work Unit	*ET/PCU	13	5.0
	Full time general practice	1	0.4
	Medical unit	35	13.4
	*Obs and Gyn unit	13	5.0
	*OPD	19	7.3
	Paediatric	15	5.7
	Surgical Unit	38	14.5
	Public Health	17	6.5
	Other specialties	111	42.4
Engaged in active clinical service	Yes	237	90.5
	No	25	9.5
Type of hospital working, if engaged in active clinical service	National Hospital	63	26.6
	Teaching Hospital	63	26.6
	Provincial General Hospital	8	3.4
	District General Hospital	46	19.4
	Divisional Hospital	13	5.5
	Base Hospital	32	13.5
	*PMCU	4	1.7
	Specialized hospitals	3	1.3
	Central clinics	1	0.4
	National programmes	1	0.4
	*MOH	1	0.4
	Others	1	0.4
	Unanswered	1	0.4
Engaged in private practice	No	179	68.3
	Yes	83	31.7
Nature of private practice	Affiliated to private hospitals	11	13.3
	Independent general practice (part time or full time)	50	60.2
	Specialist practice	22	26.5

*PG = Post Graduate, ET/PCU = Emergency Treatment/ Primary Care Unit, Obs and Gyn = Obstetrics and Gynaecology, OPD = Outpatient Department, PMCU = Primary Medical Care Unit, MOH = Medical Officer of Health

A higher number of participants provided the correct answer for questions that assessed their knowledge of antibiotics (Table 2). The responses were somewhat divided upon asking antibiotics are used only to treat bacterial infections, in which 159 (60.7%) had answered “yes” while 103 (39.3%) had answered “no”. Out of the latter group, 48 (46.6%) had mentioned fungal and parasitic infections, 25 (24.3%) had highlighted prophylactic use, 4 (3.9%) as gastroparesis, 3 (2.9%) as hepatic encephalopathy, 3 (2.9%) as viral infections and 12 (11.7%) had highlighted other instances such as an anti-inflammatory and immunomodulation (Table 3).

Table 2

Summary of responses related to knowledge on antibiotic use and resistance.

	Yes		No		Don’t know		Unanswered
	n	%	n	%	n	%	n	%
Antibiotics used to treat bacterial infections only *	159	60.7	103	39.3	-	-	-	-
Antibiotics can prevent development of bacterial infections	223	85.1	36	13.7	3	1.1	-	-
Administration of antibiotics within one hour to a patient with sepsis is life-saving	234	89.3	12	4.6	16	6.1	-	-
Antibiotics should be prescribed to any patient with fever	1	0.4	261	99.6	-	-	-	-
Empirical antibiotics should be escalated or de-escalated based on microbiological culture results	248	94.7	12	4.6	1	0.4	1	0.4
Age and body weight are important in deciding antibiotic use	259	98.9	2	0.8	1	0.4	-	-
Any antibiotic can be used to treat any bacterial infection	4	105	258	98.5	-	-	-	-
Combinations are always better than monotherapy	49	18.7	193	73.7	19	7.3	1	0.4
All bacterial infections can be treated with the same duration of antibiotics	-	-	262	100	-	-	-	-
Antimicrobial resistance is a world-wide problem *	261	99.6	-	-	1	0.4	-	-
Antimicrobial resistance is a problem in my country *	258	98.5	1	0.4	2	0.8	1	0.4

*Not considered for the knowledge score

Table 3

Summary of responses for other antimicrobial uses, knowledge on beta lactams and causes for AMR.

		n	%
Other uses of antibiotics	Fungal and parasitic infections	48	46.6
	Gastroparesis	4	3.9
	Hepatic encephalopathy	3	2.9
	Prophylaxis	25	24.3
	Viral infections	3	2.9
	Other	12	11.7
	Unanswered	8	7.8
Selecting beta lactams	Penicillin correctly identified	245	93.5
	Cefotaxime correctly identified	152	58.0
	Meropenem correctly identified	101	38.5
	Aztreonam correctly identified	57	21.8
	Vancomycin incorrectly identified	52	19.8
	Ciprofloxacin incorrectly identified	6	2.3
Cause(s) of antimicrobial resistance	Widespread or overuse of antibiotics	258	98.5
	Usage of broad-spectrum antibiotics	173	66.0
	Genetic mutations of bacteria	243	92.7
	Patient’s poor adherence to antibiotic dosage regimens	240	91.6
	Poor hand washing practice in hospitals	94	35.9
	Poor infection control practices in hospital	142	54.2
	Adhering to antibiotic treatment guidelines	3	1.1
	Substandard quality of antibiotics	193	73.7

Upon asking to pick beta-lactams out of the choices provided (Table 3), 93 participants (35.5%) selected at least 1 beta-lactam correctly but only 44 (16.8%) selected all four (penicillin, cefotaxime, meropenem, and aztreonam). Although many knew that penicillin is a beta-lactam (n = 245, 93.5%) a lower number of participants were aware that aztreonam was a beta-lactam (n = 57, 21.8%). A considerable number of participants (n = 52, 19.8%) had incorrectly picked vancomycin as a beta-lactam. Two participants (0.8%) did not selected any option and 50 (19.2%) selected an incorrect antibiotic and only 4 (1.5%) had selected all 2 incorrect antibiotics (vancomycin and ciprofloxacin) coupled with other choices. Hundred and twelve (42.7%) had selected a beta-lactam(s) with an incorrect antibiotic(s).

A high number of participants had selected appropriate causes for antimicrobial resistance (Table 3). But it is noteworthy that only 94 (35.9%) and 142 (54.2%) had selected poor handwashing practices and poor infection control practices in hospitals as possible contributors.

Most participants were aware of MRSA (n = 160, 61.1%) while out of 101 participants who did not know the meaning, 100 participants (99%) had heard the abbreviation while only 1 (1%) had not heard the term either. A similar pattern was observed with regards to ESBLEs, although the group of participants who had neither heard nor knew the meaning is comparatively greater than that of MRSA (n = 49, 35%). It is noteworthy that only a few participants were aware of the meaning of CRE (n = 38, 14.5%) and the number of participants who had neither heard nor knew the meaning was significantly high (n = 157, 81.8%). A similar pattern was seen in VRE (Table 4).

Table 4

Summary of responses for abbreviations used for common antibiotic-resistant bacteria.

	Know the meaning n (%)	Don’t know the meaning		Unanswered n (%)
		Heard n (%)*	Not heard n (%)*
MRSA	160 (61.1%)	100 (99.0%)	1 (1%)	1 (0.4%)
CRE	38 (14.5%)	35 (18.2%)	157 (81.8%)	32 (12.2%)
ESBL	114 (43.5%)	91 (65%)	49 (35%)	8 (3/1%)
VRE	69 (26.3%)	46 (27.1%)	124 (72.9%)	23 (8.8%)

*Percentage calculated from the total number of participants who stated as “Don’t know the meaning”. MRSA–methicillin resistant Staphylococcus aureus, CRE–carbapenem resistant enterobacteriacae, ESBLE–Extended spectrum beta lactamases, VRE–vancomycin resistant enterococcus

Based on the responses for the questions to assess the knowledge on drug options for various resistant bacterial phenotypes, it is apparent that most participants (n = 169, 64.5%) had selected the correct choice for MRSA while 47 (17.9%) had selected the correct choice coupled with other choices, 36 (13.7%) had selected an incorrect choice and 10 (3.8%) had refrained from providing an answer. A similar pattern is observed with regards to ESBL although 74 participants (28.3%) had selected an incorrect choice. A striking observation was made with regards to CRE where only 32 participants (12.2%) had selected the correct choice and 58 (22.1%) and 166 (63.4%) had selected the incorrect choice and refrained from answering respectively. A similar pattern is seen with VRE as well (Table 5).

Table 5

Summary of responses for drug of choice for treating resistant bacterial phenotypes.

	MRSA		CRE		ESBL		VRE
	N	%	n	%	n	%	n	%
Correct choice	169	64.5	32	12.2	102	38.9	53	20.2
Correct choice + others	47	17.9	6	2.3	13	5.0	3	1.1
Incorrect choice	36	13.7	58	22.1	74	28.2	66	25.2
Blank	10	3.8	166	63.4	73	27.9	140	53.4

Table 6 depicts understanding and perspectives regarding antibiotic use. These questions were of agree/ disagree type. Eighty-one participants (30.9%) had agreed with the statement that they find it hard to select the correct antibiotic for a patient with a bacterial infection, while 121 (46.2%) had disagreed. Further,116 (44.3%) participants agreed to the fact that patients will feel their illness is not taken seriously if an antibiotic is not prescribed, while 88 (33.6%) had disagreed and 56 (21.4%) had neither agreed nor disagreed.

Table 6

Perceptions and understanding on antibiotic use.

	Agree		Disagree		Neither agree nor disagree		Unanswered
	n	%	n	%	n	%	n	%
I find it hard to select correct antibiotic to a patient with a bacterial infection	81	30.9	121	46.2	59	22.5	1	0.4
Prescribing antibiotics when patient does not need them, doesn’t cause any harm	19	7.3	234	89.3	9	3.4	-	-
Aware about National Guidelines on antibiotic prescription	186	71.0	47	17.9	28	10.7	1	0.4
Important to know the resistant pattern of the bacteria when selecting an antibiotic	256	97.7	3	1.1	3	1.1	-	-
More likely to prescribe antibiotics when the workload is high	67	25.6	156	59.5	37	14.4	2	0.8
Consultation is short when a prescription is issued	48	18.3	156	59.5	54	20.6	4	1.5
Patient will feel their illness is not taken seriously if an antibiotic is not prescribed	116	44.3	88	33.6	56	21.4	2	0.8
Patient education on antibiotics will have an effect on their expectations in later consultation	243	92.7	9	3.4	10	3.8	-	-
I feel confident about prescribing antibiotics	178	67.9	32	12.2	50	19.1	2	0.8
I like more education on antibiotic use, resistance and stewardship	258	98.5	-	-	3	1.1	1	0.4
I think I practice rational use of antibiotics in my hospital practice	187	71.4	18	6.9	43	16.4	-	-
I think I practice rational use of antibiotics in my private practice	86	32.8	9	3.4	28	10.7	11	4.2

When the participants were questioned about their practice of requesting cultures before commencing antibiotics, most who were involved in hospital practice had stated ‘almost always or often’ (n = 169, 64.5%) while 60 (22.9%) had stated ‘sometimes’. Out of those who stated as ‘sometimes, rarely or never’ (n = 77), 16 had mentioned that no facilities are available while most (n = 26) had mentioned a variety of reasons such as encountering common infections hence not needing culture and their practice is mainly based on clinical assessment. This was different when the same question was analyzed among those who were engaged in private practice (n = 98 responses), in which most participants (n = 47/98) had stated ‘sometimes’. Out of those who stated ‘sometimes, rarely or never’ (n = 68), 18 had stated the high cost as the reason whereas 7 as ‘no facilities available’, 4 as ‘time consuming’ and 20 a variety of reasons such as referring the patient to a hospital when needing a culture and not encountering many patients that require cultures. (Table 7).

Table 7

Summary of responses for practices regarding antibiotic use.

Statement	Response	n	%
Patients demand antibiotics from you	Almost always and often	81	30.9
	Sometimes	135	51.5
	Rarely and Never	43	16.4
	Unanswered	3	1.1
I feel under pressure if my patient expects an antibiotic prescription	Almost always and often	43	16.4
	Sometimes	94	35.9
	Rarely and never	122	46.6
	Unanswered	3	1.1
How often do you prescribe antibiotics? (not considered for scoring)	≤ once a week	58	22.1
	≥ 1/day	113	43.1
	2–4 times/week	83	31.7
	Unanswered	8	3.1
Do you select antibiotics according to local/ international guidelines?	Almost always and often	158	60.3
	Sometimes	72	27.5
	Rarely and Never	27	10.3
	Unanswered	5	1.9
Do you refer BNF (British National Formulary) when prescribing antibiotics?	Almost always and often	114	43.5
	Sometimes	111	42.4
	Rarely and Never	36	13.7
	Unanswered	1	0.4
How do you decide the dose of antibiotics (more than one answer accepted)	By remembering	173	66.0
	By referring to BNF or another formulary	161	61.5
	By following my seniors’ practice	59	22.5
	By following my colleagues	27	10.3
Do you consider AMR when prescribing?	No	65	24.8
	Yes	163	62.2
	Unanswered	34	13.0
Do you order cultures before commencing antibiotics in your hospital practice?	Almost always and often	169	64.5
	Sometimes	60	22.9
	Rarely and never	17	6.4
	Not engaged in hospital practice	16	6.1
If ’Sometimes’, ’Rarely’ or ’Never’, what is the reason? (not considered for scoring)	High cost	8
	High number of patients	1
	No facilities	16
	Other	26
	Time consuming	5
	Unanswered	21
Do you request for cultures before commencing antibiotics in your private practice? *	Almost always and often	30	30.6*
	Sometimes	47	48.0*
	Rarely and never	21	21.4*
	Not engaged in private practice	151
	Unanswered	13
If ’Sometimes’, ’Rarely’ or ’Never’, what is the reason? (not considered for scoring)	High cost	18
	No facilities	7
	Other	20
	Time consuming	4
	Unanswered	19
Do you practice de-escalation therapy?	Yes	164	62.6
	No	78	29.8
	Unanswered	20	7.6
Do you practice IV to oral switch?	Yes	212	80.9
	No	42	16.0
	Unanswered	8	3.1

*In the demographic section only 83 people acknowledged to engaging in private practice. However, 98 have described their culture related practices in private practice. % calculated with 98 as denominator

Knowledge score

The maximum knowledge score achieved by a participant was 98.31% whereas the minimum was 46.55%. The mean knowledge score was 71.27% with a standard deviation (SD) of 10.83. The median knowledge score was 71.18% and the inter quartile range (IQR) was 64.4%– 79.7%. Of all 262 participants, 120 (45.8%) had achieved more than the mean value whereas 257 (98.09%) participants had obtained a knowledge score > = 50%.

Practice score

The mean practice scores achieved was 65.33% (SD 18.16). The median practice score was 66.67% (IQR of 53.3%– 80%). One-hundred and thirty-seven (52.3%) participants had achieved a score more than the mean value whereas 213 (81.3%) participants had achieved a score >50%.

Comparison of knowledge and practice scores across groups

The knowledge and practice scores were significantly correlated [Spearman correlation, p<0.001, r = 0.343 (Bias corrected 95% CI 0.237–0.448)]. However, the knowledge and practice scores were significantly different (p<0.001, Related sample Wilcoxon Signed Rank Test).

Knowledge scores (mean 74.62% and 69.31%, t-test p<0.001) and practice scores (Median 70.59% and 64.71%, Mann Whitney U test p = 0.001) were significantly higher in those with postgraduate qualifications (Table 8).

Table 8

Comparison of knowledge and practice scores across different groups.

	Sex		Difference	Involved in or completed post-graduate studies		Difference	Years in practice			Difference
	Male (%)	Female (%)		Yes (%)	No (%)		0–5 (%)	6–10 (%)	>10 (%)
Mean knowledge score (SD)	71.26 (10.87)	71.27 (10.86)	0.997 (T-test)	74.62 (10.33)	68.39 (10.48)	<0.001 (T-test)	71.29 (9.85)	73.77 (12.08)	68.96 (11.08)	0.038 (One-Way ANOVA)
Median (IQR)	71.19 (64.41–77.97)	71.19 (64.41–79.67)		74.58 (66.10–81.36)	67.80 (62.71–74.57)		71.19 (64.40–79.67)	72.88 (66.10 -–82.20)	67.80 (61.02–77.97)
Maximum	98.31	93.22		98.31	91.53		93.22	98.31	94.92
Minimum	45.76	47.46		47.46	45.76		50.85	49.15	45.76
Mean practice score (SD)	65.63 (17.72)	64.88 (18.87)	0.821 (Mann-Whitney U test)	69.31 (18.61)	61.91 (17.11)	0.001 (Mann-Whitney U test)	67.61 (15.77)	65.79 (18.38)	60.72 (21.22)	0.062 (Kruskal–Wallis test)
Median (IQR)	66.67 (53.85–80.00)	66.67 (53.33–80.00)		70.59 (58.82–82.35)	64.71 (52.94–73.33)		66.67 (60.00–80.00)	64.71 (53.33–80.00)	60.18 (46.67–76.47)
Maximum	100	100		100	93.33		100	94.12	100
Minimum	17.65	00		00	17.65		20	26.67	00

Table 9 portrays how the participants gather and expand their knowledge on antibiotics. Most (n = 161, 60.8%) had stated that they update themselves on antibiotic prescription by self-study. Other methods being working with seniors (n = 155,58.5%), working with peers (n = 86, 32.5%) and through teaching sessions (n = 109, 41.1%). Most participants had learnt proper antibiotic prescription from consultants (n = 137,52.3%). 63 (24%) had stated that they prefer online courses as a useful means to conduct training programmes related to antibiotic prescription while 49 (18.7%) stated as lectures, 28 (10.7%) as workshops, 16(6.1%) as clinical cases, 13(5%) as continuous medical education and 4 (1.5%) as newsletters. 71 (27.1%) had refrained from providing an answer.

Table 9

Expanding knowledge on antibiotics.

		n	%
How do you update yourself on antibiotic prescription on most occasions	By working with peers	86	32.5
	By working with seniors	155	58.5
	Through teaching sessions	109	41.1
	Self-study	161	60.8
From whom did you learn the most on proper antibiotic prescription?	Consultants	137	52.3
	Peers	22	8.4
	Self-study	92	35.1
	Other	10	3.8
	Unanswered	1	0.4
What type of training programs would you like, related to antibiotic prescription?	Workshops	28	10.7
	Lectures	49	18.7
	Clinical cases	16	6.1
	Online course	63	24.0
	CME	13	5.0
	Newsletters	4	1.5
	Unanswered	71	27.1

Discussion

This study explores the knowledge and attitude among Sri Lankan doctors with regards to antimicrobial use and resistance. As stated previously, AMR is one of the major threats to global health [1] and the significance of this study lies in the study findings and public health importance in a country where inappropriate antibiotic use is common [10].

Our study highlights relatively higher knowledge scores and comparatively lower practice scores achieved by the participants. A total of 98.09% had achieved a knowledge score more than 50%. The mean value of the percentage score achieved is 71.27% and the median, 71.18%. This is in contrast to the practices score in which 81.3% had achieved a score more than 50%. Here, the mean value was 65.33% and the median, 66.67%.

Most number of participants had 0–5 years of experience hence they are relatively fresh out of undergraduate studies, indicating they may still recall the content from undergraduate curricula while good practices may not be internalized.

Most did not have post graduate qualifications (53.8%) and this indicates that their knowledge updates were from CME (Continuous Medical Education) or if not from the undergraduate studies.

The high knowledge scores exhibited by the study participants is in agreement with another study conducted among junior doctors in primary care centers and hospitals in Crete, Greece [18]. Similar patterns were observed in other studies such as a study conducted among medical doctors in Khyber Pakhtun Khawah, Pakistan [8] and among 2500 Chinese medical and non-medical students [19].

While the overall scores were high, individual questions revealed that some core knowledge on antibiotics, resistance mechanisms and prescription have room for improvement.

Upon asking to select beta lactams, only 35.5% selected at least 1 and only 16.8% selected all four. Some (19.8%) incorrectly picked vancomycin as a beta lactam and only 21.8% knew aztreonam is a beta lactam.

The reason for this result, however, is not clear, but this could have been contributed by the lack of knowledge the participants possess regarding basic classification of beta lactams. Vancomycin is not uncommon in ward practice, and it is a non-beta lactam glycopeptide. Aztreonam on the other hand is not commonly heard of during undergraduate training and since most participants are relatively fresh undergraduates with 0–5 years of experience, the numbers are understandable.

With regards to knowledge in terms associated with common antibiotic-resistant bacteria majority (61.1%) were aware of MRSA. However, the numbers were far less in CRE (14.5%) and VRE (26.3%). These numbers contradict with a cross-sectional survey conducted among Italian young doctors in which 94% participants declared to know what VRE are, 90% to know CRE, 92.9% to know ESBLs and 99% to know MRSA [20]. The difference between these results may be related to their personal experience with patients infected with these multi-drug resistant organisms. Gaps were identified regarding drug choices for treating resistant bacterial phenotypes in which 38.9%, 20.2%, and 12.2% had correctly identified the antibiotics for ESBL, VRE and CRE. This highlights the need for CPD on antimicrobial resistance for updating practice.

A notable number of participants had stated that they find it hard to select the correct antibiotic (30.9%). Majority of our study participants were relatively younger, with 0–5 years experience. These rates are in agreement with previous studies conducted in Greece which identified that younger doctors lacked confidence in antibiotic prescription, particularly in instances where complex decision making is required [18]. Similar findings were present even decades ago where residents have expressed lack of confidence in antibiotic selection [21]. This highlights the need for continuous training, discussions, peer learning and a multi-disciplinary approach to antibiotic use throughout the working career of a medical officer.

The causes for AMR were correctly identified by most participants. However, it is surprising to see that only 35.9% had identified poor handwashing as a contributory factor. Handwashing is one of the most important ways of preventing illness [22] and a cornerstone among efforts to reduce antibiotic resistance [23]. A study reveals that higher knowledge of these factors were positively correlated with having post-graduate training [24].

The knowledge score was significantly associated with post graduate status of the participant and higher knowledge scores were observed among those were either pursuing or had completed their post graduate studies. This outcome is not a surprise since much knowledge and skills are acquired during post graduate training and rational use of antibiotics are learned in depth during clinical training. This knowledge score differed significantly with the duration of medical practice and highest scores were seen among 6–10 years category.

Majority (71%) are aware of national guidelines but only 14.1% would almost always practice selecting antibiotics based on them. These numbers however are not confined to this study. A study conducted in a tertiary care institution at the Caribbean reveals that only 41% and 35% would consult national and institutional guidelines respectively [24]. However, certain studies highlight the importance of treatment guidelines [3, 8] and some demand the need to develop local guidelines [25], a burden Sri Lanka doesn’t have to undergo since the Sri Lanka College of Microbiologists in collaboration with other professional colleges and Ministry of Health has published a local guideline already [26].

Majority (97.7%) agreed that knowing the resistant pattern of an organism is beneficial but 64.5% would almost always and often and 22.9% would sometimes order cultures in their hospital practice. Cost and lack of facilities were identified as key reasons for this drawback. Not all hospitals in Sri Lanka are equipped with microbiology laboratory facilities although all cater to a large number of patients on a daily basis. Hence, most practicing doctors in wards would prescribe antibiotics based on their knowledge and past experiences even if a culture is deemed necessary. It is well known that narrowing of antimicrobial spectrum based on culture results is an effective antibiotic management strategy in preventing AMR [27].

A high number of respondents (66%) agreed that they decide the antibiotic dose based on what they remember. In fact, this is the highest preference of majority while 61.5% refer to BNF or another formulary when they need to decide the dose. This preference might pave the way to medical errors since memory is not always accurate and reliable, ultimately leading to poor healing, long hospital stay, high cost and development of AMR at the end of the spectrum. This finding is in line with another study where only 7% of physicians would always take cultures for a suspected infection. The authors had observed that the prohibitive cost and delay in retrieving microbiology reports in some areas have adversely affected perceptions of the value of obtaining routine cultures [24].

Notably, 24.8% of the participants do not consider AMR when prescribing antibiotics, and needless to say this plays a pivotal role in worsening of AMR. This finding maybe due to lack of knowledge and large number of patients who require medical attention, hence limiting the time allocated for an individual patient.

It is commendable to see that majority practice de-escalation therapy (62.6%) and IV to oral switch (80.9%). It is up to the local guidelines to provide precise indications concerning intravenous–oral switch criteria, antibiotic combination choice criteria, and optimal durations of antibiotic treatments [3].

The practices score was also significantly associated with the postgraduate status of the participant just as the knowledge scores. This outcome maybe contributed by reasons stated previously which are common in both instances.

The knowledge and practice scores correlated significantly. This implies that possessing high and up-to-date knowledge in antibiotic use has a significant impact on correct antibiotic practices.

Our study has certain limitations. The practice was assessed using a questionnaire than using direct observations. However, we believe that the anonymous nature of the questionnaire would have made more people to answer the questions truthfully. Further, even though we validated and piloted the tool, we did not calculate the reliability, validity or other measures statistically.

Conclusions and recommendations

This study concludes that the practitioners generally had a good level of knowledge regarding antibiotic use. However, there is room for improvement in knowledge regarding beta lactams and selection of antibiotics for different resistant phenotypes.

Comparatively lower practices scores regarding antibiotic use were noted. The knowledge and practices of a doctor are positively correlated with the post graduate status highlighting the fact that knowledge and practices are refined during post graduate training. The knowledge score was also associated with duration of clinical practice.

This study also reflects the need to regulate the prescription of antibiotics and initiate training programmes for doctors in order to improve their practices. This also highlights potential causes for AMR hence aforementioned training programmes should focus more on behavior rather than knowledge alone. Furthermore, doctors must be encouraged to use local guidelines and CME is an essential step in promoting rational use of antibiotics, despite Sri Lanka being a country where CME is not mandatory for license renewal.

Mark allocation for knowledge scores.

(DOCX)

Click here for additional data file.

Mark allocation for practice scores.

(DOCX)

Click here for additional data file.

(XLSX)

Click here for additional data file.

29 Nov 2021

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Pathiyil Ravi Shankar

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please amend your current ethics statement to address the following concerns:

a) Did participants provide their written or verbal informed consent to participate in this study?

b) If consent was verbal, please explain i) why written consent was not obtained, ii) how you documented participant consent, and iii) whether the ethics committees/IRB approved this consent procedure.

3. Thank you for stating the following in the Competing Interests/Financial Disclosure * (delete as necessary) section:

(VL has received funding from Pfizer for a non-related study. Others declare no competing interests)

We note that you received funding from a commercial source: (Pfizer)

Please provide an amended Competing Interests Statement that explicitly states this commercial funder, along with any other relevant declarations relating to employment, consultancy, patents, products in development, marketed products, etc.

Within this Competing Interests Statement, please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your amended Competing Interests Statement within your cover letter. We will change the online submission form on your behalf.

4. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

Reviewer #3: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: PONE-D-21-34680

Research Article

Knowledge and practices on antibiotic use among Sri Lankan doctors

Reviewer’s Comment

General Comments: The manuscript requires copyediting for both content and English language.

Specific Comments:

Abstract:

Introduction: It would be better to avoid using abbreviations in abstract section. If it is used, follow only the internationally adopted abbreviations. I also suggest the author to use the term ‘antimicrobial’ rather than antibiotic.

Methods: The information provided in method section lack vital information about how the questionnaire was developed? Who were the study participants? How questionnaire was applied? How validation of the questionnaire was performed? How the scoring was executed? How validated response can be obtained through google form? Likert scale questions are generally used to assess the attitude of the respondents, but here it has been used for knowledge/practice?

Key words: The authors should make sure the keywords are MeSH words, particularly knowledge and practices,

Main body:

Introduction: Reorganize this section in different paragraph with background information, problem statement, rationale of the study

Methodology: It will be worthwhile mentioning the qualification and experiences of the experts who were involved in validation of the questionnaire. How reliability of the questionnaire was performed? Why different scoring was used, e.g. 0.25 score, 1 score, negative score, 0 score, 2 score? These are too confusing! The term ‘mark/s’ should better be replaced with the term ‘score/s’.

The information provided in method section lack vital information about how the questionnaire was developed? Who were the study participants? How questionnaire was applied? One respondent can fill the google questionnaire multiple times. How validated response can be obtained through google form? Likert scale questions are generally used to assess the attitude of the respondents, but here it has been used for knowledge/practice? Duration and location of the study?

Results: the number of unanswered responses are seen in few cases only, it would be better to merge them as ‘Don’t know/Unanswered’(table 2). Why the heading for Table 4 is below the table? Full form of the abbreviations used in table 4 should be mentioned in footer of the table. Table 6 is about the perception on antibiotic use, whereas there is no any word about perception in the title of the manuscript? It is quite unusual to use 4 points likert scale (table 6). It can be made into 3 scales (agree, Neutral/unanswered, disagree).

Practice can better be determined through observation of real prescriptions (table 7) not through asking the questions. The responses for this section through questionnaire particularly online will be biases-how many will say that they do malpractice???

It would be meaningful to use the median score rather than mean score.

There are nine tables in this manuscript. It is advisable to reduce the number of tables if feasible through merging or modifying the tables.

Discussion: Discussion require more comparisons and clarifications. Study limitations and recommendations are vital components of a manuscript.

Conclusions: This section should be short and conclusive with key message from the findings.

References: Make sure the references and their citations in the text are as per the journal’s requirements.

Reviewer #2: 1. Title - did the authors also study the aspect related to resistance?

2. Abstract - ABR or AMR?

3. The objective in the abstract, main text as well as findings and conclusions should be consistent; look at the ms very carefully - the terms use, resistance, prescription etc were used inconsistently

4. Abstract - methods - too brief, how was the data analyzed?

5. This topic has been studied in the past and many articles have been published. What the study adds to the literature, practice and policy? What is new?

6. main text - obj - I thought the study also look at knowledge and practice related ABc resistance!

7. Methods - was poorly written - non structured and non systematic way of writing the methods section - difficult for readers to follow and not easy to be replicated; what was the sample size and power of the study? How the respondents were selected? the main domain - what is the possible min and max values/scores for practice? what are the possible min and max scores for knowledge?; Was any psychometric measures done on the tool? Was the study piloted? alpha value? what software was used for the analysis? mean values should be followed by sd, and median with IQR

8. any limitations of the study? what are the study implications? any recommendations?

Reviewer #3: Authors have done a good work. There are some comments and suggestions in the manuscript submitted. Please go through the comments and suggestions and do the necessary corrections for a revised submission.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Mukhtar Ansari

Reviewer #2: Yes: Mohamed Izham Mohamed Ibrahim

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Reviewer Comment.doc

Click here for additional data file.

21 Dec 2021

PONE-D-21-34680

Knowledge and practices on antibiotic use among Sri Lankan doctors

PLOS ONE

1. Formatting has been done according to the two style guides provided

2. Please amend your current ethics statement to address the following concerns:

a) Did participants provide their written or verbal informed consent to participate in this study?

b) If consent was verbal, please explain i) why written consent was not obtained, ii) how you documented participant consent, and iii) whether the ethics committees/IRB approved this consent procedure.

The questionnaire contained the following segment at the start of the questionnaire. Ethical clearance application stated that informed consent would be taken by participants through online means. The explicit way of doing was not included in the ethics application.

Relevant section from ethics application

“An informed consent will be obtained at the start of the survey. All information regarding the study, participants’ rights and researcher’s contact details are provided on the first page of the online survey link directed via mobile application-WhatsApp/Viber. Information on personal identity such as name, address, contact details and citizenship ID number, SLMC reg number shall not be obtained in the questionnaire.

Participants are assured that their responses are treated anonymously and no individual will be identified to maintain the confidentiality of the information. The collected data and records shall be held by the members of the research team, made available only to the facilitator, ethical committee and other relevant departments and will not be made accessible to the public. Only the final report will be published.

Any participant has the right to continue or to refuse answering the questionnaire. Only the subjects who have given consent will be proceeded to the survey.”

The first section of the online survey

“This questionnaire is part of a study to assess knowledge, attitudes and practices on antibiotic use and resistance among medical doctors in Sri Lanka with the aim of developing targeted educational programmes.

This study is being conducted by Dr.Kaushika Jayawardene, Dr.Veranja Liyanapathirana and Dr.Asanka Tennegedara from Department of Microbiology, Faculty of Medicine, University of Peradeniya.

Ethical clearance was obtained from the Ethics Review Committee, Faculty of Medicine, University of Peradeniya under the Project No.2020/EC/36.

We invite you to kindly complete this questionnaire by clicking the appropriate boxes.

We invite you to answer this questionnaire individually, without discussing or referring to content.

Your responses, will be treated as confidential and analyzed data will be used for the sole purpose of scientific communications and content development.

Further information on the study is available from https://drive.google.com/file/d/1ALCJrQkzIJwLdF-mPacLkm5zXxtYnkGS/view?usp=sharing

Your clicking the "next" button here will be taken as giving consent to participate in this study.

THANK YOU for choosing to fill the questionnaire independently without any help.”

As this was clearly stated at the start of the questionnaire and the participants had access to the information sheet, along with the fact that no personal identifiers were collected or analyzed and that participants were medical officers who are sufficiently competent to comprehend the introductory section of the questionnaire, authors feel that this information and way of obtaining consent is sufficient.

3. Thank you for stating the following in the Competing Interests/Financial Disclosure * (delete as necessary) section:

(VL has received funding from Pfizer for a non-related study. Others declare no competing interests)

We note that you received funding from a commercial source: (Pfizer)

Please provide an amended Competing Interests Statement that explicitly states this commercial funder, along with any other relevant declarations relating to employment, consultancy, patents, products in development, marketed products, etc.

Within this Competing Interests Statement, please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your amended Competing Interests Statement within your cover letter. We will change the online submission form on your behalf.

The amended competing interests statement included within the cover letter reads as “VL has received funding from Pfizer for a non-related study. Others declare no competing interests. "This does not alter our adherence to PLOS ONE policies on sharing data and materials.”

I would also like to emphasize that this study received not funding from anywhere.

4. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

The two supplementary tables were renamed. And the list was included at the end of the manuscript before the references.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

Reviewer #3: I Don't Know

________________________________________

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No – Authors would like to inform that the completed database has been uploaded as a supplementary material which the reviewer may have missed

Reviewer #3: Yes

________________________________________

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

________________________________________

5. Review Comments to the Author

Reviewer #1: PONE-D-21-34680

Research Article

Knowledge and practices on antibiotic use among Sri Lankan doctors

Reviewer’s Comment

General Comments: The manuscript requires copyediting for both content and English language.

Thank you, we have gone through and improved the manuscript where possible.

Specific Comments:

Abstract:

Introduction: It would be better to avoid using abbreviations in abstract section. If it is used, follow only the internationally adopted abbreviations. I also suggest the author to use the term ‘antimicrobial’ rather than antibiotic.

Thank you. We have replaced antibiotic resistance with antimicrobial resistance and used the abbreviation AMR, which is widely known now.

Methods: The information provided in method section lack vital information about how the questionnaire was developed? Who were the study participants? How questionnaire was applied? How validation of the questionnaire was performed? How the scoring was executed? How validated response can be obtained through google form? Likert scale questions are generally used to assess the attitude of the respondents, but here it has been used for knowledge/practice?

Thank you. We have included the sentence, “The Google sheet generated was used for data analysis”. Given the restriction on the number of wording, authors are unable to include all requested details in the methods section within the abstract. Details on how the marking was done is given in the methods section in detail along with detailed mark break-down in the supplementary material. Participants are indicated to in the objectives, therefore a repetition was not done in the methods section.

Key words: The authors should make sure the keywords are MeSH words, particularly knowledge and practices,

Thank you. The keywords were updated to

Health Knowledge, Attitudes, Practice

Antibiotics

Main body:

Introduction: Reorganize this section in different paragraph with background information, problem statement, rationale of the study

Authors feel that the current organization of the introduction includes background information, a problem statement and the rationale therefore no change was done to the introduction.

Methodology: It will be worthwhile mentioning the qualification and experiences of the experts who were involved in validation of the questionnaire.

The methods section already mentioned that the questionnaire was validated by a medical educationist and two consultant microbiologists. Now it reads as “It was validated by a Specialist Medical Educationist and two Consultant Microbiologists with experience in similar studies and relevant clinical practice respectively.”

How reliability of the questionnaire was performed? Why different scoring was used, e.g. 0.25 score, 1 score, negative score, 0 score, 2 score? These are too confusing! The term ‘mark/s’ should better be replaced with the term ‘score/s’.

Thank you for pointing this out. We accept that the assignment of marks may be a bit confusing on first reading. However, the supplementary tables indicates the allocation in a clear way. These were decided up on by all researchers. We have replaced the mark/s with score/scores where appropriate but retained the word mark where it is more appropriate.

The information provided in method section lack vital information about how the questionnaire was developed? Who were the study participants? How questionnaire was applied? One respondent can fill the google questionnaire multiple times. How validated response can be obtained through google form? Duration and location of the study?

Thank you. The methods section now reads – “This was a cross-sectional study conducted using recruiting 262 participants from various fields of Medicine working in different units in Sri Lanka. The ethical clearance was obtained from the Ethics Review Committee, Faculty of Medicine, University of Peradeniya, Sri Lanka (2020/EC/36) and participants indicated consent for participation through pressing the next button as informed through instructions. The questionnaire was Google form-based. It was validated by a Specialist Medical Educationist and two Consultant Microbiologists with experience in similar studies and relevant clinical practice respectively. It was disseminated to doctors through hospital based social media groups. All doctors practicing in Sri Lanka, registered with the Sri Lanka Medical Council were eligible to answer the questions. One attempt was allowed per-email address; therefore a single response was taken per-participant. The questionnaire was open to accept answers from 15/10/2021 to 12/01/2021”. We hope this gives sufficient information.

Likert scale questions are generally used to assess the attitude of the respondents, but here it has been used for knowledge/practice?

Thank you. While Likert scales are usually used for attitudes, they are been used to measure knowledge and practices previously too. We have converted the marks and the methods of conversion are also given.

Results: the number of unanswered responses are seen in few cases only, it would be better to merge them as ‘Don’t know/Unanswered’(table 2).

Thank you for the suggestion. As for the rest of the analysis, the unanswered segments vary, and it is important to identify them, we prefer to keep the unanswered segment separately in Table 2 as well.

Why the heading for Table 4 is below the table? Full form of the abbreviations used in table 4 should be mentioned in footer of the table.

The title of table 4 is placed above it. It may be due to a technical error in generating the pdf that reviewer has seen it such. The footnote defining abbreviations were added.

Table 6 is about the perception on antibiotic use, whereas there is no any word about perception in the title of the manuscript? It is quite unusual to use 4 points likert scale (table 6). It can be made into 3 scales (agree, Neutral/unanswered, disagree).

Thank you very much. The title was changed to include perceptions. We feel that it is important to highlight the non-response rate, therefore would like to keep the two columns separately, unless the editors would like the two columns to be combined as well.

Practice can better be determined through observation of real prescriptions (table 7) not through asking the questions. The responses for this section through questionnaire particularly online will be biases-how many will say that they do malpractice???

Thank you for pointing this out. However, even directly observed studies will have this bias. We believe that the fact that participants have acknowledged some malpractices and the anonymous nature of the questionnaire will improve correct responses. However, we have now included this as a shortcoming of the study.

It would be meaningful to use the median score rather than mean score.

We agree, therefore we have given both the mean and the median scores in all places.

There are nine tables in this manuscript. It is advisable to reduce the number of tables if feasible through merging or modifying the tables.

Thank you. We have limited the tables as much as possible to get the nine tables. Editorial recommendations on which further tables to merge would be most welcome.

Discussion: Discussion require more comparisons and clarifications. Study limitations and recommendations are vital components of a manuscript.

Thank you. We have now included a limitations section and the recommendations were already included in the conclusion section.

Conclusions: This section should be short and conclusive with key message from the findings.

Thank you for pointing this out. We believe that it contains the main messages only and includes the recommendations and have renamed the section as conclusions and recommendations. If the editors wishes us to do, we can modify the section and remove the recommendations.

References: Make sure the references and their citations in the text are as per the journal’s requirements.

Reviewer #2:

1. Title - did the authors also study the aspect related to resistance?

We did not ask about resistance from the point of view of discussing about resistance. So would like to keep it as antibiotics. However, in response to reviewer 1, we have added “perceptions” to the title

2. Abstract - ABR or AMR?

Thank you. ABR has been replaced with AMR

3. The objective in the abstract, main text as well as findings and conclusions should be consistent; look at them very carefully - the terms use, resistance, prescription etc were used inconsistently

Thank you for pointing out, we have gone through the manuscript again and modified where we felt it was needed.

4. Abstract - methods - too brief, how was the data analyzed?

We expanded the methods section to include administration of the questionnaire and some other aspects. We feel that the data analysis is already sufficiently described. However, if the editor wishes so, we would be able to expand it a bit more.

5. This topic has been studied in the past and many articles have been published. What the study adds to the literature, practice and policy? What is new?

6. main text - obj - I thought the study also look at knowledge and practice related ABc resistance!

We look at the antibiotic use, in the context of AMR. This was added to the text. The main focus is actually not AMR but the current level of KAP on antibiotics in the context of AMR.

7. Methods - was poorly written - non structured and non systematic way of writing the methods section - difficult for readers to follow and not easy to be replicated;

Thank you, we have revised the methods section to be more systematic. If the editors wishes us to do so, we are agreeable to move the description on the scoring system to the supplementary material, along with the two tables.

what was the sample size and power of the study?

The sample size was calculated to be 384, but we analyzed the received samples, and we have now stated that convenience sampling was used. It is globally knowns that physicians are a very difficult group to do surveys on, therefore, authors feel that the sample size available is adequate.

Cho YI, Johnson TP, Vangeest JB. Enhancing surveys of health care professionals: a meta-analysis of techniques to improve response. Eval Health Prof. 2013 Sep;36(3):382-407. doi: 10.1177/0163278713496425. PMID: 23975761.

Taylor T, Scott A. Do Physicians Prefer to Complete Online or Mail Surveys? Findings From a National Longitudinal Survey. Eval Health Prof. 2019 Mar;42(1):41-70. doi: 10.1177/0163278718807744. Epub 2018 Nov 1. PMID: 30384770.

How the respondents were selected?

As stated above, and in the methods section, anyone who was lisenced to practice medicine in Sri Lanka were eligible to participate. This is indicated in the method section.

the main domain - what is the possible min and max values/scores for practice? what are the possible min and max scores for knowledge?;

Thank you for the comment. The following paragraph has been added to the methods section “The maximum mark achievable for knowledge score was 14.75 while it differed for the practice score depending on if participants were engaged in hospital practice and/or private practice. All total marks were converted to %, so the maximum achievable mark was 100% while the minimum was 0%.”

Further, this was already presented in the results section but we deleted it after including in the methods section.

Was any psychometric measures done on the tool? Was the study piloted? alpha value?

The tool was piloted, it is mentioned in the methods section. No psychometric analysis was done. We did not measure the alpha value. The use of alpha value for validation of questionnaires is being debated. As our study population is relatively homogenous, authors feel that calculation of alpha value is not needed.

Taber, K.S. The Use of Cronbach’s Alpha When Developing and Reporting Research Instruments in Science Education. Res Sci Educ 48, 1273–1296 (2018). https://doi.org/10.1007/s11165-016-9602-2

What software was used for the analysis?

Thank you, this was included in the methods section now. SPSS version 23 was used for data analysis.

Mean values should be followed by sd, and median with IQR

Mean values were given with the SD, it was added to where it was not given, IQR was added to the median

8. any limitations of the study? what are the study implications? any recommendations?

Limitations section was added while the conclusion section was changed to conclusions and recommendations

Reviewer #3: Authors have done a good work. There are some comments and suggestions in the manuscript submitted. Please go through the comments and suggestions and do the necessary corrections for a revised submission.

Thank you. However, the manuscript with comments was not available either as an attachment, within the email or in the journal’s submission portal.

________________________________________

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Mukhtar Ansari

Reviewer #2: Yes: Mohamed Izham Mohamed Ibrahim

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Reviewer Comment.doc

Click here for additional data file.

27 Dec 2021

3 Jan 2022

Author response to reviewer comments

Comments to the Author

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Author response: We have corrected the manuscript and hope the language it clearer now

Reviewer #3: Yes

________________________________________

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Authors are advised to address all of the comments. The write up of the manuscript still requires language editing.

Author response: We could not find any reviewer comments to address. We had given our responses to the reviewer comments in our previous revisions, we would be most grateful if the reviewer can specify which comments need further addressing.

Thank you for suggesting language edits, we have attended to it. We hope the reviewer and the editors are satisfied with the changes made and look forward to any specific comments that may still need corrections.

Reviewer #3: The authors have answered to all the queries made by the reviewers. The manuscript now in a better form and ready to be accepted.

Submitted filename: 03 01 2021 Point by point rebuttal.docx

Click here for additional data file.

13 Jan 2022

Knowledge, perceptions and practices on antibiotic use among Sri Lankan doctors

PONE-D-21-34680R2

Dear Dr. Liyanapathirana,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Pathiyil Ravi Shankar

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have addressed all of the comments, and the manuscript looks fine after the revisions have been made.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Mukhtar Ansari

19 Jan 2022

PONE-D-21-34680R2

Knowledge, perceptions and practices on antibiotic use among Sri Lankan doctors

Dear Dr. Liyanapathirana:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Pathiyil Ravi Shankar

Academic Editor

PLOS ONE