Factors influencing preoperative chest radiography request for elective endoscopic procedures among medical personnel.

BACKGROUND: Chest radiography is not routinely recommended before elective endoscopies. A high incidence of perioperative chest radiography requests was observed at our institution. This study aims to investigate factors influencing preoperative chest radiography request for patients undergoing elective gastrointestinal (GI) endoscopies.
METHODS: This cross-sectional clinical study recruited 264 participants from different medical specialties who were responsible for preoperative endoscopic chest x-ray (CXR) ordering including anesthesiologists, surgeons and gastroenterologists. They completed questionnaires exploring their general knowledge and attitudes about preoperative chest radiography. Demographic characteristic of the participants affecting the knowledge on preoperative chest radiography was determined. A Structural Equation Model (SEM) was constructed from validated conceptual framework to find causal relationships between hypothesized factors and intention for preoperative endoscopic chest radiography request. Statistical analyses were performed using the SPSS software version 18.0 and Analysis of Moment Structures (AMOS) version 18.0.
RESULTS: The questionnaire response rate was 53.79%. Baseline general knowledge on preoperative chest radiography of the participants was comparable. The SEM results showed unsupported relationship between hypothesized factors and the intention for preprocedural GI endoscopic CXR request (p < 0.1).
CONCLUSIONS: General knowledge of medical personnel on tuberculosis needs improvement. To rectify the unnecessary chest radiography request before elective GI endoscopic procedures, awareness of the patients' health conditions, adherence to the hospital's policy and realizing of possible patient-related mishaps are not the determinants for preprocedural endoscopic chest radiography request. Future works are required to explore other alternative factors involved for reducing chest radiography requests which are not indicated.

Introduction

Preoperative assessments are considered prerequisite for patients undergoing surgical procedures in elective or emergency settings. Adequate tests are helpful in identifying and determining risks, optimizing anesthetic techniques to reduce morbidity and mortality, directing postoperative management, and preventing prosecutions in case there are any adverse events during the operations [1]. However, preoperative screening can be costly, resource intensive, time-consuming, and uncomfortable for patients. Ideally, the ordering of preprocedural testing should be based on data gathered from patients, comprising their histories, comorbidities, and significant findings from physical examination (American Society for Gastrointestinal Endoscopy (ASGE) guidelines [2]. Notwithstanding the paramount need for preoperative testing, several studies have concluded that no benefit is provided by routine testing in cases of elective, low-risk, ambulatory surgery for either adult or pediatric patients [3, 4].

Chest radiography is one of the frequent paradigms of preoperative tests covered by recent preoperative guidelines for elective surgery. It is recommended for patients over 60 years of age, especially those with a history of smoking, an American Society of Anesthesiologists (ASA) physical status of 3 or higher, a respiratory tract infection, signs and symptoms of cardiopulmonary disease, and decompensated heart failure [5-8]. In contrast, chest radiography is not routinely recommended before certain elective surgeries, especially endoscopies, because the incidence of radiographic figures that alters clinical outcomes is as low as 0.1% among all abnormal images [9, 10] and the endoscopic procedures themselves are regarded as low risk, having a cardiac event risk of less than 1%. Previous studies reported that there were insufficient data to determine the benefits of routine laboratory testing before endoscopic procedures; nevertheless, surgeons were inclined to unnecessarily request routine laboratory and preoperative screening tests [6, 11–14]. Less than one per cent of tests from all patients have been reported to reveal abnormalities that could affect perioperative outcomes [15]. Routine testing in cases of low-risk surgery may result in extra testing, exposure to radiation, surgery cancellation, increased patient anxiety, and additional hospital expenses [4, 5, 16, 17]. In our institution, Siriraj Hospital, approximately 78% of the patients scheduled for non-neurological and non-cardiovascular-thoracic operations were proceeded to preoperative chest radiography [18]. In addition, from the chart reviews of the patients scheduled for elective gastrointestinal endoscopic procedures in Siriraj Hospital, 52.1% were reported with abnormal chest radiography which cardiomegaly predominated. However, active pulmonary lesion accounted for only 0.2% [19, 20]. Still, a number of physicians express concern about tuberculosis in patients awaiting scheduled surgeries, especially in endemic areas where the prevalence is high, such as in Asian countries [21]. As evidenced by the Global Tuberculosis Report of the World Health Organization 2016, Thailand is among the 14 countries with the highest disease severity [21, 22]. Although there are around 120,000 new cases per annum (equivalent to 171 cases per 100,000 of the Thai population), only 55.3% of the cases are detectable. It is therefore still controversial whether chest radiographic examinations form a useful part of preoperative evaluations and should be no longer considered as mandatory by preoperative guidelines.

In an investigation of the causes of unnecessary testing, Brown et. al. found several influencing factors, such as practice traditions, the belief that other physicians want the tests done, medicolegal concerns, a desire to avoid surgical delays or cancellations, and a lack of awareness of evidence and guidelines [23]. However, the study depended on the semi-structured format interview and thus no model on factors affecting preoperative chest radiography request was proposed. We hypothesized that knowing the root causes that resulted in preoperative chest radiography request would aid in developing measures or guidelines to curb unnecessary requests. Our study was designed with the primary objective to determine the factors that influenced medical personnel in the residency and fellowship training programs about the importance and necessity of preoperative chest radiography request in patients undergoing gastrointestinal endoscopy. Secondary objective of the study was to assess the baseline general knowledge of medical personnel on preoperative chest x-ray (CXR) and tuberculosis.

Materials and methods

Study design and participants

This cross-sectional clinical study was conducted at a tertiary-care, university-based hospital. Two hundred and sixty-four participants were recruited from the residency and fellowship training programs of the Departments of Anesthesiology (80 individuals), Surgery (171 individuals), and Gastrointestinal Medicine (13 individuals), all under the Faculty of Medicine. The inclusion criteria were medical personnel who agreed to participate and were able to comprehend Thai language. The exclusion criteria were medical personnel who refused to participate in the study. Sample size calculation was performed based on primary objective which was to identify factors contributing to preoperative endoscopic CXR request. The calculation formula for Structural Equation Model (SEM) [24]; anticipated effect size 0.5, desired statistical power level 0.8, number of latent variables 4 and probability level 0.05, recommended minimal sample size (n) of 116.

The conceptual framework was constructed based on previous study on the causes of unnecessary preoperative testing (Fig 1) [23]. The model proposed 3 hypotheses which might reflect the motives of medical personnel toward preoperative CXR request for GI endoscopy. The 3 hypotheses included H1: awareness of patients’ history and co-existing diseases (PHC), H2: adherence to the hospital guideline and policy (HGP) and H3: prevention of patient-related complications (PPC). The questionnaire was subsequently generated by 5 specialists from different fields of expertise (namely, anesthesiology, surgery, and psychology) with reference to ASA and The National Institute for Care and Health Excellence (NICE) guidelines for preoperative CXR [5, 7], to assess knowledge and attitudes about preoperative chest radiography of medical personnel. The questionnaire was originally created in Thai language (S1 Appendix; translated to English with the assistance of content expert for publication). The aim of the questionnaire was to determine the factors involved in decision-making by the residents and fellows with regard to their preoperative chest radiography requests. The Index of Item-Objective Congruence (IOC) of the questionnaire was 0.91. Internal consistency of the questionnaire was confirmed by anesthesia alumni and residents from other institutions (Cronbach’s alpha coefficient 0.896).

Fig 1

Model of research.

The questionnaire was categorized into 3 main sections. Part 1 dealt with general information on the participants, comprising their age, sex, education, the number of gastrointestinal endoscopic treatments during the preceding 3 months, responsibility for ordering preoperative CXR, and participation in preoperative CXR evaluation courses. Part 2 focused on the participants’ basic knowledge of the necessity and importance of preprocedural or preoperative chest radiography, drawing upon information contained in the guidelines of ASA and NICE, UK. The knowledge test consisted of 18 questions about preoperative CXR for general and ambulatory surgery, CXR indications and knowledge of tuberculosis. Finally, Part 3 addressed the participants’ attitudes towards preprocedural or preoperative requests for chest radiography. Each item under the basic knowledge category was a yes–no question, while the items under the attitude category were rated in Likert-scale fashion, such as “extremely important” (5), “very important” (4), “moderately important” (3), “slightly important” (2), and “very slightly important” (1).

The general knowledge questions about preoperative CXR and tuberculosis were categorized into 3 main topics. The first, “patient’s age”, dealt with the appropriate age for preoperative CXR (question number 9, 10, 12; Q9, Q 10, Q12). The second class of questions, “low-risk surgery and health concerns” (Q14, Q17, Q20), examined the need for CXRs in cases of low-risk surgical procedures or patient health conditions, such as heart disease and asthma. The final category, “TB awareness” (Q24, Q25, Q26), related to the need for preoperative CXRs in particular situations, such as patients with a TB history or as part of TB surveillance activities. Other knowledge questions which did not involve in ambulatory or low-risk surgery were excluded from the analysis. The attitude section consisted of 8 major questions (Q); 4 major questions (Q27-30) comprised 21 sub-questions (Q27.1–27.6, Q28.1–28.5, Q29.1–29.5 and Q30.1–30.5) and 4 other questions without sub-questions (Q31-34). The attitude questions determined the opinion of the medical personnel toward the necessity of preoperative CXR request and the opinion toward Siriraj pre-anesthesia clinic (SiPAC) recommendation in term of proper CXR request. The attitude questions were categorized into 3 main topics based on proposed model of research (Fig 1) i.e. PHC (Q27.1–27.5), HGP (Q29.4–29.6) and PPC (Q29.1–29.3). The PHC questions contained the question contents similar to those in the knowledge questions but with Likert-scales rather than yes-no answers. Also, a topic related to necessity of chest radiography before elective GI endoscopy following institutional (Siriraj Preanesthetic Clinic; SiPAC) guideline or as individual’s opinion (CXR) (Q31-34) was regarded as outcome determinant. Other attitudes questions which were not related to endoscopic surgery were excluded from the analysis.

Data collection

The questionnaires were delivered to the residents and fellows in closed envelopes to the participant’s departments of original affiliation in Siriraj Hospital and let the residents or fellows voluntarily participate. Informed consents were obtained from the participants before their entry into the study. A research assistant requested the participants to return the questionnaires within 2 weeks in provided closed containers. If the participants failed to return the questionnaires within the time limit, they would be asked to return the questionnaires one more time. After the participants had completed the questionnaires, the items were calculated and summarized. The protocol for the study was approved by the Siriraj Institutional Review Board, Faculty of Medicine, Siriraj Hospital, Mahidol University, Bangkok, Thailand (Ethical number SI 651/2017).

Statistical analysis

Descriptive statistic was used to define the demographic characteristics of the respondents by mean, percentage, and standard deviation. Statistical significances of general knowledge score among participants were determined by chi-square test. A p-value of less than 0.05 was regarded as statistically significant. A Confirmatory Factor Analysis (CFA) and a Structural Equation Model (SEM) using the maximum likelihood (ML) method were performed. Internal consistency of factors for CFA was assessed by Cronbach’s alpha coefficient. The statistical analyses were performed using SPSS Statistics for Windows, version 18.0 (SPSS Inc., Chicago, Ill., USA). Analysis of Moment Structures (AMOS) version 18.0 was used for CFA and SEM analyses.

Results

The questionnaires were distributed to the medical personnel during 4 December 2017–30 January 2018. Although 264 medical personnel were recruited to the study, only 142 (53.79%) returned completed questionnaires. Of those respondents, 54 were from the Department of Anesthesiology, 77 from Surgery, and 11 from Gastrointestinal (GI) Medicine. Surgeons represented the major subpopulation group (54.22%), followed by anesthesiologists (38.03%) and gastroenterologists (7.75%). Residents were classified according to year of training. Anesthesiology residents attended a 3-year training program whereas Surgery residents completed 4–5 years of training. Some of the participants were Surgery fellows, the ones who further their studies on surgical sub-specialties. Gastroenterologists were medical personnel who participated in the sub-specialty fellowship training program. Table 1 details the baseline characteristics of the participants (age, gender, position in training programs, preoperative CXR course attendance, being a key individual to order preoperative CXRs, and GI endoscopic treatment experience during the preceding 3 months).

Table 1

Demographic data of the questionnaire respondents.

Variables	Mean ± SD*/n (%)
	Anesthesiologists(n = 54)	Surgeons(n = 77)	Gastroenterologists(n = 11)	Total(n = 142)
Age (Years)	28.24 ± 1.44	29.17 ± 2.35	31.27 ± 1.74	28.98 ± 2.15
Sex
Female	46 (85.2)	24 (31.2)	5 (45.5)	75 (52.8)
Male	8 (14.8)	53 (68.8)	6 (54.5)	67 (47.2)
Position
Resident**	54 (100)	69 (89.6)	-	123 (86.6)
1st	16 (29.6)	21 (27.3)	-	42 (29.6)
2nd	17 (31.5)	23 (29.9)	-	46 (32.4)
3rd	21 (38.9)	18 (23.4)	-	39 (27.5)
4th	-	11 (14.3)	-	11 (7.7)
5th	-	4 (5.2)	-	4 (2.8)
Fellow	-	8 (10.4)	11 (100)	19 (13.4)
Having attended preoperative CXR course	26 (48.1)	36 (46.8)	5 (45.5)	68 (47.9)
Being a key individual for ordering preoperative Chest x-ray	24 (44.4)	71 (92.2)	11 (100)	106 (74.6)
Being a part of Gastrointestinal endoscopic treatment in the preceding 3 months
0–10 cases	42 (77.7)	50 (64.9)	1 (9.0)	93 (65.5)
>10 cases	12 (22.3)	27 (35.1)	10 (91.0)	49 (34.5)

* SD, Standard deviation

**Classified according to year of training

Data presented as mean ± SD or number (%).

Overall baseline general knowledge on preoperative CXR and tuberculosis among 3 medical specialties did not vary. However, in the case of Q9, which asked whether all patients of any age who are undergoing surgery should have preoperative CXRs, the gastroenterologists had a significantly higher proportion of correct answers (36.4%) than the other two work groups. Another noticeable difference was observed with Q14, which asked whether a patient with non-active asthma requires preoperative chest radiography. For that question (Q14), surgeons had the highest proportion of correct answers (79.2%) (Table 2). The questions that belonged to each topic which was regarded as factor in the conceptual framework (Fig 1) expressed good internal consistency as determined by Cronbach’s alpha and were then subjected to Confirmatory Factor Analysis (CFA) (Table 3). The CFA results demonstrated a fitness of data into the hypothetical model: root mean square error of approximation (RMSEA) = 0.063 (< 0.08); comparative fit index (CFI) = 0.944 (≥ 0.90); goodness of fit index (GFI)_ = 0.906 (≥ 0.90); adjusted goodness of fit index (AGFI) = 0.852 (≥ 0.80); and Chi-square/df = 1.559 (< 3).

Table 2

The respondents’ scores for the chest x-ray knowledge questions.

Topics of general knowledge questions	Respondents with correct answers: n (%)			p-value
	Anesthesiologists(n = 54)	Surgeons(n = 77)	Gastroenterologists(n = 11)
Topics related to patient’s age
Q9 Every patient	4 (7.4)	23 (29.9)	4 (36.4)	0.004*
Q10 Over 45 years old	48 (88.9)	68 (88.3)	10 (90.9)	0.967
Q12 All pediatric patients	50 (92.6)	69 (89.6)	9 (81.8)	0.536
Topics related to low-risk surgery and health concerns
Q14 Colonoscopy in non-active asthma	31 (57.4)	61 (79.2)	4 (36.4)	0.002*
Q17 Cataract surgery in CKD requiring hemodialysis	41 (75.9)	44 (57.1)	6 (54.5)	0.069
Q20 EGD in no underlying disease	38 (70.4)	58 (75.3)	6 (54.5)	0.342
Topics related to TB awareness
Q24 Prevalence of TB	14 (25.9)	28 (36.4)	1 (9.1)	0.124
Q25 Incidence of TB	37 (68.5)	56 (72.7)	9 (81.8)	0.649
Q26 Specificity of chest x-ray for TB	13 (24.1)	20 (26.0)	2 (18.2)	0.848

*Significant at p < 0.05 by Chi-square test

Abbreviations: Q, Question; CXR, Chest x-ray; CKD, Chronic kidney disease; EGD, Esophagogastroduodenoscopy; TB, Tuberculosis

Table 3

Factors used for the Confirmatory Factor Analysis (CFA).

Factors	Question Number	Measurement variable	Cronbach’s Alpha
Patients’ history and co-existing diseases(PHC)	Q27.1	History of pulmonary tuberculosis	0.823
	Q27.2	History of heart disease
	Q27.3	History of chronic obstructive pulmonary disease
	Q27.4	History of upper respiratory tract infection
	Q27.5	Smoking
Hospital guideline and policy(HGP)	Q29.4	Prevention of prosecution should there be adverse events during the operation	0.715
	Q29.5	Following the hospital’s policy
	Q27.6	Healthy patient older than 45 years with no underlying diseases
Prevention of patient-related complications(PPC)	Q29.1	Avoidance of operation cancellation by surgeons and anesthesiologists	0.641
	Q29.2	Tuberculosis surveillance
	Q29.3	Prevention of risks or complications during the operation
Chest radiography request before elective GI endoscopy(CXR)	Q31	Do you consider that the current preoperative evaluation guidelines of Siriraj Preanesthetic Clinic (SiPAC) could reduce and prevent complications during the operations	0.616
	Q32	Do you consider that following the current SiPAC preoperative evaluation guidelines could prevent cancellations of operations by surgeons and anesthesiologists?
	Q33	Do general medical personnel strictly follow the SiPAC preoperative evaluation guidelines
	Q34	Are you concerned about adverse events or complications during an operation if chest radiography for the patient is not available preoperatively

Abbreviation: GI, Gastrointestinal; SiPAC, Siriraj Preanesthetic Clinic

Structural Equation Model (SEM) was established to demonstrate the associations between hypothesized factors with the outcome, the intention to request preoperative CXR for elective endoscopic procedures (Fig 2). The SEM was assessed for fitness of data after model adjustment which resulted in acceptable values for all indices: RMSEA = 0.066 (< 0.08); CFI = 0.936 (≥ 0.90); GFI_ = 0.901 (≥ 0.90); AGFI = 0.848 (≥ 0.80); and Chi-square/df = 1.918 (< 3) (Table 4). Among all hypothesized factors including PHC, HGP and PPC, none of them was statistically associated with an intention for preoperative endoscopic CXR request as determined by the p-value > 0.05 (Table 5).

Fig 2

The Structural Equation Model (SEM) used for analyzing the causal relationships between factors and chest radiography request before elective GI endoscopic procedures.

Table 4

Fit indices for the Structural Equation Model (SEM) before and after adjustment.

Index	Criterion	Statistical values obtained from analysis
		Before adjustment	After adjustment
χ2/df (CMIN/df)	<3	2.512	1.918
GFI	≥0.90	0.836	0.901
AGFI	≥0.80	0.766	0.848
CFI	≥0.90	0.832	0.936
TLI	≥0.90	0.790	0.914
RMSEA	<0.08	0.104	0.066
Model summary		Lack of Fit	Acceptable Model Fit

Table 5

Hypothesis result of the structural model.

Hypothesis	Standardized path coefficients (β)	T-value	p-value	Test result
H1: Patients’ history and co-existing diseases	0.127	1.676**	0.094	Un-Supported
(PHC) => CXR
H2: Hospital guideline and policy	0.038	0.347	0.728	Un-Supported
(HGP) => CXR
H3: Prevention of patient-related complications	-0.162	-0.950	0.342	Un-Supported
(PPC) => CXR

Remark:

***p<0.01,

**p<0.1 and

*p<0.05

Discussion

Our study investigated the factors influencing preprocedural CXR request for gastrointestinal endoscopy among medical personnel. Overall baseline general knowledge on preoperative CXR was comparable between different medical specialties. The validated questionnaire incorporated sets of questions that led to the factors relating to preoperative CXR request. The factors according to the hypothetical model in association with the intention to perform preoperative CXR for GI endoscopy were validated by CFA and further analyzed with SEM. Nevertheless, no associations between hypothesized factors (PHC, HGP and PPC) and outcome as determined by CXR request (CXR) were observed.

It is well documented that preoperative chest radiography, especially in ambulatory settings, might not be essential. This is because only 12% of patients are found to have abnormal CXR findings and, interestingly, as few as 0.03% need further investigation and treatment [9]. Chest radiography is therefore recommended for patients over 60 years of age and is not routinely required for before elective endoscopic procedures [2, 6, 7]. For other minor surgeries, the requirement of preoperative chest radiography could be omitted because it did not alter patient management and rate of postoperative pulmonary complications or adverse events [25-28]. Instead, vigilant history taking and physical examination might be adequate for preoperative preparation [29]. Despite the establishment of international guidelines on preoperative chest radiography over the past 3 decades, there are still no official institutional and Thai national guidelines being issued. In addition, proper guidelines for the preoperative preparation of elective ambulatory cases are yet to be established. As a result, a trend toward unnecessary chest radiography requirement before minor elective operation especially gastrointestinal endoscopy is frequently observed among medical profession. This may be due to the fact that Thailand is an endemic area of tuberculosis as defined by World Health Organization (WHO) [21]. Other possible reasons included intention to complete preoperative preparation or avoidance of case cancellation according to lack of adequate preoperative testing [23]. For these reasons, it would be invaluable to clarify the rationale for preoperative CXR requests. Our study focused on a different population of medical professions—anesthesiologists, surgeons, and gastroenterologists—who had experience with patient evaluations and the requesting of tests preoperatively. We assessed their knowledge and attitudes by analyzing their answers to a validated questionnaire. Taken previous established guidelines and research together [5, 7, 23], we generated a questionnaire that comprised the general knowledge and attitude towards preoperative CXR. Even though our focus was on elective GI endoscopic procedures, other questions irrelevant to endoscopy were present in order to prevent bias when the participants answer the questionnaire.

Referring to the general knowledge questions, most of the surgeons and gastroenterologists were responsible for ordering preoperative CXRs, which might be why they scored higher than anesthesiologists on Q9 (whether all patients undergoing surgery needed preoperative CXRs). Also, surgeons were striking leaders in terms of the percentage of correct answers to Q14 (whether a non-active asthmatic patient requires a preoperative CXR). The medical personnel who had been specially trained in medicine had had a greater chance of encountering tuberculosis patients who, typically, do not show clinical symptoms yet are still contagious. Therefore, the reason why the gastroenterologists exhibited the lowest percentage of staff with correct answers to the questions related to the prevalence and specificity of CXR for tuberculosis (Q24 and Q26) might be because they are strongly concerned about patients at risk. However, the number of gastroenterologists was much less than the numbers for the staff in the 2 other groups, which is consistent with the limited number of training positions in gastroenterology. Therefore, better results and statistical significance might be achieved with an increased number of gastrointestinal-medicine respondents. The results of our analysis revealed no striking differences in the knowledge of all 3 groups of respondents. Nevertheless importance of TB for preoperative-CXR ordering needed to be highlighted among all medical professions according to relatively lower proportion of participants with correct answers comparing with other topics. The concept of identifying factors responsible for preoperative endoscopic CXR request was to find solutions for our institution to limit the unnecessary test ordering for better cost effectiveness and to reduce risk of radiation exposure. In spite of good model fit for both CFA and SEM, however, the hypothesized factors failed to express association with the outcome meaning that there might be other factors involved.

Our study is the first to find factors for endoscopic CXR request. Similar study regarding an evaluation of factors influencing preoperative testing prior to low-risk surgery was reported previously. The results from retrospective study analysis by Bayesian generalized linear mixed model suggested that institution size was the factor associated with excessive preoperative blood testing. Larger institutions were assumed to have enough resource for routine blood tests and such testing might be general practice of the institutions [30]. This result supported the speculation that there could be other factors besides patients’ characteristics and physicians’ perspectives that accounted for the behavior.

There were some limitations of the study we would like to address. Apart from the hypothesized factors, other non-clinical influences were not introduced to the framework model. As clinical criteria were regarded as important guidance for clinical decision-making, a variety of non-clinical aspects might be involved such as patient’s worries, physician’s time constraints and physician’s personal experience or belief [31]. Had these non-clinical factors been considered, significant relationship with the wish for preoperative endoscopic CXR might have been achieved. Another limitation was the actual practice of each individual on CXR request was not monitored therefore their actions might not be well predicted. Our future plan is to directly observe the actual practice on preoperative chest x-ray of the medical personnel after their knowledge and attitudes have been evaluated. Future research could determine alternative factors affecting the actual practices relating to preoperative-CXR ordering. Anyhow, knowledge and attitudes should still be properly enhanced by an establishment of official institutional and/or national guidelines.

Conclusion

The SEM of the conceptual framework was confirmed with good model fit. However, it failed to demonstrate the relationships between our hypothesized factor i.e. PHC, HGP and PPC and intention to request preoperative endoscopic CXR. Other non-clinical factors might be involved thus requiring further study.

Questionnaire for assessment of knowledge and attitudes of medical personnel about preprocedural or preoperative requests for chest radiography.

(PDF)

Click here for additional data file.

Raw data of knowledge and attitudes about preoperative chest radiography.

(XLSX)

Click here for additional data file.

12 Aug 2020

PONE-D-20-04685

Medical Personnel Knowledge and Attitudes About Preoperative Chest Radiography for Elective Endoscopic Procedures

PLOS ONE

Dear Dr. Siriussawakul,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Sep 24 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

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: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Nancy Beam, PhD

Staff Editor

PLOS ONE

Journal Requirements:

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

1. 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. In ethics statement in the manuscript and in the online submission form, please provide additional information about the patient records used in your retrospective study. Specifically, please ensure that you have discussed whether all data were fully anonymized before you accessed them and/or whether the IRB or ethics committee waived the requirement for informed consent. If patients provided informed written consent to have data from their medical records used in research, please include this information.

3. Please provide additional details regarding medical personnel consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified (1) whether consent was informed and (2) what type you obtained (for instance, written or verbal, and if verbal, how it was documented and witnessed). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this 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: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

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: Yes

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: Yes

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: General comments

- The methods used in this study are mostly sound. In parts, there has been inappropriate use of statistical tests and over-interpretation of results.

- The manuscript is written well and is largely intelligible throughout.

- It is helpful that the authors have included the raw data and the questionnaire used in the study.

- The discussion has new results that are better placed in the results section, does not reference sufficient literature to contextualise the results and leaves the reader without a discussion of the limitations or a conclusion.

- The results are, at times, difficult to read owing to long sentences and ambiguous references to questions in the questionnaire.

- The manuscript could be improved overall by being more explicit about what the main findings were and why these findings are relevant to the field.

- Almost half of the references used in the manuscript are more than 15 years old. Has the literature evolved since then? Have the methods used in pre-operative/pre-endoscopy investigation changed since then? Does this affect the assertions you make in your manuscript? Having conducted a quick literature search, I believe there have been several articles and international guidelines published on this topic over the past 15 years.

Introduction

- Well written. Well done.

- Establishes background, importance of work, problem and niche for solution.

- The wording of “Barely 1% of the tests from overall patients have been reported to reveal abnormalities that would affect perioperative outcomes” is a bit strange to me. My suggestion to improve this would be “Less than one per cent of tests from all patients have been reported to reveal abnormalities that could affect perioperative outcomes”

Methods

- Was the questionnaire validated and/or compared to previously designed questionnaires? If it was validated, then how was it validated?

- What method was used to generate the index of consistency? This should be described in full. Does 0.91 allow for group comparisons and/or individual comparisons? Assuming the use of Cronbach’s alpha, 0.91 would allow individual comparisons over time – have you anonymised but maintained an identifiable code for the participants to do so? It would be worth mentioning the answers to these questions.

- The authors should mention whether the questionnaire was carried out in English alone, English with Thai verbal help, spoken in Thai (with answers captured in English) and/or translated from Thai.

- How did the authors capture whether a respondent believed an attitude item was neither important nor unimportant? How did the authors capture whether a respondent was unsure of an attitude-related item – was this differentiated from a blank answer?

- The attitude section of the questionnaire in the manuscript includes 8 questions with 24 sub-questions but the manuscript says otherwise. Could you please update the manuscript to reflect this, unless I have misunderstood.

- I am very concerned by the high number of statistical comparisons that are being conducted in the subgroup analyses – have you performed a multiple comparisons correction method? If so, which one and why? For example, in table 5 the are approximately 85 comparisons. Using a standard p<0.05 alpha level cut-off would suggest that <4 of those comparisons could be false positives. Four of the total of 8 ‘statistically significant’ results presented in the table may be false positives. This is particularly relevant given that these comparisons are part of a sub-group analysis.

- Why was a Spearman’s rank correlation used for correlation of knowledge and attitudes if this is thought of as a non-parametric measure of correlation? I would suggest using Pearson’s correlation coefficient.

- Please provide the method used for sample size calculation of the retrospective chart review (could be provided as an appendix). The original article is in Thai and I can’t read it, apologies.

- Could you please clarify what you mean by “The preoperative chest radiography statuses of the patients were registered”? What do you mean by status?

- The questionnaire attitudes section gives the five Likert scale responses to answer this question “Do you consider that the current preoperative evaluation guidelines of Siriraj Preanesthetic Clinic (SiPAC) could reduce and prevent complications during the operations?” All possible answers to this question do not make sense – please provide any verbal/written feedback that was given to clarify this.

Results

- Table 1 and 2 notes: “*Significant at p < 0.05”. Please indicate what statistical test was used for this. I suggest you make captions stand-alone, such that they can be read without having to reference other sections (eg. methods).

- Table 1: Did you ask or allow respondents to identify as a non-binary gender? It is important to include these categories, even if there were no respondents that identified. Also, in the sub-group analysis tables you stratify according to sex but asked about gender in the questionnaire. Please update to reflect the information you received.

- In this section of the results, you have re-stated what is inherent in the table. There is no need to ‘double report’ this. Please summarise this or remove it from the text.

o “There were significant mean age differences between the 3 groups of medical professions (anesthesiologists, surgeons, and gastroenterologists), with the highest mean age being for gastroenterologists (31.27 ± 1.74 years), followed by surgeons (29.17 ± 2.35) and anesthesiologists (28.24 ± 1.44). The overall proportions of male and female participants were 47.2% and 52.8%, respectively.”

- The whole paragraph on demographic characteristics is effectively conveyed in the table. My suggestion would be to significantly cut the paragraph on demographic characteristics, for readability.

- The paragraph starting with “The knowledge and attitudes about preoperative chest radiography…” is largely method-related and I believe it would be better suited in the methods section. How does this paragraph differ to the questionnaire information you have detailed in the methods section?

- Were the sub-group analyses identified a priori or were they post hoc? Either way, I believe some justification is necessary as to why the sub-group analyses have been performed and some sort of adjustment for multiple comparisons must be made. As an aside, the other pertinent question is ‘do these analyses add to the paper and its overall message?’ My impression from having read the paper is no – I would suggest you make it more explicit as to why these analyses are important.

- Could you please detail what is meant by “good” and “poor” attitudes? What method was used to assess this?

- The idea of correlating attitude and knowledge scores seems problematic as I am unclear the exact method you have used to do this. Attitudes, while they have been captured here semi-quantitatively, are not quantitative and often knowledge is insufficient to change attitudes. I believe that correlating the two is fundamentally flawed.

- No data is presented in the results section regarding the retrospective patient chart review. Please include this in the results section.

Discussion

- The article would be improved by adding an initial short paragraph to the discussion that summarises the main findings of the study.

- Paragraph 1 of the discussion brings in new results and is better placed in the results section.

- Paragraphs 2 and 3 of the discussion summarise and attempt to rationalise the results. These two paragraphs would be strengthened by i) making the paragraphs more succinct by focusing on the most salient points, ii) by comparing and contrasting your results with the literature (how does your research fit into the broader literature on CXR prior to endoscopy?) and iii) justifying your assertions based on other published literature.

- The authors do not discuss the limitations of their study and the manuscript be strengthened by such a discussion.

- The authors do not provide a concluding paragraph emphasising the importance and utility of this research. The manuscript would be enhanced by this.

o How will this research be used at Mahidol University hospitals? Will there be an intervention carried out to improve the knowledge and attitudes of physicians?

- “This might be because of a better decision-making ability of female respondents” Did you assess decision-making ability? While this may be true, I do not see how your results justify this assertion. Please provide additional reasoning for this.

- Why is “inactive pulmonary infiltration” non-significant? This could represent latent tuberculosis which may significant in certain patient populations. Could you please clarify?

- Limitations

o Understand the knowledge and attitudes of practitioners does not always reliably predict their actions and course of practice. Have they addressed this?

o Low completion rate – why? Will this bias results?

o “Our study focused on a diverse population of medical professions—anesthesiologists, surgeons, and gastroenterologists—who had experience with patient evaluations and the requesting of tests preoperatively.” I disagree that this is a diverse population of medical professions – they are all specialties that have the potential to routinely order, and need knowledge of, endoscopies. Is there any potential bias that could result due to this?

Minor textual amendments:

- Page 5 of manuscript “2,030 outpatient charts produced January–November 2017” should read “2,030 outpatient charts produced between January and November 2017”

- Page 6 of manuscript “others dealt with general, preoperative, CXR indications, and knowledge of tuberculosis.” This sentence is unclear. Could you please clarify it.

- Page 6 of manuscript “The attitude test consisted of 11 questions” should read “The attitude section consisted of 11 questions.”

- Page 11 of manuscript “Our retrospective data indicated that that there” should read “Our retrospective data indicated that there”, delete the extra ‘that’

Reviewer #2: Dear author manuscript is well written but has some minor corrections to be done. the discussion part is however well written but results from other studies were not compared to the present study either similar / dissimilar.

Reviewer #3: Thank you for the opportunity to review this manuscript. The advantage of this study was conducted scientifically and systematically. The manuscript is well written and described clearly. Several suggestions to improve this manuscript are listed below.

The methods section of the abstract needs to contain more information, especially regarding the study location and duration. It will reflect the inference population. Research tool (the questionnaire) needs to be elaborated further as to whether a newly developed questionnaire or a validated questionnaire was used.

In this study, two study designs were employed, a prospective clinical study and an observational retrospective study. Is there any reason why it cannot be called a cross-sectional study since all the information is obtained at a single point of time with no element of follow-up?

What is the basis of taking 264 participants? Was it based on any sample size calculation? In this study, elaborate on how a sample size of 264 was determined. Any exclusion criteria for study participants? What does it mean by "incomplete medical records"? Who decides whether an x-ray is uninterpretable? Include a reference number for ethical clearance.

Define "resident" and "fellow".

This study uses a newly developed questionnaire to assess knowledge and attitude towards preoperative chest radiography for elective endoscopic procedures. It is important that the questionnaire is validated before it is being used for the actual study. Furthermore, the readers must be convinced that the questionnaire undergoes a proper questionnaire validation process as it will determine the validity of the results. "The Index of consistency score of the questionnaire was 0.91". Does this indicate internal consistency reliability? The assessment of content validity needs further elaboration (e.g. assessment of content validity index by a group of independent experts). Any reason why factor analysis (EFA and CFA) was not performed?

Elaborate sampling method. Any method employed to ensure the representativeness of the study participants in terms of the department and work experience?

What does it mean by "Statistical significances were determined by chi-squared tests"? Statistical significance refer to the p-value and can be obtained from all statistical test (not just the chi-squared test).

Rephrase "The data with a normal distribution were analyzed and reported as percentage, mean, and standard deviation". All data were analyzed. Data with normal distribution needs to be presented as mean and its standard deviation, whereas data with skewed distribution needs to be resented as median and interquartile range. Frequency and its percentage are used to describe categorical variables.

The sample size calculation needs to be placed under "study design and participants" and not under the method of statistical analysis.

The knowledge test consisted of 18 questions; 7 were about surgery in ambulatory settings, whereas the others dealt with general, preoperative, CXR indications, and knowledge of tuberculosis. The attitude test consisted of 11 questions. Elaborate method to obtain the total score for each domain.

The response rate in this study is low (53.79%). In terms of statistical power, was it still adequate? Discuss the response rate in relation to external validity of this study.

The description regarding the questionnaire and the methods of scoring for each item (paragraph two of the results section) needs to be placed under the methods section.

Lengthy text presentation of the statistical results. Most of the information can be obtained from the tables, and presenting the results in both text and tables are redundant.

In table 1, 2, indicate the statistical analysis conducted to obtain each p-value. Here, the comparison of numerical variables between three independent groups was conducted using one-way ANOVA, and for categorical variables using either the Chi-squared test or Fisher exact test. Methods of these statistical analyses need to be included in the methods section.

**********

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: Andrés Noé

Reviewer #2: Yes: Aisha Wali

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: plos one review.docx

Click here for additional data file.

24 Sep 2020

Manuscript number: PONE-D-20-04685

Journal: PLOS ONE

Title: Medical Personnel Knowledge and Attitudes About Preoperative Chest Radiography for Elective Endoscopic Procedures

Dear editor,

Thank you for giving a chance to improve our manuscript. Hopefully, you would appreciate our revised version. We described the change in response to the reviewers and in the manuscript.

Regards

Arunotai Siriussawakul

Associate professor

Department of Anesthesiology,

Siriraj Hospital, Mahidol University, Thailand.

Journal Requirements:

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

1. 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

Answer: We did as suggestion.

2. In ethics statement in the manuscript and in the online submission form, please provide additional information about the patient records used in your retrospective study. Specifically, please ensure that you have discussed whether all data were fully anonymized before you accessed them and/or whether the IRB or ethics committee waived the requirement for informed consent. If patients provided informed written consent to have data from their medical records used in research, please include this information.

Answer: We have decided not to include this set of data in the manuscript because it did not add to our primary or secondary objective of the study. Also, part of the data was discussed in our previous publication which was mentioned in the introduction part as ref 20.

3. Please provide additional details regarding medical personnel consent. In the ethics statement in the Methods and online submission information, please ensure that you have specified (1) whether consent was informed and (2) what type you obtained (for instance, written or verbal, and if verbal, how it was documented and witnessed). If your study included minors, state whether you obtained consent from parents or guardians. If the need for consent was waived by the ethics committee, please include this information.

Answer: The research assistance applied the questionnaires in closed enveloped to the residents and fellows at their departments of affiliation. The document explaining the details and objectives of the study was attached with the questionnaire. The persons who would like to participate voluntarily took the questionnaire. They were asked to return the questionnaires within 2 weeks in provided closed containers. If not, they would be asked to return the questionnaires again by the research assistance. The IRB committee had agreed to let us proceed with the study (ethic no. Si 651/2017).

________________________________________

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: No

Reviewer #2: Yes

Reviewer #3: Yes

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: Yes

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: Yes

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: General comments

- The methods used in this study are mostly sound. In parts, there has been inappropriate use of statistical tests and over-interpretation of results.

Answer: We have revised the data and used appropriate analysis as suggestion.

- The manuscript is written well and is largely intelligible throughout.

Answer: Thank you very much.

- It is helpful that the authors have included the raw data and the questionnaire used in the study.

Answer: Thank you.

- The discussion has new results that are better placed in the results section, does not reference sufficient literature to contextualize the results and leaves the reader without a discussion of the limitations or a conclusion.

Answer: We have removed that new result from the result section and added details including limitations and conclusion to the discussion section. Also, more related literatures were added.

- The results are, at times, difficult to read owing to long sentences and ambiguous references to questions in the questionnaire.

Answer: We have combined the tables with legends and supplementary appendix to the manuscript. Hopefully it will ease reviewing the result section.

- The manuscript could be improved overall by being more explicit about what the main findings were and why these findings are relevant to the field.

Answer: We have made new analysis which pointed out the factors that reflected intention for preoperative endoscopic chest radiography request.

- Almost half of the references used in the manuscript are more than 15 years old. Has the literature evolved since then? Have the methods used in pre-operative/pre-endoscopy investigation changed since then? Does this affect the assertions you make in your manuscript? Having conducted a quick literature search, I believe there have been several articles and international guidelines published on this topic over the past 15 years.

Answer: We have added updated references as suggestion.

Introduction

- Well written. Well done.

Answer: Thank you very much. We appreciate your comment.

- Establishes background, importance of work, problem and niche for solution.

Answer: Thank you.

- The wording of “Barely 1% of the tests from overall patients have been reported to reveal abnormalities that would affect perioperative outcomes” is a bit strange to me. My suggestion to improve this would be “Less than one per cent of tests from all patients have been reported to reveal abnormalities that could affect perioperative outcomes”

Answer: We did as suggestion.

Methods

- Was the questionnaire validated and/or compared to previously designed questionnaires? If it was validated, then how was it validated?

Answer: Thank you for your nice suggestion. The questionnaire was newly created. We validated the questionnaire before the data collection by Index of item-objective congruence (IOC) by 5 content experts in the field. We also test the reliability using Cronbach’s alpha assessment from Anesthesia alumni and residents from other institutions which were not included in the study. The details are mentioned in paragraph 2 of the methods section

- What method was used to generate the index of consistency? This should be described in full. Does 0.91 allow for group comparisons and/or individual comparisons? Assuming the use of Cronbach’s alpha, 0.91 would allow individual comparisons over time – have you anonymised but maintained an identifiable code for the participants to do so? It would be worth mentioning the answers to these questions.

Answer: Actually, we validated the questionnaire in this study by Index of Item-Objective Congruence (IOC) by 5 experts in the field. The IOC was 0.91 therefore good content validity was achieved. We had as well assessed the internal consistence of the questionnaire by using Cronbach’s alpha which was 0.896 (good reliability). We have already mentioned this in the materials and methods section.

- The authors should mention whether the questionnaire was carried out in English alone, English with Thai verbal help, spoken in Thai (with answers captured in English) and/or translated from Thai.

Answer: The original questionnaire was written in Thai. We translated it to English for publication with the assistance of the experts in the field.

- How did the authors capture whether a respondent believed an attitude item was neither important nor unimportant? How did the authors capture whether a respondent was unsure of an attitude-related item – was this differentiated from a blank answer?

Answer: We agreed with the points you raised. Attitude questions were in Likert pattern thus it would be quite difficult to determine which ones were important or unimportant. Therefore, we re-analyzed the data by using confirmatory factor analysis (CFA) and constructed structural equation model (SEM) to find out the factors that affected the preoperative chest x-ray request of medical personnel.

- The attitude section of the questionnaire in the manuscript includes 8 questions with 24 sub-questions but the manuscript says otherwise. Could you please update the manuscript to reflect this, unless I have misunderstood.

Answer: We have made a correction. Original attitude questionnaire comprised 8 major questions. We, however, picked only 11 sub-questions and 4 major questions that related to elective gastrointestinal endoscopic procedures and tuberculosis awareness. We have clearly elaborated this in the methods.

- I am very concerned by the high number of statistical comparisons that are being conducted in the subgroup analyses – have you performed a multiple comparisons correction method? If so, which one and why? For example, in table 5 there are approximately 85 comparisons. Using a standard p<0.05 alpha level cut-off would suggest that <4 of those comparisons could be false positives. Four of the total of 8 ‘statistically significant’ results presented in the table may be false positives. This is particularly relevant given that these comparisons are part of a sub-group analysis.

Answer: We agreed that sub-group analyses might well create false positive results therefore we decided to remove them from the result section. Additionally, we re-analyzed the data with more appropriate statistics method according to the objectives of the study.

- Please provide the method used for sample size calculation of the retrospective chart review

Answer: We decided not to include this information because it did not answer the research question. In addition, our members in the research group have used part of this information in the publication which is cited in this study.

- Could you please clarify what you mean by “The preoperative chest radiography statuses of the patients were registered”? What do you mean by status?

Answer: We decided not to include this set of data in the manuscript because it did not match with the objectives of the study.

- The questionnaire attitudes section gives the five Likert scale responses to answer this question “Do you consider that the current preoperative evaluation guidelines of Siriraj Preanesthetic Clinic (SiPAC) could reduce and prevent complications during the operations?” All possible answers to this question do not make sense – please provide any verbal/written feedback that was given to clarify this.

Answer: The Siriraj Preanesthetic Care Unit (SiPAC) was set up in our institution to provide preoperative evaluation and preparation according to the medical condition of the patient as well as laboratory investigation. This setting is related to the hospital policy to reduce perioperative risk and improve cost effectiveness of patient treatment. However, the system for evaluation of hospital economy in term of patient testing for preoperative preparation is yet to be established. The question arises whether there is a policy for reduction of unnecessary laboratory testing including chest radiography. We, therefore, constructed the research framework the identify factors that influence chest x-ray request.

Results

- Table 1 and 2 notes: “*Significant at p < 0.05”. Please indicate what statistical test was used for this. I suggest you make captions stand-alone, such that they can be read without having to reference other sections (eg. methods).

Answer: The tables were re-designed and rearranged in the new format. The footnote stated the statistics used and the abbreviations in the table. At present, table is 1 the demographic characteristics of the patients. The results are reported as descriptive statistics with mean±SD or number (%). Table 2 displays the baseline knowledge about preoperative chest x-ray among medical personnel. The significance is determined by p < 0.05 using Chi-square. We do hope these will make the results more understandable.

- Table 1: Did you ask or allow respondents to identify as a non-binary gender? It is important to include these categories, even if there were no respondents that identified. Also, in the sub-group analysis tables you stratify according to sex but asked about gender in the questionnaire. Please update to reflect the information you received.

Answer: All participant (142 persons) identified themselves as female (75 persons) or male (67 persons). We have already changed the word “gender” to sex in the questionnaire.

- In this section of the results, you have re-stated what is inherent in the table. There is no need to ‘double report’ this. Please summarise this or remove it from the text.

o “There were significant mean age differences between the 3 groups of medical professions (anesthesiologists, surgeons, and gastroenterologists), with the highest mean age being for gastroenterologists (31.27 ± 1.74 years), followed by surgeons (29.17 ± 2.35) and anesthesiologists (28.24 ± 1.44). The overall proportions of male and female participants were 47.2% and 52.8%, respectively.”

Answer: We have removed this from the text as suggestion.

- The whole paragraph on demographic characteristics is effectively conveyed in the table. My suggestion would be to significantly cut the paragraph on demographic characteristics, for readability.

Answer: According to the fact that table 1 had already demonstrated complete demographic data of the participants, we decided to remove unnecessary paragraph on demographic characteristics as suggestion.

- The paragraph starting with “The knowledge and attitudes about preoperative chest radiography…” is largely method-related and I believe it would be better suited in the methods section. How does this paragraph differ to the questionnaire information you have detailed in the methods section?

Answer: It is not different from what is written in methods section. We have moved the paragraph to the methods section as suggestion.

- Were the sub-group analyses identified a priori or were they post hoc? Either way, I believe some justification is necessary as to why the sub-group analyses have been performed and some sort of adjustment for multiple comparisons must be made. As an aside, the other pertinent question is ‘do these analyses add to the paper and its overall message?’ My impression from having read the paper is no – I would suggest you make it more explicit as to why these analyses are important.

Answer: The sub-group was post-hoc analysis. However, we agreed that they might not add to the paper. We decided to remove the sub-group analysis and perform confirmatory factor analysis (CFA) and constructed structural equation model (SEM) as mentioned later on. The demographic data of the participants were reported descriptively.

- Could you please detail what is meant by “good” and “poor” attitudes? What method was used to assess this?

Answer: Attitude questions were in Likert pattern thus it would be quite difficult to determine which ones were poor or good. Therefore, we re-analyzed the data by using confirmatory factor analysis (CFA) and constructed structural equation model (SEM) to find out the factors that affected the chest x-ray request of medical personnel. By SEM analysis we thought it would be more appropriate for the analysis of factors that affected chest x-ray request of medical personnel. The information from SEM would be better to obtain the important data on what was the key element for appropriate CXR request.

- The idea of correlating attitude and knowledge scores seems problematic as I am unclear the exact method you have used to do this. Attitudes, while they have been captured here semi-quantitatively, are not quantitative and often knowledge is insufficient to change attitudes. I believe that correlating the two is fundamentally flawed.

Answer: We agreed with your suggestion that the idea of correlating knowledge and attitudes may not be appropriate. Therefore, we decided separate the questionnaire into 2 sets including 1. Knowledge part which was later analyzed and reported as demographic data and 2. Attitude part containing questions with scaled items to perform confirmatory factor analysis (CFA) of the measured variables derived from the questionnaire. Subsequently, we generated structural equation model (SEM) to find relationships between the latent variables with the outcome which was intention of CXR request for preprocedural gastrointestinal endoscopic procedures.

- No data is presented in the results section regarding the retrospective patient chart review. Please include this in the results section.

Answer: We have removed this from the manuscript

Discussion

- The article would be improved by adding an initial short paragraph to the discussion that summarises the main findings of the study.

Answer: We have added summary of main findings as suggestion.

- Paragraph 1 of the discussion brings in new results and is better placed in the results section.

Answer: We have removed the paragraph 1 from the study because it is actually a part of our published data which we have lately added to the introduction (ref. 19).

- Paragraphs 2 and 3 of the discussion summarise and attempt to rationalise the results. These two paragraphs would be strengthened by i) making the paragraphs more succinct by focusing on the most salient points, ii) by comparing and contrasting your results with the literature (how does your research fit into the broader literature on CXR prior to endoscopy?) and iii) justifying your assertions based on other published literature.

Answer: We did as suggestion.

- The authors do not discuss the limitations of their study and the manuscript be strengthened by such a discussion.

Answer: We added the limitation accordingly.

- The authors do not provide a concluding paragraph emphasising the importance and utility of this research. The manuscript would be enhanced by this.

How will this research be used at Mahidol University hospitals?

Answer: We have added the paragraph “The concept of identifying factors responsible for preoperative endoscopic CXR request was to find solutions for our institution to limit the unnecessary test ordering for better cost effectiveness and reduced risk of radiation exposure.” in the discussion section.

Will there be an intervention carried out to improve the knowledge and attitudes of physicians?

Answer: Future plans will be providing proper knowledge and attitudes to medical personnel and closely follow up on each individual on real practice on preoperative chest x-ray ordering for the patients. We expect that with good knowledge and attitude, unnecessary chest x-ray ordering will tend to decline. However, the true factors influencing chest x-ray request need to be identified.

- “This might be because of a better decision-making ability of female respondents” Did you assess decision-making ability? While this may be true, I do not see how your results justify this assertion. Please provide additional reasoning for this.

Answer: We did not assess decision-making ability in this study. It was just our speculation. We agreed with your suggestion that the results did not indicate the decision-making ability therefore we removed the sentence form the discussion. This could be the one of the limitations of the study in the sense that we could not identify real practice of each individual therefore his/her decision making would not be revealed.

- Why is “inactive pulmonary infiltration” non-significant? This could represent latent tuberculosis which may significant in certain patient populations. Could you please clarify?

Answer: We have deleted this from the manuscript.

Limitations

o Understand the knowledge and attitudes of practitioners does not always reliably predict their actions and course of practice. Have they addressed this?

Answer: We were well aware of this point. Our future plan is directly observing the actual practice on preoperative chest x-ray of the medical personnel. We expect to see minimal unnecessary chest x-ray ordering in the ones with good knowledge and attitudes.

o Low completion rate – why? Will this bias results?

Answer: The participants volunteered to answers the questionnaires. We could not force the ones who refused to participate according to the ethics we declared to the IRB committee. Having said that, based on sample calculation in Structural Equation Mode (SEM) analysis, the minimal sample size required for reliable outcome is 116 (ref.23). We were able to get 142 respondents which was enough for analysis.

o “Our study focused on a diverse population of medical professions—anesthesiologists, surgeons, and gastroenterologists—who had experience with patient evaluations and the requesting of tests preoperatively.” I disagree that this is a diverse population of medical professions – they are all specialties that have the potential to routinely order, and need knowledge of, endoscopies. Is there any potential bias that could result due to this?

Answer: We agreed and so we change the word “diverse” to different.

Minor textual amendments:

- Page 5 of manuscript “2,030 outpatient charts produced January–November 2017” should read “2,030 outpatient charts produced between January and November 2017”

Answer: The sentence was deleted because we did not include this result in the manuscript.

- Page 6 of manuscript “others dealt with general, preoperative, CXR indications, and knowledge of tuberculosis.” This sentence is unclear. Could you please clarify it.

Answer: The sentenced was deleted.

- Page 6 of manuscript “The attitude test consisted of 11 questions” should read “The attitude section consisted of 11 questions.”

Answer: We have already changed the word “test” to section.

- Page 11 of manuscript “Our retrospective data indicated that that there” should read “Our retrospective data indicated that there”, delete the extra ‘that’

Answer: The sentenced was re-written.

Reviewer #2:

Dear author manuscript is well written but has some minor corrections to be done. the discussion part is however well written but results from other studies were not compared to the present study either similar / dissimilar.

Answer: Thank you very much. We have re-written the discussion paragraph comparing our study with others.

Reviewer #3:

Thank you for the opportunity to review this manuscript. The advantage of this study was conducted scientifically and systematically. The manuscript is well written and described clearly. Several suggestions to improve this manuscript are listed below.

The methods section of the abstract needs to contain more information, especially regarding the study location and duration. It will reflect the inference population. Research tool (the questionnaire) needs to be elaborated further as to whether a newly developed questionnaire or a validated questionnaire was used.

Answer: We added more details in the Materials and Methods section under Study design and participants and Data collection parts.

In this study, two study designs were employed, a prospective clinical study and an observational retrospective study. Is there any reason why it cannot be called a cross-sectional study since all the information is obtained at a single point of time with no element of follow-up?

Answer: The main study is a cross-sectional clinical study. We have removed the retrospective data out of the study because it was a part of our recently published data (we added this as ref.19 and 20 in the introduction section ) which did not add to our result interpretation.

What is the basis of taking 264 participants? Was it based on any sample size calculation? In this study, elaborate on how a sample size of 264 was determined. Any exclusion criteria for study participants? What does it mean by "incomplete medical records"?

Answer: The number 264 was from total medical personnel from residency and fellowship training programs in Siriraj Hospital, Mahidol University. Based on our analysis of the Structural Equation Model (SEM), the minimal sample size which resulted in reliable outcome was 116 (ref. 24) as explained in the Materials and methods section, study design and participants part. Exclusion criteria were medical personnel who refuse to participate and the ones who did not comprehend Thai given that original questionnaire was written in Thai. However, the questionnaires were sent to all potential participants because we thought that medical personnel might be busy and only some of them could dedicate time to participate. The phrase “incomplete medical records” was not mentioned when we re-wrote the manuscript.

Who decides whether an x-ray is uninterpretable?

Answer: We have deleted this from the manuscript.

Include a reference number for ethical clearance.

Answer: We have added the refence number as suggestion.

Define "resident" and "fellow".

Answer: We have made an explanation in paragraph 1 of the result section.

This study uses a newly developed questionnaire to assess knowledge and attitude towards preoperative chest radiography for elective endoscopic procedures. It is important that the questionnaire is validated before it is being used for the actual study. Furthermore, the readers must be convinced that the questionnaire undergoes a proper questionnaire validation process as it will determine the validity of the results. "The Index of consistency score of the questionnaire was 0.91". Does this indicate internal consistency reliability? The assessment of content validity needs further elaboration (e.g. assessment of content validity index by a group of independent experts). Any reason why factor analysis (EFA and CFA) was not performed?

Answer: We have re-analyzed the data. The details are explained in the Materials and methods section, paragraph 2. CFA and SEM were also performed as mentioned in paragraph 2 and 3 of result section. We also invited co-author, Dr. Somsit Daungekanong, who is an expert on statistical analysis to assist with the re-analysis and data interpretation.

Elaborate sampling method. Any method employed to ensure the representativeness of the study participants in terms of the department and work experience?

Answer: We agreed to what you suggested. Anyway, we only have 264 trainees in our institution. Therefore, we decided to use the overall results to guide for the improvement of practice on CXR request and official institution guideline establishment in the future.

What does it mean by "Statistical significances were determined by chi-squared tests"? Statistical significance refer to the p-value and can be obtained from all statistical test (not just the chi-squared test).

Answer: We have re-analyzed the data and used Chi-square test only in Table 2.

Rephrase "The data with a normal distribution were analyzed and reported as percentage, mean, and standard deviation". All data were analyzed. Data with normal distribution needs to be presented as mean and its standard deviation, whereas data with skewed distribution needs to be resented as median and interquartile range. Frequency and its percentage are used to describe categorical variables.

Answer: We re-analyzed the data and employed descriptive statistics for the demographic characteristics of the medical personnel as shown in Table 1. The results were described by mean±SD and percentage.

The sample size calculation needs to be placed under "study design and participants" and not under the method of statistical analysis.

Answer: We have moved the sample calculation to the study design and participants as suggestion.

The knowledge test consisted of 18 questions; 7 were about surgery in ambulatory settings, whereas the others dealt with general, preoperative, CXR indications, and knowledge of tuberculosis. The attitude test consisted of 11 questions. Elaborate method to obtain the total score for each domain.

Answer: We have re-analyzed the data for better result interpretation and rationale of the study. We adopted IOC and Cronbach’s alpha to determine the validity and reliability of the questionnaire. With the use of Daniel Soper’s equation model analysis, we were able to obtain the good fit of the model to define the relationship between latent variables and the chest x-ray request (outcome).

The response rate in this study is low (53.79%). In terms of statistical power, was it still adequate? Discuss the response rate in relation to external validity of this study.

Answer: Actually, based on sample size calculation for SEM (ref. 24), the number of participants we obtained was enough for analysis. Even though the response rate was 53.79%, we did not adopt the result from the analysis to determine the knowledge and attitudes of other groups of medical personnel. Instead, we aimed to use the analysis for an improvement of preoperative evaluation study for proper chest x-ray request in the future.

The description regarding the questionnaire and the methods of scoring for each item (paragraph two of the results section) needs to be placed under the methods section.

Answer: According to the re-analysis of the data for better result interpretation with more appropriate statistics use, we had removed this part.

Lengthy text presentation of the statistical results. Most of the information can be obtained from the tables, and presenting the results in both text and tables are redundant.

Answer: We have cut the redundant phrases out from the results section.

In table 1, 2, indicate the statistical analysis conducted to obtain each p-value. Here, the comparison of numerical variables between three independent groups was conducted using one-way ANOVA, and for categorical variables using either the Chi-squared test or Fisher exact test. Methods of these statistical analyses need to be included in the methods section.

Answer: We did as suggestion.

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: Andrés Noé

Reviewer #2: Yes: Aisha Wali

Reviewer #3: No

________________________________________

**We would like the express our sincere gratitude toward your invaluable reviews. We have made some changes as follows;

1. The article name is changed to “Factors influencing preoperative chest radiography request for elective endoscopic procedures among medical personnel” considering that it reflected our primary objective of the study.

2. We have added to the methods that Hypothetical framework of possible factors that influenced medical personnel of chest x-ray request was constructed. The model validity was confirmed by Confirmatory Factor Analysis (CFA) and subsequently subjected to Structural Equation Model (SEM) to determine any associations. By doing so, we are affirmed that the statistical analysis of variables was appropriate as you have suggested.

3. General knowledge on preoperative chest x-ray was compared among medical personnel by using Chi-square test.

4. The multiple comparisons of subgroup analyses of knowledge and attitudes among medical personnel were removed according to possible false positive results as the reviewer suggested.

5. Correlation between knowledge and attitude by Spearmen’s rank was removed. Instead we determined the association of hypothesized factor with the intention to request for chest x-rays by SEM as mentioned earlier.

Submitted filename: Response to reviewer .docx

Click here for additional data file.

28 Oct 2020

Factors influencing preoperative chest radiography request for elective endoscopic procedures among medical personnel

PONE-D-20-04685R1

Dear Dr. Siriussawakul,

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,

Sanjiv Mahadeva, MRCP, MD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

The authors have made substantive revision based on the original 3 reviewers comments - these revisions are satisfactory

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 #2: 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 #2: Yes

**********

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

Reviewer #2: Yes

**********

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 #2: 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 #2: 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 #2: (No Response)

**********

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 #2: Yes: Aisha Wali

4 Nov 2020

PONE-D-20-04685R1

Factors influencing preoperative chest radiography request for elective endoscopic procedures among medical personnel

Dear Dr. Siriussawakul:

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. Sanjiv Mahadeva

Academic Editor

PLOS ONE