Methods
A cross-sectional survey was conducted amongst Otorhinolaryngologists (ORLs) practicing in India by widely advertising on various social media platforms, an online hosted digital questionnaire (Appendix 1) . Participation in the anonymous survey was voluntary, an informed consent was taken from all the participants at the beginning of the survey and participants could terminate the survey at any time they desired. The survey was conducted between 18th April 2020 and 10th May 2020. India was under lockdown since 25th March 2020. This was the time when government was contemplating on how to implement a phased un-lock, which eventually started on 8th June 2020. Hence the timing of survey coincides with the time when the ORLs must be grappling with the questions on how to restart their practice and how to reshape their standard operating procedures.
The questionnaire captured demographic parameters like age, gender and location of the practicing ORLs. It took into account the difference in risk perceptions depending upon the workplace and technical title of the respondent. Few of the questions were aimed to assess the personnel protective equipment (PPE) used by the doctors and its impact on economics of running a clinical practice. The data including the technical title/professional designation was self-reported by the participants. Validated scales to know about stress were not used as that would have made the questionnaire very long. A scale of 1 to 10 was used to measure the stress which was divided in to low (1-3), intermediate (4-7) and high (8-10) categories. The questionnaire was pretested, face and content validity was checked by the experts. The survey began with a question asking, whether you are an ORL and those who answered in negative were automatically eliminated from the survey. Data collected was exported to Microsoft Excel 2013 (Microsoft Corporation) software (Appendix 2, in PDF) , cleaned and analyzed using Statistical Package for Social Sciences (SPSS) version 24. Proportions of different categories were estimated. Differences in proportions between groups were considered significant at 95% confidence level.