Authors’ reply to Shrivastava

The article titled “Awareness about COVID19 among patients with cancer attending the department of radiation oncology: A prospective observational study – ACROS,” published in the journal Cancer Research, Statistics and Treatment, provides deep insight and a comprehensive analysis of the knowledge, attitude, and practice regarding precautionary measures among cancer patients during the late stages of the COVID-19 pandemic.[1] A letter to the editor from Shrivastava[2] has identified discrepancies in the study, which we aim to clarify. The final questionnaire used in the study was titled the “ACROS questionnaire” (with “ACROS” as an acronym from the study’s title, underlined and marked in bold in the article’s introduction). We would like to further clarify that participants from the pilot survey were excluded from the final analysis. The questionnaire was interview-based rather than self-administered, as out of 274, 251 (91.6%) of patients were from lower socioeconomic backgrounds, and 119 (43.4%) were illiterate. Therefore, regional language adaptations were not required. Validation was completed and mentioned in the study.[1] We categorized the knowledge score as adequate (score ≥ 7) or inadequate (score < 7). This analysis is based on the approach reported by Singh et al.,[3] who proposed that categorization can be determined using the median value. Two open-ended questions, that we did not include in the final analysis, were as follows: How frequently do you go outside? What is the ideal physical distance that should be maintained? We understand the importance of above-said questions and therefore, included them in the final questionnaire. However, most patients gave vague answers to the first question. Upon further inquiry, it was revealed that many left their homes only to visit the hospital, which introduced recall bias. Additionally, none of the participants answered the second question correctly. As a result, these responses were excluded from the analysis. The patient footfall during that period was both unpredictable and below average. As a result, a convenient sampling strategy was adopted.[1] We acknowledge the concern raised by Srivastava[2] regarding statistical tests used in our study. We used non-parametric tests, such as Kruskal-Wallis test and Spearman’s correlation, as normality tests (Shapiro-Wilk and Kolmogorov-Smirnov) indicated a non-normal distribution. A patient pool of 280 did not necessarily mean normality. Regression analysis helped to determine associations between independent and dependent variables.[4] It also helped us in determining which of the independent variables better predicted the outcome of the dependent variable. Hence, binary regression (for binary variables) was used for the above. Thus we considered the former as an appropriate method of data analysis. The Chi-square test is a test of association between two nominal/ordinal variables and is usually calculated and displayed in a tabular form along with the variables (in a cross-tabular form).[5] Hence, the test was not mentioned in the statistical analysis. However, we thank the author for pointing this out. Our study concluded that cancer patients in rural India, particularly those from lower socioeconomic backgrounds, showed declining concern about COVID-19 and ignored preventive measures as cases decreased.[1] Despite some limitations, already addressed by us in the main manuscript, our study addressed a unique and important issue that could shape social and health policies for future pandemic mitigation.[2] Therefore, although we appreciate the feedback, we feel that the letter to the editor would have been much more objective, if the many positive aspects of the study (critical issues addressed and the valuable insights provided) were highlighted, along with the constructive criticism. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.