Critical comment on “Demographic profile and trend in burn injury: A clinical study in South Karnataka”

Dear Editor, I appreciate the article “Demographic profile and trend in burn injury: A clinical study in South Karnataka” published in Muller Journal of Medical Sciences and Research 16(1):p 41-45, January–June 2025. However, I have a few concerns about the Results and Discussion mentioned in the study. Authors have mentioned in the Materials and Methods section of the “ABSTRACT” that “The Z-test and Microsoft Excel were used to evaluate and analyze patient data related to burns, including age, sex, etiology, manner of burn injury, percentage of body surface area involved, length of hospital stay, number of patients undergoing surgical care, discharge, and mortality.” Similarly, in the “Statistical analysis” portion of the main study, the authors have mentioned about these statistical tests. In actuality, in the Results section, no such statistical test has been used by the authors except the percentages. Inappropriate statistical techniques can lead to improper research study conclusions, give rise to errors, and hence undermine the significance of the research paper.[1] Statistical tests build scientific rigor by providing a sound framework for testing hypotheses, establishing the significance of the study results, and drawing valid and reliable conclusions from data. Results would have been more authentic if the authors had used a logistic regression test to find out the association of age and complications in burn patients [Table 4] and binary logistic regression to find out the association between age and prognosis [Table 5]. Prognosis has 2 components, i.e., improved and poor, and hence, a binary logistic regression will be used. Similarly, a Chi-square test of significance could have been applied to find out the association of the degree of burn (ordinal variable) in patients with type of treatment, such as conservative management, surgical interventions, ventilator, and inotropic support (nominal variable), in Table 7. Similarly, a Chi-square test of significance is required for Table 10, signifying the relationship of gender versus prognosis. There is no relevance to mention the word “statistically significant” inside Tables 4 and 10 when, in actuality, no statistical test of significance is used. It is not at all meaningful and logical to conclude the results merely based on percentages. The authors mentioned regarding using “z” test along with Microsoft Excel to calculate results. Microsoft Excel is not a statistical test rather a spreadsheet program that allows researchers to organize, format, and calculate data in a spreadsheet. Authors mentioned in the Results section, “Statistically significant, correlating with previous studies.”[2] This needs to be written in the Discussion part and not in the Results section. It is prudent to describe the results of so-called previous studies. In the Discussion section, authors have mentioned in the first paragraph, “The age group involved mostly 0–10 years accounted for 27.4%, correlating with previous studies. Cause of burns mostly thermal burns correlating with previous studies.” The references are missing for both these studies, and the authors need to mention the results of the so-called “previous studies.” All the previous studies with similar results or contrary results need to be cited carefully, mentioning their references. It is prudent to use the terms “similar” and “contrary” in Discussion, and there is no need to overstate or rewrite the results in the Discussion section. The Discussion section should focus on the larger implications of the study in terms of known research.[3] Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.