Detailed Limitations of Study Help Readers to Interpret Result Effectively.

Sir,

We read a brief report contributed by Alshehri et al. in March-April 2017 issue of your journal.[1] Authors conducted a cross-sectional survey to determine the use of prescription and over-the-counter (OTC) drugs for acne among students of Saudi Arabia. The article was relatively well presented and the result of the study was clear. However, we wanted to share some of our views about the materials and methods of the study.

The sample size was calculated as 379. However, authors distributed questionnaire to all 30789 students through electronic media and printed questionnaire. Hence, we can assume that the sample size was 30789. In addition, authors stated that survey response rate could not be calculated as the questionnaire might reach “a much larger sample of students.” In the preceding text, they already mentioned that they distributed the questionnaire to “all the university students.” Hence, from our end, we calculated the survey response rate and found it to be 1.36% [(420/30789) × 100]. Now, the result of the study lacks justification. This level of survey response rate may not represent all university students and their use of medication for acne.[2] Adding this limitation of study could enrich the presentation.

Among the 420 respondents, 211 were diagnosed with acne by dermatologist and 209 were not diagnosed by dermatologist. From this data, it was evident that, among 420 respondents, all had acne. Authors did not declare that only filled questionnaire of respondents with acne were analyzed. Hence, from this study, we interpreted that university students in the Kingdom of Saudi Arabia had 100% prevalence of acne. This was quite a surprising result. Authors forgot to mention this finding of their study.

According to authors, the survey instrument was a validated questionnaire. Authors did not mention any details about it. The questionnaire might be a validated questionnaire from similar previous studies or it might be drafted by the authors. To ascertain validity, authors used Cronbach's alpha and obtained alpha as 0.405. Cronbach's alpha actually measures the internal consistency of scale. It is an index of reliability not validity.[3] In addition, alpha <0.7 is not an acceptable level of internal consistency.[4] Hence, the survey instrument was a questionable one. Again, authors did not carry out a pretest of the questionnaire. Author's declaration about these limitations might help us to interpret results with caution.

Writing a brief report is sometime difficult than original article because authors are bound to write it within certain word limits, usually 1500 words. However, this challenge is a platform to prove expertise in presenting the study concisely. Adding limitations and confounding factors of the study is highly desirable in any study.[5] It helps in proper interpretation of results of the study by the readers.

“The reply from the authors are available at Alshehri MD, Almutairi AT, Alomran AM, Alrashed BA, Kaliyadan F. Reply to: Detailed limitations of study help readers to interpret results effectively. Indian Dermatol Online J 2017;8:296”

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.