Control Group and Refusal Rate Critiques about Hepatitis C Virus Infection in HIV Positive Cases.

Sir,

The paper published in this journal about prevalence and risk factors of hepatitis C virus (HCV) infection in human immunodeficiency virus HIV positive cases of a center in Iran(1) has some minor critiques.

Without a suitable control group, we cannot conclude that even high frequency behaviors like imprisonment (88.9%), injecting drug use (79.2%), or any other variable can be considered as risk factor. We should define risk factors in comparison with a basic population. Maybe in some populations, such behaviors (variables) are more prevalent and these variables can be considered as protective factor. Percentages higher than 50% for example, do not mean that this variable is a risk factor. Authors did not even compare the total frequency of these demographics and risk factors with other subgroups in their study and mentioned percentages in both HCV positive and negative cases. In addition, comparison of HCV positive and negative cases is not completely logical specifically when the number of two groups is imbalanced with 200 HCV-negative cases in comparison with only 26 HCV-positive ones according to ELISA. In addition, the nature of sampling shows that the controls are not representative of the HCV-negative cases because of nonprobability sampling and high refusal rate. However, the study base of HCV-positive and HCV-negative cases is similar and this similarity rationalizes selecting HCV-negative samples as controls;(2) but, with considering above mentioned issues.

I know that research in HIV field is most of the time along with considerable missing data due to nonresponse or refusal to participate and even dropout. However, high refusal rate may distort the results. Authors are encouraged to discuss this issue or at least do sensitivity analysis.