Heterogeneity in studies of p16 in cervical lesions in different Malaysian institutions: Time to consider...

Clinical decision making becomes difficult when clinical and methodological heterogeneity does not permit synthesis of results of multiple small studies.

For studies done in Malaysia, to identity the sample sizes and heterogeneity present in the various studies which used p16 in evaluating lesions of the cervix. To evaluate if it would be possible for a single study to answer the various questions posed by the original authors. To highlight areas where the design features of future studies can be optimised.

Various databases were searched using synonyms for p16 AND cervix AND Malaysia. These were assessed for broad conformity to a Diagnostic Test Accuracy format. Methodological and clinical heterogeneity indicators were extracted into standardised fields.

There were 5 studies eligible for inclusion. Each sought to study different aspects of the disease such as diagnostic grade stratification and pathogenesis. The study type broadly conformed to a Diagnostic Test Accuracy format. The study design used was either consecutive or non-consecutive. Sample size ranged from 75 to 201. Clinical heterogeneity was present in the choice of controls with some using normal and some using inflamed tissue. Methodological heterogeneity in applying the reference test, index test and different antibody clones were present.

There was both clinical and methodological heterogeneity making synthesis of studies difficult. It is possible to design a study which would be able to answer all the questions posed by the original authors with internal validity while at the same time increasing sample size.