In evidence-based practice, research findings greatly influence professional activity, recommendations, clinical therapies, and drug prescriptions. Scientific evidence that supports decision making in the health sector should be based on the results of high-quality clinical trials (Sackett et al., 1996; Chiappelli, 2019). Therefore, bias, data manipulation, or limitations greatly influence research findings, thereby questioning their validity (Hinton et al., 2015; Papageorgiou et al., 2015) (see Table 4, Table 5).

The study design most suited for evidence-based clinical practice in terms of efficacy and clinical interventions is the randomized clinical trial (RCT) (Needleman et al., 2005). However, when performed inappropriately, it generates misinterpretations and erroneous conclusions in a systematic review. In an RCT, the primary or main outcome of the research must be defined. This will greatly determine the representative population sample required to answer the research question, beyond the other outcomes, known as secondary outcomes (Pannuti et al., 2020).

To perform a study appropriately, systematic errors or biases should be minimized. Selective outcome reporting (SOR) is a type of bias that occurs when the primary outcome of a clinical trial is modified or unpublished, or when a new outcome is reported (Mathieu et al., 2009; Chan et al., 2004). SOR occurs frequently (Hannink et al., 2013; Wayant et al., 2017), and is associated with significant results (Zhang et al., 2017; Aggarwal and Oremus, 2019). Perhaps this is because studies presenting results that are inconsistent with their hypotheses are difficult to publish (Pedrazzi et al., 2020). SORs have been observed in systematic reviews relating to oral health and to orthodontic/dental implant clinical trials (Koufatzidou et al., 2019; Sendyk et al., 2019). However, to the best of our knowledge, no study has evaluated SOR in oral/maxillofacial surgery clinical trials.

The extraction of impacted third molars is the most common oral and maxillofacial surgical procedure. Furthermore, clinical research models based on third molar extractions are used to investigate interventions and/or new drugs or therapies for pain and inflammation in medicine and dentistry. These models offer reproducibility, a short period of evaluation of acute inflammatory pain, and an adequate sample of young and healthy patients. Postoperative pain and inflammation can be measured in different ways: visual analog scales, analgesic consumption, degree of facial edema, oral aperture limitation, quality of tissue repair, and levels of perioperative anxiety (Silva et al., 2011).

Thus, our study aimed to evaluate the prevalence of SOR in published clinical trials involving the extraction of lower third molars to investigate a model, and to discuss the impairment effects of SOR on evidence-based clinical practice.