Background
Obesity is one of the most common health problems leading to a variety of complications including type 2 diabetes and cardiovascular diseases. With increasing publication in obestiy research, the appropriate interpretation and reporting of study results is all the more important. SPIN can be defined as specific reporting strategies, intentionally or non-intentionally, that could lead the readers to misinterpret the results. This study aimed at identifying the prevalence and types of SPIN in randomized controlled trials (RCTs) of obesity with statistically non-significant results for primary outcomes to provide adequate reporting directions.

Methods
RCTs on obesity with statistically non-significant primary outcomes published from July 2015 to June 2016 were retrieved from PubMed using the key words related to "obesity", "complementary and alternative medicine", “western medicine”, “dietary therapy and exercise”, and “randomized controlled trials”. All included RCTs were classified into 3 intervention categories, i.e. western medicine, complementary and alternative medicine, and dietary supplement and exercise. The identification and classification of SPIN in the included articles was performed by independent two researchers. In addition, the association between the prevalence of SPIN and various factors including country, sample size, protocol registration, funding and selection bias was analyzed.

Results
Among 46 RCTs with statistically non-significant primary outcomes, 32 studies (69.6%) were assessed as having at least one SPIN in title, abstract or main text. Of these, 9 (28.1%) articles were on CAM, 7 (21.9%) on western medicine and 16 (50.0%) on dietary supplement and exercise. There was no statistically significant difference in the association between the prevalence of SPIN and the types of interventions (P=0.069). The most common type of SPIN was ‘focusing on statistically significant within-group comparison’ in Results section of abstract (13/24, 54.2%) and main text (15/18, 83.3%), and ‘focusing only on treatment effectiveness with no consideration of statistical significance’ in Conclusions section of abstract (20/30, 66.7%) and main text (16/29, 55.2%). Studies where random sequence generation was appropriately done was less likely to have SPIN (P=0.046).

Conclusions
As a majority of obesity RCTs have SPIN, researchers should pay more attention to adequately interpreting and reporting statistically non-significant primary outcomes.