Article type
Abstract
"Background:
With the rapid increase in the number of conducted and published randomized controlled trials (RCTs), systematic reviews (SR) encounter a substantial influx of studies. Dealing with meta-analysis and other synthesis becomes challenging due to incomplete reporting of methodology. These trials are assessed as unclear risk of bias and been down certainty.
Objectives:
To verify the integrity of unclear reported trials and the eligibility for inclusion in SR.
Methods:
Based on one SR of RCTs from 2013-2019, we conducted the telephone interviews with all authors of 283 trials that only reported “random”, with the questions asked in the following order, including the specific randomization methods, concealment methods and other details. We searched the contact number based on their affiliates and email addresses. If any author provided incorrect information regarding the randomization methods, we would terminate the interview.
Results:
The authors of 147 (51.94%) trials were failed to be located based on their affiliations, and inquiries made at hospital reception desks or departments confirmed that these individuals did not exist. The authors of 122 trials (43.11%) were successfully contacted; however, they declined to answer any questions regarding their articles. In response to our inquiries, the authors of 11 trials (3.89%) stated that their studies employed semi-randomization or without randomization. Among these 11 respondents, 9 trials were published by a single author. One trial (0.35%), published by a single author, indicated an inability to recall the randomization method used and refused to answer other questions. For another trial (0.35%), also published by a single author, we reached out to the department's reception desk only to learn that the author had gone abroad for advanced studies and could not be contacted. Only one trial (0.35%), authored by three individuals, acknowledged using a random table but declined answering further queries.
Conclusions:
Among 283 trials, we only could include one trial that utilized a specific randomization method.So we suggest that the journal editors should manage the reporting quality of RCTs following the CONSORT. To prevent research waste and unnecessary publications, it is necessary to improve selection criteria."
With the rapid increase in the number of conducted and published randomized controlled trials (RCTs), systematic reviews (SR) encounter a substantial influx of studies. Dealing with meta-analysis and other synthesis becomes challenging due to incomplete reporting of methodology. These trials are assessed as unclear risk of bias and been down certainty.
Objectives:
To verify the integrity of unclear reported trials and the eligibility for inclusion in SR.
Methods:
Based on one SR of RCTs from 2013-2019, we conducted the telephone interviews with all authors of 283 trials that only reported “random”, with the questions asked in the following order, including the specific randomization methods, concealment methods and other details. We searched the contact number based on their affiliates and email addresses. If any author provided incorrect information regarding the randomization methods, we would terminate the interview.
Results:
The authors of 147 (51.94%) trials were failed to be located based on their affiliations, and inquiries made at hospital reception desks or departments confirmed that these individuals did not exist. The authors of 122 trials (43.11%) were successfully contacted; however, they declined to answer any questions regarding their articles. In response to our inquiries, the authors of 11 trials (3.89%) stated that their studies employed semi-randomization or without randomization. Among these 11 respondents, 9 trials were published by a single author. One trial (0.35%), published by a single author, indicated an inability to recall the randomization method used and refused to answer other questions. For another trial (0.35%), also published by a single author, we reached out to the department's reception desk only to learn that the author had gone abroad for advanced studies and could not be contacted. Only one trial (0.35%), authored by three individuals, acknowledged using a random table but declined answering further queries.
Conclusions:
Among 283 trials, we only could include one trial that utilized a specific randomization method.So we suggest that the journal editors should manage the reporting quality of RCTs following the CONSORT. To prevent research waste and unnecessary publications, it is necessary to improve selection criteria."