Article type
Year
Abstract
Background:
It is difficult to effectively evaluate the effects of interventions based on available studies. The systematic review/meta-analysis is considered to be the best available evidence, but this is not the same as 'sufficient evidence' or 'strong evidence'.
Objectives:
We took the systematic review of iguratimod in the treatment of rheumatoid arthritis as an example to find out whether sufficient sample sizes can support high-quality evidence.
Methods:
We retrieved studies from databases including PubMed, the Cochrane Library, CBM and CNKI from inception to January 2018. Two review authors independently extract data. We also used TSA v0.9 software.
Results:
A total of 43 randomised controlled trials (RCTs) were included with 4054 participants, 70 of which were withdrawn. In the iguratimod and methotrexate groups, trial sequential analysis of overall response rate showed that the cumulative Z-curve did not cross the trial sequential analysis threshold value until the included sample size was more than the required information size (RIS). This suggested that the result of the meta-analysis was a false-positive and more trials were need to confirm efficacy (Figure 1). The combination of iguratimod and methotrexate groups and methotrexate groups included four RCTs with 440 participants. Trial sequential analysis of the overall response rate showed that the cumulative Z-curve did not cross both the trial sequential analysis threshold value and RIS, which suggested that there was no significant difference between the intervention group and control group and more trials were still needed (Figure 2).
Conclusions:
If a statistically significant result is not obtained, whether the intervention is truly invalid or there is a false-negative result caused by an insufficient sample size cannot be confirmed. Therefore, the sample size is one of the factors that leads to lower power and lack of credibility. It is difficult to use such findings for clinical decision-making, as they cannot support high-quality evidence and are also a waste of scientific research and medical resources.
Patient or healthcare consumer involvement:
In clinical trials, verifying the efficacy of a drug requires more patient involvement and better estimation of sample sizes to save patient resources and improve the accuracy of trial results.
It is difficult to effectively evaluate the effects of interventions based on available studies. The systematic review/meta-analysis is considered to be the best available evidence, but this is not the same as 'sufficient evidence' or 'strong evidence'.
Objectives:
We took the systematic review of iguratimod in the treatment of rheumatoid arthritis as an example to find out whether sufficient sample sizes can support high-quality evidence.
Methods:
We retrieved studies from databases including PubMed, the Cochrane Library, CBM and CNKI from inception to January 2018. Two review authors independently extract data. We also used TSA v0.9 software.
Results:
A total of 43 randomised controlled trials (RCTs) were included with 4054 participants, 70 of which were withdrawn. In the iguratimod and methotrexate groups, trial sequential analysis of overall response rate showed that the cumulative Z-curve did not cross the trial sequential analysis threshold value until the included sample size was more than the required information size (RIS). This suggested that the result of the meta-analysis was a false-positive and more trials were need to confirm efficacy (Figure 1). The combination of iguratimod and methotrexate groups and methotrexate groups included four RCTs with 440 participants. Trial sequential analysis of the overall response rate showed that the cumulative Z-curve did not cross both the trial sequential analysis threshold value and RIS, which suggested that there was no significant difference between the intervention group and control group and more trials were still needed (Figure 2).
Conclusions:
If a statistically significant result is not obtained, whether the intervention is truly invalid or there is a false-negative result caused by an insufficient sample size cannot be confirmed. Therefore, the sample size is one of the factors that leads to lower power and lack of credibility. It is difficult to use such findings for clinical decision-making, as they cannot support high-quality evidence and are also a waste of scientific research and medical resources.
Patient or healthcare consumer involvement:
In clinical trials, verifying the efficacy of a drug requires more patient involvement and better estimation of sample sizes to save patient resources and improve the accuracy of trial results.