Article type
Year
Abstract
Background: Clinical interpretation of changes measured on a scale is dependent on knowing the minimum clinically important difference (MCID) for that scale: the threshold above which we (e.g. clinicians, patients, and researchers) perceive a difference in an outcome. Until now, approaches to determining the MCID were based upon individual studies or surveys of experts. However, the comparison of treatment effects derived from a pairwise meta-analysis or network meta-analysis (NMA) to all trial-specific results in a meta-analysis could improve our clinical understanding of treatment effects derived from meta-analysis models. Furthermore, the calculation of a MCID based on a systematic review could enhance clinical decision-making when the MCID for a scale is unknown.
Objectives: To demonstrate how a distribution-based approach of pooled standard deviations (SDs) can be used to estimate MCIDs.
Methods: We approximated MCIDs using a distribution-based approach that pooled SDs associated with baseline mean or mean change values for two scales (i.e. Mini-Mental State Exam [MMSE] and Alzheimer Disease Assessment Scale – Cognitive Subscale [ADAS-Cog]), as reported in parallel randomized controlled trials (RCTs) that were included in a systematic review of cognitive enhancing medications for dementia (i.e. cholinesterase inhibitors and memantine). We excluded RCTs that did not report baseline or mean change SD values. We derived MCIDs at 0.4 and 0.5 SDs of the pooled SD.
Results: We showed that MCIDs derived with a distribution-based approach approximated published MCIDs for the MMSE and ADAS-Cog. For the MMSE (51 RCTs, 12449 patients), we estimated a MCID of 1.6 at 0.4 SDs and 2 at 0.5 SDs based on baseline SDs and we estimated a MCID of 1.4 at 0.4 SDs and 1.8 at 0.5 SDs based on mean change SDs . For the ADAS-Cog (37 RCTs, 10006 patients), we estimated a MCID of 4 at 0.4 SDs and 5 at 0.5 SDs based on baseline SDs and we estimated a MCID of 2.6 at 0.4 SDs and 3.2 at 0.5 SDs based on mean change SDs. MCIDs were unchanged when we excluded studies where SDs were estimated from other measures of uncertainty (e.g. standard error, 95% confidence interval).
Conclusions: A distribution-based approach using data included in a systematic review can approximate MCIDs. Our approach performed better when we derived MCIDs from baseline as opposed to mean change SDs . This approach could facilitate clinical interpretation of outcome measures reported in RCTs and systematic reviews of interventions. Future research should focus on the generalizability of this method to other clinical scenarios.
Patient or healthcare consumer involvement: Two clinicians (Straus and Watt) were involved in the design and interpretation of study results.
Objectives: To demonstrate how a distribution-based approach of pooled standard deviations (SDs) can be used to estimate MCIDs.
Methods: We approximated MCIDs using a distribution-based approach that pooled SDs associated with baseline mean or mean change values for two scales (i.e. Mini-Mental State Exam [MMSE] and Alzheimer Disease Assessment Scale – Cognitive Subscale [ADAS-Cog]), as reported in parallel randomized controlled trials (RCTs) that were included in a systematic review of cognitive enhancing medications for dementia (i.e. cholinesterase inhibitors and memantine). We excluded RCTs that did not report baseline or mean change SD values. We derived MCIDs at 0.4 and 0.5 SDs of the pooled SD.
Results: We showed that MCIDs derived with a distribution-based approach approximated published MCIDs for the MMSE and ADAS-Cog. For the MMSE (51 RCTs, 12449 patients), we estimated a MCID of 1.6 at 0.4 SDs and 2 at 0.5 SDs based on baseline SDs and we estimated a MCID of 1.4 at 0.4 SDs and 1.8 at 0.5 SDs based on mean change SDs . For the ADAS-Cog (37 RCTs, 10006 patients), we estimated a MCID of 4 at 0.4 SDs and 5 at 0.5 SDs based on baseline SDs and we estimated a MCID of 2.6 at 0.4 SDs and 3.2 at 0.5 SDs based on mean change SDs. MCIDs were unchanged when we excluded studies where SDs were estimated from other measures of uncertainty (e.g. standard error, 95% confidence interval).
Conclusions: A distribution-based approach using data included in a systematic review can approximate MCIDs. Our approach performed better when we derived MCIDs from baseline as opposed to mean change SDs . This approach could facilitate clinical interpretation of outcome measures reported in RCTs and systematic reviews of interventions. Future research should focus on the generalizability of this method to other clinical scenarios.
Patient or healthcare consumer involvement: Two clinicians (Straus and Watt) were involved in the design and interpretation of study results.