BACKGROUND: Clinicians find standardized mean differences (SMDs) calculated from continuous outcomes difficult to interpret. Our objective was to determine theperformance of methods in converting SMDs or means to odds ratios of treatmentresponse and numbers needed to treat (NNTs) as more intuitive measures oftreatment effect. METHODS: Meta-epidemiological study of large-scale trials (≥100patients per group) comparing active treatment with placebo, sham ornon-intervention control. Trials had to use pain or global symptoms as continuousoutcomes and report both the percentage of patients with treatment response andmean pain or symptom scores per group. For each trial, we calculated odds ratios of observed treatment response and NNTs and approximated these estimates fromSMDs or means using all five currently available conversion methods by Hasselbladand Hedges (HH), Cox and Snell (CS), Furukawa (FU), Suissa (SU) and Kraemer andKupfer (KK). We compared observed and approximated values within trials byderiving pooled ratios of odds ratios (RORs) and differences in NNTs. ROR <1 and positive differences in NNTs imply that approximations are more conservative thanestimates calculated from observed treatment response. As measures of agreement, we calculated intraclass correlation coefficients. RESULTS: A total of 29 trials in 13 654 patients were included. Four out of five methods were suitable (HH, CS,FU, SU), with RORs between 0.92 for SU [95% confidence interval (95% CI),0.86-0.99] and 0.97 for HH (95% CI, 0.91-1.04) and differences in NNTs between0.5 (95% CI, -0.1 to -1.6) and 1.3 (95% CI, 0.4-2.1). Intraclass correlationcoefficients were ≥0.90 for these four methods, but ≤0.76 for the fifth method byKK (P for differences ≤0.027). CONCLUSIONS: The methods by HH, CS, FU and SU are suitable to convert summary treatment effects calculated from continuous outcomesinto odds ratios of treatment response and NNTs, whereas the method by KK isunsuitable.
Methods to convert continuous outcomes into odds ratios of treatment response and numbers needed to treat: meta-epidemiological study
Rutjes A;
2012-01-01
Abstract
BACKGROUND: Clinicians find standardized mean differences (SMDs) calculated from continuous outcomes difficult to interpret. Our objective was to determine theperformance of methods in converting SMDs or means to odds ratios of treatmentresponse and numbers needed to treat (NNTs) as more intuitive measures oftreatment effect. METHODS: Meta-epidemiological study of large-scale trials (≥100patients per group) comparing active treatment with placebo, sham ornon-intervention control. Trials had to use pain or global symptoms as continuousoutcomes and report both the percentage of patients with treatment response andmean pain or symptom scores per group. For each trial, we calculated odds ratios of observed treatment response and NNTs and approximated these estimates fromSMDs or means using all five currently available conversion methods by Hasselbladand Hedges (HH), Cox and Snell (CS), Furukawa (FU), Suissa (SU) and Kraemer andKupfer (KK). We compared observed and approximated values within trials byderiving pooled ratios of odds ratios (RORs) and differences in NNTs. ROR <1 and positive differences in NNTs imply that approximations are more conservative thanestimates calculated from observed treatment response. As measures of agreement, we calculated intraclass correlation coefficients. RESULTS: A total of 29 trials in 13 654 patients were included. Four out of five methods were suitable (HH, CS,FU, SU), with RORs between 0.92 for SU [95% confidence interval (95% CI),0.86-0.99] and 0.97 for HH (95% CI, 0.91-1.04) and differences in NNTs between0.5 (95% CI, -0.1 to -1.6) and 1.3 (95% CI, 0.4-2.1). Intraclass correlationcoefficients were ≥0.90 for these four methods, but ≤0.76 for the fifth method byKK (P for differences ≤0.027). CONCLUSIONS: The methods by HH, CS, FU and SU are suitable to convert summary treatment effects calculated from continuous outcomesinto odds ratios of treatment response and NNTs, whereas the method by KK isunsuitable.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
