Absolute risks rather than incidence rates should be used to estimate the number needed to treat from time-to-event data

Bender, Ralf; Kromp, Mandy; Kiefer, Corinna; Sturtz, Sibylle

doi:10.1016/j.jclinepi.2013.01.011

Cited by 10 publications

(4 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More recently, however, the authors of a study comparing the risk difference approach (reciprocal of risk differences estimated by survival time methods) and the incidence difference approach (reciprocal of incidence rates differences) concluded that the methods based on incidence rates often lead to misleading NNT estimates and recommended the use of survival time methods to estimate NNTs in RCTs with time-to-event outcomes [28]. The incidence difference approach still can be used in the case of small baseline risks, strong treatment effects, and exponentially distributed survival times [28]. Nevertheless, Girerd et al argued that the two methods measure different things, but both are valid and provide complementary information regarding the absolute effect of an intervention, highlighting that the incidence rate approach assesses person-years rather than persons [55].…”

Section: Discussionmentioning

confidence: 99%

“…This calculating method estimates the number of person-times (e.g. patient-years), not the absolute number of persons, needed to observe one less (or one more) event in the treatment group than in the control group [28, 29, 54–56]. This estimate is different from the “classical” person-based NNT, and therefore may be difficult to interpret [56].…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Number needed to treat (NNT) in clinical literature: an appraisal

Mendes

Alves

Marques

2017

BMC Med

132

View full text Add to dashboard Cite

BackgroundThe number needed to treat (NNT) is an absolute effect measure that has been used to assess beneficial and harmful effects of medical interventions. Several methods can be used to calculate NNTs, and they should be applied depending on the different study characteristics, such as the design and type of variable used to measure outcomes. Whether or not the most recommended methods have been applied to calculate NNTs in studies published in the medical literature is yet to be determined. The aim of this study is to assess whether the methods used to calculate NNTs in studies published in medical journals are in line with basic methodological recommendations.MethodsThe top 25 high-impact factor journals in the “General and/or Internal Medicine” category were screened to identify studies assessing pharmacological interventions and reporting NNTs. Studies were categorized according to their design and the type of variables. NNTs were assessed for completeness (baseline risk, time horizon, and confidence intervals [CIs]). The methods used for calculating NNTs in selected studies were compared to basic methodological recommendations published in the literature. Data were analyzed using descriptive statistics.ResultsThe search returned 138 citations, of which 51 were selected. Most were meta-analyses (n = 23, 45.1%), followed by clinical trials (n = 17, 33.3%), cohort (n = 9, 17.6%), and case–control studies (n = 2, 3.9%). Binary variables were more common (n = 41, 80.4%) than time-to-event (n = 10, 19.6%) outcomes. Twenty-six studies (51.0%) reported only NNT to benefit (NNTB), 14 (27.5%) reported both NNTB and NNT to harm (NNTH), and 11 (21.6%) reported only NNTH. Baseline risk (n = 37, 72.5%), time horizon (n = 38, 74.5%), and CI (n = 32, 62.7%) for NNTs were not always reported. Basic methodological recommendations to calculate NNTs were not followed in 15 studies (29.4%). The proportion of studies applying non-recommended methods was particularly high for meta-analyses (n = 13, 56.5%).ConclusionsA considerable proportion of studies, particularly meta-analyses, applied methods that are not in line with basic methodological recommendations. Despite their usefulness in assisting clinical decisions, NNTs are uninterpretable if incompletely reported, and they may be misleading if calculating methods are inadequate to study designs and variables under evaluation. Further research is needed to confirm the present findings.Electronic supplementary materialThe online version of this article (doi:10.1186/s12916-017-0875-8) contains supplementary material, which is available to authorized users.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Number needed to treat (NNT) in clinical literature: an appraisal

Mendes

Alves

Marques

2017

BMC Med

132

View full text Add to dashboard Cite

show abstract

“…ARR and NNT were estimated from Kaplan-Meier curves at the median of follow-up time (4.4 years). 30 Tests for interaction of treatment effect over the subgroups were obtained by the Cox regression model for the HR and a Cochran’s Q test for the ARR. The threshold for significance for treatment effect was set at 0.05 for the main analysis and subgroup analysis.…”

Section: Methodsmentioning

confidence: 99%

Effectiveness of blood pressure-lowering drug treatment by levels of absolute risk: post hoc analysis of the Australian National Blood Pressure Study

Breslin

Doust

et al. 2018

BMJ Open

View full text Add to dashboard Cite

ObjectivesIn many current guidelines, blood pressure (BP)-lowering drug treatment for primary prevention of cardiovascular disease (CVD) is based on absolute risk. However, in clinical practice, therapeutic decisions are often based on BP levels alone. We sought to investigate which approach was superior by conducting a post hoc analysis of the Australian National Blood Pressure (ANBP) cohort, a seminal study establishing the efficacy of BP lowering in ‘mild hypertensive’ persons.DesignA post hoc subgroup analysis of the ANBP trial results by baseline absolute risk tertile.Setting and participants3244 participants aged 35–69 years in a community-based randomised placebo controlled trial of blood pressure-lowering medication.InterventionsChlorothiazide500 mg versus placebo.Primary outcome measuresAll-cause mortality and non-fatal events (non-fatal CVD, congestive cardiac failure, renal failure, hypertensive retinopathy or encephalopathy).ResultsTreatment effects were assessed by HR, absolute risk reduction and number needed to treat. Participants had an average 5-year CVD risk in the intermediate range (10.5±6.5) with moderately elevated BP (mean 159/103 mmHg) and were middle aged (52±8 years). In a subgroup analysis, the relative effects (HR) and absolute effects (absolute risk reduction and number needed to treat) did not statistically differ across the three risk groups except for the absolute benefit in all-cause mortality (p for heterogeneity=0.04). With respect to absolute benefit, drug treatment significantly reduced the number of events in the high-risk group regarding any event with a number needed to treat of 18 (10 to 64), death from any cause with 45 (25 to 196) and major CVD events with 23 (12 to 193).ConclusionOur analysis confirms that the benefit of treatment was substantial only in the high-risk tertile, reaffirming the rationale of treating elevated blood pressure in the setting of all risk factors rather than in isolation.

show abstract

“…In the case of a single study with individual participant data and incomplete observation, reviewers should apply methods for survival data. Methods based upon Kaplan-Meier curves (7,9) and the Cox regression model (10)(11)(12) are available to perform point and interval estimates for the risk difference at different time-points.…”

Section: Direct Calculation Of the Absolute Effectmentioning

confidence: 99%