Assessing causal relationships between treatments and clinical outcomes: always read the fine print

Abstract

Changes in clinical practice should be driven by relevant and reliable evidence. Hence, adoption of a new therapy requires demonstrating that it provides (causes) benefit. Such evidence is generally obtained from intent-to-treat analyses of randomized clinical trials (RCTs). In this paper, we review other approaches to assessing the causal relationship between treatments and outcomes: (1) inference from non-randomized (observational) studies, (2) analysis of randomized studies where patients received treatments other than those to which they were randomized and (3) analysis of studies where the outcome of interest is sometimes unobservable because of a competing event (competing risks). We conclude that for the practice-changing demonstration of a favorable benefitto-risk ratio, the gold standard is the intent-to-treat analysis of RCTs. At the same time, we illustrate how careful application of special statistical methods for assessment of treatment-outcome causation can be instrumental in complementing existing randomized evidence and guiding design of future research.

Targeted Provision of Oral Iron: The Evolution of a Practical Screening Option

“…An efficacious intervention sometimes, however, implies a concomitant harmful consequence for the treated party or adverse outcomes induced in some members of the larger population (7, 8). In fact, in the universal public health prophylaxis scenario, some people may not benefit at all, and others may only be harmed, also without receiving any benefit.…”

First‐line antibiotic treatment in patients with localized extragastric mucosa‐associated lymphoid tissue lymphoma

“…Meanwhile, the clinical response rates (CR, PR, and ORR) and the clinical outcomes (7‐year EFS and OS) of patients who received first‐line antibiotics were not inferior to those of our localized cohort of patients with extragastric MALT lymphoma who received conventional treatment. However, the ORRs and clinical outcomes of our patients who received explorative treatment (first‐line antibiotics) should be cautiously compared with those of similar patients (no difference in age, sex, and stage) who received conventional treatment owing to the selection bias caused by the use of antibiotic treatment, and the heterogeneity of the primary location of extragastric MALT lymphomas may interfere with the presentation of comparisons between these two subgroups [ 31 , 32 ].…”

Bayesian analysis of longitudinal studies with treatment by indication

“…This type of confounding can generally be avoided by comparing groups of subjects who presented with similar indications at similar times after their initial diagnosis; however, these variables are typically not recorded for subjects who do not receive treatment. This issue is especially common in studies that rely on data from electronic medical records, administrative databases, and health registries (Byar, 1980;Levine and Julian, 2008;Freidlin and Korn, 2012). Thus, for the group of untreated subjects, it is unclear whether treatment was indicated at an unknown time during the study and was purposefully withheld at that time or if treatment was simply never considered due to lack of indications.…”