Abstract:Conventional observational epidemiology has an unenviable reputation for generating false-positive findings, or “scares,” as others call them. In 1993, for example, the New York Times reported that “vitamin E greatly reduces the risk of heart disease” following simultaneous publication of 2 observational studies in the New England Journal of Medicine demonstrating that use of vitamin E supplements, even for just a few years, was associated with a substantially lower risk of coronary heart disease (CHD). Randomized controlled trials (RCTs)—testing precisely the same hypothesis—revealed no reduction in risk at all. As I write this, UK news media are reporting that “junk food makes your kids dumb,” in response to a paper reporting that children given a poor diet at age 3 had lower IQ scores 5 years later. In the latter case, however, no RCTs are ever likely to be carried out, and the status of the finding will remain liminal. The ratio of false-positive to false-negative (FP:FN) published findings in traditional epidemiology is very high, John Ioannidis and colleagues argue. But what exactly are “false-positive” findings in the context of conventional observational epidemiology, and how can their prevalence be quantified? My suspicion is that use of vitamin E supplements would—in most contexts—be associated with lower risk of CHD, because of substantial confounding by a myriad of measured and unmeasured factors related to such a socially and behaviorally patterned exposure. The same will apply to children fed on junk food and their later intelligence. The inability to substantially “control” statistically for confounding in many situations—due to measurement error in assessed confounders and omission of others—remains underappreciated. In the sense that these associations do exist, they should perhaps not be called false positives; they are false positives only if they are taken to be indicators of underlying causal effects. Noncausal but replicable observational associations could clearly be of considerable value through prediction of disease risk, and targeting of preventive measures to those who can benefit most. For the holy grail of epidemiology—identifying causes of disease—they are a disappointment, however. A wide range of approaches—from formally comparing associations in contexts where confounding structures differ, utilizing correlates of the exposure under study that are not plausible causes and natural experiments, through to formal instrumental variables methods—offer greater hope for reliable causal inference than plowing on with traditional approaches and keeping the FP:FN ratio high. Ioannidis and colleagues reiterate the low positive predictive value of a nominally “significant” P, something I (along with a generation of epidemiologists, I imagine) first encountered in Michael Oakes’ seminal “Statistical Inference, although (as has been

Heterodera glycines in Indiana: I. Reproduction of Geographical Isolates on Soybean Differentials.

Analgesic effect of parecoxib sodium and oxycodone for postoperative pain after total knee arthroplasty

Pre-emptive injection of peri-articular-multimodal drug for post-operative pain management in total knee arthroplasty: a double-blind randomized clinical trial

Study of Efficacy of Extraosseous Local Infiltration of Multimodal Drug Cocktail for Pain Management After Total Joint Arthroplasty in Lower Limb

A Review on the Efficacy of Extraosseous Local Infiltration of Multimodal Drug Cocktail for Pain Management After Total Knee or Hip Arthroplasty

Childhood abuse in adults in primary care: Empirical findings and clinical implications

Efficacy and safety of Postoperative Intravenous Parecoxib sodium Followed by ORal CElecoxib (PIPFORCE) post-total knee arthroplasty in patients with osteoarthritis: a study protocol for a multicentre, double-blind, parallel-group trial.

Postoperative Intravenous Parecoxib Sodium Followed by Oral Celecoxib Post Total Knee Arthroplasty in Osteoarthritis Patients (PIPFORCE): a Multicentre, Double-Blind, Randomised, Placebo-Controlled Trial

Patient-controlled epidural analgesia or multimodal pain regimen with periarticular injection after total hip arthroplasty: a randomized, double-blind, placebo-controlled study

Pre-emptive use of prolonged release oxycodone/naloxone combo significantly reduces postoperative pain after total hip arthroplasty

Random allocation in observational data: how small but robust effects could facilitate hypothesis-free causal inference.

Brief Report: The Euro‐Lupus Low‐Dose Intravenous Cyclophosphamide Regimen Does Not Impact the Ovarian Reserve, as Measured by Serum Levels of Anti–Müllerian Hormone

Intravenous Vs Oral Acetaminophen As a Component of Multimodal Analgesia after Total Hip Arthroplasty: A Randomized, Blinded Trial.

Efficacy of Multimodal Cocktail Periarticular Injection with or Without Steroid in Total Knee Arthroplasty

Seizure outcome in patients undergoing temporal lobe resections for epilepsy.

Pain management after total hip arthroplasty: comparative study of analgesic efficacy and tolerability between oral tramadol/dexketoprofen and injectable paracetamol + tramadol

Safety and Efficacy of Perioperative Intravenous Meloxicam for Moderate-to-Severe Pain Management in Total Knee Arthroplasty: A Randomized Clinical Trial

Can Multimodal Analgesia Reduce Postoperative Opioid Consumption in Patients Undergoing Shoulder Arthroscopy? A Retrospective Study

Perioperative Pain Management in Total Knee Arthroplasty: A Narrative Review of Current Multimodal Analgesia Protocols

Efficacy of intravenous acetaminophen in multimodal management for pain relief following total knee arthroplasty: a meta-analysis

4He(4He,3He)5He(g.s.) reaction at 118 MeV, and its distorted wave Born approximation interpretation.