Antipsychotics for the treatment of schizophrenia: likelihood to be helped or harmed, understanding proximal and distal benefits and risks

Abstract

Benefit-risk decisions are the central part of the philosophy of evidence-based medicine. Although number needed to treat (NNT) and number needed to harm (NNH) can quantify differences between two antipsychotics in terms of benefits and risks for the treatment of schizophrenia, these benefits and risks can take on greatly differing degrees of importance or relevance depending on the subjective point of view of the patient and clinician, baseline risks, severity of the underlying illness, as well as the time horizon when these effects emerge. The metric of likelihood to be helped or harmed, the ratio NNH to NNT, can be helpful in quantifying the benefit:risk ratio, provided that the outcomes are carefully matched in terms of both importance to the clinician and the patient and whether they are proximal or distal. The examples provided are extensions to the initially published NNT and NNH analyses conducted by the first author.

Important clinical features of atypical antipsychotics in acute bipolar depression that inform routine clinical care: a review of pivotal studies with number needed to treat

“…The likelihood to be helped or harmed (LHH), the ratio of NNH to NNT, has been used to quantify the benefi ts and risks for evidencebased medicine [11,15,31] . The LHH can provide an estimate of the trade-offs between benefi ts and risks.…”

The ABC's of dopamine receptor partial agonists - aripiprazole, brexpiprazole and cariprazine: the 15-min challenge to sort these agents out

“…There are several important caveats. The harms discussed are primarily from acute studies and do not reflect effects that can take time to become manifest, such as tardive dyskinesia, the long‐term accumulation of body weight, or the development of insulin resistance/type 2 diabetes mellitus . The data presented are from carefully conducted registration trials that enrolled subjects who fulfilled restrictive inclusion/exclusion criteria.…”

A Systematic Review of Duloxetine for Osteoarthritic Pain: What is the Number Needed to Treat, Number Needed to Harm, and Likelihood to be Helped or Harmed?

“…18 Likelihood to be helped or harmed can be a useful way of synthesizing data regarding benefits and risks. Likelihood to be helped or harmed is calculated using the following formula: LHH = NNH/NNT.…”

“…3 Approximately 16.3% (39/239) of the patients who received duloxetine in the 13-week placebo-controlled trials for chronic pain due to OA discontinued treatment due to an adverse reaction, compared with 5.6% (14/248) for placebo, for an NNH of 10 (95% CI, [7][8][9][10][11][12][13][14][15][16][17][18][19][20]. Among all patients receiving duloxetine across indications, the most commonly observed adverse reactions (incidence of $ 5% and at least twice the incidence in placebo patients) were nausea, dry mouth, somnolence, fatigue, constipation, decreased appetite, and hyperhidrosis.…”