OUTLOOK Towards better patient care: drugs to avoid in 2018 ABSTRACT there are no available treatments capable of improv- ing prognosis or quality of life, beyond their pla- ● To make it easier to choose quality care, and to cebo effect. prevent disproportionate harms to patients, Prescrire has published its annual update of drugs to avoid in the name of better care. his is Prescrire’s sixth consecutive annual review ● Prescrire’s assessments of the harm-benefit bal- of “drugs to avoid”, which includes document- ance of drugs in given situations are based on a Ted cases of drugs more dangerous than ben- rigorous procedure that includes a systematic and eficial (1,2). The aim is to make it easier to choose reproducible literature search, identification of safe, effective treatments, primarily to avoid expos- patient-relevant outcomes, prioritisation of the ing patients to unacceptable harms. This review is supporting data based on the strength of evidence, confined to drugs that should be avoided in all the comparison with standard treatments, and an ana- clinical situations for which they are authorised in lysis of both known and potential adverse effects. France or in the European Union. Drugs whose harm-benefit balance is unfavourable in a particular ● This annual review of drugs to avoid covers all situation are not included in our annual reviews of the drugs examined by Prescrire between 2010 and drugs to avoid if they have a favourable harm-benefit 2017 that are authorised in the European Union or balance in a different situation. in France. We identified 90 drugs (79 of which are marketed in France) that are more harmful than beneficial in all the indications for which they have A reliable, rigorous and independent been authorised. methodology ● In most cases, when drug therapy is really neces- What data sources and methodology do we use to sary, other drugs with a better harm-benefit balance assess a drug’s harm-benefit balance? are available. The following review concerns drugs and indica- tions on which we published detailed analyses in ● Even in serious situations, when no effective our French edition over an eight-year period, from treatment exists, there is no justification for pre- 2010 to 2017. Some drugs and indications were scribing a drug with no proven efficacy that pro- examined for the first time, while others were vokes severe adverse effects. It is sometimes re-evaluated as new data on efficacy or adverse acceptable to test these drugs in clinical trials, but effects became available. patients must be informed of the uncertainty over All our publications are intended to provide health their harm-benefit balance and of the trial’s objec- professionals (and thereby their patients) with the tives. Tailored supportive care should be used when clear, independent, reliable and up-to-date infor- PRESCRIRE INTERNATIONAL • APRIL 2018 • VOLUME 27 N° 192 • PAGE 107-1 Downloaded from english.prescrire.org on 30/09/2021 Copyright(c)Prescrire. For personal use only. OUTLOOK mation they need, free from conflicts of interest and Empirical data and personal experience: commercial pressures. risk of bias. Empirical assessment of a drug’s Prescrire is structured in such a way as to guaran- harm- benefit balance based on individual experience tee the quality of the information provided to our can help to guide further research but is subject to subscribers. The Editorial Staff comprise a broad major bias and represents only weak evidence (3,4). range of health professionals working in various For example, it can be difficult to attribute a specif- sectors and free from conflicts of interest. We also ic outcome to a particular drug, as other factors call on an extensive network of external reviewers must be taken into account, including the natural (specialists, methodologists, and practitioners rep- history of the disease, the placebo effect, the effect resentative of our readership), and each article un- of another treatment the patient may not have dergoes multiple quality controls and cross- checking mentioned, or a change in lifestyle or diet. Similar- at each step of the editorial process (see About ly, a doctor who sees an improvement in certain Prescrire > How we work at english.prescrire.org). patients may be unaware that many other patients Our editorial process is a collective one, as symbol- have been harmed by the same treatment (3). ised by the “©Prescrire” signature. The best way to minimise subjective bias caused Prescrire is also fiercely independent. Our work by non-comparative evaluation of a few patients is is funded solely and entirely by our subscribers. No to prioritise well-conducted clinical studies, particu- company, professional organisation, insurance larly double-blind, randomised trials versus standard system, government agency or health authority has care (3,4). any financial influence whatsoever over the content of our publications. Serious conditions with no effective treat- Comparison with standard treatments. The ment: patients should be informed of the harm-benefit balance of a given drug has to be consequences of interventions. When faced continually re-evaluated as new data on efficacy or with a serious condition for which there is no effect- adverse effects become available. Likewise, treatment ive treatment, some patients opt to forgo treatment options evolve as new drugs arrive on the market. while others are willing to try any drug that might Not all drugs are equal: some offer a therapeutic bring them even temporary relief, despite a risk of advantage, while others are more harmful than serious adverse effects. beneficial and should not be used (3). When the short-term prognosis is poor, some All Prescrire’s assessments of drugs and indica- health professionals may propose “last-chance” tions are based on a systematic and reproducible treatments without fully informing the patient of literature search. The resulting data are then analysed the harms, either intentionally or unwittingly. collectively by our Editorial Staff, using an estab- But patients in this situation must not be treated lished procedure: as guinea pigs. It is very useful to enrol patients – Efficacy data are prioritised: most weight is given into clinical trials provided they are informed of the to studies providing robust supporting evidence, harms and the uncertain nature of the benefits, and i.e. well-conducted, double-blind, randomised con- that the trial results are published in order to advance trolled trials; medical knowledge. – The drug is compared with a carefully chosen However, patients must be made aware that they standard treatment, if one exists (not necessarily a have the option of refusing to participate in clinical drug); trials or to receive last-chance treatments with an – The accent is placed on those clinical endpoints uncertain harm-benefit balance. They must also be most relevant to the patients concerned. This means reassured that, if they do refuse, they will not be that we often ignore surrogate endpoints such as abandoned but will continue to receive the best laboratory markers that have not been shown to available care. Even though they are not aimed at correlate with a favourable clinical outcome (4,5). modifying the outcome of the underlying disease, supportive care and symptomatic treatment are Careful analysis of adverse effects. Adverse useful elements of patient care. effects can be more difficult to analyse, as they are By their very nature, clinical trials involve a high often less thoroughly documented than efficacy, degree of uncertainty. In contrast, drugs used for and this discrepancy must be taken into account. routine care must have an acceptable harm-benefit The adverse effect profile of each drug is assessed balance. Marketing authorisation should only be by examining data from clinical trials and animal granted on the basis of proven efficacy relative to pharmacotoxicology studies, and any pharmaco- standard care, and an acceptable adverse effect logical affiliation. profile: in general, little, if any, extra information on The fact that a new drug has been granted mar- efficacy is collected once marketing authorisation keting authorisation does not signify that its has been granted (3). harm-benefit balance has been fully documented. Indeed, rare but serious adverse effects may only emerge after several years of routine use (3). PRESCRIRE INTERNATIONAL • APRIL 2018 • VOLUME 27 N° 192 • PAGE 107-2 Downloaded from english.prescrire.org on 30/09/2021 Copyright(c)Prescrire. For personal use only. OUTLOOK 90 authorised drugs that are more ● Ivabradine, an inhibitor of the cardiac If current, dangerous than beneficial can cause visual disturbances, cardiovascular dis- orders (including myocardial infarction), potential- As of early 2018, based on the drugs, examined by ly severe bradycardia and other cardiac arrhythmias. Prescrire between 2010 and 2017, that are authorised It has no advantages in either angina or heart failure in France or in the European Union, 90 drugs were (Prescrire Int n° 88, 110, 118, 155, 165; Rev Prescrire identified that are more dangerous than beneficial n° 403). Established treatments shown to be effect- in all their authorised indications. 79 of these drugs ive in angina include beta-blockers and the calcium are marketed in France (a,b). channel blockers amlodipine and verapamil. There They are listed below, based first on the thera- are also better options for heart failure: one is to peutic area in which they are used and then in al- refrain from adding another drug to an optimised phabetical order of their international nonproprietary treatment regimen; another is to use a beta-blocker names (INNs). with a proven impact on mortality. These 90 drugs comprise: ● Nicorandil, a vasodilator with solely symptomat- – Active substances with adverse effects that, given ic efficacy as a preventive treatment in effort an gina, the clinical situations in which they are used, are can cause severe mucocutaneous ulceration disproportionate to the benefits they provide; (Pres crire Int n° 81, 95, 110, 132).