Editor’s view—beware claims of efficacy in frailty in cardiovascular trials

I strongly recommend having a read of Lucy Pollock’s excellent editorial on how we are all being potentially mis-sold messages on the efficacy of cardiovascular treatments in people with frailty [1]. It is so easy to assume that trials that report sub-group analyses by frailty do so using some validated measure of frailty, especially when they claim to use familiar tools to measure frailty, such as the ‘Frailty Index.’ However, the term is not exactly copyrighted, and many trials devise their own, non-validated indices that are of dubious value to say the least. The editorial accompanies an important piece of work that we published in our last issue [2] that uncovers widespread over-use of cardiovascular indices as components of frailty indices, on occasion even outscoring much more important indicators of frailty, such as cognitive impairment. This was especially prevalent in trials in cardiovascular conditions such as heart failure. This issue strikes at the heart of evidence-based medicine and undermines the credibility of research. So perhaps giving a frail patient with newly diagnosed heart failure quadruple therapy is not such a good idea after all? We should not have to go back to revisit these questions and re-examine the evidence in forensic detail if researchers were more honest in their approach. For now, beware claims of efficacy in frailty. I recommend having a good look at how frailty was measured—if this is not transparently reported, assume this was probably not what Age & Ageing readers call frailty. Although we are the journal of the British Geriatrics Society, we take great pride in being a truly international journal. One ongoing lament is that, despite great willingness over many years, we do not publish many papers from low and middle-income countries (LMIC). One reason might be that our editorial team and excellent peer reviewers, who are mostly from higher-income countries, may not always be in tune with the challenges of research in this setting. I am pleased to have recently appointed a new associate editor for LMIC submissions, Dr Leon Geffen, who will handle selected manuscripts. Meanwhile, we publish a meta-analysis of epidemiological studies of late-life depression in sub-Saharan Africa, which finds a high prevalence throughout the region [3]. The highest rates were in East Africa and in low-income countries, with the authors recommending targeted mental health policies for older adults in these settings. I thought the answer to this question was probably ‘no,’ but also thought I was not sure. Now, thanks to a systematic review and meta-analysis of randomised controlled trials published in our journal [4] I am certain that I am not sure! The authors found ten trials, but together they only recruited a measly 516 participants and they found high levels of heterogeneity in results, suggesting caution is needed before making any conclusions. The mean improvement in sleep in those receiving the intervention was an extra 12 minutes per night (P < .001) and there was a somewhat surprising improvement in mini-mental state examination scores of 1.8 points (P = .02) and in a measure of neuropsychiatric symptoms. The authors find these fall short of clinically important thresholds and I agree with them in recommending larger trials be undertaken to confirm effects before we change our practice. For now, I will stick to recommending watching Aberdeen FC’s matches, but only on television—it is always freezing in Pittodrie. Social media (and sometimes the mainstream media, too) can be awash with stories of serious side effects of vaccines, even though they are very rare. I wonder if any outlets will pick up the results of a meta-analysis by Stefania Maggi and colleagues in our latest issue [5] that showed recipients of the herpes zoster vaccine were half as likely to be diagnosed with Alzheimer’s disease and 25% less likely to be diagnosed with any dementia. Some other vaccines were also associated with lower incidence of dementia. The pro-inflammatory potential of many viral infections and their propensity to cause cardiovascular disease and neurodegeneration is probably under-appreciated, but this work suggests some of that could be mitigated through vaccination. Although there may be an element of selection bias driving the results, this paper suggests adult vaccination may have an important role to play in limiting dementia prevalence. Given falling vaccination uptake and the low rates of adult vaccination in many countries, it is important to spread the message that vaccination has a role in promoting healthy ageing beyond infection control. I always enjoy a paper that focuses on person-centredness and reminds us what matters most to the people we serve. We have previously published some outstanding papers on this and I am pleased to have two additional publications that add to the evidence base. Firstly, a paper from the Netherlands explores what happens if we ask attendants at the Emergency Department, ‘What matters most to you?’ [6]. They found this was a feasible and worthwhile exercise, with younger patients prioritising medical concerns whilst older ones prioritised relationships, functional abilities and emotional well-being over medical aspects. This may not be a surprise to many readers, but it is nevertheless satisfying to see concrete evidence of a shift in patient priorities with age that is consistent with our own experience. This matters because it mandates a change in the way we assess and manage patients at the front door, depending on their priorities, a world away from the horrible conveyer belt of disease-centric ‘pathways’ that seems to dominate acute and emergency medicine. In Parkinson’s disease, Wilson and colleagues synthesise 26 qualitative reports on the experience and perceived impacts of treatments [7]. Their proposed conceptual framework representing the treatment experience provides patients, carers and clinicians with the key considerations required to personalise treatments for Parkinson’s disease. To finish off, we have a selection of studies that test tools for predicting future events. When used sensibly, these are a valuable adjunct to medical management as patients value information on prognosis and a steer on likely future eventualities to guide their own treatment. Predicting adverse drug reactions is a fool’s game in my experience, possibly because there are so many different potential events and they do not all share common risk factors (e.g. the risk factors for developing drug-related diarrhoea or constipation are unlikely to be same). However, in the latest issue, we published a study that reasonably predicts severe toxicity in older patients receiving systemic anti-cancer therapy for solid tumours [8]. Ning and colleagues provide preliminary data that shows using electroencephalography to measure occipital alpha power can significantly improve prediction of post-operative delirium [9]. Their findings will need to be confirmed in a larger study. Finally, Bian and colleagues suggest measures of inflammation are a better predictor of future major acute cardiovascular events than lipid levels in older people with established atherosclerotic disease [10]. They suggest targeting inflammation may be more effective than lipid-lowering therapy for secondary prevention in this group. None declared.