Ann Robinson’s research reviews—24 September 2018

Ann Robinson reviews the latest research from the top medical journals

Non, Je ne regrette rien
I’ve been a GP for so long that I can’t remember why I chose it as a specialty in the first place. All I know is that I have no regrets—not often anyway. One useful piece of information in making an informed career choice would be to know which specialties in medicine have the highest rates of burnout and career regret.

A prospective cohort study of 3588 US second year residents (equivalent to foundation year 2 in the UK) found nearly half (45.2%) reported symptoms of burnout and 14% regretted their career choices. But both measures of burnout (“I feel burned out from my work” and “I’ve become more callous towards people since I started this job”) and regret (“Would I choose this career again?”) varied hugely depending on which specialties the residents were working in. Doctors in urology, neurology, emergency medicine, and surgery were more fed up than those in dermatology, pathology, and internal medicine. Residents had filled in a questionnaire while still at medical school; those with higher self-reported levels of anxiety as students were more likely to report burnout and regret once working, and those with high levels of empathy were less prone to regret. How can it be that so many of these young doctors, who are only at the start of their careers, are so unhappy in their career choice?

A linked editorial wisely points out that a huge meta-analysis (182 studies, 45 countries) of burnout in the same journal says that reaching any conclusions about prevalence is impossible given the heterogeneity in study design, methods of measuring, and response rates. It warns the medical profession against taking “a self-reported complaint of unhappiness and dissatisfaction” and turning it into “a call for action on what is claimed to be a national epidemic that purportedly affects half to two-thirds of practicing physicians.” But that doesn’t mean we can dismiss the subject altogether; more work is badly needed to understand what makes for a happy doctor who can flourish at work and get to the end of their career without major regrets.

Doctors doctor while scribes scribe
Here’s one practical idea to prevent burnout. A scribe! How biblical that sounds, conjuring an image of a bearded man wielding a quill and parchment. In reality, scribes wear modern dress and act as a personal assistant, entering information into the electronic health records (EHR) while the doctor gets on with the job; interacting and listening to the patient without needing to “feed the machine.” Despite the challenges of keeping the EHR up to date,  I suspect that most of us would not want to go back to the pre-EHR days of illegible entries, difficulty retrieving information, and lack of audit trails (although I’m reliably informed that there are still some UK hospital that use paper notes!). A 12 month crossover study in the US randomly assigned medical scribes to some primary care physicians (PCPs) and not others and then swapped them over every three months for a year. The PCPs completed a survey at the end of each three month period and, unsurprisingly, the periods with a scribe resulted in less time catching up on EHR work and higher rates of spending at least 75% of the consultation interacting with the patient rather than staring at the computer. The patients mostly welcomed a scribe and only 2.4% expressed unease. I’m not convinced that I’d like a scribe; writing up a consultation helps me to organise my thoughts and check the management plan. But I guess you get used to thinking out loud while someone else enters your thoughts.  

Obesity: what’s to be done?
We all know that the population is getting fatter; 35% of men and 40% of women in the United States are obese and the UK is not far behind. But what on earth can we offer our obese patients? What works apart from bariatric surgery? The US Preventive Services Task Force (USPSTF) has reviewed the evidence on behavioral and pharmacotherapy interventions that are available in primary care in the US. It concludes with “moderate certainty” that behavioural interventions have a “moderate net benefit” in obese adults (BMI >30). It’s hardly a magic bullet but its the best that we can offer without surgery. Disappointingly, the nature of the data means that you can’t say how many sessions and which format works best (face to face or remote, group or individual). Combining drugs with behavioural interventions is probably more effective than behavioural interventions alone in achieving and maintaining weight loss. But, the authors warn, the trials including pharmacotherapy have highly selective inclusion criteria, high attrition rates, and poor long term follow-up. So the USPSTF sticks with its recommendation that behavioral interventions should be the primary focus of effective interventions for weight loss in adults. It’s so much easier said than done.

Could Lorcaserin reduce obesity?
In the UK, even if we want to prescribe drugs to aid weight loss, it’s orlistat or nothing. Patients can buy it over the counter or online as Alli. Lorcaserin, a novel selective serotonin 2C receptor agonist that modulates appetite, is available in the US but not yet approved in the EU. It works—to an extent—with average weight loss of 2.8kg more than placebo over a 40 month period. But cardiovascular safety has been a problem with obesity drugs such as sibutramine, and this pharma funded, international randomised controlled trial (RCT) looked at 12 000 overweight or obese patients with known cardiovascular disease or multiple risk factors to see whether daily lorcaserin increased the risk of a major cardiovascular event, compared to placebo. Happily it didn’t; rates of a major CV event were around 4% in both groups. Just over a third of those on lorcaserin had lost at least 5% of their body weight at one year (nearly two thirds hadn’t, so it’s not that good), compared to less than a fifth in the placebo group. Lorcaserin is expensive and not stunningly effective but at least it appears to be safe.  

Rivaroxaban doesn’t prevent venous thromboembolism after medical admission
It’s good to know what doesn’t work. Patients who have been in hospital for a medical (rather than surgical) reason remain at increased risk of venous thromboembolism (VTE) when they get home. Should they be given thromboprophylaxis? After all, it’s easy enough now that we have direct acting oral anticoagulants (DOACs, the drugs formerly known as NOACs) such as rivaroxaban and don’t have to faff about with warfarin monitoring. An RCT of over 12 000 patients at increased risk of VTE (using the IMPROVE score) compared patients given rivaroxaban to those given placebo for 45 days after discharge. Only around 1% developed symptomatic or fatal VTE in this high risk group, whether they were on rivaroxaban or not. Very few developed major bleeding in either group (0.28% in the rivaroxaban group, 0.15% on placebo). So the absolute risk of VTE is not very high and rivaroxaban doesn’t reduce it. But it is good to have further confirmation that if we do need to prescribe a DOAC for a proven VTE, the incidence of major bleeding risk in this study was reassuringly low.

Common treatments of basal cell carcinoma are much of a muchness
Basal cell carcinoma (BCC) will hardly ever kill you but we treat them because if left untreated, they can cause disfigurement, disability, and poor quality of life. But which treatment option would you opt for if you developed a BCC? Most interventions for BCC haven’t been compared in head-to-head randomised trials. This review of studies evaluated the comparative effectiveness and safety of treatments of primary BCC in adults. The results support the effectiveness of commonly used methods for low risk BCC; surgery, external beam radiation, topical imiquimod, and curettage and diathermy. The ideal management of high risk BCC subtypes remains uncertain.  Further research is required, and I’m none the wiser as to which treatment I’d opt for if/when I get a BCC; I’d probably leave it to the dermatologist to decide.

Does gluten in pregnancy make kids more prone to type 1 diabetes?
Type 1 diabetes is increasing faster than genetic drift (the evolutionary change in frequency of a gene variant due to chance) would explain. So what else is going on? Environmental factors are likely to be contributing and attention has focused on the role of gluten proteins (in wheat, rye, and barley) in the development of diabetes. Gluten proteins are more immunogenic than other dietary proteins because they contain lots of the hydrophobic amino acids proline and glutamine that don’t degrade readily in the intestine.  And prenatal exposure to gluten could be relevant to type 1 diabetes development, because the process leading to islet autoimmunity may begin in fetal life. So the question is: does eating a high gluten diet while pregnant increase the risk of having a child who develops type 1 diabetes? In this prospective cohort study, over 66 000 Danish women kept detailed food diaries while pregnant. Interestingly, the ones who ate most gluten (and calories) had the lowest BMI’s. Overall, 0.37% of the offspring had developed type 1 diabetes by the age of 15.6. The risk increased proportionally with increased maternal gluten intake during pregnancy, which certainly warrants further study. Major limitations to the study were that the absolute risk of type 1 diabetes was very low so numbers were small and we don’t know how mums fed their offspring once they were born. It’s bound to make headlines but it’s way too soon to load yet another dietary admonishment on pregnant women for the time being.

A clear question, with a clear answer
Children can develop bronchiectasis from infections (pneumonia, Tb, or whooping cough); cystic fibrosis; immunodeficiency; or other causes of lung damages. In about 40% of children and adults with bronchiectasis, no cause is found.  Non severe acute exacerbations are usually treated with amoxicillin–clavulanate (Augmentin) but the authors of this Australian/New Zealand study say that azithromycin is also often prescribed because of its convenient once-daily dosing. They wondered if azithromycin was as effective as amoxicillin-clavulanate, couldn’t find any head-to-head studies, and so did one. Simple question with a simple answer; azithromycin is basically fine as an alternative. By 21 days of treatment, 83% of exacerbations had resolved in both groups (though more quickly with amoxicillin-clavulanate). Azithromycin is a good option for children who are allergic to penicillin and once-daily dosing is a big advantage in some families. It must be balanced against a risk of treatment failure, longer exacerbation duration, and the risk of inducing macrolide resistance, say the authors. If only all research asked a clear, relevant question that yielded a straightforward, practical answer.

Ann Robinson is an NHS GP and health writer/broadcaster. She works within her local community and is a trustee of the Anthony Nolan charity.