Ann Robinson’s research reviews—3 September 2021

Ann Robinson reviews the latest research from the top medical journals

The rise and rise of childhood diabetes in the US

This vast observational, cross sectional, multicentre study of 3.47 million US teenagers under the age of 19 years, found that the estimated prevalence of type 1 and type 2 diabetes increased substantially between 2001 and 2017 (type 1 from 1.48/1000 to 2.15/1000 and type 2 from 0.34/1000 to 0.67/1000). There are a few caveats: some cases of relatively asymptomatic type 2 diabetes may have been missed, there were few Asian or Pacific Islanders included in the study so prevalence rates may not be accurate, and trends between 2017 and now may have changed (although there’s no reason they should have done). Developing diabetes in childhood or adolescence is known to increase the risk of early complications, comorbidities, and premature death, particularly in those with type 2 diabetes and from ethnic minority groups. Type 2 diabetes remains less common than type 1 in the young, but rates are rising and it’s easier to miss. The ongoing rise in prevalence of both types of diabetes among young people is of grave concern.

JAMA doi:10.1001/jama.2021.11165

Antipsychotics in pregnancy—no need to stop

A large and well conducted cohort study of 411 251 mother-child pairs in Hong Kong found that there was no increased risk of attention deficit/hyperactivity disorder (ADHD), autism spectrum disorders (ASD), prematurity, or small for gestational age babies born to mothers who took antipsychotic drugs during their pregnancies. The pregnancies occurred between 2001 and 2015, with follow-up until 2019. Having a mother with a serious psychiatric disorder such as bipolar disorder or schizophrenia was associated with a significantly increased risk of ADHD and ASD in the children, but not with prematurity or being small for gestational age. The study provides useful clarity and reassurance for clinicians and their patients: women who are taking antipsychotics do not necessarily need to stop treatment when they become pregnant. A linked editorial emphasises that psychiatric illness in pregnancy, especially if untreated, is a high risk factor for both mother and child, who need ongoing medical and psychosocial support both before and well beyond pregnancy and delivery.

JAMA Intern Med doi:10.1001/jamainternmed.2021.4571

JAMA Intern Med doi:10.1001/jamainternmed.2021.4562

Pfizer jab: reassurance about safety

This Israeli study provides useful reassurance in the face of vaccine sceptics and doubters. Around 885 000 people vaccinated with the BNT162b2 mRNA vaccine were compared with a similar number of unvaccinated matched people. Lymphadenopathy was more common in the vaccinated group, and there were more cases of myocarditis, appendicitis, and shingles. A different cohort of people who developed covid-19 itself had a much higher incidence of myocarditis compared with matched uninfected people. They also had higher incidence of other adverse events, such as acute kidney injury, pulmonary embolism, deep vein thrombosis, myocardial infarction, pericarditis, and intracranial haemorrhage. The study was, in effect, two studies in one: a direct comparison of risks of vaccine versus covid infection wasn’t possible as the two cohorts were different, although the order of magnitude of adverse events is probably valid. Vaccination has been well shown in other studies to substantially reduce the risk of hospital admission, severe disease, and death from covid-19, and this study usefully examined the risks of non-fatal but serious complications of vaccination. Nothing that works is ever entirely risk-free, but this offers reassurance about the low level of risk associated with the Pfizer vaccine and contextualises vaccine risks such as myocarditis by showing how much more common it is in those who actually get covid. Whether die-hard antivaxxers will be swayed is another matter.

N Engl J Med doi:10.1056/NEJMoa2110475

Covid survivors in Wuhan are mostly fine after a year

This large ambidirectional (with both retrospective and prospective phases) cohort study of covid-19 survivors who were hospitalised in Wuhan over five months in 2020 provides some good news. Within a year of follow-up, most (88%) employees had returned to work and serious health impairment was rare. Self reported health improved between six and 12 months after discharge. At six months, 68% of patients reported one or more symptoms. This reduced to 49% at 12 months. Fatigue or muscle weakness was reported by 52% of patients at six months, and 20% at 12 months. Interestingly, anxiety and depression scores were slightly higher at 12 than six months. Predictably, those who had needed critical care had a high rate of persistent lung diffusion impairment (54%) and imaging abnormalities. The authors say, optimistically, that it will take some people longer to regain their baseline health status. The rates of return to work and self reported return to baseline status may not be generalisable to all populations.

Lancet doi:10.1016/S0140-6736(21)01755-4

Ann Robinson is an NHS GP and health writer and broadcaster.