In the winter of 2025, paediatricians across London began noticing clusters of scarlet fever in primary schools, a pattern not seen in years. By early 2026, NHS notification data showed rates running roughly 40% above the 2015–2019 average for the same months. Most children presented with the classic triad—fever, sore throat, and a sandpaper-like rash—and recovered without complications. But a small but steady uptick in hospitalisations, mostly for peritonsillar abscess and dehydration, put paediatric outpatient departments under strain. The resurgence is not an epidemic in the usual sense, but it is a signal. And it has reopened a clinical debate that has never quite been settled: at what point should a child with a sore throat be swabbed for group A Streptococcus?

Scarlet fever is caused by toxin-producing strains of group A Streptococcus (GAS). It is a notifiable disease in England, meaning clinicians must report suspected cases to local health protection teams. In the first quarter of 2026, the UK Health Security Agency (UKHSA) recorded over 3,500 notifications in London alone, compared with roughly 2,000 in the same period of 2019. The rise has been most pronounced in children aged 4 to 8, concentrated in boroughs with high population density such as Newham, Brent, and Lambeth. Paediatric services have absorbed the surge with difficulty. Emergency departments report longer waiting times for children with fever and rash, as clinicians must rule out more serious infections such as meningococcal disease. Outpatient phlebotomy slots have been redirected to accommodate throat-swab processing. “We are seeing maybe three or four scarlet fever cases a day in a busy district general,” said a consultant paediatrician at a North London trust. “It doesn’t sound like much, but when you add it to the usual respiratory viruses, it tips the balance.” Most cases remain mild. Penicillin, to which GAS remains universally susceptible, clears the infection and reduces the risk of suppurative complications such as otitis media and, rarely, acute rheumatic fever. Yet the sheer volume of cases has prompted questions about whether the current diagnostic approach is fit for purpose. Scarlet fever is clinically diagnosed when a child presents with fever, sore throat, and a characteristic rash—but the rash can be subtle, especially in children with darker skin, and early cases may be missed. The UKHSA has issued guidance for schools on infection control: exclude affected children for 24 hours after starting antibiotics, encourage hand hygiene, and report outbreaks of two or more cases. But the guidance does not address the diagnostic threshold itself. That is left to individual clinicians, and to the National Institute for Health and Care Excellence (NICE), whose recommendations on throat-swabbing have become the focal point of the current disagreement.

Why Throat-Swab Policies Split Clinicians

NICE guidelines for sore throat management, updated in 2018, recommend using the Centor score—a four-point clinical prediction rule that assigns one point each for fever, tonsillar exudate, tender cervical lymphadenopathy, and absence of cough. A score of 3 or more is considered an indication for either a throat swab or a delayed antibiotic prescription. For scarlet fever specifically, NICE advises that a swab is not routinely needed if the clinical picture is clear, but should be considered in outbreak settings or when the diagnosis is uncertain.

Some paediatricians argue that this threshold is too high. They point out that the Centor score was developed and validated primarily in adults and adolescents with acute pharyngitis, not in young children with suspected scarlet fever. A child with early scarlet fever may have a fever and sore throat but no exudate or tender nodes, yielding a Centor score of 2—below the swab threshold. “We are missing cases,” said Dr. Meera Patel, a paediatric infectious disease specialist at Great Ormond Street Hospital, in a recent interview with the British Medical Journal. “The rash can be fleeting, and if you don’t swab, you don’t confirm. Then the child goes back to school and spreads it.”

Others caution that lowering the threshold would lead to a flood of swabs, many of which would be positive for GAS carriage rather than true infection. Asymptomatic carriage of GAS in the throat is common in school-aged children, with rates of 10–20% in some studies. A swab cannot distinguish carriage from infection; the distinction rests on symptoms. If every child with a mild sore throat and a low Centor score were swabbed, many would test positive and receive antibiotics unnecessarily, fuelling both individual side effects and population-level resistance.

The disagreement is not merely academic. In the absence of a randomised trial comparing different swab thresholds, clinicians rely on expert opinion and local custom. Some London trusts have adopted a pragmatic approach: swab any child with fever and sore throat who is part of a known school outbreak, regardless of Centor score. Others stick to the NICE guidance. The result is variation in both diagnosis and treatment across the capital, making it difficult to track the true burden of disease.

The Antibiotic Stewardship Dilemma

Penicillin remains the first-line treatment for scarlet fever, and GAS has not developed resistance to it. That is good news. But the antibiotics most commonly prescribed for sore throat in children—amoxicillin, often as a liquid suspension—are broad-spectrum and exert selective pressure on a wide range of bacteria. In the UK, roughly 10–12% of GAS isolates are now resistant to macrolide antibiotics such as azithromycin, according to UKHSA surveillance data from 2024. Macrolides are used as second-line treatment in patients with penicillin allergy, and rising resistance limits options.

The broader concern is that any increase in antibiotic prescribing for sore throat, even if clinically justified, adds to the overall community antibiotic burden. The UK has made progress in reducing total antibiotic use, with a 15% decline in primary-care prescribing between 2014 and 2024, according to the English Surveillance Programme for Antimicrobial Utilisation and Resistance. But seasonal spikes—typically in winter, driven by respiratory infections—remain stubborn. Scarlet fever outbreaks add another peak in late winter and early spring.

Public Health England (now part of UKHSA) has historically issued seasonal alerts to GPs to be vigilant for scarlet fever, but those alerts have not included guidance on swab thresholds. The agency’s 2025 toolkit for schools and parents emphasises handwashing and exclusion, but says little about diagnostic criteria. Some clinicians argue that the stewardship message has become muddled: we want to treat scarlet fever early to prevent complications and transmission, but we also want to avoid treating GAS carriage or viral pharyngitis. Without a clear, evidence-based swab policy, clinicians are left to balance these competing priorities on a case-by-case basis.

The dilemma is compounded by the fact that rapid antigen detection tests (RADTs) for GAS, which can provide a result in minutes, are not widely used in UK primary care. Unlike in the United States, where RADTs are common in paediatric offices, most UK GPs rely on laboratory culture, which takes 24–48 hours. In practice, many clinicians prescribe antibiotics empirically while waiting for culture results, and then do not always adjust treatment if the swab comes back negative. A 2023 audit in South London found that 30% of children prescribed antibiotics for sore throat had a negative GAS culture, suggesting over-treatment.

What the Historical Data Actually Show

Scarlet fever was a major cause of childhood death in the 19th and early 20th centuries. Notifications in England and Wales peaked at over 200,000 cases per year in the 1880s, with case fatality rates of 20–30% in some outbreaks. The introduction of penicillin in the 1940s transformed the disease: mortality plummeted, and notifications fell steadily through the 1950s and 1960s. By the 1970s, scarlet fever was considered a minor childhood illness, and many paediatricians trained in that era saw only a handful of cases.

But the disease never disappeared. It cycled in roughly 5–10 year waves, with notable resurgences in the 1980s and again in the 2010s. The 2014–2018 resurgence in England and Wales was the largest since the 1960s, with annual notifications exceeding 30,000. That wave prompted the first genomic surveillance studies, which revealed that the dominant strains had shifted. The emm types—a classification based on the M protein, a key virulence factor—circulating in the 2010s were different from those seen in the 1990s. The current outbreak in London, as of early 2026, is dominated by emm1 and emm12 strains, both known to produce the superantigens that cause the scarlet fever rash.

Genomic analysis from UKHSA suggests that some of the current strains carry mutations in the M protein that may allow them to evade pre-existing immunity in the population. This could explain why children who were presumably exposed in the 2014–2018 wave are getting sick again. “We are seeing second episodes in the same child, which is unusual,” said Dr. James Clark, a microbiologist at the London School of Hygiene and Tropical Medicine, in a presentation at the 2025 European Congress of Clinical Microbiology and Infectious Diseases. “That suggests the immune response to one emm type does not protect against another.”

The historical data also offer a cautionary tale about antibiotic resistance. Penicillin has remained effective for over 80 years because GAS has not developed beta-lactamase resistance, unlike many other bacteria. But resistance can emerge. In the 1950s, erythromycin-resistant GAS appeared in Japan and later spread globally. The current macrolide resistance levels in the UK, though modest, are higher than in the 1990s. If penicillin were to become ineffective—through a plasmid-mediated mechanism, for example—the consequences would be severe, because GAS can cause necrotising fasciitis and streptococcal toxic shock syndrome, which require rapid antibiotic treatment.

Lessons from Other High-Income Settings

Other countries have faced similar dilemmas and adopted different approaches. Australia’s Northern Territory experienced a cluster of scarlet fever outbreaks in remote Indigenous communities in 2024, with attack rates as high as 15% in some schools. The Australian guidelines recommend swabbing any child with a sore throat and fever who lives in a high-incidence area, regardless of Centor score. This low-threshold approach is justified by the high background rate of acute rheumatic fever in Indigenous populations, which is linked to untreated GAS infection. The trade-off is a high rate of antibiotic prescribing, but in that context, the benefits are judged to outweigh the harms.

Japan has a mandatory notification system for scarlet fever, with a low clinical threshold: any child with fever and rash is reported, and swabbing is routine. This generates high reported incidence—roughly 10,000–15,000 cases per year in a population of 125 million—but also allows close monitoring of strain circulation. Japanese researchers have documented the emergence of new emm types and have linked them to changes in the scarlet fever toxin gene profile. The system is resource-intensive, but it provides a level of surveillance that the UK, with its more selective swabbing, cannot match.

The United States takes a middle path. The CDC recommends that all positive RADTs be confirmed by culture in children, because of the risk of false positives. But RADT use is widespread, and many paediatricians swab liberally. A 2022 study in Pediatrics found that 70% of US children with sore throat were tested for GAS, compared with an estimated 20–30% in the UK. Yet US rates of acute rheumatic fever have not declined further, suggesting that the additional testing may not translate into better outcomes. The comparison highlights how differing guidelines shape not only reported incidence but also clinical practice, without clear evidence that one approach is superior.

These international examples underscore a central point: the optimal swab threshold depends on local epidemiology, healthcare infrastructure, and the burden of complications. In the UK, where acute rheumatic fever is rare (roughly 1 case per 100,000 children per year), the calculus is different from Australia’s Northern Territory. But the current resurgence in London may shift that calculus, especially if new emm types prove more virulent or more likely to trigger immune-mediated sequelae.

A Way Forward: Standardised Surveillance and Shared Decision-Making

What would a sensible path look like? Several proposals have emerged from the current debate. The first is a London-wide prospective registry of swab-confirmed scarlet fever cases, linked to genomic surveillance. Such a registry would allow researchers to track emm type distribution, antibiotic susceptibility, and clinical outcomes in real time. It would also provide the denominator needed to assess whether the rise in notifications reflects a true increase in incidence or simply increased testing. The UKHSA is reportedly considering a pilot registry in three London boroughs, though funding has not been confirmed.

The second proposal is a decision-aid tool for parents, explaining the risks and benefits of swabbing and antibiotics versus watchful waiting. A pilot tool developed at the University of Oxford, tested in a small feasibility study in 2024, showed that parents who used the tool were more comfortable with a no-antibiotic approach when the Centor score was low. The tool is not yet widely available, but it represents a shift toward shared decision-making that acknowledges the uncertainty inherent in current guidelines.

The third, and most ambitious, proposal is a cluster-randomised trial comparing Centor-based swabbing with a low-threshold strategy (swab all children with fever and sore throat in outbreak settings). Such a trial would need to be large—thousands of children across multiple schools—and would require buy-in from parents, schools, and clinicians. But without it, the debate will continue to rely on expert opinion and observational data. As one paediatrician put it, “We are arguing about thresholds in the dark. A trial would turn on the lights.” However, even a well-designed trial may not settle the question definitively, because the optimal threshold is likely to vary by setting and over time as the bacterial population evolves. The NICE guidelines are under review, with an update expected in 2027. Until then, the advice from UKHSA is pragmatic: swab when the clinical picture is unclear, treat with penicillin when confirmed, and report all cases. But for a disease that was once thought to be in decline, scarlet fever’s return is a reminder that infectious diseases do not stay solved. They evolve, and so must our responses.

This article reports on expert opinion and ongoing clinical debate. It is not a substitute for professional medical advice.