Design: Population-based matched cohort study using linked primary care, hospital and mortality data
Population: 991 patients with generalised pustular psoriasis among 25.8 million people in England, 2008 to 2022
Primary outcome: All-cause and cause-specific mortality, and years of life lost
Effect: All-cause mortality adjusted HR 3.19 (95% CI 2.58 to 3.91); sepsis 9.76 (4.92 to 19.35)

Linked primary care, hospital and mortality records covering 25.8 million people in England identified 991 patients with generalised pustular psoriasis between 2008 and 2022. Prevalence rose from 20.9 to 32.5 per million over that period and incidence began climbing after 2016, reaching 4.9 per million person-years.

Against matched comparators, all-cause mortality was more than three times higher. The cause-specific pattern is the part to remember: excess deaths from neoplasms, respiratory, digestive and circulatory disease, and a near-tenfold excess from sepsis. A skin disease whose barrier fails over the whole body surface kills the way a burn does.

Two demographic findings matter for Indian and diaspora practice. Sixty-three per cent of patients were female, and onset came earlier in Asian patients than in White patients, against a mean onset age of 51 years overall. Younger age at diagnosis carried the greatest loss of life expectancy, so a young patient with a first flare is not the reassuring presentation.

Treat a generalised pustular flare as a systemic emergency: fluid balance, temperature, and cultures before assuming the fever is inflammatory.
Have a low threshold for antibiotics while awaiting cultures in an extensive flare.
Screen for and manage cardiovascular and metabolic risk in the interval periods, not only the skin.
Ask about IL36RN-related family history in younger and Asian patients presenting early.
Record which flares required admission; frequency of admission is the practical marker of severity here.

Why it matters

It puts a mortality figure on a diagnosis often managed as a dermatological emergency and then discharged as a skin problem.

The statistics, in plain English

Adjusted hazard ratios compare the rate of death over follow-up, matched up to 10 comparators per patient and adjusted for deprivation and recorded comorbidity. The sepsis estimate (9.76) has a very wide interval, 4.92 to 19.35, because deaths coded to sepsis in 991 patients are few — the direction is certain, the magnitude is not. Rising prevalence in a rare disease usually reflects better coding as much as more disease, so read the mortality figures rather than the trend.