Legionnaires’ Disease in Lazio: Age, Sex, Smoking and Rainfall Drive Risk

Legionnaires’ disease, a severe pneumonia caused by Legionella bacteria, has been climbing across much of Europe and North America over the past decade, yet the combined contribution of demographic, behavioural, geographic and meteorological factors has remained poorly understood. A new study from the Lazio Region of Italy, published in BMC Public Health, offers one of the most detailed regional analyses to date, drawing on nearly a decade of surveillance data to disentangle how individual characteristics, place of residence and weather patterns shape the incidence of this waterborne infection.

The research team, led by investigators at the National Institute for Infectious Diseases Lazzaro Spallanzani IRCCS in Rome, analysed 1,932 reported cases of Legionnaires’ disease in Lazio between 2015 and 2023. Individual data came from the regional Legionnaires’ disease surveillance database, which also captured information on smoking and alcohol consumption as well as comorbidities. Daily rainfall and temperature records were retrieved from the regional hydrometeorological network, allowing the researchers to link weather conditions to disease occurrence at a fine temporal and geographic resolution.

To quantify the associations, the team calculated Legionnaires’ disease incidence rates per 100,000 person-years and fitted a negative binomial multivariate regression model with eight predictors selected on the basis of the literature and exploratory data analysis. Rather than treating each case as an isolated observation, the researchers aggregated individual-level values of six categorical predictors into strata, cross-combining predictor levels at month level, and then linked rainfall and temperature averages for each stratum to the corresponding model stratum. The model offset incorporated regional prevalence estimates of smoking, alcohol use and comorbidities applied to population denominators, a technical choice designed to account for the fact that these risk behaviours and conditions are unevenly distributed across the population at risk.

The temporal trend was striking. Incidence rose steadily from 2017 to 2019, then fell during the COVID-19 pandemic, and climbed again in the post-pandemic period. This pattern suggests that the non-pharmaceutical interventions implemented during the pandemic, including lockdowns, travel restrictions and reduced use of shared water systems in hotels and workplaces, may have temporarily suppressed transmission opportunities, with cases rebounding once normal activity resumed.

In the final regression model, three factors emerged as strongly and significantly associated with disease incidence. People aged 70 and older faced an incidence rate ratio of 17.1 compared with younger adults, with a 95 percent confidence interval of 11.0 to 27.2, while those aged 50 to 69 had an incidence rate ratio of 4.68, with a confidence interval of 3.03 to 7.43. Male sex carried an incidence rate ratio of 5.22, with a confidence interval of 3.61 to 7.80, and reported comorbidity was associated with an incidence rate ratio of 4.61, with a confidence interval of 3.41 to 6.18. These effect sizes are large by epidemiological standards and confirm that age, sex and underlying health status remain the dominant determinants of who develops Legionnaires’ disease after exposure.

Beyond these individual factors, the study identified several environmental and behavioural associations. Smoking or alcohol consumption was linked to an incidence rate ratio of 2.05, with a confidence interval of 1.53 to 2.75. Domicile or residence in the Rome metropolitan area carried an incidence rate ratio of 2.03, with a confidence interval of 1.58 to 2.61, reflecting the dense urban infrastructure and complex water systems of the capital. Average monthly precipitation was also directly associated with incidence, with an incidence rate ratio of 1.15 per unit increase, with a confidence interval of 1.11 to 1.20, suggesting that rainfall may facilitate the aerosolisation and dispersal of Legionella from contaminated water sources.

The COVID-19 period itself was significantly inversely associated with incidence, with an incidence rate ratio of 0.45 and a confidence interval of 0.32 to 0.65, reinforcing the interpretation that pandemic restrictions temporarily reduced exposure. The model also revealed significant two-way interactions that add nuance to the main effects. The interaction between the COVID-19 period and monthly average minimum temperature, between the 50 to 69 age group and smoking or alcohol consumption, and between the Rome metropolitan area and monthly average rainfall all produced incidence rate ratios significantly greater than one, indicating that these combinations amplified risk beyond what either factor alone would predict.

Conversely, several interactions produced incidence rate ratios significantly below one. The combination of male sex with either the 50 to 69 or 70 and older age groups, the combination of older age with smoking or alcohol consumption, and the combination of older age with comorbidity all showed attenuated effects, a pattern that may reflect saturation, where the baseline risk in older individuals is already so high that additional risk factors contribute proportionally less. The interactions between the COVID-19 or post-COVID-19 periods and the Rome metropolitan area also showed incidence rate ratios below one, suggesting that the pandemic’s protective effect was partly concentrated in the capital’s dense urban environment.

The authors conclude that these findings underscore the multifactorial nature of Legionnaires’ disease occurrence and may help inform targeted surveillance and prevention strategies. By quantifying how age, sex, comorbidity, behavioural factors, urban residence and meteorological conditions combine to drive incidence, the study provides regional public health authorities with a more precise evidence base for identifying high-risk strata and timing interventions. As climate variability alters rainfall and temperature patterns across the Mediterranean, understanding the meteorological determinants of Legionella transmission becomes increasingly important for protecting vulnerable populations in Lazio and beyond.

Subject of Research: Legionnaires' disease epidemiology and public health

Article Title: Unraveling legionellosis dynamics in the Lazio Region: effects of individual, geographical and meteorological determinants

Article References: Spaziante, M., D’Amato, M., Leone, S., Bongiovanni, A., De Carli, G., Tonziello, G., Malatesta, G. N., Agresta, A., De Santis, C., Vantaggio, V., Nicastri, E., Palazzolo, C., Asta, F., Iacovacci, S., Siddu, A., Girardi, E., Nisii, C., D’Arezzo, S., Fontana, C., … Vairo, F. (2026). Unraveling legionellosis dynamics in the Lazio Region: effects of individual, geographical and meteorological determinants. BMC Public Health. https://doi.org/10.1186/s12889-026-29880-y

Image Credits: AI Generated

DOI: 10.1186/s12889-026-29880-y

Keywords: Legionnaires' disease, Legionella, epidemiology, public health, meteorological factors, Lazio Region, Unraveling, legionellosis, dynamics, Lazio, Region, effects