Cyclical and seasonal variations in the incidence of type 1 diabetes mellitus between 1985 and 2016 in Bauru, São Paulo, Brazil

TL;DR

The study found cyclical and seasonal patterns in type 1 diabetes incidence in Brazil, with higher rates among girls and during colder months.

Contribution

The study identifies a 7.5-year cyclical pattern and seasonal variation in T1DM incidence, particularly in girls.

Findings

A significant 18% cyclical variation in T1DM incidence every 7.5 years was observed.

Girls showed a 22.9% variation every 5 years, with seasonal peaks during colder months.

Higher incidence rates coincided with H1N1 and dengue outbreaks and lower temperatures.

Abstract

To evaluate cyclical and seasonal variation in the incidence of type 1
diabetes mellitus (T1DM) from 1985 to 2016 in Bauru, São Paulo,
Brazil.

This was a retrospective longitudinal study. Clinical data were collected for
individuals known to have T1DM, who aged from 0-14 years, residing in Bauru,
São Paulo State, and followed at a local endocrinology clinic from
1985 to 2016. Incidence rates were calculated annually and grouped into
quadrennial intervals. Trends were analyzed using Joinpoint Regression to
estimate annual percentage changes. Poisson regression models assessed
cyclical and seasonal patterns over various periods (3- to 7.5-year cycles).
Seasonal variation was evaluated using the Akaike Information Criterion and
chi-square likelihood ratios to assess model fit.

Among the 298 included patients, the mean annual incidence was 12.1 per
100,000 person-years (95% CI: 10.7-13.4), with an average annual increase of
2.77% (95% CI: 1.3-4.3%). A significant cyclical variation of 18% every 7.5
years was observed, with girls exhibiting a 22.9% variation every 5 years.
No cyclical pattern was identified for boys. Seasonal analysis revealed
higher amplitudes among girls (±26.4%) and in the 5-9.99-year age
group (±26.2%), predominantly during colder months.

T1DM cyclical variations with a 7.5-year cycle were observed, with girls
showing a pronounced variation and a distinct 5-year cycle. Seasonal
variations were found among girls, particularly in the 5-9.99-year age
group. Outbreaks of H1N1 and dengue, along with the lowest temperatures,
coincided with higher incidence rates, aligning with the 7.5-year cycles.
Targeted health policies are needed to mitigate the impact of these factors,
supporting surveillance, early diagnosis, and preventive strategies for
T1DM.

Figures