Time-series studies in the field of air pollution aim to test and quantify potential short-term relationships between daily time series of ambient air pollution levels and daily time series of health indicators. This type of study has sometimes been subject to misunderstandings regarding the methodology, particularly concerning the lack of need to account for individual factors and personal exposure to indoor pollution. Adjusting for these individual confounding factors, as is typically done in "classical" epidemiological studies (case-control studies, cohort studies), is inappropriate in time-series studies based on aggregated data. The same does not apply to third-party factors that may vary over time with air pollution levels (weather conditions, viral epidemics, changes in healthcare structures, etc.), which must be accounted for during analysis either indirectly, through time modeling, or directly, through nonlinear modeling. Over the past decade, numerous studies using time-series analysis have been published. They highlight the existence of short-term associations between commonly observed daily levels of air pollution and respiratory mortality. The consistency of the many results published in the international literature supports the conclusion that the short-term relationships between air pollution and respiratory mortality are unbiased.