Assessing seasonality and the role of its potential drivers in environmental epidemiology: a tutorial

Abstract

Several methods have been used to assess the seasonality of health outcomes in epidemiological studies. However, little information is available on the methods to study the changes in seasonality before and after adjusting for environmental or other known seasonally varying factors. Such investigations will help us understand the role of these factors in seasonal variation in health outcomes and further identify currently unknown or unmeasured risk factors. This tutorial illustrates a statistical procedure for examining the seasonality of health outcomes and their changes, after adjusting for potential environmental drivers by assessing and comparing shape, timings and size. We recommend a three-step procedure, each carried out and compared before and after adjustment: (i) inspecting the fitted seasonal curve to determine the broad shape of seasonality; (ii) identifying the peak and trough of seasonality to determine the timings of seasonality; and (iii) estimating the peak-to-trough ratio and attributable fraction to measure the size of seasonality. Reporting changes in these features on adjusting for potential drivers allows readers to understand their role in seasonality and the nature of any residual seasonal pattern. Furthermore, the proposed approach can be extended to other health outcomes and environmental drivers.

Should We Adjust for Season in Time-Series Studies of the Short-Term Association Between Temperature and Mortality?

“…We assessed seasonality in mortality in Detroit and Miami before and after the adjustment of temperature over lag days 0–21 described above, while the seasonality of mortality was assessed using a cyclic spline function with 4 df . 15 Here, we observed a reduction though not elimination in seasonality after the adjustment (Figure 4), suggesting that temperature can be a partial mediator (Figure 3B) or a confounder (Figure 3A) for the season-mortality association. It also suggests that Figure 3C is not plausible since the seasonality in mortality did not eliminate after temperature adjustment.…”

Heat-related mortality and ambulance transport after a power outage in the Tokyo metropolitan area

“…A natural cubic spline with three equally spaced knots for the day of the year (falsenormalDnormalOnormalYfalse), consecutive numbers representing each day from July to September in 2019) was used to model seasonality following the standard model specification in the epidemiological literature. 36,37 Using this approach, we estimated the RRs for heat (95th vs. 50th percentile of daily average temperature) at different ER values (0%, 10%, and 20%). The ratios of RRs (RRRs) were computed to compare the risks associated with various ER values to the risks associated with no ER in the absence of a power outage.…”

Incidence and seasonality of Kawasaki disease in children in the Philippines, and its association with ambient air temperature

“…We used a pair of sine and cosine functions to assess seasonality of KD in annual cycle based on monthly data ( 27 ) using Equation 1 : where was the monthly KD cases on time t following a quasi-Poisson distribution; was the intercept; was the radians transformed from , where , since we were estimating the annual cycle using monthly data; c and s were the coefficients derived from , respectively; was a linear function controlling for the long-term trend; was the error term. We summarized the seasonality of KD from its shape, timings (i.e., peak and trough), and size: (a) we used sine and cosine coefficients to derive predicted seasonal shape, (b) the month with maximum and minimum estimates of KD cases were identified as peak and trough, respectively, and (c) the difference in the maximum estimates of KD cases and the minimum estimates of KD cases was calculated to measure the amplitude (both in relative and absolute scale) ( 29 ).…”