Smoothing 3D Drug Overdose Mortality Rates and Displaying Patterns with SAS/JMP Software

• Activity Number: 445 - GOVT CSpeed 2
• Type: Contributed
• Date/Time: Thursday, August 12, 2021 : 4:00 PM to 5:50 PM
• Sponsor: Government Statistics Section
• Author(s): Nancy Han* and Rong Wei
• Companies: National Center for Health Statistics, CDC and National Center for Health Statistics, CDC
• Keywords: drug overdose death; JMP three-dimensional plot; mortality changes in age and calendar year; two-dimensional data smoothing

Abstract

The visual display of data is a critical component for demonstrating changes in population health. One area of current interest is examining trends in drug overdose death rates by age and calendar year. The traditional method uses 2-dimensional line graphs with age defined as a categorical variable. Mortality statistics are often presented in pre-determined age categories of 5 or 10 years, and any change that may occur within these categories is masked. If more granular measures of age and time intervals are plotted to show changes in a continuous manner, we may obtain more precise and unbiased information from the same data. In this study, drug overdose mortality data based on the National Vital Statistics System from 1999-2019 are used as an example to show temporal changes by age in a continuous manner with a SAS/JMP 3-dimensional data plot. The plot provides information on the peak changes in death rates within the dimensions of age and year. Smoothing death rates at both the age and year dimensions are carried out with the SAS/TPSPLINE procedures. This smoothing is assessed by checking predicted values, residuals, quantile plots, and the fitting of statistical parameters.