Comparing Results Between Survey Analysis Software and NCI’s Joinpoint Software for Trend Analyses of Survey Data

• Activity Number: 476 - Innovations in Analytic Approaches for Survey Data
• Type: Contributed
• Date/Time: Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
• Sponsor: Survey Research Methods Section
• Author(s): Jennifer Rammon* and Kate Hubbard and Deanna Kruszon-Moran and Robin Cohen and Margaret Carroll
• Companies: National Center for Health Statistics/CDC and National Center for Health Statistics/CDC and National Center for Health Statistics and National Center for Health Statistics/CDC and National Center for Health Statistics/CDC
• Keywords: trend analyses; survey data; NHIS; NHANES; Joinpoint; piece-wise regression

Abstract

The National Center for Health Statistics Guidelines for Analysis of Trends recommends that analysts use record-level data and survey analysis software (such as SUDAAN, the R survey package, STATA, or SAS-survey) to fit desired trend models of survey data. A partial exception to using record-level survey data is suggested when changes in trend will be assessed using joinpoint regression models fit with NCI’s Joinpoint software. In this case, the guidelines recommend that analysts use NCI’s Joinpoint software with aggregated data to identify the number and location of joinpoints, but that they use survey analysis software with record-level data to obtain final slope estimates and to conduct tests of hypothesis for the model identified by the Joinpoint software. In practice, there are times when the results produced by NCI’s Joinpoint software are incongruent with the results produced by survey analysis software. We identify cases where these inconsistencies could occur, discuss plausible explanations for the discrepancies, and provide examples from the National Health and Nutrition Examination Survey (NHANES) and the National Health Interview Survey (NHIS).