Statistical significance

When trends are analyzed in Health, United States, terms such as “stable,” “no clear trend,” and “did not change significantly” indicate that the slope of the trend line was not significantly different from zero. Terms such as “increase” and “decrease” indicate that a significant trend was found.

In addition to trend testing, comparisons are often made between estimates at two time points or for two populations. For data sources with standard errors, the difference between two estimates was assessed for statistical significance using z tests at the 0.05 alpha level. For data sources with no standard errors, relative differences were assessed using the methods recommended by the data source. The terms “higher” or “highest” and “lower” or “lowest” indicate that the estimates were significantly different, while the terms “similar,” “no difference,” and “no change” indicate that the estimates were not significantly different. Lack of comment does not necessarily indicate that trends or differences were tested and found to be not significant. Data sets include point estimates and standard errors, when available, for data users who would like to perform additional statistical tests.

Statistical significance of differences or trends is partly a function of sample size (that is, the larger the sample, the smaller the change that can be detected), and statistical significance does not always indicate public health significance. Moreover, a small sample size may result in statistically nonsignificant results despite the existence of potentially meaningful differences. For more information, see Wasserstein RL, Lazar NA. The ASA’s statement on p-values: Context, process, and purpose. Am Stat 70(2):129–33. 2016. (Also see Sources and Definitions, Joinpoint trend analysis software; Statistical testing.)