An Introduction to Time-Trend Analysis

Abstract

Healthcare professionals often are presented with data that appear to indicate an upward or downward trend over time. For example, admissions of acquired immunodeficiency syndrome (AIDS) patients appear to be increasing, cesarean section rates appear to be decreasing, or nosocomial pneumonia rates appear to be increasing. Critical decisions sometimes are based on such trends, which often are presented without a statistical analysis. Those responsible for decision making may be left wondering whether these apparent trends represent only chance variation. Graphs showing trends over time generally present one of three kinds of outcome data: counts (eg, three AIDS admissions), proportions (eg, 10 cesarean sections per 100 total deliveries), or person-time data (eg, 13 cases of nosocomial pneumonia per 10,000 patient days). Using familiar examples and a minimum of technical language, we illustrate the analysis of time trends.