Joinpoint regression analysis with time-on-study as time-scale. Application to three Italian population-based cohort studies

Abstract


 Background: Joinpoint regression analysis is usually applied to study varying trends over time in order to identify the time point(s) in which the trend significantly changes. We illustrate three epidemiological investigations in which this methodology was applied with time-on-study as time-scale.
 Methods: Data were retrieved from the healthcare utilization databases of Lombardy Region (Italy). We investigated the trend of the: (1) mortality rate among centenarians hospitalized for hip fracture (2004-2011); (2) proportion of persistent patients after the initial prescription of antihypertensive drugs during the first year of treatment according to gender (2005); prescription rate of statins in the year before and after the hospital admission among patients hospitalized for a transient ischemic attack (2008-2009).
 Results: The following results were obtained: (1) a joinpoint was identified in the fourth month, showing an increased risk of death during the three months after hip fracture hospitalization; (2) the proportion of patients still under antihypertensive treatment falls until the fifth month, remaining stable afterwards; there was evidence that the prevalence of patients who discontinued the treatment was significantly higher among women than men; (3) during the year after the transient ischemic attack episode, monthly rate of available statins was double than the previous year with a significant decrease in the first four months.
 Conclusions: The joinpoint regression analysis can be a useful tool in epidemiologic framework when a temporal trend is the objective of the investigation since it allows to make inference by means of a quantitative method rather than a qualitative evaluation.