Drug utilization researchers are often interested in evaluating prescribing and medication use patterns and trends over a specified period of time. Joinpoint regression is a useful methodology to identify any deviations in secular trends without a preconceived notion of where these break points might occur. This article provides a tutorial on the use of joinpoint regression, within Joinpoint software, for the analysis of drug utilization data.
The statistical considerations for whether a joinpoint regression analytical technique is a suitable approach are discussed. Then, we offer a tutorial as an introduction on conducting joinpoint regression (within Joinpoint software) through a step-by-step application, which is a case study developed using opioid prescribing data from the United States. Data were obtained from public files available through the Centers for Disease Control and Prevention from 2006 to 2018. The tutorial provides parameters and sample data needed to replicate the case study and it concludes with general considerations for the reporting of results using joinpoint regression in drug utilization research.
The case study evaluated the trend of opioid prescribing in the United States from 2006 to 2018, where time points of significant variation (one in 2012 and another in 2016) are detected and interpreted.
Joinpoint regression is a helpful methodology for drug utilization for the purposes of conducting descriptive analyses. This tool also assists with corroborating assumptions and identifying parameters for fitting other models such as interrupted time series. The technique and accompanying software are user-friendly; however, researchers interested in using joinpoint regression should exercise caution and follow best practices for correct measurement of drug utilization.

OBJECTIVE: The utilization of healthcare services and costs among patients with cancer is often estimated by the phase of care: initial, interim, or terminal. Although their durations are often set arbitrarily, we sought to establish data-driven phases of care using joinpoint regression in an advanced melanoma population as a case example.
METHODS: A retrospective claims database study was conducted to assess the costs of advanced melanoma from distant metastasis diagnosis to death during January 2010-September 2014. Joinpoint regression analysis was applied to identify the best-fitting points, where statistically significant changes in the trend of average monthly costs occurred. To identify the initial phase, average monthly costs were modeled from metastasis diagnosis to death; and were modeled backward from death to metastasis diagnosis for the terminal phase. Points of monthly cost trend inflection denoted ending and starting points. The months between represented the interim phase.
RESULTS: A total of 1,671 patients with advanced melanoma who died met the eligibility criteria. Initial phase was identified as the 5-month period starting with diagnosis of metastasis, after which there was a sharp, significant decline in monthly cost trend (monthly percent change [MPC] = -13.0%; 95% CI = -16.9% to -8.8%). Terminal phase was defined as the 5-month period before death (MPC = -14.0%; 95% CI = -17.6% to -10.2%).
CONCLUSIONS: The claims-based algorithm may under-estimate patients due to misclassifications, and may over-estimate terminal phase costs because hospital and emergency visits were used as a death proxy. Also, recently approved therapies were not included, which may under-estimate advanced melanoma costs.
CONCLUSIONS: In this advanced melanoma population, optimal duration of the initial and terminal phases of care was 5 months immediately after diagnosis of metastasis and before death, respectively. Joinpoint regression can be used to provide data-supported phase of cancer care durations, but should be combined with clinical judgement.