HSR&D Home » Research » RRP 09-407 – HSR&D Study
Longer-Term Outcomes of Participation in Hypertension Self-Management
George Lee Jackson, PhD MHA
Durham VA Medical Center, Durham, NC
Funding Period: June 2010 - May 2011
A series of clinical trials have demonstrated that tailored self-management support/disease management programs can significantly improve diabetes and blood pressure (BP) control by the end of the study period. However, few studies have examined whether the health benefits of trial participation persist in the longer-term after trial conclusion. The purpose of this study was to examine the longer-term health and economic outcomes of trial participants to determine whether outcomes observed years after trial conclusion were similar to outcomes at study end. The first trial, the Diabetes Group Visits Trial [VA HSR&D IIR 03-084], found that patients with co-morbid hyperglycemia and hypertension who received group medical visits for 12 months had a reduction in systolic BP (top blood pressure number) of 7.3mmHg compared to usual care controls. The second trial, the Veteran Study To Improve The Control of Hypertension (V-STITCH) [VA HSR&D IIR 20-034], found that patients with hypertension who received bi-monthly nurse-delivered behavioral telephone calls for 24 months were more likely to have controlled hypertension at trial end than usual care patients (10.5% more intervention patients had BP control than control patients). The third trial, Hypertension Intervention Nurse Telemedicine Study (HINTS) [VA HSR&D IIR 04-426], randomly allocated patients to one of four 18-month intervention arms (usual care, tailored nurse-administered behavioral adherence intervention, medication management, and a combined behavioral adherence and medication management intervention). Compared to usual care, both the behavioral and medication management interventions showed significant improvement in blood pressure control compared to usual care at one year (approximately 13% greater for both groups than usual care), but differences were not statistically significant at 18 months (end of the trial).
In Aim 1, we assessed whether systolic BP among intervention patients after the trial conclusion persist at levels observed at the study end, and whether systolic BP remains better than that of control group patients. In Aim 2, we assessed whether healthcare utilization and expenditures among intervention patients after the trial conclusion persist at levels observed at the study end, and whether healthcare utilization and expenditures differ from control group patients after the trial conclusion.
We analyzed systolic BP, Veterans Affairs (VA) utilization, and VA expenditures among participants in three completed HSR&D sponsored clinical trials. First, we conducted descriptive analyses for each study over the entire longer-term follow-up period of 18 months. Second, we conducted study specific analyses of each outcome to understand the longer-term trends in systolic BP, utilization and expenditures for each study cohort. Information on patient cost and utilization was obtained from the centralized VA Decision Support System that captures data from across the VA healthcare system. For all three trials, we examined outpatient (primary care, specialty care, total outpatient), overall costs, and probability of having an inpatient admission in the pre-intervention, intervention, and post-intervention periods via generalized linear models (GLM) fit with generalized estimating equations. We also used different modeling strategies (negative binomial or Poisson models depending on the outcome) to examine the number of primary care, specialty care, and total overall outpatient visits during the same time periods. We controlled for treatment arm, indicators for each half-year time period, and their interactions in separate analyses of each trial. The potential of longer-term clinical benefits was examined using blood pressures measured at VA outpatient visits. As a result, each patient had a varying number of these outcome assessments at varying intervals of time. We used descriptive statistics by six-month intervals and spline smoother lines to understand average trends over time. We then used linear mixed models to estimate trajectories and test differences for intervention patients as compared to control patients.
Economic Results: For the Diabetes Group Visits Trial, overall expenditures were significantly lower (approximately $7,500) for intervention patients compared to control patients 12-18 months after the trial ended. These differences were likely driven by a lower probability of inpatient admissions (12% lower for intervention patients), which appears to have offset higher primary care costs (approximately $140) during the period. In the less resource intensive V-STITCH trial, utilization and costs were similar between the two groups during and after the study, with the exception of higher total overall costs for intervention patients 12-18 months after the study (approximately $3,100 higher for intervention patients). This difference was not noted when those patients who died during the time periods were excluded from the analysis. Longer-term economic differences were not noted between intervention groups in the HINTS Trial.
Clinical Results: Based on clinic blood pressure readings, as opposed to ones collected as part of study data-collection visits, there was not a statistically significant difference in systolic BP readings between group visit and usual care patients. However, the descriptive statistics showed systolic BP being approximately 2-5 mmHg lower for group visit patients compared to usual care patients over the ensuing 18 months post-intervention. There was no difference in clinic systolic BP measurements for the V-STITCH study. For the HINTS Trial, patients in the combined behavioral and medication arm had increasing lower systolic BP when compared to usual care patients during the period 18-months after trial completion, to the point that the difference was greater than 5 mmHg more than a year after completion.
Hypertension is the most widely recognized modifiable risk factor for stroke, myocardial infarction, peripheral vascular disease, heart failure, and end-stage renal disease. In the VA healthcare system, hypertension is the most common chronic condition with a prevalence of 37% among veterans. VA HSR&D has developed and tested a number of innovative programs aimed at supporting patient self-management of hypertension and other chronic illnesses. However, there has generally been a lack of information as to the longer-term impacts of these programs after the completion of clinical trials. Results of this study indicate longer-term potential economic benefits of group visits and longer-term clinical benefits of combining behavioral telephone calls with medication management. Further, this project demonstrates the utility of taking advantage of electronic health record information to examine both clinical and economic outcomes of patients who have completed trial interventions. Such analyses can help answer questions such as: 1) can we expect patient benefits to be maintained over time?; 2) what will be the impact on healthcare utilization after the program is over?; and 3) will increasing resources be needed over time because patients will need to undergo additional care?
External Links for this Project
Dimensions for VADimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.
If you have VA-Intranet access, click here for more information vaww.hsrd.research.va.gov/dimensions/
VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address. Search Dimensions for this project
DRA: Cardiovascular Disease
DRE: Treatment - Observational, Prevention, Prognosis, Treatment - Comparative Effectiveness
Keywords: Hypertension, Primary care, Self-care
MeSH Terms: none