HSR&D Home » Research » IIR 08-028 – HSR&D Study
Complex Antithrombolic Therapy in Older Veterans: GI Risk and Preference
Neena Susan Abraham, MD MSc
Michael E. DeBakey VA Medical Center, Houston, TX
Funding Period: August 2009 - July 2012
Anticoagulants and antiplatelets are commonly prescribed in combination to older veterans for secondary cardioprophylaxis (i.e. complex antithrombotic therapy, CAT). These agents are independently associated with clinically significant upper gastrointestinal events (UGIE), such as bleeding and perforation from ulcers of the stomach and duodenum, the magnitude of which remains unknown. Furthermore, little is known regarding preferences of elderly veterans for the risk-benefit of CAT strategies. To provide a basis for individualization of CAT strategies, we quantified national CAT-related bleeding events (i.e., UGIE, lower GI events [LGIE], transfusions and hospitalizations) and assessed preferences of elderly cardiac veterans when patient preference for risk aversion, cardiovascular benefit and medication benefit was explicitly considered. We then assessed how preference-elicitation influenced patient CAT adherence.
1- To quantify CAT-related UGIE, LGIE, transfusions, and hospitalizations among a large cohort of elderly cardiac veterans.
2- To elicit preferences for CAT strategies among older cardiovascular patients.
3- To explore how preference elicitation influenced 1-year medication adherence.
To quantify CAT-related bleeding events (i.e., UGIE, LGIE, transfusions and hospitalizations), we conducted a retrospective cohort study among elderly veterans prescribed a CAT strategy, using medical and pharmacy data from a merged VA-Medicare dataset from 10/01/02-09/30/08. A time-dependent analysis assessed the risks associated with CAT strategies.
Adaptive conjoint analysis (ACA) was used to define patient preferences for CAT strategies considering trade-offs for UGIE (as quantified in the first specific aim), cardiovascular benefits and medication strategy attributes (i.e., complexity of administration, need for follow-up, etc). Elderly multi-morbid cardiovascular patients from two VAMC sites (Houston and Connecticut) completed the interactive ACA computer survey. Previously obtained qualitative data regarding patient perceptions of CAT risks-benefits, burden of self-management and risk communication was used to create the ACA survey. VA pharmacy data, was used to assess each patient's adherence to their CAT strategy prior to and 1 year following preference elicitation. The post-ACA prescription strategy was categorized as concordant or discordant with the patient's elicited preferences and adherence assessed based on these strata.
Specific Aim #1: CAT-related bleeding events are a significant and clinically relevant risk. The incidence rates for UGIE, transfusions, and hospitalizations were greatest with triple therapy (TRIP; i.e. anticoagulant, antiplatelet and aspirin) at 27.6/1000 PY (95% CI: 24.8-30.6), 73.9/1000 PY (95% CI: 69.1-79.0), 118.3/1000 PY (95% CI: 112.1-124.8) and 800.8/1000 PY (95% CI: 773.3-829.1). Adjusted models revealed a 40-60% increased UGIE risk associated with CAT; a 30% increased LGIE risk with anticoagulant and antiplatelet dual therapy; a 3- to 6-fold increased risk in transfusions associated with CAT; and 30% increased risk of hospitalization associated with dual therapy and a 40% increase in risk with triple therapy.
Specific Aim #2: Participants (N=201) preferred to maximize cardiovascular benefit (reduced 5-year risk of MI and CVA) while minimizing potential CAT-related adverse events (intracranial hemorrhage and upper GI bleeding risk). There was little variation regarding dominant preferences after Monte-Carlo simulations. Alterations in lifestyle (i.e., changes in physical activity), nuisance side-effects (i.e., stomach discomfort) and complexity of CAT regimen (i.e., dual vs. triple CAT therapy with or without a gastroprotective agent) were important, but less so, than risk-benefit attributes. Rank order of the relative importances differed slightly by age. Among respondents 60-69 years, prevention of MI and CVA had greater relative importance than avoidance of neurologic (intracerebral hemorrhage [ICH} or CVA) or UGIE complications. Those aged 70-79 years valued CVA risk prevention more than MI risk prevention. The eldest respondents (80 years and older) showed increasing concern regarding neurologically induced incapacity with dominant preferences for reduction of CVA and ICH risk. Among all ages, prevention of UGIE was viewed as important, but its relative importance ranked after CV benefit and before the burden of medication self-management.
A 15% increase in medication adherence was observed in patients prescribed a preference-concordant CAT strategy and a 6% increase in adherence was observed in patients prescribed a preference-discordant strategy. Qualitative analysis of exit interviews (n=56) suggested preference elicitation contributed to enhanced cognitive engagement, awareness and activation of patients to discuss CAT preferences with their physicians.
This project fills three fundamental gaps in our knowledge: 1) quantification of the real-life risk of CAT-associated events (i.e., UGIE, LGIE, hospitalization and transfusion) among elderly veterans; 2) quantification of how patients trade-off between the potential cardioprotective benefit of CAT and the risk of UGIE; and most importantly, 3) how preferences influence medication adherence in a longitudinal fashion.
Our results reveal CAT-related bleeding events, including UGIE, LGIE, transfusions and hospitalizations, are a clinically relevant risk. VA providers should be cognizant of the magnitude of bleeding risk and counsel elderly veterans accordingly. The preference data suggested there was little variation among elderly veterans' dominant preferences- lowering CV risk was valued more strongly than UGIE bleeding aversion. However, the oldest elderly patients focus more on preserving function over survival.
Pharmacoepidemiology data demonstrated a 20% improvement in CAT adherence among patients prescribed a preference-concordant strategy, and a 6% improvement among the preference-discordant group. We hypothesize preference elicitation served as an "educational moment" for patients; by working through trade-offs, patients actively engaged in clarifying their preferences, explicitly learning about risks and benefits and the burden of CAT self-management. We further hypothesize the ACA task improved adherence through behavioral and perceptual mediators, including 1) enhanced patient cognitive engagement and activation, 2) patient reassurance with prescribed goals of therapy, and 3) enhanced "buy-in" regarding the prescription decision that exceeded their pre-survey perception.
The promotion of shared-decision making and patient-centered care is fundamental to the Department of Veterans Affairs. We have shown preference-elicitation, active patient engagement in decision-making, improvement of patient education regarding risks-benefits and veteran empowerment to voice concerns and desires regarding CAT-related outcomes positively impacts subsequent CAT medication adherence. We plan to use our results to develop educational and behavioral tools to educate and support patients in the process of making medication decisions in the ambulatory care setting.
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: Aging, Older Veterans' Health and Care, Health Systems, Cardiovascular Disease
DRE: Treatment - Observational
Keywords: Chronic disease (other & unspecified), Outcomes, Patient preferences
MeSH Terms: none