PPO 15-429
Understanding veterans' preferences for smoking cessation treatment in primary care
David A. Katz, MD MSc Iowa City VA Health Care System, Iowa City, IA Iowa City, IA Funding Period: June 2016 - August 2017 Portfolio Assignment: Mental and Behavioral Health |
BACKGROUND/RATIONALE:
Most veterans are reluctant to talk to a cessation counselor and many do not adhere to prescribed pharmacotherapy for smoking cessation (or prefer to quit "cold turkey"). Behavioral choice theory suggests that promoting patients' ability to choose and incorporating patients' preferences for available options will improve satisfaction with treatment decisions and adherence to recommended treatments; however, little is known about the value that veterans place on specific attributes of smoking cessation treatment. The evaluation of patient preferences using a discrete choice experiment (DCE) is based on the premise that the value of tobacco treatment can be represented by a limited number of attributes, which can be broken down into levels that patients value. OBJECTIVE(S): The primary aims of this study are: 1) to identify the most salient attributes of tobacco treatment that are important to veterans and to describe the decision making process of veterans in choosing between hypothetical tobacco treatments, and 2) to measure the relative importance of key attributes of tobacco treatment in VA primary care outpatients. A secondary aim of this study is to assess the stability of patients' preferences for tobacco treatment. METHODS: To address aim 1, we coded 29 in-depth interviews of current smokers that were conducted in FY2014-15 during pilot testing of the initial DCE questionnaires (based on literature review) in the Iowa City VA Healthcare System. These interviews included a "think aloud" component in order to gain insight into the thought processes used by veterans when completing DCE choice tasks. The qualitative research team used constant comparison techniques to characterize the issues raised by subjects. We reviewed the transcripts for evidence of non-compensatory decision making (e.g., choice is dominated by a single high priority attribute) and evaluated "opt-out" choices to identify strong preferences for (or aversion to) particular treatment attributes. We also asked if important attributes were missing. Findings from these interviews were used to revise the initial DCE questionnaires. To address aim 2, a research assistant administered the revised DCE questionnaires in person to 123 primary care patients in two VA medical centers who smoked at least one cigarette per day on average and were planning to quit smoking within the next 6 months. Patients were randomized to cessation counseling or drug therapy versions of the DCE questionnaire, which included 14 choice sets (each set containing two hypothetical treatments and an opt-out option). Test-retest reliability was assessed by repeating the questionnaire in a subsample of 30 subjects. We used a hierarchical Bayesian approach with a multinomial logit model to estimate part-worth utilities for each treatment attribute; the relative importance of each attribute in the study sample was calculated. FINDINGS/RESULTS: Aim 1: Most subjects were able to complete the DCE questionnaire without difficulty; however, not all subjects were fully engaged, some showed signs of cognitive overload, and some had difficulty with specific scenarios. Three common patterns of decision making emerged during analysis of the think-aloud data: 1) counting "pros" and "cons" of each treatment alternative, 2) focusing primarily on a single "must-have" attribute, and conversely, 3) a "rule out" attribute. One common strategy for comparing treatment options was to simply count the number of preferred counseling attributes within each alternative. Most subjects clearly considered two or more attributes in choosing between treatments. In 17% of cases, however, subjects zeroed in on a single attribute and were not willing to trade, regardless of the values of other attributes. To further reduce complexity of the choice task, subjects routinely discounted the importance of certain attributes or used a "rule out" strategy to eliminate a given treatment choice if it included an undesirable attribute. Cognitive interviews provided valuable insights into the comprehension and interpretation of DCE attributes, the decision processes used by subjects during completion of choice tasks, and the underlying reasons for noncompensatory decision making and "opt-out" choices. Based on cognitive interviews, we made revisions to both DCE questionnaires, including changes to the description of attribute levels and changes in wording to improve clarity and specificity. Aim 2: Of 812 potential candidates, 123 patients were determined to be eligible and agreed to participate (61 counseling, 62 medication). Patients were predominantly middle-aged (mean 59.2 years) and 87% were white; 36 and 81% had previously tried to quit with counseling and medication, respectively. Mean part-worth utilities and 95% credible intervals were significantly greater than (or less than) zero for the following attributes of counseling (relative importance is shown in parentheses): higher quit rate (29%), emphasis on the veteran's choice on when and how to quit (15%), counselor who always listens carefully and explains things clearly (13%), counselor whom one sees frequently in clinic (10%), and receiving printed materials on smoking cessation rather than internet based information or text messages (10%). For medication, mean part-worth utilities and 95% credible intervals were significantly greater than (or less than) zero for the following attributes (relative importance is shown in parentheses): low risk of physical side effects (30%), higher quit rate (20%), zero copayment (17%), monthly check-in calls to assess adverse effects (11%), and low weight gain (10%). Test-retest reliability at 2 weeks was fair (kappa=0.39 for both counseling and medication). IMPACT: This project demonstrates the feasibility of using cognitive interviewing techniques to develop and refine DCE questionnaires and then using DCE to evaluate veterans' preferences for tobacco treatment. DCE data can be used to detect and model underlying latent segments of respondents who share similar preferences and may respond to more personalized tobacco treatment strategies. Study models can also be used to evaluate choice propensity (or market share) of different treatment configurations that may interest VA operational partners. Finally, this pilot project lays the groundwork for future development of a patient decision aid that can promote shared decision making with the primary care (PACT) team, improve the quality of decision making, and actively engage patients in tobacco treatment. External Links for this ProjectNIH ReporterGrant Number: I21HX002079-01Link: https://reporter.nih.gov/project-details/9084387 Dimensions for VADimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.Learn more about Dimensions for VA. 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 PUBLICATIONS:Journal Articles
DRA:
Substance Use Disorders, Health Systems Science
DRE: Technology Development and Assessment Keywords: none MeSH Terms: none |