Veteran Perceptions, Interest, and Use of Complementary and Alternative Medicine

Demographic Information

Age range was assessed to avoid collecting identifying information. The questionnaire also included gender, military era/deployment, employment status, and race and ethnicity.

• Self-Rated Health (SF1). Self-rated health was assessed with a widely used single-item question that correlates highly with actual overall health and with function and quality of life.^17,18 Respondents were asked to rate their health as excellent (5), very good, good, fair, or poor (1).
•  Pain Screen (PEG 3-item scale). This 3-item screen has shown reliability and validity and is comparable to longer pain questionnaires.¹⁹ Respondents were asked to rate 3 measures of their pain and its consequences on a scale of 0 (no pain or no interference from pain) to 10 (worst pain or interference). Responses were averaged to determine pain score.
•  PTSD Screen. This 4-item PTSD screen was developed for primary care and is widely used in VA settings.²⁰ For each item, respondents were asked to check off whether they have had specific PTSD symptoms within the past month. The screen was considered positive with 3 of 4 affirmative responses.
•  Anxiety and Depression Screen (PHQ-4). This 4-item scale combines the brief 2-item scales for screening anxiety and depression in primary care.²¹ For each depression or anxiety symptom, respondents selected from “not at all,” (1) “several days ”(2), “more days than not,” (3) and “nearly every day.” (4) For each 2-item screen, a sum of 5 or more indicated a positive screen.
• Self-Efficacy for Health Management (modified). The original 6-item self-efficacy screen was developed to test self-efficacy in managing chronic disease.²² Since not all participants in the current study were expected to have a chronic disease, the questions were modified to address more general self-efficacy for health management. Although the scale had not been adapted in this way or validated with this change, other authors have similarly adapted it to address specific chronic diseases with satisfactory results.^23,24 For each item, respondents were asked to rate their confidence in their ability to manage aspects of their health on a scale of 1 (not at all confident) to 10 (very confident). Participants could also check “not applicable” for items that did not apply to their health concerns, and these items were not counted in the average score.
•  Familiarity With and Interest in CAM. The authors developed a checklist to assess whether participants had heard of, tried, or were practicing the 4 CAM techniques featured in the SWK and to gauge their interest in learning about them (ie, meditation/guided imagery, breathing exercises, yoga, tai chi or qigong). For each technique, respondents selected that they have “never heard of,” “heard of but never tried,” “have done this in the past,” or “are currently doing.” For some analyses, the first 2 and last 2 options were combined to determine whether respondents had done each practice. They were also asked to check off whether they would like to learn more about the practice and whether they would like to try it with an instructor and/or try it on their own. For some analyses, each technique was looked at separately, whereas for others, the 4 techniques were combined to determine whether they had tried or were currently doing any of them.
• Barriers to Practice. The authors developed a checklist of 10 barriers to practicing CAM techniques based on research but with adjustments to the specific practices and population under investigation.²⁵ The checklist included an open-end response to allow respondents to add barriers. The barrier list was a checklist and not a validated scale.
•  Perceived Benefits of CAM. The authors developed 2 questions to assess the perceived benefits of these techniques on functionality and overall wellness, rated on a Likert scale from 1 (no benefits) to 10 (very much).

Statistical Analysis

Survey instruments were scored according to generally accepted and published practices. Item-level analysis was performed to identify missing responses and describe the sample. Summary statistics were reported. Pearson product moment correlation was used to detect associations between continuous variables. Analysis of variance (ANOVA) was used to detect associations between dichotomous and continuous variables. Chi-square tests were used to detect associations between categorical variables, specifically looking at clinically meaningful differences between veterans who had experience with or interest in trying independent CAM practices and those who did not. Linear regression analysis was used to determine significant associations between participant characteristics and the belief that independent CAM practices would be helpful with daily function.

Results

The response rate for returning surveys was low (n = 134; 18.2%). Surveys distributed by peers in the community setting had the highest response rate (38%), followed by surveys distributed in primary care (23%).