US Veterans Use Vitamins and Supplements as Substitutes for Prescription Medication

Goldstein, Jennifer N. MD*; Long, Judith A. MD†,‡; Arevalo, Doris BA§; Ibrahim, Said A. MD; Mao, Jun J. MD

doi: 10.1097/MLR.0000000000000199
Original Research

Background: Vitamins and supplements are the most commonly used form of complementary and alternative medicine in the United States. Growing research suggests that patients substitute vitamins and supplements for their prescription medications. The reasons might include cost of prescription medications and discordant patient and doctor health belief systems.

Objectives: To investigate the prevalence of substitution of vitamins and supplements for prescription medications among veterans who receive care in the VA health care system and whether substitution is associated with prescription rationing due to cost, treatment beliefs, or distrust of the health system.

Research Design: Cross-sectional observational survey.

Subjects: Primary care patients (n=275) at the Philadelphia VA Medical Center.

Measures: Medication substitution, prescription medication rationing, treatment beliefs, and health system distrust were measured with structured instruments. Multivariate logistic regression was performed with substitution as the dependent variable.

Results: A significant number of primary care patients in the VA system use vitamins and supplements 206 (75%). The prevalence of medication substitution is high 48 (18%). Medication substitution is strongly associated with prescription rationing due to cost (adjusted odds ratio 6.3, 95% confidence interval: 2.0–19.5, P=0.001). Similarly, greater belief in complementary and alternative approaches to care positively predicts medication substitution (adjusted odds ratio 1.08, 95% confidence interval: 1.01–1.15, P=0.011). There is no significant association between health system distrust and likelihood of medication substitution.

Conclusions: Medication substitution is prevalent in this sample of inner city primary care patients who receive care in the VA system. Cost of prescriptions and belief in the value of complementary and alternative approaches to care appear to be associated with this patient-driven treatment decision.

*Department of Medicine, Christiana Hospital, Newark, DE

Philadelphia VA Medical Center

Department of Medicine, University of Pennsylvania Philadelphia, PA

§University of Illinois, Urbana-Champaign, Urbana, IL

Philadelphia VA Medical Center, Center for Health Equity and Promotion

Department of Family and Community Medicine, University of Pennsylvania, Philadelphia, PA

A portion of our results was presented in a poster presentation at the Society for General Internal Medicine Mid-Atlantic Conference on March 14, 2014 and the National Society for General Internal Medicine Conference on April 25, 2014.

Dr Long and Dr Ibrahim are employed by the VA Medical Center in Philadelphia. The views expressed in this paper are those of the authors and do not necessarily represent the views or policies of the Department of Veterans Affairs or the United States Government.

The Ruth L. Kirschstein National Research Service Award (NRSA): Grant #: T32HP10026 funded the fellowship of J.N.G. and therefore this research.

The authors declare no conflict of interest.

Reprints: Jennifer N. Goldstein, MD, Department of Medicine, Christiana Hospital, Ammon Bldg, Suite 2E72A, 4755 Ogletown Stanton Rd, Newark, DE 19718. E-mail:

Article Outline

Veterans who receive primary care in the VA health system have a high burden of chronic disease.1,2 Vitamins and nutritional supplements, defined as natural products by the National Center for Complementary and Alternative Medicine (NCCAM),3 are the most commonly used form of CAM,4 specifically among patients with chronic conditions and those who take prescription medications.5,6 Although many patients use vitamins and supplements in addition to their prescription medications, nearly 1 in 5 Americans use vitamins and supplements instead of the medications prescribed by their physicians.7 Little is known about why patients substitute vitamins and supplements for prescriptions. Some studies have linked both the use of CAM and medication nonadherence to increased financial burdens4,8 suggesting that some patients may see alternative treatments as a less expensive substitute for conventional care. In addition, a study of veterans suggests that those who use CAM are more likely to have a greater desire for a holistic approach to health care and distrust of the health system.9 The decision to substitute may therefore be influenced by cost, treatment beliefs, and/or health system distrust.

Given the high burden of chronic illness and its associated treatments in the VA primary care population,1,2 we might expect an even higher proportion of these patients to substitute than the general population. The primary objective of this study was to determine the prevalence of substitution within a primary care population and the most common chronic diseases for which patients substitute supplements for medication. In addition, we sought to determine how health care–related cost concerns, treatment beliefs, and health system distrust relate to substitution.

Back to Top | Article Outline


Study Sample

We performed a cross-sectional survey study of a convenience sample of primary care patients in the VA Medical Center in Philadelphia, PA. Questionnaires were distributed by 2 trained research assistants in the waiting rooms of 4 designated recruitment areas: the Primary Care Practice, Women’s Veterans Health Center, phlebotomy laboratory, and radiology. All questionnaires were completed by the patients in the waiting room or examination room and were collected on site by the research assistant. In preferentially capturing patients with more visits, this approach naturally emphasizes those patients with more chronic illness. To participate in the study, patients had to be 18 years or older, English speaking, and have a primary care physician at the Philadelphia VA Medical Center. Informed consent was obtained, and participants received a $5 gift card for their time. The study was approved by the Institutional Review Board at the Philadelphia VA Medical Center.

Back to Top | Article Outline

Study Measures

The primary outcome measure was substitution of vitamins and supplements for prescription medicine, measured by the question: “in the past 3 months, have you taken a dietary supplement to treat or prevent a health problem or condition instead of taking a prescription drug” adapted from the FDA Health and Diet Survey 2008.7 In addition, participants were asked which medical condition they substituted for, and other general questions about their use of multivitamins, single-ingredient vitamins and minerals, and nutritional supplements.

Health care–related cost concerns were measured with the question: “in the past 3 months, have you ever taken less of any medications than prescribed by your doctor because of the cost?” which is similar to questions used by other studies.10,11

Treatment beliefs were measured using the Complementary and Alternative Medicine Beliefs Inventory (CAMBI), a 17-item scale, which measures 3 domains associated with CAM use: belief in natural treatments, holistic approaches to treatment, and participation in treatment.12 A higher score on the CAMBI reflects a greater belief in alternative medicine. As has been done previously, a 5-point Likert scale was used instead of the original 7-point Likert scale to maintain consistency with the other scales in the survey.13 Four items were negatively phrased and reverse-scored. The minimum score for the scale was 17 and the maximum was 85. Internal consistency of the total scale was calculated for our patient sample and was acceptable (Cronbach α=0.77).

Health care system distrust was measured using the Health Care System Distrust Scale, a 9-item scale that measures 2 domains of distrust related to the health system: competence and values. All items are scored on a 5-point Likert scale with a minimum score of 9 and a maximum score of 45.14 Internal consistency for the total scale was calculated for our patient sample and was acceptable (Cronbach α=0.87).

Participants reported date of birth, race, ethnicity, annual income, and level of education. They also reported whether or not they obtained all of their medications from the VA. Two variables were obtained from the patient electronic medical record: sex and whether the respondent was exempt from paying a copayment for prescription medication. Veterans may be deemed copay exempt if they have a disability or illness related to their military service or if they fall below a specified income threshold. The questionnaire was written at a middle school reading level (Flesch-Kincaid Readability Grade Level of 6.99, FOG level of 7.64).

Back to Top | Article Outline

Statistical Analysis

All data were analyzed using STATA 12.1 (College Station, TX). Bivariate analysis using the Pearson χ2 test was performed to compare nonscaled covariates between substituters and nonsubstituters. Because of the small sample size and non-normal distribution of the scaled results, the Mann-Whitney U test was used for scaled items. We then constructed a multivariate model including all variables from the bivariate analysis significant at P<0.2. All analyses were 2-sided with α<0.05 indicating significance. We hypothesized a 0.33 SD difference between substituters and nonsubstituters and required a total sample of 290 to detect this difference at 0.05 level of significance.

Back to Top | Article Outline


Three hundred questionnaires were completed and collected. Twenty-five questionnaires were excluded from analysis: 5 duplicate respondents, 2 patients without primary care at the Philadelphia VA, 1 with an incomplete consent form, and 17 for answering the same question response for each question in a scale (for example, circled the number 3 all the way down the scale). This left 275 completed questionnaires for the final analysis. Patients refused to participate in the study for various reasons, including time constraints and unwillingness to share personal information.

Back to Top | Article Outline

Sample Characteristics

The majority of participants were over 40 years of age, 53% were female and 63% were African American. Over 80% of participants had greater than a high school education and more than half had an annual income of <$50,000 annually. The median score on the total CAMBI scale was 66 (interquartile range: 63–72). The median for the Health System Distrust scale was 25 (IQR: 21–29), which is similar to that of the general West Philadelphia population15 (Table 1).

Use of vitamins and supplements was high among veterans, with 206 (75%) endorsing use in the previous 3 months. Prevalence of substitution was consistent with the national average,7 with 48 (18%) reporting substitution of vitamins and supplements for prescription medication in the previous 3 months (9 participants did not answer the substitution question). Among the subgroup of substituters, 25% substituted for hyperlipidemia medications, whereas 17% did so for anxiety/depression medications and 15% for arthritis/back pain medications. Other common conditions for which patients substituted included hot flashes (15%), diabetes (10%), and hypertension (8%) (Fig. 1).

Back to Top | Article Outline

Cost, Beliefs, and Substitution

Respondents who rationed their prescriptions due to cost, obtained medications outside of the VA, and were copay exempt were significantly more likely to substitute than those who did not (21% vs. 6%, P=0.001; 31% vs. 14%, P=0.005; 81% vs. 63%, P=0.018), (Table 1). Individuals who substituted vitamins and supplements for their prescriptions had significantly higher pro-CAM beliefs (higher CAMBI score) than those who did not (median 70 vs. 66, P=0.015). Health system distrust was not significantly associated with substitution (median 25 vs. 27, P=0.060) (Table 1).

A subanalysis was performed of copay exempt veterans and revealed that income levels varied significantly between those who were nonexempt and exempt with 27%, 36%, and 36% of individuals in the low, middle, and higher income levels, respectively, in the nonexempt group, versus 35%, 17%, and 48% in the copay exempt group (P=0.002). We found no other distinguishing characteristics between copay exempt and nonexempt veterans.

Back to Top | Article Outline

Results From Multivariate Models

In multivariate modeling, substitution was strongly associated with prescription rationing [adjusted odds ratio (AOR) 6.3, 95% confidence interval (CI): 2.03–19.5, P=0.001] and obtaining medications outside of the VA (AOR 2.8, 95% CI: 1.05–7.43, P=0.038). Substitution was also strongly associated with copay exempt status (AOR 6.4, 95% CI: 1.88–21.49, P=0.003). Substitution was significantly associated with greater belief in complementary and alternative medicine (higher CAMBI score) (AOR 1.08, 95% CI: 1.01–1.15, P=0.011), but was not significantly associated with health system distrust (AOR 1.02, 95% CI: 0.96–1.08, P=0.562) (Table 2).

Back to Top | Article Outline


Veterans who receive care at the VA have access to low-cost or free prescription medications. However, many use alternative products instead. This is one of the only studies to date to examine the prevalence of substitution of vitamins and supplements for prescription medication among primary care patients. It is also one of the first to evaluate some potential key factors that motivate patients to make this treatment decision.

We found that 75% of veterans use vitamins and supplements and 18% substitute, which is the same proportion as the general population.7 Veterans substitute for serious medical conditions, most commonly, hyperlipidemia. On both bivariate and multivariate analyses we found that veterans who rationed their prescriptions because of cost issues were significantly more likely to substitute, suggesting that financial constraints may induce people to use vitamins and supplements in place of their prescription medication. As veterans are less likely to ration their prescriptions due to cost than the majority of the US population,16 our findings likely underestimate substitution for financial reasons within the general population.

Interestingly, we also found that veterans who were exempt from paying copays for their prescriptions were significantly more likely to substitute. This runs counter to data that increased insurance coverage leads to increased use of health resources, and improved adherence.17,18 Income level varied significantly between copay exempt and nonexempt veterans. However, given that many individuals qualify for copay exemption on the basis of disability rather than income level, such variability is expected. Furthermore, as none of the copay exempt veterans were paying for their prescriptions, the issue of income is probably mute.

Previous studies have found that Veterans who are copay-exempt are, in general, less compliant than their copay-paying counterparts.19 Our analysis supports that copay-exempt individuals are less compliant with their prescriptions in that they substitute vitamins and supplements for their prescriptions more than their nonexempt counterparts. However, our survey did not illuminate specific factors that explained the difference in behavior between copay exempt and nonexempt veterans. This requires future study to elucidate the causes underlying the observed differences.

Our findings that pro-CAM beliefs were associated with substitution are not surprising, as such beliefs have been associated with increased CAM use in previous studies.12,20

We found that levels of health system distrust in the VA were similar to that of the general West Philadelphia population,15 although distrust was not significantly associated with substitution. Distrust of the health system has been cited in previous studies of veterans who use CAM.9 It has also been associated with patients who exclusively use alternative medicine (have no ties to Western medicine).20 However, no other study has looked at distrust and substitution. Thus, although distrust might drive overall use of CAM, it does not seem to drive substitution among veterans who use Western medications.

There are several notable limitations to our study. First, our sample size is relatively small, thus potentially lowering the accuracy of the effect size. Second, we studied a convenience sample of veterans who receive primary care at a major inner city VA hospital. Therefore, our findings might not be generalizable to the general population. On the other hand, our sample was approximately half female, and majority low income and African American, which is representative of the general West Philadelphia population from which the VA draws most of its patients.

Back to Top | Article Outline


In summary, we found that the vast majority of veterans in an inner city VA Medical Center use vitamins and supplements, and a substantial proportion substitute them for conventional prescriptions. Although the specific reasons for substitution are unknown for each veteran, it appears that cost and treatment beliefs influence their decision. These are 2 areas that are amenable to further study and intervention. Veterans who use vitamins and supplements because they cannot afford their prescriptions should be provided with support and assistance through the VA to ensure that they are receiving appropriate care. Those who substitute because of their beliefs in complementary and alternative approaches to care may benefit from discussions with a pharmacist or other care provider to elicit specific concerns or motivations that drive their behaviors. Previous studies have found that long-term educational interventions, which provide patient-centered advice, have been successful in improving medication adherence.21–23 Understanding the specific substitutions taking place and the motivations behind them could foster a better understanding of whether the substitution is potentially dangerous. It could also improve medication reconciliation and screening for potential pharmaceutical/supplement interactions. Such patient-driven approaches to care are supported within the medical home model of the VA, the Patient Aligned Care Team Model (PACT), which has been implemented in all VA primary care clinics nationwide.24 Most importantly, VA providers should be aware of the high prevalence of substitution, and should ask about the use of vitamins and supplements when addressing medication adherence with patients.

Back to Top | Article Outline


The authors would like to acknowledge David A. Asch, MD, for his thoughtful comments and edits.

Back to Top | Article Outline


1. Yoon J, Zulman D, Scott JY, et al.. Costs Associated With Multimorbidity Among VA Patients. Med Care. 2014;52:S31–S36.
2. Yu W, Ravelo A, Wagner TH, et al.. Prevalence and costs of chronic conditions in the VA health care system. Med Care Res Rev. 2003;60:146S–167SS.
3. National Institutes of Health. Complementary, Alternative, or Integrative Health: What’s in a Name? 2008. Available at: Accessed March 14, 2014.
4. Barnes PM, Bloom B, Nahin RL. Complementary and Alternative Medicine Use Among Adults and Children: United States. 2008.Hyattsville, MD: National Center for Complementary and Alternative Medicine, National Institutes of Health.
5. Gardiner P, Graham RE, Legedza AT, et al.. Factors associated with dietary supplement use among prescription medication users. Arch Intern Med. 2006;166:1968–1974.
6. Nahin RL, Pecha M, Welmerink DB, et al.. Concomitant use of prescription drugs and dietary supplements in ambulatory elderly people. J Am Geriatr Soc. 2009;57:1197–1205.
7. US Food and Drug Administration. 2008 Health and Diet Survey: Topline Frequencies (Weighted). Available at: Accessed March 4, 2014.
8. Pagan JA, Pauly MV. Access to conventional medical care and the use of complementary and alternative medicine. Health Aff. 2005;24:255–262.
9. Kroesen K, Baldwin CM, Brooks AJ, et al.. US military veterans’ perceptions of the conventional medical care system and their use of complementary and alternative medicine. Fam Pract. 2002;19:57–64.
10. Piette JD, Heisler M, Krein S, et al.. The role of patient-physician trust in moderating medication nonadherence due to cost pressures. Arch Intern Med. 2005;165:1749–1755.
11. Kurlander JE, Kerr EA, Krein S, et al.. Cost-related nonadherence to medications among patients with diabetes and chronic pain: factors beyond finances. Diabetes Care. 2009;32:2143–2148.
12. Bishop FL, Yardley L, Lewith G. Developing a measure of treatment beliefs: the complementary and alternative medicine beliefs inventory. Complement Ther Med. 2005;13:144–149.
13. Grzywacz JG, Neiberg R, Quandt SA, et al.. Measuring differential beliefs in complementary therapy research: an exploration of the Complementary and Alternative Medicine Beliefs Inventory (CAMBI). Complement Ther Med. 2012;20:54–60.
14. Shea JA, Micco E, Dean LT, et al.. Development of a revised Health Care System Distrust scale. J Gen Intern Med. 2008;23:727–732.
15. Armstrong K, McMurphy S, Dean LT, et al.. Differences in the patterns of health care system distrust between blacks and whites. J Gen Intern Med. 2008;23:827–833.
16. Piette JD, Heisler M. Problems due to medication costs among VA and non-VA patients with chronic illnesses. Am J Manag Care. 2004;10:861–868.
17. Elliott DJ, Robinson EJ, Anthony KB, et al.. Patient-centered outcomes of a value-based insurance design program for patients with diabetes. Popul Health Manag. 2013;16:99–106.
18. Newhouse JP, Manning WG, Morris CN, et al.. Some interim results from a controlled trial of cost sharing in health insurance. N Engl J Med. 1981;305:1501–1507.
19. Doshi JA, Zhu J, Lee BY, et al.. Impact of a prescription copayment increase on lipid-lowering medication adherence in veterans. Circulation. 2009;119:390–397.
20. Astin JA. Why patients use alternative medicine: results of a national study. JAMA. 1998;279:1548–1553.
21. Murray MD, Young J, Hoke S, et al.. Pharmacist intervention to improve medication adherence in heart failure: a randomized trial. Ann Intern Med. 2007;146:714–725.
22. McDonald HP, Garg AX, Haynes RB. Interventions to enhance patient adherence to medication prescriptions: scientific review. JAMA. 2002;288:2868–2879.
23. Clifford S, Barber N, Elliott R, et al.. Patient-centred advice is effective in improving adherence to medicines. Pharm World Sci. 2006;28:165–170.
24. Veterans Health Administration. VHA Patient Centered Medical Home Model (PACT) Concept Paper. 2011. Available at: Accessed March 11, 2014.

complementary and alternative medicine; medication adherence; chronic disease management; attitudes and beliefs; health system distrust

© 2014 by Lippincott Williams & Wilkins.