To the Editor:
Dietary and herbal supplements are widely used, perceived as safe, and promoted for therapeutic potential and correction of nutrient deficiencies. However, they may be harmful for people with CKD due to ingredients that are only partly scrutinized by regulatory agencies.1–3 Supplements may be inaccurately labeled, interact with each other and/or prescription medications, or contain nephrotoxic substances. Renally eliminated supplements and metabolites may accumulate in CKD, leading to adverse outcomes. Earlier studies examined select supplement use by people with CKD,4 but without conclusive safety data for most supplements, a broader examination of supplement use in CKD is needed to inform interventions in this population. Furthermore, whether CKD awareness translates to caution in supplement use is unknown.
Using NHANES data from 2011 to 2014, we characterized the prevalence of use by nonpregnant adults (≥20 years) of any and/or high-risk supplements (high risk defined by the National Kidney Foundation [NKF] on the basis of containing potassium, phosphorus, or other compounds potentially harmful with CKD, dialysis, or after kidney transplantation5; Table S1), focusing on differences by GFR category (G1, G2, etc; see Table 1 notes) and CKD awareness. NHANES uses stratified, multistage, probability-cluster techniques to sample populations representative of noninstitutionalized US civilians (Item S1).
Table 1.
Study Population, Stratified by GFR Category and Awareness
| G3 | G4–G5 | |||||
|---|---|---|---|---|---|---|
| Unaware (685 [6.1 %]) | Aware (97 [0.7%]) | Unaware (39 [0.2%]) | Aware (66 [0.4%]) | P | ||
| Supplement Use | ||||||
| Any | 27.3 (24.3, 30.4) | 33.1 (26.9, 39.9) | 40.5 (25.9, 57.0) | 34.6 (15.6, 60.2) | 37.4 (21.9, 55.9) | 0.07 |
| Low risk only | 24.6 (21.9, 27.6) | 30.8 (24.9, 37.4) | 37.1 (23.4, 53.3) | 34.6 (15.6, 60.2) | 37.4 (21.9, 55.9) | 0.03 |
| High riska | 2.7 (2.1, 3.4) | 2.3 (1.2, 4.5) | 3.4 (0.8, 13.8) | 0 (0, 0) | 0 (0, 0) | 0.8 |
| Decision to Take the Supplement(s)b | 0.03 | |||||
| Personal decision | 46.6 (43.0, 50.2) | 37.6 (26.2, 50.6) | 28.7 (14.0, 50.1) | 34.5 (13.6, 64.0) | 13.8 (5.7, 29.9 | |
| Clinician recommendation | 13.0 (11.4, 14.7) | 21.6 (15.3, 29.7) | 20.7 (9.1, 40.3) | 6.4 (1.1, 29.3) | 30.1 (8.6, 66.3) | |
| Both of above | 28.1 (25.1, 31.2) | 35.1 (25.2, 46.6) | 43.6 (21.9, 68.0) | 47.7 (13.4, 84.3) | 32.3 (16.5, 53.5) | |
| Don’t know | 12.3 (10.8, 14.0) | 5.6 (3.2, 9.9) | 7.0 (2.2, 19.9) | 11.4 (1.8, 48.0) | 23.8 (10.3, 45.8) | |
| Mean Age, y | 45.8 (44.9, 46.7) | 70.0 (68.8, 71.1) | 68.6 (66.2, 71.0) | 73.2 (71.3, 75.0) | 63.2 (60.0, 66.5) | <0.001 |
| Age category | <0.001 | |||||
| <45 y | 48.7 (46.2, 51.1) | 1.3 (0.6, 3.1) | 4.4 (2.1, 9.2) | 3.7 (0.5, 20.8) | 11.6 (4.9, 25.1) | |
| 45–64 y | 37.4 (35.5, 39.3) | 27.6 (22.2, 33.7) | 26.6 (15.3, 42.2) | 15 (6.1, 32.3) | 35.4 (21.0, 53.2) | |
| 65–74 y | 9.5 (8.7, 10.4) | 26.4 (23.0, 30.1) | 34.6 (24.6, 46.3) | 15.4 (6.6, 31.9) | 28.1 (16.4,43.7) | |
| ≥75 y | 4.4 (3.9, 5.1) | 44.7 (39.3, 50.1) | 34.4 (26.1, 43.7) | 65.9 (55.6, 74.9) | 24.8 (13.2, 41.8) | |
| Sex | 0.006 | |||||
| Male | 49.4 (48.2, 50.6) | 42 (37.9, 46.1) | 49.8 (37.8, 61.8) | 28.6 (16.7, 44.5) | 43.3(27.5, 60.6) | |
| Female | 50.6 (49.4, 51.8) | 58 (53.9, 62.1) | 50.2 (38.2, 62.2) | 71.4 (55.5, 83.3) | 56.7 (39.4, 72.5) | |
| Race/Ethnicity | <0.001 | |||||
| White | 66.1 (60.7,71.1) | 82.3 (77.5, 86.3) | 72.8 (61.0, 82.1) | 59.1 (40.2, 75.7) | 52.2 (37.5, 66.5) | |
| Black | 10.8 (8.4, 13.8) | 7.0 (5.0, 9.8) | 17.7 (10.3, 28.8) | 26.8 (15.0, 43.0) | 28.7 (19.6, 40.1) | |
| Hispanic | 15.1 (11.8, 19.1) | 5.9 (4.0, 8.5) | 7.4 (3.9, 13.5) | 11.2 (4.1, 27.2) | 14.7 (7.5, 26.7) | |
| Asian | 5.3 (4.3, 6.6) | 2.5 (1.6, 4.1) | 1.6 (0.5, 5.3) | 0 (0, 0) | 2.3 (0.7, 7.7) | |
| Other/multiracial | 2.6 (2.1, 3.3) | 2.3 (1.4, 3.7) | 0.5 (0.1, 3.6) | 2.9 (0.4, 20.5) | 2.1 (0.5, 9.1) | |
| Educational Attainment | 0.002 | |||||
| ≤HS | 36 (32.6, 39.6) | 41.3 (35.7, 47.1) | 46.7 (31.3, 62.7) | 81.7 (61.0, 92.8) | 52.7 (39.0, 66.0) | |
| >HS | 64 (60.4, 67.4) | 58.7 (52.9, 64.3) | 53.3 (37.3, 68.7) | 18.3 (7.2, 39.0) | 47.3 (34.0, 61.0) | |
| CVD | <0.001 | |||||
| Yes | 6.5 (5.9, 7.2) | 32.5 (28.2, 37.1) | 38.3 (27.6, 50.3) | 60.5 (30.3, 84.4) | 43.1 (30.3, 56.8) | |
| No | 93.3 (92.6, 94.0) | 67.2 (62.7, 71.5) | 59.1 (46.7, 70.5) | 39.5 (15.6, 69.7) | 56.9 (43.2, 69.7) | |
| Missing | 0.1 (0.1, 0.2) | 0.3 (0.1, 1.2) | 2.5 (0.6, 10.8) | 0 (0, 0) | 0 (0, 0) | |
| Hypertension | <0.001 | |||||
| Yes | 30.2 (28.4, 32.0) | 70 (65.6, 74.1) | 83.4 (67.8, 92.3) | 79.3 (71.1, 85.6) | 79.6 (60.2, 91.0) | |
| No | 69.8 (67.9, 71.6) | 29.5 (25.6, 33.7) | 16.6 (7.7, 32.2) | 18.7 (12.3, 27.3) | 20.4 (9.0, 39.8) | |
| Missing | 0.1 (0.0, 0.2) | 0.5 (0.1, 2.8) | 0 (0, 0) | 2 (0.3, 13.3) | 0 (0, 0) | |
| Diabetes | <0.001 | |||||
| Yes | 8.1 (7.3, 8.9) | 23.6 (20.2, 27.4) | 45.2 (35.3, 55.4) | 47.2 (23.1, 72.7) | 59.7 (46.1, 72.0) | |
| Borderline | 2.3 (1.8, 2.8) | 3.9 (2.5, 6.1) | 3.4 (0.9, 11.7) | 5.4 (1.2, 21.3) | 0 (0, 0) | |
| No | 89.6 (88.6, 90.5) | 72.4 (68.3, 76.2) | 51.4 (38.8, 63.9) | 47.4 (23.2, 72.8) | 40.3 (28.0, 53.9) | |
| Missing | 0 (0.0, 0.1) | 0.1 (0.0, 0.4) | 0 (0, 0) | 0 (0, 0) | 0 (0, 0) | |
| Arthritis | <0.001 | |||||
| Yes | 22.8 (21.2, 24.4) | 53.0 (48.8, 57.1) | 48.0 (35.9, 60.4) | 56.9 (33.6, 77.5) | 49.5 (38.3, 60.8) | |
| No | 77.1 (75.4, 78.7) | 47.0 (42.9, 51.2) | 51.5 (39.2, 63.6) | 43.1 (22.5, 66.4) | 47.5 (35.5, 59.8) | |
| Missing | 0.2 (0.1, 0.3) | 0 (0.0, 0.3) | 0.5 (0.1, 3.5) | 0 (0, 0) | 3.0 (0.4, 19.2) | |
| Health Insurance | <0.001 | |||||
| Private | 53.3 (50.4, 56.3) | 23.4 (18.0, 29.8) | 23.3 (12.6, 39.2) | 6.4 (1.7, 21.5) | 5.2 (2.0, 12.6) | |
| Medicare | 12.6 (11.5, 13.7) | 59 (52.9, 64.9) | 55.9 (41.2, 69.7) | 67.4 (49.0, 81.6) | 55.9 (40.0, 70.6) | |
| Medicaid | 5.5 (4.6, 6.7) | 4.3 (2.7, 7.0) | 8.2 (4.0, 15.9) | 2.0 (0.4, 8.7) | 5.6 (1.7, 16.6) | |
| Dual eligible | 1.3 (1.0, 1.8) | 5.2 (3.6, 7.5) | 3.4 (1.3, 8.2) | 18.4 (8.0, 36.7) | 22.1 (10.4, 41.1) | |
| Other public | 6.7 (5.8, 7.8) | 5.0 (3.5, 7.1) | 2.8 (1.0, 7.8) | 1.9 (0.2, 13.3) | 8.8 (3.0, 23.4) | |
| Missing | 20.5 (18.6, 22.6) | 3.1 (1.7, 5.4) | 6.4 (2.3, 16.6) | 3.9 (0.7, 19.9) | 2.4 (0.5, 10.9) | |
Note: Values are percentages, with 95% confidence interval (CI) in parentheses; missing includes refused or don’t know. All percentages were calculated within eGFR/supplement awareness categories, taking into account complex survey design. Raw numbers are omitted because they do not directly correspond to the percentages due to weighting. G1–G2 refers to all people with eGFR ≥ 60 mL/min/1.73 m2 including normal kidney function.
Abbreviations and definitions: CKD, chronic kidney disease; CVD, cardiovascular disease (a composite of self-reported coronary heart disease, angina, heart attack, heart failure, and/or stroke); G, glomerular filtration rate (GFR) category, based on estimated GFR calculated using the CKD-EPI equation (G1, G2, G3, G4, and G5 corresponding to estimated GFRs of >90, 60–89, 30–59, 15–29, and <15 mL/min/1.73 m2, respectively); HS, high school.
With or without additional low-risk supplement(s).
The proportions of people taking supplements based on their own decision versus per recommendation of health care provider were calculated from among those taking (any) supplements only, not the base cohort.
Supplement use was determined by the response to “Have you ever used or taken any dietary supplements in the past month?” Independent variables, ascertained by survey response, included age, sex, race/ethnicity, education level, insurance status, diabetes, arthritis, hypertension, and CVD. CKD awareness was assessed by response to “Have you ever been told that you had weak or failing kidneys?” (Item S2).
Descriptive summaries were estimated using complex survey design (SVY TABOUT, Stata 14; StataCorp LP). Logistic regression was performed to calculate the predicted probabilities of supplement use by GFR/CKD-awareness categories and adjusting for the independent variables above. A missing indicator was created for missing, don’t know, and refused responses. Analyses used NHANES-recommended methods to produce nationally representative estimates (SURVEYFREQ, SAS, version 9.4; SAS Institute Inc).
The cohort included 10,005 individuals, of whom 6.8% and 0.6% were categorized as G3 and G4–G5, respectively (Table 1). There was greater CKD awareness with lower GFRs: 10.3% (95% CI, 8.0%–13.0%) with G3 and 63.1% (95% CI, 48.9%–75.4%) with G4–G5.
Overall, 27.8% (95% CI, 24.9%–30.9%) of US adults reported taking 1 or more dietary supplement, translating to 58.7 (95% CI, 49.7–67.8) million people; 2.6% (95% CI, 2.1%–3.1%) took high-risk supplements. Supplements used most frequently are summarized in Tables 2 and S2. There was no difference in crude rates of any supplement use across GFR/CKD-awareness categories, though there was greater use of low-risk supplements with worse GFR (Table 1). CKD-aware patients with G4 to G5 took supplements primarily related to kidney health, rather than general vitamins or antacids (Table S2). No patients with G4 to G5 and few CKD-aware patients with G3 took high-risk supplements (Table 1). Supplement use was higher with CKD than among people with diabetes, hypertension, arthritis, or CVD (Table S3).
Table 2.
Specific High-Risk Supplements Used Most Frequently by US Adults
| Weighted % | |
|---|---|
| G1–G2 | |
| Flaxseed oil | 16.1 |
| Evening primrose oil | 5.9 |
| GNC Men’s Mega Men Sport creatine precursor B | 4.6 |
| Odorless garlic (bulb) | 4.6 |
| Milk thistle seed extract | 4.4 |
| Milk thistle (seed) | 3.9 |
| Garlic | 2.2 |
| Dandelion leaf extract | 2.1 |
| Borage seed oil | 2.1 |
| Garlic (bulb) | 2.1 |
| G3, CKD-unaware | |
| Flaxseed oil | 1.3 |
| Turmeric extract (root) | 0.6 |
| Dandelion root | 0.6 |
| Concentrated garlic bulb | 0.5 |
| Horsetail stem extract | 0.4 |
| Sunflower oil | 0.4 |
| Turmeric concentrate (rhizome) | 0.4 |
| Thisilyn milk thistle extract (seed) | 0.4 |
| Bitter melon extract (fruit) | 0.3 |
| Papaya juice powder | 0.3 |
| G3, CKD-aware | |
| Milk thistle seed extract | 0.4 |
| Papaya fruit | 0.3 |
Note: The top 10 high-risk supplements, as defined by the NKF,5 in each GFR/awareness category are shown as a weighted percentage of people in that group. G1–G2 refers to all people with eGFR ≥ 60 mL/min/1.73 m2 including normal kidney function. There were no individuals with G4 to G5 (CKD unaware or aware) taking high-risk supplements.
The majority of people took supplements on their own (vs being advised to do so by a health care provider), except people with G4–G5 were less likely to take supplements on their own if they were CKD aware (13.8% [95% CI, 5.7%–29.9%] vs 34.5% [95% CI, 13.6%–64.0%] for unaware; Table 1). Reasons for supplement use were similar across GFR categories, with the exception of taking supplements for healthy joints/arthritis (most common with G3; none with G4–G5), anemia (most common among CKD-unaware people with G4–G5), kidney/bladder health (most common among CKD-aware people with G4–G5), or for other unspecified reasons (most common among CKD-aware people with G4–G5; Table S4).
Adjusted probabilities of supplement use were similar irrespective of GFR/CKD-awareness categories (Fig S1). Factors with greater odds of supplement use were age older than 44 years, female sex, and higher education (Table S5). People of Hispanic ethnicity were less likely to take supplements. There was no association with insurance type or presence of examined comorbid conditions.
Nearly one-third of US adults, or more than 58.7 million people, reported taking dietary supplements in 2011 to 2014. This is likely an underestimate because data were self-reported. Whether health care providers warned against supplement use is also unknown. Participants with decreased GFRs shared similar exposures to supplements overall, but no one with G4–G5 took high-risk supplements and they were less likely to take low-risk supplements on their own (as opposed to a health care provider’s recommendation). Nonetheless, supplements not specified by the NKF as high risk may still be harmful. Further research into supplement safety is needed, and people with CKD may benefit from periodic medication/supplement review and education on the acceptability and safety of supplements, particularly given the high prevalence of supplement use based on personal decision and not clinician’s advice.3
Supplementary Material
Acknowledgements:
We thank Molly M. Jeffery, PhD, and Rachel Giblon (both Mayo Clinic) for statistical guidance.
Support: This work was supported by the Mayo Clinic Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery. Ms Kurani is supported by NIH CTSA no. TL1TR002380. Drs McCoy and Hickson are supported by NIH/NIDDK grants K23 DK114497 and K23 DK109134, respectively. Dr Thorsteinsdottir is supported by NIH/NIA grant K23 AG051679. The funders had no role in study design, data collection, analysis, reporting, or the decision to submit for publication.
Footnotes
Financial Disclosure: The authors declare that they have no relevant financial interests.
Publisher's Disclaimer: Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
References
- 1.Marcus DM, Grollman AP. The consequences of ineffective regulation of dietary supplements. Arch Intern Med. 2012;172(13): 1035–1036. [DOI] [PubMed] [Google Scholar]
- 2.Geller AI, Shehab N, Weidle NJ, et al. Emergency department visits for adverse events related to dietary supplements. N Engl J Med. 2015;373(16):1531–1540. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Inker LA, Astor BC, Fox CH, et al. KDOQI US commentary on the 2012 KDIGO clinical practice guideline for the evaluation and management of CKD. Am J Kidney Dis. 2014;63(5):713–735. [DOI] [PubMed] [Google Scholar]
- 4.Grubbs V, Plantinga LC, Tuot DS, et al. Americans’ use of dietary supplements that are potentially harmful in CKD. Am J Kidney Dis. 2013;61(5):739–747. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.NKF. Herbal supplements and kidney disease. 2015. https://www.kidney.org/atoz/content/herbalsupp. Accessed March 15, 2019.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.