Prevalence and trends in the use of any supplement, MVMM, any vitamin, and any mineral supplement among US adults aged 20 years or older with diabetes from 1999 to 2014* †
| 30-day prevalence of dietary supplement use, Weighted % (Weighted 95% CI)‡ | P value for trend | 2013–2014 vs 1999–2000 | |||||||||
| 1999–2000 (n=577) | 2001–2002 (n=631) | 2003–2004 (n=656) | 2005–2006 (n=687) | 2007–2008 (n=1089) | 2009–2010 (n=1072) | 2011–2012 (n=985) | 2013–2014 (n=991) | Ratio | Difference | ||
| Any supplement | 52 (45 to 59) | 50 (45 to 55) | 56 (50 to 63) | 51 (45 to 57) | 51 (47 to 55) | 55 (49 to 61) | 53 (48 to 58) | 58 (54 to 62) | 0.08 | 1.1 (1.0 to 1.3) | 6.1 (−1.4 to 14) |
| Excluding MVMM§ | 17 (12 to 21) | 15 (11 to 18) | 22 (18 to 25) | 13 (8 to 17) | 19 (16 to 22) | 23 (18 to 29) | 26 (21 to 31) | 26 (22 to 30) | <0.0001 | 1.6 (1.2 to 2.2) | 9.7 (3.8 to 16) |
| MVMM¶ | 36 (28 to 43) | 36 (31 to 41) | 35 (29 to 40) | 38 (32 to 44) | 32 (28 to 36) | 31 (28 to 35) | 27 (23 to 31) | 32 (28 to 36) | 0.006 | 0.9 (0.7 to 1.1) | −3.6 (−12 to 4.4) |
| Any vitamin | 47 (40 to 54) | 45 (39 to 50) | 48 (42 to 53) | 46 (40 to 52) | 46 (42 to 51) | 47 (42 to 51) | 49 (44 to 54) | 52 (49 to 56) | 0.03 | 1.1 (1.0 to 1.3) | 5.8 (−1.8 to 14) |
| Excluding MVMM | 11 (7 to 15) | 9 (6 to 12) | 13 (9 to 16) | 8 (4 to 11) | 14 (11 to 17) | 15 (12 to 20) | 22 (18 to 26) | 20 (17 to 24) | <0.0001 | 1.9 (1.3 to 2.7) | 9.5 (4.6 to 14) |
| Any mineral | 47 (41 to 53) | 51 (44 to 58) | 54 (46 to 61) | 53 (46 to 60) | 45 (41 to 50) | 49 (45 to 53) | 44 (39 to 48) | 51 (46 to 55) | 0.23 | 1.1 (0.9 to 1.3) | 3.7 (−3.6 to 11) |
| Excluding MVMM | 13 (9 to 17) | 16 (12 to 21) | 19 (15 to 24) | 16 (11 to 20) | 14 (10 to 17) | 18 (15 to 22) | 17 (13 to 21) | 19 (15 to 23) | 0.15 | 1.4 (0.99 to 2.1) | 5.8 (0.2 to 12) |
*Results are presented for both of the overall use and use when excluding MVMM.
†Survey-weighted logistic regression was used to calculate p for linear trend across waves. A decrease corresponds to a ratio <1 (or a difference <0) and p for trend <0.05 when comparing 2013–2014 to 1999–2000. An increase represents a ratio >1 (or a difference >0) and p for trend <0.05. Stable represents a p for trend ≥0.05.
‡Sample weights were considered for analyses to account for complex survey design. All relative SEs are <30%, which is consistent with National Health and Nutrition Examination Survey (NHANES) reporting guidelines.
§The trend markedly varied for overall use and use when excluding MVMM.
¶MVMM is defined as containing ≥10 vitamins and/or minerals.
MVMM, multivitamin/multimineral.