Estimation of trends in serum and RBC folate in the U.S. population from pre- to postfortification using assay-adjusted data from the NHANES 1988-2010

Abstract

The NHANES has monitored folate status of the U.S. population from prefortification (1988-1994) to postfortification (1999-2010) by measuring serum and RBC folate concentrations. The Bio-Rad radioassay (BR) was used from 1988 to 2006, and the microbiologic assay (MBA) was used from 2007 to 2010. The MBA produces higher concentrations than the BR and is considered to be more accurate. Thus, to bridge assay differences and to examine folate trends over time, we adjusted the BR results to be comparable to the MBA results. Postfortification, assay-adjusted serum and RBC folate concentrations were 2.5 times and 1.5 times prefortification concentrations, respectively, and showed a significant linear trend (P < 0.001) to slightly lower concentrations during 1999-2010. The postfortification prevalence of low serum (<10 nmol/L) or RBC (<340 nmol/L) folate concentrations was ≤ 1%, regardless of demographic subgroup, compared with 24% for serum folate and 3.5% for RBC folate prefortification, with substantial variation among demographic subgroups. The central 95% reference intervals for serum and RBC folate varied by demographic subgroup during both pre- and postfortification periods. Age and dietary supplement use had the greatest effects on prevalence estimates of low folate concentrations during the prefortification period. In summary, the MBA-equivalent blood folate concentrations in the U.S. population showed first a sharp increase from pre- to postfortification, then showed a slight decrease (17% for serum and 12% for RBC folate) during the 12-y postfortification period. The MBA-equivalent pre- and postfortification reference concentrations will inform countries that plan folic acid fortification or that need to evaluate its impact.

FIGURE 1

Selected percentiles for pre- and postfortification serum (A) and RBC (B) folate concentrations by survey period (NHANES 1988–2010). Concentrations measured from 1988–1994 and from 1999–2006 by the Bio-Rad radioassay were adjusted to make them comparable to the 2007–2010 concentrations measured by microbiologic assay (13, 16, 17). Error bars represent 95% CIs. Sample sizes (n) for serum folate were as follows: 23,359 (1988–1994), 7411 (1999–2000), 8242 (2001–2002), 7692 (2003–2004), 7639 (2005–2006), 7705 (2007–2008), and 8184 (2009–2010). Sample sizes (n) for RBC folate were as follows: 22,846 (1988–1994), 7393 (1999–2000), 8220 (2001–2002), 7618 (2003–2004), 7578 (2005–2006), 7728 (2007–2008), and 8222 (2009–2010). pctl, percentile.

FIGURE 2

Pre- and postfortification 95th percentile serum (A) and RBC (B) folate concentrations for users (solid line) and nonusers (dashed line) of dietary supplements by survey period (NHANES 1988–2010). Concentrations measured from 1988 to 1994 and from 1999 to 2006 by the Bio-Rad radioassay were adjusted to make them comparable to the 2007–2010 concentrations measured by microbiologic assay (13, 16, 17). Error bars represent 95% CIs. Sample sizes (n) were the same as shown in Figure 1. pctl, percentile.

FIGURE 3

Frequency distribution curves for serum (A) and RBC (B) folate concentrations for prefortification (1988–1994), early postfortification (1999–2004), and late postfortification (2005–2010) time periods (NHANES 1988–2010). Concentrations measured from 1988–2006 by the Bio-Rad radioassay were adjusted by regressing them to be comparable to the 2007–2010 concentrations measured by microbiologic assay (13, 16, 17). The vertical dashed lines represent the 10-nmol/L low serum (A) and 340-nmol/L low RBC (B) folate cutoffs. Sample sizes (n) for serum folate were as follows: 23,359 (1988–1994), 23,345 (1999–2004), and 23,528 (2005–2010). Sample sizes (n) for RBC folate were as follows: 22,846 (1988–1994), 23,231 (1999–2004), and 23,528 (2005–2010).