Skip to main content

Table 4 Population attributable risk (PAR) and population attributable risk percentage (PAR%) for high serum methylmalonic acid (MMA) in the National Health and Nutrition Examination Surveys (NHANES), 1999-20041

From: Population prevalence, attributable risk, and attributable risk percentage for high methylmalonic acid concentrations in the post-folic acid fortification period in the US

Characteristic

Cases2

PAR3

PAR%4

Race-ethnicity5

   

   Non-Hispanic white (n = 8170)

356

0.774

24.8

   Non non-Hispanic white (n = 10399)6, 7

230

--

--

Age5

   

   < 60 y (n = 14142)6

204

--

--

   60 y (n = 4427)

382

1.262

40.5

Supplement use8, 9

   

   Yes (n = 7681)6

252

--

--

   No (n = 10888)

334

0.368

11.8

Serum creatinine8, 10

   

   < 130 μmol/L (n = 18238)6

477

--

--

   130 μmol/L (n = 331)

109

0.415

13.3

Serum vitamin B-128, 11

   

   < 148 pmol/L (n = 349)

111

0.506

16.2

   148 pmol/L (n = 18220)6

475

--

--

  1. 1 n = 18569; NHANES 1999-2000, 2001-2002, and 2003-2004 were combined into one analytic data set, 1999-2004. NHANESs, 1999-2004 were conducted after the folic acid fortification commenced. PAR and PAR% for sex were not presented because sex variable was not significantly related to high serum MMA in the logistic regression model (P = 0.98).
  2. 2Number of cases with serum MMA > 350 nmol/L
  3. 3Prevalence of a condition/disease in the population due to the presence of risk factor or prevalence of a condition/disease in the population that would be reduced if risk factor was removed. PAR = (Prevalence of high MMATotal sample × 100) - (Prevalence of high serum MMAReferent group × 100). Prevalence of high serum MMATotal sample = Cases of high serum MMATotal sample/Sample sizeTotal. Prevalence of high serum MMAReferent group = Cases of high serum MMAReferent group/Sample sizeReferent group. PAR and PAR% were calculated based on weighted sample. Weighted sample was used to account for differential probabilities of selection and adjustments for non-coverage and non-response bias.
  4. 4Percent of prevalence of a condition/disease in the population due to presence of risk factor or percent of prevalence of a condition/disease in the population that would be reduced if risk factor was removed. PAR% was calculated based on weighted sample size. PAR% = (Prevalence of high serum MMATotal sample - Prevalence of high serum MMAReferent group ÷ Prevalence of high serum MMATotal sample) x100
  5. 5Non-modifiable risk factor for high serum MMA
  6. 6Referent group
  7. 7In order to achieve a dichotomous variable for race-ethnicity, non-Hispanic black and Mexican American/Hispanic were combined into one category, non-non-Hispanic white (non-white). Cases of serum MMA > 350 nmol/L for non-Hispanic black and Mexican American/Hispanic were 65 and 165, respectively.
  8. 8Modifiable risk factor for high serum MMA
  9. 9Persons who took vitamin/mineral supplements 1 month prior to the survey
  10. 10Kidney dysfunction was defined as having serum creatinine ≥ 130 μmol/L
  11. 11Vitamin B-12 deficiency was defined as having serum vitamin B-12 < 148 pmol/L
Back to article page

Nutrition & Metabolism

ISSN: 1743-7075

Contact us