Table 2.

ANA prevalence in the U.S. female population ages 20 years and older in NHANES (1999–2004) by reproductive history, and PORs

PremenopausalPostmenopausal
WeightedPORWeightedPOR
NN ANA+a% ANA+ (95% CI)(95% CI)bNaN ANA+a% ANA+ (95% CI)(95% CI)b
Age at menarche, y
 <1029316.8 (5.0–43.6)0.50 (0.16–1.54)70
 10–123446818.6 (14.3–23.9)(Reference)2954715.6 (11.3–21.2)(Reference)
 13–153717020.0 (15.3–25.6)0.97 (0.65–1.45)3646017.3 (12.9–22.7)1.10 (0.73–1.66)
 16–2038917.1 (7.3–35.1)1.10 (0.45–2.69)541935.5 (21.7–52.4)3.04 (1.56–5.90)
  Pc0.9540.019
Ever had a live birth
 No (nulliparous)1962511.1 (7.5–16.2)(Reference)881414.3 (8.4–23.3)(Reference)
 Yes (parous)58612522.4 (18.6–26.7)2.01 (1.20–3.36)63111218.0 (14.5–22.2)1.03 (0.53–2.02)
Pc<0.0010.399
Number of births
 11402821.4 (13.9–31.4)(Reference)811419.4 (10.7–32.6)(Reference)
 22105022.1 (17.0–28.3)1.20 (0.67–2.14)1552716.5 (9.9–26.3)1.03 (0.43–2.47)
 31312620.9 (12.7–32.3)0.99 (0.47–2.07)1402515.1 (9.6–22.9)1.10 (0.47–2.58)
 4+1052127.8 (16.2–43.5)1.10 (0.50–2.40)2554621.3 (14.6–29.9)1.24 (0.52–2.92)
  Pc0.8480.9480.6190.579
Years since last birth
 ≤11012224.3 (15.7–35.7)(Reference)
 >1–51102120.0 (13.2–29.1)0.67 (0.33–1.36)
 >5–101172825.1 (16.4–36.3)1.09 (0.55–2.15)8235.1 (8.2–76.6)(Reference)
 >10–151102223.4 (13.9–36.8)0.82 (0.43–1.57)20519.8 (7.0–45.0)0.84 (0.06–11.8)
 >15–20781519.4 (10.3–33.4)0.76 (0.27–2.13)31413.6 (4.6–34.2)0.94 (0.08–11.3)
 >20701721.3 (11.7–35.4)1.03 (0.42–2.51)56910118.0 (13.8–23.2)0.81 (0.08–7.94)
  Pc0.9670.9340.8200.889
Age at menopause (natural; y)
 <3517522.3 (8.6–46.7)2.12 (0.66–6.79)
 35–447865.0 (1.5–15.2)0.55 (0.18–1.67)
 45–491122421.6 (13.9–32.0)1.58 (0.70–3.58)
 50–541862916.1 (8.9–27.3)(Reference)
 55+691220.9 (11.6–34.8)1.06 (0.48–2.34)
  Pc0.0380.554

Abbreviation: +, positive.

  • aN is the number of subjects within the sample, not an estimated count for the U.S. population; % is adjusted for sample weights; individual totals for each variable may vary due to the number of missing or “don't know” responses.

  • bPOR and 95% CIs were estimated by survey logistic regression, adjusting for age, race/ethnicity, and PIR. Values were not calculated (—) where numerically infeasible.

  • cP value for the Wald χ2 test for general association (missing/don't knows excluded), or linear trend test in regression models for POR.