Skip to main content

Table 3 Associationa of density estimates (percent density, dense area, and non-dense area) and categoriesb of the four acquisition parameters

From: The influence of mammogram acquisition on the mammographic density and breast cancer association in the mayo mammography health study cohort

  Spearman correlation coefficient(P value) Category 1 Category 2 Category 3   
    22-25 26-27 28-32   
  Cases
(n = 254)c
Subcohort
(n = 1,910)c
Mean SD Mean SD Mean SD   
   Percent density 0.09 (0.15)d 0.04 (0.07)d 17.1 13.8 17.3 14.8 19.0 14.3   
   Dense area 0.28 (< 0.001)d 0.16 (< 0.001)d 2,070 1,603 2,404 1,864 2,987 2,238   
   Non-dense area 0.20 (0.002)d 0.16 (< 0.001)d 12,349 6,393 14,386 7,512 15,405 8,289   
    Quartile 1 Quartile 2 Quartile 3 Quartile 4
mAs    9-126 127-173 174-194 194-371
    Mean SD Mean SD Mean SD Mean SD
   Percent density -0.03 (0.60) -0.06 (0.008) 18.4 14.3 18.4 14.9 17.1 13.9 16.3 13.3
   Dense area 0.14 (0.02) 0.09 (< 0.001) 1,958 1,481 2,432 1,843 2,415 1,867 2,534 2,086
   Non-dense area 0.24 (< 0.001) 0.25 (< 0.001) 10,906 6,034 13,412 7,284 14,214 7,091 15,404 7,571
    Quartile 1 Quartile 2 Quartile 3 Quartile 4
Thickness, mm    6-45 46-54 55-62 62-94
   Percent density -0.25 (< 0.001) -0.27 (< 0.001) 23.3 16.5 17.7 13.2 15.7 12.7 13.0 11.3
   Dense area -0.01 (0.89) -0.12 (< 0.001) 2,512 1,846 2,425 1,857 2,261 1,798 2,113 1,844
   Non-dense area 0.41 (< 0.001) 0.35 (< 0.001) 10,211 6,210 13,604 7,142 14,152 6,632 16,236 7,494
    Quartile 1 Quartile 2 Quartile 3 Quartile 4
Compression force, pounds   6.6-21 22-24 25-28 28-49
   Percent density -0.14 (0.02) -0.16 (< 0.001) 20.5 15.1 18 14.5 16.9 14.1 14.2 11.6
   Dense area -0.01 (0.88) -0.02 (0.39) 2,360 1,761 2,258 1,695 2,407 1,999 2,283 1,880
   Non-dense area 0.26 (< 0.001) 0.31 (< 0.001) 10,859 5,945 12,788 6,749 14,153 6,759 16,536 8,160
  1. aMean and SD of density by categories are estimated using the subcohort only. Correlation coefficients estimated on cases and subcohort separately. bQuartiles were used where appropriate. The distribution of voltage peak (kVp) was highly skewed and was categorized into three levels. cOwing to missing acquisition parameters, 18 cases and 47 controls that were included in the analyses depicted in Table 2 are not included here. dCorrelation coefficients for kVp are based on the three categories and not continuous measures. mAs, milliampere-seconds; SD, standard deviation.