The current round of model building uses the collapsed MJ exposure variable
This model includes
use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr i.EDUC_cat i.race_eth indfmpir ridageyr,rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
i.EDUC_cat _IEDUC_cat_1-5 (naturally coded; _IEDUC_cat_1 omitted)
i.race_eth _Irace_eth_1-4 (naturally coded; _Irace_eth_1 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 68,611
Number of PSUs = 214 Population size = 298,035,877
Subpop. no. obs = 19,441
Subpop. size = 133,015,285
Design df = 109
F( 36, 74) = 45.02
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.081166 .0900856 0.94 0.351 .9165826 1.275302
_IMJ2_2 | 1.351174 .1727591 2.35 0.020 1.048712 1.74087
gndr | 2.254758 .1120625 16.36 0.000 2.043243 2.48817
_IEDUC_cat_2 | 1.155943 .1317258 1.27 0.206 .9222501 1.448853
_IEDUC_cat_3 | .976727 .1205645 -0.19 0.849 .7647574 1.247449
_IEDUC_cat_4 | 1.038433 .1273142 0.31 0.759 .814419 1.324065
_IEDUC_cat_5 | .7381725 .0899919 -2.49 0.014 .5797241 .9399274
_Irace_eth_2 | 1.252365 .0695511 4.05 0.000 1.121832 1.398085
_Irace_eth_3 | .9957338 .0752287 -0.06 0.955 .8572592 1.156576
_Irace_eth_4 | .9073141 .0639002 -1.38 0.170 .7891079 1.043227
indfmpir | 1.018176 .0192621 0.95 0.343 .9807059 1.057078
ridageyr | 1.028041 .0030318 9.38 0.000 1.022049 1.034067
_cons | .071281 .0125789 -14.97 0.000 .0502431 .1011278
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | 1.002551 .0850181 0.03 0.976 .8474471 1.186042
_IMJ2_2 | .9995363 .1156582 -0.00 0.997 .7946918 1.257183
gndr | 2.435001 .1257827 17.23 0.000 2.19804 2.697506
_IEDUC_cat_2 | 1.075327 .1304962 0.60 0.551 .8454413 1.367722
_IEDUC_cat_3 | 1.198871 .1313026 1.66 0.101 .9649411 1.489513
_IEDUC_cat_4 | 1.194915 .1367742 1.56 0.123 .9523829 1.499209
_IEDUC_cat_5 | .8448599 .1001706 -1.42 0.158 .6679273 1.068662
_Irace_eth_2 | 1.356215 .0935342 4.42 0.000 1.182945 1.554864
_Irace_eth_3 | .8784635 .0642784 -1.77 0.079 .7598727 1.015562
_Irace_eth_4 | .9112657 .0655358 -1.29 0.199 .7902083 1.050869
indfmpir | 1.006747 .0172752 0.39 0.696 .973084 1.041575
ridageyr | 1.048406 .0025344 19.56 0.000 1.043395 1.053442
_cons | .0360908 .005909 -20.29 0.000 .0260896 .0499257
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | 1.010595 .1083398 0.10 0.922 .8171472 1.249838
_IMJ2_2 | .9817426 .1704985 -0.11 0.916 .6958418 1.385112
gndr | 2.799287 .1927263 14.95 0.000 2.442226 3.208553
_IEDUC_cat_2 | 1.187676 .177494 1.15 0.252 .8832035 1.597111
_IEDUC_cat_3 | 1.105081 .1608308 0.69 0.494 .8281739 1.474574
_IEDUC_cat_4 | 1.151613 .1893584 0.86 0.392 .8313294 1.595291
_IEDUC_cat_5 | .72398 .1324902 -1.76 0.080 .5037391 1.040513
_Irace_eth_2 | 2.486514 .1937666 11.69 0.000 2.130662 2.901798
_Irace_eth_3 | 1.001051 .0924942 0.01 0.991 .8335364 1.20223
_Irace_eth_4 | 1.13396 .1153303 1.24 0.219 .9269448 1.387209
indfmpir | .9452285 .0212171 -2.51 0.014 .9040985 .9882296
ridageyr | 1.088587 .0030177 30.62 0.000 1.082622 1.094585
_cons | .003742 .0006595 -31.71 0.000 .0026387 .0053065
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.Store full model RRs in a matrix for future:
data.0 <- c(1.081166 ,1.351174,1.002551 ,.9995363,1.010595,.9817426)
RR0 <- matrix(data.0, nrow = 3, ncol = 2, byrow = TRUE)
RR0 [,1] [,2]
[1,] 1.081166 1.3511740
[2,] 1.002551 0.9995363
[3,] 1.010595 0.9817426As gender is a political variable, it will not be considered for removal.
use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr i.race_eth indfmpir ridageyr,rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
i.race_eth _Irace_eth_1-4 (naturally coded; _Irace_eth_1 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 68,616
Number of PSUs = 214 Population size = 298,070,653
Subpop. no. obs = 19,446
Subpop. size = 133,050,061
Design df = 109
F( 24, 86) = 75.53
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.114028 .0923798 1.30 0.196 .9451893 1.313027
_IMJ2_2 | 1.406301 .1799681 2.66 0.009 1.091251 1.812307
gndr | 2.276105 .1132495 16.53 0.000 2.062361 2.512002
_Irace_eth_2 | 1.272604 .0704743 4.35 0.000 1.140319 1.420235
_Irace_eth_3 | 1.038213 .0726173 0.54 0.593 .9038184 1.192592
_Irace_eth_4 | .892116 .0637167 -1.60 0.113 .7743624 1.027776
indfmpir | .9802786 .0172675 -1.13 0.261 .9466455 1.015107
ridageyr | 1.028201 .002957 9.67 0.000 1.022357 1.034079
_cons | .0730937 .0088738 -21.55 0.000 .0574621 .0929776
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | 1.030628 .087858 0.35 0.724 .8704119 1.220335
_IMJ2_2 | 1.038105 .1213104 0.32 0.750 .8234843 1.308661
gndr | 2.450987 .1241066 17.70 0.000 2.216952 2.709728
_Irace_eth_2 | 1.371535 .0952682 4.55 0.000 1.195137 1.573968
_Irace_eth_3 | .8777422 .0619099 -1.85 0.067 .7632294 1.009436
_Irace_eth_4 | .8865064 .062838 -1.70 0.092 .7703162 1.020222
indfmpir | .9769956 .0146415 -1.55 0.123 .9484034 1.00645
ridageyr | 1.048482 .0025674 19.33 0.000 1.043406 1.053583
_cons | .0414227 .0049295 -26.75 0.000 .0327194 .0524411
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | 1.049833 .1107816 0.46 0.646 .8517074 1.294047
_IMJ2_2 | 1.027226 .181276 0.15 0.879 .7240473 1.457354
gndr | 2.826599 .1947695 15.08 0.000 2.465771 3.240227
_Irace_eth_2 | 2.532578 .19621 11.99 0.000 2.172082 2.952905
_Irace_eth_3 | 1.025286 .0919623 0.28 0.781 .8583018 1.224758
_Irace_eth_4 | 1.108953 .1104565 1.04 0.301 .9102858 1.350978
indfmpir | .9028455 .0175096 -5.27 0.000 .8688005 .9382246
ridageyr | 1.089011 .0030916 30.04 0.000 1.0829 1.095155
_cons | .0041487 .0005716 -39.81 0.000 .0031573 .0054515
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.E <- c( 1.114028 ,1.406301 ,1.030628 ,1.038105 ,1.049833 ,1.027226 )
RR.E <- matrix(data.E, nrow = 3, ncol = 2, byrow = TRUE)
RR.E [,1] [,2]
[1,] 1.114028 1.406301
[2,] 1.030628 1.038105
[3,] 1.049833 1.027226D_RR <- (RR0-RR.E)/RR0
100*D_RR [,1] [,2]
[1,] -3.039496 -4.079933
[2,] -2.800556 -3.858659
[3,] -3.882663 -4.632925100*mean(abs(D_RR))[1] 3.715706use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr i.EDUC_cat indfmpir ridageyr,rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
i.EDUC_cat _IEDUC_cat_1-5 (naturally coded; _IEDUC_cat_1 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 68,611
Number of PSUs = 214 Population size = 298,035,877
Subpop. no. obs = 19,441
Subpop. size = 133,015,285
Design df = 109
F( 27, 83) = 50.92
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.097158 .0905562 1.12 0.264 .9315897 1.292152
_IMJ2_2 | 1.368162 .173845 2.47 0.015 1.063569 1.759987
gndr | 2.240263 .1112499 16.24 0.000 2.030273 2.471973
_IEDUC_cat_2 | 1.195458 .1333261 1.60 0.112 .958377 1.491187
_IEDUC_cat_3 | 1.010424 .1205671 0.09 0.931 .7976184 1.280006
_IEDUC_cat_4 | 1.073598 .1243135 0.61 0.541 .8534409 1.350549
_IEDUC_cat_5 | .7568343 .0870842 -2.42 0.017 .6025024 .9506986
indfmpir | 1.014387 .0197172 0.73 0.464 .9760511 1.054228
ridageyr | 1.028115 .00298 9.57 0.000 1.022226 1.034038
_cons | .0705759 .0112729 -16.60 0.000 .0514246 .0968597
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | 1.029004 .086671 0.34 0.735 .8707979 1.215954
_IMJ2_2 | 1.022993 .1175023 0.20 0.843 .8147132 1.284518
gndr | 2.409643 .1242765 17.05 0.000 2.175502 2.668984
_IEDUC_cat_2 | 1.170981 .1417908 1.30 0.195 .9211342 1.488595
_IEDUC_cat_3 | 1.31705 .140536 2.58 0.011 1.065994 1.627232
_IEDUC_cat_4 | 1.316915 .1457929 2.49 0.014 1.057463 1.640025
_IEDUC_cat_5 | .9242253 .1071901 -0.68 0.498 .734427 1.163073
indfmpir | 1.003123 .0164963 0.19 0.850 .9709549 1.036357
ridageyr | 1.048673 .002499 19.94 0.000 1.043732 1.053638
_cons | .033369 .0048066 -23.60 0.000 .0250816 .0443946
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | 1.047966 .1097063 0.45 0.655 .8516059 1.289601
_IMJ2_2 | .9823627 .1725749 -0.10 0.920 .6935212 1.391502
gndr | 2.721885 .1877448 14.52 0.000 2.374095 3.120623
_IEDUC_cat_2 | 1.352227 .1948216 2.09 0.039 1.016334 1.799132
_IEDUC_cat_3 | 1.232868 .1688532 1.53 0.129 .9397832 1.617357
_IEDUC_cat_4 | 1.292752 .1990743 1.67 0.098 .9527195 1.754146
_IEDUC_cat_5 | .8005932 .1390504 -1.28 0.203 .5674295 1.129567
indfmpir | .9183684 .020075 -3.90 0.000 .8794301 .9590308
ridageyr | 1.08761 .0030514 29.93 0.000 1.081579 1.093674
_cons | .0044405 .0007085 -33.95 0.000 .0032367 .0060922
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.R <-c(1.097158 ,1.368162 ,1.029004 ,1.022993,1.047966 ,.9823627)
RR.R <- matrix(data.R, nrow = 3, ncol = 2, byrow = TRUE)
RR.R [,1] [,2]
[1,] 1.097158 1.3681620
[2,] 1.029004 1.0229930
[3,] 1.047966 0.9823627D_RR <- (RR0-RR.R)/RR0
100*D_RR [,1] [,2]
[1,] -1.479144 -1.2572770
[2,] -2.638569 -2.3467582
[3,] -3.697921 -0.0631632100*mean(abs(D_RR))[1] 1.913805use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr i.EDUC_cat i.race_eth ridageyr,rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
i.EDUC_cat _IEDUC_cat_1-5 (naturally coded; _IEDUC_cat_1 omitted)
i.race_eth _Irace_eth_1-4 (naturally coded; _Irace_eth_1 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 70,182
Number of PSUs = 214 Population size = 306,497,065
Subpop. no. obs = 21,012
Subpop. size = 141,476,473
Design df = 109
F( 33, 77) = 51.18
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.100949 .0897144 1.18 0.240 .9367537 1.293924
_IMJ2_2 | 1.245111 .1567605 1.74 0.084 .9701481 1.598005
gndr | 2.233306 .1054658 17.01 0.000 2.03376 2.452431
_IEDUC_cat_2 | 1.147835 .1247987 1.27 0.207 .9253228 1.423854
_IEDUC_cat_3 | 1.033004 .1174961 0.29 0.776 .824513 1.294215
_IEDUC_cat_4 | 1.071073 .1239685 0.59 0.554 .8515162 1.347241
_IEDUC_cat_5 | .7463866 .0853624 -2.56 0.012 .5950057 .9362818
_Irace_eth_2 | 1.224852 .0722405 3.44 0.001 1.089726 1.376735
_Irace_eth_3 | .957524 .0727378 -0.57 0.569 .8236879 1.113106
_Irace_eth_4 | .8930853 .0654535 -1.54 0.126 .7723404 1.032707
ridageyr | 1.028046 .00289 9.84 0.000 1.022334 1.03379
_cons | .0739079 .0127284 -15.13 0.000 .0525353 .1039753
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | 1.020784 .087193 0.24 0.810 .8618072 1.209088
_IMJ2_2 | .9092115 .1031391 -0.84 0.403 .7261433 1.138433
gndr | 2.391754 .1199918 17.38 0.000 2.165376 2.6418
_IEDUC_cat_2 | 1.04338 .1140454 0.39 0.698 .8401532 1.295766
_IEDUC_cat_3 | 1.195339 .121385 1.76 0.082 .9774227 1.46184
_IEDUC_cat_4 | 1.168547 .1205868 1.51 0.134 .9524043 1.433742
_IEDUC_cat_5 | .8168633 .0878183 -1.88 0.063 .6601037 1.01085
_Irace_eth_2 | 1.372931 .0855804 5.08 0.000 1.213372 1.553471
_Irace_eth_3 | .8625425 .0607601 -2.10 0.038 .7501466 .9917789
_Irace_eth_4 | .9167614 .0620634 -1.28 0.202 .8016491 1.048403
ridageyr | 1.048983 .0025377 19.77 0.000 1.043965 1.054024
_cons | .0371727 .0058772 -20.82 0.000 .0271728 .0508528
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | 1.072668 .1106669 0.68 0.498 .8743012 1.316041
_IMJ2_2 | 1.017075 .1751403 0.10 0.922 .7229874 1.430789
gndr | 2.753539 .1856125 15.03 0.000 2.409177 3.147124
_IEDUC_cat_2 | 1.066084 .1504759 0.45 0.651 .8059292 1.410217
_IEDUC_cat_3 | 1.007851 .138477 0.06 0.955 .7675906 1.323314
_IEDUC_cat_4 | .9931289 .148096 -0.05 0.963 .7390076 1.334634
_IEDUC_cat_5 | .5810971 .0932102 -3.38 0.001 .4228437 .7985784
_Irace_eth_2 | 2.619565 .2030051 12.43 0.000 2.246592 3.054458
_Irace_eth_3 | 1.044106 .0969141 0.46 0.643 .8686578 1.25499
_Irace_eth_4 | 1.169005 .1077067 1.69 0.093 .973891 1.40321
ridageyr | 1.088866 .0028171 32.91 0.000 1.083297 1.094464
_cons | .0035082 .000636 -31.18 0.000 .0024492 .005025
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.I <- c(1.100949,1.245111 ,1.020784 ,.9092115,1.072668,1.017075)
RR.I <- matrix(data.I, nrow = 3, ncol = 2, byrow = TRUE)
RR.I [,1] [,2]
[1,] 1.100949 1.2451110
[2,] 1.020784 0.9092115
[3,] 1.072668 1.0170750D_RR <- (RR0-RR.I)/RR0
100*D_RR [,1] [,2]
[1,] -1.829784 7.849692
[2,] -1.818661 9.036670
[3,] -6.142223 -3.598947100*mean(abs(D_RR))[1] 5.045996use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr i.EDUC_cat i.race_eth indfmpir,rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
i.EDUC_cat _IEDUC_cat_1-5 (naturally coded; _IEDUC_cat_1 omitted)
i.race_eth _Irace_eth_1-4 (naturally coded; _Irace_eth_1 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 68,611
Number of PSUs = 214 Population size = 298,035,877
Subpop. no. obs = 19,441
Subpop. size = 133,015,285
Design df = 109
F( 33, 77) = 25.79
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | .9571534 .0776759 -0.54 0.591 .8149453 1.124177
_IMJ2_2 | 1.216895 .1511794 1.58 0.117 .9513013 1.55664
gndr | 2.163565 .1082695 15.42 0.000 1.959276 2.389154
_IEDUC_cat_2 | 1.058413 .1193842 0.50 0.616 .8463808 1.323564
_IEDUC_cat_3 | .8682652 .1058654 -1.16 0.249 .6818712 1.105611
_IEDUC_cat_4 | .8815593 .1039136 -1.07 0.287 .6978948 1.113559
_IEDUC_cat_5 | .6338603 .0768632 -3.76 0.000 .4984439 .8060663
_Irace_eth_2 | 1.215292 .0679261 3.49 0.001 1.087854 1.35766
_Irace_eth_3 | .9164525 .0683261 -1.17 0.244 .7905625 1.062389
_Irace_eth_4 | .8588513 .0603294 -2.17 0.032 .7472306 .9871459
indfmpir | 1.064036 .0195913 3.37 0.001 1.025907 1.103583
_cons | .2188138 .0253005 -13.14 0.000 .1739999 .2751696
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | .8176119 .0695524 -2.37 0.020 .6907557 .967765
_IMJ2_2 | .8307283 .1001639 -1.54 0.127 .6541455 1.054979
gndr | 2.266055 .1150376 16.11 0.000 2.049149 2.50592
_IEDUC_cat_2 | .9237482 .1098926 -0.67 0.506 .7297172 1.169372
_IEDUC_cat_3 | .9832259 .1084227 -0.15 0.878 .7901972 1.223407
_IEDUC_cat_4 | .9108924 .1044319 -0.81 0.417 .7257434 1.143276
_IEDUC_cat_5 | .651794 .0768885 -3.63 0.000 .5159072 .8234725
_Irace_eth_2 | 1.279481 .0851099 3.71 0.000 1.121442 1.459791
_Irace_eth_3 | .7535341 .0524306 -4.07 0.000 .6564652 .8649561
_Irace_eth_4 | .8260827 .0578152 -2.73 0.007 .7190871 .9489985
indfmpir | 1.08352 .0190219 4.57 0.000 1.046468 1.121885
_cons | .2604361 .0335856 -10.43 0.000 .2016967 .3362819
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | .719317 .0737898 -3.21 0.002 .586977 .8814944
_IMJ2_2 | .6899826 .1133243 -2.26 0.026 .4982707 .9554566
gndr | 2.467555 .163268 13.65 0.000 2.164283 2.813323
_IEDUC_cat_2 | .9170404 .1321449 -0.60 0.549 .6892137 1.220178
_IEDUC_cat_3 | .7931513 .1117647 -1.64 0.103 .5998805 1.04869
_IEDUC_cat_4 | .736807 .1160692 -1.94 0.055 .539211 1.006813
_IEDUC_cat_5 | .4658245 .0814792 -4.37 0.000 .3293546 .6588413
_Irace_eth_2 | 2.217334 .1618013 10.91 0.000 1.91876 2.562368
_Irace_eth_3 | .7404327 .0628494 -3.54 0.001 .6257816 .8760894
_Irace_eth_4 | .9471997 .0926923 -0.55 0.580 .7802045 1.149939
indfmpir | 1.064766 .0238115 2.81 0.006 1.018603 1.113021
_cons | .157771 .0222775 -13.08 0.000 .1192578 .2087215
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.A <- c(.9571534,1.216895,.8176119,.8307283,.719317,.6899826)
RR.A <- matrix(data.A, nrow = 3, ncol = 2, byrow = TRUE)
RR.A [,1] [,2]
[1,] 0.9571534 1.2168950
[2,] 0.8176119 0.8307283
[3,] 0.7193170 0.6899826D_RR <- (RR0-RR.A)/RR0
100*D_RR [,1] [,2]
[1,] 11.47026 9.93795
[2,] 18.44685 16.88863
[3,] 28.82243 29.71858100*mean(abs(D_RR))[1] 19.21412Of the variables above, race has the least average change in RR; it will be excluded from the next cycle.
RR.1 <- RR.Rc(111,222,222,222,333,222)
use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr indfmpir ridageyr,rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 68,616
Number of PSUs = 214 Population size = 298,070,653
Subpop. no. obs = 19,446
Subpop. size = 133,050,061
Design df = 109
F( 15, 95) = 103.68
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.130602 .0929024 1.49 0.138 .9606839 1.330573
_IMJ2_2 | 1.424398 .1812572 2.78 0.006 1.106875 1.833008
gndr | 2.262358 .1124539 16.42 0.000 2.050105 2.496587
indfmpir | .9747516 .0176818 -1.41 0.161 .9403293 1.010434
ridageyr | 1.02822 .0029215 9.79 0.000 1.022446 1.034027
_cons | .0753972 .0084576 -23.04 0.000 .0603672 .0941693
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | 1.062758 .090334 0.72 0.475 .8979879 1.257762
_IMJ2_2 | 1.068938 .1237659 0.58 0.566 .8497483 1.344666
gndr | 2.420194 .1221912 17.51 0.000 2.189737 2.674904
indfmpir | .9756942 .013897 -1.73 0.087 .9485359 1.00363
ridageyr | 1.04869 .0025453 19.59 0.000 1.043658 1.053747
_cons | .0416069 .0044129 -29.98 0.000 .0337187 .0513403
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | 1.095178 .1124804 0.89 0.378 .8934709 1.342421
_IMJ2_2 | 1.032786 .1841635 0.18 0.857 .725308 1.470612
gndr | 2.742077 .1894962 14.60 0.000 2.391087 3.144589
indfmpir | .877164 .0165874 -6.93 0.000 .8448967 .9106636
ridageyr | 1.08794 .0031174 29.41 0.000 1.081779 1.094136
_cons | .0055111 .0006687 -42.87 0.000 .0043332 .0070093
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.E <- c(1.130602 ,1.424398 ,1.062758 ,1.068938 ,1.095178,1.032786 )
RR.E <- matrix(data.E, nrow = 3, ncol = 2, byrow = TRUE)
RR.E [,1] [,2]
[1,] 1.130602 1.424398
[2,] 1.062758 1.068938
[3,] 1.095178 1.032786D_RR <- (RR.1-RR.E)/RR.1
100*D_RR [,1] [,2]
[1,] -3.048239 -4.110332
[2,] -3.280259 -4.491233
[3,] -4.505108 -5.132860100*mean(abs(D_RR))[1] 4.094672use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr i.EDUC_cat ridageyr,rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
i.EDUC_cat _IEDUC_cat_1-5 (naturally coded; _IEDUC_cat_1 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 70,182
Number of PSUs = 214 Population size = 306,497,065
Subpop. no. obs = 21,012
Subpop. size = 141,476,473
Design df = 109
F( 24, 86) = 63.78
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.120333 .0904202 1.41 0.162 .9547212 1.314672
_IMJ2_2 | 1.265921 .1579794 1.89 0.061 .9885289 1.621154
gndr | 2.217553 .1046931 16.87 0.000 2.019467 2.435069
_IEDUC_cat_2 | 1.200042 .1260383 1.74 0.085 .9745244 1.477748
_IEDUC_cat_3 | 1.083982 .1179357 0.74 0.460 .8737215 1.344842
_IEDUC_cat_4 | 1.123472 .1212864 1.08 0.283 .9070632 1.391512
_IEDUC_cat_5 | .7745404 .0834627 -2.37 0.019 .6255914 .9589532
ridageyr | 1.028163 .0028324 10.08 0.000 1.022564 1.033792
_cons | .0707093 .0107022 -17.50 0.000 .0523838 .0954457
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | 1.050681 .0891855 0.58 0.561 .8879875 1.243183
_IMJ2_2 | .9334824 .1047636 -0.61 0.541 .7473162 1.166025
gndr | 2.36321 .1179903 17.23 0.000 2.140555 2.609025
_IEDUC_cat_2 | 1.143247 .1228093 1.25 0.215 .9240093 1.414504
_IEDUC_cat_3 | 1.322551 .1285041 2.88 0.005 1.090883 1.603419
_IEDUC_cat_4 | 1.296456 .1259608 2.67 0.009 1.069371 1.571763
_IEDUC_cat_5 | .8956914 .0900997 -1.10 0.276 .7337916 1.093312
ridageyr | 1.04917 .0024816 20.29 0.000 1.044263 1.0541
_cons | .0339541 .0045839 -25.06 0.000 .0259829 .0443709
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | 1.124236 .113975 1.16 0.251 .9195893 1.374425
_IMJ2_2 | 1.028818 .176697 0.17 0.869 .7319902 1.446011
gndr | 2.65939 .1794499 14.50 0.000 2.326483 3.039933
_IEDUC_cat_2 | 1.18867 .1625432 1.26 0.209 .9064798 1.558707
_IEDUC_cat_3 | 1.071445 .1370717 0.54 0.591 .8314801 1.380664
_IEDUC_cat_4 | 1.050041 .1449439 0.35 0.724 .7987121 1.380455
_IEDUC_cat_5 | .5855999 .0870506 -3.60 0.000 .4361619 .7862386
ridageyr | 1.08675 .0028158 32.11 0.000 1.081184 1.092346
_cons | .0043306 .0006861 -34.35 0.000 .0031635 .0059282
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.I <- c(1.120333 ,1.265921 ,1.050681 ,.9334824,1.124236 ,1.028818 )
RR.I <- matrix(data.I, nrow = 3, ncol = 2, byrow = TRUE)
RR.I [,1] [,2]
[1,] 1.120333 1.2659210
[2,] 1.050681 0.9334824
[3,] 1.124236 1.0288180D_RR <- (RR.1-RR.I)/RR.1
100*D_RR [,1] [,2]
[1,] -2.112276 7.472872
[2,] -2.106600 8.749874
[3,] -7.277908 -4.728936100*mean(abs(D_RR))[1] 5.408078use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr i.EDUC_cat indfmpir,rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
i.EDUC_cat _IEDUC_cat_1-5 (naturally coded; _IEDUC_cat_1 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 68,611
Number of PSUs = 214 Population size = 298,035,877
Subpop. no. obs = 19,441
Subpop. size = 133,015,285
Design df = 109
F( 24, 86) = 23.15
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | .9741474 .0785225 -0.32 0.746 .8303135 1.142897
_IMJ2_2 | 1.237044 .1532037 1.72 0.089 .9677943 1.5812
gndr | 2.146704 .1072406 15.29 0.000 1.94434 2.37013
_IEDUC_cat_2 | 1.122798 .1244916 1.04 0.299 .9012888 1.398747
_IEDUC_cat_3 | .9274312 .1100272 -0.64 0.527 .7331021 1.173273
_IEDUC_cat_4 | .9412803 .1055646 -0.54 0.591 .7536765 1.175582
_IEDUC_cat_5 | .671777 .0772405 -3.46 0.001 .5348793 .8437125
indfmpir | 1.063945 .02008 3.28 0.001 1.024882 1.104496
_cons | .2045592 .0220203 -14.74 0.000 .1652572 .253208
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | .8432503 .071412 -2.01 0.047 .7129546 .9973581
_IMJ2_2 | .8569868 .1026542 -1.29 0.200 .6758779 1.086626
gndr | 2.237223 .113104 15.93 0.000 2.02392 2.473007
_IEDUC_cat_2 | 1.051634 .1236446 0.43 0.669 .8330332 1.3276
_IEDUC_cat_3 | 1.143662 .1224207 1.25 0.213 .9250389 1.413955
_IEDUC_cat_4 | 1.063303 .11727 0.56 0.579 .8545264 1.323088
_IEDUC_cat_5 | .7559177 .086523 -2.44 0.016 .6024922 .9484133
indfmpir | 1.087224 .0182319 4.99 0.000 1.051682 1.123966
_cons | .2181626 .0251307 -13.22 0.000 .1736311 .2741152
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | .7669189 .0762031 -2.67 0.009 .6298286 .9338488
_IMJ2_2 | .7231507 .1181014 -1.98 0.050 .5231836 .9995475
gndr | 2.393776 .1586051 13.17 0.000 2.099192 2.729701
_IEDUC_cat_2 | 1.138728 .1575779 0.94 0.350 .8655811 1.49807
_IEDUC_cat_3 | .9949971 .1334031 -0.04 0.970 .7628106 1.297857
_IEDUC_cat_4 | .932527 .1381744 -0.47 0.638 .6952184 1.25084
_IEDUC_cat_5 | .579824 .0959272 -3.29 0.001 .4177253 .8048254
indfmpir | 1.047147 .0226207 2.13 0.035 1.00326 1.092955
_cons | .1448468 .0178198 -15.70 0.000 .1135047 .1848435
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.A <- c(.9741474,1.237044 ,.8432503,.8569868,.7669189,.7231507)
RR.A <- matrix(data.A, nrow = 3, ncol = 2, byrow = TRUE)
RR.A [,1] [,2]
[1,] 0.9741474 1.2370440
[2,] 0.8432503 0.8569868
[3,] 0.7669189 0.7231507D_RR <- (RR.1-RR.A)/RR.1
100*D_RR [,1] [,2]
[1,] 11.21175 9.583514
[2,] 18.05180 16.227501
[3,] 26.81834 26.386588100*mean(abs(D_RR))[1] 18.04658Of the variables above, education has the least average change in RR; it will be excluded from the next cycle.
RR.2 <- RR.Ec(111,222,222,222,333,222)
use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr ridageyr,rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 70,190
Number of PSUs = 214 Population size = 306,545,053
Subpop. no. obs = 21,020
Subpop. size = 141,524,461
Design df = 109
F( 12, 98) = 135.85
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.174563 .0941209 2.01 0.047 1.002078 1.376737
_IMJ2_2 | 1.347399 .168559 2.38 0.019 1.051516 1.72654
gndr | 2.222371 .1048229 16.93 0.000 2.024031 2.440147
ridageyr | 1.026825 .0027857 9.76 0.000 1.021319 1.032361
_cons | .0735952 .0079714 -24.09 0.000 .0593769 .0912182
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | 1.100645 .093988 1.12 0.264 .929275 1.303618
_IMJ2_2 | .991614 .1125176 -0.07 0.941 .7919053 1.241687
gndr | 2.362098 .1160804 17.49 0.000 2.142879 2.603742
ridageyr | 1.048138 .0025136 19.60 0.000 1.043168 1.053131
_cons | .0398168 .0040783 -31.47 0.000 .0325014 .0487787
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | 1.218877 .1194473 2.02 0.046 1.003708 1.480172
_IMJ2_2 | 1.122065 .1956506 0.66 0.510 .7942012 1.585278
gndr | 2.668387 .1793587 14.60 0.000 2.335565 3.048636
ridageyr | 1.085236 .0027698 32.05 0.000 1.07976 1.090739
_cons | .0040827 .0005002 -44.90 0.000 .0032025 .0052047
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.I <- c(1.174563 ,1.347399 ,1.100645,.991614,1.218877 ,1.122065 )
RR.I <- matrix(data.I, nrow = 3, ncol = 2, byrow = TRUE)
RR.I [,1] [,2]
[1,] 1.174563 1.347399
[2,] 1.100645 0.991614
[3,] 1.218877 1.122065D_RR <- (RR.2-RR.I)/RR.2
100*D_RR [,1] [,2]
[1,] -3.888283 5.405722
[2,] -3.564970 7.233722
[3,] -11.294876 -8.644482100*mean(abs(D_RR))[1] 6.672009use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr indfmpir,rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 68,616
Number of PSUs = 214 Population size = 298,070,653
Subpop. no. obs = 19,446
Subpop. size = 133,050,061
Design df = 109
F( 12, 98) = 43.93
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.001713 .0803987 0.02 0.983 .8543932 1.174434
_IMJ2_2 | 1.28306 .1587967 2.01 0.046 1.003959 1.639751
gndr | 2.17632 .108407 15.61 0.000 1.971726 2.402143
indfmpir | 1.016521 .0181404 0.92 0.361 .9811959 1.053118
_cons | .2004787 .0112409 -28.66 0.000 .179393 .2240428
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | .8688678 .0744245 -1.64 0.104 .7332024 1.029635
_IMJ2_2 | .8923507 .1071301 -0.95 0.345 .7033936 1.132069
gndr | 2.26393 .1112692 16.63 0.000 2.053799 2.495561
indfmpir | 1.046684 .015605 3.06 0.003 1.016208 1.078074
_cons | .2348403 .0130155 -26.14 0.000 .2104103 .2621067
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | .7980226 .0777503 -2.32 0.022 .6578892 .9680051
_IMJ2_2 | .7616256 .125088 -1.66 0.100 .550012 1.054656
gndr | 2.44473 .1616624 13.52 0.000 2.144429 2.787085
indfmpir | .9830073 .0186337 -0.90 0.368 .946761 1.020641
_cons | .1453275 .0107124 -26.17 0.000 .125574 .1681884
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.A <- c(1.001713 ,1.28306 ,.8688678,.8923507,.7980226,.7616256)
RR.A <- matrix(data.A, nrow = 3, ncol = 2, byrow = TRUE)
RR.A [,1] [,2]
[1,] 1.0017130 1.2830600
[2,] 0.8688678 0.8923507
[3,] 0.7980226 0.7616256D_RR <- (RR.2-RR.A)/RR.2
100*D_RR [,1] [,2]
[1,] 11.40003 9.922648
[2,] 18.24406 16.519882
[3,] 27.13307 26.255236100*mean(abs(D_RR))[1] 18.24582Of the variables above, income has the least average change in RR; it will be excluded from the next cycle.
RR.3 <- RR.Ic(111,222,222,222,333,222)
use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr,rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 70,190
Number of PSUs = 214 Population size = 306,545,053
Subpop. no. obs = 21,020
Subpop. size = 141,524,461
Design df = 109
F( 9, 101) = 58.58
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.024774 .0793717 0.32 0.753 .8789415 1.194803
_IMJ2_2 | 1.191097 .1424676 1.46 0.147 .9397054 1.509742
gndr | 2.149441 .1016438 16.18 0.000 1.957139 2.360638
_cons | .2106798 .0092474 -35.48 0.000 .1931263 .2298288
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | .8685494 .0741751 -1.65 0.102 .7333053 1.028737
_IMJ2_2 | .7976393 .0922267 -1.96 0.053 .6342811 1.00307
gndr | 2.2241 .1066723 16.67 0.000 2.022417 2.445896
_cons | .2738866 .0110967 -31.96 0.000 .2527532 .2967871
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | .8321702 .0773206 -1.98 0.051 .692206 1.000435
_IMJ2_2 | .7859162 .1250278 -1.51 0.133 .5733796 1.077234
gndr | 2.410192 .1540981 13.76 0.000 2.123333 2.735804
_cons | .1379807 .0075147 -36.37 0.000 .1238625 .1537082
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.A <- c(1.024774 ,1.191097 ,.8685494,.7976393,.8321702,.7859162)
RR.A <- matrix(data.A, nrow = 3, ncol = 2, byrow = TRUE)
RR.A [,1] [,2]
[1,] 1.0247740 1.1910970
[2,] 0.8685494 0.7976393
[3,] 0.8321702 0.7859162D_RR <- (RR.3-RR.A)/RR.3
100*D_RR [,1] [,2]
[1,] 12.75274 11.60028
[2,] 21.08724 19.56151
[3,] 31.72648 29.95805100*mean(abs(D_RR))[1] 21.11438Of the variables above, no average change of less than 10% is found. Therefore the model from deletion cycle 3 is the final model.
use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr ridageyr, rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 70,190
Number of PSUs = 214 Population size = 306,545,053
Subpop. no. obs = 21,020
Subpop. size = 141,524,461
Design df = 109
F( 12, 98) = 135.85
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.174563 .0941209 2.01 0.047 1.002078 1.376737
_IMJ2_2 | 1.347399 .168559 2.38 0.019 1.051516 1.72654
gndr | 2.222371 .1048229 16.93 0.000 2.024031 2.440147
ridageyr | 1.026825 .0027857 9.76 0.000 1.021319 1.032361
_cons | .0735952 .0079714 -24.09 0.000 .0593769 .0912182
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | 1.100645 .093988 1.12 0.264 .929275 1.303618
_IMJ2_2 | .991614 .1125176 -0.07 0.941 .7919053 1.241687
gndr | 2.362098 .1160804 17.49 0.000 2.142879 2.603742
ridageyr | 1.048138 .0025136 19.60 0.000 1.043168 1.053131
_cons | .0398168 .0040783 -31.47 0.000 .0325014 .0487787
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | 1.218877 .1194473 2.02 0.046 1.003708 1.480172
_IMJ2_2 | 1.122065 .1956506 0.66 0.510 .7942012 1.585278
gndr | 2.668387 .1793587 14.60 0.000 2.335565 3.048636
ridageyr | 1.085236 .0027698 32.05 0.000 1.07976 1.090739
_cons | .0040827 .0005002 -44.90 0.000 .0032025 .0052047
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr ridageyr i.SMK_cat, rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
i.SMK_cat _ISMK_cat_0-4 (naturally coded; _ISMK_cat_0 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 70,139
Number of PSUs = 214 Population size = 306,306,164
Subpop. no. obs = 20,969
Subpop. size = 141,285,572
Design df = 109
F( 24, 86) = 66.50
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.12941 .0954849 1.44 0.153 .9551684 1.335438
_IMJ2_2 | 1.264867 .1629115 1.82 0.071 .9798995 1.632706
gndr | 2.204669 .1036174 16.82 0.000 2.008577 2.419904
ridageyr | 1.02604 .0028595 9.22 0.000 1.020388 1.031723
_ISMK_cat_1 | 1.05086 .0791107 0.66 0.511 .9052022 1.219957
_ISMK_cat_2 | 1.049977 .0767022 0.67 0.506 .9084489 1.213555
_ISMK_cat_3 | 1.224928 .1043444 2.38 0.019 1.034636 1.450218
_ISMK_cat_4 | 1.32306 .2455705 1.51 0.134 .9158327 1.911363
_cons | .0736428 .0080861 -23.76 0.000 .0592405 .0915467
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | 1.098333 .0956865 1.08 0.284 .9241552 1.305338
_IMJ2_2 | .9869493 .1101641 -0.12 0.907 .7910727 1.231327
gndr | 2.352808 .1174591 17.14 0.000 2.131155 2.597515
ridageyr | 1.047727 .0026051 18.75 0.000 1.042577 1.052903
_ISMK_cat_1 | 1.118922 .0783662 1.60 0.112 .9739012 1.285538
_ISMK_cat_2 | 1.058459 .0799253 0.75 0.453 .9113335 1.229336
_ISMK_cat_3 | .9497629 .1059026 -0.46 0.645 .7614424 1.184659
_ISMK_cat_4 | .9404479 .1465534 -0.39 0.694 .6905574 1.280766
_cons | .039537 .0041616 -30.69 0.000 .0320923 .0487086
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | 1.158231 .1181643 1.44 0.153 .9461927 1.417787
_IMJ2_2 | 1.043562 .1869673 0.24 0.812 .7316487 1.488448
gndr | 2.643404 .1772368 14.50 0.000 2.314467 3.019092
ridageyr | 1.084459 .0027338 32.16 0.000 1.079054 1.089891
_ISMK_cat_1 | 1.134997 .0918373 1.56 0.120 .9668234 1.332423
_ISMK_cat_2 | 1.22129 .1060835 2.30 0.023 1.028139 1.450728
_ISMK_cat_3 | 1.036988 .1039378 0.36 0.718 .8501577 1.264876
_ISMK_cat_4 | 1.415986 .2678241 1.84 0.069 .9733133 2.059991
_cons | .003986 .0004991 -44.12 0.000 .0031099 .0051089
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.S <- c(1.12941 ,1.264867 ,1.098333 ,.9869493,1.158231 ,1.043562 )
RR.S <- matrix(data.S, nrow = 3, ncol = 2, byrow = TRUE)
RR.S [,1] [,2]
[1,] 1.129410 1.2648670
[2,] 1.098333 0.9869493
[3,] 1.158231 1.0435620D_RR <- (RR.3-RR.S)/RR.3
100*D_RR [,1] [,2]
[1,] 3.8442382 6.1252829
[2,] 0.2100586 0.4704149
[3,] 4.9755636 6.9962970100*mean(abs(D_RR))[1] 3.770309use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr ridageyr i.AL_cat, rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
i.AL_cat _IAL_cat_0-2 (naturally coded; _IAL_cat_0 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 68,710
Number of PSUs = 214 Population size = 297,316,373
Subpop. no. obs = 19,540
Subpop. size = 132,295,780
Design df = 109
F( 18, 92) = 88.81
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.104569 .0871709 1.26 0.210 .9446335 1.291584
_IMJ2_2 | 1.192533 .1537121 1.37 0.175 .9236827 1.539637
gndr | 2.290824 .1174497 16.17 0.000 2.069479 2.535844
ridageyr | 1.029449 .0030117 9.92 0.000 1.023497 1.035435
_IAL_cat_1 | 1.075182 .0758649 1.03 0.307 .9348603 1.236565
_IAL_cat_2 | 1.363427 .1136229 3.72 0.000 1.155844 1.608289
_cons | .061846 .0075042 -22.94 0.000 .0486262 .0786599
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | 1.056009 .0921643 0.62 0.534 .8882681 1.255426
_IMJ2_2 | .9712096 .1171703 -0.24 0.809 .7646593 1.233553
gndr | 2.431123 .1269229 17.02 0.000 2.192143 2.696156
ridageyr | 1.050909 .0026987 19.34 0.000 1.045574 1.056271
_IAL_cat_1 | 1.113318 .0766831 1.56 0.122 .9712518 1.276164
_IAL_cat_2 | 1.353463 .1041646 3.93 0.000 1.161987 1.576491
_cons | .0328064 .0037875 -29.60 0.000 .0260967 .0412413
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | 1.074217 .1160176 0.66 0.509 .8672182 1.330625
_IMJ2_2 | 1.078072 .1937506 0.42 0.677 .7550105 1.539369
gndr | 2.75633 .199093 14.04 0.000 2.388679 3.180569
ridageyr | 1.090339 .0031013 30.41 0.000 1.08421 1.096503
_IAL_cat_1 | 1.246252 .0911208 3.01 0.003 1.078129 1.440592
_IAL_cat_2 | 1.699178 .1691396 5.33 0.000 1.394946 2.069762
_cons | .0028743 .0004133 -40.70 0.000 .0021616 .0038219
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.Al <- c(1.104569 ,1.192533 ,1.056009 ,.9712096,1.074217 ,1.078072 )
RR.Al <- matrix(data.Al, nrow = 3, ncol = 2, byrow = TRUE)
RR.Al [,1] [,2]
[1,] 1.104569 1.1925330
[2,] 1.056009 0.9712096
[3,] 1.074217 1.0780720D_RR <- (RR.3-RR.Al)/RR.3
100*D_RR [,1] [,2]
[1,] 5.959152 11.493700
[2,] 4.055440 2.057696
[3,] 11.868302 3.920718100*mean(abs(D_RR))[1] 6.559168use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr ridageyr bmxbmi, rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 70,080
Number of PSUs = 214 Population size = 305,959,182
Subpop. no. obs = 20,910
Subpop. size = 140,938,589
Design df = 109
F( 15, 95) = 118.50
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.280623 .1074311 2.95 0.004 1.084458 1.512272
_IMJ2_2 | 1.460632 .1885991 2.93 0.004 1.130833 1.886614
gndr | 2.31898 .1099931 17.73 0.000 2.110911 2.547559
ridageyr | 1.024742 .002853 8.78 0.000 1.019103 1.030412
bmxbmi | 1.063064 .004757 13.67 0.000 1.053677 1.072534
_cons | .0133227 .0022324 -25.77 0.000 .0095578 .0185706
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | 1.225362 .1047028 2.38 0.019 1.034465 1.451487
_IMJ2_2 | 1.098991 .1294114 0.80 0.425 .8702339 1.387882
gndr | 2.509066 .1301915 17.73 0.000 2.263856 2.780837
ridageyr | 1.046205 .002573 18.37 0.000 1.041117 1.051317
bmxbmi | 1.077954 .0041276 19.60 0.000 1.069804 1.086166
_cons | .0046248 .0007999 -31.08 0.000 .0032825 .0065158
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | 1.424801 .1414582 3.57 0.001 1.170297 1.734653
_IMJ2_2 | 1.31562 .2241156 1.61 0.110 .9386436 1.843998
gndr | 2.970084 .2024381 15.97 0.000 2.594779 3.399673
ridageyr | 1.084132 .0029276 29.91 0.000 1.078345 1.08995
bmxbmi | 1.104941 .0056387 19.55 0.000 1.093822 1.116173
_cons | .0002026 .0000449 -38.34 0.000 .0001306 .0003145
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.B <- c(1.280623 ,1.460632 ,1.225362 ,1.098991 ,1.424801 ,1.31562 )
RR.B <- matrix(data.B, nrow = 3, ncol = 2, byrow = TRUE)
RR.B [,1] [,2]
[1,] 1.280623 1.460632
[2,] 1.225362 1.098991
[3,] 1.424801 1.315620D_RR <- (RR.3-RR.B)/RR.3
100*D_RR [,1] [,2]
[1,] -9.029741 -8.403821
[2,] -11.331265 -10.828508
[3,] -16.894568 -17.249892100*mean(abs(D_RR))[1] 12.28963use "data\NHANES0518_new.dta", clear
svyset sdmvpsu [pw=wtmec12yr], strata(sdmvstra)
xi: svy,subpop(if include==1): mlogit BP_cat i.MJ2 gndr ridageyr hei2015, rrr pweight: wtmec12yr
VCE: linearized
Single unit: missing
Strata 1: sdmvstra
SU 1: sdmvpsu
FPC 1: <zero>
i.MJ2 _IMJ2_0-2 (naturally coded; _IMJ2_0 omitted)
(running mlogit on estimation sample)
Survey: Multinomial logistic regression
Number of strata = 105 Number of obs = 68,641
Number of PSUs = 214 Population size = 297,338,173
Subpop. no. obs = 19,471
Subpop. size = 132,317,580
Design df = 109
F( 15, 95) = 99.37
Prob > F = 0.0000
------------------------------------------------------------------------------
| Linearized
BP_cat | RRR Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
Normal | (base outcome)
-------------+----------------------------------------------------------------
Elevated |
_IMJ2_1 | 1.214348 .104401 2.26 0.026 1.024097 1.439942
_IMJ2_2 | 1.364246 .1833728 2.31 0.023 1.04519 1.780696
gndr | 2.180875 .1078477 15.77 0.000 1.977265 2.405451
ridageyr | 1.029401 .0029725 10.04 0.000 1.023527 1.03531
hei2015 | .9926355 .002118 -3.46 0.001 .9884466 .9968421
_cons | .0987333 .0137519 -16.62 0.000 .074916 .1301226
-------------+----------------------------------------------------------------
Stage_1_HTN |
_IMJ2_1 | 1.127368 .1005711 1.34 0.182 .944667 1.345405
_IMJ2_2 | 1.010226 .1298268 0.08 0.937 .7830693 1.303277
gndr | 2.244915 .1104387 16.44 0.000 2.036362 2.474827
ridageyr | 1.050878 .0028085 18.57 0.000 1.045326 1.056459
hei2015 | .9889045 .0020313 -5.43 0.000 .9848868 .9929387
_cons | .0649984 .0084111 -21.12 0.000 .0502941 .0840017
-------------+----------------------------------------------------------------
Stage_2_HTN |
_IMJ2_1 | 1.238724 .1266002 2.09 0.039 1.011587 1.516861
_IMJ2_2 | 1.062457 .2060491 0.31 0.755 .7234008 1.560427
gndr | 2.514559 .1714962 13.52 0.000 2.196631 2.878503
ridageyr | 1.089829 .0030827 30.41 0.000 1.083736 1.095956
hei2015 | .9829453 .0023304 -7.26 0.000 .9783374 .987575
_cons | .0085479 .0013591 -29.95 0.000 .0062373 .0117143
------------------------------------------------------------------------------
Note: _cons estimates baseline relative risk for each outcome.data.H <- c(1.214348 ,1.364246 ,1.127368 ,1.010226 ,1.238724 ,1.062457 )
RR.H <- matrix(data.H, nrow = 3, ncol = 2, byrow = TRUE)
RR.H [,1] [,2]
[1,] 1.214348 1.364246
[2,] 1.127368 1.010226
[3,] 1.238724 1.062457D_RR <- (RR.3-RR.H)/RR.3
100*D_RR [,1] [,2]
[1,] -3.387217 -1.250335
[2,] -2.427940 -1.876940
[3,] -1.628302 5.312348100*mean(abs(D_RR))[1] 2.64718Only BMI produced a mean change in RR greater than 10%; alcohol use produced a mean change in RR of less than 10%, but individual changes were greater than 10%; addition of smoking or healthy eating index did not produce any changes in RR greater than 10%. Therefore model 3 will include alcohol use and BMI, but not HEI or smoking.
Model 1 will be the crude model, examining BP category only as a function of MJ use category; model 2 will include gender and age as confounders; and model 3 will include gender and age, as well as all mediators (tobacco use, alcohol use, & BMI).