Stata 计算 患病率比（prevalence ratio）

*** M-H relative risk

1. cs ill exposed [fweight = freq], by(confounder)

复制代码

*** Poisson regression unadjusted
' y) ^! ]8 y5 }: X$ e; Q: L( r

1. poisson ill exposed confounder [fweight = freq], irr

复制代码

*** Poisson regression adjusted by chi-squared

1. glm ill exposed confounder [fweight = freq], family(poisson) scale(x2) eform

复制代码

" G$ I5 y1 M3 L) w2 @*** Poisson regression adjusted by deviance
# z  ]( y- b9 W$ z: u1 A

1. glm ill exposed confounder [fweight = freq], family(poisson) scale(dev) eform irls

复制代码

8 O# _" n5 i6 C9 N5 ?*** Poisson regression with robust variance
7 l( \5 w) z0 E% G3 p3 |

1. poisson ill exposed confounder [fweight = freq], irr r

复制代码

3 n- I) k# z5 t% q7 b( m*** Log-binomial regression

1. glm ill exposed confounder [fweight = freq], family(binomial) link(log) eform

复制代码

*** Odds ratio from logistic regression

1. logistic ill exposed confounder [fweight = freq]

复制代码

" @0 Y8 \7 l0 M1 e' H9 UA comparison of two methods for estimating prevalence ratios.
  C/ u1 [5 f. M+ u6 Thttps://www.ncbi.nlm.nih.gov/m/p ... m=/14567763/related
/ p6 ^8 `5 u7 Z- a! ]- D
Poisson regression with or without robust
* U+ D9 R* u- Y& Uhttps://www.statalist.org/forums ... ussion/general/5912

1 S# a- S/ {/ S- L" v, yAlternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio
6 B6 d$ h7 b8 \7 [8 khttps://www.statalist.org/forums/filedata/fetch?id=5927

. Q! T! P5 \% w