The National Health and Nutrition Examination Survey (NHANES) is a cross-sectional population-based survey designed to collect information on the health and nutrition of the U.S. household population.
The address is: https://wwwn.cdc.gov/nchs/nhanes/Default.aspx
"In nhanes Database Mining Tutorial 5", we have already introduced how to draw how to combine the effect values of 5 interpolated data, and then draw the restricted cubic spline diagram (RCS) of multiple linear regression. Many fans asked in private messages in the background, why only a1 data is used, can all the data be used, I thought about it later, we can use all the imputation data to do a sensitivity analysis to enhance the results of our article credibility.
There are 5 curves in the figure above, representing 5 interpolation data, and the trend of the lines is very close, indicating that the conclusions of our interpolation data are very reliable, which will help to enhance the reliability of the conclusions of our article.
Okay, let’s not talk nonsense, let’s start right away, assuming you have read the previous article "Nhanes Database Mining Tutorial 3-Multiple Imputation of Data", and have generated 5 data (a1-a5) through interpolation and analyzed them, we will directly enter the drawing link.
Let's draw a1 first, import data and R package
library(survey)
library(rms)
bc<-read.csv("E:/nhanes/a1.csv",sep=',',header=TRUE)
Let me introduce the data, SEQN: serial number, RIAGENDR, # sex, RIDAGEYR, # age, RIDRETH1, # race, DMDMARTL, # marital status, WTINT2YR, WTMEC2YR, # weight, SDMVPSU, # psu, SDMVSTRA, # strata, LBDGLUSI, # blood sugar mmol representation, LBDINSI, # insulin (pmmol/L), PHAFSTHR # meal After blood sugar, LBXGH #glycosylated hemoglobin, SPXNFEV1, #FEV1: forced expiratory volume in the first second, SPXNFVC #FVC: forced vital capacity, ml (estimated lung capacity), LBDGLTSI #2 hours after meal blood sugar.
We choose lines 1-16, followed by missing data analysis, we do not need to
convert categorical variables into factors
bc$RIAGENDR<-as.factor(bc$RIAGENDR)
bc$RIDRETH1<-as.factor(bc$RIDRETH1)
bc$DMDMARTL<-as.factor(bc$DMDMARTL)
This article is a reprinted article, the original address is as follows: https://mp.weixin.qq.com/s?__biz=MzI1NjM3NTE1NQ==&mid=2247488693&idx=1&sn=965b582465c489f5e86812397a35bf0c&chksm=ea26f4a9dd517dbf6342f 7e0909fe4b9436bf8ba87d83673691e6c6cb9d9831a983a4e6ab15f#rd