NHANES 数据集描述性统计分析 - 纤维化分组比较 - 常规

NHANES 数据集描述性统计分析 - 纤维化分组比较

本页面展示了来自NHANES数据集的描述性统计分析结果，包括年龄、性别、种族、教育水平、吸烟状况、饮酒状况、BMI、腰围、高血压、糖尿病指标、ALT、HDL、TG、PLT、WBC等指标。数据根据纤维化分组进行比较，并显示了每个组别的统计结果和p值。

总体描述性统计

tbl.all <- tbl_svysummary(NHANES_design,  
                           include = c(RIDAGEYR, sex.group, race.group, DMDEDUC2, smoke.group, drink.group,bmi.group, BMXWAIST, hyper, diabetes.index, LBXSATSI, LBDHDD, LBXTR, LBXPLTSI, LBXWBCSI, LBXIN, LBDGLUSI, ms.group, homa.ir,homa.ir.quantile,tyg,LUXCAPM,wb,cap.302.group),
                           label = list(RIDAGEYR ~ 'Age (years)', race.group ~ 'Race', DMDEDUC2 ~ 'Education level', LBXSATSI ~ 'ALT', LBDHDD ~ 'HDL', LBXTR ~ 'TG', LBXPLTSI ~ 'PLT', LBXWBCSI ~ 'WBC'),
                           statistic = list(all_continuous()  ~ "{median} ({p25}, {p75})",
                                            all_categorical() ~ "{n_unweighted} ({p}%)“),
                           sort = list(RIDAGEYR ~ "frequency", race.group ~ "frequency", DMDEDUC2 ~ "alphanumeric", smoke.group ~ "frequency", bmi.group ~ "alphanumeric", hyper ~ "frequency", diabetes.index ~ "frequency", LBXSATSI ~ "frequency", LBDHDD ~ "frequency", LBXTR ~ "frequency", LBXPLTSI ~ "frequency", LBXWBCSI ~ "frequency", LBXIN ~ "frequency", LBDGLUSI ~ "frequency", homa.ir ~ "frequency", lsm.8.2.group ~ "frequency", homa.ir.quantile ~ "alphanumeric"),
                           missing = 'no') %>%
   add_n(statistic = "{N_nonmiss_unweighted}", # 默认为 "{n}, {N_miss_unweighted}"
         col_label = "**N**",
         footnote = TRUE)

纤维化分组描述性统计

纤维化分组 1

#tab.1(lsm.7.group)
tbl.1 <- tbl_svysummary(NHANES_design,  
                         by = lsm.7.group,
                         include = c(RIDAGEYR, sex.group, race.group, DMDEDUC2, smoke.group, drink.group,bmi.group, BMXWAIST, hyper, diabetes.index, LBXSATSI, LBDHDD, LBXTR, LBXPLTSI, LBXWBCSI, LBXIN, LBDGLUSI, ms.group, homa.ir,homa.ir.quantile,tyg,LUXCAPM,wb,cap.302.group),
                         label = list(RIDAGEYR ~ 'Age (years)', race.group ~ 'Race', DMDEDUC2 ~ 'Education level', LBXSATSI ~ 'ALT', LBDHDD ~ 'HDL', LBXTR ~ 'TG', LBXPLTSI ~ 'PLT', LBXWBCSI ~ 'WBC'),
                         statistic = list(all_continuous()  ~ "{median} ({p25}, {p75})",
                                          all_categorical() ~ "{n_unweighted} ({p}%)“),
                         sort = list(RIDAGEYR ~ "frequency", race.group ~ "frequency", DMDEDUC2 ~ "alphanumeric", smoke.group ~ "frequency", bmi.group ~ "alphanumeric", hyper ~ "frequency", diabetes.index ~ "frequency", LBXSATSI ~ "frequency", LBDHDD ~ "frequency", LBXTR ~ "frequency", LBXPLTSI ~ "frequency", LBXWBCSI ~ "frequency", LBXIN ~ "frequency", LBDGLUSI ~ "frequency", homa.ir ~ "frequency", lsm.8.2.group ~ "frequency", homa.ir.quantile ~ "alphanumeric"),
                         missing = 'no') %>%
   modify_stat(p ~ format(p, digits = 3))

纤维化分组 2

#tab.2(lsm.8.2.group)
tbl.2 <- tbl_svysummary(NHANES_design,  
                         by = lsm.8.2.group,
                         include = c(RIDAGEYR, sex.group, race.group, DMDEDUC2, smoke.group, drink.group,bmi.group, BMXWAIST, hyper, diabetes.index, LBXSATSI, LBDHDD, LBXTR, LBXPLTSI, LBXWBCSI, LBXIN, LBDGLUSI, ms.group, homa.ir,homa.ir.quantile,tyg,LUXCAPM,wb,cap.302.group),
                         label = list(RIDAGEYR ~ 'Age (years)', race.group ~ 'Race', DMDEDUC2 ~ 'Education level', LBXSATSI ~ 'ALT', LBDHDD ~ 'HDL', LBXTR ~ 'TG', LBXPLTSI ~ 'PLT', LBXWBCSI ~ 'WBC'),
                         statistic = list(all_continuous()  ~ "{median} ({p25}, {p75})",
                                          all_categorical() ~ "{n_unweighted} ({p}%)“),
                         sort = list(RIDAGEYR ~ "frequency", race.group ~ "frequency", DMDEDUC2 ~ "alphanumeric", smoke.group ~ "frequency", bmi.group ~ "alphanumeric", hyper ~ "frequency", diabetes.index ~ "frequency", LBXSATSI ~ "frequency", LBDHDD ~ "frequency", LBXTR ~ "frequency", LBXPLTSI ~ "frequency", LBXWBCSI ~ "frequency", LBXIN ~ "frequency", LBDGLUSI ~ "frequency", homa.ir ~ "frequency", lsm.8.2.group ~ "frequency", homa.ir.quantile ~ "alphanumeric"),
                         missing = 'no') %>%
   modify_stat(p ~ format(p, digits = 3))

合并结果

#合并
tab1_merge<-tbl_merge(
   tbls = list(tbl.all, tbl.1, tbl.2),
   tab_spanner = c("**Overall**", "**Fibrosis**", "**Significal fibrosis**"))%>%
   modify_header(
     all_stat_cols() ~ "**{level}**, N = {n_unweighted} ({style_percent(p)}%)“) %>% 
   bold_labels()

结论

通过对NHANES数据集进行描述性统计分析，我们可以观察到不同纤维化分组之间在年龄、性别、种族、教育水平、吸烟状况、饮酒状况、BMI、腰围、高血压、糖尿病指标、ALT、HDL、TG、PLT、WBC等指标上的差异。这些差异可能反映了纤维化与相关因素之间的关系。

注：

本页面使用R语言和gtsummary包进行数据分析和可视化。
lsm.7.group 和 lsm.8.2.group 分别代表两种不同的纤维化分组。
modify_stat() 函数用于修改p值的格式，将其保留三位小数。
tbl_merge() 函数用于合并不同分组的统计结果。
modify_header() 函数用于修改表格标题。
bold_labels() 函数用于将表格标签设置为粗体。

数据来源：

NHANES 数据集

更多信息：

联系我们：

邮箱地址
[网站地址](https://www.yourwebsite.com