These are WTEs in post. Also calcualte the ratio of these (senior/junior)
There are already variables measuring numbers with different qualifications.
Here we calculate proportion of WTE staff in post with these qualifications.
Possible hypothesis is that human capital matters most in units that have lower overall staffing levels, so divide staffing into quartiles. (Staffing is number of WTE direct care nurses per bed.)
Other variables to use that could be related to human capital:
## glmer(formula = diedicu ~ ave.cost.nurse + IMlo + prop.occ.c +
## admissions.perbed.perday + I(transin.perbed.perday * 7) +
## total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.621 0.381
## ave.cost.nurse 0.009 0.013
## IMlo 0.966 0.012
## prop.occ.c 0.256 0.096
## admissions.perbed.perday 0.692 0.822
## I(transin.perbed.perday * 7) 1.789 0.652
## total.support.staff.perbed 0.154 0.193
## N.dc.perbed -0.122 0.040
## clinical.NHD.perbed -0.175 0.058
## intensivistYes 0.006 0.109
## IMlomean.x -0.054 0.020
## I(avyulos/100) 0.492 0.170
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.276
## Residual 1.000
## ---
## number of obs: 36935, groups: trust.code, 63
## AIC = 22536.6, DIC = 22511
## deviance = 22510.6
## glmer(formula = diedicu ~ ave.cost.nurse * N.dc.quart + IMlo +
## prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.578 0.445
## ave.cost.nurse 0.007 0.015
## N.dc.quart(4.25,4.82] -0.563 1.155
## N.dc.quart(4.82,5.68] 0.146 0.579
## N.dc.quart(5.68,14.2] -0.178 1.096
## IMlo 0.966 0.012
## prop.occ.c 0.253 0.096
## admissions.perbed.perday 0.628 0.899
## I(transin.perbed.perday * 7) 1.567 0.714
## total.support.staff.perbed 0.112 0.205
## N.dc.perbed -0.134 0.061
## clinical.NHD.perbed -0.158 0.065
## intensivistYes -0.018 0.113
## IMlomean.x -0.054 0.020
## I(avyulos/100) 0.481 0.176
## ave.cost.nurse:N.dc.quart(4.25,4.82] 0.032 0.054
## ave.cost.nurse:N.dc.quart(4.82,5.68] -0.003 0.025
## ave.cost.nurse:N.dc.quart(5.68,14.2] 0.015 0.054
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.276
## Residual 1.000
## ---
## number of obs: 36935, groups: trust.code, 63
## AIC = 22547.5, DIC = 22510
## deviance = 22509.5
## Data: dta4
## Subset: ss8
## Models:
## cost.icu: diedicu ~ ave.cost.nurse + IMlo + prop.occ.c + admissions.perbed.perday +
## cost.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## cost.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## cost.icu: I(avyulos/100) + (1 | trust.code)
## cost.icu.int: diedicu ~ ave.cost.nurse * N.dc.quart + IMlo + prop.occ.c + admissions.perbed.perday +
## cost.icu.int: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## cost.icu.int: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## cost.icu.int: I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## cost.icu 13 22537 22647 -11255
## cost.icu.int 19 22547 22709 -11255 1.11 6 0.98
## glmer(formula = diedicu ~ ratio.pbq.wte + IMlo + prop.occ.c +
## admissions.perbed.perday + I(transin.perbed.perday * 7) +
## total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.239 0.258
## ratio.pbq.wte -0.142 0.133
## IMlo 0.965 0.012
## prop.occ.c 0.212 0.093
## admissions.perbed.perday 0.428 0.797
## I(transin.perbed.perday * 7) 1.613 0.625
## total.support.staff.perbed 0.168 0.187
## N.dc.perbed -0.104 0.037
## clinical.NHD.perbed -0.170 0.058
## intensivistYes -0.043 0.106
## IMlomean.x -0.051 0.020
## I(avyulos/100) 0.413 0.165
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.273
## Residual 1.000
## ---
## number of obs: 38168, groups: trust.code, 65
## AIC = 23317.7, DIC = 23292
## deviance = 23291.7
## glmer(formula = diedicu ~ ratio.pbq.wte * N.dc.quart + IMlo +
## prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -0.653 0.339
## ratio.pbq.wte -1.007 0.310
## N.dc.quart(4.25,4.82] -0.587 0.316
## N.dc.quart(4.82,5.68] -0.936 0.346
## N.dc.quart(5.68,14.2] -1.065 0.383
## IMlo 0.965 0.012
## prop.occ.c 0.220 0.093
## admissions.perbed.perday -0.037 0.781
## I(transin.perbed.perday * 7) 1.383 0.586
## total.support.staff.perbed 0.158 0.175
## N.dc.perbed -0.034 0.059
## clinical.NHD.perbed -0.206 0.056
## intensivistYes -0.060 0.098
## IMlomean.x -0.051 0.020
## I(avyulos/100) 0.322 0.159
## ratio.pbq.wte:N.dc.quart(4.25,4.82] 0.783 0.367
## ratio.pbq.wte:N.dc.quart(4.82,5.68] 1.152 0.392
## ratio.pbq.wte:N.dc.quart(5.68,14.2] 1.285 0.390
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.241
## Residual 1.000
## ---
## number of obs: 38168, groups: trust.code, 65
## AIC = 23318.6, DIC = 23281
## deviance = 23280.6
## Data: dta4
## Subset: ss8
## Models:
## pbq.icu: diedicu ~ ratio.pbq.wte + IMlo + prop.occ.c + admissions.perbed.perday +
## pbq.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## pbq.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## pbq.icu: I(avyulos/100) + (1 | trust.code)
## pbq.icu.int: diedicu ~ ratio.pbq.wte * N.dc.quart + IMlo + prop.occ.c + admissions.perbed.perday +
## pbq.icu.int: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## pbq.icu.int: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## pbq.icu.int: I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## pbq.icu 13 23318 23429 -11646
## pbq.icu.int 19 23319 23481 -11640 11.1 6 0.086 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## glmer(formula = diedicu ~ ratio.ot.total.spend + IMlo + prop.occ.c +
## admissions.perbed.perday + I(transin.perbed.perday * 7) +
## total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.433 0.237
## ratio.ot.total.spend 0.910 4.803
## IMlo 0.966 0.012
## prop.occ.c 0.260 0.096
## admissions.perbed.perday 0.760 0.833
## I(transin.perbed.perday * 7) 1.849 0.652
## total.support.staff.perbed 0.147 0.195
## N.dc.perbed -0.128 0.040
## clinical.NHD.perbed -0.173 0.058
## intensivistYes -0.003 0.109
## IMlomean.x -0.054 0.020
## I(avyulos/100) 0.487 0.176
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.276
## Residual 1.000
## ---
## number of obs: 36935, groups: trust.code, 63
## AIC = 22537, DIC = 22511
## deviance = 22511.0
## glmer(formula = diedicu ~ ratio.ot.total.spend * N.dc.quart +
## IMlo + prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.555 0.283
## ratio.ot.total.spend 1.766 6.985
## N.dc.quart(4.25,4.82] 0.042 0.131
## N.dc.quart(4.82,5.68] 0.030 0.143
## N.dc.quart(5.68,14.2] 0.139 0.193
## IMlo 0.965 0.012
## prop.occ.c 0.254 0.096
## admissions.perbed.perday 1.217 0.844
## I(transin.perbed.perday * 7) 1.810 0.643
## total.support.staff.perbed 0.151 0.195
## N.dc.perbed -0.146 0.060
## clinical.NHD.perbed -0.156 0.058
## intensivistYes -0.001 0.111
## IMlomean.x -0.056 0.020
## I(avyulos/100) 0.552 0.177
## ratio.ot.total.spend:N.dc.quart(4.25,4.82] 4.840 11.774
## ratio.ot.total.spend:N.dc.quart(4.82,5.68] 6.869 11.078
## ratio.ot.total.spend:N.dc.quart(5.68,14.2] -10.195 10.691
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.265
## Residual 1.000
## ---
## number of obs: 36935, groups: trust.code, 63
## AIC = 22544, DIC = 22506
## deviance = 22506.0
## Data: dta4
## Subset: ss8
## Models:
## ot.icu: diedicu ~ ratio.ot.total.spend + IMlo + prop.occ.c + admissions.perbed.perday +
## ot.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## ot.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## ot.icu: I(avyulos/100) + (1 | trust.code)
## ot.icu.int: diedicu ~ ratio.ot.total.spend * N.dc.quart + IMlo + prop.occ.c +
## ot.icu.int: admissions.perbed.perday + I(transin.perbed.perday * 7) +
## ot.icu.int: total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## ot.icu.int: intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## ot.icu 13 22537 22648 -11255
## ot.icu.int 19 22544 22706 -11253 4.98 6 0.55
## glmer(formula = diedicu ~ ratio.bank.total.spend + IMlo + prop.occ.c +
## admissions.perbed.perday + I(transin.perbed.perday * 7) +
## total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.447 0.230
## ratio.bank.total.spend -0.864 1.213
## IMlo 0.966 0.012
## prop.occ.c 0.262 0.096
## admissions.perbed.perday 0.771 0.820
## I(transin.perbed.perday * 7) 1.879 0.649
## total.support.staff.perbed 0.145 0.193
## N.dc.perbed -0.129 0.040
## clinical.NHD.perbed -0.167 0.059
## intensivistYes 0.004 0.108
## IMlomean.x -0.054 0.020
## I(avyulos/100) 0.520 0.178
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.275
## Residual 1.000
## ---
## number of obs: 36935, groups: trust.code, 63
## AIC = 22536.5, DIC = 22511
## deviance = 22510.5
## glmer(formula = diedicu ~ ratio.bank.total.spend * N.dc.quart +
## IMlo + prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.477 0.269
## ratio.bank.total.spend 1.346 4.184
## N.dc.quart(4.25,4.82] 0.142 0.139
## N.dc.quart(4.82,5.68] 0.117 0.155
## N.dc.quart(5.68,14.2] 0.047 0.218
## IMlo 0.966 0.012
## prop.occ.c 0.261 0.096
## admissions.perbed.perday 0.979 0.831
## I(transin.perbed.perday * 7) 2.074 0.675
## total.support.staff.perbed 0.147 0.197
## N.dc.perbed -0.144 0.061
## clinical.NHD.perbed -0.169 0.061
## intensivistYes 0.021 0.110
## IMlomean.x -0.054 0.020
## I(avyulos/100) 0.485 0.185
## ratio.bank.total.spend:N.dc.quart(4.25,4.82] -3.174 4.594
## ratio.bank.total.spend:N.dc.quart(4.82,5.68] -3.244 4.420
## ratio.bank.total.spend:N.dc.quart(5.68,14.2] -0.099 4.534
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.271
## Residual 1.000
## ---
## number of obs: 36935, groups: trust.code, 63
## AIC = 22545.8, DIC = 22508
## deviance = 22507.8
## Data: dta4
## Subset: ss8
## Models:
## bank.icu: diedicu ~ ratio.bank.total.spend + IMlo + prop.occ.c + admissions.perbed.perday +
## bank.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## bank.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## bank.icu: I(avyulos/100) + (1 | trust.code)
## bank.icu.int: diedicu ~ ratio.bank.total.spend * N.dc.quart + IMlo + prop.occ.c +
## bank.icu.int: admissions.perbed.perday + I(transin.perbed.perday * 7) +
## bank.icu.int: total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## bank.icu.int: intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## bank.icu 13 22537 22647 -11255
## bank.icu.int 19 22546 22708 -11254 2.73 6 0.84
## glmer(formula = diedicu ~ ratio.agency.total.spend + IMlo + prop.occ.c +
## admissions.perbed.perday + I(transin.perbed.perday * 7) +
## total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.413 0.228
## ratio.agency.total.spend -0.266 0.728
## IMlo 0.966 0.012
## prop.occ.c 0.258 0.096
## admissions.perbed.perday 0.754 0.822
## I(transin.perbed.perday * 7) 1.838 0.649
## total.support.staff.perbed 0.152 0.194
## N.dc.perbed -0.128 0.040
## clinical.NHD.perbed -0.175 0.058
## intensivistYes 0.005 0.110
## IMlomean.x -0.054 0.020
## I(avyulos/100) 0.485 0.170
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.276
## Residual 1.000
## ---
## number of obs: 36935, groups: trust.code, 63
## AIC = 22536.9, DIC = 22511
## deviance = 22510.9
## glmer(formula = diedicu ~ ratio.agency.total.spend * N.dc.quart +
## IMlo + prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.416 0.270
## ratio.agency.total.spend -0.274 0.864
## N.dc.quart(4.25,4.82] 0.080 0.137
## N.dc.quart(4.82,5.68] 0.040 0.157
## N.dc.quart(5.68,14.2] 0.072 0.198
## IMlo 0.966 0.012
## prop.occ.c 0.257 0.096
## admissions.perbed.perday 0.814 0.839
## I(transin.perbed.perday * 7) 1.784 0.674
## total.support.staff.perbed 0.142 0.197
## N.dc.perbed -0.139 0.060
## clinical.NHD.perbed -0.173 0.061
## intensivistYes -0.004 0.114
## IMlomean.x -0.055 0.020
## I(avyulos/100) 0.482 0.173
## ratio.agency.total.spend:N.dc.quart(4.25,4.82] -0.047 2.246
## ratio.agency.total.spend:N.dc.quart(4.82,5.68] 1.168 2.990
## ratio.agency.total.spend:N.dc.quart(5.68,14.2] 0.385 2.578
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.275
## Residual 1.000
## ---
## number of obs: 36935, groups: trust.code, 63
## AIC = 22548.2, DIC = 22510
## deviance = 22510.2
## Data: dta4
## Subset: ss8
## Models:
## agency.icu: diedicu ~ ratio.agency.total.spend + IMlo + prop.occ.c + admissions.perbed.perday +
## agency.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## agency.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## agency.icu: I(avyulos/100) + (1 | trust.code)
## agency.icu.int: diedicu ~ ratio.agency.total.spend * N.dc.quart + IMlo + prop.occ.c +
## agency.icu.int: admissions.perbed.perday + I(transin.perbed.perday * 7) +
## agency.icu.int: total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## agency.icu.int: intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## agency.icu 13 22537 22648 -11255
## agency.icu.int 19 22548 22710 -11255 0.66 6 1
## glmer(formula = diedicu ~ ratio.senior.junior.wte + IMlo + prop.occ.c +
## admissions.perbed.perday + I(transin.perbed.perday * 7) +
## total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.294 0.250
## ratio.senior.junior.wte 0.017 0.308
## IMlo 0.970 0.013
## prop.occ.c 0.204 0.096
## admissions.perbed.perday -0.014 0.920
## I(transin.perbed.perday * 7) 1.643 0.651
## total.support.staff.perbed 0.213 0.196
## N.dc.perbed -0.083 0.043
## clinical.NHD.perbed -0.178 0.060
## intensivistYes -0.034 0.108
## IMlomean.x -0.046 0.021
## I(avyulos/100) 0.327 0.189
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.283
## Residual 1.000
## ---
## number of obs: 36343, groups: trust.code, 61
## AIC = 21993.2, DIC = 21967
## deviance = 21967.2
## glmer(formula = diedicu ~ ratio.senior.junior.wte * N.dc.quart +
## IMlo + prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.205 0.342
## ratio.senior.junior.wte -1.766 1.306
## N.dc.quart(4.25,4.82] -0.307 0.272
## N.dc.quart(4.82,5.68] -0.485 0.303
## N.dc.quart(5.68,14.2] -0.313 0.286
## IMlo 0.970 0.013
## prop.occ.c 0.213 0.096
## admissions.perbed.perday -0.084 0.934
## I(transin.perbed.perday * 7) 1.464 0.662
## total.support.staff.perbed 0.184 0.201
## N.dc.perbed -0.046 0.067
## clinical.NHD.perbed -0.171 0.066
## intensivistYes -0.061 0.110
## IMlomean.x -0.046 0.021
## I(avyulos/100) 0.344 0.196
## ratio.senior.junior.wte:N.dc.quart(4.25,4.82] 2.387 1.578
## ratio.senior.junior.wte:N.dc.quart(4.82,5.68] 3.188 1.639
## ratio.senior.junior.wte:N.dc.quart(5.68,14.2] 1.770 1.357
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.279
## Residual 1.000
## ---
## number of obs: 36343, groups: trust.code, 61
## AIC = 22000.5, DIC = 21963
## deviance = 21962.5
## Data: dta4
## Subset: ss8
## Models:
## senior.icu: diedicu ~ ratio.senior.junior.wte + IMlo + prop.occ.c + admissions.perbed.perday +
## senior.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## senior.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## senior.icu: I(avyulos/100) + (1 | trust.code)
## senior.icu.int: diedicu ~ ratio.senior.junior.wte * N.dc.quart + IMlo + prop.occ.c +
## senior.icu.int: admissions.perbed.perday + I(transin.perbed.perday * 7) +
## senior.icu.int: total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## senior.icu.int: intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## senior.icu 13 21993 22104 -10984
## senior.icu.int 19 22000 22162 -10981 4.71 6 0.58
## glmer(formula = diedicu ~ ratio.enb6 + IMlo + prop.occ.c + admissions.perbed.perday +
## I(transin.perbed.perday * 7) + total.support.staff.perbed +
## N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## I(avyulos/100) + (1 | trust.code), data = dta4, family = binomial(),
## subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.217 0.264
## ratio.enb6 -0.202 0.176
## IMlo 0.965 0.012
## prop.occ.c 0.211 0.093
## admissions.perbed.perday 0.432 0.794
## I(transin.perbed.perday * 7) 1.673 0.625
## total.support.staff.perbed 0.156 0.186
## N.dc.perbed -0.106 0.037
## clinical.NHD.perbed -0.168 0.057
## intensivistYes -0.037 0.105
## IMlomean.x -0.052 0.020
## I(avyulos/100) 0.408 0.165
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.272
## Residual 1.000
## ---
## number of obs: 38168, groups: trust.code, 65
## AIC = 23317.6, DIC = 23292
## deviance = 23291.6
## glmer(formula = diedicu ~ ratio.enb6 * N.dc.quart + IMlo + prop.occ.c +
## admissions.perbed.perday + I(transin.perbed.perday * 7) +
## total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -0.855 0.370
## ratio.enb6 -0.804 0.418
## N.dc.quart(4.25,4.82] -0.373 0.336
## N.dc.quart(4.82,5.68] -0.374 0.347
## N.dc.quart(5.68,14.2] -0.436 0.384
## IMlo 0.965 0.012
## prop.occ.c 0.217 0.094
## admissions.perbed.perday 0.175 0.824
## I(transin.perbed.perday * 7) 1.545 0.625
## total.support.staff.perbed 0.159 0.186
## N.dc.perbed -0.089 0.058
## clinical.NHD.perbed -0.178 0.059
## intensivistYes -0.052 0.105
## IMlomean.x -0.052 0.020
## I(avyulos/100) 0.359 0.170
## ratio.enb6:N.dc.quart(4.25,4.82] 0.724 0.502
## ratio.enb6:N.dc.quart(4.82,5.68] 0.669 0.522
## ratio.enb6:N.dc.quart(5.68,14.2] 0.786 0.527
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.264
## Residual 1.000
## ---
## number of obs: 38168, groups: trust.code, 65
## AIC = 23326.6, DIC = 23289
## deviance = 23288.6
## Data: dta4
## Subset: ss8
## Models:
## enb6.icu: diedicu ~ ratio.enb6 + IMlo + prop.occ.c + admissions.perbed.perday +
## enb6.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## enb6.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## enb6.icu: I(avyulos/100) + (1 | trust.code)
## enb6.icu.int: diedicu ~ ratio.enb6 * N.dc.quart + IMlo + prop.occ.c + admissions.perbed.perday +
## enb6.icu.int: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## enb6.icu.int: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## enb6.icu.int: I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## enb6.icu 13 23318 23429 -11646
## enb6.icu.int 19 23327 23489 -11644 2.96 6 0.81
## glmer(formula = diedicu ~ ratio.diploma + IMlo + prop.occ.c +
## admissions.perbed.perday + I(transin.perbed.perday * 7) +
## total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.367 0.229
## ratio.diploma -0.066 0.269
## IMlo 0.965 0.012
## prop.occ.c 0.213 0.093
## admissions.perbed.perday 0.495 0.800
## I(transin.perbed.perday * 7) 1.604 0.635
## total.support.staff.perbed 0.159 0.189
## N.dc.perbed -0.105 0.037
## clinical.NHD.perbed -0.165 0.058
## intensivistYes -0.031 0.106
## IMlomean.x -0.051 0.020
## I(avyulos/100) 0.431 0.165
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.276
## Residual 1.000
## ---
## number of obs: 38168, groups: trust.code, 65
## AIC = 23318.8, DIC = 23293
## deviance = 23292.8
## glmer(formula = diedicu ~ ratio.diploma * N.dc.quart + IMlo +
## prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.267 0.260
## ratio.diploma -1.644 0.622
## N.dc.quart(4.25,4.82] -0.084 0.151
## N.dc.quart(4.82,5.68] -0.344 0.180
## N.dc.quart(5.68,14.2] -0.317 0.219
## IMlo 0.965 0.012
## prop.occ.c 0.203 0.093
## admissions.perbed.perday 0.541 0.774
## I(transin.perbed.perday * 7) 1.362 0.594
## total.support.staff.perbed 0.134 0.176
## N.dc.perbed -0.067 0.056
## clinical.NHD.perbed -0.184 0.055
## intensivistYes -0.079 0.098
## IMlomean.x -0.050 0.020
## I(avyulos/100) 0.395 0.157
## ratio.diploma:N.dc.quart(4.25,4.82] 1.019 0.737
## ratio.diploma:N.dc.quart(4.82,5.68] 2.565 0.826
## ratio.diploma:N.dc.quart(5.68,14.2] 2.368 0.759
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.243
## Residual 1.000
## ---
## number of obs: 38168, groups: trust.code, 65
## AIC = 23316.8, DIC = 23279
## deviance = 23278.8
## Data: dta4
## Subset: ss8
## Models:
## diploma.icu: diedicu ~ ratio.diploma + IMlo + prop.occ.c + admissions.perbed.perday +
## diploma.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## diploma.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## diploma.icu: I(avyulos/100) + (1 | trust.code)
## diploma.icu.int: diedicu ~ ratio.diploma * N.dc.quart + IMlo + prop.occ.c + admissions.perbed.perday +
## diploma.icu.int: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## diploma.icu.int: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## diploma.icu.int: I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## diploma.icu 13 23319 23430 -11646
## diploma.icu.int 19 23317 23479 -11639 14 6 0.03 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## glmer(formula = diedicu ~ ratio.degree + IMlo + prop.occ.c +
## admissions.perbed.perday + I(transin.perbed.perday * 7) +
## total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.367 0.226
## ratio.degree -0.428 0.768
## IMlo 0.965 0.012
## prop.occ.c 0.215 0.094
## admissions.perbed.perday 0.526 0.798
## I(transin.perbed.perday * 7) 1.606 0.630
## total.support.staff.perbed 0.149 0.187
## N.dc.perbed -0.106 0.037
## clinical.NHD.perbed -0.167 0.058
## intensivistYes -0.039 0.107
## IMlomean.x -0.051 0.020
## I(avyulos/100) 0.438 0.165
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.275
## Residual 1.000
## ---
## number of obs: 38168, groups: trust.code, 65
## AIC = 23318.6, DIC = 23293
## deviance = 23292.6
## glmer(formula = diedicu ~ ratio.degree * N.dc.quart + IMlo +
## prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.367 0.261
## ratio.degree -1.157 1.243
## N.dc.quart(4.25,4.82] 0.055 0.140
## N.dc.quart(4.82,5.68] -0.003 0.161
## N.dc.quart(5.68,14.2] -0.011 0.206
## IMlo 0.965 0.012
## prop.occ.c 0.214 0.094
## admissions.perbed.perday 0.624 0.803
## I(transin.perbed.perday * 7) 1.579 0.641
## total.support.staff.perbed 0.168 0.187
## N.dc.perbed -0.106 0.057
## clinical.NHD.perbed -0.162 0.059
## intensivistYes -0.052 0.107
## IMlomean.x -0.051 0.020
## I(avyulos/100) 0.421 0.165
## ratio.degree:N.dc.quart(4.25,4.82] -5.734 5.578
## ratio.degree:N.dc.quart(4.82,5.68] 0.870 1.571
## ratio.degree:N.dc.quart(5.68,14.2] 2.139 1.916
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.269
## Residual 1.000
## ---
## number of obs: 38168, groups: trust.code, 65
## AIC = 23327.6, DIC = 23290
## deviance = 23289.6
## Data: dta4
## Subset: ss8
## Models:
## degree.icu: diedicu ~ ratio.degree + IMlo + prop.occ.c + admissions.perbed.perday +
## degree.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## degree.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## degree.icu: I(avyulos/100) + (1 | trust.code)
## degree.icu.int: diedicu ~ ratio.degree * N.dc.quart + IMlo + prop.occ.c + admissions.perbed.perday +
## degree.icu.int: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## degree.icu.int: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## degree.icu.int: I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## degree.icu 13 23319 23430 -11646
## degree.icu.int 19 23328 23490 -11645 3.01 6 0.81
## glmer(formula = diedicu ~ ratio.enb6 * N.dc.three + IMlo + prop.occ.c +
## admissions.perbed.perday + I(transin.perbed.perday * 7) +
## total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -0.835 0.353
## ratio.enb6 -0.621 0.348
## N.dc.three(4.48,5.51] -0.415 0.277
## N.dc.three(5.51,14.2] -0.272 0.331
## IMlo 0.965 0.012
## prop.occ.c 0.207 0.093
## admissions.perbed.perday 0.116 0.802
## I(transin.perbed.perday * 7) 1.660 0.608
## total.support.staff.perbed 0.188 0.180
## N.dc.perbed -0.109 0.055
## clinical.NHD.perbed -0.177 0.058
## intensivistYes -0.038 0.100
## IMlomean.x -0.052 0.020
## I(avyulos/100) 0.363 0.165
## ratio.enb6:N.dc.three(4.48,5.51] 0.617 0.417
## ratio.enb6:N.dc.three(5.51,14.2] 0.513 0.455
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.257
## Residual 1.000
## ---
## number of obs: 38168, groups: trust.code, 65
## AIC = 23322.5, DIC = 23289
## deviance = 23288.5
## Data: dta4
## Subset: ss8
## Models:
## enb6.icu: diedicu ~ ratio.enb6 + IMlo + prop.occ.c + admissions.perbed.perday +
## enb6.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## enb6.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## enb6.icu: I(avyulos/100) + (1 | trust.code)
## enb6.icu.thr: diedicu ~ ratio.enb6 * N.dc.three + IMlo + prop.occ.c + admissions.perbed.perday +
## enb6.icu.thr: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## enb6.icu.thr: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## enb6.icu.thr: I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## enb6.icu 13 23318 23429 -11646
## enb6.icu.thr 17 23323 23468 -11644 3.07 4 0.55
## glmer(formula = diedicu ~ ave.cost.nurse * N.dc.three + IMlo +
## prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.470 0.413
## ave.cost.nurse 0.007 0.014
## N.dc.three(4.48,5.51] -0.252 0.535
## N.dc.three(5.51,14.2] 0.068 1.002
## IMlo 0.966 0.012
## prop.occ.c 0.250 0.096
## admissions.perbed.perday 0.645 0.842
## I(transin.perbed.perday * 7) 1.883 0.641
## total.support.staff.perbed 0.155 0.193
## N.dc.perbed -0.156 0.057
## clinical.NHD.perbed -0.159 0.062
## intensivistYes 0.006 0.106
## IMlomean.x -0.054 0.020
## I(avyulos/100) 0.504 0.167
## ave.cost.nurse:N.dc.three(4.48,5.51] 0.011 0.024
## ave.cost.nurse:N.dc.three(5.51,14.2] 0.002 0.048
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.266
## Residual 1.000
## ---
## number of obs: 36935, groups: trust.code, 63
## AIC = 22543.1, DIC = 22509
## deviance = 22509.1
## Data: dta4
## Subset: ss8
## Models:
## cost.icu: diedicu ~ ave.cost.nurse + IMlo + prop.occ.c + admissions.perbed.perday +
## cost.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## cost.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## cost.icu: I(avyulos/100) + (1 | trust.code)
## cost.icu.int3: diedicu ~ ave.cost.nurse * N.dc.three + IMlo + prop.occ.c + admissions.perbed.perday +
## cost.icu.int3: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## cost.icu.int3: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## cost.icu.int3: I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## cost.icu 13 22537 22647 -11255
## cost.icu.int3 17 22543 22688 -11255 1.51 4 0.83
## glmer(formula = diedicu ~ ratio.pbq.wte * N.dc.three + IMlo +
## prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -0.535 0.305
## ratio.pbq.wte -1.017 0.247
## N.dc.three(4.48,5.51] -0.926 0.260
## N.dc.three(5.51,14.2] -1.035 0.311
## IMlo 0.965 0.012
## prop.occ.c 0.211 0.093
## admissions.perbed.perday -0.239 0.732
## I(transin.perbed.perday * 7) 1.310 0.557
## total.support.staff.perbed 0.261 0.168
## N.dc.perbed -0.038 0.054
## clinical.NHD.perbed -0.214 0.054
## intensivistYes -0.072 0.092
## IMlomean.x -0.051 0.020
## I(avyulos/100) 0.296 0.151
## ratio.pbq.wte:N.dc.three(4.48,5.51] 1.069 0.304
## ratio.pbq.wte:N.dc.three(5.51,14.2] 1.271 0.322
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.227
## Residual 1.000
## ---
## number of obs: 38168, groups: trust.code, 65
## AIC = 23309.5, DIC = 23276
## deviance = 23275.5
## Data: dta4
## Subset: ss8
## Models:
## pbq.icu: diedicu ~ ratio.pbq.wte + IMlo + prop.occ.c + admissions.perbed.perday +
## pbq.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## pbq.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## pbq.icu: I(avyulos/100) + (1 | trust.code)
## pbq.icu.int3: diedicu ~ ratio.pbq.wte * N.dc.three + IMlo + prop.occ.c + admissions.perbed.perday +
## pbq.icu.int3: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## pbq.icu.int3: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## pbq.icu.int3: I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## pbq.icu 13 23318 23429 -11646
## pbq.icu.int3 17 23310 23455 -11638 16.2 4 0.0028 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## glmer(formula = diedicu ~ ratio.ot.total.spend * N.dc.three +
## IMlo + prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.433 0.279
## ratio.ot.total.spend 3.480 5.940
## N.dc.three(4.48,5.51] 0.005 0.113
## N.dc.three(5.51,14.2] 0.165 0.156
## IMlo 0.965 0.012
## prop.occ.c 0.249 0.096
## admissions.perbed.perday 1.081 0.839
## I(transin.perbed.perday * 7) 1.919 0.637
## total.support.staff.perbed 0.177 0.192
## N.dc.perbed -0.166 0.057
## clinical.NHD.perbed -0.154 0.058
## intensivistYes 0.020 0.107
## IMlomean.x -0.056 0.020
## I(avyulos/100) 0.536 0.175
## ratio.ot.total.spend:N.dc.three(4.48,5.51] 3.485 11.185
## ratio.ot.total.spend:N.dc.three(5.51,14.2] -11.642 7.611
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.264
## Residual 1.000
## ---
## number of obs: 36935, groups: trust.code, 63
## AIC = 22540.9, DIC = 22507
## deviance = 22506.9
## Data: dta4
## Subset: ss8
## Models:
## ot.icu: diedicu ~ ratio.ot.total.spend + IMlo + prop.occ.c + admissions.perbed.perday +
## ot.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## ot.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## ot.icu: I(avyulos/100) + (1 | trust.code)
## ot.icu.int3: diedicu ~ ratio.ot.total.spend * N.dc.three + IMlo + prop.occ.c +
## ot.icu.int3: admissions.perbed.perday + I(transin.perbed.perday * 7) +
## ot.icu.int3: total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## ot.icu.int3: intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## ot.icu 13 22537 22648 -11255
## ot.icu.int3 17 22541 22686 -11253 4.06 4 0.4
## glmer(formula = diedicu ~ ratio.bank.total.spend * N.dc.three +
## IMlo + prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.325 0.261
## ratio.bank.total.spend -1.375 3.872
## N.dc.three(4.48,5.51] 0.010 0.127
## N.dc.three(5.51,14.2] 0.089 0.164
## IMlo 0.966 0.012
## prop.occ.c 0.258 0.096
## admissions.perbed.perday 0.764 0.814
## I(transin.perbed.perday * 7) 2.040 0.684
## total.support.staff.perbed 0.170 0.193
## N.dc.perbed -0.165 0.057
## clinical.NHD.perbed -0.159 0.059
## intensivistYes 0.006 0.108
## IMlomean.x -0.054 0.020
## I(avyulos/100) 0.520 0.177
## ratio.bank.total.spend:N.dc.three(4.48,5.51] 0.260 4.129
## ratio.bank.total.spend:N.dc.three(5.51,14.2] 1.309 4.143
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.267
## Residual 1.000
## ---
## number of obs: 36935, groups: trust.code, 63
## AIC = 22543.2, DIC = 22509
## deviance = 22509.2
## Data: dta4
## Subset: ss8
## Models:
## bank.icu: diedicu ~ ratio.bank.total.spend + IMlo + prop.occ.c + admissions.perbed.perday +
## bank.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## bank.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## bank.icu: I(avyulos/100) + (1 | trust.code)
## bank.icu.int3: diedicu ~ ratio.bank.total.spend * N.dc.three + IMlo + prop.occ.c +
## bank.icu.int3: admissions.perbed.perday + I(transin.perbed.perday * 7) +
## bank.icu.int3: total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## bank.icu.int3: intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## bank.icu 13 22537 22647 -11255
## bank.icu.int3 17 22543 22688 -11255 1.32 4 0.86
## glmer(formula = diedicu ~ ratio.agency.total.spend * N.dc.three +
## IMlo + prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.287 0.262
## ratio.agency.total.spend -0.389 0.813
## N.dc.three(4.48,5.51] -0.042 0.123
## N.dc.three(5.51,14.2] 0.077 0.156
## IMlo 0.966 0.012
## prop.occ.c 0.250 0.096
## admissions.perbed.perday 0.664 0.814
## I(transin.perbed.perday * 7) 1.880 0.665
## total.support.staff.perbed 0.165 0.191
## N.dc.perbed -0.154 0.057
## clinical.NHD.perbed -0.168 0.059
## intensivistYes -0.013 0.112
## IMlomean.x -0.055 0.020
## I(avyulos/100) 0.476 0.169
## ratio.agency.total.spend:N.dc.three(4.48,5.51] 0.946 2.151
## ratio.agency.total.spend:N.dc.three(5.51,14.2] 0.796 2.427
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.267
## Residual 1.000
## ---
## number of obs: 36935, groups: trust.code, 63
## AIC = 22543.5, DIC = 22510
## deviance = 22509.5
## Data: dta4
## Subset: ss8
## Models:
## agency.icu: diedicu ~ ratio.agency.total.spend + IMlo + prop.occ.c + admissions.perbed.perday +
## agency.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## agency.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## agency.icu: I(avyulos/100) + (1 | trust.code)
## agency.icu.int3: diedicu ~ ratio.agency.total.spend * N.dc.three + IMlo + prop.occ.c +
## agency.icu.int3: admissions.perbed.perday + I(transin.perbed.perday * 7) +
## agency.icu.int3: total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## agency.icu.int3: intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## agency.icu 13 22537 22648 -11255
## agency.icu.int3 17 22544 22688 -11255 1.35 4 0.85
## glmer(formula = diedicu ~ ratio.senior.junior.wte * N.dc.three +
## IMlo + prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.195 0.324
## ratio.senior.junior.wte -0.192 0.784
## N.dc.three(4.48,5.51] -0.157 0.221
## N.dc.three(5.51,14.2] 0.004 0.212
## IMlo 0.970 0.013
## prop.occ.c 0.202 0.096
## admissions.perbed.perday -0.081 0.927
## I(transin.perbed.perday * 7) 1.644 0.656
## total.support.staff.perbed 0.222 0.192
## N.dc.perbed -0.094 0.062
## clinical.NHD.perbed -0.172 0.060
## intensivistYes -0.037 0.106
## IMlomean.x -0.047 0.021
## I(avyulos/100) 0.327 0.191
## ratio.senior.junior.wte:N.dc.three(4.48,5.51] 0.787 1.201
## ratio.senior.junior.wte:N.dc.three(5.51,14.2] 0.179 0.812
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.274
## Residual 1.000
## ---
## number of obs: 36343, groups: trust.code, 61
## AIC = 22000.2, DIC = 21966
## deviance = 21966.2
## Data: dta4
## Subset: ss8
## Models:
## senior.icu: diedicu ~ ratio.senior.junior.wte + IMlo + prop.occ.c + admissions.perbed.perday +
## senior.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## senior.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## senior.icu: I(avyulos/100) + (1 | trust.code)
## senior.icu.int3: diedicu ~ ratio.senior.junior.wte * N.dc.three + IMlo + prop.occ.c +
## senior.icu.int3: admissions.perbed.perday + I(transin.perbed.perday * 7) +
## senior.icu.int3: total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## senior.icu.int3: intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## senior.icu 13 21993 22104 -10984
## senior.icu.int3 17 22000 22145 -10983 0.99 4 0.91
## glmer(formula = diedicu ~ ratio.diploma * N.dc.three + IMlo +
## prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.271 0.244
## ratio.diploma -1.747 0.450
## N.dc.three(4.48,5.51] -0.337 0.129
## N.dc.three(5.51,14.2] -0.362 0.169
## IMlo 0.965 0.012
## prop.occ.c 0.204 0.093
## admissions.perbed.perday 0.540 0.726
## I(transin.perbed.perday * 7) 1.383 0.559
## total.support.staff.perbed 0.229 0.170
## N.dc.perbed -0.066 0.051
## clinical.NHD.perbed -0.186 0.053
## intensivistYes -0.094 0.093
## IMlomean.x -0.049 0.020
## I(avyulos/100) 0.433 0.147
## ratio.diploma:N.dc.three(4.48,5.51] 2.003 0.606
## ratio.diploma:N.dc.three(5.51,14.2] 2.523 0.572
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.228
## Residual 1.000
## ---
## number of obs: 38168, groups: trust.code, 65
## AIC = 23307.1, DIC = 23273
## deviance = 23273.1
## Data: dta4
## Subset: ss8
## Models:
## diploma.icu: diedicu ~ ratio.diploma + IMlo + prop.occ.c + admissions.perbed.perday +
## diploma.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## diploma.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## diploma.icu: I(avyulos/100) + (1 | trust.code)
## diploma.icu.int3: diedicu ~ ratio.diploma * N.dc.three + IMlo + prop.occ.c + admissions.perbed.perday +
## diploma.icu.int3: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## diploma.icu.int3: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## diploma.icu.int3: I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## diploma.icu 13 23319 23430 -11646
## diploma.icu.int3 17 23307 23452 -11637 19.7 4 0.00056 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## glmer(formula = diedicu ~ ratio.degree * N.dc.three + IMlo +
## prop.occ.c + admissions.perbed.perday + I(transin.perbed.perday *
## 7) + total.support.staff.perbed + N.dc.perbed + clinical.NHD.perbed +
## intensivist + IMlomean.x + I(avyulos/100) + (1 | trust.code),
## data = dta4, family = binomial(), subset = ss8, na.action = na.omit)
## coef.est coef.se
## (Intercept) -1.286 0.257
## ratio.degree -1.517 1.152
## N.dc.three(4.48,5.51] -0.079 0.118
## N.dc.three(5.51,14.2] -0.003 0.161
## IMlo 0.965 0.012
## prop.occ.c 0.213 0.094
## admissions.perbed.perday 0.527 0.782
## I(transin.perbed.perday * 7) 1.660 0.617
## total.support.staff.perbed 0.160 0.183
## N.dc.perbed -0.120 0.054
## clinical.NHD.perbed -0.148 0.058
## intensivistYes -0.047 0.104
## IMlomean.x -0.051 0.020
## I(avyulos/100) 0.434 0.162
## ratio.degree:N.dc.three(4.48,5.51] 0.631 1.643
## ratio.degree:N.dc.three(5.51,14.2] 2.110 1.591
##
## Error terms:
## Groups Name Std.Dev.
## trust.code (Intercept) 0.264
## Residual 1.000
## ---
## number of obs: 38168, groups: trust.code, 65
## AIC = 23323.6, DIC = 23290
## deviance = 23289.6
## Data: dta4
## Subset: ss8
## Models:
## degree.icu: diedicu ~ ratio.degree + IMlo + prop.occ.c + admissions.perbed.perday +
## degree.icu: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## degree.icu: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## degree.icu: I(avyulos/100) + (1 | trust.code)
## degree.icu.int3: diedicu ~ ratio.degree * N.dc.three + IMlo + prop.occ.c + admissions.perbed.perday +
## degree.icu.int3: I(transin.perbed.perday * 7) + total.support.staff.perbed +
## degree.icu.int3: N.dc.perbed + clinical.NHD.perbed + intensivist + IMlomean.x +
## degree.icu.int3: I(avyulos/100) + (1 | trust.code)
## Df AIC BIC logLik Chisq Chi Df Pr(>Chisq)
## degree.icu 13 23319 23430 -11646
## degree.icu.int3 17 23324 23469 -11645 2.95 4 0.57