*Purpose: This program creates the tables for the hospital staffing analysis using alternative ratios and different sets of hospitals.
*Author: Martha Johnson
*Date: 1/13/2020


** Contents **


*1) Exclude H&H hospitals, original ratios
*2) All hospitals, H&H ratios
*3) Exclude H&H hospitals, H&H ratios


*set directories
global input "U:\NY Nurse Project\input"
global output "U:\NY Nurse Project\output"


** 1) Exclude H&H hospitals, original ratios


*load analysis data
use "$input\hospital_data", clear

*drop H&H hospitals
drop if inlist(fac_facid,1165,1172,1186,1294,1301,1438)
drop if inlist(fac_facid,1445,1454,1486,1626,1633,1692)


** Calculate nurse FTE needs from patient days. **


*1 patient per nurse
foreach prefix in pediatric_surgical surgical_adult {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*1)
}

*2 patients per nurse
foreach prefix in critical_care_adult critical_care_peds leveliiiiv_crit_care {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*2)
}

*3 patients per nurse
foreach prefix in medical_peds peds_mixed_acuity medsurg_peds_combd ///
	stepdn_telemtry_peds step_dn_telemtry_adult adult_mixed_acuity ///
	other_ed neonate_mixed_acuity leveli_contg_care levelii_int obstetrics {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*3)
}

*4 patients per nurse
foreach prefix in adult_psych adolescent_psych childadolest_psych child_psych medsurg_combd_adult medical_adult ///
	geropsych specialty_psych other_unit_type_psy behavl_healthchemicl multiple_unit_types {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*4)
}

*5 patients per nurse
foreach prefix in adult_rehab pediatric_rehab rehab_ped_mixd_acuity rehab_mixd_acuity_ad skilled_nursing {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*5)
}

*6 patients per nurse
foreach prefix in well_baby_nursery {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*6)
}


** Calculate actual nurse FTEs from annual nursing patient-care hours **


*FTE nurses in each service unit
foreach var of varlist *_nh {
	gen `var'_act=`var'/(31.5*50)
}
rename *_nh_act *_act


** Calculate gaps **


foreach prefix in adolescent_psych adult_mixed_acuity adult_psych adult_rehab behavl_healthchemicl child_psych childadolest_psych critical_care_adult critical_care_peds geropsych leveli_contg_care levelii_int leveliiiiv_crit_care medsurg_combd_adult medical_adult surgical_adult medical_peds medsurg_peds_combd multiple_unit_types neonate_mixed_acuity obstetrics other_ed other_unit_type_psy pediatric_rehab pediatric_surgical peds_mixed_acuity rehab_mixd_acuity_ad rehab_ped_mixd_acuity skilled_nursing specialty_psych step_dn_telemtry_adult stepdn_telemtry_peds well_baby_nursery {
	gen `prefix'_gap=max(0,`prefix'_need-`prefix'_act)
}


** Estimate RN and LPN shares


*sum nurse hours by nurse type
egen hrs_all=rowtotal(*_rns *_lpns)
egen hrs_rns=rowtotal(*_rns)
egen hrs_lpns=rowtotal(*_lpns)
egen hrs_uaps=rowtotal(*_uaps)

*percentage of each position type
gen pctrn=hrs_rns/hrs_all
gen pctlpn=hrs_lpns/hrs_all

*For two hospitals not reporting nurse hours, replace with average
foreach type in rn lpn {
	sum pct`type'
	local avpct`type'=r(mean)
	replace pct`type'=`avpct`type'' if mi(pct`type')
}

*sum gaps across service units and multiply by rn and lpn percentages
egen totgap=rowtotal(*gap)
gen rnlpn_need=totgap
gen rn_need=totgap*pctrn
gen lpn_need=totgap*pctlpn


** Calculate total FTE nurses and patients


*total FTE nurses across service units:
egen tot_nh=rowtotal(*_nh)
gen fte_nh=tot_nh/(31.5*50)
gen fte_rns=fte_nh*pctrn
gen fte_lpns=fte_nh*pctlpn

*total patient FTEs across all service units
egen tot_pv=rowtotal(*_pv)
gen fte_pats=(tot_pv*24)/(31.5*50)


** Output data for report


** Table 1 **

collapse (count) reporters=beds (sum) beds rnlpn_need rn_need lpn_need fte_rns fte_lpns fte_pats, by(phipregion)

*for appendix table, adjust FTE definition to 37.5 hours of patient care per week
gen rn_need_app=rn_need*(31.5/37.5)
gen lpn_need_app=lpn_need*(31.5/37.5)

*label variables for table
la var phipregion "Region"
la var reporters "Reporters"
la var beds "Certified beds"
la var rnlpn_need "RN/LPN need"
la var rn_need "RN need"
la var lpn_need "LPN need"
la var fte_rns "RNs"
la var fte_lpns "LPNs"
la var fte_pats "Patients"
la var rn_need_app "RN need for appendix"
la var lpn_need_app "LPN need for appendix"

*export table
export excel using "$output\Hospital_Tables", sheet("Table1_nonhh_hosp_orig_ratios", replace) firstrow(varlabels)


** Table 2 **


*drop unnecessary variables
drop fte_pats rnlpn_need *_app

*merge in wage data
merge 1:1 phipregion using "$input\wages", keepusing(hosp_wage_rn hosp_wage_lpn)
keep if _m==3
drop _m

*adjust to 2019 dollars (1.055 is based on first 3 quarters of 2017 and 2019 employment cost index from BLS for health industries, series CIU1026200000000I (K))
foreach var of varlist *wage* {
	replace `var'=`var'*1.055
}

*existing wage costs
gen cost=hosp_wage_rn*fte_rns*36*52+hosp_wage_lpn*fte_lpns*36*52
sum cost
local cost=r(sum)
display `cost'

*new wages, lower bound and upper bound
foreach type in rn lpn {

	*lower bound
	gen wage_cost_old`type's_lb=fte_`type's*0.05*hosp_wage_`type'*36*52
	gen wage_cost_new`type's_lb=`type'_need*1.05*hosp_wage_`type'*36*52
	gen wage_cost_tot`type's_lb=wage_cost_old`type's_lb+wage_cost_new`type's_lb
	
	*upper bound
	gen wage_cost_old`type's_ub=fte_`type's*0.15*hosp_wage_`type'*36*52
	gen wage_cost_new`type's_ub=`type'_need*1.15*hosp_wage_`type'*36*52
	gen wage_cost_tot`type's_ub=wage_cost_old`type's_ub+wage_cost_new`type's_ub
	
}

*sum rn and lpn costs
foreach type1 in old new tot {
	foreach type2 in lb ub {
		gen wage_cost_`type1'_`type2'=wage_cost_`type1'rns_`type2'+wage_cost_`type1'lpns_`type2'
	}
}

*for appendix table, adjust FTE definition to 37.5 hours of patient care per week and 40 hours paid per week
gen wage_cost_tot_lb_app=wage_cost_tot_lb*(31.5/37.5)*(40/36)
gen wage_cost_tot_ub_app=wage_cost_tot_ub*(31.5/37.5)*(40/36)

*format
format wage_cost* %15.0fc

*sort
sort phipregion

*order
keep phipregion wage_cost_old_lb wage_cost_old_ub wage_cost_new_lb wage_cost_new_ub wage_cost_tot_lb wage_cost_tot_ub wage_cost_tot_lb_app wage_cost_tot_ub_app
order phipregion wage_cost_old_lb wage_cost_old_ub wage_cost_new_lb wage_cost_new_ub wage_cost_tot_lb wage_cost_tot_ub wage_cost_tot_lb_app wage_cost_tot_ub_app

*label variables for table
la var phipregion "Region"
la var wage_cost_old_lb "Lower Bound Wage Cost Existing Nurses"
la var wage_cost_new_lb "Lower Bound Wage Cost New Nurses"
la var wage_cost_tot_lb "Lower Bound Wage Cost Total"
la var wage_cost_old_ub "Upper Bound Wage Cost Existing Nurses"
la var wage_cost_new_ub "Upper Bound Wage Cost New Nurses"
la var wage_cost_tot_ub "Upper Bound Wage Cost Total"
la var wage_cost_tot_lb_app "Lower Bound Wage Cost Total for Appendix"
la var wage_cost_tot_ub_app "Upper Bound Wage Cost Total for Appendix"

*export table
export excel using "$output\Hospital_Tables", sheet("Table2_nonhh_hosp_orig_ratios", replace) firstrow(varlabels)


** 2) All hospitals, H&H ratios


*load analysis data
use "$input\hospital_data", clear


** Calculate nurse FTE needs from patient days. **


*1 patient per nurse
foreach prefix in pediatric_surgical surgical_adult {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*1)
}

*2 patients per nurse
foreach prefix in critical_care_adult critical_care_peds leveliiiiv_crit_care ///
	obstetrics {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*2)
}

*3 patients per nurse
foreach prefix in neonate_mixed_acuity leveli_contg_care levelii_int  {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*3)
}

*4 patients per nurse
foreach prefix in step_dn_telemtry_adult stepdn_telemtry_peds {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*4)
}

*5 patients per nurse
foreach prefix in other_ed {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*5)
}

*6 patients per nurse
foreach prefix in skilled_nursing well_baby_nursery medsurg_combd_adult medical_adult ///
	medical_peds medsurg_peds_combd peds_mixed_acuity adult_mixed_acuity {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*6)
}

*7 patients per nurse
foreach prefix in adult_rehab pediatric_rehab rehab_ped_mixd_acuity ///
	rehab_mixd_acuity_ad adult_psych adolescent_psych childadolest_psych ///
	child_psych geropsych specialty_psych other_unit_type_psy ///
	behavl_healthchemicl multiple_unit_types {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*7)
}


** Calculate actual nurse FTEs from annual nursing patient-care hours **


*FTE nurses in each service unit
foreach var of varlist *_nh {
	gen `var'_act=`var'/(31.5*50)
}
rename *_nh_act *_act


** Calculate gaps **


foreach prefix in adolescent_psych adult_mixed_acuity adult_psych adult_rehab behavl_healthchemicl child_psych childadolest_psych critical_care_adult critical_care_peds geropsych leveli_contg_care levelii_int leveliiiiv_crit_care medsurg_combd_adult medical_adult surgical_adult medical_peds medsurg_peds_combd multiple_unit_types neonate_mixed_acuity obstetrics other_ed other_unit_type_psy pediatric_rehab pediatric_surgical peds_mixed_acuity rehab_mixd_acuity_ad rehab_ped_mixd_acuity skilled_nursing specialty_psych step_dn_telemtry_adult stepdn_telemtry_peds well_baby_nursery {
	gen `prefix'_gap=max(0,`prefix'_need-`prefix'_act)
}


** Estimate RN and LPN shares


*sum nurse hours by nurse type
egen hrs_all=rowtotal(*_rns *_lpns)
egen hrs_rns=rowtotal(*_rns)
egen hrs_lpns=rowtotal(*_lpns)
egen hrs_uaps=rowtotal(*_uaps)

*percentage of each position type
gen pctrn=hrs_rns/hrs_all
gen pctlpn=hrs_lpns/hrs_all

*For two hospitals not reporting nurse hours, replace with average
foreach type in rn lpn {
	sum pct`type'
	local avpct`type'=r(mean)
	replace pct`type'=`avpct`type'' if mi(pct`type')
}

*sum gaps across service units and multiply by rn and lpn percentages
egen totgap=rowtotal(*gap)
gen rnlpn_need=totgap
gen rn_need=totgap*pctrn
gen lpn_need=totgap*pctlpn


** Calculate total FTE nurses and patients


*total FTE nurses across service units:
egen tot_nh=rowtotal(*_nh)
gen fte_nh=tot_nh/(31.5*50)
gen fte_rns=fte_nh*pctrn
gen fte_lpns=fte_nh*pctlpn

*total patient FTEs across all service units
egen tot_pv=rowtotal(*_pv)
gen fte_pats=(tot_pv*24)/(31.5*50)


** Output data for report


** Table 1 **


collapse (count) reporters=beds (sum) beds rnlpn_need rn_need lpn_need fte_rns fte_lpns fte_pats, by(phipregion)

*for appendix table, adjust FTE definition to 37.5 hours of patient care per week
gen rn_need_app=rn_need*(31.5/37.5)
gen lpn_need_app=lpn_need*(31.5/37.5)

*label variables for table
la var phipregion "Region"
la var reporters "Reporters"
la var beds "Certified beds"
la var rnlpn_need "RN/LPN need"
la var rn_need "RN need"
la var lpn_need "LPN need"
la var fte_rns "RNs"
la var fte_lpns "LPNs"
la var fte_pats "Patients"
la var rn_need_app "RN need for appendix"
la var lpn_need_app "LPN need for appendix"

*export table
export excel using "$output\Hospital_Tables", sheet("Table1_all_hosp_hh_ratios", replace) firstrow(varlabels)


** Table 2 **


*drop unnecessary variables
drop fte_pats rnlpn_need *_app

*merge in wage data
merge 1:1 phipregion using "$input\wages", keepusing(hosp_wage_rn hosp_wage_lpn)
keep if _m==3
drop _m

*adjust to 2019 dollars (1.055 is based on first 3 quarters of 2017 and 2019 employment cost index from BLS for health industries, series CIU1026200000000I (K))
foreach var of varlist *wage* {
	replace `var'=`var'*1.055
}

*existing wage costs
gen cost=hosp_wage_rn*fte_rns*36*52+hosp_wage_lpn*fte_lpns*36*52
sum cost
local cost=r(sum)
display `cost'

*new wages, lower bound and upper bound
foreach type in rn lpn {

	*lower bound
	gen wage_cost_old`type's_lb=fte_`type's*0.05*hosp_wage_`type'*36*52
	gen wage_cost_new`type's_lb=`type'_need*1.05*hosp_wage_`type'*36*52
	gen wage_cost_tot`type's_lb=wage_cost_old`type's_lb+wage_cost_new`type's_lb
	
	*upper bound
	gen wage_cost_old`type's_ub=fte_`type's*0.15*hosp_wage_`type'*36*52
	gen wage_cost_new`type's_ub=`type'_need*1.15*hosp_wage_`type'*36*52
	gen wage_cost_tot`type's_ub=wage_cost_old`type's_ub+wage_cost_new`type's_ub
	
}

*sum rn and lpn costs
foreach type1 in old new tot {
	foreach type2 in lb ub {
		gen wage_cost_`type1'_`type2'=wage_cost_`type1'rns_`type2'+wage_cost_`type1'lpns_`type2'
	}
}

*for appendix table, adjust FTE definition to 37.5 hours of patient care per week and 40 hours paid per week
gen wage_cost_tot_lb_app=wage_cost_tot_lb*(31.5/37.5)*(40/36)
gen wage_cost_tot_ub_app=wage_cost_tot_ub*(31.5/37.5)*(40/36)

*format
format wage_cost* %15.0fc

*sort
sort phipregion

*order
keep phipregion wage_cost_old_lb wage_cost_old_ub wage_cost_new_lb wage_cost_new_ub wage_cost_tot_lb wage_cost_tot_ub wage_cost_tot_lb_app wage_cost_tot_ub_app
order phipregion wage_cost_old_lb wage_cost_old_ub wage_cost_new_lb wage_cost_new_ub wage_cost_tot_lb wage_cost_tot_ub wage_cost_tot_lb_app wage_cost_tot_ub_app

*label variables for table
la var phipregion "Region"
la var wage_cost_old_lb "Lower Bound Wage Cost Existing Nurses"
la var wage_cost_new_lb "Lower Bound Wage Cost New Nurses"
la var wage_cost_tot_lb "Lower Bound Wage Cost Total"
la var wage_cost_old_ub "Upper Bound Wage Cost Existing Nurses"
la var wage_cost_new_ub "Upper Bound Wage Cost New Nurses"
la var wage_cost_tot_ub "Upper Bound Wage Cost Total"
la var wage_cost_tot_lb_app "Lower Bound Wage Cost Total for Appendix"
la var wage_cost_tot_ub_app "Upper Bound Wage Cost Total for Appendix"

*export table
export excel using "$output\Hospital_Tables", sheet("Table2_all_hosp_hh_ratios", replace) firstrow(varlabels)


** 3) Exclude H&H hospitals, H&H ratios


*load analysis data
use "$input\hospital_data", clear

*drop H&H hospitals
drop if inlist(fac_facid,1165,1172,1186,1294,1301,1438)
drop if inlist(fac_facid,1445,1454,1486,1626,1633,1692)

** Calculate nurse FTE needs from patient days. **


*1 patient per nurse
foreach prefix in pediatric_surgical surgical_adult {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*1)
}

*2 patients per nurse
foreach prefix in critical_care_adult critical_care_peds leveliiiiv_crit_care ///
	obstetrics {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*2)
}

*3 patients per nurse
foreach prefix in neonate_mixed_acuity leveli_contg_care levelii_int  {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*3)
}

*4 patients per nurse
foreach prefix in step_dn_telemtry_adult stepdn_telemtry_peds {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*4)
}

*5 patients per nurse
foreach prefix in other_ed {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*5)
}

*6 patients per nurse
foreach prefix in skilled_nursing well_baby_nursery medsurg_combd_adult medical_adult ///
	medical_peds medsurg_peds_combd peds_mixed_acuity adult_mixed_acuity {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*6)
}

*7 patients per nurse
foreach prefix in adult_rehab pediatric_rehab rehab_ped_mixd_acuity ///
	rehab_mixd_acuity_ad adult_psych adolescent_psych childadolest_psych ///
	child_psych geropsych specialty_psych other_unit_type_psy ///
	behavl_healthchemicl multiple_unit_types {
	gen `prefix'_need=`prefix'_pv*24/(31.5*50*7)
}


** Calculate actual nurse FTEs from annual nursing patient-care hours **


*FTE nurses in each service unit
foreach var of varlist *_nh {
	gen `var'_act=`var'/(31.5*50)
}
rename *_nh_act *_act


** Calculate gaps **


foreach prefix in adolescent_psych adult_mixed_acuity adult_psych adult_rehab behavl_healthchemicl child_psych childadolest_psych critical_care_adult critical_care_peds geropsych leveli_contg_care levelii_int leveliiiiv_crit_care medsurg_combd_adult medical_adult surgical_adult medical_peds medsurg_peds_combd multiple_unit_types neonate_mixed_acuity obstetrics other_ed other_unit_type_psy pediatric_rehab pediatric_surgical peds_mixed_acuity rehab_mixd_acuity_ad rehab_ped_mixd_acuity skilled_nursing specialty_psych step_dn_telemtry_adult stepdn_telemtry_peds well_baby_nursery {
	gen `prefix'_gap=max(0,`prefix'_need-`prefix'_act)
}


** Estimate RN and LPN shares


*sum nurse hours by nurse type
egen hrs_all=rowtotal(*_rns *_lpns)
egen hrs_rns=rowtotal(*_rns)
egen hrs_lpns=rowtotal(*_lpns)
egen hrs_uaps=rowtotal(*_uaps)

*percentage of each position type
gen pctrn=hrs_rns/hrs_all
gen pctlpn=hrs_lpns/hrs_all

*For two hospitals not reporting nurse hours, replace with average
foreach type in rn lpn {
	sum pct`type'
	local avpct`type'=r(mean)
	replace pct`type'=`avpct`type'' if mi(pct`type')
}

*sum gaps across service units and multiply by rn and lpn percentages
egen totgap=rowtotal(*gap)
gen rnlpn_need=totgap
gen rn_need=totgap*pctrn
gen lpn_need=totgap*pctlpn


** Calculate total FTE nurses and patients


*total FTE nurses across service units:
egen tot_nh=rowtotal(*_nh)
gen fte_nh=tot_nh/(31.5*50)
gen fte_rns=fte_nh*pctrn
gen fte_lpns=fte_nh*pctlpn

*total patient FTEs across all service units
egen tot_pv=rowtotal(*_pv)
gen fte_pats=(tot_pv*24)/(31.5*50)


** Output data for report


** Table 1 **

collapse (count) reporters=beds (sum) beds rnlpn_need rn_need lpn_need fte_rns fte_lpns fte_pats, by(phipregion)

*for appendix table, adjust FTE definition to 37.5 hours of patient care per week
gen rn_need_app=rn_need*(31.5/37.5)
gen lpn_need_app=lpn_need*(31.5/37.5)

*label variables for table
la var phipregion "Region"
la var reporters "Reporters"
la var beds "Certified beds"
la var rnlpn_need "RN/LPN need"
la var rn_need "RN need"
la var lpn_need "LPN need"
la var fte_rns "RNs"
la var fte_lpns "LPNs"
la var fte_pats "Patients"
la var rn_need_app "RN need for appendix"
la var lpn_need_app "LPN need for appendix"

*export table
export excel using "$output\Hospital_Tables", sheet("Table1_nonhh_hosp_hh_ratios", replace) firstrow(varlabels)


** Table 2 **


*drop unnecessary variables
drop fte_pats rnlpn_need *_app

*merge in wage data
merge 1:1 phipregion using "$input\wages", keepusing(hosp_wage_rn hosp_wage_lpn)
keep if _m==3
drop _m

*adjust to 2019 dollars (1.055 is based on first 3 quarters of 2017 and 2019 employment cost index from BLS for health industries, series CIU1026200000000I (K))
foreach var of varlist *wage* {
	replace `var'=`var'*1.055
}

*existing wage costs
gen cost=hosp_wage_rn*fte_rns*36*52+hosp_wage_lpn*fte_lpns*36*52
sum cost
local cost=r(sum)
display `cost'

*new wages, lower bound and upper bound
foreach type in rn lpn {

	*lower bound
	gen wage_cost_old`type's_lb=fte_`type's*0.05*hosp_wage_`type'*36*52
	gen wage_cost_new`type's_lb=`type'_need*1.05*hosp_wage_`type'*36*52
	gen wage_cost_tot`type's_lb=wage_cost_old`type's_lb+wage_cost_new`type's_lb
	
	*upper bound
	gen wage_cost_old`type's_ub=fte_`type's*0.15*hosp_wage_`type'*36*52
	gen wage_cost_new`type's_ub=`type'_need*1.15*hosp_wage_`type'*36*52
	gen wage_cost_tot`type's_ub=wage_cost_old`type's_ub+wage_cost_new`type's_ub
	
}

*sum rn and lpn costs
foreach type1 in old new tot {
	foreach type2 in lb ub {
		gen wage_cost_`type1'_`type2'=wage_cost_`type1'rns_`type2'+wage_cost_`type1'lpns_`type2'
	}
}

*for appendix table, adjust FTE definition to 37.5 hours of patient care per week and 40 hours paid per week
gen wage_cost_tot_lb_app=wage_cost_tot_lb*(31.5/37.5)*(40/36)
gen wage_cost_tot_ub_app=wage_cost_tot_ub*(31.5/37.5)*(40/36)

*format
format wage_cost* %15.0fc

*sort
sort phipregion

*order
keep phipregion wage_cost_old_lb wage_cost_old_ub wage_cost_new_lb wage_cost_new_ub wage_cost_tot_lb wage_cost_tot_ub wage_cost_tot_lb_app wage_cost_tot_ub_app
order phipregion wage_cost_old_lb wage_cost_old_ub wage_cost_new_lb wage_cost_new_ub wage_cost_tot_lb wage_cost_tot_ub wage_cost_tot_lb_app wage_cost_tot_ub_app

*label variables for table
la var phipregion "Region"
la var wage_cost_old_lb "Lower Bound Wage Cost Existing Nurses"
la var wage_cost_new_lb "Lower Bound Wage Cost New Nurses"
la var wage_cost_tot_lb "Lower Bound Wage Cost Total"
la var wage_cost_old_ub "Upper Bound Wage Cost Existing Nurses"
la var wage_cost_new_ub "Upper Bound Wage Cost New Nurses"
la var wage_cost_tot_ub "Upper Bound Wage Cost Total"
la var wage_cost_tot_lb_app "Lower Bound Wage Cost Total for Appendix"
la var wage_cost_tot_ub_app "Upper Bound Wage Cost Total for Appendix"

*export table
export excel using "$output\Hospital_Tables", sheet("Table2_nonhh_hosp_hh_ratios", replace) firstrow(varlabels)