Analysis of LSOA results: mRS mean shift

file_name = 'lsoa_base'
limit_to_england = False

data_field = 'mean_shift'
data_name = 'mRS mean shift'

Proportions of stroke types: Analysis of SAMueL data

Data from SAMueL using NIHSS 11+ as a surrogate for LVO:

Admission type	All arrivals	Arrival within 6 hrs known onset	Arrival within 4 hrs known onset
Proportion all admissions	100	42.9	37.1
Proportion haemorrhagic	11.5	13.6	14.1
Proportion ischaemic	88.5	86.4	85.9
Proportion ischaemic with NIHSS 0-10	74.9	67.4	65.7
Proportion ischaemic with NIHSS 11+	25.1	32.6	34.3

For original analysis see: https://samuel-book.github.io/samuel-1/descriptive_stats/10_using_nihss_10_for_lvo.html

Import packages

import contextily as ctx
import geopandas
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import os

pd.set_option('display.max_rows', 150)

Load data

Load shape file

filename = "zip://./data/Lower_layer_super_output_areas_(E+W)_2011_Boundaries_(Generalised_Clipped)_V2.zip"
lsoa_gdf = geopandas.read_file(filename, crs='EPSG:27700')
lsoa_gdf.head()

	LSOA11CD	LSOA11NM	LSOA11NMW	geometry
0	E01000001	City of London 001A	City of London 001A	POLYGON ((532105.092 182011.230, 532162.491 18...
1	E01000002	City of London 001B	City of London 001B	POLYGON ((532746.813 181786.891, 532671.688 18...
2	E01000003	City of London 001C	City of London 001C	POLYGON ((532135.145 182198.119, 532158.250 18...
3	E01000005	City of London 001E	City of London 001E	POLYGON ((533807.946 180767.770, 533649.063 18...
4	E01000006	Barking and Dagenham 016A	Barking and Dagenham 016A	POLYGON ((545122.049 184314.931, 545271.917 18...

# Load country outline
outline = geopandas.read_file('./data/Countries_(December_2022)_GB_BFC.zip')
if limit_to_england:
    mask = outline['CTRY22NM'] == 'England'
    outline = outline[mask]
else:
    mask = (outline['CTRY22NM'] == 'England') | (outline['CTRY22NM'] == 'Wales')
    outline = outline[mask]

outline

	FID	CTRY22CD	CTRY22NM	CTRY22NMW	BNG_E	BNG_N	LONG	LAT	GlobalID	SHAPE_Leng	SHAPE_Area	geometry
0	1	E92000001	England	Lloegr	394883	370883	-2.07811	53.23497	{096C3540-B006-4D2D-BCDF-B3E945B80D51}	1.358939e+07	1.304623e+11	MULTIPOLYGON (((83962.840 5401.150, 83970.680 ...
2	3	W92000004	Wales	Cymru	263405	242881	-3.99417	52.06741	{1DFA050A-C94B-4E58-8210-DE8274C7C33E}	3.273688e+06	2.078260e+10	MULTIPOLYGON (((322081.699 165165.901, 322082....

Load hospital info

hospitals_gdf = geopandas.read_file('./data/stroke_hospitals_2022.csv')
hospitals_gdf.geometry = geopandas.points_from_xy(
        hospitals_gdf.Easting, hospitals_gdf.Northing)
# Set crs for geometry, using epsg4326 for lat/long
hospitals_gdf = hospitals_gdf.set_crs(epsg=27700)

if limit_to_england:
    mask = hospitals_gdf['Country'] ==  'England'
    hospitals_gdf = hospitals_gdf[mask]
hospitals_gdf.head()

	Postcode	Hospital_name	Use_IVT	Use_MT	Use_MSU	Country	Strategic Clinical Network	Health Board / Trust	Stroke Team	SSNAP name	...	ivt_rate	Easting	Northing	long	lat	Neuroscience	30 England Thrombectomy Example	hospital_city	Notes	geometry
0	RM70AG	RM70AG	1	1	1	England	London SCN	Barking	Havering and Redbridge University Hospitals N...	Queens Hospital Romford HASU	...	11.9	551118	187780	0.179030640661934	51.5686465521504	1	0	Romford		POINT (551118.000 187780.000)
1	E11BB	E11BB	1	1	1	England	London SCN	Barts Health NHS Trust	The Royal London Hospital	Royal London Hospital HASU	...	13.4	534829	181798	-0.0581329916047372	51.5190178361295	1	1	Royal London		POINT (534829.000 181798.000)
2	SW66SX	SW66SX	1	1	1	England	London SCN	Imperial College Healthcare NHS Trust	Charing Cross Hospital, London	Charing Cross Hospital HASU	...	9.9	524226	176487	-0.212736111308184	51.4737165811988	1	1	Charing Cross		POINT (524226.000 176487.000)
3	SE59RW	SE59RW	1	1	1	England	London SCN	King's College Hospital NHS Foundation Trust	King's College Hospital, London	King's College Hospital HASU	...	15	532536	176228	-0.0932514945350423	51.4695052907757	1	0	Kings College		POINT (532536.000 176228.000)
4	BR68ND	BR68ND	1	0	0	England	London SCN	King's College Hospital NHS Foundation Trust	Princess Royal University Hospital	Princess Royal University Hospital HASU	...	13.3	543443	165032	0.0591464390000555	51.3662434074157	0	0	Princess Royal		POINT (543443.000 165032.000)

5 rows × 22 columns

Load LSOA model output data

lsoa_data = pd.read_csv(f'./output/{file_name}.csv')
lsoa_data.head(3).T

	0	1	2
lsoa	Adur 001A	Adur 001B	Adur 001C
closest_ivt_unit	BN25BE	BN25BE	BN112DH
closest_ivt_time	17.6	18.7	17.6
closest_mt_unit	BN25BE	BN25BE	BN25BE
closest_mt_time	17.6	18.7	19.8
transfer_mt_unit	BN25BE	BN25BE	BN25BE
transfer_mt_time	0.0	0.0	31.6
mt_transfer_required	False	False	True
msu_unit	BN25BE	BN25BE	BN25BE
msu_time	17.6	18.7	19.8
ivt_drip_ship	107.6	108.7	107.6
ivt_mothership	107.6	108.7	109.8
mt_drip_ship	167.6	168.7	259.2
mt_mothership	167.6	168.7	169.8
drip_ship_lvo_untreated_probs	[0.05 0.08 0.14 0.16 0.24 0.14 0.19]	[0.05 0.08 0.14 0.16 0.24 0.14 0.19]	[0.05 0.08 0.14 0.16 0.24 0.14 0.19]
drip_ship_nlvo_untreated_probs	[0.2 0.26 0.12 0.13 0.14 0.06 0.09]	[0.2 0.26 0.12 0.13 0.14 0.06 0.09]	[0.2 0.26 0.12 0.13 0.14 0.06 0.09]
drip_ship_lvo_ivt_probs	[0.09 0.09 0.13 0.15 0.22 0.12 0.2 ]	[0.09 0.09 0.13 0.15 0.22 0.12 0.2 ]	[0.09 0.09 0.13 0.15 0.22 0.12 0.2 ]
drip_ship_lvo_mt_probs	[0.18 0.12 0.14 0.17 0.18 0.09 0.12]	[0.18 0.12 0.14 0.17 0.18 0.09 0.12]	[0.13 0.11 0.14 0.17 0.2 0.1 0.15]
drip_ship_nlvo_ivt_probs	[0.36 0.22 0.11 0.12 0.1 0.03 0.06]	[0.35 0.23 0.11 0.12 0.1 0.03 0.06]	[0.36 0.22 0.11 0.12 0.1 0.03 0.06]
drip_ship_lvo_untreated_mean_utility	0.3319	0.3319	0.3319
drip_ship_nlvo_untreated_mean_utility	0.5997	0.5997	0.5997
drip_ship_lvo_ivt_mean_utility	0.3664	0.3664	0.3664
drip_ship_lvo_mt_mean_utility	0.4962	0.4962	0.441
drip_ship_nlvo_ivt_mean_utility	0.7045	0.7036	0.7045
drip_ship_lvo_ivt_added_utility	0.0345	0.0345	0.0345
drip_ship_lvo_mt_added_utility	0.1643	0.1643	0.1091
drip_ship_nlvo_ivt_added_utility	0.1048	0.1039	0.1048
drip_ship_lvo_untreated_cum_probs	[0.05 0.13 0.27 0.43 0.67 0.81 1. ]	[0.05 0.13 0.27 0.43 0.67 0.81 1. ]	[0.05 0.13 0.27 0.43 0.67 0.81 1. ]
drip_ship_nlvo_untreated_cum_probs	[0.2 0.46 0.58 0.71 0.85 0.91 1. ]	[0.2 0.46 0.58 0.71 0.85 0.91 1. ]	[0.2 0.46 0.58 0.71 0.85 0.91 1. ]
drip_ship_lvo_ivt_cum_probs	[0.09 0.18 0.31 0.46 0.68 0.8 1. ]	[0.09 0.18 0.31 0.46 0.68 0.8 1. ]	[0.09 0.18 0.31 0.46 0.68 0.8 1. ]
drip_ship_lvo_mt_cum_probs	[0.18 0.3 0.44 0.61 0.79 0.88 1. ]	[0.18 0.3 0.44 0.61 0.79 0.88 1. ]	[0.13 0.24 0.38 0.55 0.75 0.85 1. ]
drip_ship_nlvo_ivt_cum_probs	[0.36 0.58 0.69 0.81 0.91 0.94 1. ]	[0.35 0.58 0.69 0.81 0.91 0.94 1. ]	[0.36 0.58 0.69 0.81 0.91 0.94 1. ]
drip_ship_lvo_untreated_mrs<=2	0.27	0.27	0.27
drip_ship_nlvo_untreated_mrs<=2	0.58	0.58	0.58
drip_ship_lvo_ivt_mrs<=2	0.31	0.31	0.31
drip_ship_lvo_mt_mrs<=2	0.44	0.44	0.38
drip_ship_nlvo_ivt_mrs<=2	0.69	0.69	0.69
drip_ship_lvo_ivt_shift	[ 0.04 0.01 -0.01 -0.01 -0.02 -0.02 0.01]	[ 0.04 0.01 -0.01 -0.01 -0.02 -0.02 0.01]	[ 0.04 0.01 -0.01 -0.01 -0.02 -0.02 0.01]
drip_ship_lvo_mt_shift	[ 0.13 0.04 0. 0.01 -0.06 -0.05 -0.07]	[ 0.13 0.04 0. 0.01 -0.06 -0.05 -0.07]	[ 0.08 0.03 0. 0.01 -0.04 -0.04 -0.04]
drip_ship_nlvo_ivt_shift	[ 0.16 -0.04 -0.01 -0.01 -0.04 -0.03 -0.03]	[ 0.15 -0.03 -0.01 -0.01 -0.04 -0.03 -0.03]	[ 0.16 -0.04 -0.01 -0.01 -0.04 -0.03 -0.03]
drip_ship_lvo_untreated_mean_mRS	3.64	3.64	3.64
drip_ship_nlvo_untreated_mean_mRS	2.29	2.29	2.29
drip_ship_lvo_ivt_mean_mRS	3.48	3.48	3.48
drip_ship_lvo_mt_mean_mRS	2.8	2.8	3.1
drip_ship_nlvo_ivt_mean_mRS	1.71	1.72	1.71
drip_ship_lvo_ivt_mean_shift	-0.16	-0.16	-0.16
drip_ship_lvo_mt_mean_shift	-0.84	-0.84	-0.54
drip_ship_nlvo_ivt_mean_shift	-0.58	-0.57	-0.58
drip_ship_lvo_ivt_improved	0.17	0.17	0.17
drip_ship_lvo_mt_improved	0.75	0.75	0.54
drip_ship_nlvo_ivt_improved	0.58	0.57	0.58
mothership_lvo_untreated_probs	[0.05 0.08 0.14 0.16 0.24 0.14 0.19]	[0.05 0.08 0.14 0.16 0.24 0.14 0.19]	[0.05 0.08 0.14 0.16 0.24 0.14 0.19]
mothership_nlvo_untreated_probs	[0.2 0.26 0.12 0.13 0.14 0.06 0.09]	[0.2 0.26 0.12 0.13 0.14 0.06 0.09]	[0.2 0.26 0.12 0.13 0.14 0.06 0.09]
mothership_lvo_ivt_probs	[0.09 0.09 0.13 0.15 0.22 0.12 0.2 ]	[0.09 0.09 0.13 0.15 0.22 0.12 0.2 ]	[0.09 0.09 0.13 0.15 0.22 0.12 0.2 ]
mothership_lvo_mt_probs	[0.18 0.12 0.14 0.17 0.18 0.09 0.12]	[0.18 0.12 0.14 0.17 0.18 0.09 0.12]	[0.18 0.12 0.14 0.17 0.18 0.09 0.12]
mothership_nlvo_ivt_probs	[0.36 0.22 0.11 0.12 0.1 0.03 0.06]	[0.35 0.23 0.11 0.12 0.1 0.03 0.06]	[0.35 0.23 0.11 0.12 0.1 0.03 0.06]
mothership_lvo_untreated_mean_utility	0.3319	0.3319	0.3319
mothership_nlvo_untreated_mean_utility	0.5997	0.5997	0.5997
mothership_lvo_ivt_mean_utility	0.3664	0.3664	0.3664
mothership_lvo_mt_mean_utility	0.4962	0.4962	0.4962
mothership_nlvo_ivt_mean_utility	0.7045	0.7036	0.7036
mothership_lvo_ivt_added_utility	0.0345	0.0345	0.0345
mothership_lvo_mt_added_utility	0.1643	0.1643	0.1643
mothership_nlvo_ivt_added_utility	0.1048	0.1039	0.1039
mothership_lvo_untreated_cum_probs	[0.05 0.13 0.27 0.43 0.67 0.81 1. ]	[0.05 0.13 0.27 0.43 0.67 0.81 1. ]	[0.05 0.13 0.27 0.43 0.67 0.81 1. ]
mothership_nlvo_untreated_cum_probs	[0.2 0.46 0.58 0.71 0.85 0.91 1. ]	[0.2 0.46 0.58 0.71 0.85 0.91 1. ]	[0.2 0.46 0.58 0.71 0.85 0.91 1. ]
mothership_lvo_ivt_cum_probs	[0.09 0.18 0.31 0.46 0.68 0.8 1. ]	[0.09 0.18 0.31 0.46 0.68 0.8 1. ]	[0.09 0.18 0.31 0.46 0.68 0.8 1. ]
mothership_lvo_mt_cum_probs	[0.18 0.3 0.44 0.61 0.79 0.88 1. ]	[0.18 0.3 0.44 0.61 0.79 0.88 1. ]	[0.18 0.3 0.44 0.61 0.79 0.88 1. ]
mothership_nlvo_ivt_cum_probs	[0.36 0.58 0.69 0.81 0.91 0.94 1. ]	[0.35 0.58 0.69 0.81 0.91 0.94 1. ]	[0.35 0.58 0.69 0.81 0.91 0.94 1. ]
mothership_lvo_untreated_mrs<=2	0.27	0.27	0.27
mothership_nlvo_untreated_mrs<=2	0.58	0.58	0.58
mothership_lvo_ivt_mrs<=2	0.31	0.31	0.31
mothership_lvo_mt_mrs<=2	0.44	0.44	0.44
mothership_nlvo_ivt_mrs<=2	0.69	0.69	0.69
mothership_lvo_ivt_shift	[ 0.04 0.01 -0.01 -0.01 -0.02 -0.02 0.01]	[ 0.04 0.01 -0.01 -0.01 -0.02 -0.02 0.01]	[ 0.04 0.01 -0.01 -0.01 -0.02 -0.02 0.01]
mothership_lvo_mt_shift	[ 0.13 0.04 0. 0.01 -0.06 -0.05 -0.07]	[ 0.13 0.04 0. 0.01 -0.06 -0.05 -0.07]	[ 0.13 0.04 0. 0.01 -0.06 -0.05 -0.07]
mothership_nlvo_ivt_shift	[ 0.16 -0.04 -0.01 -0.01 -0.04 -0.03 -0.03]	[ 0.15 -0.03 -0.01 -0.01 -0.04 -0.03 -0.03]	[ 0.15 -0.03 -0.01 -0.01 -0.04 -0.03 -0.03]
mothership_lvo_untreated_mean_mRS	3.64	3.64	3.64
mothership_nlvo_untreated_mean_mRS	2.29	2.29	2.29
mothership_lvo_ivt_mean_mRS	3.48	3.48	3.48
mothership_lvo_mt_mean_mRS	2.8	2.8	2.8
mothership_nlvo_ivt_mean_mRS	1.71	1.72	1.72
mothership_lvo_ivt_mean_shift	-0.16	-0.16	-0.16
mothership_lvo_mt_mean_shift	-0.84	-0.84	-0.84
mothership_nlvo_ivt_mean_shift	-0.58	-0.57	-0.57
mothership_lvo_ivt_improved	0.17	0.17	0.17
mothership_lvo_mt_improved	0.75	0.75	0.75
mothership_nlvo_ivt_improved	0.58	0.57	0.57
drip_ship_nlvo_ivt_added_mrs<=2	0.11	0.11	0.11
drip_ship_lvo_ivt_added_mrs<=2	0.04	0.04	0.04
drip_ship_lvo_mt_added_mrs<=2	0.17	0.17	0.11
mothership_nlvo_ivt_added_mrs<=2	0.11	0.11	0.11
mothership_lvo_ivt_added_mrs<=2	0.04	0.04	0.04
mothership_lvo_mt_added_mrs<=2	0.17	0.17	0.17
weighted_drip_ship_added_utility	0.031414	0.031311	0.025895
weighted_mothership_added_utility	0.031414	0.031311	0.031311
weighted_drip_ship_added_mrs<=2	0.03305	0.03305	0.02705
weighted_mothership_added_mrs<=2	0.03305	0.03305	0.03305
weighted_drip_ship_mean_shift	-0.1643	-0.16315	-0.1343
weighted_mothership_mean_shift	-0.1643	-0.16315	-0.16315
LSOA_predicted_admissions	2.550427	2.550427	2.550427

# Merge with shape file

lsoa_data_gdf = lsoa_gdf.merge(lsoa_data, left_on='LSOA11NM', right_on='lsoa', how='right')
lsoa_data_gdf.head()

	LSOA11CD	LSOA11NM	LSOA11NMW	geometry	lsoa	closest_ivt_unit	closest_ivt_time	closest_mt_unit	closest_mt_time	transfer_mt_unit	...	mothership_nlvo_ivt_added_mrs<=2	mothership_lvo_ivt_added_mrs<=2	mothership_lvo_mt_added_mrs<=2	weighted_drip_ship_added_utility	weighted_mothership_added_utility	weighted_drip_ship_added_mrs<=2	weighted_mothership_added_mrs<=2	weighted_drip_ship_mean_shift	weighted_mothership_mean_shift	LSOA_predicted_admissions
0	E01031349	Adur 001A	Adur 001A	POLYGON ((525328.061 105725.349, 525436.843 10...	Adur 001A	BN25BE	17.6	BN25BE	17.6	BN25BE	...	0.11	0.04	0.17	0.031414	0.031414	0.03305	0.03305	-0.16430	-0.16430	2.550427
1	E01031350	Adur 001B	Adur 001B	POLYGON ((524713.079 106781.551, 524753.189 10...	Adur 001B	BN25BE	18.7	BN25BE	18.7	BN25BE	...	0.11	0.04	0.17	0.031311	0.031311	0.03305	0.03305	-0.16315	-0.16315	2.550427
2	E01031351	Adur 001C	Adur 001C	POLYGON ((523976.094 108414.594, 524038.094 10...	Adur 001C	BN112DH	17.6	BN25BE	19.8	BN25BE	...	0.11	0.04	0.17	0.025895	0.031311	0.02705	0.03305	-0.13430	-0.16315	2.550427
3	E01031352	Adur 001D	Adur 001D	POLYGON ((524263.733 106628.407, 524317.000 10...	Adur 001D	BN112DH	17.6	BN25BE	19.8	BN25BE	...	0.11	0.04	0.17	0.025895	0.031311	0.02705	0.03305	-0.13430	-0.16315	2.550427
4	E01031370	Adur 001E	Adur 001E	POLYGON ((522978.319 106148.604, 523260.859 10...	Adur 001E	BN112DH	16.5	BN25BE	19.8	BN25BE	...	0.11	0.04	0.17	0.026245	0.031311	0.02705	0.03305	-0.13530	-0.16315	2.550427

5 rows × 105 columns

Calculate difference between Mothership and Drip and Ship

lsoa_data_gdf[f'nlvo_ivt_{data_field}_mothership_minus_dripship'] = \
    lsoa_data_gdf[f'mothership_nlvo_ivt_{data_field}'] - lsoa_data_gdf[f'drip_ship_nlvo_ivt_{data_field}']

lsoa_data_gdf[f'lvo_mt_{data_field}_mothership_minus_dripship'] = \
    lsoa_data_gdf[f'mothership_lvo_mt_{data_field}'] - lsoa_data_gdf[f'drip_ship_lvo_mt_{data_field}']

Plots

fig = plt.figure(figsize=(18,12))

################### Plot data values ###################

cols = [f'drip_ship_nlvo_ivt_{data_field}',
    f'mothership_nlvo_ivt_{data_field}',
    f'drip_ship_lvo_mt_{data_field}',
    f'mothership_lvo_mt_{data_field}']

titles = [f'Drip and ship nLVO {data_name}',
    f'Mothership nLVO {data_name}',
    f'Drip and ship LVO {data_name}',    
    f'Mothership LVO {data_name}']

# Indexes for subplots (to leave gaps on right for diff between Mothership and Drip and Ship)
indexes = [1, 2, 4, 5]

# Find maximum of data
vmin = np.min([lsoa_data_gdf[col].min() for col in cols])
vmax = np.max([lsoa_data_gdf[col].max() for col in cols])

for i in range(4):
    ax = fig.add_subplot(2, 3, indexes[i])

    # Plot data
    lsoa_data_gdf.plot(ax=ax, # Set which axes to use for plot (only one here)
                 column=cols[i], # Column to apply colour
                 antialiased=False, # Avoids artifact boundry lines
                 edgecolor='face', # Make LSOA boundry same colour as area
                 vmin=vmin, # Manual scale min (remove to make automatic)
                 vmax=vmax, # Manual scale max (remove to make automatic)
                 cmap='inferno_r', # Colour map to use
                 # Adjust size of colourmap key, and add label
                 legend_kwds={'shrink':0.5, 'label':f'{data_name.capitalize()}'},
                 # Set to display legend
                 legend=True)
    
    # Add country boarder
    outline.plot(ax=ax, edgecolor='k', facecolor='None', linewidth=1.0)
    
    # Add hospitals
    mask = hospitals_gdf['Use_IVT'] == '1'
    hospitals_gdf[mask].plot(ax=ax, edgecolor='k', facecolor='w', markersize=40, marker='o')
    mask = hospitals_gdf['Use_MT'] == '1'
    hospitals_gdf[mask].plot(ax=ax, edgecolor='k', facecolor='r', markersize=40, marker='o')

    # Tighten map around mainland England and Wales
    ax.set_xlim(120000)

    ax.set_title(titles[i])
    ax.set_axis_off() # Turn of axis line and numbers

################### Plot differences between mothership and drip and ship ###################

cols = [f'nlvo_ivt_{data_field}_mothership_minus_dripship',
    f'lvo_mt_{data_field}_mothership_minus_dripship']

# Find absolute maximum of data extent
vmax = np.max((np.abs([lsoa_data_gdf[col].min() for col in cols]), 
               np.abs([lsoa_data_gdf[col].max() for col in cols])))
vmin = -vmax

titles = [f'nLVO advantage of Mothership ({data_name})',
          f'LVO advantage of Mothership ({data_name})']

# Indexes for subplots
indexes = [3, 6]

for i in range(2):
    ax = fig.add_subplot(2, 3, indexes[i])

    lsoa_data_gdf.plot(ax=ax, # Set which axes to use for plot (only one here)
                 column=cols[i], # Column to apply colour
                 antialiased=False, # Avoids artifact boundry lines
                 edgecolor='face', # Make LSOA boundry same colour as area
                 vmin=vmin, # Manual scale min (remove to make automatic)
                 vmax=vmax, # Manual scale max (remove to make automatic)
                 cmap='bwr', # Colour map to use
                 # Adjust size of colourmap key, and add label
                 legend_kwds={'shrink':0.5, 'label':f'Advantage of Mothership ({data_name.capitalize()})'},
                 # Set to display legend
                 legend=True)

    # Add country boarder
    outline.plot(ax=ax, edgecolor='k', facecolor='None', linewidth=1.0)
    
    # Add hospitals
    mask = hospitals_gdf['Use_IVT'] == '1'
    hospitals_gdf[mask].plot(ax=ax, edgecolor='k', facecolor='w', markersize=40, marker='o')
    mask = hospitals_gdf['Use_MT'] == '1'
    hospitals_gdf[mask].plot(ax=ax, edgecolor='k', facecolor='r', markersize=40, marker='o')

    ax.set_title(titles[i])
    ax.set_axis_off() # Turn of axis line and numbers    

    # Tighten map around mainland England and Wales
    ax.set_xlim(120000)
    
plt.tight_layout(pad=1)
plt.savefig(f'./output/{data_field}_six_in_one.jpg', dpi=300, bbox_inches='tight')
plt.show()

Plot weighted benefit

# Check weighted data exists
col = f'weighted_drip_ship_{data_field}'

if col not in lsoa_data_gdf.columns:
    print ('Required data does not exist')
else:
    lsoa_data_gdf[f'weighted_mothership_advantage_{data_field}'] = (
        lsoa_data_gdf[f'weighted_mothership_{data_field}'] - 
        lsoa_data_gdf[f'weighted_drip_ship_{data_field}'])

    fig = plt.figure(figsize=(18,6))

    # Set up columns to map
    cols = [
        f'weighted_drip_ship_{data_field}', 
        f'weighted_mothership_{data_field}',
        f'weighted_mothership_advantage_{data_field}']

    titles = [
        f'Weighted drip and ship for {data_name}',
        f'Weighted mothership for {data_name}',
        f'Weighted advantage of mothership for {data_name}']

    cmaps = ['inferno_r', 'inferno_r', 'bwr']

    # Find absolute maximum of data extent
    vmax = np.max([lsoa_data_gdf[col].max() for col in cols[0:2]])
    vmin = np.min([lsoa_data_gdf[col].min() for col in cols[0:2]])

    for i, col in enumerate(cols):
        ax = fig.add_subplot(1, 3, i+1)

        # Radjust vmin and max for advantage plot
        if i == 2:
            vmax = lsoa_data_gdf[col].max()
            vmin = lsoa_data_gdf[col].min()
            vmax = max(abs(vmax), abs(vmin))
            vmin = -vmax

        # Plot data
        lsoa_data_gdf.plot(ax=ax, # Set which axes to use for plot (only one here)
                    column=col, # Column to apply colour
                    antialiased=False, # Avoids artifact boundry lines
                    edgecolor='face', # Make LSOA boundry same colour as area
                    vmin=vmin, # Manual scale min (remove to make automatic)
                    vmax=vmax, # Manual scale max (remove to make automatic)
                    cmap=cmaps[i], # Colour map to use
                    # Adjust size of colourmap key, and add label
                    legend_kwds={'shrink':0.5, 'label':f'{data_name.capitalize()}'},
                    # Set to display legend
                    legend=True)
        
        # Add country boarder
        outline.plot(ax=ax, edgecolor='k', facecolor='None', linewidth=1.0)
        
        # Add hospitals
        mask = hospitals_gdf['Use_IVT'] == '1'
        hospitals_gdf[mask].plot(ax=ax, edgecolor='k', facecolor='w', markersize=40, marker='o')
        mask = hospitals_gdf['Use_MT'] == '1'
        hospitals_gdf[mask].plot(ax=ax, edgecolor='k', facecolor='r', markersize=40, marker='o')

        # Tighten map around mainland England and Wales
        ax.set_xlim(120000)

        # Add title
        ax.set_title(titles[i])
        ax.set_axis_off() # Turn of axis line and numbers
    
    plt.tight_layout(pad=1)
    plt.savefig(f'./output/{data_field}_weighted_results.jpg', dpi=300, bbox_inches='tight')
    plt.show()