Excess deaths from COVID-19 correlate with age and socio-economic status. A database study in the Stockholm region

Figures & data

Table 1. Mortality and age categories.

	March				April				May
	2016–2019 n, mean (95% CI)	2020	Excess mortality	Significance^a (chi-square)^b	2016–2019 n, mean (95% CI)	2020	Excess mortality	Significance^a (chi-square)^b	2016–2019 n, mean (95% CI)	2020	Excess mortality	Significance^a (chi-square) ^b
Monthly mortality	1474 (1326–1622)	1819	23%	*** (38.9)	1376 (1300–1452)	2934	113%	***(1020.5)	1287 (1244–1330)	1860	44%	*** (147.4)
Age categories
60–69 years	154 (134–174)	153	−1%		145 (126–165)	246	69%	**(44.8)	151 (125–177)	162	7%
70–79 years	347 (302–391)	378	9%		323 (261–384)	632	96%	***(156.7)	306 (279–333)	406	33%	*** (13.7)
≥80 years	840 (768–911)	1141	36%	*** (46.6)	779 (712–846)	1886	142%	***(821.1)	694 (645–743)	1156	67%	*** (168.1)

The proportions of deaths were significantly higher for march, april, and may (23%, 113%, and 44%, respectively). the chi-square values were well below the limit for p < 0.001 (chi-square = 10.83) in all monthly comparisons.

^a Levels of significance: *p < 0.05; **p < 0.01; ***p < 0.001.

^b In all chi-square comparisons: 1 df (degree of freedom), meaning that the chi-square value for p < 0.05 is 3.84; p < 0.01 is 6.64; and for p < 0.001 is 10.83.

Table 2. Proportion of all deaths (all causes), and excess deaths (all causes) in March–May 2020, in comparison with data for respective month in 2016–2019, in nursing homes versus in all other places of death.

	Nursing homes			All others
Month	2016–2019 (95% CI)	2020	Excess mortality	2016–2019 (95% CI)	2020	Excess mortality
March	531 (482–580)	590	11%^a	944 (828–1059)	1229	30%^a
April	475 (437–511)	1269	167%^a	898 (852–951)	1665	85%^a
May	428 (368–488)	625	46%^a	855 (795–924)	1235	44%^a

^a p < 0.05 in all comparisons between 2016–2019 and 2020.

Table 3. Age at death.

	Cause of death		p Value
	COVID-19, Mean (median)	Other causes, Mean (median)
All patients (years)	81.0 (83)	79.2 (82)	<0.0001
Female patients (years)	84.2 (86)	82.0 (86)	<0.0001
Male patients (years)	78.3 (80)	76.3 (79)	<0.001

Patients dying from COVID-19 were significantly older than patients dying from other causes.

Table 4. COVID-19-related deaths per month during march–may 2020, in relation to socio-economic mosaic groups 1 and 3.

Month	Mosaic group 1	Mosaic group 3	Chi-square (1 df)	p Value
March	36/703478	91/715494	22.9	<0.0001
April	230/703576	451/715666	68.0	<0.0001
May	163/703508	226/715875	9.2	0.003

There were more COVID-19-related deaths in mosaic group 3.

Table 5. COVID-19-related deaths/1000 inhabitants in april 2020.

Age group (years)	Mosaic group	Deaths n	Deaths/1000 inhabitants	Chi-square (Mosaic 1 versus 3)	p Value (Mosaic 1 versus 3)
40–59	1	11	0.05
	2	14	0.06
	3	26	0.15	8.6	0.003
60–69	1	17	0.25
	2	28	0.31
	3	54	0.86	22.4	<0.00001
70–79	1	58	1
	2	95	1.19
	3	110	2.12	22.5	<0.00001
≥80	1	143	5
	2	274	6.98
	3	265	7.88	19.8	<0.00001

For each age group, the proportion of deaths was significantly higher in mosaic group 3, compared to mosaic group 1.

Table 6. Proportion of nursing home residents with at least one change of place of care during their last 2 weeks of life.

2016–2019, March–May, % (95% CI) with changes in place of care	2020, March–May, % with changes in place of care	Chi-square (1 df)	p Value
28.3% (26.7%–30.0%)	15.2%	162.0	<0.0001

March–May 2016–2019 is compared with March–May 2020.

Table 7. Proportion of nursing home residents with at least one change of place of care during their last 2 weeks of life, March–May 2020.

Month	COVID-19 deaths, n (%) with changes in place of care	Other causes, n (%) with changes in place of care	Chi-square (1 df)	p Value
March	22/39 (56%)	111/551 (20%)	27.9	<0.0001
April	99/449 (22%)	43/820 (5%)	80.8	<0.0001
May	48/213 (22%)	55/412 (13%)	8.6	0.003
March–May	169/701 (24%)	209/1783 (12%)	59.8	<0.0001

Residents dying from COVID-19 are compared with residents dying from other causes.

Table 8. Proportion of nursing home residents who eventually died in acute hospitals or geriatric wards, March–May 2020.

Months	COVID-19 deaths, n (%) hospital deaths	Other causes, n (%) hospital deaths	Chi-square (1 df)	p Value
March–May	132/701 (19%)	96/1687 (5%)	109.1	<0.0001

Residents dying from COVID-19 are compared with residents dying from other causes.