Mortality data on deaths due to pneumonia & influenza through the week ending 2020-04-18

Dan Weinberger,1 Jenny Chen,7 Ted Cohen,1 Forrest W. Crawford,2 Farzad Mostashari,3 Don Olson,4 Virginia E Pitzer,1 Nicholas G Reich,5 Marcus Russi,1 Lone Simonsen,6 Annie Watkins,1 Cecile Viboud7

1Department of Epidemiology of Microbial Diseases and the Public Health Modeling Unit, Yale School of Public Health, New Haven, CT 2Department of Biostatistics and the Public Health Modeling Unit, Yale School of Public Health, New Haven, CT; Yale Departments of Ecology and Evolutionary Biology, Statistics & Data Science, Yale School of Management 3Aledade, Inc 4Department of Health and Mental Hygiene, New York City, NY 5Department of Biostatistics and Epidemiology, School of Public Health and Health Sciences, University of Massachusetts, Amherst, MA 6Department of Science and Environment, Roskilde University, Denmark 7Division of International Epidemiology and Population Studies, Fogarty International Center, National Institutes of Health, Bethesda, MD

Abstract

Background Efforts to track the severity and public health impact of the novel coronavirus, COVID-19, in the US have been hampered by testing issues, reporting lags, and inconsistency between states. Evaluating unexplained increases in deaths attributed to broad outcomes, such as pneumonia and influenza or due to all causes, can provide a more complete and consistent picture of the burden caused by COVID-19.

Methods We evaluated increases in the occurrence of deaths due to any cause above a seasonal baseline (adjusted for influenza activity) or deaths due to coronaivirus/P&I (P&I&C) for March-early-April across the United States. These estimates are compared with reported deaths due to COVID-19 and with testing data.

Results There were notable increases in the rate of death due to all-causes and due to P&I&C in March and early-April 2020. In a number of states, these deaths pre-dated increases in COVID-19 testing rates and were not counted in official records at the time as related to COVID-19. There was substantial variability between states in the discrepancy between reported rates of death due to COVID-19 and the estimated burden of excess deaths due to P&I. The increase in all-cause deaths is 1.5 times higher than the official tally of COVID-19 reported deaths during the study period.

Conclusions Excess deaths provide an estimate of full COVID-19 burden and indicate that official tallies likely undercount deaths due to the virus.

National analysis of P&I

Excess deaths due to pneumonia/influenza/covid nationally in the specified time period

#>    state      2.5%   50%     97.5%
#> 43    US 42883.975 44220 45625.000
#> 33    NY 15602.000 15851 16111.000
#> 30    NJ  5412.000  5552  5661.000
#> 22    MI  2335.000  2444  2545.000
#> 19    MA  1909.000  1997  2078.000
#> 37    PA  1566.000  1673  1769.025
#> 5     CA  1398.000  1654  1886.000
#> 14    IL  1355.000  1458  1552.000
#> 9     FL  1285.000  1416  1542.000
#> 18    LA  1047.000  1096  1142.000
#> 42    TX   814.000   943  1070.000
#> 10    GA   648.000   729   797.000
#> 34    OH   622.000   713   807.000
#> 20    MD   642.000   705   759.000
#> 15    IN   572.000   653   729.000
#> 47    WA   563.000   647   723.000
#> 6     CO   553.000   615   671.000
#> 45    VA   303.000   375   440.000
#> 24    MO   250.975   329   395.000
#> 48    WI   258.000   327   388.000
#> 4     AZ   192.000   275   357.000
#> 2     AL   182.000   244   301.000
#> 23    MN   169.000   228   282.000
#> 25    MS   183.000   228   272.000
#> 32    NV   116.000   161   202.000
#> 41    TN    75.000   156   234.000
#> 35    OK    95.000   154   215.000
#> 17    KY    45.975   120   191.000
#> 12    IA    51.000   113   159.000
#> 39    SC    47.975   113   175.000
#> 36    OR    58.000   109   158.000
#> 16    KS    49.000   105   152.000
#> 38    RI    75.000   104   127.000
#> 7     DC    62.000    91   114.000
#> 3     AR    40.000    86   131.000
#> 8     DE    36.000    62    83.000
#> 31    NM    28.000    61    92.000
#> 29    NH    25.000    57    84.000
#> 44    UT    24.000    57    89.000
#> 13    ID    18.000    48    74.000
#> 27    ND    20.000    42    61.000
#> 49    WV   -26.000    35    79.000
#> 46    VT     8.000    33    53.000
#> 21    ME   -23.025    22    58.000
#> 26    MT    -5.000    22    46.000
#> 50    WY     1.000    21    36.000
#> 28    NE   -29.000    12    44.000
#> 11    HI   -18.000     9    31.000
#> 40    SD   -30.000    -1    25.000
#> 1     AK   -28.000    -2    16.000

National analysis of all-cause

Total excess all cause deaths

#> [1] "67500(60300,74500)"

Figure 1: Observed weekly death rate due to any cause vs seasonal baseline (+/-95% Prediction Interval)

The black line shows the observed proportion of deaths that were due to any cause per week. The red line and shaded area represent the 95% Prediction Interval. The latest P&I data is for the week ending 2020-04-18.

Figure 2: Observed weekly death rate due to pneumonia/influenza/covid-19 vs seasonal baseline for pneumonia&influenza (+/-95% Prediction Interval)

The black line shows the observed proportion of deaths that were due to Pneumonia & Influenza & Covid-19 (P&I&C) per week. The red line and shaded area represent the 95% Prediction Interval. The latest P&I&C data is for the week ending 2020-04-18.

Figure 3: Reported number of COVID-19 deaths,compared with the excess deaths due to pneumonia and influenza in each week, by state.

The red line shows the number of excess P&I cases +/-95% prediction intervals. The blue solid line shows the reported number of COVID-19 deaths for the same week (as compiled by covidtracking.com), and the dotted blue line shows the reported COVID-19 deaths for weeks in which the CDC data were not yet reliable. The grey dashed line represents number of tests performed per-capita in that week.

Table 1: Excess P&I deaths and reported deaths due to COVID-19

Observed and Excess deaths due to COVID-19, pneumonia/influenza/Covid-19, and all-causes COVID-19, from March 1, 2020 through Apr 18, 2020
State COVID-19 deaths (U07.1) Excess Pneumonia/Influenza/Covid-19 deaths Excess all-cause deaths Excess all-cause deaths/100,000
US 38490 44220 (42880,45620) 67500 (60300,74500) 20.6 (18.4,22.7)
NY 16042 15850 (15600,16110) 24200 (23500,24800) 124.3 (120.9,127.4)
NJ 5052 5550 (5410,5660) 8200 (7900,8600) 92.8 (88.8,96.9)
MA 1866 2000 (1910,2080) 2900 (2500,3200) 41.7 (36.9,46.5)
DC 83 90 (60,110) 200 (200,300) 34 (23.2,43.8)
MI 2217 2440 (2340,2540) 3100 (2700,3600) 31.5 (27.4,35.5)
LA 1057 1100 (1050,1140) 1000 (700,1200) 20.5 (14.9,25.8)
IL 1210 1460 (1360,1550) 2100 (1600,2500) 16.3 (12.8,19.5)
CO 553 620 (550,670) 900 (700,1100) 15.4 (11.4,18.9)
VT 0 30 (10,50) 100 (0,200) 14.9 (3,25.8)
MD 603 700 (640,760) 900 (600,1100) 14.4 (10.2,18.8)
FL 899 1420 (1280,1540) 2900 (2100,3600) 13.4 (9.6,16.9)
WI 233 330 (260,390) 600 (300,900) 9.6 (4.4,14.9)
TN 132 160 (80,230) 500 (200,800) 7.5 (2.5,12.2)
AZ 238 280 (190,360) 500 (200,900) 7.1 (2.4,11.8)
VA 317 380 (300,440) 600 (300,900) 6.9 (3,10.4)
CA 1256 1650 (1400,1890) 2700 (1500,3900) 6.8 (3.7,9.9)
SC 147 110 (50,180) 300 (100,600) 6.8 (2.1,11.7)
IN 580 650 (570,730) 400 (100,700) 6.5 (2.2,11)
WA 595 650 (560,720) 400 (100,800) 5.7 (1.3,10.1)
PA 1512 1670 (1570,1770) 600 (100,1200) 4.9 (0.4,9)
TX 513 940 (810,1070) 1200 (500,2000) 4.3 (1.7,6.8)
UT 11 60 (20,90) 100 (0,300) 4.1 (-0.2,8.1)
MS 180 230 (180,270) 100 (-100,300) 3.5 (-2.9,10)
NV 144 160 (120,200) 100 (-100,300) 3.3 (-2.4,8.5)
MO 197 330 (250,400) 200 (-100,500) 3.2 (-2.3,8.4)
RI 95 100 (80,130) 0 (-100,100) 1 (-9.3,10.8)
ID 31 50 (20,70) 0 (-100,100) -0.4 (-6.5,5.8)
GA 593 730 (650,800) -200 (-600,200) -1.6 (-5.2,2)
MN 143 230 (170,280) -100 (-400,200) -2 (-6.5,2.7)
HI 0 10 (-20,30) 0 (-100,100) -2.6 (-9.8,4.2)
DE 48 60 (40,80) 0 (-100,100) -3.1 (-13.2,6.4)
MT 0 20 (0,50) 0 (-100,100) -3.1 (-12.4,5.8)
NH 43 60 (20,80) 0 (-200,100) -3.7 (-11.6,4.2)
NE 10 10 (-30,40) -100 (-200,0) -4.2 (-11.7,2.4)
ME 0 20 (-20,60) -100 (-200,100) -4.4 (-13.2,4.2)
OR 77 110 (60,160) -200 (-400,0) -4.7 (-10.2,0.6)
OH 460 710 (620,810) -600 (-1200,-100) -5.4 (-9.9,-1)
AR 22 90 (40,130) -200 (-400,0) -5.4 (-12.2,1.2)
IA 72 110 (50,160) -200 (-400,0) -6.6 (-13.5,-0.6)
AL 188 240 (180,300) -300 (-700,0) -7 (-13.3,-0.3)
KS 89 100 (50,150) -200 (-400,-100) -7.6 (-14.2,-2)
SD 0 0 (-30,20) -100 (-200,0) -7.7 (-19,3.7)
OK 139 150 (100,220) -400 (-600,-100) -9 (-14.8,-3.5)
WV 0 40 (-30,80) -200 (-400,0) -9.5 (-20.4,0)
KY 130 120 (50,190) -500 (-800,-300) -11.8 (-17.4,-6)
WY 0 20 (0,40) -100 (-200,0) -12.8 (-26.1,-0.7)
NM 47 60 (30,90) -400 (-600,-300) -19.4 (-27,-12.1)
AK 0 0 (-30,20) -200 (-200,-100) -21.1 (-32.8,-10.4)
ND 0 40 (20,60) -200 (-300,-100) -23.9 (-34.6,-13.8)

Figure 4: Compare Excess P&I mortality vs Excess ILI

Here we compare the observed vs expected number of deaths due to pneumonia and influenza in each week compare to the observed vs expected number of outpatient visits for influenza-like illness (ILI) in each week. we would expect ILI (blue line) to increase earlier than deaths (red line)

Figure S1: Evaluation of reporting Delays

## Figure S2: Excess all-cause deaths for NY state (incl NYC) vs Excess deaths due to pneumonia and influenza

#>       42 
#> 1.307107
#>       42 
#> 1.763441
#>       42 
#> 1.415217

#>       42 
#> 1.270297

Figure S3: NYC only data P&I&C vs all-cause excess deaths

Ratio of all-cause vs P&I deaths in NYC. In NYC, only 25% of excess deaths were coded as P&I

#> [1] 2.566667 1.828571 1.711870 1.564897 1.397906

Table S1: Do not adjust for flu: P&I

Comparison of excess deaths due to pneumonia/influenza/Covid-19, as estimated from a regression that adjusts for influenza with a regression that does not
State Excess P&I&C, adjusted for influenza Excess P&I&C, unadjusted for influenza
1 AK 0(-30,20) 0(-10,20)
2 AL 240(180,300) 240(170,290)
3 AR 90(40,130) 90(40,130)
4 AZ 280(190,360) 280(190,360)
5 CA 1650(1400,1890) 1600(1370,1820)
6 CO 620(550,670) 580(520,640)
7 DC 90(60,110) 100(80,120)
8 DE 60(40,80) 70(50,90)
9 FL 1420(1280,1540) 1250(1110,1380)
10 GA 730(650,800) 700(620,780)
11 HI 10(-20,30) 10(-20,30)
12 IA 110(50,160) 90(40,130)
13 ID 50(20,70) 40(10,70)
14 IL 1460(1360,1550) 1410(1290,1510)
15 IN 650(570,730) 640(550,720)
16 KS 100(50,150) 110(50,150)
17 KY 120(50,190) 140(70,200)
18 LA 1100(1050,1140) 1080(1020,1130)
19 MA 2000(1910,2080) 1920(1820,2010)
20 MD 700(640,760) 700(640,750)
21 ME 20(-20,60) 10(-30,40)
22 MI 2440(2340,2540) 2370(2260,2470)
23 MN 230(170,280) 180(120,240)
24 MO 330(250,400) 300(220,370)
25 MS 230(180,270) 230(180,270)
26 MT 20(0,50) 20(0,50)
27 ND 40(20,60) 40(20,60)
28 NE 10(-30,40) 10(-20,40)
29 NH 60(20,80) 50(20,80)
30 NJ 5550(5410,5660) 5420(5310,5530)
31 NM 60(30,90) 50(20,80)
32 NV 160(120,200) 170(120,210)
33 NY 15850(15600,16110) 15580(15330,15810)
34 OH 710(620,810) 630(530,720)
35 OK 150(100,220) 160(110,220)
36 OR 110(60,160) 90(40,140)
37 PA 1670(1570,1770) 1590(1480,1690)
38 RI 100(80,130) 90(60,110)
39 SC 110(50,180) 120(60,180)
40 SD 0(-30,20) 10(-20,30)
41 TN 160(80,230) 160(80,240)
42 TX 940(810,1070) 760(600,920)
43 US 44220(42880,45620) 40990(39490,42440)
44 UT 60(20,90) 50(20,80)
45 VA 380(300,440) 360(290,420)
46 VT 30(10,50) 30(10,50)
47 WA 650(560,720) 600(520,680)
48 WI 330(260,390) 280(210,340)
49 WV 40(-30,80) 30(-20,70)
50 WY 20(0,40) 20(0,40)

Table S2: Do not adjust for flu: all-cause

#>  [1] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
#> [15] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
#> [29] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
#> [43] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
Comparison of excess deaths due to any cause, as estimated from a regression that adjusts for influenza with a regression that does not, with or without adjustments for reporting delays
State Excess deaths, adjusted for influenza Excess deaths, unadjusted for influenza Excess deaths, adjusted for reporting delay, not influenza
1 AK -150 (-240,-80) -110 (-170,-50) -40 (-110,20)
2 AL -340 (-650,-10) -320 (-620,-20) 70 (-230,370)
3 AR -160 (-370,40) -160 (-360,30) -60 (-260,130)
4 AZ 520 (180,860) 520 (160,870) 680 (320,1030)
5 CA 2700 (1480,3900) 2120 (950,3260) 2740 (1550,3870)
6 CO 880 (660,1090) 780 (590,960) 860 (670,1040)
7 DC 240 (160,310) 220 (150,280) 290 (220,360)
8 DE -30 (-130,60) -30 (-120,60) 440 (260,690)
9 FL 2870 (2070,3640) 1760 (1030,2480) 2030 (1290,2740)
10 GA -170 (-550,210) -140 (-520,220) 780 (390,1150)
11 HI -40 (-140,60) -40 (-140,50) -30 (-120,70)
12 IA -210 (-430,-20) -240 (-420,-70) -110 (-290,60)
13 ID -10 (-120,100) 20 (-90,130) 40 (-70,140)
14 IL 2070 (1620,2460) 1750 (1300,2190) 1950 (1500,2400)
15 IN 440 (140,740) 410 (130,700) 850 (560,1140)
16 KS -220 (-410,-60) -250 (-420,-90) -180 (-350,-20)
17 KY -530 (-780,-270) -580 (-820,-360) -10 (-250,230)
18 LA 950 (690,1200) 880 (610,1140) 1720 (1440,1990)
19 MA 2880 (2540,3210) 2520 (2170,2850) 2700 (2350,3030)
20 MD 870 (610,1130) 860 (600,1100) 960 (700,1210)
21 ME -60 (-180,60) -90 (-210,20) -80 (-200,30)
22 MI 3150 (2740,3550) 2600 (2180,2990) 3450 (2940,4070)
23 MN -110 (-370,150) -100 (-330,120) 40 (-200,260)
24 MO 200 (-140,510) -90 (-420,250) 540 (200,870)
25 MS 100 (-90,300) 120 (-80,300) 240 (40,430)
26 MT -30 (-130,60) -30 (-120,60) 0 (-100,90)
27 ND -180 (-260,-100) -180 (-260,-100) -160 (-240,-90)
28 NE -80 (-230,50) -110 (-240,20) -70 (-200,50)
29 NH -50 (-160,60) -50 (-160,40) -40 (-150,50)
30 NJ 8240 (7890,8600) 7880 (7540,8220) 8320 (7960,8680)
31 NM -410 (-570,-250) -370 (-510,-230) -170 (-310,-30)
32 NV 100 (-70,260) 70 (-90,230) 170 (10,330)
33 NY 24190 (23510,24790) 23520 (22880,24140) 24020 (23370,24650)
34 OH -630 (-1160,-110) -1380 (-1900,-860) -630 (-1150,-100)
35 OK -360 (-580,-140) -370 (-590,-140) -50 (-280,180)
36 OR -200 (-430,30) -170 (-400,40) 20 (-200,240)
37 PA 620 (60,1150) -240 (-770,310) 230 (-330,810)
38 RI 10 (-100,110) -40 (-140,50) 50 (-50,140)
39 SC 350 (110,600) 380 (130,620) 630 (370,870)
40 SD -70 (-170,30) -70 (-170,20) -40 (-140,50)
41 TN 510 (170,830) 540 (210,860) 790 (450,1120)
42 TX 1240 (500,1970) 300 (-590,1110) 1080 (190,1890)
43 US 67490 (60330,74510) 48150 (40850,55050) 55750 (48310,62860)
44 UT 130 (-10,260) 140 (20,260) 170 (40,280)
45 VA 590 (260,890) 370 (60,670) 720 (410,1030)
46 VT 90 (20,160) 90 (20,160) 100 (30,170)
47 WA 440 (100,770) 370 (60,670) 550 (230,850)
48 WI 560 (250,860) 110 (-200,390) 220 (-90,500)
49 WV -170 (-370,0) -250 (-400,-100) 70 (-100,240)
50 WY -70 (-150,0) -70 (-140,0) -60 (-130,10)