Cause of Death

Rank the causes not by raw count but by age-adjusted rate - deaths per 100,000, standardized to a fixed age structure so the roll is not just a map of who is old. Heart disease and cancer still head the ledger, as they have for a century. But 3 of these entries - overdose, chronic liver disease, suicide - belong to the despair thread this page follows, and they are the ones that have moved. The next panel shows how far.

1. 01 Heart disease 167.2 702,880
2. 02 Cancer 142.3 608,371
3. 03 Unintentional injury 64.7 227,039
4. 04 COVID-19 44.5 186,552
5. 05 Stroke 39.5 165,393
6. 06 Chronic lower respiratory 34.3 147,382
7. 07 Alzheimer disease 28.9 120,122
8. 08 Diabetes 24.1 101,209
9. 09 Chronic liver disease despair 13.3 54,803
10. 10 Kidney disease 12.8 57,937
11. 11 Suicide despair 14.2 49,476
12. 12 Drug overdose despair 32.6 107,941

Note: drug overdose is shown as its own line for the theme, but on the death certificate most overdoses are coded inside Unintentional injury - so it is partly double-counted against that row, not additive to the national total.

The top of the bill is the oldest news in medicine: heart disease and cancer have led it for a century, and they lead it still. The story is underneath. Read each cause by its place on the roll at the start of the span and again at its close - and watch the despair thread walk up from the foot. Overdose alone climbs 4 places, from the bottom of the list to its middle, while the settled killers ease down as the country gets better at the diseases it already knew how to fight.

The reordering as a table

Cause	1999 rate	2023 rate	Rank move
Heart disease	266.5	167.2	0
Cancer	200.8	142.3	0
Stroke	61.6	39.5	-1
Unintentional injury	35.3	64.7	+2
Chronic lower respiratory	45.4	34.3	-1
Diabetes	25.0	24.1	-2
Alzheimer disease	16.5	28.9	0
Chronic liver diseasedespair	9.6	13.3	-1
Suicidedespair	10.5	14.2	-1
Drug overdosedespair	6.1	32.6	+4

Rank is within these ten causes only, ordered by age-adjusted rate at each moment; a positive move is a climb toward the head of the bill. Rates are illustrative stand-ins in the CDC WONDER shape - see Methodology.

Here is the reading the title promises. Each state is shaded by its combined drug-induced and alcohol-induced death rate - age-adjusted, per 100,000. It is a rate, not a count, so a small state carrying a concentrated toll is not hidden behind Texas or California. Nationally the combined rate runs about 45.7 per 100,000; Appalachia and the desert Southwest sit far above it, each for its own reason - opioids in the coal country, alcohol across the Mountain West.

Every state, in a table

State	Drug	Alcohol	Combined
West Virginia WV	80.9	17.0	97.9
New Mexico NM	43.0	34.0	77.0
District of Columbia DC	63.0	12.0	75.0
Louisiana LA	55.9	16.0	71.9
Tennessee TN	51.7	15.0	66.7
Delaware DE	54.4	12.0	66.4
Alaska AK	35.0	28.0	63.0
Nevada NV	41.0	22.0	63.0
Kentucky KY	44.7	14.0	58.7
Maine ME	43.0	15.0	58.0
Ohio OH	45.7	12.0	57.7
Oregon OR	35.0	22.0	57.0
Arizona AZ	35.0	21.0	56.0
South Carolina SC	40.0	13.0	53.0
Rhode Island RI	40.0	13.0	53.0
Pennsylvania PA	42.0	11.0	53.0
Montana MT	22.0	30.0	52.0
Maryland MD	42.0	10.0	52.0
Missouri MO	38.0	13.0	51.0
Colorado CO	30.0	20.0	50.0
Connecticut CT	38.0	12.0	50.0
Florida FL	37.0	13.0	50.0
North Carolina NC	38.0	12.0	50.0
Washington WA	33.0	17.0	50.0
Wyoming WY	22.0	28.0	50.0
Indiana IN	37.0	11.0	48.0
Vermont VT	32.0	16.0	48.0
Massachusetts MA	36.0	11.0	47.0
New Hampshire NH	33.0	14.0	47.0
Michigan MI	34.0	12.0	46.0
Oklahoma OK	30.0	16.0	46.0
Illinois IL	33.0	12.0	45.0
Wisconsin WI	30.0	15.0	45.0
Virginia VA	33.0	11.0	44.0
Alabama AL	32.0	12.0	44.0
New York NY	33.0	10.0	43.0
California CA	27.0	16.0	43.0
New Jersey NJ	33.0	9.0	42.0
North Dakota ND	20.0	22.0	42.0
Arkansas AR	27.0	13.0	40.0
South Dakota SD	13.0	27.0	40.0
Mississippi MS	27.0	12.0	39.0
Kansas KS	24.0	14.0	38.0
Minnesota MN	23.0	15.0	38.0
Idaho ID	20.0	18.0	38.0
Utah UT	22.0	15.0	37.0
Georgia GA	25.0	11.0	36.0
Texas TX	19.0	13.0	32.0
Hawaii HI	17.0	14.0	31.0
Iowa IA	15.0	15.0	30.0
Nebraska NE	11.0	14.0	25.0
United States US	32.6	13.1	45.7

Rates are age-adjusted deaths per 100,000. Drug-induced and alcohol-induced causes are the CDC WONDER induced-cause groupings; combined is their sum. See Methodology for the exact ICD-10 sets and the swap-point to the live WONDER export.

"Deaths of despair" is one phrase for two epidemics that barely overlap on the map. Set each state by its drug-induced rate across the bottom and its alcohol-induced rate up the side. The diagonal is parity - equal loss to each. Below it sit the opioid states, West Virginia furthest out; above it, a cluster where drink is the heavier killer - the Mountain West and Northern Plains, Montana and the Dakotas. 5 states lose more to alcohol than to drugs, and they are not the states the overdose headlines name.

The corners, in a table

State	Drug	Alcohol	Which is worse
West Virginia WV	80.9	17.0	drugs
District of Columbia DC	63.0	12.0	drugs
Louisiana LA	55.9	16.0	drugs
New Mexico NM	43.0	34.0	drugs
Alaska AK	35.0	28.0	drugs
Nebraska NE	11.0	14.0	alcohol
Montana MT	22.0	30.0	alcohol
South Dakota SD	13.0	27.0	alcohol

A representative set of the corner states; the full fifty are in the map table above. Rates are age-adjusted deaths per 100,000, illustrative stand-ins in the CDC WONDER shape.

Three ways to die of despair, tracked across a quarter century of death certificates. Suicide drifts up and holds. Alcohol climbs, then jumps in the pandemic. But the drug-overdose line does something the others never do: after 2019 it turns almost vertical, more than 5.1x its 1999 rate, as illicit fentanyl replaces the pill and the needle. This is the shape the phrase "deaths of despair" was coined to describe.

The curve as a table (selected years)

Year	Drug	Alcohol	Suicide
1999	6.1	6.8	10.5
2001	6.8	7.0	10.7
2004	9.4	7.2	11.0
2007	11.9	7.5	11.5
2010	12.3	7.6	12.4
2013	13.8	8.0	13.0
2016	19.8	9.1	13.9
2019	21.6	10.4	14.5
2022	32.6	13.1	14.3
2023	31.3	12.6	14.2

The overdose line did not just rise - its cause changed underneath it. In 2010 the pile was prescription opioids and a little heroin. By 2023 it is a wall of illicit fentanyl, now joined by a second front of methamphetamine and cocaine. Read the oxblood block grow; that is the third opioid wave, the one no pill count predicts.

Drug-class rates as a table

Year	Illicit fentanyl	Methamphetamine	Cocaine	Rx opioids	Heroin
2010	1.0	0.6	1.5	4.7	1.0
2013	1.0	0.8	1.6	5.1	2.7
2016	6.2	2.1	3.2	5.2	4.9
2019	11.4	5.0	4.5	4.2	4.4
2022	22.7	9.0	8.3	3.5	1.9
2023	22.0	9.5	8.6	3.2	1.3

Age-adjusted deaths per 100,000 in which each drug was involved (multiple-cause coding). Because most overdoses now involve more than one drug, a year's classes sum to MORE than its all-drug total - the stack shows the growing pile of involvements, not a partition of deaths. See Methodology.

Most causes of death climb steadily with age - the old die of the body wearing out. Overdose does the opposite. It crests in the 35-44 band and falls away after, hollowing out the working years and leaving a generation-shaped hole in the middle of the population. This inverted curve is why the epidemic costs so many years of life per death: it kills the not-yet-old, the parents and workers, and leaves the very young and the very old largely untouched.

The age curve as a table

Age band	Rate per 100k
15-24	12
25-34	44
35-44	62
45-54	52
55-64	41
65-74	16
75+	5

A death at eighty-five and a death at thirty-five are one line each on the bill, but they are not the same loss. Count instead the years of life each death cuts short, and the ledger reweights toward the young. Heart disease is a smaller share of the lost years than of the deaths; overdose and suicide are far larger. Overdose takes 9% of these deaths but 15% of the lost years - the clearest measure of why the despair epidemic reads as a national wound and not a rounding error.

Deaths and lost years as a table

Cause	Deaths %	Years lost %	Gap
Cancer	25	24	-1
Heart disease	29	18	-11
Unintentional injury	11	16	+5
Drug overdose	9	15	+6
Suicide	4	8	+4
Chronic liver disease	4	6	+2
Stroke	7	5	-2
Alzheimer disease	6	4	-2
Diabetes	5	4	-1

Shares are of the totals across the causes shown, not of all deaths. Years of life lost is illustrative here - computed from a death's age against a fixed reference age - and stands in for a WONDER age-by-cause export. A positive gap means the cause takes a larger share of the lost years than of the deaths. See Methodology.