Deaths by Cause (OASIS Detailed Causes, Nested)

Limitations

Deaths by OASIS detailed cause (nested) and county of residence. Causes nest: each of the 15 groups contains the causes listed in it and all is every death, so never add causes together. The NCHS and Georgia rankable topics count the same deaths: never add across them. OASIS withholds rates on 1-4 deaths (NULL). race_unknown is our count, with no rate. Race groups are NULL in 52 county-years (2000-2006) with over 20% unknown race or ethnicity; small groups' series break before 2015, not trends. Age-adjusted rates are OASIS's: weighted averages of age-specific rates with the year 2000 U.S. standard million as weights, so places and groups with different age mixes compare fairly; they never add or average across rows. OASIS publishes every count, however small, and withholds rates based on 1-4 events (NULL here); the platform adds no small-cell suppression of its own and never recomputes a withheld rate. A group with no residents in a place (num_population 0) has no rate (NULL): OASIS shows 0 there, but a rate on no population is undefined and would rank as the lowest (our mask; the count stays 0 unless the race mask NULLs it). 60% of rows with at least one death have no age-adjusted rate, mostly 1-4 deaths in small counties, small groups and rare causes. Within a group the listed causes never overlap but need not add up to it: they do exactly only in endocrine_nutritional_and_metabolic_diseases, mental_and_behavioral_disorders, nervous_system_diseases, genitourinary_diseases; elsewhere the rest of the group is in no listed cause (heart failure in major_cardiovascular_diseases, undetermined-intent injuries and complications of medical care in external_causes). The 15 groups never overlap either, and some deaths are in none of them (ill-defined and unknown causes, skin diseases, pregnancy, diseases of veins, in situ and benign neoplasms): 1,706 to 5,034 of Georgia's deaths a year, 2000-2025. That number jumps for one year in 2008 (5,034), 2012 (4,646), 2015 (4,551), against about 2,500 to 3,300 in the years around them, probably deaths whose cause was still pending or unknown (our reading); the NCHS and Georgia rankable topics' all_other_causes jump with it. OASIS's labels are kept in the cause description, but some codes say what the label hides: meningococcal_infection is OASIS's Meningitis (A39 only); nervous_system_malformations is OASIS's Neural Tube Defects (Q00-Q07, every malformation of the nervous system, not only neural tube defects); substance_use_disorders (OASIS: Disorders Related to Drug Use) counts deaths from alcohol and drug use disorders, not overdoses, which are in unintentional_poisoning, suicide and homicide or, when the intent is undetermined, only in external_causes; covid_19 is 0 before 2020 and counted inside infectious_and_parasitic_diseases. Dementia deaths moved between causes in 2014-2016 (our reading: a change in how they were coded, not in how many people died): Georgia's alzheimers_disease nearly doubled, from 2,042 in 2013 to 4,017 in 2016, while other_mental_and_behavioral_disorders, mostly unspecified dementia, fell from 4,626 to 2,716, so their sum barely changed (6,668 and 6,733); their groups shift with them (mental_and_behavioral_disorders 4,925 to 3,062, nervous_system_diseases 3,888 to 6,694), so don't read these series across 2014-2016 as trends. From 2014 on, dementia deaths also moved into senile degeneration of the brain (G31.1, in other_nervous_system_diseases here, whose deaths rose from 1,215 in 2013 to 5,849 in 2025): 5 Georgia deaths in 2013, 3,947 in 2025 (dementia_deaths), while alzheimers_disease's age-adjusted death rate fell from 47.8 (2018) to 30.9 (2025) and unspecified dementia fell too; the three together held steady, so this is probably coding as well (our reading): don't read alzheimers_disease or other_nervous_system_diseases as a trend after 2013. Race groups follow the platform convention from OASIS's race x ethnicity cross: hispanic is Hispanic of any race; white, black, asian, native_american (American Indian or Alaska Native), pacific_islander (Native Hawaiian or Other Pacific Islander) and multiracial are non-Hispanic. race_unknown is our calculation, the total minus those seven (unknown ethnicity, or non-Hispanic of unknown race), so the race groups add up to all wherever the race mask (below) leaves them; it has no population and so no rate. Hispanic ethnicity is under-reported on death certificates, so hispanic counts and rates may be low. The smaller race groups' series break in ways that look like changes in how race was recorded, not in mortality (our reading): Georgia's multiracial deaths were 0-16 a year in 2000-2007, then 253 in 2008, 8 in 2009 and 325 in 2012; native_american's age-adjusted rate fell from 1,227.8 (2005) to 89.3 (2008); asian deaths went 404, 191, 417 in 2005-2007; and 1,027-1,960 deaths a year in 2000-2006 are race_unknown (mostly unknown ethnicity), against at most 371 since, so the non-Hispanic groups are undercounted then. Multiracial rates stay far below every other group's (Georgia 2024, all causes: 102.2 against 762.2 for all races), probably because death certificates record more than one race far less often than population estimates do (our reading): don't compare them with other groups, and don't read the small groups' changes before 2015 as trends. Where race or ethnicity is unknown for more than 20% of a place-year's deaths (race_unknown / all), the race groups' counts and rates for that place and year are NULL (our mask, OASIS's own 20% threshold for missing items): each group's count would miss its share of the unknowns, so its rate runs low and its trend follows the recording, not the group. The all, sex and race_unknown rows are kept, so the unknown share stays visible. It applies to 52 county-years (10 counties, 2000-2006, never Georgia), on every cause: in those years a race group's county deaths can add up to less than Georgia's race row, which counts every county. num_population is OASIS's own population (its population module) for the place, year and group, the denominator of every crude rate (num_deaths / num_population x 100,000 reproduces each published crude rate); the race groups' populations add to all, as do male and female. Deaths are placed by county of residence as OASIS geocodes it, so about 4% land in a different county than in the federal NCHS files. 2025, the newest year, was added to OASIS on 2026-07-29; OASIS marks no year provisional. Counties add up to Georgia, except a race group's deaths in a year the race mask NULLs some of its county rows. Not served: OASIS's SES-vulnerability rows (its SES 'rates' divide by the whole place's population), age and education breakdowns, and the significance-flag and percent measures. tract_geoid is always NULL.