Claim at v1.1.0
split-temp-asthma-ems
8d6f6609df92db24fac43a44b1691342aa07da6e7babbacca38ee1d927fd6abf
Data
{
"has_outcome": true,
"kind": "confirmation",
"lineage": {
"audit": "docs/lineage_audit.md; data/derived/lineage_audit.json",
"class": "a",
"class_meaning": "exposure or treatment only, not the outcome",
"classes": [
"a"
],
"early_sources": [
{
"class": "a",
"dataset": "GHCN-Daily v3 (NCEI), doi:10.7289/V5D21VHZ",
"evidence": "Daily temperature (GHCN-Daily) is the exposure, not the outcome. It is weather, fetched for earlier heat claims and reused.",
"fetched": "2026-11-12T20:23:55-05:00"
}
]
},
"outcome": "split-temp-asthma-ems",
"prereg_branch": {
"name": "prereg/split-temp-asthma-ems",
"sha": "15e54e1e7aab3891c95332349640c46aa5d01b95"
},
"prior": 0.45,
"registered_at": "2026-11-12T22:49:54-0500",
"registrant": "Claude (split-literature study, graph discovery)",
"registration": {
"alternatives": [
"Respiratory infections: cold spells coincide with influenza and RSV weeks; month fixed effects remove only part of this. Named, not separated.",
"Call-taking behaviour: on very cold or very hot days people call EMS for different reasons or call types are assigned differently; the difficulty-breathing codes (DIFFBR, RESPIR) could absorb asthma calls.",
"Air pollution: ozone on hot days and particulates on stagnant cold days; not adjusted, so the estimate is the total effect of weather.",
"Pollen: warm spring days release tree pollen (the Bronx study in the graph, PMID 30207818), which would make April-May slopes positive for a reason other than temperature."
],
"graph_evidence": [
{
"concepts": [
"temperature",
"asthma"
],
"direction": "split",
"n_findings": 30,
"n_studies": 17,
"pubmed_ids": [
24932702,
30641275,
35817164,
36240934,
39251927,
20708875,
30207818,
31651344,
33230642,
34599899,
36435499,
23035136,
21432566,
29529393,
7889426,
14982513,
32213400
],
"relationship_id": "4FaXcuNJRt|EAshH_Ko3h",
"sign_split": {
"minus": 8,
"plus": 22,
"sig_minus": 4,
"sig_plus": 16
},
"summary": "The graph's 22 positive and 8 negative findings do not disagree about temperature level once the variable names and abstracts are read. I recoded each finding by what was measured. 17 findings are short-term effects of temperature level; 16 of them (7 studies: Taipei, Hong Kong twice, Hefei, Lanzhou, a US emergency department, the Bronx) say colder days or falling temperatures bring more asthma visits or admissions, and they are stored with a positive sign because the variable is named 'low temperature' or 'cold (5th percentile)'; one (Guangzhou-type subtropical city, respiratory morbidity) says heat. 9 findings are about temperature change or variability, not level, and disagree by season: in Shanghai a day-to-day rise raises childhood asthma visits in the warm season (RR 1.98) and a drop lowers them (RR 0.31), with no effect in the cool season; Hefei and a dry-climate city find variability effects mainly in the cool season; a US case-crossover study finds a negligible inverse effect. 3 are cross-sectional comparisons of climates (prevalence is higher in warmer places in China, Costa Rica and Italy), a different question. One is an interaction with air quality. Eight of the 30 rows are four results from PMID 35817164 stored twice."
}
],
"id": "split-temp-asthma-ems",
"kind": "confirmation",
"literature": [],
"plan": "Outcome: daily count of FDNY EMS Incident Dispatch Data (76xm-jjuj) rows with initial_call_type in ('ASTHMB','ASTHFC'), by date of incident_datetime, 2006-01-01 to 2026-08-31, one SoQL group-by query through concord/socrata.py. Exposure: Central Park (USW00094728) daily mean temperature (TMAX+TMIN)/2 from concord/noaa.py, averaged over the day and the three days before it, in units of 5 C. Model: Poisson regression of the daily count on the exposure, with fixed effects for each year-month (so that season and trend are removed and only day-to-day variation within a month is used), day of week and US federal holidays; standard errors Newey-West with 7 lags. The model is fitted separately to October-March and April-September, and once to all days with an exposure-by-half-year interaction for the difference. Sensitivity, reported but not deciding: final_call_type instead of initial; lag 0 only; lags 0 to 7; excluding March 2020 to June 2021; daily minimum temperature. Exploratory: slope by calendar month. Disclosure of what the project had seen before this registration. Graph side: all findings of both relationships and their PubMed abstracts were read (data/derived/graph/splits/findings.json, abstracts.json). City side: Central Park daily temperature (NOAA GHCN-Daily) has been used with shootings, heat-illness ED visits and EMS dispatches of type SHOT, STAB, DRUG, DRUGFC, EDP, EDPC, EDPM. For EMS dispatches (76xm-jjuj) the project has seen the count of every initial_call_type for calendar 2023 (ASTHMB 7,130; ASTHFC 2,813) and the date range of the file (2005-01-01 to 2026-09-30). No daily series of asthma dispatches has been queried by any agent (data/cache/socrata holds four 76xm-jjuj queries, none with an asthma code). For the COVID-19 daily counts (rc75-m7u3) I queried the column list and the first and last date (2020-02-29, 2025-10-13; 2,054 rows) and nothing else; no other agent has queried it.",
"prediction": {
"direction": "-",
"magnitude": "Cool half-year: incidence rate ratio of asthma dispatches per 5 C higher mean temperature over lags 0 to 3 days between 0.95 and 0.99. Warm half-year: rate ratio between 1.00 and 1.05. Difference of log rate ratios (warm minus cool) positive.",
"moderator_prediction": "Secondary, scored separately and not deciding the verdict: the warm-minus-cool difference of log rate ratios is positive with a 95% interval excluding 0. Reasoning from the moderators: the cold effect in the literature is a cool-season, low-percentile effect, and the only seasonal analysis of rises (Shanghai) finds them harmful in the warm season only.",
"refute_if": "The cool half-year rate ratio per +5 C is above 1 and its 95% interval excludes 1.",
"support_if": "The cool half-year rate ratio per +5 C is below 1 and its 95% interval excludes 1."
},
"prior": 0.45,
"registered_at": "2026-11-12T22:49:54-0500",
"registrant": "Claude (split-literature study, graph discovery)",
"sha256": "75d1a5bf8b737e3e3f6821d4e7524a010ad94f6fe2d39dbedae37ce7d9182fe7",
"statement": "In New York City, within the same month and year, days in the cool half of the year (October to March) that are colder have more FDNY EMS dispatches for asthma, and the slope is more positive (or less negative) in the warm half of the year (April to September)."
},
"registration_commit": {
"commit": "a588c84415ad5612b91449e102c1369458917cf1",
"committed_at": "2026-11-12T22:49:54-05:00"
},
"registration_file_sha256": "860b0029fa6d7da542248e5584eec97440a476877d95346b586f8d5b49f67e3e",
"registration_path": "claims/registered/split-temp-asthma-ems.json",
"registration_sha256": "75d1a5bf8b737e3e3f6821d4e7524a010ad94f6fe2d39dbedae37ce7d9182fe7",
"registration_text": "{\n \"id\": \"split-temp-asthma-ems\",\n \"kind\": \"confirmation\",\n \"statement\": \"In New York City, within the same month and year, days in the cool half of the year (October to March) that are colder have more FDNY EMS dispatches for asthma, and the slope is more positive (or less negative) in the warm half of the year (April to September).\",\n \"graph_evidence\": [\n {\n \"relationship_id\": \"4FaXcuNJRt|EAshH_Ko3h\",\n \"concepts\": [\n \"temperature\",\n \"asthma\"\n ],\n \"n_findings\": 30,\n \"n_studies\": 17,\n \"sign_split\": {\n \"plus\": 22,\n \"minus\": 8,\n \"sig_plus\": 16,\n \"sig_minus\": 4\n },\n \"direction\": \"split\",\n \"summary\": \"The graph's 22 positive and 8 negative findings do not disagree about temperature level once the variable names and abstracts are read. I recoded each finding by what was measured. 17 findings are short-term effects of temperature level; 16 of them (7 studies: Taipei, Hong Kong twice, Hefei, Lanzhou, a US emergency department, the Bronx) say colder days or falling temperatures bring more asthma visits or admissions, and they are stored with a positive sign because the variable is named 'low temperature' or 'cold (5th percentile)'; one (Guangzhou-type subtropical city, respiratory morbidity) says heat. 9 findings are about temperature change or variability, not level, and disagree by season: in Shanghai a day-to-day rise raises childhood asthma visits in the warm season (RR 1.98) and a drop lowers them (RR 0.31), with no effect in the cool season; Hefei and a dry-climate city find variability effects mainly in the cool season; a US case-crossover study finds a negligible inverse effect. 3 are cross-sectional comparisons of climates (prevalence is higher in warmer places in China, Costa Rica and Italy), a different question. One is an interaction with air quality. Eight of the 30 rows are four results from PMID 35817164 stored twice.\",\n \"pubmed_ids\": [\n 24932702,\n 30641275,\n 35817164,\n 36240934,\n 39251927,\n 20708875,\n 30207818,\n 31651344,\n 33230642,\n 34599899,\n 36435499,\n 23035136,\n 21432566,\n 29529393,\n 7889426,\n 14982513,\n 32213400\n ]\n }\n ],\n \"literature\": [],\n \"prediction\": {\n \"direction\": \"-\",\n \"magnitude\": \"Cool half-year: incidence rate ratio of asthma dispatches per 5 C higher mean temperature over lags 0 to 3 days between 0.95 and 0.99. Warm half-year: rate ratio between 1.00 and 1.05. Difference of log rate ratios (warm minus cool) positive.\",\n \"support_if\": \"The cool half-year rate ratio per +5 C is below 1 and its 95% interval excludes 1.\",\n \"refute_if\": \"The cool half-year rate ratio per +5 C is above 1 and its 95% interval excludes 1.\",\n \"moderator_prediction\": \"Secondary, scored separately and not deciding the verdict: the warm-minus-cool difference of log rate ratios is positive with a 95% interval excluding 0. Reasoning from the moderators: the cold effect in the literature is a cool-season, low-percentile effect, and the only seasonal analysis of rises (Shanghai) finds them harmful in the warm season only.\"\n },\n \"plan\": \"Outcome: daily count of FDNY EMS Incident Dispatch Data (76xm-jjuj) rows with initial_call_type in ('ASTHMB','ASTHFC'), by date of incident_datetime, 2006-01-01 to 2026-08-31, one SoQL group-by query through concord/socrata.py. Exposure: Central Park (USW00094728) daily mean temperature (TMAX+TMIN)/2 from concord/noaa.py, averaged over the day and the three days before it, in units of 5 C. Model: Poisson regression of the daily count on the exposure, with fixed effects for each year-month (so that season and trend are removed and only day-to-day variation within a month is used), day of week and US federal holidays; standard errors Newey-West with 7 lags. The model is fitted separately to October-March and April-September, and once to all days with an exposure-by-half-year interaction for the difference. Sensitivity, reported but not deciding: final_call_type instead of initial; lag 0 only; lags 0 to 7; excluding March 2020 to June 2021; daily minimum temperature. Exploratory: slope by calendar month. Disclosure of what the project had seen before this registration. Graph side: all findings of both relationships and their PubMed abstracts were read (data/derived/graph/splits/findings.json, abstracts.json). City side: Central Park daily temperature (NOAA GHCN-Daily) has been used with shootings, heat-illness ED visits and EMS dispatches of type SHOT, STAB, DRUG, DRUGFC, EDP, EDPC, EDPM. For EMS dispatches (76xm-jjuj) the project has seen the count of every initial_call_type for calendar 2023 (ASTHMB 7,130; ASTHFC 2,813) and the date range of the file (2005-01-01 to 2026-09-30). No daily series of asthma dispatches has been queried by any agent (data/cache/socrata holds four 76xm-jjuj queries, none with an asthma code). For the COVID-19 daily counts (rc75-m7u3) I queried the column list and the first and last date (2020-02-29, 2025-10-13; 2,054 rows) and nothing else; no other agent has queried it.\",\n \"alternatives\": [\n \"Respiratory infections: cold spells coincide with influenza and RSV weeks; month fixed effects remove only part of this. Named, not separated.\",\n \"Call-taking behaviour: on very cold or very hot days people call EMS for different reasons or call types are assigned differently; the difficulty-breathing codes (DIFFBR, RESPIR) could absorb asthma calls.\",\n \"Air pollution: ozone on hot days and particulates on stagnant cold days; not adjusted, so the estimate is the total effect of weather.\",\n \"Pollen: warm spring days release tree pollen (the Bronx study in the graph, PMID 30207818), which would make April-May slopes positive for a reason other than temperature.\"\n ],\n \"prior\": 0.45,\n \"registrant\": \"Claude (split-literature study, graph discovery)\",\n \"registered_at\": \"2026-11-12T22:49:54-0500\",\n \"sha256\": \"75d1a5bf8b737e3e3f6821d4e7524a010ad94f6fe2d39dbedae37ce7d9182fe7\"\n}",
"statement": "In New York City, within the same month and year, days in the cool half of the year (October to March) that are colder have more FDNY EMS dispatches for asthma, and the slope is more positive (or less negative) in the warm half of the year (April to September)."
}History in this collection
- addedv1.0.0Nov 14, 2026this version