Research method
Logistic Regression
Logistic regression models the log-odds of a binary outcome — preterm versus term, high cholesterol versus not, preeclampsia versus not — as a function of predictors. Results are usually odds ratios per IQR or per standardised unit. Close cousins in this set include conditional logistic models in matched case-control designs, elastic-net screening followed by unpenalized logistic, and modified Poisson models that report risk ratios instead of odds. An OR is not automatically a risk ratio, especially when the outcome is common.
Environmental papers use logistic (and related binary) models when the scientific question is a yes/no event under a pollution or PFAS contrast. The method answers 'how do the odds of this outcome change with this assigned exposure after measured confounders?' Its main limitation is that a selected AOR cannot prove the chemical caused the event, correlated traffic or PFAS layers remain, and swapping OR for RR, or PAH stations for a city-wide surface, changes what the number means.
Evidence
What the evidence shows
Drawn from 8 studies in this library. Each finding starts with a plain-language takeaway, then the denser detail. Supports means evidence for a finding; Challenges means evidence against a stated position; Qualifies marks scope with a short note on each study’s contribution. Challenged positions are labeled — they are not findings.
A Los Angeles case-control of births near MATES monitors (preterm prevalence 9.2%) used stations, land-use regression and source-apportioned PM in conditional logistic models. Total PAHs and several PAH species showed ~30% odds increases (benzo(a)pyrene 13%), but PAH samples were small and limited to two stations. IQR increases in ammonium nitrate PM₂.₅ and 9–13% odds rises appeared for benzene, diesel/EC/OC and biomass-burning PM₂.₅.
PROTECT also studied preterm birth (~9.1% of 1,092 pregnancies) but with modified Poisson risk ratios, not logistic ORs. Mean pregnancy PM₂.₅ was 6.8 μg/m³; an IQR increase gave RR 1.014 (individual-adjusted) and 1.012 (also municipal-adjusted). Phthalate metabolites did not modify the association. Monitor-based municipality assignment likely biases toward the null.
In Ronneby versus Karlshamn, regression on ln-PFAS found cholesterol and LDL 7–9% higher in the exposed town and odds of high cholesterol about 1.3 per ln-PFAS unit. PFOS (median 157 ng/ml) and PFHxS (median 136 ng/ml) were correlated at r ≈ 0.9, so the OR is not compound-specific.
Shanghai deliveries quantified eight PFAS in cord blood. HDP, preeclampsia and gestational hypertension rates were 6.1%, 2.8% and 3.3%. PFOA had the highest median (6.98 ng/ml), but elastic-net selected PFBS (median 0.047 ng/ml) for HDP/preeclampsia. Unpenalized logistic AORs were 1.64 for HDP and 1.81 for preeclampsia per standardised PFBS unit. Cord blood is not a full pregnancy-long exposure history.
An administrative neurologic cohort reported 13,170 non-Alzheimer dementia, 4,201 Parkinson, 1,277 Alzheimer and 658 MS cases. Road proximity raised risks; greenness attenuated NAD/PD HRs by 0.3–6.2% and AD ORs by 11–28%. Air pollution associated with NAD and PD, not AD or MS; noise effects were generally null — diagnosis-specific binary/survival models, not one 'neurologic OR'.
NICHD Scandinavian SGA-study pairs found 55 of 412 children (12%) overweight or obese at age 5. Increasing maternal PFAS associated with higher BMI and triceps z-scores and with overweight, but associations differed by country and concentration; organochlorines were less consistent. Swedish NFI lichen occurrence, modeled 10 years apart, fell for three genera (Alectoria relative loss ~1.70%/year; more than half of Alectoria records lost in region 3) — presence/absence change, not human disease odds.
Open questions
Tensions and limits
Some items are genuine disagreements on the same question. Others mark different assays, populations, or outcomes — limits on how far one study travels — not a forced fight between papers.
Two preterm-birth analyses both sit near a 9% outcome and do not report the same contrast. Los Angeles finds ~30% PAH odds increases (benzo(a)pyrene 13%) plus 9–13% for several PM sources, with PAH data from only two stations. PROTECT reports RR 1.014 per IQR of pregnancy PM₂.₅ at a mean of 6.8 μg/m³. Odds ratios and risk ratios are not interchangeable, and the exposure products (PAH stations versus municipality monitors) are not the same assignment.
- LA air toxics, especially PAHs, track preterm birth
- Prenatal PM2.5 linked to preterm birth in Puerto Rico
Study Role Design N Population Outcome LA air toxics, especially PAHs, track preterm birth Supports Case-controlPopulation-based case-control of LA preterm birth near MATES monitors with traffic toxics N=241415 · 241,415 births after exclusions from 276,891 birth records; analyses often restricted nearer monitors Births in Los Angeles County near MATES air-toxics monitors Preterm birth odds related to PAHs, traffic markers, and source-apportioned PM2.5 Prenatal PM2.5 linked to preterm birth in Puerto Rico Supports CohortPROTECT pregnancy cohort with pregnancy-average PM2.5 and preterm-birth risk ratios N=1092 · 1,092 PROTECT participants with live singleton births in 21 municipalities Pregnant women in the PROTECT cohort in Puerto Rico Preterm birth associated with prenatal PM2.5 (with phthalate-interaction tests) PFAS logistic results do not agree on which compound or outcome carries the signal. Ronneby's high-cholesterol OR is about 1.3 per ln-PFAS in a PFOS/PFHxS-dominated mixture. Shanghai selected low-concentration PFBS (AOR 1.64/1.81) while the highest-median compound, PFOA at 6.98 ng/ml, was not the selected HDP predictor. Age-5 overweight associations differed by country. A single 'PFAS odds ratio' is not portable.
- PFAS in Ronneby water tracks higher cholesterol
- Cord-blood PFBS linked to preeclampsia in Shanghai
- Prenatal PFAS tracked 5-year child overweight risk
Study Role Design N Population Outcome PFAS in Ronneby water tracks higher cholesterol Supports Cross-sectionalOpen sampling of Ronneby/Karlshamn adults comparing serum PFAS and lipids by exposure group N=1945 · 1,945 adults aged 20–60 Adults from PFAS-contaminated Ronneby and control Karlshamn, Sweden Serum cholesterol/LDL associated with PFOS and PFHxS Cord-blood PFBS linked to preeclampsia in Shanghai Supports CohortShanghai delivery cohort with cord-blood PFAS and recorded hypertensive disorders of pregnancy N=687 · 687 women delivering singletons at two Shanghai hospitals (13 excluded for missing covariates in some models) Women delivering at two Shanghai hospitals, 2011–2012 Hypertensive disorders of pregnancy and preeclampsia related to cord PFAS (esp. PFBS) Prenatal PFAS tracked 5-year child overweight risk Supports CohortProspective Scandinavian SGA-study follow-up of maternal PFAS/POPs to age-5 adiposity N=412 · 412 mother–child pairs from Norway and Sweden Scandinavian mother–child pairs in the NICHD SGA study Age-5 BMI, skinfolds, and overweight/obesity related to prenatal PFAS/POPs
Common misconceptions
An odds ratio of 1.3 is essentially a 30% increase in risk.
OR approximates RR only when the outcome is rare. Preterm prevalence was 9.2% in Los Angeles and 9.1% in PROTECT; high cholesterol in Ronneby is not a rare event. PROTECT reports RRs near 1.01 per IQR — a different scale from a 30% PAH OR.
Elastic-net selected PFBS, so PFOA is unimportant in that Shanghai sample.
PFOA had the highest median (6.98 ng/ml); PFBS (0.047 ng/ml) was the selected HDP/preeclampsia predictor after penalisation. Selection is a prediction step among correlated PFAS, not a toxicological clearance of PFOA, and cord blood is not pregnancy-long exposure.
Logistic occurrence models of canopy lichens measure biomass loss in the upper canopy.
Swedish NFI presence on host trees does not measure upper-canopy biomass, and interacting drivers (logging, N deposition, climate) mean no single lever is shown to reverse the ~1.70%/year Alectoria loss.
Exam-style questions
Short-answer questions that ask you to explain or compare, not recall.
Los Angeles reports ~30% PAH odds increases at 9.2% preterm prevalence; PROTECT reports RR 1.014 per IQR. What must you state before comparing those two numbers?
OR versus RR, PAH species from two stations versus pregnancy-average PM₂.₅ from municipal monitors, and case-control conditional logistic versus modified Poisson. A 30% OR is not a 1.4% risk increase in the other study.
Why does an AOR of 1.81 per standardised PFBS unit not prove that PFBS caused preeclampsia?
The model is observational on cord-blood concentrations after elastic-net selection. Cord blood is not a full pregnancy exposure history; residual confounding remains; and PFOA, the highest-median PFAS, was not the selected predictor — so the AOR is not a compound-isolated causal effect.
Greenness attenuated AD ORs by 11–28% but NAD/PD HRs by only 0.3–6.2%, and air pollution associated with NAD and PD but not AD or MS. What does that tell you about a single logistic 'dementia' model?
Pooling diagnoses would hide outcome-specific associations and different attenuation. Record-ascertained subtypes can also be misclassified, and traffic co-exposures remain correlated.
Ronneby reports odds of high cholesterol about 1.3 per ln-PFAS with PFOS–PFHxS r ≈ 0.9. What happens if you interpret that as a PFOS-only effect?
You claim compound specificity the correlation does not support. The OR is a mixture contrast between exposure groups, and cross-sectional lipids still do not prove later cardiovascular disease.
The studies
8 studies in this library bear on Logistic Regression, ordered by citations.
- Roads and air pollution track neurologic disease
In Metro Vancouver, living near roads raised incidence of non-Alzheimer dementia, Parkinson’s, Alzheimer’s, and MS; air pollution linked to NAD and PD, noise was null.
- 2012 Colorado wildfire PM raised asthma visits
During a 32-day 2012 fire period, modeled PM2.5 associated with asthma/wheeze and COPD ED visits; cardiovascular associations were largely null.
- PFAS in Ronneby water tracks higher cholesterol
Swedish adults on PFAS-contaminated water had 7–9% higher total cholesterol and LDL, with high odds of cholesterol >6.2 mmol/L as PFOS/PFHxS rose.
- Cord-blood PFBS linked to preeclampsia in Shanghai
Among Shanghai births, higher standardized cord PFBS associated with preeclampsia (AOR 1.81) and overall hypertensive disorders of pregnancy (AOR 1.64) in a dose-response pattern.
- LA air toxics, especially PAHs, track preterm birth
In Los Angeles, entire-pregnancy PAHs associated with ~30% higher preterm-birth odds; ammonium nitrate PM2.5 and several traffic markers also raised odds.
- Prenatal PFAS tracked 5-year child overweight risk
In 412 Scandinavian mother–child pairs, higher maternal PFAS associated with higher child BMI and triceps skinfold z-scores and with overweight/obesity at age 5, with geographic differences.
- Prenatal PM2.5 linked to preterm birth in Puerto Rico
In the PROTECT birth cohort, an IQR rise in pregnancy-average PM2.5 was associated with higher preterm-birth risk among relatively healthy Puerto Rican women.
- Boreal canopy lichens declined from several drivers
Over a decade of Swedish forest inventory, hair lichens Alectoria, Usnea, and Bryoria became less common, with losses tied to forestry, nitrogen, and climate together—not one driver alone.
Learn alongside