Table 4.

Associations Between Sepsis Hospitalizations and Risk of Subsequent Fatal Coronary Heart Disease Events

	Sepsis Cases	Nonsepsis Controls	Risk of Fatal Coronary Heart Disease, Hazard Ratio (95% Confidence Interval)
	Fatal CHD Events /No. at Risk (%)	Fatal CHD Events /No. at Risk (%)	Crude	Model 1	Model 2	Model 3	Model 4
Traditional Cox proportional hazards model
	75/1070 (7.01)	185/5350 (3.46)	2.63 (1.97–3.52)	2.70 (1.99–3.67)	2.60 (1.91–3.55)	2.07 (1.42–3.03)	2.05 (1.51–2.78)
Piecewise Cox proportional hazards model
0–1 year	25/1070 (2.86)	38/5350 (0.71)	3.45 (2.07–5.77)	4.20 (2.35–7.49)	4.50 (2.49–8.14)	3.12 (1.35–7.23)	3.07 (1.81–5.20)
1–4 years	39/779 (5.01)	87/4630 (1.88)	3.56 (2.36–5.36)	3.66 (2.35–5.69)	3.54 (2.25–5.56)	3.29 (1.89–5.74)	2.36 (1.51–3.67)
4+ years	11/329 (3.34)	60/2210 (2.71)	1.11 (0.56–2.21)	1.63 (0.74–3.60)	1.39 (0.61–3.19)	1.15 (0.34–3.94)	0.92 (0.45–1.89)

Fatal CHD included acute CHD deaths and death ≤28 days of an acute myocardial infarction. Model 1: adjusted for sex, race, education, income, and region. Model 2: additional adjustment for tobacco use. Model 3: additional adjustment for atrial fibrillation, chronic lung disease, diabetes, hypertension, obesity, peripheral artery disease, stroke, chronic kidney disease, and albumin-to-creatinine ratio. Model 4: adjusted only for REasons for Geographic And Racial Differences in Stroke sepsis risk score.

Abbreviation: CHD, coronary heart disease.