Skip to main content
. 2024 Jan 8;11:1310823. doi: 10.3389/fpubh.2023.1310823

Table 3.

Results of the mixed effects Bayesian INLA Cox proportional hazard regression model in the female breast cancer cohort, mutually adjusted and adjusted for baseline age class, year, month, patient ID, municipality of residence, and tumor morphology.

Mixed effects INLA
Cox proportional hazard regression
Female breast cancer cohort
All-cause death
N = 2,837; person-years = 10,445; n = 447
Fixed effects HR 95% CrI
Residence in SIN
Extra-SIN 1.00 (ref)
SIN 1.22 1.01–1.48
Pathological/clinical staging
TNM I 1.00 (ref)
TNM II 1.26 0.88–1.79
TNM III 2.77 1.93–3.97
TNM IV 17.05 11.94–24.34
NA 4.13 2.87–5.95
Histological grading
Grade 1 1.00 (ref)
Grade 2 1.93 0.94–3.96
Grade 3 2.50 1.20–5.23
NA 2.18 1.02–4.66
Proliferation index
Ki-67 ≤ 20% 1.00 (ref)
Ki-67 21–50% 1.42 1.10–1.83
Ki-67 > 50% 1.81 1.29–2.55
NA 1.09 0.41–2.89
Hormonal receptors status
ER- and PR- 1.00 (ref)
ER+ and/or PR+ 0.61 0.45–0.81
NA 0.88 0.33–2.30
Epidermal receptor status
HER2/neu- 1.00 (ref)
HER2/neu+ 0.59 0.44–0.79
NA 1.40 0.87–2.23
Topography
C50.0–1 1.00 (ref)
C50.2–5 0.88 0.65–1.18
C50.6–8 0.81 0.56–1.17
NA 0.95 0.67–1.34
Laterality
Right 1.00 (ref)
Left 1.03 0.84–1.26
Bilateral 1.65 1.01–2.68
NA 1.28 0.80–2.03

Province of Taranto, 2015–2020, follow-up to 31/12/2021. Time: days of follow-up. Outcome (incidence): all-cause death. N: incident cases and non-cases. n: incident cases.