Skip to main content

Table 3 Model characteristics

From: Cost-effectiveness of prostate cancer screening: a systematic review of decision-analytical models

Study Model-type Natural history model By (TNM) stage of cancer? TNM staging used Differentiation by Gleason grade? Gleason grading used Time horizon Deterministic sensitivity analysis Probabilistic sensitivity analysis
Chilcott et al. [23] Individual patient simulation No Yes Localised (T1–2), Locally advanced (T3–4) and metastatic Yes (G < 7, G = 7, G > 7) lifetime Yes Yes
Heijnsdijk et al. [25] Individual patient simulation Yes -MISCAN Yes 18 stages: each tumour stage (T1,2 etc) modelled individually Yes (G < 7, G = 7, G > 7) lifetime Yes No
Hummel and Chilcott [24] Individual patient simulation and cohort No Yes Localised (T1–2), Locally advanced (T3–4) and metastatic Yes (G < 7, G = 7, G > 7) lifetime Yes No
Keller et al. [29] Cohort model No Yes Low (≤T1a), intermediate (≤T2b), high risk (>T2b) and metastatic Yes G ≤ 6, G ≤ 7, G > 7 up to 70 years Yes Yes
Kobayashi et al. [27] Markov cycle tree (cohort) No Yes Localised (T1–2), Locally advanced (T3–4) and metastatic No up to 80 years old Yes No
Martin et al. [30] Cohort model No No None No 50 years (lifetime) Yes No
Pataky et al. [26] Individual patient simulation Yes -adapted FHCRC Yes Locoregional (≤T2a vs > T2)
distant disease
Assumed Yesa Not reported assumed lifetime Yes No
Roth et al. [28] Individual patient simulation Yes -adapted FHCRC Yes Locoregional (≤T2a vs > T2)
distant disease
Yesa Indirectly (8–10/ 2–7) lifetime Yes Yes
Wolstenholme et al. [31] Cohort model No Yes Localised (T1–2), Locally advanced (T3–4) and metastatic No lifetime(100 years old) Yes Yes
Shteynshlyuger & Andriole, [22] Population based model Yes No details provided lifetime Yes No
  1. a Based on values derived from the natural history, but Gleason is not explicitly modelled in the model-based economic evaluation