. Author manuscript; available in PMC: 2009 Jun 25.

Published in final edited form as: J Am Stat Assoc. 2009 Jun 1;104(486):512–523. doi: 10.1198/jasa.2009.0017

Table 4.

Summary of parameter estimates using the nonlinear random effects models and the Bayesian hierarchical models

Random effects models	Non-linear Random Effects Models^* Using NLMIXED				Bayesian hierarchical models using WinBUGS
Random effects models	I None	IIa ε_i	IIIc ε_i, μ_iB	IVa ε_i, μ_iB, μ_iA	I None	IIa ε_i	IIIc ε_i, μ_iB	IVa ε_i, μ_iB, μ_iA
MSI specificity	₉₀₂920₉₃₇	₈₉₃912₉₃₂	₈₈₉909₉₂₉	₈₉₄914₉₃₄	₈₉₈917₉₃₆	₈₉₈916₉₃₄	₈₉₃912₉₃₈	₈₉₅914₉₃₉
MSI sensitivity	₇₃₅819₉₀₄	₈₉₂978₁₀₀₀	₈₈₀982₁₀₀₀	₉₂₂968₁₀₀₀	₇₄₅842₉₅₁	₈₄₃934₉₈₅	₈₇₂957₉₉₆	₇₄₀922₉₉₀
MUT specificity	₁₀₀₀1000₁₀₀₀	₉₀₆953₉₉₉	₈₉₈952₁₀₀₀	₉₄₇986₁₀₀₀	₉₁₇980₉₉₈	₉₂₆968₉₉₆	₉₁₆958₉₉₀	₉₃₅981₉₉₈
MUT sensitivity	₅₆₄622₆₇₉	₅₃₆594₆₅₃	₅₀₈656₈₀₅	₄₉₅641₇₈₆	₅₆₅621₆₇₈	₅₃₇590₆₄₅	₅₃₁630₇₂₆	₄₈₈645₈₀₁
Family-recruitment prevalence	₄₉₁555₆₁₈	₃₁₈495₆₇₃	₃₀₂465₆₂₉	₄₄₂532₆₂₂	₄₆₀536₆₁₀	₃₅₄520₆₉₄	₃₂₈476₆₄₁	₃₈₀532₆₈₅
Registry-recruitment prevalence	₂₈47₆₅	₀18₄₃	₀17₃₉	₈28₄₇	₂₄42₆₆	₉29₇₅	₉27₇₂	₁₁33₈₂
σ_ε (prevalence)	—	₄₄₅1034₁₆₂₄	₃₅₄904₁₄₅₄	₃₁₁601₈₉₀	—	₆₇₂1060₁₇₉₀	₆₀₅959₁₆₂₆	₄₉₇798₁₃₈₄
σ_{μ_A} (MSI sensitivity)	—	—	—	₈₈₀2529₄₁₇₇	—	—	—	₇₃₇1649₃₇₆₆
σ_{ν_A} (MSI specificity)	—	—	—	—	—	—	—	—
σ_{μ_B} (MUT sensitivity)	—	—	₁₁₁742₁₃₇₅	₁₃₇756₁₃₇₅	—	—	₆₀₁944₁₆₅₀	₅₉₇932₁₆₁₀
σ_{ν_B} (MUT specificity)	—	—	—	—	—	—	—	—

NOTE: The triple notation of _LP_U denotes the point estimate P with 95% confidence limits (L, U) for the nonlinear random effects models, or the posterior median P with 95% equal tailed credible limits (L, U) using Bayesian hierarchical models. The numbers have been multiplied by 1,000 for presentation.

95% confidence intervals based on normal approximation.