|
|
|
|
|
|
| \begin{table}[h!] \begin{adjustbox}{max width=1.2\textwidth} \begin{threeparttable} |
| \caption{Doctor's Expected Utility from Different Prescription Choices} |
| \label{tab:Doctor's-expected-utility} { |
| \begin{tabular}{|c|c|c|} |
| \hline |
| & {\footnotesize{}Utility from treatment use} & {\footnotesize{}Utility loss from gatekeeping cost}\tabularnewline |
| \hline |
| |
| \hline |
| \multicolumn{3}{|l|}{{\footnotesize{}No treatment}}\tabularnewline |
| \hline |
| {\footnotesize{}(1) C \& DV} & {\footnotesize{}0} & {\footnotesize{}$-g\int_{L_{P}}U_{l}(\pi,P)dF(\pi|\gamma,\eta)-g\int_{H_{P}}U_{v}(\pi)dF(\pi|\gamma,\eta)$}\tabularnewline |
| \hline |
| {\footnotesize{}(2) PV} & {\footnotesize{}0} & {\footnotesize{}$-g\int_{L_{0}}U_{l}(\pi,0)dF(\pi|\gamma,\eta)-g\int_{H_{0}}U_{v}(\pi)dF(\pi|\gamma,\eta)$}\tabularnewline |
| \hline |
| |
| \multicolumn{3}{|l|}{{\footnotesize{}Severe treatment}}\tabularnewline |
| \hline |
| {\footnotesize{}(3) C \& DV} & {\footnotesize{}$P(\pi\in L2_{P}\cup H_{P})V_{v}(\hat{\pi})$} & {\footnotesize{}$-g\int_{L1_{P}}U_{l}(\pi,P)dF(\pi|\gamma,\eta)-g\int_{L2_{P}}\left[U_{l}(\pi,P)-U_{v}(\pi)\right]dF(\pi|\gamma,\eta)$}\tabularnewline |
| \hline |
| {\footnotesize{}(4) PV} & {\footnotesize{}$P(\pi\in L2_{0}\cup H_{0})V_{v}(\hat{\pi})$} & {\footnotesize{}$-g\int_{L1_{0}}U_{l}(\pi,0)dF(\pi|\gamma,\eta)-g\int_{L2_{0}}\left[U_{l}(\pi,0)-U_{v}(\pi)\right]dF(\pi|\gamma,\eta)$}\tabularnewline |
| \hline |
| |
| \multicolumn{3}{|l|}{{\footnotesize{}Simple treatment}}\tabularnewline |
| \hline |
| {\footnotesize{}(5) C \& (DV/no voucher)} & {\footnotesize{}$P(\pi\in L_{P}\cup H_{P})V_{l}(\hat{\pi},P)$} & {\footnotesize{}$-g\int_{H_{P}}\left[U_{v}(\pi)-U_{l}(\pi,P)\right]dF(\pi|\gamma,\eta)$}\tabularnewline |
| \hline |
| {\footnotesize{}(6) (PV \& DV)/voucher} & {\footnotesize{}$P(\pi\in L_{0}\cup H_{0})V_{l}(\hat{\pi},0)$} & {\footnotesize{}$-g\int_{H_{0}}\left[U_{v}(\pi)-U_{l}(\pi,0)\right]dF(\pi|\gamma,\eta)$}\tabularnewline |
| \hline |
| \multirow{2}{*}{{\footnotesize{}(7) PV/no voucher}} & \multirow{2}{*}{{\footnotesize{}$P(\pi\in L_{P}\cup H_{P})V_{l}(\hat{\pi},P)$}} & {\footnotesize{}$-g\int_{L1_{0}\backslash L1_{P}}U_{l}(\pi,0)dF(\pi |\gamma,\eta)-g\int_{L1_{P}\cup L2_{0}}\left[U_{l}(\pi,0)-U_{l}(\pi,P)\right]dF(\pi |\gamma,\eta)$}\tabularnewline |
| & & {\footnotesize{}$-g\int_{H_{0}}\left[U_{v}(\pi)-U_{l}(\pi,P)\right]dF(\pi|\gamma,\eta)$}\tabularnewline |
| \hline |
| |
| \multicolumn{3}{|l|}{{\footnotesize{}Patient choice between simple and severe treatment}}\tabularnewline |
| \hline |
| {\footnotesize{}(8) C \& (DV/no voucher)} & {\footnotesize{}$P(\pi\in L_{P})V_{l}(\hat{\pi},P)+P(\pi\in H_{P})V_{v}(\hat{\pi})$} & 0 |
| \tabularnewline |
| \hline |
| {\footnotesize{}(9) C \& (PV \& DV/voucher)} & {\footnotesize{}$P(\pi\in L_{0})V_{l}(\hat{\pi},0)+P(\pi\in H_{0})V_{v}(\hat{\pi})$} & 0 |
| \tabularnewline |
| \hline |
| \multirow{2}{*}{{\footnotesize{}(10) PV/no voucher}} & \multirow{2}{*}{{\footnotesize{}$P(\pi\in L_{P})V_{l}(\hat{\pi},P)+P(\pi\in H_{P})V_{v}(\hat{\pi})$}} & |
| {\footnotesize{}$-g\int_{L_{0}}U_{l}(\pi,0)dF(\pi|\gamma,\eta)$}\tabularnewline |
| & & {\footnotesize{}$+g\int_{L_{P}}U_{l}(\pi,P)dF(\pi|\gamma,\eta)+g\int_{H_{P}\backslash H_0}U_{v}(\pi)dF(\pi |\gamma,\eta)$}\tabularnewline |
| \hline |
|
|
| \end{tabular} |
| } \begin{tablenotes}[flushleft] \small |
| \item \emph{Notes}: C denotes control, DV denotes voucher given to doctor, PV denotes voucher given to patient. "No voucher" and "voucher" indicate if the doctor offers the subsidy when prescribing simple treatment. |
| \end{tablenotes} \end{threeparttable} \end{adjustbox} \end{table} |
|
|
|
|
|
|