Data Set subject Unique observation identifier treatment - 1=received the treatment; 0=did not receive the treatment time - 1=time 1 (pre-period, or baseline); 2=time 2 (post-period, or follow-up) SEX-1-patient was male; 0=patient was female AGE - age in years num chron: total number of chronic conditions urban_rural: 1=live in an urban area; 0=live in or non-urban (rural, or suburban) area unemp_rate: unemployment rate in the subject's zip code med_hh_income: median household income in the subject's zip code disease burden: general score of disease states at baseline - higher scores indicate higher illness burden IP_visits_pmpm: number of inpatient admissions IP_cost PMPM: medical costs associated with inpatient admissions IP_cost PMPM1: medical costs associated with inpatient admissions +$1.00 for cost analyses using log link and gamma distribution IP_visit_flag; 1=had at least 1 inpatient (IP) admission; 0=did not have an IP admission total_cost_pmpm: total medical costs total_cost_PMPM1: total medical costs associated with ER visits +$1.00 for cost analyses using log link and gamma distribution Data Set The GENMOD Procedure Model Information DID.DIFF_IN_DIFF_HWK Distribution Link Function Dependent Variable Gamma Log total_cost_pmpm1 Number of Observations Read 200000 Number of Observations Used 200000 Class Level Information Class Levels Values treatment 2 01 time 2 12 Parameter Information Parameter Effect treatment time Prm1 Intercept Prm2 time 1 Prm3 time 2 Prm4 treatment 0 Prm5 treatment 1 Prm6 treatment*time 0 1 Prm7 treatment*time 0 2 Prm8 treatment*time 1 1 Prm9 treatment*time 1 2 Criteria For Assessing Goodness Of Fit Criterion DF Value Value/DF Deviance 2E5 243909.5404 1.2196 Scaled Deviance 2E5 232237.0246 1.1612 Pearson Chi-Square 2E5 345372.9584 1.7269 Scaled Pearson X2 2E5 328844.8173 1.6443 Log Likelihood -1463716.261 Full Log Likelihood -1463716.261 AIC (smaller is better) 2927442.5211 AICC (smaller is better) 2927442.5214 BIC (smaller is better) 2927493.5515 Algorithm converged. Analysis Of Maximum Likelihood Parameter Estimates Parameter DF Estimate Standard Error Wald 95% Confidence Limits Wald Chi-Square Pr > ChiSq Intercept 1 6.2842 0.0056 6.2733 6.2952 1268158 treatment Least Squares Means Standard Pr> Exponentiated Exponentiated treatment Estimate Error 0 6.3320 z Value 0.002815 2249.48 z Alpha Lower Upper total_cost_pmpm1 LS-Mean 6.34 LS-Means for treatment With 95% Confidence Limits 6.33 6.32 6.31 6.30 6.29 0 treatment 1 time Least Squares Means Standard Pr> Exponentiated Exponentiated time Estimate 1 6.3178 Error 0.003427 1843.31 total_cost_pmpm1 LS-Mean 6.30 6.31 6.32 LS-Means for time With 95% Confidence Limits 1 time 2 treatment time Least Squares Means time treatment Estimate Standard Error Pr> Exponentiated Exponentiated z Value |z Alpha Lower Upper Exponentiated Lower Upper 1 0 2 0 1 1 6.3333 6.3047 6.3308 0.003981 1590.31 total_cost_pmpm1 LS-Mean 6.34 6.32 6.30 6.28 LS-Means for treatment*time With 95% Confidence Limits 01 02 11 12 treatment*time Data Set The GENMOD Procedure Model Information DID.DIFF_IN_DIFF_HWK Distribution Link Function Dependent Variable Normal Identity total_cost_pmpm1 Number of Observations Read 200000 Number of Observations Used 200000 Class Level Information Class Levels Values treatment 2 01 time 2 12 Parameter Information Parameter Effect treatment time Prm1 Intercept Prm2 time 1 Prm3 time 2 Prm4 treatment 0 Prm5 treatment 1 Prm6 treatment*time 0 1 Prm7 treatment*time 0 2 Prm8 treatment*time 1 1 Prm9 treatment*time 1 2 Criteria For Assessing Goodness Of Fit Criterion DF Value Deviance 2E5 106515288908 Scaled Deviance 2E5 200000.0000 Value/DF 532587.0963 1.0000 Pearson Chi-Square 2E5 106515288908 532587.0963 Scaled Pearson X2 Log Likelihood 2E5 200000.0000 -1602335.879 Full Log Likelihood -1602335.879 AIC (smaller is better) 3204681.7581 AICC (smaller is better) 3204681.7584 BIC (smaller is better) 3204732.7885 Algorithm converged. 1.0000 Analysis Of Maximum Likelihood Parameter Estimates Standard Parameter DF Estimate Error Wald 95% Confidence Limits Wald Chi-Square Pr> ChiSq Intercept 1 536.0576 3.9738 528.2690 543.8461 18197.3 total_cost_pmpm1 LS-Mean 540 550 560 LS-Means for treatment With 95% Confidence Limits 0 treatment 1 total_cost_pmpm1 LS-Mean 560 555 550 545 time Estimate 1 554.37 2 549.53 time Least Squares Means Standard Error z Value Prz Alpha Lower Upper 2.4407 227.14 total_cost_pmpm1 LS-Mean 570 560 550 540 530 treatment*time Least Squares Means time treatment 1 0 Estimate 561.60 Standard Error z Value Pr> z Alpha 2.8348 198.11