effect of cms nonpayment policy on ssi 819


erroneously smaller standard errors.20 Thus, we chose to report unweighted results in our regression models to reduce type 1 error, as others have done.19 However, we reported weighted results for population demographics and other descriptive statistics to obtain nationally representative estimates. For our sensitivity analysis, we included an intervention


washout period in our model from January to December 2008 to account for anticipatory or lagged responses to CMS policy. Second, we ran a hierarchical regression model with hospital random effects to cluster patients by hospital and to detect any differential effect. This analysis was limited to data prior to 2012 due to NIS sampling redesign. Third, for patients undergoing lumbar laminectomies or total shoulder and elbow replacement procedures not targeted by CMS policy (identified using International Classification of Disease, Ninth Revision codes), we compared the Medicare group against 3 different control groups: non-Medicare 60–64-year-old patients, 65–69-year-old patients with private insurance, and Medicare 60–80-year-old patients (Table 1). These procedures had been used as com- parisons in previous studies examining SSIs.12 Patient visits that included a CMS targeted procedure were excluded from the control group. Finally, we ran the difference-in-differences models in 3 all-capture state inpatient databases (SID) from California, Florida, and New York.5 These states were chosen because of data availability and sizeable state populations. SID databases had the added value of tracking patient visits across different hospitals and facilities within the state. Thus, we ran another analysis that included readmissions due to SSIs that occurred within 30 days postdischarge. Prepolicy parallelism was tested in all models to ensure that difference-in-difference assumptions were satisfied. Analyses were performed using SAS 9.4 software (SAS Institute, Cary, NC) and STATA v14 software (StataCorp, College Station, TX). This study was determined by


our institutional review board to be exempt from the need for approval.


results


We identified 1,753,854 orthopedic discharges (unweighted: 367,017) and 10,211 SSIs (unweighted: 2,136) in our popula- tion. The probability of an SSI pre- and postpolicy were 0.7% and 0.5%, respectively, for the Medicare population and 0.6% and 0.5%, respectively, for the non-Medicare population. Figure 1 presents weighted quarterly data from 2000 to 2013 on number of SSIs per 1,000 procedures for Medicare and non-Medicare populations. This figure shows qualitatively that rates of SSI were decreasing steadily in parallel. In addition, our parallel trends test showed that trends prior to October 2008 SSI rates did not differ significantly (P=.6287), which satisfies difference-in-difference assumptions. The demo- graphics of the 2 groups were clinically similar despite having significant P values (Table 2). Table 3 shows that the number of SSIs per 1,000 orthopedic procedures decreased by 1.8 and 1.0 for Medicare and non- Medicare populations, respectively, after policy implementa- tion. Therefore, the absolute difference was −0.8 SSI per 1,000 orthopedic procedures. The adjusted regression-based RR estimates of SSI were similar, with decreases of 0.7 (95% confidence interval [CI], 0.6–0.8) for the Medicare population and 0.8 (95% CI, 0.7–0.9) for the non-Medicare population. The estimated policy effect to reduce SSI among the Medicare population compared to the control population was not significant (RR, 0.9; 95% CI, 0.8–1.1). Our results were robust across a series of sensitivity analyses.


First, we tested our model with an intervention washout period for 2008 and did not find a significant intervention effect


figure 1. Crude surgical site infection (SSI) rates from 2000 to 2013 by payer status. SSI rates were calculated as the number of SSIs that occurred per 1,000 orthopedic procedures performed for each discharge quarter. National Inpatient Sample data sampling weights were applied to account for variations in sampling method over the years. The vertical line represents the Centers for Medicare and Medicaid Services (CMS) policy implementation date. Trend lines (dashed: Medicare; solid: non-Medicare) for both populations were not adjusted for patient and hospital variables.


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60  |  Page 61  |  Page 62  |  Page 63  |  Page 64  |  Page 65  |  Page 66  |  Page 67  |  Page 68  |  Page 69  |  Page 70  |  Page 71  |  Page 72  |  Page 73  |  Page 74  |  Page 75  |  Page 76  |  Page 77  |  Page 78  |  Page 79  |  Page 80  |  Page 81  |  Page 82  |  Page 83  |  Page 84  |  Page 85  |  Page 86  |  Page 87  |  Page 88  |  Page 89  |  Page 90  |  Page 91  |  Page 92  |  Page 93  |  Page 94  |  Page 95  |  Page 96  |  Page 97  |  Page 98  |  Page 99  |  Page 100  |  Page 101  |  Page 102  |  Page 103  |  Page 104  |  Page 105  |  Page 106  |  Page 107  |  Page 108  |  Page 109  |  Page 110  |  Page 111  |  Page 112  |  Page 113  |  Page 114  |  Page 115  |  Page 116  |  Page 117  |  Page 118  |  Page 119  |  Page 120  |  Page 121  |  Page 122  |  Page 123  |  Page 124  |  Page 125  |  Page 126  |  Page 127  |  Page 128  |  Page 129  |  Page 130  |  Page 131  |  Page 132  |  Page 133  |  Page 134  |  Page 135  |  Page 136