Good afternoon. I'm Robert Bynum, the PGY-6 at the University of Arizona in Tucson, down the street from Dr. Patel. I forgot the slide, but we have no disclosures. First, a special thank you to the research team, Sarah, who is on call today so that I can be here, and Dr. Dumont, who is my mentor. The Patient Protection and Affordable Care Act was introduced on March 23rd of 2010 and promised broad sweeping health care changes. Some of the provisions listed here potentially impacted trauma patients who are disproportionately uninsured and young. In 2010, the age 26 expansion was applied. And in 2014, the individual mandate, the Medicaid expansion, eligibility expansion, and the small business expansion were also phased in. These provisions increased the numbers of patients who had health insurance. Shown here is a heartening trend from our data detailing that the number of uninsured patients has decreased steadily since the introduction of the PPACA. The slope of this trend changed from 2013 to 2014, concurrent with the individual mandate and the other provisions. Prior studies have shown that there is a bias in the treatment of patients who have no insurance when examining with severe TBI. The Hefferman study showed that uninsured status leads to decreased rates of discharge to rehab. And there is a trend towards higher mortality in uninsured patients with severe TBI. This trend was again confirmed in this Haines study. And then the Rensen study showed a decrease in utilization of CT head with uninsured patients. Another study published in 2017 looked at the racial and insurance disparities in severe TBI patients and found an increase in the mortality and decrease in the procedural rates for uninsured patients. And finally, a McCutcheon paper from 2015 out of UCSD with data from before the enactment of the PPACA showed a decreased odds ratio of 0.76 for surgery for uninsured patients. Giving these data, we hypothesize that the bias in surgical treatment and inpatient mortality of severe TBI patients without insurance has diminished since the introduction of the PPACA. So to investigate this hypothesis, we queried the national inpatient sample for all patients over the age of 18 with a diagnosis of traumatic intracerebral hemorrhage from 2011 to 2016. We compared the surgical treatment and inpatient mortality against patient's insurance status. And we grouped our patients into 2011 to 2013, and then 2014 to 2016 years after the phase end of the individual mandate. Multivariate analysis was also performed controlling for gender, age, severity score, and teaching hospital status. Now, this is a busy slide. These are the demographic characteristics of our uninsured patient sample. I want to draw attention to a few things. First, the mean age of uninsured people increased over the time divisions. Secondly, the percentage of uninsured women increased, implying that more men have obtained insurance since the introduction of the PPACA. Also, the distribution of uninsured people when grouped by region shifted in an unexpected way. The percentage of people that did not have insurance that lived in the South increased, while in all of the regions it decreased. We weren't able to tease out the reasons for this from our data, but we suspect that it may have something to do with the Medicaid expansion in certain regions going into effect before they went into effect in the South. So certain states may not have actually expanded as rapidly as others. So the meat of our study, the surgical rates for severe TBI patients increased over the study period in both insured and uninsured groups, but the increase was disparate. From 2011 to 2014, the rates were equivalent, about 7% to 9%. However, in 2015 and 16, though the overall rate of surgery increased, there was a discrepancy between the two groups. And when demographic data is controlled for, uninsured patients had a surgical rate and procedural rate of 8% to 10%, and insured patients had a higher procedural rate of 10% to 11%. When the odds ratio for this data is shown and, sorry, when the odds ratio for the previously shown data is calculated, we see that the odds ratio for surgery from 2011 to 2013 was non-significantly less for uninsured patients than for insured patients. However, the odds ratio for surgery from 2014 to 2016 for uninsured patients is 0.78, which is similar to the odds ratio previously reported by the McCutcheon study for prior to the enactment of the PPACA. Also shown here, the odds ratio for inpatient mortality is higher in both groups, sorry, in both time frames for uninsured patients. So uninsured patients are more likely to die while inpatient across all years of our study. Our data show, unsurprisingly, that the number of uninsured patients has decreased after the introduction of the PPACA. We also show that uninsured patients with severe TBI after 2014 are less likely to undergo surgery. Our odds ratio is 0.78, which is comparable to the odds ratio previously published. The reason for this is elusive. We can't determine that from our data set. We also, incidentally, we also show that the odds ratio of inpatient mortality for uninsured patients is higher than for insured patients, and that this did not change with the introduction of the 2014 individual mandate. Therefore, we see the same bias that existed previous to the introduction of the PPACA than the same bias exists after the introduction that existed previously. Therefore, we reject our hypothesis. The limitations of this study are derived from the National Inpatient Sample being an administrative data set that relies heavily on self-reporting. It does not keep track of, nor does it verify individual data points. It is incumbent upon the reporting institution to do that. Also, hospital-level decisions are not included in our analysis, and we think that that may be the culprit for some of these biases. But we don't have a good answer for that either, because there's no way, from our database at least, to discover those individual hospital-level decisions. Thank you for your time. Come see us in Tucson in our new hospital when the weather's bearable, sometime between March and April. Any questions? Any questions? So my question is just to add granularity to the question. Yes, ma'am. It's just severe intracranial. Yes. So did you differentiate between subdural and neuronal? No, no, no, just an intracranial hemorrhage. Because of the coding, when it gets too granular, the numbers start to tease out and become less meaningful. OK. And did you have all the same substance there, between the time periods that were occurring? So it's the National Inpatient Sample. So it's, yeah, so it's a large national database. Yeah, yeah. OK. Is the National Inpatient Database equivalent to Vizient data, or is it completely different? I am not sure what the Vizient data is. So Vizient data looks at all the academic centers from across the country and contributes to a database that looks at administrative data for all diagnoses. And you're able to drill down to granular data, as well. And I think that that data, because it's reported and we all use it for studying questions like this, it's quite reliable. But I'm not sure if it's the same. So the National Inpatient Database includes both community teaching, non-teaching, rural, urban. It's a much broader database than what it sounds like this Vizient database is. And you just looked at TBI. Yes, ma'am, severe TBI. Not diffuse axonal injury versus, you know. So there might have been in that study, your data, more incidents of non-operative trauma than operative trauma. So they all had an intracranial hemorrhage of some sort. And then, yes. Yes, sir. I just want to ask you about, just a small trick about the decision-making. If some physician decided to do operation for this patient and the administrative or the hospital resources didn't can cope with, in this case, including the virus, or have a deficiency in material or resources to do such an operation. So that's an interesting question. And that's what we think may be what is behind this bias. Because at least in our institution, when patients come in with trauma, we have no idea what their insurance status is. And that's actually difficult for the physicians and those making the clinical decisions to discover it. It takes some doing to look through the EMR to find out what their insurance is. And frequently, we have no clue. It may be that in some hospitals, there just aren't the resources for that. We don't know that. And that particular, your question is, I think, a very important one that is actually underlying this bias. Why are certain patients that may be surgical in some hospitals not surgical in others? And it may be that there just aren't the resources at those places. We don't know that. Yes, sir. Thank you very much. Very interesting. Thank you.

Robert Bynum

PGY-6

University of Arizona

Patient Protection and Affordable Care Act

trauma patients