WEBVTT 00:00:05.270 --> 00:00:07.310 position:50% align:middle As she said, my name is Jacqueline Nikpour. 00:00:07.310 --> 00:00:12.520 position:50% align:middle I am currently finishing up my postdoctoral fellowship at the University of Pennsylvania, 00:00:12.520 --> 00:00:19.290 position:50% align:middle but I actually have the distinct pleasure of doing this work that I'm going to share today while I was still a 00:00:19.290 --> 00:00:22.450 position:50% align:middle doctoral student at Duke University School of Nursing. 00:00:22.450 --> 00:00:31.180 position:50% align:middle NCSBN funded this work as part of their Centre of Regulatory Excellence in engaging predoctoral scholars, 00:00:31.180 --> 00:00:34.050 position:50% align:middle early career folks in getting involved in some of their work. 00:00:34.050 --> 00:00:37.750 position:50% align:middle So, I am extraordinarily grateful to NCSBN for their support. 00:00:37.750 --> 00:00:43.470 position:50% align:middle And I also want to just acknowledge some of the folks, both from Duke and from Penn, 00:00:43.470 --> 00:00:48.620 position:50% align:middle who have been part of my training, been part of my research, 00:00:48.620 --> 00:00:53.060 position:50% align:middle and have been colleagues at the Veterans Health Administration, which is a big part of what I'm going 00:00:53.060 --> 00:00:54.850 position:50% align:middle to be talking about today. 00:00:54.850 --> 00:01:02.610 position:50% align:middle So, as Daisy mentioned, I'm a primary care researcher pretty much for 90% 00:01:02.610 --> 00:01:04.210 position:50% align:middle of the work that I do. 00:01:04.210 --> 00:01:12.130 position:50% align:middle And I'm really interested in the nursing workforce and how nurses can be a part of this whole era of primary 00:01:12.130 --> 00:01:15.350 position:50% align:middle care transformation and healthcare reform that we're part of. 00:01:15.350 --> 00:01:19.220 position:50% align:middle I have spent a lot of time in the nursing world and a lot of time in the health policy world, 00:01:19.220 --> 00:01:23.050 position:50% align:middle and I've seen how little the two of them talk, and it's a problem. 00:01:23.050 --> 00:01:28.290 position:50% align:middle So, I'm hopeful that this will be salient to folks on, kind of, both sides of that aisle. 00:01:28.290 --> 00:01:33.910 position:50% align:middle So, I want to just quickly acknowledge my dissertation committee and co-authors, Dr. Marion Broome, 00:01:33.910 --> 00:01:39.556 position:50% align:middle Susan Silva, and Kelli Allen as well as some of my colleagues from the Veterans Health Administration 00:01:39.556 --> 00:01:43.650 position:50% align:middle as well as my funding from the Center for Health Outcomes and Policy Research. 00:01:43.650 --> 00:01:49.240 position:50% align:middle I have the pleasure of working with Dr. Aiken who you all heard from yesterday and Dr. Brooks Carthon who you 00:01:49.240 --> 00:01:52.920 position:50% align:middle heard from this morning, and it's been really wonderful. 00:01:52.920 --> 00:01:58.490 position:50% align:middle As well as the Robert Wood Johnson Foundation Future of Nursing Scholars, which also funded me during my PhD 00:01:58.490 --> 00:02:02.550 position:50% align:middle program and throughout my post-doc. 00:02:02.550 --> 00:02:05.950 position:50% align:middle So, a little bit of background about what I'm going to talk about today. 00:02:05.950 --> 00:02:11.320 position:50% align:middle We're really going to focus on chronic pain and the role of nurse practitioners in caring for chronic pain 00:02:11.320 --> 00:02:12.870 position:50% align:middle patients in the U.S. 00:02:12.870 --> 00:02:19.450 position:50% align:middle So, chronic, non-cancer, musculoskeletal pain will impact about 100 million 00:02:19.450 --> 00:02:23.540 position:50% align:middle adult Americans at some point in their lifetime, so that's a third of the U.S. 00:02:23.540 --> 00:02:25.140 position:50% align:middle population. 00:02:25.140 --> 00:02:30.900 position:50% align:middle And this is one of the fastest growing reasons why people are seeking healthcare. 00:02:30.900 --> 00:02:35.880 position:50% align:middle Chronic pain is the third highest category of all healthcare spending in the U.S. 00:02:35.880 --> 00:02:41.660 position:50% align:middle And as we've, kind of, come out of this era of pain is the fifth vital sign, 00:02:41.660 --> 00:02:47.700 position:50% align:middle and we don't want to just prescribe an opioid medication for every person who comes in saying they 00:02:47.700 --> 00:02:52.960 position:50% align:middle have chronic pain for three months or more, the guidelines for treating patients with chronic pain 00:02:52.960 --> 00:02:56.030 position:50% align:middle in primary care have really evolved over the last decade. 00:02:56.030 --> 00:03:01.080 position:50% align:middle So, not only are we thinking about treating patients with opioid medications where clinically 00:03:01.080 --> 00:03:05.310 position:50% align:middle indicated inappropriate, but we're also thinking about non-opioid medications, 00:03:05.310 --> 00:03:09.410 position:50% align:middle things like NSAIDs that are prescription-based, anticonvulsants, 00:03:09.410 --> 00:03:12.550 position:50% align:middle muscular relaxants, even antidepressants. 00:03:12.550 --> 00:03:17.340 position:50% align:middle We're also thinking about the physical and emotional and mental, genetic, 00:03:17.340 --> 00:03:23.810 position:50% align:middle environmental components of chronic pain that make it a particularly difficult condition to treat. 00:03:23.810 --> 00:03:27.190 position:50% align:middle It's not like there's a simple cause or a simple solution. 00:03:27.190 --> 00:03:33.356 position:50% align:middle It really is, kind of, a full body and mind connection and experience for the people 00:03:33.356 --> 00:03:36.410 position:50% align:middle who live with chronic pain. 00:03:36.410 --> 00:03:39.640 position:50% align:middle Chronic pain care can be really difficult to access. 00:03:39.640 --> 00:03:45.400 position:50% align:middle One of our speakers earlier talked about health professional shortage areas, and many patients, 00:03:45.400 --> 00:03:49.900 position:50% align:middle especially in rural areas, who may or may not have access to a primary 00:03:49.900 --> 00:03:54.750 position:50% align:middle care provider, often don't have access to a pain specialist or other resources. 00:03:54.750 --> 00:03:59.730 position:50% align:middle Or other providers like a physical therapist, or a chiropractor, or acupuncturist, 00:03:59.730 --> 00:04:04.350 position:50% align:middle or non-pharmacologic modalities of care can be really hard to access in areas, 00:04:04.350 --> 00:04:09.550 position:50% align:middle especially where you can already barely access primary care to begin with. 00:04:09.550 --> 00:04:13.500 position:50% align:middle And again, chronic pain is most often seen in a primary care setting. 00:04:13.500 --> 00:04:19.380 position:50% align:middle It's where it often presents itself or where patients often go for their first line, kind of, treatment. 00:04:19.380 --> 00:04:25.290 position:50% align:middle And if there's no one else to refer these patients out to, then a lot of the times primary care providers are 00:04:25.290 --> 00:04:30.230 position:50% align:middle the ones responsible for managing these patients' chronic pain over an extended period of time, 00:04:30.230 --> 00:04:32.820 position:50% align:middle often months or even years. 00:04:32.820 --> 00:04:38.550 position:50% align:middle All right, so chronic pain is a particularly difficult issue amongst the veteran population. 00:04:38.550 --> 00:04:44.270 position:50% align:middle So, veterans compared to the general population have higher rates of chronic pain as well as higher rates 00:04:44.270 --> 00:04:48.580 position:50% align:middle of opioid overdose and associated mental health conditions. 00:04:48.580 --> 00:04:53.670 position:50% align:middle We are in this era where we've got a rapid expansion of younger veterans who are returning 00:04:53.670 --> 00:04:55.560 position:50% align:middle from overseas conflicts. 00:04:55.560 --> 00:05:02.070 position:50% align:middle And the VA is an interesting kind of institution because it's not like the private sector where you 00:05:02.070 --> 00:05:08.310 position:50% align:middle cannot see patients who are Medicaid-insured or not accept certain types of insurance. 00:05:08.310 --> 00:05:15.060 position:50% align:middle They have a legal obligation to care for all patients who qualify for VA care regardless of their background 00:05:15.060 --> 00:05:16.970 position:50% align:middle or insurance status. 00:05:16.970 --> 00:05:22.750 position:50% align:middle VA also operates on a fixed budgetary model that results in lower salaries for providers compared to the 00:05:22.750 --> 00:05:28.310 position:50% align:middle private sector and, as a result, has a real difficulty expanding to or adapting to these 00:05:28.310 --> 00:05:32.370 position:50% align:middle rapid patient increases. 00:05:32.370 --> 00:05:35.340 position:50% align:middle And so this is where nurse practitioners really come in. 00:05:35.340 --> 00:05:41.480 position:50% align:middle As we're thinking about not only the provider shortage and the role of NPs in treating patients who are 00:05:41.480 --> 00:05:46.320 position:50% align:middle in health professional shortage areas, but as we think about chronic pain and what it is and 00:05:46.320 --> 00:05:52.240 position:50% align:middle how we experience it, all of the various biopsychosocial elements of chronic 00:05:52.240 --> 00:05:58.380 position:50% align:middle pain require a really holistic approach to treatment, which is exactly how NPs are educated, and trained, 00:05:58.380 --> 00:06:00.060 position:50% align:middle and practiced every day. 00:06:00.060 --> 00:06:02.480 position:50% align:middle We, as nurses, see the full patient. 00:06:02.480 --> 00:06:07.360 position:50% align:middle We see all the different kinds of factors that can influence a patient's health status and their 00:06:07.360 --> 00:06:10.370 position:50% align:middle perception of pain, and we treat these patients accordingly. 00:06:10.370 --> 00:06:16.110 position:50% align:middle So, NPs are really ideally aligned, especially in the VA and especially in rural areas, 00:06:16.110 --> 00:06:20.650 position:50% align:middle to care for patients with chronic pain experiences. 00:06:20.650 --> 00:06:25.990 position:50% align:middle But despite all this, as we all know, there are still major barriers in a lot of states, 00:06:25.990 --> 00:06:32.710 position:50% align:middle especially in the private sector in regards to what a nurse practitioner is legally obliged or allowed to do 00:06:32.710 --> 00:06:37.230 position:50% align:middle and what the practice that they work in will allow them to do, what we call scope of practice restrictions. 00:06:37.230 --> 00:06:43.470 position:50% align:middle So, in the VA, which is where we're going to be talking about today, nurse practitioners since 2016 have had 00:06:43.470 --> 00:06:46.870 position:50% align:middle full scope of practice, meaning they are not required to have physician 00:06:46.870 --> 00:06:50.830 position:50% align:middle oversight or physician sign-off on the care that they deliver. 00:06:50.830 --> 00:06:56.020 position:50% align:middle But in the private sector, many states still have a patchwork kind of regulatory 00:06:56.020 --> 00:06:59.370 position:50% align:middle environment for the care that MPs are able to deliver. 00:06:59.370 --> 00:07:03.940 position:50% align:middle So, if you live where I live, which is in a semi-rural area that's on the border 00:07:03.940 --> 00:07:08.400 position:50% align:middle of three states, you can go to three different states within a five-mile radius, 00:07:08.400 --> 00:07:11.280 position:50% align:middle depending on where you're practicing, if you've got multiple employers, 00:07:11.280 --> 00:07:17.030 position:50% align:middle and the care that you are allowed and able to deliver is going to vary tremendously. 00:07:17.030 --> 00:07:18.910 position:50% align:middle So, we know that this is bad for patients. 00:07:18.910 --> 00:07:22.640 position:50% align:middle We know that this is bad for NPs, creates an undue administrative burden, 00:07:22.640 --> 00:07:31.290 position:50% align:middle and ultimately results in time and resources taken away from providing high-quality patient care. 00:07:31.290 --> 00:07:37.800 position:50% align:middle And so one of the things that we wanted to look at going into this study was one of the barriers for nurse 00:07:37.800 --> 00:07:42.100 position:50% align:middle practitioners in the private sector to achieving full scope of practice. 00:07:42.100 --> 00:07:49.880 position:50% align:middle Is this concern of policymakers and legislators and lobbyists who fund policymakers and legislators that 00:07:49.880 --> 00:07:55.480 position:50% align:middle nurse practitioners receiving full scope of practice from a policy standpoint would result 00:07:55.480 --> 00:08:03.890 position:50% align:middle in overprescribing opioid medications and contributing to the ongoing opioid overdose crisis? 00:08:03.890 --> 00:08:08.880 position:50% align:middle I hear a couple, like, groans in the crowd because you and I know that that 00:08:08.880 --> 00:08:13.140 position:50% align:middle isn't what's happening and that's not what nurse practitioners are doing, 00:08:13.140 --> 00:08:20.950 position:50% align:middle but this perception of NPs as being less qualified or unable to deliver effective high-quality and safe care 00:08:20.950 --> 00:08:26.450 position:50% align:middle is really, kind of, manifesting itself as we exist in this ongoing opioid overdose crisis. 00:08:26.450 --> 00:08:35.800 position:50% align:middle And it's, kind of, presented an opportunity for detractors to degrade from the value of the profession. 00:08:35.800 --> 00:08:41.903 position:50% align:middle But we really didn't have the evidence when this concern first started coming out around 2017, 2018, 00:08:41.903 --> 00:08:49.680 position:50% align:middle to say, yeah, NPs are prescribing more or less opioid medications compared to a physician colleague. 00:08:49.680 --> 00:08:51.840 position:50% align:middle And so we wanted to address that evidence. 00:08:51.840 --> 00:08:58.220 position:50% align:middle And thinking of the theme of our conference today from data to policy, we really wanted to address a timely 00:08:58.220 --> 00:09:05.530 position:50% align:middle and salient policy issue of will expanding nurse practitioner scope practice result in this influx 00:09:05.530 --> 00:09:09.650 position:50% align:middle of unnecessary opioid overprescription. 00:09:09.650 --> 00:09:14.330 position:50% align:middle I should give a disclaimer that this was pre-COVID and obviously things have changed a lot, 00:09:14.330 --> 00:09:20.710 position:50% align:middle but this is still a concern that's remaining, and we'll talk about what that looks like. 00:09:20.710 --> 00:09:25.840 position:50% align:middle So, our aims for this study were, first, to take data from the VA, which again, 00:09:25.840 --> 00:09:32.930 position:50% align:middle full scope of practice, exists and say, okay, in a national health system where NPs are able 00:09:32.930 --> 00:09:40.190 position:50% align:middle to deliver high-quality care without oversight of a physician, are there differences in the amount, 00:09:40.190 --> 00:09:46.110 position:50% align:middle and the dosage, and the length of an opioid prescription between a physician, a nurse practitioner, 00:09:46.110 --> 00:09:48.850 position:50% align:middle and with your physician assistants in there as well? 00:09:48.850 --> 00:09:53.980 position:50% align:middle We wanted to see not only are there differences in how these different types of providers are caring 00:09:53.980 --> 00:10:00.050 position:50% align:middle for chronic pain patients in terms of opioids, but also were there differences in how long a patient 00:10:00.050 --> 00:10:05.660 position:50% align:middle was prescribed an opioid medication for and the actual dosage that they received on a daily basis? 00:10:05.660 --> 00:10:12.530 position:50% align:middle So, CDC says that 90, what we call, morphine milligram equivalents or more is considered a 00:10:12.530 --> 00:10:18.930 position:50% align:middle high-dose opioid and should be used in extremely sparing conditions. 00:10:18.930 --> 00:10:21.910 position:50% align:middle We also wanted to see if there were differences in non-opioid medications, 00:10:21.910 --> 00:10:25.596 position:50% align:middle which are increasingly being recommended, again, by the CDC guidelines, 00:10:25.596 --> 00:10:30.035 position:50% align:middle the National Academy of Medicine, and other governing agencies, to see, okay, 00:10:30.035 --> 00:10:36.320 position:50% align:middle are there differences between these groups and how we're caring for chronic pain as, sort of, a whole. 00:10:36.320 --> 00:10:37.280 position:50% align:middle So, that was our first goal. 00:10:37.280 --> 00:10:42.770 position:50% align:middle Our second goal was to see are there differences between these groups of providers in some of the 00:10:42.770 --> 00:10:48.940 position:50% align:middle demographic and clinical characteristics that might be interacting with how a provider would 00:10:48.940 --> 00:10:50.010 position:50% align:middle treat chronic pain. 00:10:50.010 --> 00:10:55.570 position:50% align:middle So, for example, if there's maybe one provider group that's caring for sicker patients or more 00:10:55.570 --> 00:10:57.830 position:50% align:middle complex patients, what does that look like? 00:10:57.830 --> 00:11:00.990 position:50% align:middle How is that impacting how these providers are caring for these patients? 00:11:00.990 --> 00:11:07.290 position:50% align:middle So, we wanted to look across the groups and then within the groups to see how are we managing chronic pain 00:11:07.290 --> 00:11:10.250 position:50% align:middle in this national health system. 00:11:10.250 --> 00:11:17.530 position:50% align:middle This was a descriptive correlational study of chronic pain patients in the VA nationally and their 00:11:17.530 --> 00:11:19.550 position:50% align:middle primary care providers. 00:11:19.550 --> 00:11:23.600 position:50% align:middle We used what was called summary records or what we dubbed summary records. 00:11:23.600 --> 00:11:31.130 position:50% align:middle So, what we did is we took data from clinical databases across the VA which are stored in, what they call, 00:11:31.130 --> 00:11:33.470 position:50% align:middle their corporate data warehouse. 00:11:33.470 --> 00:11:37.021 position:50% align:middle And for each patient, we took all of their visits for the year and, kind of, 00:11:37.021 --> 00:11:40.070 position:50% align:middle collapsed them down into one summary record. 00:11:40.070 --> 00:11:45.740 position:50% align:middle So, each patient that we were looking at, we were, sort of, looking at the full, kind of, 00:11:45.740 --> 00:11:49.190 position:50% align:middle year-long care that they received, including multiple visits, 00:11:49.190 --> 00:11:50.990 position:50% align:middle potentially with multiple providers. 00:11:50.990 --> 00:11:53.330 position:50% align:middle We'll talk about that in a minute. 00:11:53.330 --> 00:11:58.930 position:50% align:middle This was a study that took place from...with data from October 2015 to September 2016. 00:11:58.930 --> 00:12:04.160 position:50% align:middle VA operates on an October to September fiscal year schedule. 00:12:04.160 --> 00:12:09.010 position:50% align:middle And you'll note that, within this time, this is around the time when NPs were starting to have 00:12:09.010 --> 00:12:13.518 position:50% align:middle been granted full scope of practice around, I think it was, March of 2016. 00:12:13.518 --> 00:12:17.390 position:50% align:middle So, kind of, right smack dab in the middle. 00:12:17.390 --> 00:12:24.240 position:50% align:middle Before this date, there were already a number of VAs that were expanding scope of practice for NPs and who 00:12:24.240 --> 00:12:30.830 position:50% align:middle this regulatory change didn't really affect because NPs were already, kind of, practicing independently. 00:12:30.830 --> 00:12:33.670 position:50% align:middle But as this regulation, kind of, took hold over the coming months, 00:12:33.670 --> 00:12:42.350 position:50% align:middle more and more practices were adopting policies so that NPs did not require additional physician oversight. 00:12:42.350 --> 00:12:48.420 position:50% align:middle And so the amount of change that took place, kind of, varied, but overall in the VA compared to the 00:12:48.420 --> 00:12:55.690 position:50% align:middle private sector, NPs had significantly more autonomy during this time period. 00:12:55.690 --> 00:12:58.600 position:50% align:middle Our primary outcomes, again, were the prescriptions of opioid and 00:12:58.600 --> 00:13:02.880 position:50% align:middle non-opioid medications, and we looked at opioid dosage and the length 00:13:02.880 --> 00:13:07.230 position:50% align:middle of prescription for those medications. 00:13:07.230 --> 00:13:11.960 position:50% align:middle These next couple of slides are really fancy ways of describing our methodology, 00:13:11.960 --> 00:13:14.320 position:50% align:middle but I'm going to simplify it as much as I can. 00:13:14.320 --> 00:13:16.730 position:50% align:middle So, we took these two major data sources. 00:13:16.730 --> 00:13:23.160 position:50% align:middle So, we took the clinical data from this corporate data warehouse that had prescription data for the year that 00:13:23.160 --> 00:13:27.340 position:50% align:middle had comorbidities and patient characteristics. 00:13:27.340 --> 00:13:32.730 position:50% align:middle And we merged that with what's called SHEP, or the Survey of Health Experience of Patients. 00:13:32.730 --> 00:13:37.100 position:50% align:middle This is basically for non-VA folks, HCAHPS or CAHPS for the VA. 00:13:37.100 --> 00:13:39.270 position:50% align:middle It's a patient satisfaction survey. 00:13:39.270 --> 00:13:44.759 position:50% align:middle But what we got from this survey was, number one, the patient ID. 00:13:44.759 --> 00:13:50.650 position:50% align:middle So, we were able to connect it along with the type of provider that was their assigned primary care provider 00:13:50.650 --> 00:13:55.630 position:50% align:middle and who they had a visit with, which is why they got that survey for that patient. 00:13:55.630 --> 00:13:59.160 position:50% align:middle So, at least one visit with their assigned primary care provider. 00:13:59.160 --> 00:14:02.421 position:50% align:middle They may have had other visits with other providers during that timeframe, 00:14:02.421 --> 00:14:08.600 position:50% align:middle but the person who they're getting these surveys for is really the person who's responsible for managing the 00:14:08.600 --> 00:14:09.920 position:50% align:middle majority of their care. 00:14:09.920 --> 00:14:13.410 position:50% align:middle And they may have someone step in for them if they're not in for an appointment, 00:14:13.410 --> 00:14:20.270 position:50% align:middle but that's really the person that holds decision-making responsibility for the long-term for these patients. 00:14:20.270 --> 00:14:27.230 position:50% align:middle We got patient demographics, we got VA, facility, and state, and we also had patients self-reporting 00:14:27.230 --> 00:14:32.266 position:50% align:middle their overall health and their overall mental health as part of the survey. 00:14:32.266 --> 00:14:39.330 position:50% align:middle So, these two data sources were merged together, and we took from all these different smaller databases 00:14:39.330 --> 00:14:42.450 position:50% align:middle within the corporate data warehouse, brought these all together, 00:14:42.450 --> 00:14:48.370 position:50% align:middle created this merged analysis data set, merged that with the SHEP records, 00:14:48.370 --> 00:14:54.220 position:50% align:middle and we ended up with about 275,000 patients who had an assigned primary care provider, 00:14:54.220 --> 00:14:57.500 position:50% align:middle some of who did and did not have chronic pain. 00:14:57.500 --> 00:15:02.990 position:50% align:middle We further then whittled that down based on some of our inclusion and exclusion criteria, 00:15:02.990 --> 00:15:07.360 position:50% align:middle so patients who had no pain diagnosis were removed. 00:15:07.360 --> 00:15:13.030 position:50% align:middle We had four pain diagnoses that we were looking at, osteoarthritis, back pain, neck pain... 00:15:13.030 --> 00:15:17.768 position:50% align:middle or, I'm sorry, upper back pain, lower back pain, neck pain, and osteoarthritis. 00:15:17.768 --> 00:15:24.470 position:50% align:middle And then we took away patients who had a comorbidity of kidney failure, of liver failure, and of cancer, 00:15:24.470 --> 00:15:30.080 position:50% align:middle because these are diagnoses that are going to substantially impact the patient's course of treatment 00:15:30.080 --> 00:15:33.880 position:50% align:middle beyond just what type of provider is caring for them. 00:15:33.880 --> 00:15:41.150 position:50% align:middle We removed patients who were in, I believe it was, nine states where NPs did not have authority at the 00:15:41.150 --> 00:15:46.438 position:50% align:middle state level to prescribe Schedule II medications, which includes your opioid classes. 00:15:46.438 --> 00:15:49.920 position:50% align:middle And this is something that isn't governed by the VA. 00:15:49.920 --> 00:15:55.116 position:50% align:middle The VA, kind of, covered in their regulatory changes physician oversight, 00:15:55.116 --> 00:16:00.400 position:50% align:middle but the ability to prescribe certain medications is still determined by the state. 00:16:00.400 --> 00:16:05.700 position:50% align:middle So, patients who were in states where NPs or PAs could not prescribe these medications were taken out because 00:16:05.700 --> 00:16:08.540 position:50% align:middle there's really nothing to compare it to. 00:16:08.540 --> 00:16:12.430 position:50% align:middle And then, if they had missing comorbidity data or it was not clear who their assigned primary care 00:16:12.430 --> 00:16:14.470 position:50% align:middle provider was, these patients were also removed. 00:16:14.470 --> 00:16:21.290 position:50% align:middle So, we ended up with just under about 40,000 patients who had a chronic pain diagnosis and were cared 00:16:21.290 --> 00:16:27.700 position:50% align:middle for in the VA in this time period between 2015 and 2016 by a physician, a nurse practitioner, 00:16:27.700 --> 00:16:31.770 position:50% align:middle or a physician assistant. 00:16:31.770 --> 00:16:36.560 position:50% align:middle Within our data, we had the majority of patients being cared for by a physician, 00:16:36.560 --> 00:16:44.750 position:50% align:middle and about 8,400 patients being cared for by an NP, and just under 3,000 being cared for by a PA. 00:16:44.750 --> 00:16:49.485 position:50% align:middle Again, we looked at different patient characteristics and outcomes in terms of opioid 00:16:49.485 --> 00:16:52.160 position:50% align:middle and non-opioid prescriptions. 00:16:52.160 --> 00:16:57.630 position:50% align:middle Demographics we looked at included age, race, ethnicity, assigned sex at birth, 00:16:57.630 --> 00:17:03.010 position:50% align:middle and their education level of whether or not this is a person who had secondary education or not. 00:17:03.010 --> 00:17:06.929 position:50% align:middle We can talk about that a little bit more, but we were interested to see if there were 00:17:06.929 --> 00:17:11.669 position:50% align:middle differences among, let's say, patients who may be more likely to be employed 00:17:11.669 --> 00:17:17.650 position:50% align:middle in a manual labor field and may be at higher risk for receiving a chronic pain diagnosis 00:17:17.650 --> 00:17:21.150 position:50% align:middle and certain medications. 00:17:21.150 --> 00:17:26.050 position:50% align:middle This is a very simplified way of looking at our sample characteristics. 00:17:26.050 --> 00:17:31.690 position:50% align:middle So, within each column, you'll see the patient characteristics that were 00:17:31.690 --> 00:17:36.540 position:50% align:middle statistically significantly more likely to be found amongst that provider group. 00:17:36.540 --> 00:17:42.737 position:50% align:middle So, physicians were more likely to see patients who were age 65 plus, who had some 00:17:42.737 --> 00:17:48.440 position:50% align:middle post-secondary education, who self-reported fair or poor health and fair or poor 00:17:48.440 --> 00:17:53.870 position:50% align:middle mental health, who had a comorbidity of hypertension, and who had certain mental health conditions 00:17:53.870 --> 00:17:57.080 position:50% align:middle like psychoses or substance abuse. 00:17:57.080 --> 00:18:04.340 position:50% align:middle Nurse practitioners were more likely to care for female veterans, veterans who were non-Hispanic White, 00:18:04.340 --> 00:18:09.300 position:50% align:middle patients who had five or more comorbidities, and comorbidities like congestive heart 00:18:09.300 --> 00:18:11.700 position:50% align:middle failure and hypothyroidism. 00:18:11.700 --> 00:18:18.020 position:50% align:middle So, one of the things I thought was really interesting here is we hear this argument all the time of, "Oh, 00:18:18.020 --> 00:18:21.040 position:50% align:middle physicians treat more complex patients than NPs." 00:18:21.040 --> 00:18:27.950 position:50% align:middle Well, these are patients who are more like...or these are providers who are treating patients with all these, 00:18:27.950 --> 00:18:29.430 position:50% align:middle you know, long lists of comorbidities. 00:18:29.430 --> 00:18:33.350 position:50% align:middle So, that isn't necessarily the case all the time. 00:18:33.350 --> 00:18:38.240 position:50% align:middle And then PAs were also statistically more likely to care for White patients and patients who had a 00:18:38.240 --> 00:18:42.220 position:50% align:middle diagnosis of osteoarthritis. 00:18:42.220 --> 00:18:46.160 position:50% align:middle And if you take one thing away from my talk today, I want it to be from this slide. 00:18:46.160 --> 00:18:53.460 position:50% align:middle So, this was after adjusting for all of our covariates, all our demographics, and clinical characteristics. 00:18:53.460 --> 00:19:01.140 position:50% align:middle We wanted to know is there a difference in physicians and NPs in terms of who is prescribing higher or more 00:19:01.140 --> 00:19:04.510 position:50% align:middle amounts of opioids and are there differences in dosage and length. 00:19:04.510 --> 00:19:09.830 position:50% align:middle So, we actually found that compared to physicians, nurse practitioners had lower odds 00:19:09.830 --> 00:19:11.411 position:50% align:middle of prescribing an opioid. 00:19:11.411 --> 00:19:20.150 position:50% align:middle They had about 12.8% lower odds of prescribing opioid medication compared to their physician colleagues, 00:19:20.150 --> 00:19:22.780 position:50% align:middle and this was statistically significant. 00:19:22.780 --> 00:19:28.320 position:50% align:middle Physician assistants also had lower odds of prescribing an opioid compared to physicians. 00:19:28.320 --> 00:19:34.270 position:50% align:middle About 16.3% lower odds, but there was no difference statistically between nurse 00:19:34.270 --> 00:19:36.370 position:50% align:middle practitioners and physician assistants. 00:19:36.370 --> 00:19:40.484 position:50% align:middle And across all three groups, there were no significant differences in the amount 00:19:40.484 --> 00:19:44.790 position:50% align:middle of high-dose or long-term, long-term meaning 90 days or more, 00:19:44.790 --> 00:19:47.390 position:50% align:middle opioid medications between the three provider groups. 00:19:47.390 --> 00:19:53.630 position:50% align:middle So, again, thinking of our conference theme from data to policy, this is a clear, 00:19:53.630 --> 00:20:00.100 position:50% align:middle actionable piece of evidence that can be shown to policymakers that we have been taking to policymakers, 00:20:00.100 --> 00:20:07.340 position:50% align:middle to say, "Hey, if you're concerned about how NPs are caring for patients who have chronic pain and this 00:20:07.340 --> 00:20:10.570 position:50% align:middle opioid crisis that we're in, we found on a national level that this 00:20:10.570 --> 00:20:11.420 position:50% align:middle is not happening. 00:20:11.420 --> 00:20:17.012 position:50% align:middle In fact, they have lower odds of prescribing an opioid." 00:20:17.012 --> 00:20:21.830 position:50% align:middle And if you take a second piece away from my talk today, I want it to be from this slide. 00:20:21.830 --> 00:20:27.260 position:50% align:middle So, we not only were interested in the provider group, but all the characteristics of patients within and 00:20:27.260 --> 00:20:28.790 position:50% align:middle between those provider groups. 00:20:28.790 --> 00:20:35.940 position:50% align:middle So, we found that patients who were more...who had higher odds of receiving an opioid were 00:20:35.940 --> 00:20:42.020 position:50% align:middle about middle-aged in that 41 to 64 age group and had no post-secondary education. 00:20:42.020 --> 00:20:46.780 position:50% align:middle So, we may be thinking, and this is, kind of, me just speculating, 00:20:46.780 --> 00:20:56.610 position:50% align:middle of these middle-aged White adults who are potentially in a rural area, may not have a ton of pain specialists 00:20:56.610 --> 00:21:02.150 position:50% align:middle who can provide them with all these resources to treat chronic pain in a holistic manner, 00:21:02.150 --> 00:21:04.330 position:50% align:middle may be working in manual labor fields. 00:21:04.330 --> 00:21:08.670 position:50% align:middle But the thing that I really want you to take away from this slide is that we are still seeing racial 00:21:08.670 --> 00:21:13.270 position:50% align:middle disparities in prescriptions of opioid medications, and this is across provider groups. 00:21:13.270 --> 00:21:19.740 position:50% align:middle So, in the physician literature for a number of years now, there's been documentation of racial disparities 00:21:19.740 --> 00:21:25.500 position:50% align:middle in who we're prescribing or who physicians are prescribing an opioid medication to. 00:21:25.500 --> 00:21:30.960 position:50% align:middle And the evidence has suggested that Black patients compared to White patients, Hispanic patients, 00:21:30.960 --> 00:21:36.180 position:50% align:middle and patients of a different race or ethnicity, are less likely to get an opioid medication, 00:21:36.180 --> 00:21:42.900 position:50% align:middle even if they're of the same clinical status and other demographics as White patients or patients of a 00:21:42.900 --> 00:21:44.900 position:50% align:middle different race or ethnicity. 00:21:44.900 --> 00:21:49.400 position:50% align:middle And so really to me what this says is that we're asking the wrong question. 00:21:49.400 --> 00:21:52.840 position:50% align:middle We're asking, are NPs overprescribing opioids? 00:21:52.840 --> 00:21:56.520 position:50% align:middle We should be asking, are NPs equitably prescribing opioids? 00:21:56.520 --> 00:22:04.160 position:50% align:middle Because remember, in our sample, we had nurse practitioners as the provider group that 00:22:04.160 --> 00:22:09.430 position:50% align:middle was more likely to see White patients along with PAs compared to physicians. 00:22:09.430 --> 00:22:15.260 position:50% align:middle And so this is really a problem that we need to be thinking about as we educate and train the 00:22:15.260 --> 00:22:22.980 position:50% align:middle next generation of primary care nurse practitioners is providing chronic pain care in an equitable manner and 00:22:22.980 --> 00:22:26.300 position:50% align:middle focusing less, again, on are we overprescribing. 00:22:26.300 --> 00:22:30.340 position:50% align:middle So, asking the right questions is really key for us going forward. 00:22:30.340 --> 00:22:34.270 position:50% align:middle And then looking at non-opioid prescriptions, we found pretty similar patterns. 00:22:34.270 --> 00:22:40.770 position:50% align:middle So, I thought it was interesting that not only did physicians have higher odds of prescribing an opioid, 00:22:40.770 --> 00:22:46.110 position:50% align:middle but physicians also had higher odds of prescribing a non-opioid medication for chronic pain. 00:22:46.110 --> 00:22:49.950 position:50% align:middle So, these are your anticonvulsants, your antidepressants, 00:22:49.950 --> 00:22:56.050 position:50% align:middle because there are certain antidepressants like Cymbalta that have pain management properties, and there's that, 00:22:56.050 --> 00:22:59.290 position:50% align:middle you know, link between mental health and chronic pain. 00:22:59.290 --> 00:23:04.720 position:50% align:middle Prescription NSAIDs, prescription acetaminophen, muscular relaxants, things along those lines, 00:23:04.720 --> 00:23:10.120 position:50% align:middle physicians actually had higher odds of prescribing these medications compared to NPs and PAs as well, 00:23:10.120 --> 00:23:14.375 position:50% align:middle and there was no significant difference between NPs and PAs. 00:23:14.375 --> 00:23:22.670 position:50% align:middle I will add the caveat that with the tens of thousands of patients that we had, these effect sizes, 00:23:22.670 --> 00:23:29.290 position:50% align:middle these statistical effect sizes may not really be clinically meaningful in the day-to-day, 00:23:29.290 --> 00:23:30.970 position:50% align:middle kind of, practice. 00:23:30.970 --> 00:23:36.130 position:50% align:middle And so this just may be us looking at something with a microscope, but nevertheless, it is interesting. 00:23:36.130 --> 00:23:40.850 position:50% align:middle And I'd be curious, and unfortunately, we weren't able to get this data in this study, 00:23:40.850 --> 00:23:48.459 position:50% align:middle if NPs or PAs were more likely to refer patients to physical therapy, or to a chiropractor, 00:23:48.459 --> 00:23:50.190 position:50% align:middle or an acupuncturist. 00:23:50.190 --> 00:23:57.180 position:50% align:middle And I know, from some of my separate qualitative work, that NPs who we interviewed to see how they're managing 00:23:57.180 --> 00:24:02.570 position:50% align:middle chronic pain patients, especially in rural areas, I would hear from NPs who would say, "Well, 00:24:02.570 --> 00:24:07.600 position:50% align:middle I'm the only healthcare provider in my county, and so we don't have a lot of resources to manage 00:24:07.600 --> 00:24:14.430 position:50% align:middle chronic pain, but you know what, the pool at the senior center has free swimming classes 00:24:14.430 --> 00:24:17.982 position:50% align:middle on Saturdays if you're over age 65, and that's a really good way to, kind of, 00:24:17.982 --> 00:24:23.220 position:50% align:middle get their mobility," or our physical therapist or our massage therapist works on a sliding scale. 00:24:23.220 --> 00:24:27.700 position:50% align:middle So, really, kind of, using the resources in their communities to effectively 00:24:27.700 --> 00:24:34.790 position:50% align:middle treat chronic pain in accordance with how we as healthcare professionals and our governing bodies 00:24:34.790 --> 00:24:39.560 position:50% align:middle like the National Academy and the CDC are saying we should be treating chronic pain, 00:24:39.560 --> 00:24:40.391 position:50% align:middle that's really what we're seeing. 00:24:40.391 --> 00:24:47.550 position:50% align:middle And I'd be interested to see if this will come up in a national database as well for a future study. 00:24:47.550 --> 00:24:52.490 position:50% align:middle And again, with our non-opioid prescriptions, we're seeing that Black patients compared to all of our 00:24:52.490 --> 00:24:59.789 position:50% align:middle other patients were at higher odds of receiving a non-opioid prescription compared to White patients, 00:24:59.789 --> 00:25:03.820 position:50% align:middle Hispanic patients, or patients of a different racial or ethnic background. 00:25:03.820 --> 00:25:10.870 position:50% align:middle So, again, we're not asking the right question of, are we treating chronic pain patients 00:25:10.870 --> 00:25:12.400 position:50% align:middle by overprescribing opioids? 00:25:12.400 --> 00:25:14.330 position:50% align:middle We really need to, kind of, look within ourselves and say, 00:25:14.330 --> 00:25:20.710 position:50% align:middle "How can we provide the most equitable pain care to all of our constituents possible?" 00:25:20.710 --> 00:25:24.970 position:50% align:middle So, just to, kind of, recap, patients who were cared for by physicians had higher 00:25:24.970 --> 00:25:29.670 position:50% align:middle odds of both an opioid and a non-opioid prescription compared to those of nurse practitioners or 00:25:29.670 --> 00:25:34.140 position:50% align:middle physician assistants, but these effect sizes might be pretty small 00:25:34.140 --> 00:25:35.250 position:50% align:middle in clinical practice. 00:25:35.250 --> 00:25:42.480 position:50% align:middle And so for a policymaker, this is evidence, again, that nurse practitioner care is equal or, 00:25:42.480 --> 00:25:47.658 position:50% align:middle in some cases, you're seeing higher satisfaction, or it may even be better than the care 00:25:47.658 --> 00:25:49.324 position:50% align:middle that's delivered by other providers. 00:25:49.324 --> 00:25:54.630 position:50% align:middle So, key evidence to, kind of, support the case for full practice reform. 00:25:54.630 --> 00:25:59.641 position:50% align:middle Patients who were non-Hispanic, White, middle-aged with no post-secondary education were some 00:25:59.641 --> 00:26:04.800 position:50% align:middle of the most likely patients to receive an opioid compared to patients who were non-Hispanic Black, 00:26:04.800 --> 00:26:09.810 position:50% align:middle and younger, and female were more likely to be prescribed a non-opioid medication. 00:26:09.810 --> 00:26:15.600 position:50% align:middle So, again, how are we taking steps to ensure that we're treating pain patients equitably? 00:26:15.600 --> 00:26:20.310 position:50% align:middle And dosing and length of prescription for opioid medications did not differ amongst provider groups. 00:26:20.310 --> 00:26:26.260 position:50% align:middle So, I think this is a great way to think about our theme from data that policy and, kind of, 00:26:26.260 --> 00:26:32.810 position:50% align:middle take this with us as we go forward and I'm happy to take any questions and engage in a dialogue for the 00:26:32.810 --> 00:26:34.035 position:50% align:middle next three minutes. 00:26:34.035 --> 00:26:35.056 position:50% align:middle I'm sorry, Monica. 00:26:35.056 --> 00:26:36.514 position:50% align:middle Thank you so much.