WEBVTT 00:00:00.360 --> 00:00:02.972 position:50% align:middle - [Woman] Ulrike Muench is an associate professor 00:00:02.972 --> 00:00:07.040 position:50% align:middle in the School of Nursing at the University of California, San Francisco, 00:00:07.100 --> 00:00:10.160 position:50% align:middle and a nurse and nurse practitioner by training. 00:00:10.160 --> 00:00:15.520 position:50% align:middle Her research examines the evolving roles and practice patterns of health care professionals, 00:00:15.520 --> 00:00:18.480 position:50% align:middle especially nurses and nurse practitioners. 00:00:18.480 --> 00:00:23.900 position:50% align:middle She uses big data and large surveys with interdisciplinary social science methods 00:00:23.900 --> 00:00:27.420 position:50% align:middle to inform local, state, and national policies. 00:00:27.420 --> 00:00:31.447 position:50% align:middle Her research has been published in leading health care journals, including JAMA, 00:00:31.447 --> 00:00:35.009 position:50% align:middle New England Journal of Medicine, and Health Affairs. 00:00:41.360 --> 00:00:42.910 position:50% align:middle - [Ulrike] Hello, everyone. 00:00:42.910 --> 00:00:47.972 position:50% align:middle My name is Ulrike Muench, and I'm an associate professor in the School of Nursing at UCSF, 00:00:47.972 --> 00:00:52.367 position:50% align:middle and affiliated faculty at the Philip R. Lee Institute for Health Policy Studies. 00:00:52.367 --> 00:00:58.170 position:50% align:middle Today, I'm sharing with you preliminary results from our study in which we examined opioid prescribing 00:00:58.170 --> 00:01:04.600 position:50% align:middle patterns of nurse practitioners and the association with state scope of practice regulations. 00:01:04.600 --> 00:01:08.590 position:50% align:middle I'm grateful for the support from the National Council of State Boards of Nursing, 00:01:08.590 --> 00:01:14.270 position:50% align:middle and I would also like to thank my collaborators on this research, Dr. Joanne Spetz, Dr. Jennifer Perloff, 00:01:14.270 --> 00:01:17.570 position:50% align:middle and Dr. Cindy Thomas, as well as Matthew Jura. 00:01:18.600 --> 00:01:22.810 position:50% align:middle To begin, I would like to give you some background on what motivated the study. 00:01:22.810 --> 00:01:28.231 position:50% align:middle My team and I began studying prescribing patterns of nurse practitioners several years ago through 00:01:28.231 --> 00:01:30.937 position:50% align:middle a grant from the National Council of State Boards of Nursing. 00:01:30.937 --> 00:01:36.000 position:50% align:middle While a large body of research existed on quality outcomes of NPs and MDs, 00:01:36.000 --> 00:01:39.730 position:50% align:middle prescribing patterns were rarely the focus of these studies. 00:01:39.730 --> 00:01:44.575 position:50% align:middle This is somewhat surprising since much of the scope of practice policy debates centers around 00:01:44.575 --> 00:01:49.400 position:50% align:middle prescriptive authority and controlled substance prescribing of nurse practitioners. 00:01:49.400 --> 00:01:54.920 position:50% align:middle In our early research on prescribing patterns, we compared in a national sample of Medicare patients 00:01:54.920 --> 00:02:00.752 position:50% align:middle the prescribing patterns of beneficiaries who receive care predominantly from NPs compared 00:02:00.752 --> 00:02:07.192 position:50% align:middle to beneficiaries who receive their care predominantly from MDs, and examined how the volume and types 00:02:07.192 --> 00:02:12.959 position:50% align:middle of prescriptions differed in these two groups of patients across the top 20 drug classes in 00:02:12.959 --> 00:02:18.750 position:50% align:middle primary care. We've found that both types of providers prescribe very similarly for these 00:02:18.750 --> 00:02:24.182 position:50% align:middle common drug classes. The table shows the number in share prescriptions as well as the average 00:02:24.182 --> 00:02:27.404 position:50% align:middle number of prescriptions per beneficiary. 00:02:27.500 --> 00:02:33.654 position:50% align:middle You can see that for antihypertensives, for example, both the NP and the primary care physician 00:02:33.654 --> 00:02:40.615 position:50% align:middle beneficiary groups received antihypertensives as the most common drug class, with a total share of 9.7% 00:02:40.615 --> 00:02:47.520 position:50% align:middle and 10.5%. For the number of prescriptions per beneficiaries, both groups received, on average, 00:02:47.520 --> 00:02:50.143 position:50% align:middle about 10.7 prescriptions. 00:02:51.370 --> 00:02:55.780 position:50% align:middle The pattern was very similar for the other drug classes as well. 00:02:55.780 --> 00:03:00.489 position:50% align:middle Following these analyses on prescribing patterns across the most common medication classes 00:03:00.489 --> 00:03:07.254 position:50% align:middle in primary care, we examined differences in adherence for three chronic medication classes, 00:03:07.254 --> 00:03:13.000 position:50% align:middle anti-diabetic medications, renin-angiotensin system antagonists, and statins. 00:03:13.000 --> 00:03:16.480 position:50% align:middle We've found no difference in adherence for two classes. 00:03:16.480 --> 00:03:24.060 position:50% align:middle For statins, however, 73.8% of NP patients and 74.8% of MD patients 00:03:24.060 --> 00:03:26.130 position:50% align:middle had good adherence. 00:03:26.130 --> 00:03:31.949 position:50% align:middle While this finding was statistically significant, it is likely not a clinically meaningful difference. 00:03:32.370 --> 00:03:37.091 position:50% align:middle Around the same time when we were working on these analyses, researchers and policymakers 00:03:37.091 --> 00:03:42.150 position:50% align:middle had begun to examine the role of clinicians in contributing to the opioid epidemic. 00:03:42.150 --> 00:03:47.710 position:50% align:middle This also brought renewed interest to the policy debate about prescriptive authority for NPs, 00:03:47.710 --> 00:03:53.832 position:50% align:middle specifically with regards to the prescribing of controlled substances without physician oversight. 00:03:54.350 --> 00:04:02.600 position:50% align:middle The opioid epidemic was at its peak between 2010 and 2012, and you can see that overdose deaths 00:04:02.600 --> 00:04:07.373 position:50% align:middle from prescription opioids have not significantly decreased since then. 00:04:07.760 --> 00:04:11.770 position:50% align:middle Overdose deaths involving any opioids remain high as well. 00:04:11.770 --> 00:04:16.282 position:50% align:middle There was much interest in understanding the role of clinician's opioid prescribing patterns 00:04:16.282 --> 00:04:18.890 position:50% align:middle in the opioid epidemic. 00:04:19.650 --> 00:04:29.872 position:50% align:middle Two studies published in 2015 and 2018 using data from 2012, and 2016, and 2017 drew attention to 00:04:29.872 --> 00:04:34.960 position:50% align:middle the fact that the majority of opioid prescriptions are written in primary care. 00:04:34.960 --> 00:04:41.250 position:50% align:middle The study on the left counted prescriptions from NPs and PAs together. And the study on the right, 00:04:41.250 --> 00:04:45.520 position:50% align:middle the later time period, showed that the share of opioid prescription from NPs 00:04:45.520 --> 00:04:50.743 position:50% align:middle and dentists had increased somewhat by 2016-2017. 00:04:51.935 --> 00:04:57.710 position:50% align:middle While studies began emerging on opioid prescribing outcomes of physicians for different specialties 00:04:57.710 --> 00:05:03.280 position:50% align:middle and conditions, evidence of NP opioid prescribing was still lacking. 00:05:03.280 --> 00:05:09.760 position:50% align:middle Two studies examined the associations of opioids with state scope of practice regulations. 00:05:09.760 --> 00:05:15.320 position:50% align:middle Schirle and McCabe found that states that required physician oversight in prescriptive authority had 00:05:15.320 --> 00:05:20.610 position:50% align:middle higher average number of opioids per 100 patients. And the study by Ladd and colleagues, 00:05:20.610 --> 00:05:25.520 position:50% align:middle which examined aggregated number of opioid prescriptions at the state level, found that 00:05:25.520 --> 00:05:29.600 position:50% align:middle for both NPs and MDs, the number of opioids were higher in states that did 00:05:29.600 --> 00:05:33.570 position:50% align:middle not require physician oversight in practice and prescribing. 00:05:34.990 --> 00:05:41.104 position:50% align:middle In 2019, my team and I published the first study examining opioid prescribing patterns of patients 00:05:41.104 --> 00:05:46.913 position:50% align:middle that were managed by MDs and NPs using Medicare data. 00:05:47.010 --> 00:05:53.240 position:50% align:middle We conducted the analysis only in the 14 states that had adopted full practice authority for NPs, 00:05:53.240 --> 00:05:58.624 position:50% align:middle including no physician oversight requirements for the prescribing of controlled substances. 00:05:58.670 --> 00:06:03.507 position:50% align:middle We conducted a longitudinal propensity score weighted adjusted analysis using 00:06:03.507 --> 00:06:06.136 position:50% align:middle data from two time periods. 00:06:06.136 --> 00:06:12.140 position:50% align:middle The outcomes that we measured were any opioid, morphine milligram equivalents of greater 100 00:06:12.140 --> 00:06:18.630 position:50% align:middle milligrams per day, a 7-day opioid prescription overlap of 2 opioids, a 7-day opioid and 00:06:18.630 --> 00:06:23.720 position:50% align:middle benzodiazepine overlap, as well as whether beneficiary was an acute user, 00:06:23.720 --> 00:06:29.386 position:50% align:middle which was defined as under 3 months of opioid use, or a chronic user, which was defined as greater 00:06:29.386 --> 00:06:31.930 position:50% align:middle 3 months of opioid use. 00:06:31.930 --> 00:06:37.500 position:50% align:middle Our results showed that NP-managed beneficiaries were less likely to receive any opioid, 00:06:37.500 --> 00:06:42.560 position:50% align:middle were more likely to receive a high dose of morphine milligram equivalent of greater 100 milligrams 00:06:42.560 --> 00:06:48.100 position:50% align:middle per day, were less likely to have an opioid and benzodiazepine prescription overlap, 00:06:48.100 --> 00:06:50.230 position:50% align:middle and were less likely to be acute users. 00:06:50.230 --> 00:06:54.276 position:50% align:middle Our other outcomes were not statistically significant. 00:06:54.980 --> 00:07:00.220 position:50% align:middle We interpret these findings to mean that although NP patients receive fewer opioids, 00:07:00.220 --> 00:07:04.980 position:50% align:middle when they do receive opioids and when patients need continued pain management, 00:07:04.980 --> 00:07:11.060 position:50% align:middle they are slightly more likely to receive too high of a dose compared to MD-managed beneficiaries. 00:07:12.160 --> 00:07:16.772 position:50% align:middle The analysis could not determine, however, the degree to which differences in prescribing 00:07:16.772 --> 00:07:22.083 position:50% align:middle patterns were due to provider differences or different pathways of care among patients 00:07:22.083 --> 00:07:24.226 position:50% align:middle with opioid prescriptions. 00:07:24.226 --> 00:07:29.303 position:50% align:middle For example, one explanation for our findings could be that NPs are more likely to work in pain 00:07:29.303 --> 00:07:33.448 position:50% align:middle management clinics or are more often receiving referrals of patients who have 00:07:33.448 --> 00:07:35.484 position:50% align:middle chronic pain conditions. 00:07:35.640 --> 00:07:40.450 position:50% align:middle We concluded that to better understand opioid prescribing patterns of NPs, 00:07:40.450 --> 00:07:47.750 position:50% align:middle it is important to focus on subpopulations of patients, for example, opioid-naive patients, 00:07:47.750 --> 00:07:52.432 position:50% align:middle to better understand how NPs, and MDs, and primary care are managing pain in a group 00:07:52.432 --> 00:07:58.800 position:50% align:middle patients who do not have a prior history of opioid use, or how NPs and MDs prescribe opioids 00:07:58.800 --> 00:08:01.623 position:50% align:middle for patients who do have a chronic pain history. 00:08:02.872 --> 00:08:06.240 position:50% align:middle This brings us to the design of our current study. 00:08:06.240 --> 00:08:12.340 position:50% align:middle We set out with the aim to focus on opioid initiations rather than all beneficiaries who received opioids. 00:08:12.340 --> 00:08:18.130 position:50% align:middle For this study, we were able to use data on the entire Medicare population. 00:08:18.130 --> 00:08:23.890 position:50% align:middle This meant that we could observe every NP and any other clinician who had their own National Provider 00:08:23.890 --> 00:08:28.198 position:50% align:middle Identification number and all their opioid prescriptions. 00:08:28.350 --> 00:08:36.247 position:50% align:middle We identified opioid-naive beneficiaries by scanning all beneficiary prescriptions in 2017 for an opioid, 00:08:36.247 --> 00:08:41.447 position:50% align:middle and if we did not find an opioid prescription for them, they were opioid-naive. 00:08:41.447 --> 00:08:46.744 position:50% align:middle We then look in 2018 for beneficiaries that were started on an opioid. 00:08:47.790 --> 00:08:55.550 position:50% align:middle We excluded patients who were under 65 years old, had a cancer diagnosis or end-stage renal disease, 00:08:55.550 --> 00:08:59.270 position:50% align:middle and if they had a hospice claim, since pain management typically looks very different 00:08:59.270 --> 00:09:03.778 position:50% align:middle for these diagnoses and settings compared to pain management in primary care. 00:09:04.730 --> 00:09:09.760 position:50% align:middle Our measures were guided by the CDC guidelines on the prescribing of opioids for chronic pain, 00:09:09.760 --> 00:09:12.440 position:50% align:middle which were released in 2016. 00:09:12.440 --> 00:09:19.053 position:50% align:middle The guidelines included 12 recommendations, of which number 4, 5, and 6 pertain to the dose 00:09:19.053 --> 00:09:20.670 position:50% align:middle and type of opioid. 00:09:20.670 --> 00:09:25.780 position:50% align:middle We use these recommendations to specify our outcome variable of interest. 00:09:25.780 --> 00:09:31.060 position:50% align:middle For example, the guidelines recommend starting therapy with a short-acting opioid, and we 00:09:31.060 --> 00:09:36.140 position:50% align:middle identified short and long-acting opioids in our data with the goal of measuring the number and share 00:09:36.140 --> 00:09:39.610 position:50% align:middle of patients that are started on these medications. 00:09:39.610 --> 00:09:46.870 position:50% align:middle Using the lowest effective dose and rarely increasing individuals to greater than 50 milligrams per day 00:09:46.870 --> 00:09:49.477 position:50% align:middle was another recommendation. 00:09:49.490 --> 00:09:56.370 position:50% align:middle For acute pain, the guidelines suggest that three-day supply will often be sufficient. And that more 00:09:56.370 --> 00:10:01.040 position:50% align:middle than seven-day supply will rarely be needed. 00:10:01.040 --> 00:10:06.700 position:50% align:middle Of note, in response to the guidelines, 15 states have passed laws limiting first-time opioid 00:10:06.700 --> 00:10:10.188 position:50% align:middle prescription to seven days or less. 00:10:12.130 --> 00:10:18.525 position:50% align:middle We began by calculating the number of opioid initiations in 2018 for individuals who were 00:10:18.525 --> 00:10:24.470 position:50% align:middle opioid-naive in 2017 by provider specialty. 00:10:24.470 --> 00:10:30.310 position:50% align:middle We saw a total of approximately 2 million opioid initiations in 2018. 00:10:30.310 --> 00:10:37.150 position:50% align:middle Of those, 17 were prescribed by providers in general surgery. Giving us an indication that these first 00:10:37.150 --> 00:10:45.130 position:50% align:middle opioid prescriptions are likely associated with a surgical event. 00:10:45.130 --> 00:10:49.180 position:50% align:middle After general surgery, we see the physician primary care specialties 00:10:49.180 --> 00:10:56.985 position:50% align:middle of internal medicine and family medicine, followed by PAs, emergency medicine, and then NPs. 00:10:57.960 --> 00:11:02.820 position:50% align:middle If we go back to one of our slides from earlier that showed the total number of opioid prescriptions 00:11:02.820 --> 00:11:10.000 position:50% align:middle in our studies, we can see that when we focus on the initiations only, NPs no longer represent the 00:11:10.000 --> 00:11:12.077 position:50% align:middle third-largest share. 00:11:14.140 --> 00:11:19.914 position:50% align:middle Next, we excluded individuals for whom we were able to identify that they had a surgery to help us 00:11:19.914 --> 00:11:24.810 position:50% align:middle better understand what is happening with initiations in primary care. 00:11:24.810 --> 00:11:30.230 position:50% align:middle And we see that the percent of initiations from general surgery providers drops significantly, 00:11:30.230 --> 00:11:34.502 position:50% align:middle with larger shares going now to primary care specialties. 00:11:34.970 --> 00:11:42.040 position:50% align:middle And when we focus on primary care alone, we see that both PAs and NPs contribute to initiations 00:11:42.040 --> 00:11:46.280 position:50% align:middle less than physicians, which in itself isn't surprising because there are 00:11:46.280 --> 00:11:50.015 position:50% align:middle fewer NPs and PAs than primary care physicians. 00:11:50.820 --> 00:11:56.390 position:50% align:middle Next, we take a deeper dive into the different specialties just within NPs, 00:11:56.390 --> 00:12:00.530 position:50% align:middle as well as begin looking at our prescribing measures of interests. 00:12:00.530 --> 00:12:07.650 position:50% align:middle We are able to identify these specialties by using the taxonomy code for the provider that is based on the 00:12:07.650 --> 00:12:12.020 position:50% align:middle National Provider Identification number or NPI. 00:12:12.020 --> 00:12:16.443 position:50% align:middle We can obtain the NPI from the prescriber characteristics file. 00:12:18.050 --> 00:12:25.100 position:50% align:middle The first two columns show us the distributions of opioid initiations among NP specialties. 00:12:25.100 --> 00:12:29.980 position:50% align:middle Similar to when we looked at all providers, the largest shares of initiations are occurring 00:12:29.980 --> 00:12:32.670 position:50% align:middle in primary care specialties. 00:12:32.670 --> 00:12:37.280 position:50% align:middle The third column shows the average day supply on the initiations. 00:12:37.280 --> 00:12:42.670 position:50% align:middle If you remember from the guidelines, initiations are recommended to be three days in length 00:12:42.670 --> 00:12:45.050 position:50% align:middle or rarely greater than seven days. 00:12:45.050 --> 00:12:51.935 position:50% align:middle The average day supply in our data is almost nine days in length, with some of the largest days' 00:12:51.935 --> 00:12:58.480 position:50% align:middle supply coming from psych/mental health NPs with 11.8 days' supply. 00:12:59.580 --> 00:13:05.260 position:50% align:middle The next three columns show the frequency column and row percentages for the share of first opioid 00:13:05.260 --> 00:13:08.370 position:50% align:middle prescriptions that were greater than seven-day supply. 00:13:08.370 --> 00:13:15.360 position:50% align:middle 52.7% are occurring in family, which is not surprising given that it is the 00:13:15.360 --> 00:13:17.791 position:50% align:middle largest NP specialty. 00:13:17.800 --> 00:13:22.140 position:50% align:middle The row percentage tells us the within specialty percentage. 00:13:22.140 --> 00:13:29.980 position:50% align:middle For example, for all opioid initiations within family NPs, 35% of the beneficiaries received the greater 00:13:29.980 --> 00:13:34.890 position:50% align:middle than seven-day prescription, compared to gerontology where almost half of all 00:13:34.890 --> 00:13:38.286 position:50% align:middle initiations were greater seven-day supply. 00:13:39.730 --> 00:13:46.060 position:50% align:middle Next up is our morphine milligram equivalency measure, which indicates the daily morphine dose. 00:13:46.060 --> 00:13:52.170 position:50% align:middle Generally speaking, the share of beneficiaries who received MME of greater 50 milligrams from NPs 00:13:52.170 --> 00:13:58.250 position:50% align:middle was small, on average, 8.2%, with the largest percentage of 13% occurring 00:13:58.250 --> 00:13:59.896 position:50% align:middle in acute care. 00:14:02.070 --> 00:14:06.388 position:50% align:middle Moving to our short and long-acting opioid measures, the good news is that 00:14:06.388 --> 00:14:13.700 position:50% align:middle very few initiations were with long-acting opioids, only 633 prescriptions in total. 00:14:13.700 --> 00:14:19.330 position:50% align:middle So conversely, the majority of initiations were with short-acting opioids. 00:14:19.330 --> 00:14:23.584 position:50% align:middle And this was the case consistently across all NP specialties. 00:14:24.660 --> 00:14:29.767 position:50% align:middle Next, we asked what picture emerges when we examine whether these prescribing patterns look 00:14:29.767 --> 00:14:35.254 position:50% align:middle differently in full practice authority states, versus states that do not allow NPs to practice and 00:14:35.254 --> 00:14:38.987 position:50% align:middle prescribe without physician oversight? 00:14:41.300 --> 00:14:44.430 position:50% align:middle Let's begin by looking at average day supply. 00:14:44.430 --> 00:14:50.930 position:50% align:middle The blue bars represent opioid initiations from full practice authority states, and the green bars 00:14:50.930 --> 00:14:53.930 position:50% align:middle from non-full practice authority states. 00:14:53.930 --> 00:15:01.960 position:50% align:middle The top two bars are overall differences for NPs. And after that, we break it out again by NP specialty. 00:15:01.960 --> 00:15:08.571 position:50% align:middle At the bottom, I plotted the average day supply for MDs so you can see that as a comparison. 00:15:08.700 --> 00:15:13.280 position:50% align:middle This graph illustrates that, overall, there is a very small difference of slightly longer 00:15:13.280 --> 00:15:20.990 position:50% align:middle days' supply in non-full practice authority states, and that this holds true for the larger specialty family 00:15:20.990 --> 00:15:24.210 position:50% align:middle as well as the majority of other specialties. 00:15:24.210 --> 00:15:29.840 position:50% align:middle We can also see that NPs, in general, prescribe slightly higher days' supply 00:15:29.840 --> 00:15:31.920 position:50% align:middle on the first prescription. 00:15:31.920 --> 00:15:36.320 position:50% align:middle It was about 8.9 days, regardless of full practice authority status, 00:15:36.320 --> 00:15:42.424 position:50% align:middle compared to MDs who have, on average, 8.5 days' supply when they initiate an opioid. 00:15:44.461 --> 00:15:49.450 position:50% align:middle When we take a look at the average day supply of prescriptions that were longer than seven days, 00:15:49.450 --> 00:15:52.800 position:50% align:middle this pattern becomes even more pronounced. 00:15:52.800 --> 00:15:59.220 position:50% align:middle The longer days' supply are occurring now in almost every specialty in no FPA states. 00:15:59.220 --> 00:16:06.798 position:50% align:middle And NPs are prescribing almost one additional day on these greater seven-day supply prescriptions. 00:16:08.800 --> 00:16:13.774 position:50% align:middle The next graph shows the shares of beneficiary by within specialty who had greater than 00:16:13.774 --> 00:16:15.219 position:50% align:middle seven-day supply. 00:16:15.219 --> 00:16:23.470 position:50% align:middle Overall, between 34% to 35% of all NP beneficiaries have a greater than seven-day supply. 00:16:23.470 --> 00:16:29.694 position:50% align:middle And the majority of specialties are seeing a slightly higher percentage in FPA states. 00:16:31.590 --> 00:16:37.556 position:50% align:middle We now take a look at the association between full practice authority and MME greater than 00:16:37.556 --> 00:16:40.056 position:50% align:middle 50 milligrams. 00:16:40.740 --> 00:16:49.420 position:50% align:middle And we see that the pattern is similar and that we see greater shares of initiations pretty consistently 00:16:49.420 --> 00:16:54.075 position:50% align:middle coming from full practice authority states, which are the blue bars. 00:16:54.740 --> 00:17:00.422 position:50% align:middle We are currently conducting several multivariable analyses to assess if the relationship with the 00:17:00.422 --> 00:17:07.551 position:50% align:middle primary care provider predicts an opioid initiation for beneficiaries who did not have a surgical event. 00:17:08.160 --> 00:17:12.727 position:50% align:middle These regressions control for beneficiary and county-level characteristics. 00:17:13.160 --> 00:17:18.288 position:50% align:middle Our preliminary analyses suggest that in beneficiaries who receive their care predominantly 00:17:18.288 --> 00:17:25.270 position:50% align:middle from an NP compared to a physician, that they're more likely to have an opioid initiation. 00:17:25.270 --> 00:17:31.470 position:50% align:middle The effect size is very small, with the odds being 1.027. 00:17:31.470 --> 00:17:39.808 position:50% align:middle This effect stays quantitatively the same when we look at the full practice authority states versus 00:17:39.808 --> 00:17:43.200 position:50% align:middle non-full practice authority states. 00:17:43.200 --> 00:17:50.739 position:50% align:middle NPs in both types of states are slightly more likely to initiate an opioid compared to MDs. 00:17:52.409 --> 00:17:58.400 position:50% align:middle We interpret our analyses to date to mean that, for several measures, including average day supply, 00:17:58.400 --> 00:18:03.845 position:50% align:middle the likelihood of experiencing a prescription of greater seven-day supply or a prescription of 00:18:04.640 --> 00:18:11.000 position:50% align:middle greater than 50 milligrams of MME are slightly less favorable for NPs compared to physicians. 00:18:11.000 --> 00:18:16.490 position:50% align:middle Descriptive results so far indicate a possible association with scope of practice. 00:18:16.490 --> 00:18:23.070 position:50% align:middle Counterintuitively, non-full practice authority states were more likely to see longer average day supply 00:18:23.070 --> 00:18:29.663 position:50% align:middle from NPs, while full practice authority states were more often observing initiations of MME 00:18:29.663 --> 00:18:32.537 position:50% align:middle of greater 50 milligrams per day. 00:18:32.810 --> 00:18:38.795 position:50% align:middle Our preliminary regression results show that increased likelihood for an NP-managed beneficiary 00:18:38.795 --> 00:18:44.040 position:50% align:middle to experience an opioid initiation. And we observed this effect in both full practice 00:18:44.040 --> 00:18:49.960 position:50% align:middle authority states and non-full practice authority states, suggesting that full practice authority status 00:18:49.960 --> 00:18:55.894 position:50% align:middle does not play a substantial role when we examine opioid prescribing patterns of NPs. 00:18:57.100 --> 00:19:02.860 position:50% align:middle While the association with the NP provider type was small, it will be important to better understand 00:19:02.860 --> 00:19:08.223 position:50% align:middle in future analyses what drive these prescribing patterns and what we can do from a policy 00:19:08.223 --> 00:19:14.899 position:50% align:middle and nursing education perspective to support NP clinicians but, of course, also other clinicians 00:19:14.899 --> 00:19:17.840 position:50% align:middle to make changes to their prescribing patterns. 00:19:17.840 --> 00:19:23.590 position:50% align:middle This is important also from a rural/urban perspective, where NPs are more likely to be the sole provider 00:19:23.590 --> 00:19:25.441 position:50% align:middle to their patients. 00:19:25.840 --> 00:19:31.710 position:50% align:middle Our next steps are to conduct careful robustness analyses to verify these initial results are stable 00:19:31.710 --> 00:19:34.880 position:50% align:middle across multiple regression specifications. 00:19:34.950 --> 00:19:37.090 position:50% align:middle Thank you very much for your interest. 00:19:57.120 --> 00:20:03.968 position:50% align:middle An additional component that we are currently examining in our analyses is to understand 00:20:03.968 --> 00:20:13.568 position:50% align:middle what diagnoses are occurring on these first initial prescriptions to tease out better what is sort of an 00:20:13.568 --> 00:20:18.763 position:50% align:middle acute diagnosis versus, like, a chronic pain diagnosis. 00:20:23.900 --> 00:20:27.600 position:50% align:middle Okay, I'm seeing a question now from Michelle Bach [SP]. 00:20:27.600 --> 00:20:29.580 position:50% align:middle That is a really great question. 00:20:29.580 --> 00:20:36.090 position:50% align:middle The question is about whether there are any data on opioid renewals by NPs. 00:20:36.090 --> 00:20:41.620 position:50% align:middle And not that I'm aware of, but it is something that we will also be looking 00:20:41.620 --> 00:20:44.540 position:50% align:middle at as part of our analyses going forward. 00:20:44.540 --> 00:20:50.840 position:50% align:middle We are very interested in understanding how that changes by type of provider. 00:20:55.090 --> 00:21:00.790 position:50% align:middle Lee Hubbard [SP] is asking a question about the number of years in practice evaluated when we 00:21:00.790 --> 00:21:02.510 position:50% align:middle did these analyses. 00:21:02.510 --> 00:21:06.250 position:50% align:middle No, we have not done that yet. 00:21:06.250 --> 00:21:18.754 position:50% align:middle However, there is some data that we could use to look into the length of time since someone obtained 00:21:18.754 --> 00:21:27.420 position:50% align:middle their DEA number, which would allow us to get a sense of for how long an NP has been in practice. 00:21:27.420 --> 00:21:31.825 position:50% align:middle And that is also something that we would like to do going forward. 00:21:39.290 --> 00:21:44.920 position:50% align:middle I'm seeing another question here from Laura, on behalf of Lucene Pargossiene [SP] 00:21:44.920 --> 00:21:52.740 position:50% align:middle and this is a really excellent question, to what extent the variation by practices and so not 00:21:52.740 --> 00:21:58.200 position:50% align:middle just states or scope of practice is affecting these prescribing patterns. 00:21:58.200 --> 00:22:04.910 position:50% align:middle And this is also something that we are very keen to look into and that we have on our to-do list 00:22:04.910 --> 00:22:11.711 position:50% align:middle going forward, to sort of get a sense of whether it is the prescribing so with your colleagues and there 00:22:11.711 --> 00:22:18.910 position:50% align:middle are specific patterns that might happen based on who you are working with in the office. 00:22:18.910 --> 00:22:29.800 position:50% align:middle And that might have a big effect on how you're prescribing rather than our state scope 00:22:29.800 --> 00:22:33.757 position:50% align:middle of practice variations, for example. 00:22:36.150 --> 00:22:39.412 position:50% align:middle These are really excellent questions. 00:22:40.960 --> 00:22:43.650 position:50% align:middle Well, I'm not seeing any additional questions coming in. 00:22:43.650 --> 00:22:46.540 position:50% align:middle Thank you again for your interest. 00:22:46.540 --> 00:22:49.770 position:50% align:middle And have a good rest of your symposium. 00:22:49.770 --> 00:22:50.941 position:50% align:middle Thank you.