Addiction Medicine, Article

System-wide process changes improve procedural sedation billing in the pediatric emergency department

71 patients were identified as candidates. 17 (24%) denied a his- tory of opioid use, 11 (15%) declined, and the remaining 43 were ap- propriate for inclusion. Of these, the treating physician refused to prescribe naloxone for 16 (37%), 2 (5%) already possessed naloxone, and 1 (2%) clinically deteriorated and was no longer able to partici- pate. The remaining 24 (56%) were enrolled (Fig. 1).

At the 3 month follow-up, 7 (29%) participants were successfully contacted, of which 2 (29%) had chosen to fill their prescription. None reported obstacles to obtaining naloxone.


Our sample size was small and larger studies would be necessary to generalize our results. The harm reductionist team had hospital privi- leges by virtue of being medical students and did not require funding, which may not be true elsewhere. Therefore, other programs may re- quire further steps to incorporate harm reductionists into the ED. Larger studies may also be able to demonstrate the effectiveness of specific ed- ucational tools in the emergency setting and their translation to real- world overdose response.


The greatest barrier to naloxone prescription in our study was physician resistance, despite the program being approved by the de- partment chair and research director. This underscores the need to im- prove physician education about the efficacy of harm reduction. Our study implies that commonly stated objections (such as time con- straints and inadequate staffing) may be only part of the cause for phy- sician opposition to harm reduction interventions.

This study demonstrates that collaborations between an ED and community harm reductionists without formal credentials can result in delivery of meaningful overdose prevention education and naloxone to patients without compromising their care or ED throughput. As many departments lack the staff, funding, resources, and knowledge to de- velop a comprehensive harm reduction program, our approach offers a potential alternative. We encourage harm reduction programs with sufficient resources to consider reaching out to local hospitals to expand the scope of harm reduction services.

Although the majority of eligible patients received our educational intervention and a naloxone prescription, few filled that prescription. Due to internal pharmacy policies, we had been unable to provide take-home naloxone directly at discharge, and we believe this led to the low portion of participants ultimately obtaining naloxone. We rec- ommend that future similar programs provide naloxone directly to par- ticipants, and attempt to understand factors causing emergency physician hesitance to prescribe naloxone.

Funding and conflicts of interest


Kyle Barbour, MD candidate?,1 Miriam McQuade, MD candidate1 University of California, Irvine, Department of Emergency Medicine, 101 the City Drive, Route 128-01, Orange, CA 92868, United States

Orange County Needle Exchange Program (OCNEP), c/o the LGBT Center OC, 1605 N Spurgeon St, Santa Ana, CA 92701, United States

?Corresponding author at: 101 The City Drive, Route 128-01, Department of Emergency Medicine, Orange, CA 92868, United States. E-mail addresses: [email protected] (K. Barbour),

[email protected] (M. McQuade).

1 Both authors contributed equally to this work.

Shashank Somasundaram, BS Bharath Chakravarthy, MD, MPH University of California, Irvine, Department of Emergency Medicine, 101 the City Drive, Route 128-01, Orange, CA 92868, United States

E-mail addresses: [email protected] (S. Somasundaram),

[email protected] (B. Chakravarthy).

14 March 2018


  1. Chen JH, Humphreys K, Shah NH, Lembke A. Distribution of opioids by different types of Medicare prescribers. JAMA Intern Med 2016;176:259-61. 10.1001/jamainternmed.2015.6662.
  2. Rudd RA, Seth P, David F, et al. Increases in drug and opioid-involved overdose deaths — United States, 2010-2015. MMWR Morb Mortal Wkly Rep 2016;65: 1445-52.
  3. Barnett ML, Olenski AR, Jena AB. Opioid-prescribing patterns of emergency physi- cians and risk of long-term use. N Engl J Med 2017;376:663-73. 1056/NEJMsa1610524.
  4. Butler MM, Ancona RM, Beauchamp GA, et al. Emergency department Prescription opioids as an initial exposure preceding addiction. Ann Emerg Med 2016;68: 202-8.
  5. Cantrill SV, Brown MD, Carlisle RJ, et al. Clinical policy: critical issues in the prescrib- ing of opioids for adult patients in the emergency department. Ann Emerg Med 2012;60:499-525.
  6. McDonald R, Strang J. Are take-home naloxone programmes effective? Systematic review utilizing application of the Bradford Hill criteria. Addiction 2016;111: 1177-87.
  7. Masson CL, Sorensen JL, Perlman DC, et al. Hospital- versus community-based sy- ringe exchange: a randomized controlled trial. AIDS Educ Prev 2007;19:97-110.
  8. D’Onofrio G, O’Connor PG, Pantalon MV, et al. Emergency department-initiated buprenorphine/naloxone treatment for Opioid dependence: a randomized clinical trial. JAMA 2015;313:1636-44.
  9. Dwyer K, Walley AY, Langlois BK, et al. Opioid education and nasal naloxone rescue kits in the emergency department. West J Emerg Med 2015;16:381-4. https://doi. org/10.5811/westjem.2015.2.2490.
  10. Samuels EA, Dwyer K, Mello MJ, et al. Emergency department-based opioid harm re- duction: moving physicians from willing to doing. Acad Emerg Med 2016;23: 455-65.

    System-wide process changes improve procedural sedation billing in the pediatric emergency department


    Procedural sedation and analgesia is common in the emergency de- partment (ED). Data from the Nationwide Emergency Department Sam- ple found that at least 80 children are sedated in the ED every day across the United States, accounting for 0.1-1.5 sedation cases per 1000 visits [1,2]. laceration repair, fracture reduction, and abscess incision and drainage are the top conditions requiring PSA in the ED [3]. PSA has been shown to be safe and effective in the hands of trained multidisci- plinary practitioners in the ED [3,4].

    PSA is complex, requiring Close monitoring and the presence of a minimum of two providers [5-8]. Often the bedside nurse is responsible for monitoring physiologic parameters and assisting in any supportive or resuscitation measures. The PSA provider is responsible for the ad- ministration of medications and sedation oversight, while sometimes performing the procedure as well. The PSA provider must have sedation knowledge, ability to provide rescue techniques, apply monitoring de- scribed in guidelines such as those by the American Academy of Pediat- rics, and manage complications for a level deeper than the intended sedation state [8].

    PSA in the ED is resource intensive and disrupts regular patient flow. It requires parental preparation through informed consent, equipment set up, and documentation [8]. Thus, it is important to recuperate the costs through appropriate billing for both the sedation and procedure. We evaluated our sedation billing to identify the key billing gaps and underwent a quasi-experimental, quality improvement (QI) process to rectify these gaps and improve physician services charge capture.

    Material and methods

    We conducted a cross-sectional study of patients that received keta- mine as a primary agent for PSA in the EDs of two urban, academic, free- standing children’s hospitals in our hospital system. Data were divided into two groups: the pre-intervention group from July 1st 2014 through June, 30th, 2015; and the post-intervention group from August 1st, 2015 through July 31st, 2016. Data was not collected during the imple- mentation period in July 2015.

    Setting and population

    The study was conducted at our main campus (MC), a quaternary urban, level 1 trauma center with an average of 75,000 ED visits annu- ally, and our community campus (CC), a suburban community hospital with an average of 43,000 ED annual visits. These hospitals represent one of the largest free-standing children’s hospital networks in the country.

    Ketamine was the most widely used ED PSA agent administered by moderate sedation credentialed faculty. Our internal pharmacy data- base was queried to identify all patients that received ketamine for PSA. All patients that received ketamine outside the PSA indication were excluded: intubation in combination with a Neuromuscular blocker, bronchodilation in asthmatic patients, or for non-procedural analgesia. Ketamine usage was cross-referenced to physician billing.


    We evaluated patients that received PSA over the initial year to de- lineate key sources of missed billing to improve charge capture. A mul- tidisciplinary team of ED physicians, nurses and ancillary services identified several obstacles including documentation variation with an inconsistent reporting location for sedation start and stop time due to lack of provider template standardization; lack of physician sedation billing and documentation knowledge; and lack of electronic medical record diagnosis and procedure code linking.

    Subsequently, a prospective QI project was designed to measure the impact of our interventions to improve documentation and charge cap- ture for our 100 practitioners and 190 nurses in the ED. Demographic data were reviewed to include age, gender, procedure type, medication dosing, frequency of administration, provider type, financial informa- tion, and insurance status. Active moderate sedation billing codes at the time of the study included current procedural terminology (CPT) 99143-99150 [9].

    We developed a targeted bundled intervention aimed at the 3 main components attributed to cases of missed billing. The first barrier was

    Additionally, our business process reviewer (BPR) systematically reviewed charts to verify if documentation was lacking. Our BPR reviewed a daily query from T-systems (4020 McEwen Road Dallas, Texas 75244), for any delinquent documentation in all provider en- counters from the prior 24 h. PSA documentation accounted for approx- imately 15% of the 80-85 daily encounters. Our BPR reviewed PSA documentation for associated diagnosis linked to the procedure, calcu- lated sedation length, and whether a separate sedation provider was present in addition to the proceduralist. The BPR would then audit this database two to five times a week, monitoring for inadequately billed PSA due to insufficient documentation. The providers were notified within seven days of the date of service and allowed addending the chart to ensure complete documentation.

    The third component was the development of an educational initia- tive. Provider education was delivered in a multi-media manner via email, in-person lectures, and reminders on ED rounds. PowerPoint lec- tures were presented to all providers and fellows at the onset of the study, and several email blasts were sent with educational attachments in the first several weeks of July 2015. July 2015 was designated as the implementation of our 3-component bundle.

    Data collection and processing

    Our pharmacy database was queried and matched to PSA billing dur- ing the study time period. Age was positively skewed right (Shapiro- Wilk p-value b 0.001), therefore the Mann-Whitney test was utilized. Categorical comparisons were made using the Pearson Chi-Square test. A segmented Linear regression analysis was chosen for this type of data with segment parameters defined as level change after interven- tion (?2) and trend change after intervention (?3). The primary out- come was the monthly sum of PSA charges. Secondary outcome measures included coding variation, stratified by site. Statistical signifi- cance was defined as a p-value b0.05. Analyses were conducted using Statistical Package for the Social Science (SPSS), version 24 (IBM Corp., Armonk, NY) and STATA, version 13 (StataCorp LP, College Station, TX).


    A total of 1602 patients received charges for PSA with ketamine out of the 2941 PSA procedures in the ED during the study period. The aver- age age of the children undergoing sedation during the study period was 5.34 years (2.57-8.92). The pre-intervention timeframe was 07/ 01/2014 through 06/30/2015 (n = 353; 22.0%) and the post-interven- tion timeframe was 08/01/2015 through 07/31/2016 (n = 1249; 78.0%). There were no statistically significant differences in the demo- graphics of the population over the study timeframe (Table 1). There were no significant differences in the demographics between eligible patients and those excluded during the implementation period July 2015.

    Table 1

    Demographic comparison between pre and post intervention timeframes (N = 1602).

    lack of documentation standardization. The most commonly utilized

    electronic medical record (EMR) template in use did not contain a loca- tion for PSA start and stop time. This made subsequent chart review dif- ficult to locate sedation time, which is a mandatory component for the

    Pre-intervention 7/2014-6/2015 N = 353 (22.0%)

    Median (IQR) or N (%)

    Post-intervention 8/2015-7/2016

    N = 1249 (78.0%)

    Median (IQR) or N (%)


    billing codes. We modified our EMR procedure note to clearly define the PSA time.

    Second, we identified the need for timely administrative billing feed- back to assist the busy clinical providers in adequate charge capture. Fo- cused education was created to focus on the rapid identification of insufficient documentation with subsequent provider communication via email to improve correction in a timely basis. We worked with our physician services organization to assist in auditing these PSA charts.

    Age (years) 5.34 (2.57, 8.92) 5.76 (2.59, 9.52) 0.34

    Sex (female) 167 (47.3) 500 (40.0) 0.01

    Location 0.97

    Main campus 241 (68.3) 854 (68.4)

    Community campus 112 (31.7) 395 (31.6)

    Insurance status 0.20


    146 (41.4)

    572 (45.8)


    165 (46.7)

    560 (44.8)


    42 (11.9)

    117 (9.4)

    Before the intervention, there was a significant difference from month to month in sum charges (p-value b 0.001) (Fig. 1). Over time, monthly sum charges increased by $1388.81 (95% CI: 952.57-

    Table 2 Current procedural terminology (CPT) coding and frequency of ketamine charges between pre and post study timeframes (N = 1602).

    1825.06). After the intervention, average ketamine charge summation increased by $1210.02 per month. There was no significant change in the month-to-month trend after the intervention (p-value = 0.15). There were PSA charges for 329 patients in the pre-intervention com- pared to 1132 post-intervention; nearly a four-fold increase in patients being billed (Table 2). The PSA billing for the pre-implementation period had a significantly lower monthly charge mean of $6054.55

    (SD +- $5723.00) compared to a monthly charge mean of $20,600.00

    Study period CPT type No charge N (%)

    a Pearson or Fischer chi-square.

    Pre 7/14-6/15

    Same providerb

    21 (87.5)

    320 (97.3)


    Different providerc

    3 (12.5)

    9 (2.7)

    Post 8/15-7/16

    Same providerb

    84 (71.8)

    1085 (95.8)


    Different providerc

    33 (28.2)

    47 (4.2)

    b Same provider performing sedation and procedure.

    Charge N (%)


    (SD +- $4355.56) in the post period (p-value b 0.001). We also de- creased our percentage of “No Charge” by N15% from pre- to post- intervention.

    Secondary outcomes were a subgroup analysis for CPT coding. In- stead of using monthly dollar charges, each individual amount was coded as a charge (yes/no). A $0.00 was coded as not being charged, and any dollar amount N$0.00 was coded as a charge. CPT codes were collapsed into two provider categories: same physician providing both the sedation and the procedure or different physician providing the se- dation than the procedure. There were significant differences within each timeframe (pre/post) between physician type and charge (y/n) (p-value = 0.04 and p-value b 0.001, respectively) (Table 2). When a different physician provided the sedation while another performed the procedure, this code had a higher percentage of no charges than when the same physician provided both the sedation and the proce- dure. It is likely that we billed for the procedure charge only and not the sedation. When considering the effect of pre/post intervention, there was no significant difference in frequency of charges whether the provider did both the sedation and procedure or only the sedation (Breslow-Day p-value = 0.44).


    Ketamine is the most widely used dissociative, Sedative agent for children undergoing PSA in the ED [2,3]. Our bundled intervention targeted primarily ketamine moderate PSA in our ED. We reviewed pre- vious billing data to identify the most significant gaps that could im- prove our ED PSA billing. We found that our bundled intervention led to increased monthly charges for PSA, and improved charges for CPT coding.

    There was a delta of more than $4600, from pre to post-intervention for PSA billing per month. Our initial billing was – $3460.61 per monthly

    Month Frequency







    Sum of Monthly Ketamine Charges (dollars)

    0 5000 10000 15000 20000 25000

    Fig. 1. Composite monthly sedation physician charges. Month 0 is July 2014 through Month 25 July 2016. The red line indicates the July 2015 implementation period.

    c Different provider performing sedation than the procedure.

    ketamine charge. After our targeted bundled intervention, we saw monthly charges of $1210.02. We have not achieved 100% billing for PSA, however this is a significant change in the right direction. Our sec- ondary outcomes for CPT coding showed improvement in number of billed sedations, nearly four-fold from 329 to 1132. The total number of sedations hasn’t changed significantly per year, and thus our effective billing capture is substantial.

    We identified 4 major components to optimize billing: the creation of user-friendly EMR templates, monitoring provider compliance with the developed templates, physician billing education, and timely finan- cial documentation feedback by administrative support. A tiered system involving the partnership of providers and the finance team allows for improved financial outcomes in a large quaternary medical system. The ED practice setting is quite demanding, and it is unrealistic for pro- viders to bill efficiently and with 100% accuracy during their clinical shifts.

    It is essential for providers to become more cognizant and involved in billing matters in medicine, especially the ED setting. Developing pro- cesses and clinical pathways that maintain high-level clinical care while providing fiscal responsibility are the hallmark of modern medicine. This is a rapidly growing area of medical research that will involve sys- temic process changes. While we were able to improve the EMR, oppor- tunity still exists as even when highlighting a mandatory start and stop time within the uniform procedure note we were unable to create a pro- cess where the time in minutes for the sedation procedure auto-calcu- lates. Each EMR-build is hospital specific and must be tailored to its needs.

    There are several limitations to our study. We focused on a single hospital system and analyzed only our ketamine PSA billing. We were unable to conduct step-wise QI PDSA cycles over time to evaluate which aspect of our bundle resulted in the greatest improvement due to the need for a rapid intervention to improve charges. There is an increase in PSA charges beginning in March 2015, after a gap in billing was noted. This increase was likely due to preliminary discussions with leadership to increase awareness of the charge gap. Further work should be done to develop a multi-center protocol that can be implemented in various clinical settings to improve PSA billing. By standardizing charge capture for both high-volume quaternary centers as well as community hospitals that utilize a variety of sedation medications, pediatric PSA can consistently capture more of the Financial benefits to justify this resource intensive procedure.


    The implementation of a PSA bundle in our ED improved monthly charge capture by a delta of more than $4600 from pre to post-interven- tion. Our intervention bundle demonstrates that significant billing charge improvement can be obtained and sustained, with systemic pro- cess changes, provider education, uniform documentation templates, and implementation in the busy ED setting.

    Appendix A. Full segmented regression model for monthly summa- tions of Ketamine charges and the study intervention (N = 24)



    95% CI


    Intercept (?0)


    -6671.28 to -249.93


    Months (?1)




    Intervention (?2)




    Time after intervention (?3)




    Alana J. Arnold, MD, MBA? Paul E. Sirbaugh, DO, MBA Department of Pediatrics, Section of Emergency Medicine, Baylor College of

    Medicine, Houston, TX, USA

    ?Corresponding author.

    E-mail address: [email protected] (A.J. Arnold).

    Brady S. Moffett, Pharm D

    Department of Pharmacy, Texas Children’s Hospital, Houston, TX, USA

    Elizabeth A. Camp, PhD

    Department of Pediatrics, Section of Emergency Medicine, Baylor College of

    Medicine, Houston, TX, USA

    Sohail Azeem, MBA, MPH

    Clinical Business Operations, Texas Children’s Hospital, Houston, TX, USA

    LaTesha R. Smith, BS

    Clinical Care Center, Texas Children’s Hospital, Houston, TX, USA

    Corrie E. Chumpitazi, MD, MS

    Department of Pediatrics, Section of Emergency Medicine, Baylor College of

    Medicine, Houston, TX, USA

    Fentanyl-associated illness among substance users — Fulton County, Georgia, 2015

    In early 2015, Hospital A emergency physicians subjectively noticed an increase in opioid overdoses presenting to the emergency depart- ment (ED) that corresponded with an increase in fentanyl-positive substance-related deaths documented by the Fulton County medical ex- aminer (ME). This prompted Hospital A emergency physicians to begin selective fentanyl urine drug screening (UDS) for patients with clinical signs of opioid intoxication. After testing revealed that some patients had UDS positive for fentanyl, Hospital A began testing for fentanyl as part of all routine UDS in May 2015 and notified the Georgia Depart- ment of Public Health (DPH) of their findings. Fentanyl had not been commonly reported as associated with substance abuse and overdose in Georgia before this cluster. DPH initiated an epidemiologic investiga- tion to characterize events and guide Prevention efforts.

    We performed a case-control study involving ED patients who pre- sented during May 5-July 31, 2015, with acute, unintentional, substance-related illness or injury. A case was defined as UDS positive for fentanyl in addition to other opiates, cocaine, or amphetamine, in a patient who had no current or recent fentanyl prescription in evidence at the time of presentation. We also examined Fulton County ME data concerning substance-related deaths during August 1, 2013-July 31, 2015. UDS was performed on 3137 patients presenting to the ED during the study period; 254 (8%) were fentanyl positive, of which 79 were from patients who met all criteria to be classified as case-patients. Ninety-one Control subjects were also included in the study. Table 1 summarizes characteristics of case-patients and control subjects. Case- patients were more likely than control subjects to have a presenting chief complaint of drug overdose (OR: 4.5; 95% CI: 1.6-12.0) or al- tered mental status (OR: 3.8; 95% CI: 1.2-11.5), as opposed to a

    Table 1

    Characteristics of case-patients and control subjects presenting to Hospital A emergency department in Atlanta, Georgia during May 5-July31, 2015 (N = 170).

    15 March 2018

    Case-Patients (n = 79)

    Control Subjects (n = 91)

    No. % No. % Odds Ratio

    95% CI


    [1] Allareddy V, Rampa S, Auslendar M, Lowrie L. Trends in the Utilization of Moderate Sedation in Children in Emergency Departments in the USA. Oral Abstract Society

    for pediatric sedation; 2017 (Annual Meeting).








    [2] Sacchetti A, Stander E, Ferguson N, Maniar G, Valko P. Pediatric Procedural Sedation in








    the Community Emergency Department: results from the ProSCED registry. Pediatr Emerg Care Apr 2007;23(4):218-22.


    Race (N = 148)a







    [3] Bhatt M, Johnson DW, Chan J, Taljaard M, Barrowman N, Farion KJ, et al. Risk factors








    for adverse events in emergency department procedural sedation for children. Seda-








    tion Safety Study Group of Pediatric emergency research Canada (PERC). JAMA

    Presenting chief complaint

    Age range (median) 18-62


    Age group (yrs)

    b30 30-39















    Pediatr Oct 1 2017;171(10):957-64.

    [4] Grunwell JR, Travers, CE McCracken, Scherrer P, Stormorken A, Chumpitazi CE, et al. Procedural sedation outside the operating room using ketamine in 22,645 children: a report from the pediatric sedation research consortium. Pediatr Crit Care Med Dec 2016;17(12):1109-16.

    [5] Green SM, Roback MG, Kennedy RM, Krauss B. Clinical practice guideline for emer- gency department ketamine dissociative sedation: 2011 update. Ann Emerg Med May 2011;57(5):449-61.

    [6] Godwin SA, Burton JH, Gerardo CJ, Hatten BW, Mace SE, Silvers SM, et al. Clinical pol- Received naloxone 26






    icy: procedural sedation and analgesia in the emergency department. American Col-

    Arrived by EMS







    lege of Emergency Physicians. Ann Emerg Med Feb 2014;63(2):47-58.

    [7] Gross Jeffrey, et al. Practice guidelines for sedation and analgesia by non-anesthesiol-

    Admitted (e.g., medical, ICU, or psychiatric)







    Drug overdose







    Altered mental status







    Psychological concern/suicidal







    Addiction problem














    Transport, treatment, dispositionb

    ogists. American Society of Anesthesiologists Task Force on Sedation and Analgesia by Non-Anesthesiologists. Anesthesiology Apr 2002;96(4):1004-17.

    [8] Cote CJ, Wilson S, American Academy of Pediatrics, American Academy of Pediatric Dentistry. Guidelines for monitoring and management of pediatric patients before, during, and after sedation for Diagnostic and therapeutic procedures: update 2016. Pediatrics 2016;138:1.

    [9] CPT coding. AMA Website. current-procedural-terminology. Copyright 1995-2018. Accessed October 2017.

    CI, confidence interval.

    % total N 100 because of rounding.

    a Four control subjects were of other or unknown race, and 18 case-patients were of other or unknown race; calculations are for patients of white or Black race only.

    b EMS, emergency medical services; ICU, intensive care unit. Odds ratio for each vari- able is the ratio of the odds of case-patients vs control subjects who had the experience vs. odds of case-patients vs control subjects who did not have the experience.

Leave a Reply

Your email address will not be published. Required fields are marked *