Tag: Ameer Khalek

 

The management of cardiac arrest is standardized and based on consensus guidelines for all provider levels [1]. In the United States, the most commonly accepted guidelines are those of the American Heart Association. Despite efforts to disseminate these guidelines and educate the public on emergency cardiovascular care, the national rate of survival to hospital discharge for EMS-treated atraumatic cardiac arrest is only 11% [2]. There is significant room for improvement, therefore, it is important to consider innovations in this area.

I sat on the QA/QI committee for my collegiate EMS agency from 2013-2015 and had the opportunity to work with a medical director that encouraged provider involvement in quality improvement [3]. In 2014, we developed and implemented a Clinical Operating Guideline for the use of Impedance Threshold Devices (ITDs), a Class IIa (Benefit >> Risk) recommendation per the 2010 AHA guidelines. In 2015, the AHA updated their guidelines and changed the use of ITDs from a Class IIa (Benefit >> Risk) to Class III (Benefit = Risk) recommendation. By 2017, ITDs were phased out of service. What changed?

Impedance Threshold Devices

The main purpose of CPR is to restore flow of oxygenated blood to the brain and heart. There are two key theories that offer a basis for how this may occur during chest compressions [4]. The “cardiac pump” theory postulates that blood flows because the heart is squeezed between the sternum and the spine. The “thoracic pump” theory postulates that compression of the chest wall causes intrathoracic pressure to exceed extrathoracic vascular pressure and subsequent ejection of blood from the heart into systemic circulation and expiration of air from the lungs. However, the thoracic cavity pressure gradient is attenuated by passive inspiration of air with decompression (i.e., chest recoil) in cardiac arrest.

The ITD is designed to limit this.

Figure 1: Mapping Intrathoracic Pressure [5]

The ITD is a disposable one-way valve that attaches to an airway circuit and regulates intrathoracic pressures to maintain negative intrathoracic pressure during chest compressions by preventing air from entering the lungs between chest compressions, enhancing negative intrathoracic pressure. Often times ITDs will come with a flashing light to concurrently guide ventilation rate.

What Changed?

According to the AHA’s Summary of Key Issues and Major Changes, the ITD’s drop from Class IIa to III recommendation was driven by a single large multicenter randomized clinical trial known as the Resuscitation Outcomes Consortium (ROC) Prehospital Resuscitation using an Impedance Valve and Early vs. Delayed Analysis (PRIMED) trial [6]. The ROC PRIMED included ten sites across the United States and Canada (n = 8,718), and failed to demonstrate any improvement associated with the use of an ITD (compared with a sham device) as an adjunct to conventional CPR. After two years of preliminary data suggesting equivalence of risk and benefit, the National Institutes of Health (NIH) stopped enrollment in the trial [7]. The often erroneously interpreted conclusions of this trial echo throughout the annals of ancillary devices for cardiopulmonary resuscitation and, without much thought, agencies have systematically been taking them off their units. While the ROC PRIMED trial is an excellent resource for driving future academic work, the findings are not actionable and should not inform prehospital practice.

Anyone who has followed CPR research over the years knows that enormous resources are allocated to perform these trials and there are often more questions about study design than conclusions about the interventions being tested [8]. Olasveengen et al. titles this the “why-all-cardiac-arrest-trials-are-neutral”-puzzle, and highlights a few reasons why [9]. The presence of subgroups within a study with significant differences masks any potential positive and/or negative effects within an entire study population. In the ROC PRIMED trial, baseline neurologically intact survival rates ranged from 1.1% in Alabama to 8.1% in Seattle [10]. Is it fair to assume that an ITD will have equal effect in both Alabama and Seattle? Among the ten different sites, there was also variation in BLS CPR protocols being used (i.e., CPR with face mask, 30:2 compression-to-ventilation ratio). Additionally, data collection spanned benchmark outcomes such as ROSC on arrival at ED, survival to hospital admission, and survival to discharge. Researchers did not control for how closely tied these outcomes are to the quality of post-resuscitation care. More than half of included patients received substandard CPR. In a post hoc analysis of their previously published work, researchers stratified patients with high quality CPR and lower the risk of a poor outcome and found a statistically significant benefit from the ITD [11]. With variability in CPR quality, there was no surprise to find variable effect of the ITD.   

Conclusion

The quality of CPR delivered to a patient in cardiac arrest can be a significant effect modifier in clinical trials. Future work in this area should be guided by the notion that the multi-faceted nature of cardiac arrest management is prone to heterogeneity and should be controlled for when designing studies. Additionally, interventions are not mutually exclusive. Research should be refocused on building a systems-based approach to care that considers a combination of therapies.

“Why do we continue to look for a single silver bullet to treat cardiac arrest when we know that every complex disease conquered by modern medicine was done so with a multipronged approach? Isn’t it time to admit that no single intervention […] will significantly improve survival from cardiac arrest, and only a combination of therapies, implemented in a systems-based approach, is the answer?” - Dr. Keith Wesley, Minnesota’s EMS medical director [12]

Future Directions

In Resuscitation, Moore et al. (2017) compared brain blood flow between the Head Up (HUP) and Supine (SUP) body positions during a prolonged CPR effort of 15 min, using ACD CPR and ITD in a porcine model of cardiac arrest and found brain blood flow to be 2-fold higher in the HUP position. These positive findings provide strong pre-clinical support to proceed with a clinical evaluation of ACD CPR + ITD in humans in cardiac arrest [13].

In The American Journal of Emergency Medicine, Niforopoulou et al. (2017) assessed whether the use of an ITD during CPR reduces the degree of post cardiac arrest Acute Kidney Injury (AKI) in a porcine model and found that ITD during Active Compression Decompression (ACD) CPR increased 48-hour survival and decreased the degree of post-cardiac arrest AKI in the resuscitated animal [14].

---

Ameer Khalek is a MPH student at the GWU Milken Institute School of Public Health

 

February 21, 2018

Patient experience encompasses the range of interactions that patients have with their providers and healthcare system.  There are many good reasons to measure and report on patient experience data and subsequently reward providers and hospitals for better performance.  However, patient experience surveys and data collection need to be reliable and valid. Reliability and validity are critical because patient experience data are used for a variety of purposes including managerial decisions (e.g., credentialing, contract renewal, or incentive bonus programs) [1]. In addition, the Centers for Medicare & Medicaid Services (CMS) Hospital Value-Based Purchasing (VBP) plan identifies patient experience as a fundamental marker of value and integrates patient experience into public reporting. Patient experience data will become increasingly important with the implementation of Merit-Based Incentive Payment System (MIPS) under the Medicare Access and CHIP Reauthorization of 2015 (MACRA). In 2016, 44% of healthcare organizations reported having a Chief Experience Officer (CXO) added to their C-Suite with the underlying expectation to financially benefit from higher patient experience scores and positive publicity [2]. Currently, there are no patient experience mandates for emergency medicine. However, these are on the close horizon through the Emergency Department Consumer Assessment of Healthcare Providers and Systems or ‘EDCAHPS’).

A recent study by Pines et al. published in the Annals of Emergency Medicine sheds light on the current state of patient experience data reporting in emergency medicine. The study looked at month-to-month variability and construct validity in patient experience data gathered from 2012-2015 covering 1,758 facility-months and 10,328 physician-months using data from Press-Ganey Associates [5]. Using a retrospective cross-sectional study design, Dr. Pines and his team found that patient experience data varied greatly month-to-month with physician variability being considerably higher. They also found that facility-level scores have greater construct validity than physician-level ones.

Figure 1: Estimated Facility Press Ganey Rank Changes

Across facility-months, 40.8% had greater than 10 points of percentile change, 14.7% changed greater than 20 points, and 4.4% changed greater than 30. For facility-level construct validity, several facility factors predicted higher scores: teaching status; elder, male, and discharged patients without Medicaid insurance; lower patient volume; less requirement for physician night coverage; and shorter lengths of stay for discharged patients.

Figure 2: Estimated Physician Press Ganey Rank Changes

Across physician-months, 31.9% changed greater than 20 points, 21.5% changed greater than 30, and 13.6% changed greater than 40. For physician-level construct validity, younger physician age, participating in satisfaction training, increasing relative value units per visit, more commercially insured patients, higher computed tomography or magnetic resonance imaging use, working during less crowded times, and fewer night shifts predicted higher scores.

“The concept of measuring patient experience and rewarding providers who deliver a better experience is absolutely right on. No one argues with that. Yet what we found is that the data currently being gathered is not particularly reliable nor valid.” – Dr. Jesse Pines [6]

Another concern was around the data collection process which results in low numbers of responses and low response rates. For this study, patient response rates fluctuated from 3.6 – 16%. Corresponding author Dr. Arvind Venkat remarked, “Imagine you conduct a survey, and only the very happy and very unhappy return their surveys. What you get is a very biased sample. That makes it difficult to come to any meaningful conclusions from the data [7].” The researchers proposed several approaches to potentially improve response rates, but more work needs to be done in this area to substantiate practice without affecting the validity of the survey.

Ultimately, the optimization of patient experience data gathering may reduce variability in ED patient experience data and better inform decision-making and quality measurement [8]. Yet, until a solution is found to improve reliability and validity, patient experience data – particularly at the provider-level – needs an overhaul. [9].

---

Ameer Khalek is a MPH student at the GWU Milken Institute School of Public Health

 

If you haven’t heard by now, Urgent Matters holds an annual meeting at the American College of Emergency Physicians (ACEP) Scientific Assembly. While the exact theme changes from year to year, the meeting serves to bring together thought-leaders and innovators from the emergency medicine community. Thanks to Philips Blue Jay Consulting and Schumacher Clinical Partners, we are also able to acknowledge teams for developing and implementing effective solutions to common issues facing the emergency care community through our “Emergency Care Innovation of the Year Award.” This year, we received over 20 submissions and our reviewers narrowed them down to one winner and three honorable mentions.

Jesse Pines, MD MBA MSCE – Program Director, Urgent Matters

Rahul Sharma, MD MBA CPE – EM Physician-in-Chief & Medical Director, Strategic Initiatives

Hansen Hsu, MD – Director, EM Informatics & Co-Director, EM Telehealth Services

Peter Greenwald, MD MSC – Director, Telehealth QA & Co-Director, EM Telehealth Services

(Left to Right)

This year’s winner was New York Presbyterian/Weill Cornell Medical Center’s Emergency Department (ED) Telehealth Express Care Service. The ED Telehealth Express Care Service is an innovative ED-based telehealth program designed to streamline emergency care, reduce length of stay (LOS), and improve patient experience for patients with low acuity complaints.

The process for ED Telehealth Express Care patients begin as it would for any other ED patient: an in-person triage and assessment. A registered nurse speaks with the patient and categorizes the severity of their medical complaint. Next, an Advanced Practice Provider (Physician Assistant or Nurse Practitioner) evaluates the patient, performs the Medical Screening Exam which is required for all ED visits, and decides whether the Telehealth program would be medically appropriate. Qualification for the Telehealth Express Care Service is based on set criteria (available in the UM Toolkit). Participation by the patient is voluntary. If the patient declines, a physician will examine them in-person. The telehealth encounter takes place with the on-call Telehealth Physician, a board certified Emergency Medicine faculty member. Evaluation, diagnosis, treatment, and education are discussed at-length, with family members present if desired. Discharge instructions are printed directly at the end of the visit by the physician with any prescribed medications sent electronically to a pharmacy of the patient's choice.

So, why would someone physically go to an emergency department, only to be seen by a remote physician through a screen? Dr. Sharma responds in an Urgent Matters podcast:

“When was the last time you actually went to the bank and went to the teller instead of the ATM to get cash? You don’t, you only go in when you need a certified check or something complicated.”

The data supports him. The ED Express Care Service has been in place since July 2016 is available at two of six EDs at New York Presbyterian: Weill Cornell - a Level 1 Trauma Center, and Lower Manhattan - a community hospital. Together these hospitals see an annual volume of over 140,000 visits. To date, over 4,000 patients have been through the ED Express Care Service with impressive results. Express Care Service has helped reduce wait times for low-acuity patients from 2.5 hours to an average 39 minutes from arrival to discharge. The project also helps to establish primary care appointments, and twenty percent of patients receive follow-up appointments via the in-house ED patient navigator program with a primary care physician or sub-specialist prior to discharge. From the first year of data, Express Care patients, compared to those receiving a traditional fast track evaluation, were younger (median age: 38 [IQR, 27-54] vs. 43 [31-58]; P<0.001) and more likely to be male (52% vs. 46%; P<0.001), Express Care patients had less acute illness as measured by triage severity score (ESI 4 or 5: 97% vs. 84%; P<0.001), and more likely being treated for wound check/suture removal and infectious illness. Express Care patients were less likely to have x-rays preformed as part of their ED evaluation (24% vs. 42%; P<0.001). Express Care patients were treated and released more quickly than fast track patients (median time door to discharge 39 minutes as compared to 120 minute). Express Care patients were less likely to return within 72 hours, and no Express Care patient returning within 72 hours required admission to the hospital. There was a trend to higher Press Ganey satisfaction among Express care patients as compared to fast track ED patients (median: 100 [87-100] vs. 89 [74-100]; P=0.15). Another analysis listed in their submission comparing older patients to younger patients in express care found that although Express Care patients trended younger than traditional ED pathway patients on average, patients 60 and older comprised 24% of the total Express Care population, indistinguishable from the percentage of people 60 and older in the conventional treatment fast track area. Among patients 60 years and over seen in Express Care, the average age was 72. There were no significant differences by age group with respect to 72 hour returns to the ED and a low likelihood for a change in treatment plan on return. In addition to eliciting patient feedback regarding satisfaction with the ED experience, the team surveyed participating physicians about their telemedicine background and experience with the program, with most physicians reporting a significant increase in comfort and positivity about telemedicine patient encounters.Future Direction

At the Urgent Matters meeting, Dr.’s Rahul Sharma, Peter Greenwald, and Hanson Hsu spoke on plans to expand the project by incorporating nursing homes consultation, telehealth kiosks in NYC pharmacies, and paramedic telemedicine visits for CHF patients. Dr. Greenwald also briefly mentioned the value of engaging with insurance companies to provide coverage for virtual urgent care as part of a benefits package. Dr. Ali Raja, Vice Chairman of the Department of Emergency Medicine at Massachusetts General Hospital was quoted in the Wall Street Journal saying, “Ten years from now, tele-emergency medicine will be the standard around the country. We’ll still have emergency departments for those patients who are critically ill, but I think we’re all headed in this direction.”

Personally, I would like to extend my congratulations to the ED Express Care team for their innovative, patient-centered, and value-driven work. Make sure to look out for an upcoming webinar with the New York Presbyterian/Weill Cornell Medical Center team (subscribe here).

For more:

• Urgent Matters Toolkit Submission
• NYP On-Demand
• ACEP Now
• Crain’s Health Pulse
• The Wall Street Journal
• New England Journal of Medicine Catalyst.

---

Ameer Khalek is a MPH student at the GWU Milken Institute School of Public Health

August 20, 2017

Nearly 2.7 million patients present to the emergency department (ED) annually with acute low back pain (LBP). Current medical practice dictates that specific nonsteroidal anti-inflammatory drugs (NSAIDs) are an efficacious first-line therapy. However, uncertainty remains on whether the addition of other classes of medication (i.e., combination therapy) can further improve LBP outcomes. Currently, benzodiazepines (e.g., diazepam) are considered a prescriber favorite to accompany NSAID therapy and are prescribed in 300,000 US ED visits for LBP, annually.[i]

How We Got Here

 In 2014, Williams et al. published a practice-changing study in The Lancet showing that neither regular, nor as-needed acetaminophen dosing affects recovery time when compared with placebo in LBP.[ii] Since then, naproxen has become a standard over-the-counter (OTC) medication for acute LBP. Friedman et al. (2014) added to the evidence surrounding care of acute LBP by showing that under the same presentation, adding cyclobenzaprine or oxycodone/acetaminophen to naproxen therapy does not improve functional outcomes.[iii]  A 2017 study published in Annals of Emergency Medicine by Friedman et al. raises the question:

Will adding a short-term course of diazepam to naproxen yield any additional benefit to functional outcomes at one-week post-discharge from the ED in patients with acute LBP?[iv]

An Urgent Matters podcast with Dr. Friedman details the study as well as how it may or may not change emergency medicine practice.[v] 

Friedman et al. 2017

Dr. Friedman and his team conducted a randomized, double-blind, comparative efficacy clinical trial for adult patients presenting to the ED with acute (onset <2 weeks), atraumatic, non-radicular LBP and functional impairment quantified by a Roland-Morris Disability Questionnaire (RMDQ) greater than 5. The RMDQ is a validated health status measure for LBP in both research and clinical practice consisting of 24 Yes/No statements, each corresponding as one point if marked “Yes,” and no point if marked “No.” Both intervention and control group received a 10-minute LBP educational session and naproxen 500mg PO BID as-needed. The intervention group received either 28 tablets of diazepam 5mg or identical placebo to be taken BID as-needed.

• Intervention Group: Education + 500mg Naproxen BID (as needed) + 5mg Diazepam BID (as-needed)
• Control Group: Education + 500mg Naproxen BID (as needed) + Placebo BID (as-needed)

While many outcomes were measured starting at the point of discharge, the primary outcome measured was a change in RMDQ score between ED discharge and one-week follow-up. A decrease of 5 or more points when compared to the other group was considered clinically significant. The secondary outcome measures included pain intensity on a subjective four-point descriptive scale (none, mild, moderate, and severe) at one-week and three-month checkpoints and medication adherence for LBP within 72 hours of ED discharge.  Only 21% (114/545) of the patients screened were included in the study after meeting selection criteria. The mean RMDQ score of both patients randomized to the intervention and control group improved by 11 points. The study found no comparative improvement in functional outcome at one week when diazepam was added to a naproxen regimen. Additionally, pain intensity at one week and three months was no different between the two groups, with similar incidence of adverse events throughout the trial.

While limitations to the study certainly exist, results appear reasonable when taken in context of the reality behind clinical practice. Most noteworthy of the limitations is that the study was conducted in a single urban healthcare system serving a population with predominantly low socioeconomic status (SES). Lower neighborhood median household incomes and higher unemployment rates have been associated with longer length of disability (i.e., longer prognosis), potentially impacting outcome data.[vi] Secondly, in the interest of maintaining cohort homogeneity, researchers excluded many patients secondary to strict entry criteria. Therefore, there is a reasonable index of suspicion about the study’s generalizability to patients with other types of back pain. Anecdotally speaking, often times the onset of radiating pain can serve as the impetus behind seeking emergency care. Finally, it is important to note that while diazepam is also prescribed as an anxiolytic, this would have skewed data in favor of the intervention group in controlling LBP (not the case here). In the future, research should be done on the efficacy of alternative treatments for LBP such as controlled stretching and thermal pack use.

Key Takeaways

Diazepam does not have marginal effectiveness when prescribed alongside naproxen for outpatient management of atraumatic, nonradicular LBP following ED discharge when it comes to functional outcomes at one week and pain outcomes at one and three months. Patients should routinely be counseled on the potential for benign symptoms to persist, encouraged to use the recommended dose of naproxen as-needed, and return to the ED with any significant changes.

“Physicians should reassure their patients that acute and subacute low back pain usually improves over time regardless of treatment.” - ACP President Nitin Damle, MD MS

 Barriers to Implementation

The often benign symptoms of acute atraumatic, nonradicular LBP and the subjective nature of reporting pain lead to frustration for both patient and provider. Additionally, clinically irrelevant but psychologically important factors such as patients expectation for prescriptions after physician interaction are detrimental to prescriber stewardship.[vii] Physicians may also feel pressure to order additional testing – for this, the Choosing Wisely initiative is a great resource to aid physicians in identifying patients that warrant further testing.

[i] Friedman BW, Chilstrom M, Bijur PE, Gallagher EJ. Diagnostic testing and treatment of low back pain in US emergency departments. A national perspective. Spine. 2010;35(24):E1406-E1411. doi:10.1097/BRS.0b013e3181d952a5.

[ii] Efficacy of paracetamol for acute low-back pain: a double-blind, randomised controlled trial Williams, Christopher M et al. The Lancet , Volume 384 , Issue 9954 , 1586 - 1596

[iii]  Friedman BW, Dym AA, Davitt M, Holden L, Solorzano C, Esses D, Bijur PE, Gallagher EJ. Naproxen With Cyclobenzaprine, Oxycodone/Acetaminophen, or Placebo for Treating Acute Low Back PainA Randomized Clinical Trial. JAMA. 2015;314(15):1572–1580. doi:10.1001/jama.2015.13043

[iv] Diazepam Is No Better Than Placebo When Added to Naproxen for Acute Low Back Pain Friedman, Benjamin W. et al. Annals of Emergency Medicine , Volume 70 , Issue 2 , 169 - 176.e1

[v] Pines JM. – Acute Low Back Pain: Diazepam No Better Than Placebo? with Dr. Benjamin Friedman. Urgent Matters Podcast. 2017

[vi] Shraim M, Cifuentes M, Willetts JL, Marucci‐Wellman HR, Pransky G. Regional socioeconomic disparities in outcomes for workers with low back pain in the United States. American Journal of Industrial Medicine. 2017;60(5):472-483. doi:10.1002/ajim.22712.

[vii] Sirota, M., Round, T., Samaranayaka, S., & Kostopoulou, O. Expectations for antibiotics increase their prescribing: Causal evidence about localized impact. Health Psychology, 36(4), 402-409.

---

Ameer Khalek is a MPH student at the GWU Milken Institute School of Public Health

August 7, 2017

There has been a significant and steady increase in cannabis (i.e., marijuana) use in the U.S. between 2002 and 2015.[i] California was the first state to allow for medical use of cannabis after Proposition 215 in 1996. Currently, 29 states, the District of Columbia, Guam, and Puerto Rico have legalized cannabis for medical use, eight of which have also legalized recreational use. Results from a 2015 nationwide survey indicate that over 22 million Americans (>12 years) reported using cannabis within 30 days of the survey.[ii] Despite increases in use, there is a dearth of conclusive evidence around the harms and benefits of cannabis. This not only makes the topic difficult to discuss in a healthcare setting, but also contributes to a significant public health concern for at-risk populations. Recently on the Urgent Matters Podcast, Dr. Esther Choo (an emergency physician at Oregon State Health & Science University) spoke about her experiences working with a population that has access to legal recreational marijuana use.[iii] When asked if she includes marijuana use as a part of her social history she stated:

“The policy has outpaced our evidence base… We ask, and then what are we going to say?”

Dr. Choo highlights a thought-provoking dilemma that physicians face. The culture behind marijuana use is supported by public sentiments of both its health benefits as well as potential harms.  However, physicians are not equipped with the evidence base to either endorse or recommend against it. As more states legalize marijuana, evidence around both short- and long-term health outcomes is needed to better inform patient-physician interaction and accurately evaluate the public health impact of marijuana use. So, how close are we to having a pragmatic discussion centered around cannabis with our patients?

The Evidence Base Around Cannabis and Cannabinoids

A 2017 report from the National Academies of Sciences, Engineering, and Medicine offers an extensive review of relevant scientific literature surrounding the health effects of cannabis and cannabinoids published after 1999.[iv]  While there is an abundance of information available in the 468-page report, some key conclusions that I find most informative with potential to impact clinical practice are listed below. I encourage you to at least read through the annex of the report for a summary of evidence-based recommendations on cannabis/cannabinoids and their impact on various cases such as cancer, cardiometabolic risk, respiratory disease, immunity, and mental health.

There is conclusive/substantial evidence that cannabis/cannabinoids are effective for:

1. Chronic pain in adults (cannabis)
2. Chemotherapy-induced nausea & vomiting (oral cannabinoid)
3. Patient-reported multiple sclerosis spasticity symptoms (oral cannabinoid)

There is limited evidence that cannabis/cannabinoids are effective for:

1. Clinician-measured multiple sclerosis spasticity symptoms (oral cannabinoid)
2. Anxiety symptoms in individuals with social anxiety disorder (cannabidiol)
3. Posttraumatic stress disorder (nabilone)

There is no or insufficient evidence to support or refute the conclusion that cannabis or cannabinoids are an effective treatment for:

1. Cancers, including glioma (cannabinoids)
2. Symptoms of irritable bowel syndrome (dronabinol)
3. Epilepsy (cannabinoids)
4. Dystonia (nabilone and dronabinol)
5. Abstinence in the use of addictive substances (cannabinoids)

There is substantial evidence of a statistical association between cannabis smoking and:

1. Worse respiratory symptoms and more frequent chronic bronchitis episodes
2. Development of schizophrenia or other psychoses

Barriers & Future Direction

The National Academies of Sciences, Engineering, and Medicine report outlines many barriers to conducting research on the health effects of cannabis and cannabinoids. Any research around cannabis, a Schedule I substance, faces strict regulatory barriers. Additionally, researchers are often unable to obtain the quantity, quality, and type of cannabis needed to address specific research questions and draw scientific conclusions. If a physician is to prescribe cannabis, their patient should be well informed (i.e., starting dose, CBD vs. THC concentration, how to escalate safely, medication interaction, etc…). While we will have to wait for more evidence to make informed decisions on cannabis use, what about the legalization of recreational use? Does it impact healthcare? A 2017 study published in Preventive Medicine by Wang et al. raises the question:

Does the legalization of medical and recreational marijuana have an impact on hospitalizations, emergency department (ED) visits, and regional poison center (RPC) calls in Colorado?[v]

Wang et al. 2017

Dr. Wang and his team conducted a retrospective review of data collected from the Colorado Hospital Association (CHA) Discharge Databank and National Poison Data System (NPDS) data repository. The primary objective was to compare rates of hospitalizations and ED visits with marijuana-related billing codes and RPC calls with mention of marijuana. The secondary objective was to compare the primary diagnosis categories of hospitalizations and ED visits with marijuana-related billing codes to those without.

From 2000 to 2015, hospitalization rates with marijuana-related billing codes increased from 274 to 593 per 100,000. Interestingly, the prevalence of mental illness among ED visits with marijuana-related codes was 5x higher than the prevalence without – potentially due to the subsection of patients often receiving more detailed assessment around substance abuse habits, or having a greater number of visits to the ED. RPC calls remained constant from 2000 to 2009, but upon the legalization of medical marijuana in 2010 the number increased from 42 to 93. After recreational legalization in 2014, RPC calls increased from 123 to 221.

Figure 1: Rates of hospitalizations and ED visits with marijuana-related billing codes in the first three diagnosis codes in CO

Figure 2: Annual RPC human exposure calls related to marijuana in CO, by age group

Overall, legalization of both medical and recreational marijuana in Colorado is associated with an increase in hospitalizations, ED visits, and RPC calls. Since the study was conducted in a state with legalized recreational marijuana, the findings are likely generalizable to states with similar laws. However, as expected in such complex area, limitations do exist. Researchers relied solely on ICD codes and drug screens without conducting full chart reviews. Marijuana use cannot be viewed as the direct cause of most ED visits, hospitalizations, or RPC calls, as the effects are often short-lasting and not clinically relevant. The data provided does not establish causality between hospitalizations or ED visits secondary to marijuana exposure, and future research in this area should emphasize the importance of not overestimating the healthcare burden related to marijuana use. While researchers did well to establish an association between legalization of marijuana and increased healthcare utilization, there is still much work to be done in this area.

Key Terms

• Marijuana: colloquial term for the flowering plant intended for consumption. Used in discussion of Wang et al. to align with manuscript terminology.
• Cannabis: also known as marijuana, is the science-based term of flowering plant intended for consumption.
• Cannabinoid: a single class of diverse chemical compounds that act on cannabinoid receptors in the brain. The most notable cannabinoid is tetrahydrocannabinol (THC), the primary psychoactive compound in cannabis.
• Dronabinol: the international nonproprietary name (INN) for a pure isomer of THC, often used to treat anorexia in patients with HIV/AIDS as well as for refractory nausea and vomiting in people undergoing chemotherapy.
• Nabilone: a synthetic cannabinoid that mimics THC and is often used as an antiemetic and adjunct analgesic for neuropathic pain.

[i] CBHSQ (Center for Behavioral Health Statistics and Quality). 2016. Behavioral health trends in the United States: Results from the 2014 National Survey on Drug Use and Health (HHS Publication No. SMA 15-4927, NSDUH Series H-50). 

[ii] NCSL (National Conference of State Legislatures). 2016. State medical marijuana laws.

[iii] Pines JM. – Cannabis in the Emergency Department: policy vs. evidence with Dr. Esther Choo. Urgent Matters Podcast. 2017

[iv] National Academies of Sciences, Engineering, and Medicine. 2017. The Health Effects of Cannabis and Cannabinoids: The Current State of Evidence and Recommendations for Research. Washington, DC: The National Academies Press. doi:https://doi.org/10.17226/24625.

[v] Wang GS, Hall K, Vigil D, Banerji S, Monte A, VanDyke M. Marijuana and acute health care contacts in Colorado. Prev Med. 2017 Mar 30. pii: S0091-7435(17)30120-2. doi: 10.1016/j.ypmed.2017.03.022.

---

Ameer Khalek is a MPH student at the GWU Milken Institute School of Public Health