SGEM#272: Take the Money and Run without Getting a CT

Podcast: Play in new window | Download

Subscribe: RSS

Date: October 22nd, 2019

Reference: Iyengar R et al. The Effect of Financial Incentives on Patient Decisions to Undergo Low-value Head Computed Tomography Scans. AEM October 2019.

Guest Skeptic: Dr. Justin Morgenstern is an emergency physician and the creator of the excellent #FOAMed project called First10EM.com

Case: A 21-year-old comes into the emergency department after being knocked unconscious while playing rugby. The patient is now feeling great, or as they say in New Zealand “sweet as”. He had no pain, nausea, or neurologic symptoms. His exam is normal. You aren’t worried, but his dad is the coach of the American national rugby team and says that his players always get a CT when this happens. You wonder what factors might influence a patient’s preference for imaging?

Background: The CT scan is arguably one of the most important pieces of diagnostic technology that we use in emergency medicine. It allows for incredibly rapid identification of a myriad of life-threatening conditions.

However, likely because it is such a valuable tool, there seems to be little doubt that we overuse it. For example, one study that looked retrospectively at all head CTs ordered for trauma concluded that more than 1/3 were unnecessary based on the Canadian CT head rule [1].

Not only does unnecessary testing reduce efficiency and add costs, it also directly harms patients with unnecessary radiation [2]. Many imaging decisions are obvious – the patient either clearly requires or clearly does not require imaging.

One way to decrease CT scans of the head is to use a clinical decision instrument like the Canadian CT Head Rule (CCHR). The SGEM covered the classic paper on the CCHR by the Legend of Emergency Medicine Dr. Ian Stiell on SGEM#106.

We also recently reviewed a paper that looked at increasing the CCHR age criteria from 65 years of age to 75 years of age (SGEM#266). The bottom line was that this paper opens the door for further research to try to narrow the criteria in the CCHR to further reduce unnecessary head CT imaging in the emergency department. However, further, high quality prospective studies are required prior to clinical application.

There is a great deal of uncertainty in emergency medicine, which leaves a sizeable number of patients in a grey zone – where harms and benefits are closely matched, qualitatively different, or just unknown. For these patients, shared decision making is probably the best route forward.

Even when it seems clear to the physician that imaging isn’t required, we can be met with resistance from our patients. In addition, if we are working in a zero-miss culture, we may be more likely to order CT scans that are not medically necessary. Thus, it is important to know what factors influence patients’ decision to undergo CT.

This study by Iyengar and colleagues examines the impacts of financial incentives, as well as varying levels of risk and benefit, on patient preference for CT imaging in the setting of low risk head injury [3].

---

Clinical Question: Do financial incentives, together with potential risk and potential benefit information, influence patient preference for diagnostic testing?

---

Reference: Iyengar R et al. The Effect of Financial Incentives on Patient Decisions to Undergo Low-value Head Computed Tomography Scans. AEM October 2019.

• Population: A convenience sample of adult patients presenting to the University of Michigan emergency department.
  • Exclusions: Patients with chest pain or head trauma (because those were the conditions in the hypothetical cases presented). They also excluded patients with altered mental status, with contact precautions, or in resuscitation bays.
• Intervention and Comparison: Patients were all presented with a hypothetical low risk head trauma scenario. The scenario was designed such that the Canadian Head CT rule suggests against imaging. Three aspects of the scenario were randomized:
  • Benefit: This was presented as either 1% or 0.1%
  • Risk: This was presented as either 1% or 0.1%
  • Incentive: Patients were offered either $100 to forgo the CT, or $0.
    • All risk and benefit information were provided in multiple formats, include percentages (0.1%), ratios (1 in 1,000), and in visual depictions.
  • Outcome:
    • Primary Outcome: The percentage of patients that chose to receive a CT scan.
    • Secondary Outcome: They performed multiple regressions to control for potential confounders.

This is an SGEMHOP episode which means we have one of the authors on the show. Dr. William Meurer is an emergency physician. His focus is on the treatment of acute neurological emergencies, both as a researcher and clinician. He has been part of the University of Michigan Acute Stroke Team since 2006. In addition, Dr. Meurer has experience enrolling patients in acute trials and has served as a local PI for the CLEAR-ER trial (a trial enrolling acute stroke patients in the ED that tested a reperfusion strategy). He is on the executive team of the Strategies to Innovate EmeRgENcy Care Clinical Trials Network (SIREN). Dr. Meurer has other active or recently completed NIH funded clinical trials involving acute vertigo in the emergency department, hypertension, and therapeutic hypothermia after cardiac arrest.

Jessica Winkels is the second author on this AEM publication and also joins us on the podcast. She is a fourth-year medical student at the University of Michigan. Jessica is planning on going into emergency medicine after she graduates in the spring. Publishing in AEM should certainly help with her application.

Authors’ Conclusions: “Providing financial incentives to forego testing significantly decreased patient preference for testing, even when accounting for test benefit and risk. This work is preliminary, hypothetical, and requires confirmation in larger patient cohorts facing these actual decisions.”

Quality Checklist for Randomized Clinical Trials:

Although this is an RCT, it is different from our usual RCTs, and some of the questions on our check list aren’t as applicable as when we look at RCTs of therapeutic interventions. Critical appraisal is always complex, and even the two of us had to turn to a true expert in Dr. Chris Carpenter to determine which checklist was the most appropriate to use.

1. The study population included or focused on those in the emergency department. Yes
2. The patients were adequately randomized. Yes
3. The randomization process was concealed. Unsure
4. The patients were analyzed in the groups to which they were randomized. Yes
5. The study patients were recruited consecutively (i.e. no selection bias). No
6. The patients in both groups were similar with respect to prognostic factors. Unsure
7. All participants (patients, clinicians, outcome assessors) were unaware of group allocation. No
8. All groups were treated equally except for the intervention. Yes
9. Follow-up was complete (i.e. at least 80% for both groups). Yes
10. All patient-important outcomes were considered. Yes
11. The treatment effect was large enough and precise enough to be clinically significant. Yes

Key Results: They enrolled 913 patients, with a median age of 45 years of age and 56% of the population was female. The vast majority of this population identified as Caucasian and had attended at least some college. Overall, 54.2% of patients elected to receive a CT scan.

---

Decreased benefit, increased risk, and offering a cash incentive to forgo CT all decreased the desire for CT.

---

• Primary Outcome: The percentage of patients that chose to receive a CT scan.
  • If the benefit was reported as 0.1% then 49.6% of people wanted a CT, whereas if it was 1% then 58.9% wanted a CT. (OR 1.48 95% CI 1.13 – 1.92)
  • If the risk was reported as 0.1% then 59.3% of people wanted a CT, whereas if it was 1% then 49.1% wanted a CT. (OR 0.66 95% CI 0.51-0.86)
  • If no cash incentive was offered then 60% of people wanted a CT, whereas if 100$ was offered to forgo the CT then 48.3% of people wanted a CT. (OR 0.64 95% CI 0.49-0.83)
• Secondary Outcomes: The results remained consistent when adjusted for various potential confounders including age, gender, race, income, level of education, and prior history of health problems.

You can listen to the podcast on iTunes or Google Play to hear Will and Jessica answers to our ten nerdy questions.

1. Sample Size: Your sample size was based on the feasibility of medical students being able to complete a summer research project. This would give an approximate power of 85% to 90% to detect a 10% absolute change in the proportion of subjects desiring testing from a baseline test acceptance rate of 50%. Do you think that a 10% difference reflects a real clinically important difference?

2. Statistics: You performed a series of nested regression analyses for your primary statistical analysis. I’ll be honest, we got a little lost in the math. In our relatively simple mind, there were only a couple variables, with a simple yes or no answer regarding CT. It seems like presenting the raw numbers would have been easier to understand than the odds ratios that you ended up using. Can you explain your choice of statistics to me?

3. External Validity: The vast majority of this population was highly educated and white. There was also a very high percentage (24%) that worked in healthcare. How might that affect the external validity of the results?

4. External Validity 2: We was incredibly surprised than half of these patients wanted a CT. In Canada and New Zealand, a CT would not even have been offered to these patients (given that they passed the Canadian CT head rule). We often explain why a CT isn’t needed, and the vast majority are fine with that. We definitely haven’t experienced 50% of my patients asking for a CT. We therefore wonder how these results might apply in other countries.

5. Hypothetical Numbers: You chose to use hypothetical risks and benefits, rather than using known benefit and harm data. The hypothetical numbers could make these results less applicable in real clinical settings. You discuss it briefly in the paper, but could you explain the choice to use 1% and 0.1% and your numbers?

6. Real World Shared Decision Making: As you mention in the discussion, unlike the exact risk and benefit numbers you present here, it is often incredibly difficult to determine the exact risk and benefit of a test for the patient in front of you. Personally, we think that is the hardest part of this job. How do you think that uncertainty in real world practice would impact these results?

7. Health Inequities: This really isn’t a nerdy question you can answer from your data, but we wonder whether offering cash incentives could result in inequities for our patients. It seems like the $100 incentive is more likely to be enticing to someone making minimum wage than someone earning a six-figure salary. Do we want healthcare to be distributed base on something other than the benefits and harms of the intervention itself?

8. Thought Experiment or Practical Plan: We wonder whether you see this as just a thought experiment at this point, or are you thinking that people should actually institute some kind of cash incentive to reduce CT use? Where would the $100 come from? I imagine getting a patient out of the hospital earlier, and freeing up their stretcher, might actually generate more than the $100 needed to incentivize them not having the scan. Have you thought about the overall economics of this model?

9. Health Literacy: We think you did a very good job explaining the risk in multiple ways – including both numbers and images. However, one number really jumped out at me. In the group with a 0.1% benefit and a 1% harm, 50% of people still wanted a CT scan. We had explicitly told patient that their chance of harm was 10x their chance of benefit, and they still wanted to be scanned. We think that number needs attention. Does it mean that your participants really didn’t understand the numbers you were giving them? Is it just a representation of the harms and benefits being qualitatively different (the benefit is immediate whereas the harm is delayed)? Or is there something else going on, because we find that number shocking.

10. Anything Else: Is there anything else you would like the SGEMers to know about your study?

Comment on Authors’ Conclusion Compared to SGEM Conclusion: We agree with the authors conclusions as it applies to this patient population but are not sure about its external validity to other healthcare systems. It certainly is interesting and does require confirmation in other populations.

---

SGEM Bottom Line: The potential risk, the potential benefit and money can influence people’s behavior in making healthcare decisions.

---

Case Resolution: You explain to the patient that he is very low risk for a serious head injury based on the CCHR. After discussing the risks of CT and the negligible chance of benefit, the patient (and his dad) are happy to observe his symptoms and only get a CT if he gets worse.

Clinical Application: I don’t think I will be offering my patients financial incentives as part of medical decision making any time soon. However, I use shared decision making every shift. For this patient, a patient that passes the CCHR, I wouldn’t actually perform shared decision making, because I think the decision is clear. A CT isn’t needed.

But if the choice was unclear, I would perform shared decision making, presenting the risks and benefits in multiple formats, like the authors did here.

What Do I Tell My Patient?  I am not sure if you need a CT scan at this point.I think chance that we are missing an important injury is about 1%. A CT would catch that injury, but it exposes you to radiation, and so your risk might be 0.1%. Another option would be to stay in the ED for a couple more hours so I can keep an eye on you and perform some repeat neurologic testing.

Keener Kontest: There was no winner last week. In 1949 Sir Robert R. Macintosh first used a device called the gum elastic bougie as a tracheal introducer. We wanted to know what was this device designed for and what is the proper name for the device we refer to as the bougie today? The answer is the gum elastic bougie was a urinary catheter that was originally used for dilation of urethral strictures. The device we call the bougie today is correctly named the Eschmann tracheal tube introducer.

Listen to the podcast to hear this weeks’ trivia question. If you know the answer, send an email to TheSGEM@gmail.com with “keener” in the subject line. The first correct answer will receive a cool skeptical prize.

SGEMHOP: Now it is your turn SGEMers. What do you think of this episode about financial incentives and head CTs? Tweet your comments using #SGEMHOP. What questions do you have for Will, Jessica and their team? Ask them on the SGEM blog. The best social media feedback will be published in AEM.

Also, don’t forget, those of you who are subscribers to Academic Emergency Medicine can head over to the AEM home page to get CME credit for this podcast and article.

• Go to the Wiley Health Learningwebsite
• Register and create a log in
• Search for Academic Emergency Medicine – “October”
• Complete the five questions and submit your answers
• Please email Corey (coreyheitzmd@gmail.com) with any questions or difficulties.

---

Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’ Guide to Emergency Medicine.

---

References: