#114 High Value Care: Assess Quality, Mitigate Diagnostic Uncertainty, Overcome Barriers

September 17, 2018 | By Matthew Watto, MD

Kick up the quality of your care with tips and tactics from Caitlin Clancy MD, coauthor of ACP’s High Value Care Curriculum. We learn to define quality and value in healthcare; the most common barriers to high value care; use of probability and likelihood ratios to boost clinical reasoning and combat diagnostic uncertainty; some useful tools to estimate cost; sources of healthcare waste; and some general pearls on how the healthcare system works…or doesn’t. ACP members can claim free CME-MOC at acponline.org/curbsiders (goes live 0900 EST on podcast release date).

 Join our mailing list and receive a PDF copy of our show notes every Monday. Rate us on iTunes, recommend a guest or topic and give feedback at thecurbsiders@gmail.com.

Credits:
Written & Produced by: Matthew Watto MD
Hosts: Paul Williams MD, Stuart Brigham MD, Matthew Watto MD
Guest: Caitlin Clancy MD

High value show notes

Definitions
Value is equal to quality divided by cost Value = Quality / Cost

Quality refers to all the benefits, both immediate and long term, that can be derived from a particular diagnostic test or treatment.

Cost refers to the financial and non-financial harms of a particular diagnostic test or treatment e.g. missed work, physical or emotional suffering, etc.

The above definitions are according to Clancy et al. ACP HVC Curriculum 4.0 Module 1.

The problem with waste
In 2016 Healthcare spending was $3.3 trillion dollars, which made up 17.9% of the US Gross
Domestic Product (Details here at CMS.gov).

One in three healthcare dollars are waste! (Porter ME. N Engl J Med 2010). The top three contributors to spending were unnecessary services, excess administrative costs, inefficient delivery of care. Missed prevention is an honorable mention (Clancy et al. ACP HVC Curriculum 4.0 Module 1).

Clinicians don’t intend to order unnecessary services, but several factors lead to these wasteful practices (Clancy et al. ACP HVC Curriculum 4.0 Module 1).

1. Knowledge/cognitive: Diagnostic uncertainty, lack of knowledge and lack of time to look things up create discomfort and drive over ordering.
2. Process or systems related issues: It’s often easier to order something than not to order it (e.g. recurring daily labs, or repeating an echo instead of requesting records).
3. Culture: Institutional role models or local practice habits can propagate low value care. Patient pressures and trainees not wanting to disappoint their attending are also contributors.

The cost of being uninsured
In our example from the show an uninsured female comedian tore her ACL in a beer league softball game. This left her with medical bills of nearly $50,000.

The uninsured patient lacks the negotiating power (leverage) of a large private insurance company or government institution. As a result, they end up getting charged the inflated face value for diagnostic tests and procedures. This is a common cause of bankruptcy (Clancy et al. ACP HVC Curriculum 4.0 Module 2).  

The insured patient pays a monthly insurance premium that is often subsidized by their employer. When they need diagnostic testing or treatment their insurance company negotiates a lower reimbursement rate with the institution providing the needed services. The patient pays their deductible and the insurance company pays the difference. Government insurers can negotiate lower reimbursement than private insurers due to their larger size (Clancy et al. ACP HVC Curriculum 4.0 Module 2).

Large hospital systems often charge insurance companies more for their services than privately owned centers or independent practices because of their large market share and larger operating costs (Philly.com Article on Two Echocardiograms). “Economies of scale don’t translate to lower prices. With their market power bigger providers can simply demand more.” (Elisabeth Rosenthal in her book American Sickness, Rule #7). For example, if insurance company A refuses to pay the high prices charged by hospital system A (who has 60% market share in a region) then insurance company A may lose their customers to insurance company B who remains in good standing with hospital system A.

Do insured patients do better?

What we know about insured patients:

1. They have “less health-related financial stress” (NYTimes Upshot Blog 2015 by Austin Frakt).
2. They self report better health.
3. They access more services, including preventive services (Baicker et al. NEJM 2013).
4. Data shows an earlier time to diagnosis and treatment for cervical cancer (Markt et al. PLOS One. 2018 ).
5. More people with colorectal cancer go for curative surgery (Althans et al. Dis Colon Rectum. 2018 PMID 29219921.
6. Data suggests improved control of hypertension (Christopher et al. Am J Pub Health. 2016 PMC: 4695932).
7. They have improved symptoms of depression (Baicker et al. Milbank Q. 2018 PMID: 29504203).

Much of the the above data comes from the Oregon Health Insurance Experiment.

Shopping for healthcare

Shoppable tests

Doctors and patients are doing a poor job at being conscious about the cost of care. Only about 1% of healthcare consumers are getting the best prices for “shoppable” tests e.g. an elective MRI (nytimes.com article by Austin Frakt ; http://www.nber.org/papers/w24869).

Databases to search cost of testing

Here are two resources that allow you to enter a zip code and see the range of prices for a given diagnostic test. Use these resources to determine the fair market price.

1. HealthcareBluebook.com
2. FairHealthConsumer.org

Diagnostic uncertainty

Dr Clancy recommends using pretest probability, likelihood ratios and posttest probability to help mitigate diagnostic uncertainty (Clancy et al. ACP HVC Curriculum 4.0 Module 3). The example used on our show included a patient presenting to clinic with suspect heart failure. The pretest probability can be estimated based on a patient’s history, exam and initial data. Next, consider the likelihood ratio (LR) of a given diagnostic test. The magnitude of the likelihood ratio determines its effect on posttest probability. For example, a LR of 2 will raise the posttest probability by ~15%.

In general:
LR <5 = small effect on posttest probability
LR 5-10 = moderate effect on posttest probability
LR >10 = large effect on posttest probability
(source med.emory.edu)

Use the likelihood ratio nomogram to draw a straight line from the pretest probability through the LR to find posttest probability. A test isn’t worth ordering if it is unlikely to change the posttest probability or patient care. Dr Clancy recommended trying the docLogica app, which can help clinicians perform these calculations without a pencil and paper.

Barriers to High Value Care

Some of the top barriers to high value care include defensive medicine, local culture and patient expectations (Clancy et al. ACP HVC Curriculum 4.0 Module 5). On the show we discussed how to recognize these barriers and use communication and quality improvement projects to overcome them. Clinical decision tools can be added to the electronic health record to help avoid low value practices (e.g. ordering antibiotics inappropriately, making it difficult to order daily labs, or prompting the clinician with questions when ordering a blood transfusion for a hemoglobin above 7). There is some evidence that providing peer prescribing data may lower antibiotic use (Meeker et al. JAMA 2016 PMID 26864410).

Goal and Learning Objectives

Goal:

Listeners will define and recognize key aspects of high value care including: cost, quality, and barriers to high value care.

Learning objectives:

After listening to this episode listeners will…

1. Define high value care
2. Recall the magnitude and common sources of health care waste
3. Learn to assess patient costs and identify price transparency tools to determine fair market value for shoppable tests
4. Recall how insurance status effects clinical outcomes
5. Assess the value and clinical utility of diagnostic testing using pretest probability and likelihood ratios
6. Describe the barriers to high value care in clinical practice and explore ways to overcome these barriers.

Disclosures:

Dr Clancy reports no relevant financial disclosures. The Curbsiders report no relevant financial disclosures.

Time Stamps

• 00:00 Disclaimer, intro and guest bio
• 03:50 Guest one-liner, some discussion on non traditional paths, and picks of the week
• 12:10 Case 1: Defining value, cost, quality; sources of healthcare waste; and how to avoid the waste of daily labs
• 23:00 Case 2: An uninsured patient who needs major surgery; costs for uninsured versus insured patients
• 28:35 Do insured patients have better outcomes?
• 31:15 How to determine cost and “fair market price”
• 36:29 Case 3: Clinical case of suspected heart failure
• 45:45 Case 4: Clinical case of patient requesting antibiotics; barriers to high value care and how to overcome them
• 57:05 Outro

Links from the show

1. Jericho Sires (album) by The Hot Snakes
2. Love at First Sting (album) by The Black Scorpions
3. The Slow Burn (podcast) by Slate
4. American Sickness (book) by Elisabeth Rosenthal
5. docLogica app for Medical Diagnosis for the 21st Century
6. ACP’s High Value Care Curriculum (Free) https://www.acponline.org/clinical-information/high-value-care/medical-educators-resources/curriculum-for-educators-and-residents
7. Look up the fair price for medical tests and compare facilities in your area https://www.healthcarebluebook.com or https://www.fairhealthconsumer.org
8. Philly.com article on 10x different in cost of echo http://www.philly.com/philly/health/health-costs/two-echocardiograms-same-hospital-drastically-different-prices-339-and-3484-20180809.html
9. NYTimes.com article on shoppable healthcare
10. https://www.nytimes.com/2018/07/30/upshot/shopping-for-health-care-simply-doesnt-work-so-what-might.html