Low-value care and equity

Given the health risks of overuse, it’s important to understand the impact of low-value care on health equity. Are people of color at higher risk of overuse than white patients, or vice versa? The evidence on this so far is mixed. A systematic review of studies looking at low-value care by race found that white patients were more likely to receive unnecessary care, but another more recent study found higher rates of overuse for certain low-value services in Black and Hispanic patients, such as feeding tubes for dementia. Clearly, there is a lot more to be studied on this issue.

Cue a new analysis from Harvard Medical School professor Dr. Ishani Ganguli and colleagues in The BMJ. This study looked at rates of 40 low-value services in nearly 10 million Medicare patients across 595 health systems. Ganguli and colleagues compared the likelihood of receiving low-value care between Black and white Medicare patients, including a comparison of patients within the same health system. Low-value services measured included screening tests, diagnostic tests, monitoring tests, drugs, and procedures.

Here are key takeaways from their study:

• There were significantly different rates of low-value care for 29 of the 40 services measured, although most of the differences were small. Black patients had higher rates of nine low value services and white patients had higher rates of 20 low-value services. 
• White Medicare patients were more likely to receive low-value screening tests such as prostate-specific antigen testing in men over 75 (31% v 26%) and cardiac screening (5% v 2%), as well as treatments such as antibiotics for cold or ear infection (37% v 33%), and vertebroplasty (5% v 3%) which is an injection of cement into the backbone.
• Black Medicare patients were more likely to receive feeding tubes for advanced dementia (9% v 2%), two or more concurrent antipsychotic medications (8% v 5%), and certain low-value acute diagnostic tests, like imaging for uncomplicated headache (7% v 3%).
• Differences in low-value care remained even when comparing patients within the same health system. That’s important, because it indicates differences in the way patients of different races are treated by the same providers–rather than just differences in culture between the health providers they frequent. 

Drivers of disparities in low-value care

What could explain these differences? The study authors suggest some potential reasons why Black patients could be more likely to receive low-value diagnostic tests:

“Mistrust in the healthcare system because of historical and present day racism might contribute to Black adults being more receptive to diagnostic testing when acutely ill—in this scenario, it is possible that a tangible test is more reassuring than a clinician’s words and might serve to lessen valid concerns about undertreatment.”

The authors also point out that structural barriers to care for Black patients can make it hard to access care earlier and result in them arriving to the ER sicker, which could prompt low-value testing. Additionally, if patients are more likely to seek care in the ER or urgent care as opposed to primary care, they may be subject to more low-value testing from doctors who don’t know them well. Why does this matter? Low-value diagnostic testing exposes patients to radiation exposure, out-of-pocket spending, and additional follow-up testing and procedures (known as “care cascades”).

Black patients were also more likely to receive feeding tubes in the setting of advanced dementia, a practice that does not help patients live longer, and is not recommended for these patients. Differences in rates of feeding tube usage may reflect lack of trust and communication between clinicians and patients; one study found that 14% of family members of patients with feeding tubes reported that there was no discussion about feeding tube insertion, and 42% reported a discussion that was shorter than 15 minutes.  

On the flip side, why are white patients more likely to receive certain low-value screenings and treatments? The authors suggest that white patients could be more likely to request these services, or clinicians might be more likely to offer them, perhaps because of implicit biases. Black patients are less likely than white patients to receive certain cancer screenings that are considered high-value, such as age-appropriate colonoscopies and mammograms, so it makes sense that they are also less likely to receive low-value ones. 

This study provides valuable information on the intersections between health equity and value, but there’s still much more to learn. The authors suggest that health systems measure use of low-value services stratified by racial group and sex, to identify potential disparities. “These results invite further exploration of differential access by race to routine, high value primary care, patient-clinician concordance, and trust,” they write.

The post Study finds racial disparities in low-value care, even within the same health systems appeared first on Lown Institute.

The issue of overuse 

Overdiagnosis and overtreatment are issues that have troubled the healthcare system for years. The provision of care that is ineffective and even harmful to patients can lead to physical injury, wasteful spending, and emotional distress for patients. 

For example, certain types of cancer screening have the potential to prevent death from cancer (although most don’t have a proven impact on all-cause mortality). But because these tests are not perfect, they have a high rate of false positives, which lead to overtreatment and other harmful cascade events. According to a UC Davis Health study, 50% of women receiving annual breast cancer screenings for ten years experience at least one false positive mammogram.

But what if we could better target screening and reduce the rate of false positive results, ultimately improving the patient experience? That’s where AI comes in. The speed and accuracy of machine learning models trained on diverse data, coupled with the ability to recognize nuances in mammogram results, hold great promise.

How AI could reduce false positives

AI in the field of breast cancer is expansive; it has been used to read and interpret mammograms, conduct cancer risk assessments, and identify nuanced features of breast tissue like density. These tactics are primarily achieved through a form of AI called radiomics. Based on the idea that there are physical characteristics of the disease process that cannot be captured by the human eye, radiomics relies on mathematical formulas to quantify the intensity, shape, and texture of an image. 

Here’s a glimpse into what some researchers have found on the potential for AI to target screening and reduce false positives:

• 250 mammograms conducted in 2013 were re-evaluated using an AI-based, computer-aided detection (CAD) software, reviewing marked regions of interest for sensitivity and specificity rates as well as false-positive marks per image (FPPI). The findings revealed an overall 69% reduction in FPPI using the AI-based CAD versus the traditional CAD.
• An AI system applied to 9,581 mammograms was used to categorize cancer risk scores on a scale of one to ten, with higher scores indicating a greater likelihood of malignancy. For those scored between three and ten, the AI system revealed the potential to have eliminated nearly 2,000 exams and ten false positives.
• Houston Methodist Hospital created iBRISK, a breast cancer risk assessment tool supported by AI. iBRISK takes into account patients’ demographics and medical history before recommending future diagnostic testing. By targeting screening toward patients with the highest risk, we improve the chance of benefit from screening.

How AI could make overuse worse

However, even if AI were to reduce false positives in breast cancer screening, it could still exacerbate overdiagnosis. Screening tools are often judged by how successfully they identify cancers, but it’s hard to determine whether a cancer that’s detected is one that is actually harmful. In fact, pathologists often disagree amongst themselves about whether a small abnormality of cells constitutes a cancer diagnosis or not. 

As Dr. Adewole Adamson and Dr. Gil Welch wrote in a 2019 NEJM Perspective, “Diagnoses of early-stage cancer made using machine-learning algorithms will undoubtedly be more consistent and more replicable than those based on human interpretation. But they won’t necessarily be closer to the truth — that is, algorithms may not be any better than humans at determining which tumors are destined to cause symptoms or death.” Instead, these AI tools may just be able to overdiagnose at a faster rate. 

If AI detects more low-grade, non-invasive “cancers” like ductal carcinoma in situ (DCIS), that could lead to overtreatment. New research shows that outcome is likely. A 2023 study in The Lancet Digital Health examined how AI could be used to assist screening in a national breast cancer screening program to reduce workload for radiologists. In a randomized controlled trial, researchers found that screening supported by AI had a similar cancer detection rate to a standard reading by radiologists. However, AI-supported screening also detected nearly double the number of DCIS cases than standard screening. 

But it’s not all bad news! There are ways that AI could help reduce the problem of overdiagnosis. Adamson and Welch suggest that AI algorithms could be trained to uncover “gray area” cases where pathologists disagree about whether an abnormality is cancer, to increase attention to diagnostic uncertainty, improve efficiency, and help researchers understand more about how cancers develop.  

Researchers from the Australian Centre for Health Engagement, Evidence and Values suggest that autopsy studies could provide data to train AI to distinguish between fatal and non-fatal cancers, given that many older people die “with” cancer but not “because” of cancer. Data from trials of “watchful waiting” could also potentially be used to build AI algorithms that detect overdiagnosis. 

Moving towards an equitable future

In a virtual event hosted by Stat News and the Gordon and Betty Moore Foundation, Putting AI to the Test featured AI experts Dr. Vincent Liu (Kaiser Permanente), Dr. Melissa McCradden (Hospital for Sick Children), and Dr. Ziad Obermeyer (UC Berkeley) who spoke to what’s next for an ethical and equitable future of AI. 

Ultimately, the panelists reminded us that AI is merely a tool to make doctors better at what they do. AI is not a one-size-fits-all approach and it must be tailored to the settings in which it is used. For the world of breast cancer screening, this means prioritizing the creation of AI systems that perform not only well, but equitably by targeting the known racial and ethnic disparities in breast cancer. To avoid the common problem of perpetuating health disparities through AI, we should incorporate patient experiences and outcomes into algorithmic formation, said Dr. Obermeyer.

“Rather than training the AI to replicate human judgment, train the algorithm to replicate patient outcomes and longitudinal things that happen to that patient that are sources of truth.”

Dr. Ziad Obermeyer, “Putting AI to the Test”

Pew Research’s finding that three-quarters of Americans are concerned about physicians adopting AI into their practice too quickly without fully assessing its potential risks underscores the importance of putting AI to the test before adopting it on a global scale. The provision of grants – such as the one received by Rensselaer Polytechnic Institute (RPI) and Albany Medical College to study the use of AI in breast cancer treatment – and ongoing evaluation of screening and diagnostic tools ahead of their widespread adoption will be key to AI’s success. Moving forward, we must work to address the issues at the core of overdiagnosis and overtreatment, including improper risk assessment, ill-informed screening referrals, and breakdowns in communication between patients and physicians. Applying AI to a broken system will only exacerbate the issue, not solve it.

The post How AI-powered cancer screening could impact overuse appeared first on Lown Institute.

Does screening give us more years of life?

A new meta-analysis in JAMA Internal Medicine provides more information on the relationship between cancer screening and overall lifespan. Researchers from the University of Oslo looked at 18 long-term randomized clinical trials of six common cancer screening tests (Mammography screening for breast cancer; colonoscopy, sigmoidoscopy, or fecal occult blood testing for colorectal cancer; CT screening for lung cancer in smokers and former smokers; and prostate-specific antigen testing for prostate cancer) to see the impact of screening on how long participants lived. All trials had at least ten years of follow-up. In total the trials included 2.1 million participants.

The authors found that one test, sigmoidoscopy for colorectal cancer, showed a significant increase in life gained of 110 days for people who were screened, compared to those who were not screened. But the other screening tests did not show a statistically significant increase in lifespan in the screening group.

The limitations of this analysis are important to note. The meta-analysis included only randomized controlled trials that measured all-cause mortality (many do not, because it requires more participants) and had long enough follow-up, which limited it to 18 studies (a small fraction of the cancer screening trials that exist). One of the included mammogram trials was recently found to be inaccurately randomized, Yale School of Medicine Dr. F. Perry Wilson pointed out in a Medscape op-ed. However, one could argue the small sample of trials included shows a limitation of the research landscape itself– ideally we would have many more long-term randomized trials of cancer screening that are powered to detect all-cause mortality.

Why life-years gained from screening is low

How could it be that many cancer screenings don’t have an impact on overall lifespan? While screenings prevent some deaths from cancer, they don’t prevent all. For example, with no screening, an estimated 28 women out of 1,000 will die from breast cancer over their lifetime; with biennial screening starting at age 50, 21 will still die from breast cancer.

At the same time, cancer screenings have associated harms such as false positive results, overdiagnosis, and overtreatment (not to mention the financial cost of all these cascade events). Of 1,000 women who get screened for breast cancer every other year starting at age 50, we can expect there to be 1,021 false positive results, 148 unnecessary biopsies, and 10 cases of overdiagnosis. Some of these will be relatively innocuous, but others could lead to increased stress, harm from radiation, and even financial toxicity, all which can impact overall health. It could be that the benefits of screening that some people receive get washed out by the harms that others experience when looking at screening on a population level.

There are other reasons why cancer screenings may not have as large an impact on overall mortality. Treatments for some cancers have improved over the years, making it more likely that someone diagnosed with cancer will survive. And other causes of death such as heart disease have a greater impact on overall mortality than cancer.

What does this mean for me?

Do the results of this study mean that no one should get screened for cancer? The short answer is no. Even if screening has not been shown to increase lifespan on average, there will still be patients that benefit from screening and avoid early death from cancer. However, before undergoing screening, people should be aware that they are much more likely to experience a false positive or overtreatment than they are to have their life saved by the screening. Given that knowledge, many people will still choose screening, but some may not.

This risk-benefit calculation also changes based on one’s particular risk factors for cancer, as well as the type of screening. For example, colonoscopy is not just a screening test but a preventive measure as polyps are removed.

Researchers are also working on ways to increase the benefits and reduce harms from screening, so that the decision to get screened is easier. For example, Dr. Laura Esserman, a surgeon and breast cancer oncologist at the University of California San Francisco and Dr. Scott Eggener, a surgeon and urologic oncologist at the University of Chicago, suggest in a New York Times op-ed that we use personalized screening for breast cancer “based on such factors as age, genetics, lifestyle, health history and breast density.”

Esserman and Eggner recommend that very low-risk diagnoses such as ductal carcinoma in situ or a 6 on the Gleason scale be treated with active surveillance (“watch and wait” strategy”) rather than with aggressive treatment right away. They note that the classification of non-invasive, low-risk diagnoses as “cancer” adds unnecessary pressure to patients to pursue aggressive treatment, and suggest using different terminology other than the “dreaded C-word.”

Can we stop the screening train?

If cancer screening has minimal impact on overall lifespan, why is screening so prevalent? In an accompanying editorial in JAMA Internal Medicine, three of the meta-analysis authors explain that it’s hard to stop screening programs once they’ve become commonplace, because they are a win-win for many stakeholders. Screening increases volume and feeds the healthcare industry engine, creating jobs and boosting the incomes of physicians and the medical device industry. And it’s easier for policymakers to support screening programs rather than take more difficult but potentially impactful actions to prevent cancer such as regulate pollutants and other carcinogens.

Cancer screening is also unique in that there are fewer requirements for implementation compared to drug approvals. “Cancer screening—now also including novel multicancer early detection tests—can be rolled out as full-scale human experiments before basic information about test performance, benefits, harms, and cost-effectiveness is available,” the authors write.

Before we throw all our eggs in the basket for these new “liquid biopsy” tests, we have to be sure that there is a mortality benefit– a feasible ask for a multi-cancer screening test — and that conflicts of interest don’t tip the scales in the decisions to approve or cover these tests.

The post Does cancer screening help us live longer? appeared first on Lown Institute.

Why did the USPSTF depart from their previous recommendation and what are the potential harms and benefits of this change? Let’s take a closer look.

Author’s note: The USPSTF recommendations apply to anyone assigned female at birth, which includes not only women but also trans men and some nonbinary individuals. For the purposes of readability I use the term “women” throughout the blog in reference to all people for which this recommendation applies.

Why did the USPSTF lower the screening age?

The USPSTF guideline change came as a surprise to many health experts, because there have been no new clinical trials of breast cancer screening that would provide a reason for adjusting the guidelines. Rather, the USPSTF noted that the incidence of breast cancer in young women has been steadily rising, which changes the risk-benefit profile of screening. The more likely it is that someone has cancer, the more beneficial screening them for cancer will be.

Yet some healthcare researchers noted that even adjusting predictive models for higher rates of cancer in younger women, the risk-benefit profile still is not very different from the agency’s previous results from 2016.

“Change always happens over time, obviously, as the evidence evolves. At the same time, there needs to be a compelling reason and in the materials here, I don’t see a compelling reason yet. When I looked back at the 2016 modeling studies, the harm-benefit analysis was very similar,” said Ruth Etzioni of the Fred Hutchinson Cancer Center, in STAT.

The USPSTF also wanted to acknowledge that Black women are diagnosed with breast cancer at later stages and face a higher rate of mortality from breast cancer than other racial groups; thus, an earlier start date of screening for these patients could save lives and reduce racial disparities in breast cancer outcomes. But although the USPSTF ran new models exploring the benefits and harms of screening in Black women, they stopped short of recommending earlier screening for Black women specifically.

Benefits and harms of earlier screening

What are the actual numbers that drove the USPSTF’s new recommendation? The USPSTF modeling report presents a myriad of scenarios that estimate the rates at which breast cancer screening starting at different ages result in certain benefits and harms.

At a baseline, with no screening, an estimated 28 women out of 1000 will die from breast cancer over their lifetime. For every 1000 women screened every two years from age 40-74, about eight will avoid death from breast cancer over their lifetime. However, there will also be 1540 false positives, 210 unnecessary biopsies, and 12 cases of overdiagnosis (when someone is diagnosed and treated for a cancer that never would have harmed them). And among the 1000 women screened, 20 will still die from breast cancer. It’s worth mentioning that although screening prevents death from breast cancer, there is no evidence that screening mammograms reduce overall mortality (death from any cause).

Essentially this means that an individual at average risk getting screened is more likely to be treated for a cancer that never would have harmed them than they are to avoid death from breast cancer. They are more than twice as likely to die of breast cancer anyway than to catch and successfully treat a dangerous cancer. And they are far more likely to have undergo a biopsy unnecessarily or receive a false positive result than avoid death from breast cancer.

What’s the difference from the previous recommendation? Lowering the starting screening age from 50 to 40 avoids one more death from breast cancer out of 1000 women screened, but results in 519 more false positives, 62 more unnecessary biopsies, and two more cases of overdiagnosis.

It’s interesting to look at scenarios that the USPSTF did not adopt in their draft recommendation — for example, screening every year instead of every other year. Screening 1000 women every year from age 40-74 would avoid three more deaths from breast cancer compared screening every other year. However, it would result in 883 more false positive results, 71 more unnecessary biopsies, and seven more cases of overdiagnosis (see table below).

Similarly, continuing annual screening until age 79 would avoid four more breast cancer deaths per 1000 screened, compared to screening every other year until age 74. But it would increase the number of false positives by more than 1000 and the number of overdiagnosed cases by more than ten per 1000 screened.

Examining the tradeoffs

These comparisons bring up important questions of what tradeoffs we are willing to make to catch cancer early. Many would say that avoiding one more death from breast cancer is worth the harms to hundreds more from overtreatment, but it’s important to recognize that these downsides exist and to acknowledge that there is a line at which the harms outweigh the benefits.

For example, in a 2019 tongue-in-cheek article, doctors and researchers Myung Kim, Go Nishikawa, and Vinay Prasad pointed out that if we only care about the benefits of preventive care and not the harms, we would give everyone a double mastectomy to prevent all cases of breast cancer before they start. The ridiculousness of this proposal shows how we must take into account the harms of screening alongside the benefits.

It’s also important to understand that a decision made for the purposes of population health may not be acceptable to every individual. With this recommendation, the USPSTF is essentially saying that the tradeoffs of 500 more false positives, 60 biopsies, and 2 overdiagnosis cases are worth it to avoid an additional death per 1000 people from breast cancer. Across a population of millions, that’s hundreds of thousands of deaths that could be avoided. Many patients will find the risks of screening acceptable on an individual level as well. But members of the USPSTF task force also say that decisions about when to start screening are personal, and that there is always room for a discussion with your doctor. Women at average risk of breast cancer considering screening should be informed to make the best decision for them.

The public health impact of earlier screening

While models can help us understand the theoretical benefits and harms of screening, things are very different in the real world. The guideline assumes that everyone has equal access to screening and to healthcare in general, which is simply not true. About a quarter of women age 50-64 have not had a mammogram in the past two years, and 55% of uninsured women in this age group haven’t received a mammogram in the past two years. About 5% of women age 50-74 have never had a mammogram.

It’s unclear how lowering the screening age to 40 will increase access among women who don’t currently have access to care despite already being eligible for screening. In fact, as we pointed out in a previous blog on colon cancer screening, lowering the screening age could increase disparities in access if more healthy young people rush to get screened and take up the available slots.

The USPSTF is right in calling for more research to evaluate the benefits and harms of screening particularly for Black women, who are at greater risk of breast cancer mortality. However, we need far more action to reduce racial disparities than just lowering the screening age. Disparities in breast cancer deaths are a result of structural racism, which means we should be dedicating more energy to ensuring regular access to medical services, improving guideline-concordant cancer care, and addressing the social drivers of health in communities of color that may contribute to cancer like environmental conditions.

The post Earlier screening for breast cancer: Benefits and harms appeared first on Lown Institute.

Early detection is important, but how do we screen responsibly? 

It makes logical sense that catching cancer early before it spreads will reduce cancer deaths. In fact, I have personal experience with skin cancer. At 19, I had a small melanoma mole removed from my shoulder that I could barely see and would never have identified as problematic. While I’m at high risk of melanoma due to my pale skin and family history, it was not found at a dermatologist’s office. It was caught by the women’s health provider at my university during an unrelated check-up. My mole was likely melanoma in situ and presented no immediate threat to mortality, though I’m grateful to have the peace of mind that it’s gone. Essentially, my early melanoma was excised before it could have gotten worse. 

However, my experience doesn’t apply to everyone, nor does it indicate that universal screening is a beneficial policy. As Dr. Adamson notes in his editorial, melanoma screening so far has resulted in increased rates of melanoma diagnosis, without having a subsequent impact on mortality from melanoma– a classic sign of overdiagnosis. This shows that we need more research around the “natural history” of melanoma — how a cancer progresses (or does not progress) from early to late-stage, so we can better identify early-stage cancers that are dangerous. 

We can also do a better job of identifying those at higher risk of skin cancer and establishing screening guidelines for these patients. Logistically, we don’t have enough dermatologists to routinely screen everyone, so we need some way to triage. Currently, there is no consensus among dermatologists about who to screen and when, making it harder to study whether or not screening is working. 

Another consideration is the type of skin cancer. Patients with darker skin are less likely to develop UV-based skin cancer but more likely to develop acral lentiginous melanoma. Screening guidelines should match this knowledge accordingly. 

Overdiagnosis has harm too. Targeting our interventions can help.

If the benefits of screening are minimal, we have to be aware of the risks to better evaluate the balance. Screening for cancer seems may seem like a risk-free endeavor, but there are many potential physical, psychological, and financial risks. While infrequently discussed, there is the potential for harm through diagnosis. A diagnosis of cancer is scary and intimidating and can cause psychological distress. It opens the door to further cascade events such as biopsies and surgeries, which have risks of complications. And apart from the general requirements of taking time off and physically showing up to the dermatologist’s office, screening costs money. Dr. Adamson notes the need for more research on this topic to fully evaluate the benefits and harms. “In melanoma, the potential harms of screening, including those caused by overdiagnosis, have not yet been fully elucidated,” he writes.

With the current evidence, it appears that we should be moving towards a more targeted screening approach. We obviously don’t want to abandon all preventative measures – after all, the goal is early detection to save lives – but striking the balance between prevention and overdiagnosis is nuanced. More research is needed to better inform best practices for skin cancer screening and determine its value across the population. As Dr. Adamson says, “Patients deserve better.”

The post Balancing prevention and overdiagnosis in skin cancer screening appeared first on Lown Institute.

Because of this emphasis on colonoscopy, Americans may be surprised to learn that other countries consider routine colonoscopy screening for cancer to be overuse. In Australia and Ontario, Canada, for example, medical societies recommend screening using less invasive tools like a FIT test or FOTB first, and only doing a subsequent colonoscopy if the results are positive.

Cross-country differences

What explains this discrepancy in screening guidelines ? In a recent editorial in BMJ Quality and Safety, Kelsey Chalmers, Shannon Brownlee, and Vikas Saini of the Lown Institute explore why routine colonoscopy is recommended in the US as a first-line screening method, but not in other countries.

“The contrast between US standards for colorectal cancer screening and those of other countries provides insight into different nations’ consideration of costs and outcomes.”

Chalmers, Brownlee, and Saini, BMJ Quality & Safety 2023

A key issue is the lack of high-quality evidence demonstrating the superiority of colonoscopy compared to other screening methods. Although colonoscopy has the benefit of being a more sensitive measure and allows for the ability to remove polyps, the onerous preparation requirements and higher risk of complications compared to other screening methods may counteract the benefits. For example, the recent NORDICC randomized trial of colonoscopy screening in Europe found no evidence that participants invited for colonoscopy had improved mortality compared with a control group without screening, in large part due to low take-up from the invited group.

Without good evidence that colonoscopy is better than other types of screening, the values and priorities of policymakers in each country play a larger role in crafting recommendations. One of these most important priorities is cost. Colonoscopy is by far the highest-cost screening method for colon cancer, so it makes sense that other screening methods are preferred in countries that care about keeping healthcare costs low.

“Countries with limited healthcare resources or single-payer systems may be readier than the USA to define overuse based on the societal harms of delivering high-cost, high- profit care to some, while others go without beneficial care.”

Chalmers, Brownlee, and Saini, BMJ Quality & Safety 2023

Although the US is an outlier when it comes to healthcare costs, these costs are fragmented between governments, employers, insurance companies, and patients. That makes reducing low-value care a much harder sell to patients, because the cost savings don’t necessarily go into their pockets. In fact, until fairly recently, patients were charged out of pocket to get a colonoscopy after a positive screening result, creating a disincentive for going the route recommended in other countries.

Overuse still widespread

However, even in the US where colonoscopy is recommended for screening, there is still substantial overuse. Previous research on colonoscopy estimates that 17-25% of screening colonoscopies are done on patients at ages above or below national guidelines, costing $3 billion per year.

In BMJ Quality and Safety this week, researchers at the VA Ann Arbor Center for Clinical Management Research share how they developed a new metric to identify screening colonoscopy overuse using ICD-10 diagnostic codes. They applied this metric to data from the Veteran’s Affairs health system and found that 24.5% of screening colonoscopies met the criteria for probable or possible overuse.

This new metric will be incredibly useful for American researchers to identify colonoscopy overuse nationwide (keep an eye out for colonoscopy overuse in the Lown Hospitals Index soon!)

The post Why are screening colonoscopy recommendations different across countries? appeared first on Lown Institute.

The FDA guidelines contain suggested language for notifications about breast density:

“Breast tissue can be either dense or not dense. Dense tissue makes it harder to find breast cancer on a mammogram and also raises the risk of developing breast cancer. Your breast
tissue is dense. In some people with dense tissue, other imaging tests in addition to a mammogram may help find cancers. Talk to your healthcare provider about breast density, risks for breast cancer, and your individual situation.”

There are a lot of issues here. Breast density is a risk factor for developing breast cancer, but it is one of many risk factors. There may be women at much higher risk for cancer because of their age, family history, alcohol use, etc. who do not have dense breasts, while others with dense breasts are at an overall lower risk.

While the FDA keeps it simple, saying that “breast tissue can be either dense or not dense,” it’s not actually that clear cut. The American College of Obstetricians and Gynecologists (ACOG) pointed out in a comment to the FDA that “there is no standardized method for assessing breast density,” so whether or not a patient gets a classification of dense breasts depends on the opinion of the radiologist who reads the test. Breast density can also change over time, so one notification may not be true years later.

The notification language also pushes patients toward additional imaging tests by saying they “may help find cancer,” but doesn’t mention that these imaging tests also increase the risk of false positives and other cascade events. There are no trials showing that supplemental cancer screening with MRI or ultrasound improves mortality or morbidity for women with dense breasts.

Previous research on breast density notification policies show a potential for overuse. Studies of state policies notifying patients about breast density show that they strongly increase the likelihood of patients discussing supplemental screening with their doctor, and modestly increase the likelihood of supplemental screening and breast biopsies. If supplemental screening worked as intended and caught dangerous cancers early, one would expect reductions in rates of late-stage cancer in states with these breast density notifications. However, a 2017 study of these policies did not show a difference in rates of regional or metastatic disease between states with and without notifications.

Because of their high false positive rates and lack of proven benefit, ACOG does not recommend routine use of other scans like ultrasound or MRI to screen for breast cancer in women with dense breasts as their only risk factor. The US Preventive Services Task Force, an independent panel that makes evidence-based recommendations about preventive services, concluded that evidence was insufficient to recommend additional screening for women with dense breasts.

This puts doctors in a tough spot, because when patients will come to them seeking information on what to do, they will either have to tell patients to do nothing (which is likely frustrating and unsatisfying for patients) or tell them to get supplemental screening (which is not universally recommended and could put them at harm for cascade events).

“Primary care physicians in states that have passed such laws often feel poorly prepared to counsel women regarding what action to take, if any, for a woman with dense breasts and a normal mammogram.”

Dr. Kenneth Lin, Medscape

There are also significant cost implications of this policy, both on an individual and system level. An estimated 40-50% of women in the US have dense breasts, so if all of them underwent additional supplemental screening, that could have a serious impact on healthcare spending. While screening mammograms are covered under most insurance, additional MRIs can add out-of-pocket costs for patients and biopsies even more.

We know at least one group for whom this guideline change is a win-win: the makers of imaging devices, who have funded groups for years that advocate for breast density notifications.

The post New FDA guidelines on breast density notifications and implications for overuse appeared first on Lown Institute.

Low-value imaging for back pain

One of the most commonly overused tests is imaging for low back pain. Back pain is extremely prevalent– impacting an estimated 577 million people worldwide– and is the leading global cause of disability. In the face of so much pain and discomfort, it’s tempting for doctors and patients to try tests and procedures, even when they are not proven to improve outcomes. Imaging for uncomplicated low back pain tops the list of most commonly overused services in the Medicare population, impacting 9% of beneficiaries and wasting $263 million in 2018.

Unfortunately, the cost of the initial test is only the beginning. In a recent article in the Journal of General Internal Medicine, Dr. Ishani Ganguli at the Harvard Medical School and colleagues examined the prevalence and cost of cascade events resulting from imaging overuse from back pain.

They looked at about 6,000 people within Massachusetts’ Blue Cross Blue Shield plan who received an inappropriate X-ray or MRI for low back pain from 2017-2019. Compared to others who were eligible for imaging but who didn’t receive it, those who had low-value imaging were more likely to have follow-up office visits, lab tests, and surgeries.

Those who had a low-value MRI were 14 percentage points more likely to have a cascade event compared to those with no initial imaging.

After six months, those who had an MRI were 14 percentage points more likely to have a cascade event, and those who had an X-ray were 9 percentage points more likely, compared to those with no initial imaging. Results were adjusted for differences in patient age, risk, and other characteristics.

The cost of cascades

Total costs and out-of-pocket costs were both much higher for patients that had low-value imaging. Among those who had a low-value MRI, total spending on cascades was $1050 greater and out-of-pocket cost was $298 greater six months after the MRI, compared to those who did not get any imaging.

Patients with a high-deductible health plan who received low-value imaging spent $513 more out-of-pocket on cascade events than patients who did not receive imaging.

Patients with a high-deductible health plan who received low-value imaging were exposed to even higher costs, spending $513 more out-of-pocket on cascade events than patients who did not receive imaging.

Your experience may vary…

The Journal of General Internal Medicine shows how costs can skyrocket due to care cascades, even for patients with commercial insurance. However, another recent study demonstrates that the cost of imaging can vary extremely based on which insurer you have and to which hospital you go.

In the journal Radiology, John (Xuefeng) Jiang from the Department of Accounting and Information Systems at Michigan State University and colleagues examined the price variation of common scans at more than 2,000 hospitals. They found that even at the same hospital, prices for thirteen common imaging tests varied by 3.8 times depending on the insurer. Prices for head CT were the most variable, with the highest price being 5.2 times the lowest at the same hospital. The median price of MRI for low-back pain was $1,426, but the price at the same hospital varied nearly five-fold depending on the insurer.

The suffering caused by back pain undoubtedly puts pressure on clinicians to offer unnecessary tests. But both doctors and patients should be aware that these tests given “just to be safe” actually create more harm for patients — both physical and financial.

The post Imaging overuse: What care cascades may cost you appeared first on Lown Institute.

What did this trial measure?

While there have been many observational studies of colonoscopy (comparing outcomes among people who received a colonoscopy and who did not), there has never been a randomized controlled trial. That’s important because there may be differences between patients that get colonoscopies and those who do not that cannot be adjusted for in observational studies (such as access to healthcare, income, etc). By randomizing people to either get an invitation to screen or not, we can better evaluate the effectiveness of the program in a real-world setting.

Before looking at the results, it’s important to establish what this trial was designed to measure. In the NordICC trial, about 85,000 people in Poland, Norway, Sweden, and the Netherlands were randomized to either get an invitation to have a colonoscopy or not get an invitation. Cancer incidence, mortality from colon cancer, and all-cause mortality were compared after ten years.

The NordICC trial is a test of a public health intervention– if we invite people to get screened, what will be the impact on colon cancer incidence and mortality? This is why the study authors call it a “pragmatic” randomized controlled trial; they designed it to find out what will happen in the real world if they implement this screening program. This means the study takes into account aspects of screening programs that impact the outcome of the program, like how many people are able and willing to get screened. For example, a screening test that is more effective but few people want to do could have the same effect on population health as a test that is less effective but easier for more people to do.

What were the results?

The results of what are called the “intention to screen” analysis (comparing the control and intervention groups) were disappointing. While the group invited to have a colonoscopy screening had lower rates of cancer incidence (relative risk reduction of 18%, absolute rate reduction from 1.2% to 0.98%), their rates of mortality from colon cancer and all-cause mortality were not significantly different from the control group after ten years. To prevent one case of colon cancer, 455 people would need to be invited to undergo screening.

In previous observational studies of colonoscopy, the test reduced rates of colon cancer and cancer-specific mortality by 60%, so these findings were sobering for many doctors. “I think we were all expecting colonoscopy to do better,” said Samir Gupta, a gastroenterologist at the University of California, San Diego and the VA, in an interview with StatNews. The study authors speculate that there wasn’t as large a decrease as expected because colon cancer rates have declined in the trial countries and treatments for colon cancer have improved.

However, the study also showed some positive aspects of colonoscopy. The invitation to screen with colonoscopy did reduce rates of colon cancer. The rate of major bleeding from the procedure was very low (0.13%) and there were no bowel perforations (a serious adverse event). Previous studies have showed a bowel perforation rate of 3 per 10,000 procedures and an incident cardiovascular event rate of 120 per 10,000 procedures, so the lack of adverse events in this study was remarkable. It’s possible that the doctors in this trial took more care than usual, or that these procedures have become safer over time.

Low uptake problem

The biggest criticism of the study was the relatively low number of people invited to screen who went through with the procedure. Fewer than half (42%) of those invited to screen actually received a colonoscopy. In the US, about 72% of adults 50-75 are up to date with colon cancer screening. However, looking only at the study results from Norway, which had a take-up rate of 61%, the results for colon cancer incidence were similar to the full sample (relative risk reduction of 24%, absolute rate reduction from 1.57% to 1.2% over ten years).

Given the low take-up rate, the authors also conducted a “per-protocol analysis” to estimate the rate of cancer incidence and mortality if everyone who was invited to get a colonoscopy actually got one. They found that rates of colon cancer would have been lower (relative risk reduction 31%, absolute reduction 1.22% to 0.84%) as well as rates of colon cancer mortality (relative risk reduction 50%, absolute reduction 0.30% to 0.15%).

A 50% decrease in death from colon cancer sounds great, but it’s important to point out the context. The absolute risk of colon cancer death only decreased by 0.15%, compared to an 11% risk of dying from any cause. We also don’t know for sure whether this reduction is due to the impact of the procedure itself or to underlying characteristics among people who get colonoscopies.

The fact that uptake was lower than expected is itself an important finding, because it demonstrates a key limitation of colonoscopy screening programs: it doesn’t matter how much colonoscopies reduces deaths if nobody wants to get one! Using the “intention to treat” analysis, we see how well the colonoscopy screening program works in the real world, taking into account the barriers that people face to getting the screening. That allows us to better compare colonoscopies to other types of screening that may be easier for patients to undergo.

Raising questions

The NordICC trial shows that colonoscopy screening programs may not be as effective as expected, especially in comparison to other types of colon cancer screening. For example, the study authors note that the results put colonoscopy on par with flexible sigmoidoscopy, a cheaper colon cancer screening test that does not require sedation.

The results raise several questions for public health leaders to consider:

• Given the high cost of colonoscopy ($2000 on average for privately insured patients and as much as $10,000 in some cases), how much should we be investing in colonoscopy compared to other colon cancer screening methods or interventions?
• Are there ways we can better target those at higher risk of colon cancer with types of screening that have higher uptake rates than colonoscopy?
• How much are we spending on colonoscopy screening for younger low-risk people that we could use to increase awareness of early colon cancer symptoms or toward addressing the root causes of colon cancer?

The NordICC trial prompts us to reconsider colonoscopy as the “gold standard,” and to rethink how our healthcare system invests in prevention and screening around colon cancer.

The post Real-world test of colonoscopy screening program shows cancer reduction but no mortality benefit appeared first on Lown Institute.

The promise of blood testing

By now most Americans are familiar with Elizabeth Holmes and her failed company Theranos. The company promised that with only a small drop of blood, they could test for and diagnose a large selection of diseases. It was a lie. It was exposed years later, causing a national scandal and damaging the lives of trusting patients along the way.  

Other companies are still attempting to do what Theranos could not – develop a way to test blood for a spectrum of illnesses ranging from genetic abnormalities to cancer. There are various avenues to approach this task, whether it be screening for related antigens or counting blood cells. 

Multi-cancer screening tests, sometimes called liquid biopsies, are growing in popularity. These screening tests use a small amount of blood to look for markers for a variety of cancers. Currently the FDA has approved these tests for people with advanced ovarian, lung, breast, or prostate cancer to detect the recurrence of cancer and to guide therapy. Companies looking to expand the use into screening – and expand their profits – are pushing for wider adoption of blood screening tests. Is this a good idea?

The delicate balance between early detection and overuse

Our bodies have abnormalities all the time. Our cells are constantly dividing and dying as we develop with age. Cellular reproduction is bound to have errors and cancer cells will grow – but with a healthy human body, other cells involved in the immune system will quickly take care of cancer cells without the human even noticing. Sometimes they don’t, of course, and the cancerous cells multiply into benign or malignant tumors. This is where the fine line between early detection and overuse appears.

Say a patient goes in for a standard appointment, is offered a screening test, and receives an abnormal finding on their blood test –  but it’s not definitive. They have no symptoms and would not have known anything was abnormal had they not gotten the screening test. A good physician would consult the patient, likely encouraging them to get more tests to try and narrow down the diagnosis and prognosis. The patient goes for another test; again, it’s not definitive. They’ve now spent time and money on two tests that have neither informed nor improved their health. They can keep going for tests and procedures to figure out the source, or they can stop. The wisdom in their decision depends on whether or not they truly do have something harmful going on in their bodies. Stress about their unknown health status is now latched deeply to the back of their minds. 

Did these tests help the patient? The answer is unclear, and we won’t be able to step back and accurately look at this from a systemic level until we see results from the randomized controlled trials currently in progress. However, by then it might already be considered the standard of care.

Profits first and foremost

Medical testing is an expensive, lucrative business. If we’re considering funding blood screening tests via Medicare, we need to consider the costs and the benefits of this choice. Susanna Quinn, an ovarian cancer survivor, recently published a piece in the Daily Beast urging Congress to sign the Medicare Multi-Cancer Early Detection Screening Coverage Act. She argues that screening tests are a necessity, using her personal experience to elucidate the agony of going through and surviving cancer treatment.

While Quinn’s story is moving, it’s important to view her piece in the context of her potential financial conflicts. Quinn sits on the board of the Prevent Cancer Foundation, a nonprofit dedicated to early detection that receives extensive funding from pharma and device companies — including more than $1 million from Amgen, Astrazeneca, Genentech, Gilead, Merck, and Pfizer in 2020.

If Medicare begins paying for universal screening, those investments will pay out a large return on the taxpayer dime. It’s not that these efforts are not in good-faith, but there are vested financial interests pushing for early detection screening for everyone. H. Gilbert Welch, an academic physician and cancer researcher, broke down his concerns in an op-ed for Stat News.

The only thing that is clear to me is that this will cost a lot of money. That’s another thing the companies and their investors are banking on. The Galleri test costs $949 a pop and is recommended every year in those 50 and older. With 100 million Americans in this age group, that’s about $100 billion a year — 15 times the budget of the Centers for Disease Control and Prevention. And that’s not even counting the cost of all the subsequent testing and treatment that will invariably follow.

H. Gilbert Welch, STAT News

Early detection can save lives. The potential problem with preemptive, one-stop-shop blood testing is when it “diagnoses” variability in the human body. Cascade events are well-documented in the medical field and pose a significant threat to those using cancer screening blood tests. Preventative care IS public health, but it should be weighed against the sworn oath to first do no harm. The fine line between early detection and overuse is a tough one to walk. 

The post Should Medicare cover blood tests for cancer screening? appeared first on Lown Institute.