News

Ten-year dementia risk charts that combine cardiovascular and genetic risk factors have been developed by Danish researchers and graphically underline the impact of modifiable lifestyle factors, even in the face of increased genetic risk.

The research was presented at the European Atherosclerosis Society (EAS) 2020 Virtual Congress, held online this year due to the COVID-19 pandemic, and simultaneously published October 7 in the European Heart Journal.

Ida Juul Rasmussen, MD, PhD, Department of Clinical Biochemistry, Copenhagen University Hospital Rigshospitalet, combined data from more than 60,000 patients — more than 2000 with dementia — with a median follow-up of 10 years.

By combining genetic risk factors with gender and age, as well as lifestyle factors — such as diabetes, smoking, and educational level — they were able to develop a series of risk charts similar to those for cardiovascular risk from the Framingham study.

This showed that a women at least 80 years of age with the highest genetic risk factors who had diabetes, smoked, and had a low educational level would have a 66% risk for dementia over 10 years; this would drop to 45% if she did not smoke or have diabetes and had a higher educational level.

With this pattern repeated across age groups and in both men and women, Rasmussen said that “the overall message from this study is that a healthy cardiovascular risk lifestyle attenuates genetic susceptibility for dementia.”

She added that “by combining genetic risk factors and modifiable risk factors, we are indeed able to identify the individuals at the highest risk of a future dementia disease that would benefit the most” from an intensive lifestyle intervention.

Ralph L. Sacco, MD, professor and chair of neurology and executive director at the Evelyn F. McKnight Brain Institute, Leonard Miller School of Medicine, University of Miami, told theheart.org | Medscape Cardiology that an “important message” of the study is that lifestyle factors still have an appreciable impact on the risk for dementia, even in those at high genetic risk.

“I think the novelty here is the combination of cardiovascular lifestyle factors and the genetics,” Sacco said.

“In the past, there are numerous scores that have been developed on the lifestyle side, some developed on the genetic side, but this group combined both, and I think that makes it an important and well-done comprehensive risk model.”

However, Sacco noted that there were some limitations to the study, saying that the way in which some of the risk factors were defined “is pretty, I would say, crude,” citing, for example, the threshold for hypertension of 140/90 mm Hg blood pressure or medications, or 4 hours a week for high physical activity.

He also believes that the concept of the tables may not fit with the way in which, “nowadays, everybody works with an app and a portable device,” allowing them to enter continuous measures rather than rely on “categorical definitions.”

Sacco added that the very idea of using risk stratification to assign individuals to lifestyle interventions to lower their risk for dementia is a hotly debated issue.

“The high-risk approach means you screen people, and then those at highest risk you then intervene,” he explained. “However, some of the interventions here are very common and probably would have benefit across the whole population. So the other approach is the population-based approach, where everybody should exercise, not smoke, get their blood sugar under control. And those are such ubiquitous, general-population messages, I’m not sure you need to define people at high risk to promote those important behavioral changes.”

Dementia “Greatest Challenge”

Rasmussen explained that “dementia is one of the greatest challenges for health and social care in the 21st century and…by 2050, it is estimated that more than 130 million individuals will be living with dementia worldwide.”

She said that one of the main issues is that much of the underlying biology remains unknown. Consequently, “we don’t have any easily assessable biomarkers and there are no curative treatments.”

A report from the Lancet Commission on dementia prevention, intervention, and care nevertheless suggested that up to 40% of dementia cases are preventable, primarily through tackling cardiovascular risk factors.

Rasmussen said that these estimates are based on the results of the Finnish FINGER trial, which showed that, with a 2-year intervention, cognitive function could be improved or maintained in at-risk elderly individuals from the general population.

“However, the intervention was massive,” she said, “and very demanding for the participants and, of course, way too expensive to implement generally in the population. Therefore, we wanted to generate a risk-stratification tool to identify the individuals at the highest risk for a future dementia disease to include in such intensive interventions.”

They gathered data from the Copenhagen General Population Study and the Copenhagen City Heart Study, and were able include 61,664 individuals, of whom 2,158 had dementia. The median age of the participants was 60 years, and they were followed-up for a median of 10 years.

They gathered baseline data on a range of cardiovascular risk factors, including diabetes, hypertension, smoking, physical activity, and alcohol intake. Educational attainment was also determined.

Assays were also used to genotype the patients and determine apolipoprotein E (APOE) status, as well as the total number of risk alleles on genome-wide association study (GWAS).

In women, the modifiable risk factors for all-cause dementia with the highest hazard ratios for all ages were 1.54 (95% CI, 1.22 – 1.93) for diabetes versus no diabetes, 1.17 (95% CI, 1.04 – 1.32) for smoking versus no smoking, and 1.27 (95% CI, 1.13 – 1.42) for low versus high education.

The risk factors with the highest hazard ratios in men for all ages were 1.26 (95% CI, 1.01 – 1.57) for diabetes versus no diabetes, 1.35 (95% CI, 1.18 -1.55) for low versus high physical activity, and 1.38 (95% CI, 1.20 – 1.58) for low versus high education.

Regression analysis was then used generate 10-year absolute risk score stratified by gender, age, the presence or absence of diabetes, low or high educational attainment, smoking or no smoking, APOE genotype, and the number of GWAS risk alleles.

This showed that, for women, the highest 10-year risk of all-cause dementia was highest for those aged 80 to 100 years who had the APOE ε44 genotype, 22 to 31 GWAS risk alleles, diabetes, and low education, and who smoked, at 66%.

Rasmussen highlighted, however, that, for women 80 to 100 years with the APOE ε44 genotype and 22 to 31 GWAS risk alleles but who did not have diabetes, did not smoke, and had higher educational attainment, the 10-year risk for all-cause dementia was much lower, at 45%.

This pattern was repeated across the age groups, with, for example, women aged 70 to 79 years in the highest risk group having a 10-year risk for all-cause dementia of 48%, falling to 30% for those who did not smoke, did not have diabetes, and had higher educational attainment.

In men, the 10-year risk of all-cause dementia for those aged 80 to 100 years and with the APOE ε44 genotype, 22 to 31 GWAS risk alleles, diabetes, low education, and smoking, was 60%.

Again, this fell to 39% in those who did not have diabetes, did not smoke, and had higher educational attainment but were still at genetic risk. Among men 70 to 79 years, the corresponding 10-year risks for all-cause dementia were 42% and 26%.

The team was also able to use the data to develop 20-year risk charts for all-cause dementia, finding that women aged 50 to 60 years with the APOE ε44 genotype and 22 to 31 GWAS risk alleles had a 17% risk for dementia if they had midlife hypertension.

For women in the same age and genetic risk groups but without midlife hypertension, this risk dropped to 12%.

Among men, the highest risk was again in those with midlife hypertension, aged 50 to 60 years and with the APOEε44 genotype and 22 to 31 GWAS risk alleles, at 14%, falling to 10% in those without midlife hypertension.

“Potentially Important Step Forward”

In an editorial accompanying the study, Andrew Sommerlad, MD, PhD, and Naaheed Mukadam, MD, PhD, from the Division of Psychiatry, University College London, say that the development of the dementia risk charts is a “potentially important step forward.”

They argue that “it facilitates the identification of high-risk individuals based on a range of risk factors, which could help to target prevention toward those who need it most,” and therefore “yield greater impact.”

“The pessimism felt by those who assume that dementia is an inevitable consequence of ageing, or that a positive family history of dementia indicates that they will follow the same disease course, is misplaced, according to this study and previous research, which has shown the likely modifiability of dementia risk with positive lifestyle approaches, even in the face of unmodifiable age, genotype, and sex.”

Sommerlad and Mukadam add a note of caution, however, saying that “the evidence for dementia being truly preventable is, however, scarce,” adding: “Only one randomized controlled trial has successfully reduced risk from dementia, with effective management of hypertension.”

In addition, they believe that “improving public understanding of dementia is a key component of prevention,” and, due to dementia’s “economic, social, and individual costs,” it “should be a priority for all those at risk as well as policymakers and clinicians.”

The work was supported by the Danish Medical Research Council, the Research Council at Rigshospitalet, the Danish Alzheimer Research Foundation, the Lundbeck Foundation, and the M.L. Jørgensen & Gunnar Hansen’s Fund. No relevant financial relationships declared.

88th European Atherosclerosis Society (EAS) Virtual Congress 2020. Presented October 7, 2020.

Eur Heart J. Published online October 6, 2020. Abstract

For more from theheart.org | Medscape Cardiology, join us on Twitter and Facebook.

About one in four hospitals would see their penalty status change under the Hospital Readmissions Reduction Program (HRRP) if penalties were determined using “excess days in acute care” (EDAC), a more comprehensive measure of hospital use after discharge, instead of looking only at 30-day readmissions.

The EDAC measure captures all days spent in acute care settings within 30 days of discharge, including emergency department (ED) visits, observation stays, and unplanned readmissions.

In the study, published online October 13 in Annals of Internal Medicine, half of hospitals in the highest-performing group under the more blunt 30-day readmissions measure would fall to a lower-performing group if EDAC were used.

Conversely, a similar number of low-performing hospitals would jump to a higher stratum.

“We know that linking the 30-day readmission measure to penalties under the HRRP has led to intensified efforts to treat patients in the ED or as observation stays,” said lead author Rishi Wadhera, MD, MPP, MPhil Beth Israel Deaconess Medical Center, Boston. “While this makes hospitals’ readmission rates look lower, it’s not clear that these shifts are actually good for patient care.”

Given the $3 billion in penalties assessed by Centers for Medicare & Medicaid Services (CMS) under the HRRP since its implementation in 2012, a switch to EDAC could help pull some struggling hospitals back from the brink.

Of particular concern, said Wadhera, is evidence showing that small rural and less resourced hospitals might be taking the brunt of those penalties.

“Large hospitals have the capacity and infrastructure to treat patients in the ER or place them in observation status, such that these encounters won’t count as readmissions. Small rural hospitals don’t have that capacity and may be unfairly penalized,” he suggested.

This new study compared the impact of the standard 30-day readmissions measure with the EDAC measure in 3173 hospitals that participated in the HRRP in 2019.

Because of ongoing concerns regarding unintended consequences associated with the 30-day readmissions measure, including a possibility of increased mortality among patients admitted with heart failure and pneumonia, there have been numerous calls from the clinician and health policy communities, including a recent one in JAMA from Ashish K. Jha, MD, MPH, Brown University School of Public Health, Providence, Rhode Island, to replace the metric with a more comprehensive measure of hospital use.

EDAC was developed for CMS and has been publicly reported since 2017, but is not included in the HRRP. Like the 30-day readmissions measure, however, it does not fully adjust for risk factors like frailty and medical complexity.

The EDAC measure would diminish the perverse incentive to avoid readmitting patients who come back to the hospital.

“One advantage of the EDAC measure is that it’s comprehensive and less gameable, because it counts all hospital encounters after discharge, including ED visits, observation stays, and/or readmissions. The EDAC measure would diminish the perverse incentive to avoid readmitting patients who come back to the hospital and allow for fairer comparisons of hospital performance,” said Wadhera.

“But at best, switching from 30-day readmissions to the EDAC measure would be a small, incremental step forward, and would not address other major issues like inadequate risk adjustment. The HRRP is a deeply flawed program and it’s time for CMS to move on,” he added.

Value-Based Care Needs a “Course Correction”

“Although the EDAC measure captures the use of acute care hospital resources, both it and the readmissions metric are poorly designed to measure quality of care and patient safety,” writes Saul N. Weingarten MD, MPP, PhD, Tufts University, Boston, in an editorial.

He said the continued use of the readmissions measure as an indicator of quality of care or as a means to determine reimbursement is “empirically suspect.”

In comments to theheart.org | Medscape Cardiology, J. Michael McWilliams, MD, PhD, from Harvard Medical School, Boston, was appreciative of these efforts to better understand the impact of alternative payment models.

“Quality measurement is important for evaluation and monitoring progress, so the authors should be applauded for trying to move the science of measurement forward,” he said.

He’s not sold, however, on the EDAC fix, and thinks that given the “intractable problems” with HRRP, the way forward “is not to try to fix the readmissions measure but rather to abandon the program. We need to start taking about other ways to improve quality and stop expending so much energy on attempts to make pay-for-performance programs work,” he said.

It appears that the CMS might agree. Around the same time the Wadhera report was released, CMS Administrator Seema Verma told a virtual audience that the Center for Medicare & Medicaid Innovation (CMMI) needs a “course correction in model design and portfolio selection if value-based care is to advance.”

Her comments were made during an online event hosted by the CMS Health Care Payment Learning & Action Network, a group of public and private healthcare leaders working to provide thought leadership, strategic direction, and ongoing support to accelerate adoption of alternative payment models (APMs).

In her comments, reported by Home Health Care News, Verma said that of the 54 payment models developed by CMMI since its formation, only five have been shown to offer significant cost savings.

But the fixes, she suggested, revolve around making models more often mandatory and designed in ways that shift the financial risk to participants. There is also a need for better benchmarks to measure value, she said.

“It’s been a very disappointing decade for pay-for-performance programs like the HRRP. CMS wants to believe that linking payment to performance on quality measures has resulted in marked improvements in care, but a large body of evidence suggests otherwise,” said Wadhera.

“Policymakers and front-line clinicians need to work together to create more innovative approaches to motivate improvements in care delivery,” he added.

The authors reported no conflict of interest.

Ann Intern Med. Published online October 13, 2020. Abstract, Editorial