The number of people age 65 and older in the U.S. will almost double between 2012 and 2050, increasing from 43.1 million in 2012 (one in seven Americans) to 83.7 million (one in five Americans). At the same time this large demographic shift is occurring, we are also in the midst of great health care system change, as payments become value-based, and systems focus more on population health. Given the large simultaneous changes, we have an opportunity to pay closer attention to the special needs that come with an aging population and redesign the system to address them.

As we age, our bodies and our priorities change. Our bodies change by becoming more vulnerable to insult and injury, requiring deeper attention to the potential harms of care. We also accumulate multiple chronic conditions, which require complex care plans. And we are at high risk of cognitive decline and dementia, which limit our ability to self-manage and be independent. Our priorities change by focusing on the near-term instead of the long-term benefits of care, such as emphasizing quality of life and avoidance of being a burden to families, rather than lowering the risk of long-term mortality. Our current health care system is not structured to adequately address this new context for older adults.

The strength of the American health care system is its highly skilled and specialized clinicians who are able to rally the best technology to cure disease. Yet at older age, the problem is not curing a single disease, but the need to manage multiple diseases and allow people to live well with their changing health state and declining life expectancy.

Fortunately, what older adults need and want from the health care system is very well aligned with the goals of many health care reform efforts. Older adults need care that is well coordinated, avoids harm, takes place in the community rather than in an institution, and helps them achieve their individual goals for care. There is a direct line between these needs and improving quality with fewer errors and delivering care that is more patient-centered in less costly settings. All of these features are hallmarks of current health care redesign and reform initiatives.

One way to help us capitalize on this alignment between what older adults need and health care reform is to evaluate care through the lens of an aging population. In our new Dartmouth Atlas report, “Our Parents, Ourselves: Healthcare for an Aging Population,” we begin to do just that. We shift the traditional focus of the Atlas from examining variation in content and quantity of care across hospital referral regions, toward considering Medicare as the primary insurer for older adults in the United States and the unique needs of an aging population that must be addressed. Some of the key findings and their relevance for health care reform and policy are outlined below.

Volume Of Care Experienced By Older Adults

Traditionally we have focused on reporting hospital admission or clinic visit rates, but in the new report, we use a measure of “contact days,” which places the use of health care into a framework that is more centered on what people and families experience. Such measures are worth considering as we organize care around people rather than systems. Contact days are calculated as the number of days (0-365) that a person receives a service from Medicare outside of the home, including days in an inpatient setting, clinic, or having tests and procedures. Unlike traditional measures such as hospital admission rates, which report admissions per 1,000, contact days provides a number per individual that patients and families may find more meaningful.

We found that people over 65 in fee-for-service Medicare spend a startling amount of daily life absorbed in interacting with the health care system. In 2012, an average senior in the U.S. spent 17 days a year taking time to receive care; in the highest regions of East Long Island and Manhattan, New York, and Fort Lauderdale, Florida, seniors had 25 contact days per year (Exhibit 1). To put this in context, employers would have real concern if older adults were in full-time employment and absent more than one day a month for health care (or more than two in the highest regions).

Also of importance, we found that inpatient stays only averaged 4.6 days per year, which means a tremendous amount of care occurs outside of inpatient settings, and the burden of transportation or support during visits most likely falls on other family members. For patients with multiple chronic conditions, the number of contact days was nearly twice as high as average, at 33 days and, in some regions, a month and half per year (46 days).

The Role Of Specialists

Policymakers and health system leaders are pouring great effort into positioning primary care clinicians as the nexus for patient centered medical homes, accountable care organizations, and transitional care models. The idea is that primary care providers will be best able to coordinate care and thereby reduce waste and improve quality. Yet in practice, we found over 40 percent of older adults saw a sub-specialist as their predominant provider — the doctor they saw most in the outpatient setting. For example, in areas of Louisiana and Mississippi, nearly 60 percent of older adults saw specialists as their predominant provider.

While it is unclear whether these specialty doctors view themselves as delivering primary care, or whether the patients do, they have the most contact and the greatest number of opportunities to influence care. This measure serves as a caution for health systems and policymakers, as they reorganize around primary care, to ensure that specialists, who often control the flow of older patients, will participate and urge their patients to move toward primary care.

Avoidance Of Services That Provide More Risk Of Harm Than Benefit

Conveying when tests and treatments may be more harmful than beneficial is one of the most challenging conversations to have with older patients and families, especially when some of those tests and treatments have traditionally been seen by both the medical community and the public as important for maintaining health. Screening for early cancer (a cancer that will not be a threat to health for years to come) is one of those examples.

For nearly two decades, urology guidelines have recommended against blood tests (PSA) to screen for prostate cancer when a person’s life expectancy is less than 10 years. Those recommendations have gotten more strident in guidelines over the years. Despite these recommendations, the debate about PSA screening for the general population (and mammography screening as well) has raged publicly for years. Cancer-prevention messages that target younger people can create conflict and confusion for older patients and their families, and this nuance of “likelihood of benefit” for older adults can be easily lost.

Often, we report whether people receive screening but do not report the reverse — whether people avoid screening when it is unlikely to benefit them. In our new report, we found that one in five men older than 75 were screened with a PSA for occult prostate cancer in 2012 (with regional variations ranging from 10-30 percent).

One potential cause of this finding is that providers fear being penalized on quality metrics if they do not test. To mitigate these effects, quality measurement activities should include balancing measures that convey not only when screening (or other types of testing) reach the intended group, but also when it inappropriately reaches a group unlikely to benefit. This approach would penalize indiscriminate, potentially wasteful, testing.

Avoidance Of Institutional Settings Of Care

Hospitals and nursing homes are expensive, and no one wants to be in them. Yet they serve important roles for people in late life, as do skilled nursing facilities, which allow people to rehabilitate from an acute hospitalized illness. In the Medicare program, the divide between inpatient settings used for acute care, as opposed to long-term care, seems clear because Medicare does not cover long-term care. Yet in some older adult populations, that traditional divide between acute and long-term care based on who pays begins to blur when we consider what services the person needs.

For example, consider older people with a dementia diagnosis, such as Alzheimer’s Disease, who spend on average 22.5 days in a hospital or skilled nursing facility, compared to the general population of older adults, who spend 4.6 days on average. Unlike heart attacks or pneumonia that can require hospital-based procedures, the dementia diseases have no specific treatments that are delivered in the hospital. Instead, people with dementia are often hospitalized for other diseases that decompensate when their ability to self-manage fails, or the caregiver burden becomes too great, or the person approaches the end of life, with all the symptoms that entails.

These types of problems could be addressed with alternative approaches that fall between the cracks of medical care and home-based long-term care support. Yet to receive paid support, families often are limited to seeking hospital-based care. As we redesign the health care system we need to ask what services are being delivered during those inpatient stays and if people should or could be returned to a lower level of care or to their home environments with appropriate professional supports sooner.

Delivering High Quality End-Of-Life Care

It’s undeniable that many adults in the aging population are approaching their last years of life. A core feature of a health system well designed to address the needs of an aging population is the provision of high-quality end-oflife care. While strides have been made to increase access to palliative care, many areas still require improvement. One of these is how often the start of hospice is delayed until the person is within three days of dying.

On average, 17 percent of decedents are referred to hospice within three days of death. While the progression toward death in some cases may have been quick, and the referral does not represent a delay, it is unlikely to be the predominant reason when, in some regions as few as 6 percent have a late referral, and in others more than a quarter of hospice episodes begin within three days of death. Use of hospice may improve quality of dying but only if given time to do so. A high rate of hospice use does not on its own signify that we have improved care delivery at the end of life, especially if a large portion of that use is for a very brief period before death.

A Call To Action

Older adults experience some of the highest levels of contact with the health care system. Ongoing monitoring and reporting of measures meaningful for this population is one piece of a strategy to ensure that the benefits and gains of health system change reach the people who may need them most.

If we begin to observe and report on population-specific measures, we can develop the capacity not only to improve current delivery but also to inform development of new models that address the unique needs of an aging population. Finally, monitoring these measures will also enable us to learn whether vulnerable older adults experience the same gains from health care reform as other, less-vulnerable groups, and identify any unintended negative consequences.

Exhibit 1