Patient Engagement

Days Spent at Home — A Patient-Centered Goal and Outcome

Article · November 4, 2016

Interview with Dr. Thomas Lee on measures of health care quality that are most important to patients.


Most measures of the quality of health care delivery focus on what health care providers do, not what patients want. If “high-value, patient-centered care” is to be more than rhetoric, health care organizations need to measure outcomes that matter to patients. Only when they do so will care be designed and organized in ways that improve those outcomes.

To understand how this approach differs from business as usual, consider “days spent at home in the last 6 months of life.” The map shows the mean number of days spent at home by Medicare beneficiaries who died in 2012 or 2013, by hospital referral region (HRR). HRRs are regional health care markets as defined on the basis of historical hospital-usage patterns. “Days at home” was calculated as 180 days minus the number of inpatient days in an acute care facility, an inpatient rehabilitation facility, a skilled nursing facility, or an inpatient hospice unit.

Mean Number of Days Spent at Home in the Last 6 Months of Life, by Hospital Referral Region, for Medicare Beneficiaries Who Died in 2012 or 2013

Mean Number of Days Spent at Home in the Last 6 Months of Life, by Hospital Referral Region, for Medicare Beneficiaries Who Died in 2012 or 2013. Click To Enlarge.

As the map suggests, there is substantial variation in the amount of time dying patients in the United States spend at home. The mean was just 120 (of 180) days in Los Angeles and Miami and 122 in McAllen, Texas, and Chicago. In Mason City, Iowa, dying patients spent an average of 146 days at home. In Grand Junction, Colorado, the average was 143; in Rochester, Minnesota, 142. (A Supplementary Appendix providing the mean number of days for each HRR is available at

Reducing hospital-readmission rates is a laudable goal, but not exactly the same thing as trying to maximize the amount of time that frail and dying patients spend at home. Do a few more weeks at home in the last 6 months of life matter? After all, the ultimate clinical outcome was the same, since all the patients included in the analysis died. Furthermore, there are many patients who prefer not to be at home when they are sick and dying, and for whom days spent at home are days during which they have less-than-complete relief of discomfort.

Being home at the end of one’s life will never be a universal goal, but our experience and academic research suggest that, all else being equal, patients would rather be at home than in health care facilities. When surveyed about their preferences for dealing with a terminal illness, most people (86%) indicated that they would prefer to be at home in their final days. In addition, they would not want to be on a ventilator in order to gain an extra week of life, and they are not opposed to drugs that could improve symptoms but potentially shorten life. These preferences are highly consistent across regions of the country and people’s socioeconomic status.1 Despite this consistency in end-of-life preferences, there is wide regional variation in the intensity of, expenditures on, and locations of care provided during the last 6 months of life.2

Our interest in “days at home” as an outcome measure was stimulated by a comment from one person — a family member of a patient with a disabling condition, who described his emotions as he and the patient entered the front door of their home after a 6-week hospitalization. “Just to see familiar photos on the wall made us feel like we could breathe again,” he said. Like many patients and families, they wanted to maximize the number of days the patient could be at home.

A similar message emerged from focus groups conducted by the Camden Clinical Commissioning Group (CCG) in the United Kingdom, which asked frail elderly patients and their caregivers, “What is most important to you?” The priority that emerged was “time spent at home,” and when the CCG group decided to organize its clinical and community resources around that single clear goal, the results included lower use of hospitals and emergency departments, better patient experience, and higher clinician morale.3

Relationship between Number of Days with Home Health or Home Hospice Services in the Last 6 Months of Life and Total Number of Days Spent at Home in Those 6 Months

Relationship between Number of Days with Home Health or Home Hospice Services in the Last 6 Months of Life and Total Number of Days Spent at Home in Those 6 Months. Click To Enlarge.

An important message from the Camden CCG experience and from deeper analyses of Medicare data on days at home is that there is a difference between doing more activities that make good sense and organizing around a goal. One might expect that greater use of home health and hospice services would correlate with more days at home, but as the graph shows, the opposite is true. In regions where patients had more days on which they received home health services, hospice services, or both, they spent fewer total days at home (correlation coefficient, –0.53).

In short, when providers did more of one thing, they did not do less of another. There is a difference between doing more and doing better. The Camden CCG experience indicates, however, that when health care providers and community groups collaborate with the goal of increasing days spent at home, progress can be made.

There are many other important questions to explore regarding days spent at home in the last 6 months of life. For example, what socioeconomic, geographic, cultural, and other nonmedical contributors drive preferences for spending days at home? Would further analysis show that home hospice care is effective at increasing the number of days spent at home when evaluated within a narrower time frame, such as in the last 30 days of life? Furthermore, alternative types of payment contracts for health care providers may affect the number of days patients spend at home, so the effect of new payment incentives should be monitored. Ultimately, eliciting patients’ preferences and organizing care to accommodate individual patients’ goals is what matters most.

Health policy often stumbles when there is uncertainty about what we are trying to maximize in health care. It is clear that one goal is to minimize costs, but there should be counterbalancing measures to be maximized. Mortality is not a sufficient measure to define excellence in care; in fact, no single performance metric will suffice. Outcome measures that reflect what truly matters to patients can define performance in ways that increase the engagement of patients, clinicians, and provider organizations in the redesign of care.


From the Dartmouth Institute for Health Policy and Clinical Practice, Lebanon, NH (A.C.G., C.H.C.); Bayada Home Health Care, Moorestown, NJ (A.C.G.), and Harvard Medical School, Boston (T.H.L.); and Press Ganey, Wakefield (T.H.L.) — both in Massachusetts.

1. Barnato AE, Herndon MB, Anthony DL, et al. Are regional variations in end-of-life care intensity explained by patient preferences? A study of the US Medicare population. Med Care 2007;45:386-393.
CrossRef | Web of Science | Medline
2. Wennberg JE, Cooper M. The Dartmouth atlas of health care. Lebanon, NH: Dartmouth Institute (
3. Sayer C. “Time spent at home” — a patient-defined outcome. NEJM Catalyst. April 26, 2016 (

This Perspective article originally appeared in The New England Journal of Medicine.

Call for submissions:

Now inviting expert articles, longform articles, and case studies for peer review


A weekly email newsletter featuring the latest actionable ideas and practical innovations from NEJM Catalyst.

Learn More »

More From Patient Engagement
PROMs Usage Is Limited but Growing

Patient Engagement Buzz Survey: PROMs Use Is Growing, but Implementation Takes Effort

While clinical outcomes lend themselves to measurement, quantifying how an individual patient experiences symptoms or a loss of function can be more challenging. One promising approach is the use of Patient-Reported Outcome Measures (PROMs).

Taxonomy of the Patient Voice Table

Taxonomy of the Patient Voice

While health care pursues the important trend of putting patients at the center of care, the terms used to describe this goal are proliferating — and muddying the discourse. This taxonomy attempts to classify some of these terms and make some distinctions.

Agarwal01_pullquote handcrafting the patient experience

Handcrafting the Patient Experience

Health care organizations can take cues from consumer-facing companies like Airbnb to creatively insert convenience and surprise into patient encounters.

Screenshot of Tidepool daily diabetes data

A Taxonomy to Engage Patients: Objectives, Design, and Patient Activation

Health information initiatives will succeed only if they focus on patients’ motivation to engage and reflect the type of engagement they seek.

Ghafur01_pullquote -digital health health care consumer patient experience

Engaging Patients Using Digital Technology — Learning from Other Industries

Providers can benefit patients and disrupt health care by learning from the experience of other industries.

Mission Hospital Virtual Sitter - Drawing of the Virtual Bed Zone and Rails

Reducing Inpatient Falls and Injury Rates by Integrating New Technology with Workflow Redesign

How Mission Hospital scaled a virtual sitter pilot and reduced unassisted falls by 44% and fall-related injuries by 40%.

Health Care Providers Should Incentivize Patients

Survey Snapshot: Patient Financial Incentives — There Are No Quick Fixes

The NEJM Catalyst Insights Council agrees that while financial incentives are a common strategy to engage patients in healthy behaviors, they are not necessarily effective.

Support of Family and Friends Is More Effective Than Clinician Support in Realizing Health Goals - From the Patient Engagement Insights Report: Why No Single Health Incentive Works.

Patient Engagement Survey: Why No Single Health Incentive Works

Initiatives to improve patient engagement come in a variety of forms. While insurers, employers, and health care providers are all involved in using financial incentives and penalties for engagement efforts, improvement in health outcomes has been elusive. Achieving that ultimate goal will usually require a combination of financial and social approaches.

Health Systems Attending the Nudge Units in Health Care Symposium - Penn Medicine

Key Insights on Launching a Nudge Unit within a Health Care System

Leaders are finding that making higher-value choices easier through subtle changes to choice architecture can have an outsized impact on medical decision-making.

Barriers to Providing an Oustanding Patient Experience

Buzz Survey Report: Patient Experience

An independent NEJM Catalyst report sponsored by University of Utah Health on barriers to achieving an excellent patient experience.


A weekly email newsletter featuring the latest actionable ideas and practical innovations from NEJM Catalyst.

Learn More »


Patient Incentives

75 Articles

Taxonomy of the Patient Voice

While health care pursues the important trend of putting patients at the center of care,…

The “Behavioral Health Vital Signs” Initiative

A safety net system’s trauma-informed approach to integrating interpersonal violence into behavioral health programs in…

Patients As Customers

141 Articles

Information Asymmetry: The Untapped Value of…

The knowledge and preferences that patients could — and should — share with clinicians would…

Insights Council

Have a voice. Join other health care leaders effecting change, shaping tomorrow.

Apply Now