Wednesday, October 21, 2009

Hospital at Home for Decompensated Heart Failure

Imagine the year is 2022. You're a palliative care practitioner working with a hospital based team. An emergency department physician calls you to see an 83 year old man with stage D congestive heart failure who came to the hospital with a several day history of worsening dyspnea and orthopnea. He had two admissions earlier in the year. After the last one, he spent three weeks in a skilled nursing facility before returning home semi-independently. Upon evaluation of the patient, he tells you that his main goals are to be at home with his family and dog, Miffy, but he also hopes that he'll be able to sleep without awakening with dyspnea so frequently and wants to do everything "within reason" to live until his granddaughter graduates from college in four months. He has a POLST form that indicates he wishes to be DNR, and this has already been confirmed by EMS and the ED physician. Your interdisciplinary team evaluates the patient, and finds that he has excellent support from his wife and three daughters as well as a couple of neighbors. After extensive discussion, it is decided that the patient will be discharged home from the emergency department with follow-up later in the day from your colleague who is currently on service with an affiliated hospital-at-home program. Your colleague sees him daily for four days and then every other day for two weeks. He also receives daily nursing visits for the first 14 days. He returns to his baseline and afterwards, he receives intermittent home nursing.

Far-fetched? Perhaps. But a prospective, randomized controlled trial from Torino, Italy recently published in the Archives of Internal Medicine suggests that it may be possible to provide hospital-at-home care to geriatric patients with decompensated CHF without worsening six month mortality and possibly improving certain quality of life indicators (compared to routine hospital care).

The study included patients older than 75 with stage C congestive heart failure and New York Heart Association class III or IV symptoms who had an acute decompensation of CHF requiring hospitalization. Patients were excluded if they lacked family support, required mechanical ventilation, lived outside of the catchment area, or had severe comorbidities. Of 528 potential subjects, only 101 were randomized.

Outcomes:
  • 15% of patients in each group died (p = .83). This was the primary outcome.

  • Statistically significant differences in favor of hospital at home (all outcomes at 6 months after admission): Geriatric Depression Scale, Mini-nutritional Assessment, and Nottingham Health Profile (looking at overall quality of life)

  • No statistically significant difference: Measures of functionality and mini mental state exam.

  • Patients at home received a mean of 20 days of hospital-like care versus a mean of 11 days for hospitalized patients.

  • Fewer patients in the hospital at home group were readmitted to the hospital within six months, but this was not statistically significant. The number of days between hospital discharge and readmission was greater in the hospital at home group.

  • No patients in the hospital at home group were institutionalized after discharge while 16% of hospitalized patients required some form of nursing facility after hospitalization.

  • Hospital at home cost less overall ($2604.46 for each patient treated at home vs. $3027.78 for those hospitalized).

  • Stress levels were higher for family members of hospital at home patients.
Limitations in interpretation and application in the United States include the fact the study was done in Italy and the team applying the intervention is well-established (in existence for 20 years and providing care for over 1000 patients over the last ten years). An associated editorial does note that there are examples of hospital-at-home programs in the United States (Johns Hopkins, Presbyterian Healthcare Services in Albuquerque, and the VA in Portland, OR) but it's unclear how these programs differ in structure and services provided. (The Hopkins website references a study from JAGS indicating that caregiver stress may be less in their hospital-at-home program than for caregivers of hospitalized patients in contrast with the current study.)

In my hypothetical scenario above, the practitioner is a palliative care specialist. Why? Well, mainly because that's the lens through which I see the world. Certainly the outcomes that were significant carry much valence in the world of palliative care, so why not make this a palliative care intervention if that's who is interested in providing the service or if a geriatrics service isn't available? I will go to bed dreaming* of health care reform that allows for a more sensible hospice benefit which is more fluid with other home-based services (such as interventions like this) and which doesn't force the "terrible choice." The current hospice model might have a lot to offer in terms of helping to reduce the stress of family caregivers of a hospital-at-home patient.

Of special note is the fact that patients in the hospital-at-home group received acute services longer than the hospitalized group, but the latter group was at higher risk for requiring post-discharge nursing facility care. This turns the concept of "throughput" on it's head. DRG's and SNF's were introduced as cost-saving mechanisms for Medicare. A byproduct over the years is the intense pressure for hospitals to move patients rapidly through the system (not for the sake of the patient, but for the sake of the system). Could the hospital-at-home model represent another opportunity for Medicare to offer patients greater choice while perhaps saving money? The business model likely isn't there yet, but perhaps this will be a "disruptive technology" of the future. I'll remain hopeful without holding my breath.

*Okay, I really don't like dreaming about work.