Monday, August 25, 2008

Hospital Mortality Statistics and Palliative Care Teams

Several news agencies and likely your local TV news (but curiously not the NY Times, Washington Post, or Wall Street Journal) had articles in the past week highlighting the online publication of a nationwide hospital mortality database from Medicare data. The information is drawn from inpatient mortality rates from heart failure, myocardial infarctions (heart attacks) and pneumonia cases. Drew commented last year on a JAMA article last year about the usefulness of mortality rates as quality indicators.1

Note: inpatient mortality is qualified as dying within 30d of hospital admission regardless of location of care at time of death.

Let's consider the ways palliative care can affect the inpatient mortality rate for any of these three conditions:

First, does a hospital have a palliative care team?
Your hospital may fear adding a palliative care team because it may errantly believe palliative care will increase their inpatient mortality rate. But one must consider that an effective palliative care team can facilitate safe and appropriate discharge plans, that can reduce unneeded readmission due to system errors, and by connecting patients with helpful outpatient services such as hospice.

Second, does a hospital have an effective, multi-disciplinary team that has wide penetration across diagnoses and clinical services?
A palliative care team that only sees patients in the ICU or oncology floor may not have a hospital wide impact on these three diseases. So therefore any effect on the inpatient mortality rate may be minimized

Third, does your hospital administration know that any patient who is receiving goals of comfort care is excluded from this mortality rate?
Here is the language from the Joint Commission Appendix on Hospital Quality Measures:

Beginning with patients discharged in July 2006, patients receiving only comfort care (support for the dying patient) were excluded from all of the heart attack and heart failure measures. This exclusion had previously been applied only in the pneumonia measure population. With the addition of this exclusion in the heart attack and heart failure measure population, it is likely that some actual measure rates may change from previous reporting periods.

Physician/nurse practitioner/physician assistant document that the patient was receiving comfort measures only. This is commonly referred to as “palliative care” in the medical community and “comfort care” by the general public. Palliative care includes attention to the psychological and spiritual needs of the patient and support for the dying patient and the patient's family. Usual interventions are not received because a medical decision was made to limit care to comfort measures only. Comfort Measures only are not equivalent to the following: Do Not Resuscitate (DNR), living will, no code, no heroic measure.
I looked at some local hospitals for heart attack 30-day mortality. These hospitals are in the 200-300 bed range. The absolute number of heart attack mortalities in a year were in the 60-80's. Imagine the coder utilizing the palliative care modifier (v66.7 from ICD-9-CM) on some of these patients who could be seen by palliative care. Even if you saw 1 more cardiac patient a month, you could potentially decrease these absolute mortality numbers by 10-20%.

So there you have it; another way to market the benefits of palliative care to your hospital administration. Does anyone know if CAPC highlights this at all? Has anyone gone to the hospital coders or administrators to discuss this?

1. Holloway, RG; Quill TE. Mortality as a Measure of Quality: Implications for Palliative and End-of-Life Care. JAMA. 2007;298:802-804. 10.1001/jama.298.7.802 DOI

Pallimed | Blogger Template adapted from Mash2 by Bloggermint