Do-not-attempt-resuscitation orders and prognostic models for intraparenchymal hemorrhage*

Abstract

Objective Statistical models predicting outcome after intraparenchymal hemorrhage (IPH) include patients irrespective of do-not-attempt-resuscitation (DNAR) orders. We built a model to explore how the inclusion of patients with DNAR orders affects IPH prognostic models. Design Retrospective, observational cohort study from May 2001 until September 2003 Setting University-affiliated tertiary referral hospital in Seattle, Washington Patients 424 consecutive patients with spontaneous intraparenchymal hemorrhage Measurements We retrospectively abstracted information from medical records of IPH patients admitted to a single hospital. Using multivariate logistic regression of presenting clinical characteristics, but not DNAR status, we generated a prognostic score for favorable outcome (FO, defined as moderate disability or better at discharge). We compared observed probability of FO with that predicted, stratified by DNAR-status. We then generated a modified prognostic score using only non-DNAR patients. Main Results Records of 424 patients were reviewed: 44% had FO, 43% had a DNAR-order and 38% died in hospital. Observed and predicted probability of FO agreed well with all patients taken together. Observed probability of FO was significantly higher than predicted in non-DNAR patients and significantly lower in DNAR patients. Results were similar when applying a previously published and validated prognostic score. Our modified prognostic score was no longer pessimistic in non-DNAR patients, but remained overly optimistic in DNAR patients. Conclusions Although our prognostic model was well calibrated when assessing all IPH patients, predictions were significantly pessimistic in patients without, and optimistic in those with DNAR orders. Such pessimism may drive decisions to make patients DNAR in whom a FO may have been possible, thereby creating a self-fulfilling prophecy. To be most useful in clinical decision-making, IPH prognostic models should be calibrated to large IPH cohorts in whom DNAR orders were not used.

Estimating and communicating prognosis in advanced neurologic disease

“…46 There has been concern in neurologic disease, particularly in stroke, that physicians may be contributing to nocebo effects, and possibly self-fulfilling prophecies, given studies showing an association between DNR orders and a higher probability of a short-term mortality. [15][16][17] As a result, the American Stroke Association guidelines for intracerebral hemorrhage advise that the "postponement of new DNR orders until at least the second full day of hospitalization is probably recommended." 47 Providers need to be aware that short-term mortality estimates from the literature (e.g., inpatient or 30-day stroke mortality) represent a heterogeneous mix of patients with varying levels of treatment intensities.…”

“…13,14 Concerns have been raised in stroke neurology, however, about the possibility of overly pessimistic predictions if mortality/ survival estimates are used from populations with mixed treatment intensities (e.g., patients with donot-resuscitate [DNR] and non-DNR orders), leading to a "self-fulfilling prophecy." [15][16][17] "How well?" Most neurologic diseases affect function more than survival.…”

Palliative and End-of-Life Care in Stroke

“…35 Prognostic models and clinical experience for severe stroke patients may be biased by the frequent withdrawal of life-sustaining treatments, leading to a self-fulfilling prophecy in which the "true prognosis" if all life-sustaining treatment were offered is difficult to ascertain. [36][37][38] Given the frequent use of early decisions to limit life-sustaining treatment in ICH, the potential for withdrawal bias is likely far greater in ICH than in other stroke types. [39][40][41] An additional challenge comes from determining what represents a "good" outcome to an individual patient.…”

“…45,47,49 Models derived from clinical trial populations or registries based at major stroke centers are not necessarily applicable to the broader population of all stroke patients in a community, especially when one considers diverse populations with high proportions of racial and ethnic minorities. Few if any models incorporate certain important factors shown to influence outcome after stroke, such as institutional norms on approaches to end-of-life care, 39 the presence of a do-not-resuscitate (DNR) order, [36][37][38] provider values, 50 and communication regarding prognosis and treatment goals. 51 It can often be a challenge in model development to find an appropriate balance between making a model simple enough for practical bedside use while incorporating sufficient information to account for the multiple factors that influence outcome.…”

“…21,37,38,40,[109][110][111][112][113] There is concern that this association can lead to a false prognostic pessimism that may lead to premature withdrawal of life-sustaining measures and thereby to a self-fulfilling prophecy. In such studies, however, it is difficult to determine causality.…”