How reliable are your prognoses?
From a client's point of view, the antemortem differentiation of potentially reversible from irreversible illness is often the most important unknown related to clinical assessment of a patient. Client's typically ask, "Can you help him/her doc?" Our clients are concerned about the probability of their animals' recovery from diseases and disorders with or without therapy, the nature and cost of therapy, and whether recovery will be partial or complete. This phase of patient evaluation is commonly referred to as the"prognosis".
As with most aspects of patient care, formulating a prognosis requires judgment in the absence of certainty. For some patients, a prognosis is lifesaving; for others, it is a death sentence. Isn't it true that clients often make decisions about euthanasia based primarily on our prognoses?
It is therefore appropriate to ask the following questions: How reliable are our prognoses? What evidence can we cite to justify our forecasts? Do our clients fully understand the message we are trying to convey to them?
Confusion and misunderstanding often occur when two people attempt to communicate using two different languages. More commonly, confusion arises between individuals using the same language when each attach a different meaning or different definition to what appears to be a universally accepted term. To emphasize my point, I suggest that you ask your hospital staff and your clients how they would define a "guarded prognosis". What impact would their definition likely have on a client's choice of therapeutic options?
The term prognosis is derived from the Green language (pro, translated as "before"; and gnosis, translated as "to know") and means a forecast of the probable outcomes of abnormalities associated with one or more diseases, and the frequency with which they can be expected to occur. Dorland's Medical Dictionary defines prognosis as "a forecast as the probable outcome of an attack of disease" or "the prospect as to recovery from a disease as indicated by the nature and symptoms of the case". Synonyms for the term prognosis include prediction, foretelling and forecast.
In context of prognoses, effective communication between clients and doctors cannot consistently occur unless there is mutual understanding. Therefore, in place of poorly defined prognostic terms that are highly subjective, we recommend terminology that lends itself to quantification of the probability of a predicted outcome (Table 1). The goal is to help clients fully understand the message we are trying to convey to them.
To further improve communication, we recommend that the prognosis for each problem affecting patients be subcategorized according to predicted events in the immediate future (short-term prognosis), and also according to the probability of resolution of morphologic and functional abnormalities in the distant future (long-term prognosis). For example, the short-term prognosis for survival is good to excellent for most nonazotemic dogs with subclinical glomerular amyloidosis characterized by proteinuria and hypersthenuria.
However, even with contemporary supportive and symptomatic therapy, the long-term prognosis for dogs with glomerular amyloidosis has, to date, been poor to grave. In contrast, the short-term prognosis for survival of patients with non-steroidal anti-inflammatory drug induced acute oliguric renal failure and characterized by profound metabolic acidosis, severe hyperkalemia and marked hypothermia is often guarded to poor. However, with appropriate therapy the long-term prognosis for survival and recovery of adequate renal function to sustain a good quality of life is good to excellent.
Recently, considerable emphasis has been placed on the need to base our decisions about the care of patients (including prognoses) on conscientious, explicit and judicious use of current best evidence (so-called "evidence- based" medicine). Putting evidence based medicine into practice means integrating our individual clinical expertise with the best available external clinical evidence derived from systematic research. One group of investigators (McGowan et al: Journal of Infectious Diseases, Vol. 165, pages 1-13, 1992) has proposed the following scheme to score the strength and the quality of available external evidence.
Grade I evidence is defined as the highest quality evidence. It is derived from at least one properly randomized controlled clinical study.
Grade II evidence may be data obtained from:
At least one well-designed clinical study without randomization. Cohort or case-controlled analytic studies.Studies using acceptable laboratory models or simulations in the target species, preferably from more than one center.Multiple time seriesDramatic results in uncontrolled studies.
Opinions from respected authorities on the basis of clinical experience.Descriptive studiesStudies in other speciesPathophysiological justificationReports of expert committees
Additional information on how to evaluate an article on prognosis can be found at www.cche.net/usersguides/prognosis.asp.
Further information about evidence-based medicine can be found in Kirk's Current Veterinary Therapy-XIII (Polzin DJ et al: From journal to patient: Evidence-based medicine. WB Saunders, Philadelphia, pages 2-8, 2000).
Other concepts that may be useful when formulating prognoses are summarized in Table 2. Let us consider the concept of aging as a prognostic factor in greater detail. Aging is associated with a decline in total function (or functional "reserve") of several organs (for example, the kidneys) and systems (for example, the immune system). Thus, the short-term and long-term prognoses for complete recovery of "aged" patients from a variety of metabolic, degenerative, neoplastic, traumatic, infectious and toxic diseases may be less favorable than the prognoses of younger patients with comparable disorders.
This generality, however, should not be used as the primary basis for recommending "benign neglect" or euthanasia for older patients. Proper management of various illnesses and ailments affecting geriatric patients often allows them to regain and maintain adequate function of various organs and systems to sustain a good quality of life.
Once prognoses have been established, and a therapeutic plan has been formulated, follow-up evaluations of the course of the disease and its response to therapy (if any) are recommended. Therapeutic plans should include predictions of expected outcomes, and the associated risks and benefits. We should be able to clearly describe the rationale for the treatment we are recommending, and be familiar with the scientific evidence supporting our choice of therapy.
Our prediction of the progress of the patient should be discussed with the owner with the goal of helping him/her recognize deviations from the expected course of events, and also provide him/her with a balanced perspective of the significance of such deviations.
Clients should clearly understand important side effects that may be associated with therapy, their likely frequency of occurrence, and what to do if they occur. Before clients leave the hospital, they should be given a list of undesirable events that should immediately be reported to the doctor.