This review discussed the applications of illness severity scores in clinical medicine and research, reviewed the caveats pertaining to their use, and discussed appropriate construction of a score.
An illness score is a number assigned to a patient that correlates with probability that a specific outcome will follow. Illness severity scores may be diagnosis-specific or diagnosis-independent. They are not designed for use in isolation to predict outcome for individuals, but should be applied on a population basis. Illness severity scores can be used to:
• Decrease bias and confounding in research
• Provide quantitative and objective measures of patient illness
• Benchmark performance and establish protocols for triage and therapeutic management
• Improve validity of conclusions in observational studies
• Demonstrate effective randomization, which is especially important when group sizes are small in randomized trials.
Euthanasia in veterinary medicine poses a unique challenge to models developed on mortality outcome. The timing of euthanasia reflects multiple factors, including severity of patient illness, owner financial and emotional status, diagnosis of a terminal disease, subjective assessments of degree of suffering, and individual clinician perspective. If all euthanized patients are excluded, available patient data may be limited and biased. If all patients are included, the significance of a particular variable as a risk factor for death may be masked by patients euthanized for financial reasons.
Commentary: Dr. Hayes reviews the very interesting topic of illness severity scores in veterinary medicine; there are few published scores and they are rarely used. Illness severity scores are used daily in human medicine, in both clinical practice and in clinical research studies for population comparison. Objective scores allow for more accurate presentation of prognosis, institution of more invasive and intensive therapy, and assessment of more accurate outcomes in clinical research studies. Dr. Hayes has presented early validation work on both canine and feline illness severity scores at the International Veterinary Emergency and Critical Care Symposium in both 2009 and 2010, and received the “best resident abstract award” both years for her work. This is an exciting topic that will undoubtedly be used as an instrument to improve clinical care in dogs and cats, and may provide stronger statistical power for future clinical research studies.—Lisa Powell, DVM, Diplomate ACVECC
Illness severity scores in veterinary medicine: What can we learn? Hayes G, Mathews K, Kruth S, et al.
J Vet Int Med 24:457-466, 2010.