Patient with Coma | Narayan's Logistic Model | Medical Algorithm

Narayan's logistic model uses multivariate analysis to predict the probability of poor or good outcome in patients with coma.

Predictor	Value	Coefficient
age in years	same	0.061
Glasgow Coma Score	same	(-0.469)
pupillary reaction to light	0 (normal or unilaterally absent) or 1 (bilaterally absent)	1.545
oculocephalic or oculovestibular responses eye movements	0 (normal) or 1 (impaired or absent on either or both sides)	0.611
surgical mass	0 (absent) or 1 (present)	0.765

probability of poor outcome=

= (1 / (1 + (e ^ ((-1)*((sums of (value * coefficient)) - 0.674)))))

probability of good outcome = 1 - (probability of poor outcome)

where:

• 0.674 is the line intercept

• Poor outcome indicates severe disability, vegetative state or death.

• Good outcome indicates good recovery or moderate disability.

Limitations:

• Prediction is most accurate for patients at the extremes

• Prediction is least accurate for patients in the mid-range

• Patients with gunshot wounds to the head were excluded

To read more or access our algorithms and calculators, please log in or register.

Narayan's Logistic Model