Skip to main content

Table 3 Univariable and multivariable analysis for each predictor included in the model and simplified risk score to predict mortality among asphyxiated neonates admitted to NICU at FHCSH, Bahir Dar, Ethiopia, 2021

From: Incidence and development of validated mortality prediction model among asphyxiated neonates admitted to neonatal intensive care unit at Felege Hiwot Comprehensive Specialized Hospital, Bahir Dar, Northwest Ethiopia, 2021: retrospective follow-up study

Predictor Variables

Univariable

β (95% CI)

Full Model by Multivariable Analysis

Model Reduction by LR Test

SRS

β (95% CI)

P–V

P–V

β (95% CI)

Maternal age

 20–35

0

0

0

 

0

0

 < 20

0.08 (-0.38,0.54)

NA

-

 

-

-

 > 35

0.75 (.05,1.44)

0.52 (-0.39,1.44)

0.263

0.406

NA

-

Residence

 Urban

0

0

0

 

0

0

 Rural

0.67 (0.34,0.99)

0.43 (.03,0.83)

0.036

0.10

0.50 (0.12,0.88)

1

ANC Visit

 Yes

0

0

0

 

0

0

 No

0.09 (-0.70,0.89)

NA

-

-

-

-

Parity

 Multipara

0

0

   

0

 Primipara

0.04 (-0.32,0.40)

NA

-

 

-

-

 G-Multipara

0.74 (0.20,1.27)

0.16 (-0.52,0.84)

0.651

0.626

NA

-

BOH

 No

0

0

0

 

0

0

 Yes

0.52 (0.13,0.92)

0.54 (0.04,1.05)

0.035

0.160

0.60 (0.14,1.06)

1

APH

 No

0

0

0

 

0

0

 Yes

0.09 (-0.50,0.67)

NA

-

-

-

-

PROM

 No

0

0

0

 

0

0

 Yes

0 .11 (-0.31,0.54)

NA

-

-

-

-

Labor Duration

 ≤ 18 h

0

0

0

 

0

0

 > 18 h

0.06 (-0.34,0.45)

NA

-

-

-

-

Amniotic Fluid

 Clear

0

0

0

 

0

0

 Stained

0.59 (0.26,0.92)

0.60 (0.20,1.01)

0.003

0.035

0.62 (0.23,1.01)

1

Pregnancy Type

 Single

0

0

0

 

0

0

 Multiple

1.11 (0.33,1.88)

1.26 (0.33,2.20)

0.008

0.109

1.31 (0.39,2.23)

3

Place of Delivery

 Inborn

0

0

0

 

0

0

 Referred

0.03(-0.29,0.36)

NA

-

-

-

-

Gestational Age

 Term

0

0

0

 

0

0

 Preterm

0.65 (0.20,1.10)

0.32 (-0.29,0.92)

0.302

 

NA

-

 Posterm

0.41 (-0.29,1.10)

0.22 (-0.62,1.07)

0.606

0.69

NA

-

Birth Weight

 ≥ 2500(g)

0

0

0

 

0

0

 < 2500(g)

0.66 (0.30,1.03)

0.81 (0.31,1.31)

0.001

0.008

0.95 (0.51,1.40)

2

Sex of newborn

 Female

0

0

0

 

0

0

 Male

0 .37 (0.02, 0.73)

0.26 (-0.15,0.67)

0.21

0.59

NA

-

Age at admission

 ≤ 6 h

0

0

0

 

0

0

 > 6 h

0.10 (-0.24,0.44)

NA

-

-

-

-

Stage of HIE

 Stage I

0

0

   

0

 Stage II

0.77 (0.33,1.21)

0.87 (0.39,1.34)

< 0.001

< 0.001

0.80 (0.33,1.26)

2

 Stage III

2.54 (2.03,3.05)

2.66 (2.09,3.22)

< 0.001

 

2.67 (2.12,3.23)

5

Heart Rate (bpm)

 100–160

0

0

   

0

 < 100

0.69 (-0.11,1.48)

0.48 (-0.50,1.47)

0.34

0.70

NA

-

 ≥ 160

0.09 (-0.54,0.36)

NA

-

 

-

-

Temperature(oc)

 35.5–37.5

      

 < 35.5

0.60 (0.27,0.94)

0.34 (-0.07,0.74)

0.105

0.55

NA

-

 > 37.5

0.13 (-0.94, 0.68)

NA

-

 

-

-

Sepsis

 No

0

0

0

 

0

0

 Yes

0.02 (-0.32,0.36)

NA

-

-

-

-

Treatment Given

 Yes

0

0

0

 

0

0

 No

-1.33 (-3.40,0.73)

-1.10 (-3.43,1.25)

0.362

0.53

NA

-

Not cry at birth

 No

0

0

   

0

 Yes

0.56 (.18,0.95)

0.37 (-0.11,0.85)

0.134

0.407

NA

-

Failure to suck

 No

0

0

0

 

0

0

 Yes

0.60 (0.22,0.96)

0.58 (0.10,1.06)

0.018

0.10

0.69 (0.26,1.11)

1

Difficulty to Breath

 No

0

0

0

 

0

0

 Yes

0.39 (0.03,0.76)

0.02 (-0.45,0.48)

0.95

0.48

NA

-

  1. Predictors that remained in the reduced final model using the likelihood ratio test are; Residence, bad obstetric history, amniotic fluid status, multiple pregnancy, birth weight (< 2500), HIE (stage II and III), and failure to suck. Stepwise backward elimination for predictor selection was used. NA—not included in the Simplified risk score: we divided the coefficient of predictors included in the reduced model by the smallest (0.43)
  2. HIE hypoxic ischemic encephalopathy, β beta coefficient, P–V p-value, SRS simplified risk score, LR = Likelihood ratio
  3.  ⊕ Logit (P) = Log (\(\frac{\mathrm{risk of mortality }}{1-\mathrm{risk of mortality}}\)) = linear predictor = − 3.11 + 0.50*(Rural Residence) + 0.60*(Bad Obstetric History) + 0.62*(Stained Amniotic Fluid) + 1.31*(Multiple Pregnancy) + 0.95*(Birth weight < 2500 g) + 0.80 (Stage II HIE) + 2.67(Stage III HIE) + 0.69(*(Failure to suck) orby Simplified Risk Score
  4.  ⊕ Logit (P) = Log (\(\frac{\mathrm{risk of mortality }}{1-\mathrm{risk of mortality}}\)) = linear predictor = 1*(Rural Residence) + 1*(Bad Obstetric History) + 1*(Stained Amniotic Fluid) + 3*(Multiple Pregnancy) + 2*(Birth weight < 2500 g) + 2*(Stage II HIE) + 5(Stage III HIE) + 1*(Failure to suck)
  5. ⇒ Sum of risk score = 1*(Rural Residence) + 1*(Bad Obstetric History) + 1*(Stained Amniotic Fluid) + 3*(Multiple Pregnancy) + 2*(Birth weight < 2500 g) + 2*(Stage II HIE) + 5(Stage III HIE) + 1*(Failure to suck) = 16