DOI: 10.1542/peds.2003-0960-L 2004;114;e424Pediatrics

Tarnow-Mordi and Anne GreenoughSimon J. Broughton, Andrew Berry, Stephen Jacobe, Paul Cheeseman, William O.

Prediction Model for Retrieved NeonatesThe Mortality Index for Neonatal Transportation Score: A New Mortality

  

  http://pediatrics.aappublications.org/content/114/4/e424.full.html

located on the World Wide Web at: The online version of this article, along with updated information and services, is

 

of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2004 by the American Academy published, and trademarked by the American Academy of Pediatrics, 141 Northwest Pointpublication, it has been published continuously since 1948. PEDIATRICS is owned, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

at Indonesia:AAP Sponsored on May 27, 2014pediatrics.aappublications.orgDownloaded from at Indonesia:AAP Sponsored on May 27, 2014pediatrics.aappublications.orgDownloaded from

The Mortality Index for Neonatal Transportation Score: A New MortalityPrediction Model for Retrieved Neonates

Simon J. Broughton, MRCP*; Andrew Berry, FRACP‡; Stephen Jacobe, FRACP‡; Paul Cheeseman, PhD*;William O. Tarnow-Mordi, MRCP, DCH, FRCPCH§; and Neonatal Intensive Care Unit Study Group§;

Anne Greenough, MD, FRCP*

ABSTRACT. Objective. To develop a mortality pre-diction score for retrieved neonates based on the infor-mation given at the first telephone contact with a re-trieval service.

Methods. Data from the New South Wales Newbornand Pediatric Emergency Transport Service databasewere examined. Analysis was performed with the resultsfor 2504 infants (median gestational age: 36 weeks; range:24–43 weeks) who were <72 hours of age at the time ofreferral and whose outcome (neonatal death or survival)was known. The study population was divided randomlyinto 2 halves, the derivation and validation cohorts. Uni-variate analysis was performed to identify variables inthe derivation cohort related to neonatal death. The vari-ables were entered into a multivariate logistic regressionanalysis with neonatal death as the outcome. Receiveroperator characteristic (ROC) curves were constructedwith the regression model and data from the derivationcohort and then the validation cohort. The results wereused to generate an integer-based score, the MortalityIndex for Neonatal Transportation (MINT) score. ROCcurves were constructed to assess the ability of the MINTscore to predict perinatal and neonatal death.

Results. A 7-variable (Apgar score at 1 minute, birthweight, presence of a congenital anomaly, and infant’sage, pH, arterial partial pressure of oxygen, and heart rateat the time of the call) model was constructed that gen-erated areas under ROC curves of 0.82 and 0.83 for thederivation and validation cohorts, respectively. The 7variables were then used to generate the MINT score,which gave areas under ROC curves of 0.80 for bothneonatal and perinatal death.

Conclusion. Data collected at the first telephone con-tact by the referring hospital with a regionalized trans-port service can identify neonates at the greatest riskof dying. Pediatrics 2004;114:e424–e428. URL: www.pediatrics.org/cgi/doi/10.1542/peds.2003-0960-L; neonatalmortality, retrieval, neonatal transport.

ABBREVIATIONS. CRIB, Clinical Risk Index for Babies; SNAP,Score for Neonatal Acute Physiology; NETS, Newborn and Pedi-

atric Emergency Transport Service; NICUS, Neonatal IntensiveCare Unit Study; ROC, receiver operator characteristic; Pao2, ar-terial partial pressure of oxygen; Paco2, arterial partial pressure ofcarbon dioxide; MINT, Mortality Index for Neonatal Transporta-tion; Fio2, fraction of inspired oxygen; TRIPS, Transport RiskIndex of Physiologic Stability; VLBW, very low birth weight; CI,confidence interval.

Provision of the most effective neonatal trans-port service requires accurate assessment ofdisease severity and prediction of prognosis, to

facilitate appropriate triage and resource allocation.The transport process starts at the time the retrievalservice receives the first call from the referring hos-pital; therefore, it is desirable to predict outcomesaccurately at that point of contact. Unfortunately,scores developed for assessment of infants duringthe transport process have used data acquired afterthe retrieval team has arrived at the referring hospi-tal.1,2

Prediction-of-mortality scores exist for the re-trieved pediatric population and include the Pediat-ric Index of Mortality score3 and the pre-intensivecare unit Pediatric Risk of Mortality,4 but such scoresare also calculated from data obtained at the trans-port team’s first physical contact with the patient. Inaddition, neither of these scores is easily applicableto a neonatal population, because they are heavilydependent on assessment of the level of conscious-ness and pupillary signs. Other neonatal predictionscores, including the Berlin score,5 the Score for Neo-natal Acute Physiology (SNAP),6 and the ClinicalRisk Index for Babies (CRIB),7 were not developed toassess the outcomes of retrieved neonates and havefactors that limit their use in such populations. TheBerlin score requires classification of the degree ofrespiratory distress, and the SNAP has 16 variablesand thus is time-consuming to calculate. Both theCRIB and SNAP use data collected over 12 hours andthus may reflect the effects of interventions ratherthan the underlying risk at an early time point. TheNational Institute of Child Health and Human De-velopment network mortality prediction score8 andthe CRIB II score9 are both based on informationavailable shortly after birth but were developed foruse among very low birth weight (VLBW) infants,whereas retrieved infants have a wide range of birthweights and gestational ages. Therefore, the aim ofthis study was to develop a new mortality predictionscore for the retrieved neonatal population that was

From the *Department of Child Health, Guy’s, King’s, and St. Thomas’School of Medicine, King’s College, London, United Kingdom; ‡New SouthWales Newborn and Pediatric Emergency Transport Service, Wentworth-ville, Australia; and §Westmead Hospital Perinatal Centre, University ofSydney, Sydney, Australia.Accepted for publication Apr 26, 2004.doi:10.1542/peds.2003-0960-LAddress correspondence to Anne Greenough, MD, FRCP, Department ofChild Health, Kings College Hospital, Denmark Hill, London SE21 8DE,United Kingdom. E-mail: anne.greenough@kcl.ac.ukPEDIATRICS (ISSN 0031 4005). Copyright © 2004 by the American Acad-emy of Pediatrics.

e424 PEDIATRICS Vol. 114 No. 4 October 2004 www.pediatrics.org/cgi/doi/10.1542/peds.2003-0960-L at Indonesia:AAP Sponsored on May 27, 2014pediatrics.aappublications.orgDownloaded from

based on data collected at the time of the first call bythe referring hospital, when resource allocation isdecided. In addition, we wished to determinewhether the score predicted death in the VLBW pop-ulation more accurately than did gestational age orbirth weight.

METHODS

Data CollectionData were obtained from the New South Wales Newborn and

Pediatric Emergency Transport Service (NETS) database. NETSprovides an integrated transport service for New South Wales andretrieves �1650 patients each year, 50% of whom are neonates.The retrieval process is started when a clinician from the referringhospital contacts NETS. During that call, the infant’s name, dateand time of birth, medical history, and clinical data (gestationalage, birth weight, gender, and Apgar scores at 1 and 5 minutes)are recorded (time of first call data). If deemed necessary, a re-trieval team (specialist retrieval nurse and doctor) is then mobi-lized. After arrival at the referring hospital, the retrieval teamcollects additional data (time of first contact data). The infant isstabilized by the retrieval team and, when the team is ready todepart, another set of data is collected (time of stabilization data).The infant is then transported to the accepting intensive care unit,where a final set of data is collected (time of admission data). Ateach of the 4 time points, collection of the following data wasattempted: heart rate, respiratory rate, fraction of inspired oxygen(Fio2), arterial partial pressure of oxygen (Pao2), arterial partialpressure of carbon dioxide (Paco2), pH, base excess, bicarbonatelevel, oxygen saturation, and ventilator settings. Outcome data(neonatal death or survival) were obtained from the NeonatalIntensive Care Unit Study (NICUS) database, which contains dataon the outcomes of neonates admitted in New South Wales, Aus-tralia. Ethical approval for this study was obtained from WesternSydney Area Health Service, and permission to use the data fromthe NICUS database was obtained from NICUS and each of thelocal hospital consultants responsible for the NICUS database.

Model DevelopmentData for infants who were �72 hours of age at the time of the

first call, who had complete demographic data and blood gas datafor 2 of the 4 time points, and whose outcomes were known wereincluded in the analysis. The study population was divided ran-domly into 2 halves, ie, the derivation cohort and the validationcohort. The strategy described by Pollack et al10 was used to build

the predictive model.11,12 The derivation cohort was used formodel derivation, and the model’s accuracy was tested with thevalidation cohort.

Univariate analysis was used to determine whether age (inhours), gestational age, gender, birth weight, Apgar scores at 1and 5 minutes, temperature, heart rate, respiratory rate, Fio2,intubation status, Pao2, Paco2, pH, base excess, bicarbonate level,oxygen saturation, presence of a congenital abnormality, oxygen-ation index, time between the different time points in the retrievalprocess, or total retrieval time differed (P � .2, with �2 or Mann-Whitney analysis as appropriate) between the infants who diedand those who survived. The variables that did differ were en-tered into a multivariate logistic regression analysis with forwardstepwise entry, with death as the outcome. All of the variables thatdiffered between the 2 groups at P � .2 were initially entered, andthe least significant variables, identified by their logistic coeffi-cient, odds ratio, and confidence intervals (CIs), were removed 1at a time. At each stage, the resulting model was assessed withreceiver operator characteristic (ROC) curves. Goodness-of-fit test-ing (with the Hosmer-Lemeshow goodness-of-fit test13) was usedfor both the derivation and validation cohorts. A P value of �.05implied no significant difference between the observed and ex-pected values, and the goodness of fit was considered acceptable.

With the logistic coefficients, odds ratios, and CIs, integer scorepoints were assigned to each of the variables and a score wasgenerated, the Mortality Index for Neonatal Transportation(MINT) score. Cutoff points for the individual variables wereobtained by assessing the model cutoff points used in clinicalpractice and reported in the literature. ROC curves were con-structed to determine the accuracy with which the MINT scorepredicted perinatal or neonatal death for the whole populationand then for VLBW infants only. ROC curves were also con-structed to assess whether gestational age or birth weight pre-dicted death in the VLBW infant population.

PatientsDuring the study period (February 1992 to July 2001), 8806

infants were retrieved, of whom 6348 (72.1%) were �72 hours ofage at the time of the first call (Table 1). Complete demographicand physiologic data were available for 3429 infants; of those 3429infants, 2504 had outcome data recorded in the NICUS databaseand 302 died (12% mortality rate). The 2504 infants (1585 malepatients) formed the study population. The median gestationalage was 36 weeks (range: 24-43 weeks) and the birth weight was2782 g (range: 520-6140 g). The median age at the time of the firstcontact with the retrieval service was 4.5 hours (range: 0-67.9

TABLE 1. Comparison of the Characteristics of All Infants Retrieved During the Study Period, Those Retrieved in the First 72 Hours,Those Retrieved in the First 72 Hours With Complete Physiologic Data, and Those Retrieved in the First 72 Hours with Physiologic andOutcome Data (Study Population)

All RetrievedInfants

All Infants Infants Retrieved in 72 h

CompleteDemographic andPhysiologic Data

StudyPopulation

No. 8806 6348 3429 2504Age, h 7.8 (0–659) 4.7 (0–67.9) 4.5 (0–67.9) 4.5 (0–67.9)Gestational age, wk 37 (23–44) 37 (23–44) 37 (23–44) 36 (24–43)Birth weight, g 2710 (520–6140) 2800 (520–6140) 2800 (520–6140) 2782 (520–6140)Male, no. 5345 (60.6%) 3827 (60.2%) 2100 (61.2%) 1585 (63.2%)Temperature, °C 36.7 (30–40) 36.6 (30–40) 36.7 (32–40) 36.6 (32–40)Heart rate, beats per min 140 (40–300) 140 (40–300) 140 (40–256) 142 (50–236)Respiratory rate, breaths per min 60 (7–135) 62 (10–135) 68 (20–135) 68 (20–135)Fio2 0.55 (0.21–1.0) 0.6 (0.21–1.0) 0.7 (0.21–1.0) 0.7 (0.21–1.0)Oxygen saturation, % 96 (15–100) 95 (15–100) 95 (15–100) 95 (15–100)Apgar score at 1 min 6 (0–10) 6 (0–10) 6 (0–10) 6 (0–10)Apgar score at 5 min 8 (0–10) 8 (0–10) 8 (0–10) 8 (0–10)Intubated, no. 705 (8.0%) 550 (8.7%) 422 (12.3%) 313 (12.5%)Pao2, kPa 9.8 (1.1–80) 10.2 (1.1–80) 9.7 (1.1–78.5) 9.53 (1.1–78.5)Paco2, kPa 5.7 (1.1–20.9) 5.7 (1.1–20.9) 5.7 (1.1–20.9) 6 (1.1–20.9)pH 7.3 (6.47–7.9) 7.29 (6.47–7.7) 7.29 (6.47–7.63) 7.28 (6.47–7.63)Base excess �4.8 (�33 to 25) �5.0 (�33 to 20) �5 (�30 to 20) �5 (�30 to 20)Bicarbonate mmol/L 21 (1.2–40) 20.6 (2–38) 20.8 (2–40) 20.9 (2–40)

Results are presented as median (range) or number (%).

www.pediatrics.org/cgi/doi/10.1542/peds.2003-0960-L e425 at Indonesia:AAP Sponsored on May 27, 2014pediatrics.aappublications.orgDownloaded from

hours); an age of 0 was recorded when the referring hospital calledthe retrieval service before the infant was born.

RESULTSThe study population was significantly younger

than the NETS population, because only infants �72hours of age were included in the analysis (Table 1).The study patients also had significantly greater ox-ygen requirements and were more likely to be intu-bated at the time of the call.

The only significant difference between the deri-vation and validation cohorts was the proportion ofinfants who were intubated (Table 2). Univariateanalyses indicated that postnatal age, gestationalage, gender, birth weight, Apgar scores at 1 and 5minutes, respiratory rate, intubation status, Pao2,pH, base excess, and presence of a congenital abnor-mality differed (P � .2) between infants who diedand those who survived (data not shown).

From the multivariate logistic regression analysis,a 7-variable logistic regression model was generated(Table 3). The 7-variable model was shown to havethe same area under the ROC curve as a model thatcontained all of the variables. Removal of additionalvariables resulted in a smaller area under the ROCcurve; therefore, the 7-variable model was used. Theequation used to generate the probability of deathfrom the model is as follows:

logit � �25.53 � 2.50exp�02 · age � 0.29 · Apgar

score at 1 minute � 2.07exp�04 · birth weight

� 0.88exp�03 · Pao2 � 3.74 · pH � 1.75 · congenital

abnormality � 1.23 · intubation status

The probability of death (y) is given by the equationy � exp(logit)/[1 � exp(logit)]. Age is expressed inhours; for the presence of a congenital abnormalityand intubation status, yes � 1.

The model generated areas under the ROC curveof 0.82 for the generation cohort and 0.83 for thevalidation cohort. The goodness of fit (Hosmer-

Lemeshow test) for the derivation cohort was 0.364,and that for the validation cohort was 0.303.

With the information from the logistic regressionanalysis, the MINT score was derived (Table 4). Themedian MINT score for the study population was 3(range: 0-33). Eighty per cent of infants with MINTscores of �20 died (Fig 1). The MINT score had anarea under the ROC curve of 0.80 (95% CI: 0.76-0.83)for death in the perinatal period (first week afterbirth) and an area under the ROC curve of 0.80 (95%CI: 0.76-0.83) for death in the neonatal period (firstmonth after birth) (Fig 2). For VLBW infants, theMINT score had areas under the ROC curves forperinatal and neonatal death of 0.69 (95% CI: 0.60-0.77) and 0.68 (95% CI: 0.60-0.76), respectively. Ges-tational age and birth weight had areas under theROC curves of 0.64 (95% CI: 0.56-0.73) and 0.67 (95%CI: 0.59-0.76), respectively.

DISCUSSIONWe generated and validated a prediction of mor-

tality score (MINT score) for retrieved neonates thatwas based on data obtained at the first referral call.The MINT score exhibited performance similar tothat of the Transport Risk Index of Physiologic Sta-bility (TRIPS),1 which was generated with a similarpopulation (the median gestational age of the studypopulation used for generation of the TRIPS was 36weeks and the birth weight was 2610 g). The areas

TABLE 2. Characteristics of the Derivation and Validation Cohorts

Derivation Cohort Validation Cohort P

No. 1252 1252Postnatal age, h 4.7 (0–67.9) 5.1 (0–67.4) .88Gestational age, wk 36 (24–42) 36 (24–43) .70Male, no. 794 (63.4) 794 (63.6) .95Birth weight, g 2840 (580–5950) 2810 (520–5835) .75Apgar score at 1 min 6 (0–10) 6 (0–10) .62Apgar score at 5 min 8 (0–10) 8 (0–10) .67Heart rate, beats per min 142 (50–204) 142 (100–236) .71Respiratory rate breaths per min 68 (20–140) 68 (20–166) .50Fio2 0.7 (0.21–1.0) 0.7 (0.21–1.0) .20Intubated, no. 213 (17.0) 169 (13.5) .04Pao2, mm Hg 73 (16–590) 70 (8–539) .09Paco2, mm Hg 44 (11–157) 45 (8–142) .11pH 7.28 (6.53–7.63) 7.27 (6.47–7.61) .33Base excess �5 (�29 to �11) �5 (�29 to �13) .83Bicarbonate, mmol/L 20.4 (4.2–40) 21 (2.0–38) .27Oxygen saturation, % 95 (22–100) 95 (20–100) .28Congenital abnormality, no. 226 (18.0) 226 (18.0) .28Died, no. 138 (11.0) 164 (13.1) .20

The data are displayed as number (%) or median (range).

TABLE 3. Logistic Coefficients, Odds Ratios, and 95% CIs ofthe 7 Variables in the Model

Variable LogisticCoefficient

(�)

OddsRatio

95% CI

Postnatal age �0.025 0.976 0.965–0.985Apgar score at 1 min 0.294 1.341 1.254–1.435Birth weight 0.002 1.002 1.000–1.004Pao2 �0.008 0.997 0.994–0.999pH 3.744 42.284 13.689–130.618Congenital abnormality �1.747 0.174 0.122–0.249Intubated �1.227 0.292 0.203–0.423

e426 MINT SCORES at Indonesia:AAP Sponsored on May 27, 2014pediatrics.aappublications.orgDownloaded from

under the ROC curves for perinatal and neonataldeath for the TRIPS were 0.83 and 0.76 and those forthe MINT score were 0.80 and 0.80, respectively. TheTRIPS, however, is derived from data collected by amember of the transport team immediately after ar-rival at the referring hospital and immediately afterarrival at the destination hospital.1 In contrast, theMINT score uses data collected when the referringhospital first contacts the transport team via tele-phone. This is a major advantage, because decisionsare made at first contact regarding resource alloca-tion. The MINT score has an additional advantage,ie, it is based on 7 objective data items; although theTRIPS comprises only 4 items, 1 is the response tonoxious stimuli, which is subjective.

The study population differed from the total NETSpopulation, because we included only infants forwhom the transport process started at �72 hours of

age (when most neonatal retrievals occur). We alsoincluded only infants with complete demographicdata and blood gas data for at least 2 of the 4 timepoints. The study population differed significantlyfrom the NETS population with respect to the pro-portion of patients intubated and their greater oxy-gen requirements. We were thus examining the sick-est infants transported, as highlighted by themortality rate of 12%, compared with the rate of 10%for the whole NETS population. A mortality predic-tion score would be most useful for the sickest in-fants. The MINT score, however, was developedwith only 39% of the eligible cohort, and we recom-mend that it be prospectively validated before it isput into widespread use.

The MINT score comprises 7 variables, includingthe Apgar score at 1 minute. The 5-minute Apgarscore, rather than the 1-minute score, has been con-sidered to be more predictive of neonatal death.14

However, the National Institute of Child Health andHuman Development mortality prediction model8also used the 1-minute Apgar score. In this study,both 1- and 5-minute Apgar scores were available forall of the infants included and both were analyzed,but the 1-minute Apgar score performed better. ThePao2/Fio2 ratio is included in the SNAP as a mea-sure of oxygenation.6 It was selected, rather than theoxygenation index, arterial-alveolar oxygen differ-ence, or arterial-alveolar oxygen ratio, for the SNAPbecause it was statistically equivalent and avoidedthe need to determine mean airway pressure or con-current carbon dioxide tension. In our study, wefound that arterial oxygen tension, but not the in-spired oxygen concentration or oxygenation index,was significantly related to death. A possible expla-nation for the difference was that more than one-fourth of the infants were receiving 100% oxygen atthe time of the referral call. We excluded base excessfrom our score because of colinearity and because,although both pH and base excess were highly pre-dictive, pH performed better. We included congeni-tal abnormality in our analysis because congenitalabnormalities are known to have mortality effectsbeyond those indicated by physiologic derange-ments.6 Indeed, the analysis demonstrated that thepresence of a congenital abnormality was signifi-cantly associated with death, and that parameter was

Fig 1. Relationship of mortality rate to MINT score (�, number ofinfants; ■, mortality rate).

Fig 2. ROC curve for death in the neonatal period.

TABLE 4. MINT Score Point Allocation

% Died Points

pH�6.9 59.52 106.91–7.1 23.78 4�7.1 10.36 0

Age0–1 h 25.16 4�1 h 10.24 0

Apgar score at 1 min0 44.44 81 30.10 52 22.11 23 18.50 2�3 7.49 0

Birth weight�750 g 62.50 5751–1000 g 36.00 21001–1500 g 19.05 1�1500 g 10.74 0

Pao2�3 kPa 28.57 2�3 kPa 11.87 0

Congenital abnormalityYes 22.27 5No 9.55 0

Intubated at time of callYes 26.20 6No 10.04 0

Maximum 40

www.pediatrics.org/cgi/doi/10.1542/peds.2003-0960-L e427 at Indonesia:AAP Sponsored on May 27, 2014pediatrics.aappublications.orgDownloaded from

included in our model. Unfortunately, only the pres-ence or absence of a congenital abnormality wasrecorded at the time of the referral call. Therefore, wecannot comment on whether weighting for the sever-ity of the abnormality might have generated a moreaccurate score. A better standardized classificationsystem must be available,15 however, before this canbe appropriately investigated.

CONCLUSIONS

We have generated and validated an easy-to-usemortality prediction score for retrieved neonates.Such scores should not be used to ration health care;instead, a high score should be used to indicate thelevel or priority and the need for the most experi-enced transport team. All of the data used in theMINT score can be collected at the time of the firsttelephone contact by the referring hospital with thetransport team. This score might be particularly use-ful because it could facilitate more effective triage.

REFERENCES

1. Lee SK, Zupancic JA, Pendray M, et al. Transport Risk Index of Phys-iologic Stability: a practical system for assessing infant transport care.J Pediatr. 2001;139:220–226

2. Ferrara A, Atakent Y. Neonatal Stabilization Score: a quantitativemethod of auditing medical care in transported newborns weighing lessthan 1,000 g at birth. Med Care. 1986;24:179–187

3. Shann F, Pearson G, Slater A, Wilkinson K. Paediatric Index of Mortality(PIM): a mortality prediction model for children in intensive care.Intensive Care Med. 1997;23:201–207

4. Kanter RK, Edge WE, Caldwell CR, Nocera MA, Orr RA. Pediatricmortality probability estimated from pre-ICU severity of illness. Pedi-atrics. 1997;99:59–63

5. Maier RF, Rey M, Metze BC, Obladen M. Comparison of mortality risk:a score for very low birthweight infants. Arch Dis Child Fetal Neonatal Ed.1997;76:F146–F150

6. Richardson DK, Gray JE, McCormick MC, Workman K, Goldmann DA.Score for Neonatal Acute Physiology: a physiologic severity index forneonatal intensive care. Pediatrics. 1993;91:617–623

7. The International Neonatal Network. The CRIB (Clinical Risk Indexfor Babies) score: a tool for assessing initial neonatal risk and com-paring performance of neonatal intensive care units. Lancet. 1993;342:193–198

8. Horbar JD, Onstad L, Wright E. Predicting mortality risk for infantsweighing 501 to 1500 grams at birth: a National Institutes of HealthNeonatal Research Network report. Crit Care Med. 1993;21:12–18

9. Parry G, Tucker J, Tarnow-Mordi W. CRIB II: an update of the ClinicalRisk Index for Babies score. Lancet. 2003;361:1789–1791

10. Pollack MM, Ruttimann UE, Getson PR. Pediatric Risk of Mortality(PRISM) score. Crit Care Med. 1988;16:1110–1116

11. Lemeshow S, Le Gall JR. Modeling the severity of illness of ICUpatients: a systems update. JAMA. 1994;272:1049–1055

12. Laupacis A, Sekar N, Stiell IG. Clinical prediction rules: a review andsuggested modifications of methodological standards. JAMA. 1997;277:488–494

13. Hosmer DW, Hjort NL. Goodness-of-fit processes for logisticregression: simulation results. Stat Med. 2002;21:2723–2738

14. Casey BM, McIntire DD, Leveno KJ. The continuing value of the Apgarscore for the assessment of newborn infants. N Engl J Med. 2001;344:467–471

15. Richardson DK, Corcoran JD, Escobar GJ, Lee SK. SNAP-II andSNAPPE-II: simplified newborn illness severity and mortality riskscores. J Pediatr. 2001;138:92–100

e428 MINT SCORES at Indonesia:AAP Sponsored on May 27, 2014pediatrics.aappublications.orgDownloaded from

DOI: 10.1542/peds.2003-0960-L 2004;114;e424Pediatrics

Tarnow-Mordi and Anne GreenoughSimon J. Broughton, Andrew Berry, Stephen Jacobe, Paul Cheeseman, William O.

Prediction Model for Retrieved NeonatesThe Mortality Index for Neonatal Transportation Score: A New Mortality

  

ServicesUpdated Information &

tmlhttp://pediatrics.aappublications.org/content/114/4/e424.full.hincluding high resolution figures, can be found at:

References

tml#ref-list-1http://pediatrics.aappublications.org/content/114/4/e424.full.hat:This article cites 15 articles, 3 of which can be accessed free

Citations

tml#related-urlshttp://pediatrics.aappublications.org/content/114/4/e424.full.hThis article has been cited by 1 HighWire-hosted articles:

Subspecialty Collections

y_subhttp://pediatrics.aappublications.org/cgi/collection/neonatologNeonatology

orn_infant_subhttp://pediatrics.aappublications.org/cgi/collection/fetus:newbFetus/Newborn Infant

medicine_subhttp://pediatrics.aappublications.org/cgi/collection/transport_Transport Medicinethe following collection(s):This article, along with others on similar topics, appears in

Permissions & Licensing

mlhttp://pediatrics.aappublications.org/site/misc/Permissions.xhttables) or in its entirety can be found online at: Information about reproducing this article in parts (figures,

Reprints http://pediatrics.aappublications.org/site/misc/reprints.xhtml

Information about ordering reprints can be found online:

rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.Grove Village, Illinois, 60007. Copyright © 2004 by the American Academy of Pediatrics. All and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elkpublication, it has been published continuously since 1948. PEDIATRICS is owned, published, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

at Indonesia:AAP Sponsored on May 27, 2014pediatrics.aappublications.orgDownloaded from