Comprehensive
Computerized
Neonatal
Intensive
Care
Unit
Data
System
Including
Real-Time,
Computer-Generated
Daily
Progress
Notes
William W. Lowe, MD; Thomas A. Ciszek, MD; and Keith
J. Gallaher,
MDABSTRACT. A neonatal intensive care unit patient data
system, NeoData, which was developed using
microcom-puters connected by a local area network, is described.
The system allows for real-time generation of daily
progress notes, as well as admission and discharge sum-manes. It includes two databases: one for daily patient
data and one for admission/discharge summary data.
Both sets of data are easily accessible for later analysis and report generation. The daily patient data are entered
directly into a computer by the neonatal intensive care
unit medical and nurse practitioner staff; a progress note
is printed immediately thereafter for inclusion in the
patient’s chart. Data from the previous day are selectively carried forward into the current day’s note, minimizing
data entry. Several benefits are derived from this
progress note system, including legibility, tracking of
laboratory and other data, tracking of management plans and procedures due at a later date, and significant time
savings. The system has proved to be easy to learn, and
the neonatal intensive care unit staff have found it to
contribute to the efficient delivery of patient care.
Pedi-atrics 1992;89:62-66; computer program, neonatal
inten-sive care unit, database, data management, information
systems.
ABBREVIATIONS. NICU, neonatal intensive care unit; NNP,
neo-natal nurse practitioner; LAN, local area network.
Computer technology has been applied to a variety
of inpatient activities. These include calculation of
hyperalimentation infusions,’ unit dose medication
management,2 collection and management of
hospi-talization summary data,” automated generation of
official admission or discharge summaries,6’ ‘ and
charting of patient vital signs.’2’4 Several state-wide
neonatal intensive care unit (NICU) data systems
have been implemented, where software from one
center has been made available to other NICUs for
use.”
Although early projects were based on centralized
mainframe computers, since 1979 small
microcom-puter systems have often been used instead. Use of
personal computers allows for local project control
while dependence on the hospital data-processing
department and competition for computing resources
are reduced. During the 1980s, sophisticated
proprie-tary data management software became available,
From the Cape Fear Valley Medical Center, Fayetteville, NC.
Received for publication Oct 31, 1990; accepted Feb 20, 1990.
Reprint requests to (W.W.L.) Cape Fear Valley Medical Center, P0 Box 2000,
Fayetteville, NC 28302-2000.
PEDIATRICS (ISSN 0031 4005). Copyright © 1992 by the American
Acad-emy of Pediatrics.
allowing project development by computer
non-professionals.’#{176} Most of the computerized data
sys-tems described in the pediatric literature have been
designed so that the process of data entry and the
printing of hospital summaries are performed by
cler-ical staff, usually after the patient’s admission or
discharge. In the initial phase of our project, NeoData,
we developed such a discharge summary system,
which has been in use for more than 3#{189}years and
has successfully eliminated the need for dictation and
transcription of summaries.
We recently extended the system by adding the
capability for real-time production of daily progress
notes; this has proved to be reliable and time-efficient.
Instead of handwriting daily notes, information is
entered into a database and the computer prints the
progress note for that day. Because the system can
selectively carry data forward from the previous day’s
note, only the incremental daily changes need be
entered, thus decreasing data entry time. This allows a limited physician and neonatal nurse practitioner
(NNP) team to care for a larger number of babies
with more time at the bedside. Similarly, the system
can also generate admission summaries immediately
upon patient admission. The patient care team
mem-bers perform all data entry for the admission and
progress notes, resulting in improved accuracy and
eliminating reliance on clerical staff.
Any computerized system for handling patient data
will be less flexible than free-form, text-based systems
such as handwritten or dictated notes and summaries.
Through careful design of data fields, and by
includ-ing some free-form text entry fields, NeoData has
proved to be flexible enough to handle the majority
of patient problems encountered. In return for the
limitations of computerization, we have found
NeoData to provide a number of benefits, including
the following: (1) increased legibility; (2) consistent
terminology; (3) consistent note format; (4)
compila-tion of detailed daily patient data for analysis; (5)
improved tracking of data and of management plans;
and (6) perceived time savings.
HARDWARE
A real-time computerized system requires greater
hardware and software resources than a system in
which data entry is performed after discharge,
pri-manly because such a system must be used by more
than one person at a time. The change from a
single-user to a multiuser system requires multiple
micro-computers linked by a communication network, so
!tht_NlPcI 416 (Dos.?. 12/2I9O), <ENC> to #{149}SLt
NESPINATORY
Acttv#{149},j Nssptratory Probl.s, Sssrity Status,
1* RDI#{149} #{149}old
__________
Support , nasal CPA? Ga... ‘ Dat#{149} Tiu.. pa pC02 p02
Sine. , JjJ5 1, ciakaLa!L2.Q 2.LuLJA it ._41 Fi02s ,j5% Flovz_ce/.in 2___
PIP,
_
PEEP: _5 ISV, IT. 3.Oth#{149}rProcsdur#{149}s. Ondicati,
Plan.
1:
Oth#{149}r.
1NDS
7) trans&snt tschypn.a
3) r..piratory distrasa 4) SPO
5) r..piratory d.pr.ssion ‘) r,.puratory arr..t
7) raspiratory fanlur.
I) sconiu* aspiration syndroa
9) fluid aspiration syndrom.
centralized data file. Recently, networking hardware
and software for IBM-compatible microcomputers has
become widely available and increasingly reliable.
Our hardware system includes three
IBM-compat-ible computers, based on 80386SX microprocessors,
which are used for data entry and are located near
the NICU. Each has an internal hard disk for software storage, a high-resolution color monitor, a dot-matrix printer, and an uninterruptible power supply
provid-ing 5 to 10 minutes of battery-powered operation in
the event of a power outage. All patient-related data files are stored on a high-speed hard disk residing on
a fourth, 80386-microprocessor-based computer (file
server) that is operated in a ‘dedicated’ fashion, used
only to serve the file-access needs of the network.
Several other computers are used by the Neonatology clerical staff for discharge data entry. A total of eight computers are currently connected by
Ethernet-com-patible Local Area Network (LAN) hardware
(ART!-SOFT, Inc. Tucson, AZ).
Our computers were purchased from mail-order
firms selected for the price-to-performance ratio of
their equipment and for their high-quality technical
support. The data entry systems were purchased for
approximately $3000 per station, including monitor,
printer, network hardware, and uninterruptible
power supply; the file server was purchased for
ap-proximately $5000. Further details of the hardware
configuration for our system are available on request.
SOFTWARE
The computers in our system use the Microsoft
Disk Operating System version 5.0 and the LANtastic
Network Operating System version 4.0 (ARTISOFT,
mc). To handle the actual entry and processing of
patient data, we use MetaForm, a database program
developed internally. A wide range of commercial
data base programs are available for IBM-compatible
computers; however, by developing our own software
we have been able, at each step of the project, to add
those features we believe are needed to best handle
patient-related medical information.
Some of the features of MetaForm that are
partic-ularly useful in developing medical data applications are (1) a rich choice of data types (number fields, text
fields, etc), allowing for very compact computer
rep-resentation of information; (2) category lists, a pow-erful data type that allows the user to select a field
entry from a list of choices (lists may be searched
quickly and easily, may be altered as needed, and
guarantee that common terminology will be used by
all staff involved in data entry); (3) multiple field
copies (any field may be defined to allow multiple
instances of itself, eliminating the need for repeated
fields that contain the same type of information; (4)
field groups, which are related fields defined to allow
single operations on the entire group at once; (5)
calculated fields, whose values are based on data in
other fields (this feature is supported by a selection of numerical, logical, and text functions used to
con-struct formulas); (6) field conditions (any field can be
assigned a formula that determines whether the field
will be active or inactive, depending on the values in
other fields); (7) a powerful document-generation
ca-pability, allowing complex decision trees to be
exe-cuted during the composition of summaries and other
printouts; (8) a search capability, which can be used
to construct matrices of conditions for extraction of
summary data; and (9) a built-in programming
lan-guage, which can be used to design customized,
menu-based software routines to simplify standard
tasks and limit the access of users to only those
features of the program they need.
The NeoData system includes two separate data
files: NOTES, containing the daily progress
informa-tion, and NEOSUM, containing admission and
dis-charge data. Daily progress notes use the NOTES file,
discharge summaries use the NEOSUM file, and
ad-mission summaries use a subset of both. The entry of
final discharge data and the printing of discharge
summaries are performed by the Neonatology
secre-tarial staff away from the NICU, typically on the day
of discharge or on the next weekday. Of 1 1 1 8 patients
admitted to and discharged from the Neonatology
service between July 1, 1988 and October 31, 1990,
only 7 (0.63%) have required dictated addenda.
The generation of admission and progress notes is
accomplished on computers located in the NICU. To
create and print a daily progress note, the NeoData
menu allows the user to update a list of current NICU
patients, generate new notes based on those of the
previous day, edit these notes, and print them.
Ad-mission notes are processed from a similar menu. In
both cases, the data are entered directly into the
computer by the NICU physician or NNP. The work
can be performed whenever the appropriate data are
available, at any time of the day or night. If necessary,
the process may be interrupted and completed later.
For the actual data entry process, the user is pre-sented with a series of 12 data screens, displaying the
250 fields (many with multiple copies) of the NOTES
data file (Figs 1 and 2). One can move quickly and
easily from screen to screen, backward or forward.
Each screen includes one or two sections comprising
related information (Table); problem-oriented
infor-mation is organized into sections roughly by organ
system. Medications, laboratory values, and
roent-NOTES
Paq#{149}4
7-T#{149}b-991 09:30,47
Fig. 1. The fourth NOTES data screen. This shows the
RESPIRA-TORY section. Note that the cursor ison the ‘Respiratory Problems’
field, which is an example of a category list field. The window in
the lower right corner displays the list for this field, which can be
scrolled up or down to find the desired entry. Since “nasal CPAP’
is entered in the ‘Support’ field, no data can be entered into the
Edot POol 414 Do..?. 1/1/91), <tic’ to SSlt
NOTES 7-Y.b-l991 09:30:47
---.-
FL010*/ELECT*OLYTES/NUTNITION I---
--old wt. 1354qm Calculat.ons based on, Old: 1354)qm
New Vt. Il33Sq. p.r kgper day
--,
N#{149}w,Il33Nq’Iv Fluids. D.Str/sS SA sa K Ca N*tS LtpLd Rat.(24h) 1 .0ld.5 Dj9.,,Q/5,,,,, _g _jQaEq ,j,,QeEq Ja.g S.1 cc/h
_
_g I cc/hN.v,uy DjQ,j/5,,,,,, _g _L,QaEq .J.,,QsEq j3mg 4. 7cc/h _g cc/h Lyt.s/lOOcc. 2.CJatq 12.4IaEq 111R9
2.Old,,,,, D_/, _9 _aEq _mSq eq _cc/h
Nsw._ D_/_ _q _atq _s.Eq eq _cc/h Total Fluods, (-IV1)
3.Old._ D,,,,,,1_ _g _.eq _sZq eq _cc/h Old,IlSO)cc/kg/d1591
New._ D,,,,,,,,,/_ _9 _eEq _aSq eq_cc/h s.w.1S0Jcc/kg/d(kj
Total Caloroea old. IS6lcal/kg/d
New, (67)cal/kg/d
TABLE. Data Sections Within the NOTES Data Set
Identification Physical Exam Laboratory Data X-rays Medications Respiratory Cardiovascular Infectious Hematologic Jaundice
Apnea and Bradycardia
Neurologic
Other
fluids/Electrolytes/Nutrition
Discharge Planning
Addendum
rasds. Typ. atr.nqth Aaount Sch.dul. POut.
old, sromatur jQcal/oz jQ,Qcc a3h
,.
New. Dromatur#{149} j5cal/oz j,,Qcc a3h 55,.,,...
OG/Po luppl. Aaount $ch.dul. Total Intaks Sad, old, _aEq q_hr (2.3mZq/kq/d
N.,,. _.zq q_hr 12.klaEqIkq/d
rd. Old, _agq q_hr l.6)!q/kg/d
5S5 _.Eq q,,hr 3.OIaSqIkg/d
Paqa 10
Fig. 2. The 10th NOTES data screen. This shows the FLUIDS/
ELECTROLYTES/NUTRITION section. Underlined areas represent
the actual data entry fields, while brackets enclose calculated fields.
Calculated fields are used heavily in this screen to help with writing
intravenous fluid orders and to calculate total daily calories and
sodium/potassium intake.
genographic results are entered into their respective
dedicated sections. All subsequent sections include
fields for problems or diagnoses, fields for
manage-ment plans, and in most cases, a free-form field into
which textual information may be entered.
Data screens also include calculated fields, partic-ularly in the Fluids/Electrolytes/Nutrition section. Here, fluid infusion rates, fluid intakes, and caloric intakes are automatically displayed based on desired intravenous fluid constituents and feeding regimens.
These may be used to assist in the writing of fluid
orders and many are included in the printed progress note (Fig 2).
COMPUTER-GENERATED NOTES AND
SUMMARIES
Once a patient’s information for the current day is
completely updated in the computer, the progress
note is printed, signed, and placed directly on the
patient’s chart by the NNP or physician.
Continuous-feed paper is used which has been commercially
printed to duplicate the appearance of our hospital’s
standard progress sheets. Figure 3 shows an example
of a printed progress note.
Maintaining the integrity of the information in a
legal patient chart is always important. Since it is very
difficult to ensure that computerized data has not
been altered subsequent to its creation, we make no
attempt to do so in the NeoData system. The printed
note, and not the electronically stored data,
consti-tutes the legal patient record for a given day.
Correc-tions, when needed, are made to the printed note and
initialed, just as if the note had been handwritten. NeoData does not attempt to replace the paper record in any sense, legal or otherwise; it simply provides an alternative means of generating that record.
When the progress note system was implemented,
the NNP and physician staff became functional users
within several days; full proficiency was attained by
each user within 2 to 4 weeks. The system has been
successful, as measured by its rapid and continued
acceptance by the medical team members. From the
onset, one goal of the NeoData project has been to
decrease the time required to generate progress notes and to allow the available staff to care for increasing
numbers of babies. We have found that it takes from
3 to 10 minutes to select a patient in the computer
file, update the information for the day, print and
sign the note, and place it on the chart. We recently
timed the process for 1 8 typical patients; the mean
time required by an experienced person to perform
the above sequence was 5.2 ± 0.7 minutes.
We have also examined our census for two 4-month
periods a year apart, one before the system was
implemented and one after. While the mean daily
NICU census increased from 23.6 to 24.7, our NNP
staff decreased from 6.67 to 5.33 full-time
equiva-lents. Nonetheless, under the new system the
prog-ress notes are consistently being completed earlier in
the day.
In addition to the generation of progress notes and
summaries, the NeoData files are used for several
other data-processing tasks. Daily progress note
in-formation, updated during the night, is used to print
a detailed report sheet for morning rounds. This
elim-mates the recitation of numerous clinical data to the
staff coming on duty and allows rounds to focus on
patient assessment and management plans. After
dis-charge, a simple program extracts summary
infor-mation from a patient’s daily notes which simplifies
final data entry for the discharge summary. Brief
weekly patient reports are produced which are sent
to referring physicians and insurance companies as
needed. Finally, the accumulated data sets are used
for regular and efficient production of morbidity and
mortality reports and for other data searches.
DISCUSSION
In most hospitals, computer technology has had a
smaller impact on the day-to-day activities of the
medical staff than on those of clerical, administrative, and nursing staff. Clerical, laboratory, medication,
and nursing flow-sheet data tend to be objective in
nature and are usually numerical or categorical.
Med-ical staff, on the other hand, deal with a wide range of objective and subjective information. It may be the
large amounts and diverse types of data, and above
all the subjectivity of some of the information, which
have made it more difficult to apply computerized
DOE .FE PROGRESS NOTE 12/29/90 1234567
12/29/90 Day 2. Weight 1483 quo (down 60gm).
1528hri PHYSICAL EXAM:
Temp stable. Vital signs stable. Blood pressure stable. Urine output good
(2.8cc/kg/hr). Pink and quiet. Anterior fontarlelle soft S flat. Respiratory
effort minimally increased. Breath sounds equal and diffuse crackles; air
exchange good. Cardiac: normal sinus rhythm and normal pulses. No murmur heard.
Perfusion good. Abdomen: soft and full. Neuro: responsive.
I,MOR,ATORY STUDIES:
12/29/90 6 0500: Hct:46 WBc:15.6K Plat:269K 57S 8B 26L 9M
12/29/90 0500: Na:143 1C:4.2 Cl:107 C02:23 BUN: 4 cr:l.4 Ca: 8.4
12/29/90 8 0500: Bili: 5.3/0.1
Chemstrips 80-l2Omg%.
XRAYS:
CXR on 12/28/90: mild reticulogranularity and normal heart size.
MEDICATIONS:
Curr#{149}nt medications: ainpicillin 105mg IV ql2h (day 1) and gentamicin 2.6mg IV
qleh (day 1).
RESPIRATORY:
PROBLIS: Mid EDS
SUPPORT: nasal CPA (since 12/28/90); Fi02:42% PEEP:5
CBG 12/29/90 5 0111: pN:7.31 pCO2: 51 p02: 48
The plan is to support as indicated.
INFECTIOUS:
PIO$L1$: possible sepsis
DRUG LEVELS: 12/28/90: 18 hr gentamicin: 1.0
CULTURES: 12/28/90: blood: negative
The plan is to continue ampicillin B gentamicin.
HEMATOLOGIC:
Tb. last hematocrit was 46 on 12/29/90. The plan is to follow hematocrits.
JAUNDICE:
PROILS$: at risk for jaundice
The patient is currently not under phototherapy. The last bilirubin level was
5.3/0.1 on 12/29/90 6 0500. The baby’s blood type is A poe. The direct Cooinbs is
negative. The plan is to follow clinically and follow bilirubin levels.
NEUROLOGIC:
PROILDIS: at risk for lvi
The plan is to obtain a cranial ultrasound for rule out IVH on 12/31/90.
FLUIDS/ELECTROLYTES/NUTRITION:
CURRENT FLUIDS: All additives are per kg per day; Ca is elemental calcium.
Ply: D10.0/W, Na 2.OmEq, K l.OmEq, Ca 15mg.
Total fluids: 80cc/kg/day. Total calories: 32kcal/kg/day.
NEW FLUIDS: All additives are per kg per day; Ca is elemental calcium.
Ply: D10.0/W, Na 1.OmEq, K 1.OmEq, Ca 25mg.
Total fluids: 120cc/kg/day. Total calories: 48cal/kg/day. The plan is to follow electrolytes.
DISCHARGE PLANNING:
The babys follow-up care after discharge will be with Womack Peds Clinic. The
baby meets criteria for hearing screening.
Fig. 3. The printed progress note. Although most sections allow several lines of free text entry, no such fields were used in this case.
There are several reasons why computerization of
discharge summaries has been, for many NICUs, the
first step in this process. First, the process can operate
‘off-line,’ in that data entry is often performed by
clerical personnel and can be done any time after
discharge; actual patient management is not
depend-ent on the immediate completion of the
computer-related task. Second, the medical staff is freed from the unpopular job of dictating summaries, providing
a very large ‘carrot’ which increases the chance for
success of the project. Third, the department also
benefits from a database containing information
about every patient discharged which is available for
later analysis.
We have extended this concept to what we believe
is the next logical stage: computerization of the
proc-ess of writing daily progress notes. A primary
moti-vation was the need to increase the productivity of
our medical team, which in our institution is
com-posed of attending neonatologists and NNPs. Our
institution elected to commit the necessary resources
to the NeoData project in order to reduce the time
the team spends writing notes, while not sacrificing the quality of our detailed, problem-oriented medical documentation.
A characteristic of patient-related medical
infor-mation that complicates its representation in comput-erized form is its high degree of internal structure.
For instance, one may wish to allow for multiple
entries of the same field (eg, diagnoses); these field
entries need to be tightly organized on the screen. In some cases, several fields are logically connected (eg,
pH, PC02, and P02); often, one would like to allow
for multiple entries of this entire group (eg, multiple
sets of blood gas entries). Whether a given field
should have data entered into it at all may depend
on the state of another field (eg, intermittent
man-datory ventilation data need to be entered only if a
patient is supported by a ventilator); in many
situa-tions, entire data sections may need to be skipped in
this manner. Whenever possible, we have designed
MetaForm so that data sets could be organized with
these structures in mind, resulting in a much closer
representation of actual patient information.
A previously identified problem with hospital
com-puterized data systems is that of maintaining the
commitment of the primary users to the system.’3 It
was clear at the outset that our software had to be
easy to learn and had to allow efficient data entry
primary users of the system found that it directly benefited them in their daily patient care duties. The efficient user interface, the features that introduce layers of structure into the data set, and the ability to easily handle large category lists are the core proper-ties that have made our software effective.
Reliability has been a particularly critical issue for
those computerized patient data systems that have
entirely replaced a previous paper-based approach. If
the system was ‘down’ for any significant amount of
time, patient care was affected. Our system does not
replace any component of the paper-based chart;
NeoData is simply another, more efficient way to
generate paper summaries and notes. Our ultimate
protection from system malfunction is that we can
always write notes by hand.
Nonetheless, every effort has been made to
maxi-mize the reliability of the hardware and software.
Since the system was fully implemented, we have
had very few problems that could not be solved
rapidly; one half of the network was inoperative for
several hours on one occasion, and no software
prob-lem significant enough to delay the note-writing proc-ess has occurred to date. Certainly the effort involved
in managing even a modest system such as ours is
significant, and the value of having at least one person
with a strong interest in microcomputers cannot be
overemphasized. The major commitment, however,
was in developing and debugging the system; we
estimate that more than 2000 hours were expended
by one of us (W.W.L.) to program MetaForm, and
another 100 to 200 hours to design the NEOSUM and
NOTES data sets. Maintaining the system once in
place, including backing up data files, requires only
approximately 2 to 4 hours per week.
CONCLUSIONS
We have described a system for the computer
gen-eration of progress notes which is currently in use in
a busy NICU. Computerization of daily progress notes
has had several benefits, aside from the primary goal
of time savings for the medical team. Legibility is
significantly increased, the layout of the notes has
become standardized, and terminology is more
con-sistent. These improvements have already drawn the
notice and comments of the nursing staff and the
hospital utilization review personnel. The data
screens remind the user to update laboratory values,
medication lists, problem lists, and management
plans; this results in fewer omissions. The system
automatically calculates various fluid and electrolyte
quantities, reducing errors associated with manual
calculations. Finally, the resulting detailed databases facilitate utilization review and quality-assurance monitoring.
The development of the NeoData system has been
a thought-provoking experience; it has given us an
increased appreciation of the efficiency of
paper-based data systems for the day-to-day operations of
patient care. We have tried to design NeoData to
simulate, whenever possible, the use of paper forms,
while invoking the specific capabilities of the
com-puter to add functions that are simply not otherwise available. The goal of the NeoData project has been,
from the outset, to meet the specific needs of the
medical team and to facilitate the delivery of patient
care. The project has successfully achieved these
ob-jectives and has demonstrated the feasibility of com-putenzed progress note generation.
ACKNOWLEDGMENTS
We acknowledge the helpful assistance, advice, and hard work
of the NNP staff and Neonatology secretaries. We also appreciate the financial and technical support of the Cape Fear Valley Medical
Center.
REFERENCES
1. Wilson FE, Yu VYH, Hawgood S. Adamson TM, Wilkinson MH.
Corn-puterized nutritional data management in neonatal intensive care. Arch Dis Child. 1983;58:732-736
2. Adams C. Computer-generated medication administration records. Nurs
Manage. 1989;20:22-23
3. Olsson CL. A data base system for pediatric intensive care. mt IClin Monit Comput. 1984;1:73-80
4. Hamilton E, Brohan J. A versatile, comprehensive, and rapid analysis method for a perinatal database. IPerinat Med. 1987;15(suppl 1):138
5. Nakahara H, Koyanagi T, Teraoka H, Shimokawa H, Hara K, Nakano H. Microcomputer-based local area network for controlling information
on perinatal medicine. mt JBiomed Comput. 1987;21:83-93
6. Janik DS, Swamer OW, Henriksen KM, Wyman ML. A computerized
single entry system for recording and reporting data on high-risk
new-born infants. IPediatr. 1978;93:519-523
7. Janik DS, Swarner OW, Henriksen KM. Wyman ML. Computerized
newborn intensive care data recording and reporting, II: a practical microcomputer system. I Pediatr. 1979;94:328-330
8. Janik OS, Sharp EM, Forbush L Wyman ML, Jung AL. Computerized
newborn intensive care data recording, reporting, and research, III: an online system. I Pediatr. 1980;97:497-501
9. Maresh M, Beard RW, Combe D, et al. Selection of an obstetric data base for a microcomputer and its use for on-line production of birth notification forms, discharge summaries, and perinatal audit. BrIObstet Gynaecol. 1983;90:227-231
10. Finer NN, Fraser AJ. Neonatal data base and automated discharge
summary using a personal computer and proprietary software. Pediatrics.
1985;76:269-273
I 1. Poland RL, Bollinger RO, Cummings GE. Implementation of a statewide perinatal automated medical network (PAM/NET) for Michigan. Am I Perinatol. 1986;3:144-146
12. Frayer WW. Patient data management in neonatal intensive care. Clin Perinatol. 1980;7:145-154
13. Avila LS, Shabot MM. Keys to successful implementation of an ICU
patient data management system. mt / Clin Monit Comput. 1988;5:15-25
