The National Burn Information Exchange: The Use of a National Burn Registry to Evaluate and Address the Burn Problem

Burns

0039-6109/87 $0.00

+ .20

The National Burn Information Exchange The Use of a National Burn Registry to Evaluate and Address the Burn Problem

Irving Feller, M.D., * and Claudella A. Jones, R.N. t

Before 1958, there were less than 10 hospitals specializing in burn care in the United States. Few physicians recognized or responded to the magnitude and complexity of the burn problem. An in-depth understanding of the factors causing the high mortality from burn injuries was rare, and the cause of death was diagnosed commonly as burns rather than the specific condition that actually caused the death. Little information was available to study burn complications, and no valid system existed to compare mortality rates at different institutions. Burn mortality studies in literature made little, if any, distinction between minor and major burns and failed to consider patient characteristics such as age, pre burn condition, concurrent trauma, depth of burn, or treatment methods. To support the interest in burn care that was growing at the University of Michigan, a database was designed that could correlate the severity factors with the outcomes--complications and survival or death. The database was begun in 1958, and data from the charts of 500 burned patients treated at the University of Michigan Hospital between 1946 and 1958 were coded and analyzed. As with most retrospective studies, this data was incomplete. However, this effort prompted a prospective collection of data, and a local burn registry was established for patients treated at the University of Michigan and St. Joseph Mercy Hospitals in Ann Arbor. This registry continued independently until 1964 when the U.S. Public Health Service provided funds to expand the program into what is now the National Burn Information Exchange (NBIE).5 The major objectives in establishing the NBIE were: 1.

To establish standards for the burned patient's care by citing specific results that can be achieved under optimum conditions.

*Professor of Surgery and Medical Director. Burn Program, University of Michigan Burn Center, Ann Arbor, Michigan tDirector of Professional Activities, University of Michigan Burn Center, Ann Arbor, Michigan

Surgical Clinics of North America-Vol. 67, No.1, February 1987

167

168

IRVING FELLER AND CLAUDELLA

A.

JONE

2. To provide information for the improvement of the burned patient's care, by reporting those specific techniques and principles that have proved valuable in burn centers, units, and programs. 3. To further develop the expertise of the NBIE membership by comparing methods and results in larger numbers of patients. 4. To provide etiology information for the prevention of severe burns.

The number of participants increased rapidly from the original fou members (Dr. John A. Boswick, Jr., of the Sumner L. Koch Burn Unit a Cook County Hospital in Chicago; Dr. Bruce G. MacMillan of the Univer sity of Cincinnati; Dr. B. W. Haynes of the Edward Idress Evans Bur Unit at the Medical College of Virginia; and Dr. Irving Feller of th University of Michigan) to 50 active participants at the present time Information concerning more than 99,000 patients has been processed from a total of 130 different burn-care facilities during the period 1964 to 1985 Currently, active participants represent facilities with 610 beds for burne patients, or 35 per cent of the nation's 1740 hospital beds designated fo burned patients. 1 Information concerning new patients is submitted at rate of 6000 patients annually. PartiCipants in the NBIE are physicians who specialize in burn care In addition, the institution represented by the physician must have specia facilities and a consistent management plan to care for burned patients Participation is voluntary, and although members contribute a fee to offse computer costs, the bulk of financial support for the NBIE comes from th University of Michigan Burn Center and the National Institute for Bur Medicine. The NBIE processes information from three different forms: 1. The Emergent and Acute (E & A) Form shown in Figure 1 was the

first form used by the membership. Before the form was modified extensively in 1979 to reflect changes in burn management and to reduce under-reporting of treatment complications and past medical history, 49,716 cases were reported. On the newer long version of the form, 8918 cases have been reported (Fig. 2). 2. The short version of the E & A form, also part of the 1979 modification, is used only for survivors of burns of 15 per cent or less total body surface area (TBSA) (Fig. 3). The shortened form resulted in substantial time savings for the coders and for the people filling out the forms, with very little loss of information. Information concerning 22,358 patients has been received on this form. 3. The Reconstruction Form is used for subsequent hospital admissions for reconstructive surgery (Fig. 4). As of this writing, information concerning 13,671 patients has been filed on this form.

Up to 310 items of information are extracted from the E & A form and 104 items from the reconstruction form.

DATA PROCESSING AND RETRIEVAL

A series of programs l l has been written to handle the large volume o data that has been collected. The data retrieval system proper is called

Text continued on page 174

169

THE NATIONAL BURN INFORMATION EXCHANGE

NATIONAL BURN INFORMATION EXCHANGE PART

T.lIOSJ>ITAL

OM,,!.

4. SEX S.AGE

2. PATIENT'S NAME

o

Femal.

CD

I

6. HI.

I

7. WI.

.1

8. RACE

.1

3. PATIENT HOSPITAL NO.

OWhiltl

9. BLOOD TYPE

ON",.

"'Tlr. TDl'Al.:AIif

14. STAGE AT ADMISSION

12. DATE OF ADMISSION

o

Acultl (72h's,

17.

METHOD OF BURN PLACE OF OCCURRENCE HOW DID IT HAPPEN,

§"Hom, At Work O,h ••

16. % FULL THICKNESS BURN

l~.c50~:';~!~ TIONS

18. PAST MEDICAL HISTORY I.

A. Orgonis",

3.

..

2·oPo,"_o'.

7.

3·oC.o".,,"~,

L

5.

<

22. TOTAL FLUID INTAKE. FIRST THREE DAYS 1st. Day 2nd. Day

O~I

--- ----- ---

--------- --- ------ --- ----

211. TYPES OF OPERATIONS AND NUMBER OF EACH Wilt! G... .,al Ane.th •• la

I.

I.

2.

2•

3.

3.

..,

•.

3rd. Day

21. SPEC. MED.

20. ANTIBIOTICS

s

..

2.

Ele"t..,lyl. (IV)

'0 comple"<>n of autogroltingj

DR.construction

1S. TOTAL" BURN

Gluco . . (IV)

"ELL'

DEme.genl(O.721"s.)

13. DATE OF DISCHARGE

Colloid (IV)

~~l'J><:nD

THIS ADMISSION

DOthe.

11. DATE OF BURN

.1 •

5.

23. URINE OUTPUT

24, ABN. LAB. RES.

Day I.

1. _ _ _ •• _ _ _

2. 3.

2S. TRACHE. OSTOMY

2. ____ 7.

o

3- - - 8._ _ _

Ov,.

---

..- - - - -

'._ _ _

5.

5.

9~

__

No.

Dal. il V..

10.

27. WOUND CARE

With Local o. No An •• t"'sla

D'essingl: 0 Open 0 Cloled 0 Mllitiple M-dicCltions li.t.d in ord.r of impOI'tonce:

Primory E"clslon

,...

lst. AuI'01l ... fll

,,~

37.

Homograft. Reconstructiv.

'"

Oth.r

\~~!

30. TOTAL ANESTHESIA TIME ALL OPERATIONS Hr ••

31. TOTAL NO. OPEKATI"'" WITH ANEHHE>lA

o Yes o No.

133. AUTOPSY

4. CAUSES OF DEA H

Dyes

0

:

\

29. DATE OF LAST AUTOGRAFT

3 • DEATH

"a. __

~..::!..-

28. DATE OF FIRST AUTOGRAFT OR HOMOGRAFT

I.

No.

\

A-

I~- ~:f~

.I

L

~_~..........!-..........

35. TOTAL EXPENSES THIS ADMISSION 36. NO. ADMISSION FOR BURNS TO DATE

~...... l ~

)~

- ",--1'---.. ...

-

\1 I ,

M.D. S"'g9an in Cha.g.

.

.

o.lglMI {.. hit. copy) ·'.' .... n 100•. I. F.II ••. Un;v. Mich. O~pllcoM (pinkco·py) ·liIe local B",n Unh. Td"licOM (yellow COpy)· fll .. with PolI. n' R..... d.

© N.B.I.E. 1974

~ ~'"

• •

~

PARTIAL THICKNESS FULL THICKNESS

~

pj

~ l-u.e

Figure 1. Emergent and Acute Form (old version). Used from 1960 to 1978 to record information on the initial hospitalization of all patients.

I I

NATIONAL BURN I HOSPITAL fZ

e

I wf= 0 «

NAME

0',

0

Female

e

First,

8MARITALSTATUS

0

Male

0 0

Q..

Single Married

0 0

o

Divorced Separated

o

Widowed

Unknown

I·~G~

~

eOATE/TIME

§

. .~O'~.U~R~N~~~~~____- L____~~~~~~

el

::1

>

i.

o

g0

_METHOD

•

Q • Z ~ -

~

o o

---------jL-+---+--+---i

•

Time

0 am •

,~

REFERRAL

---'". 0 Direct from scene of accident 0 Transfer from (speclfy hospital) ESTIMATION OF%TOTAL. BURN ON ADM.

:;:'!~~:I!~y insufflcllncy- ~ No

not In..thlSla)

a.DlItl

Yes

.TRACHEOSTOMY

•

0

Black

o

White Other (specify)

o

Flame Hot liQu'd Hot >olid Hot gas Electricity Chemicals

o o o o o

o o o o o

o Place of employment o indoor 0 outdoor o Home o Indoor 0 outdoor o Health facility o PUblic building

Radiation Sunburn Frostbite Unknown Other

ut IB .

ONo Dvu

0

No

0

, yes

.~~::~~~v(~~U!~rdlsease _

_.-:-

Pulmonary disease Renal disease

Mo

Inll8rted

I

unknown

Mo

oaY-lvu'lb.Datl remoYld

I

Ooy

0 None

I

Vu,

•

Mo 1 Day -I Year I b. Data Mo 1 Ooy 1Vu, !n.. rtlld removed

a. Dati

_ _ .. _

4.

~_ _ _ _

•

.~.~tt.~~11~EJtAJI~s

Tralfi~ a~cident

Other accident (e.g., airPlane) Recreational facility School Other (speclfy) ____~

TYPE OF FLUID

w

:;:

mld~~Sh~ loOl2A~I:n'itht 2nd DAY

~

IOCONCURRENT INJURY No

_

Yo,

~

Drug addiction

pr;;~~~:.:.motional

a. TYpa Cl open closed dressl"'i1i multiple drlsslnllS comblnetlon

b.

Ft~uanoy

0

qd

0

bid

Crynallold (isotonic)

f-

Crystalloid (Hypertonic)

Autograft haNest

Colloid

Autograft epplicatlon

~ERATIVE

Other (speclfy) _ _ _

~_

ESChar deerldemant (exefslon)

'rI t'1

None

t" t"

d. Enzymatic ctartca.",lnt No Yes (1iCHIdfy· ...nt)

o o

t'1 ::>:l

:>Z

tI

0 other (specifV)

EXPERIENCE

c. AnimaL skin ONo DYes

e. Other (speCify) d. Artiflcal skin 0 No 0 Yes ONo OVes

DATES OF PROCEDURES without gln.ralenllthliia or with I .1 neltt! a

with .naral an.sthliia

Q

~

C

tI t'1

t"

~

Oth.r (specify) Reconstructions (thiS admlnlon) Othlr - amputation,

Z

C'l

c.Hydroth....p)' 0 No 0 V_

0 tid

PERMANENT CLOSURE

~ w

~

~:~~~o.urinary

None 0 Other (speCify)

DO~e~~I~:RE: Tr3P!~~~m~~~~~afenlde Acetate) o o Sliver Nitrate 0 Sliver Sulfadiazine o Gentamycin 0 Neosporin o WOUND CAREl SYSTIlM

Fracture

i~::~::,

. Chest

o

0 Other (specify)

r---TEMPORARY CLOSURE a. Homograft" b. Amniotic membrane ONo DYes ONo OVes •

TOTAL INTAKe:

I Is" ---l.---±=+---

~~~~:;1510ns

~ Alcoholism

o o o

agents

BURN DAV 12 DAY OF BURN

__

_

~~:~~:s

o

3 ..

a:

RACE

IbS.

SIGNIFICANT MEDICAL. HISTORY

-

:.Y:::p:~~a:t7cTIB~0:~C U~E~es ~ ;~~~~~I:~l~r 1. _ _ _ _ b. Therapeutic

fZ

V

ft. _ _ in.

_

Specify exact locallon where accident happened (e.g., guage. kitchen)

. . ESTIMATION OF% FULL·THICKNESS BURN ON ADM.

•

Months

PATIENT'S HOME ZIP CODE

fli)

PREBURN WEIGHT

BRIEFLY DESCRIBE EVENTS OF ACCIDENT:

.DATE/TIME

«

o o

Had victim any self-destructive tendencies? _

1111to

i.Agent

o o

Had victim been drinking? ~ _ _ _

requiring medical attention?

~

W

lass than 2 y ..rs _

e

PATIENT'S HOSPITAL NUMBER

-.,HEIGHT~

If 2 VNrs or old'"

b.~'. ~:: Had ::~!:::: victim :~:n;;evjOUUeCident been ' ' ' ",':in9'' .::.--====.::.J::=t==I==I==i

C)

Z

u.

o

.PREDISPOSING FACTORS

•

o

II. Cltt;' and State

C,".

SEX

.... --t

EXCHANGE (Form 79 E/A)

INFORMATION

H ... ,..'N.m.

~c.

(speclfV) _ _ _ _ __

?> o

'-

Z

t'1

TOTAL PRoeEDURU

~

OPERATIVEIEXII'IERII:NCI. ,. Olt. Qf .utotrlft _

'" TQtl1 "umlMt 01 .11 QP.rltiQns wh'Tn...'! anestl'lI151. d. Tot.i g.n.,.1 .nllSttlUI, tim. {III prO(:lId.ur.1 --

",st

b. 0It. of lut .... totr.n

... TOTAL aLOOO GIVEN THIS ADMIMIOH .~.?~!L.!eATIONS (check all that apply)

o

en

o

Z

o

~u

o

...:::;::;;

8

D

o

{specify organism). _ _ ,_ _ _ organism unknown Bacteremia (specify organism) ~ _ _ _ _ _ organism unknown Burn wound 5epsil (specily predomInant organism) _ _ _ _ _ _ __

o D D

D

o

o

0

o o o o o o o o

o o o o o o o o o

Acute liver faIlure PancreatitIs Peritonitis Hepatitis Other

00

o o o o o o o

COJltractures Cellulitis & Iymphanyltls Decubitus ulcers Graft sloughing Gangrene AllergIc Skin problems. rnh IJlfected donor site HypertrophIc scarrlng/kelQld Otl1.,,________________

Osteomyelltis/perlostlils NecrosIs Qf muscle & tendQn AmputatlQn (speclly sIte) _ _ _ _ __ Otl1er _ _ _ _ _ __ ENDOCRINE .. METABOLIC ACldosl5, metabolic & re.plratory Dlatletesmellitus, acidosis Nutritional dellclency ElectrOlyte deficiency Adrenallnsufllclency Tl1yrQld Other ___ .~ _ _ __

Cerebral edema eVA

00

o o o o o

o o o o

I.

~

IJ:I

c::

:xl

Z

00

o o o o o

~o t"'

TraumallcneufQ,ls Psycl1Qsls ,.,anagement prQblem Otl1er _ _ _ _ __

Drugreactlon Transfusion reliction Orug overdO$ll Otller

ESTIMATE%TOTALBURN ESTIMATE % FULL-THICKNESS BURN

~ PARTIAL THICKNESS

......

z

d :xl ~

DIAGRAM BURN II.

Perlpl1eral neuropatl1y Other

BLOOD" BL.OOD FORMING L.eukQpenla Coagulation defect Other _ _ _ _ _ __

00

NERVOUS SYSTEM

•

0

o o o o

~

o z

. ' FUL.L·THICKNESS

Eyelid contracture Ear, deafness ConjunctivitIs Other

t-
~

::t:

> Z

DISPOSITION

o o

TQ home To other flCUItY ~=~fY) _ _ _ __

C'l

-.- eAUSE OF OEATH

eAUTO.-sV No Vti

tol

•. PrI"'I..,.I _ _ _ _ _ _ _ _ __

Left AMA

Feed,self Dre"""seif Woundl1ealed

o o

sufficient

o

Sel1aviar problem

Wound unl1ealed (sp~..:ify%) _ _ _ %

Abnormalities (specify)

j--'-TOTAL. COST OF THIS HOSPITAL.IZATION:

8.~CX:'~~!I~~h:~!=I~ID OF PAYMENT

o

o o o o o o o

{Curhng's ulcer, etc.)

o o o o D o o o o

O.~S~::Ll~~:r:VALUATION fCI1~k ;!l~:~:t:~:~:~f o o o

00

GI ulceration

00

CAROIOVASCULAR CarCllacarrest Congestive heart failure Hypertension Thrombophlebitis Endocarditis, myocarditis Myocardial infarctlon HypotenSloJl Other _ _ _ _ _ _ __

TT"'

o o o D

o

o o o o

o

o

00

organism unknown Other infection {5Pec\lyslte _ _ _ _ __ andorganlsm) _ _ _ _ __

PneumQnia Respiratory burns Pulmonary eClem~ Respiratory Insufficiency pulmonaryembol1sm PneumothoraK Pleuraleffu,ion Other

NOYe1i

00

RENAL/GU Aculerenallaliure Urinary Iract Infectloll. (any part) Hematuria Oliguria Other _ _ _ _ _ _ _

o o o o

o

W·~~~~~:GE

~

0 D

00 D D D D

NoYes

NoYes

00

~

tol

No pilyment expected

orlg~~~ .. !~~~eF~7~;) ~Return to: 200 Nortl1 Ingalls Street Ann Arbor, Michlgiln 48104 Dupllcale(yellow) -File In your Burn U"lt TriplIcate (Plnk)---Fi\e wltl1 patient record

iI.

HO$pital

8 ~:I;:~n/BlueShleld

g ;;~~~~~n~sh~~~~:~~~~~IUIOJl

0

0

Medicaid

I N.II.E.TM ,~""oH"'. ~,. :i \

©

~

- PertoA completing

\.

D Self P"v

Private Insur"r

f""Orm----

1

NIBM 1983 \..... ,_ ...,l'TM Ann Arbor. MI '-"

I

__

Surgeon in

~l1al'9.

Figure 2. Emergent and Acute Form (new verSion). Modified in 1979, this form allows for the collection of more complete treatment information and uses a yes/no checkoff system to reduce underreporting.

~

--l ~

NATIONAL BURN

EXCHANGE (FORM 79 S-E/A)

INFORMATION

e

PATIENT'S NAME

I-'

•• Cltr'ana __ ._

. - •• -Ho_HaI MImi Last,

e

First.

PATIENT'S HOME ZIP CODE

o o •

Z

....

W

•

~~~GN~~:C~.::T MEDICAL HISTORY Cardiovascular disease Stroke (CVA) Pulmonary disease Renal disease Diabetes Anamla

H

o o o

Cancar Convulsions Alcoholism Drug addiction Previous amotional problem

Unknown Nona Other (specify)

DATil. 0" 'ROCIiOURIi.

TY"' 0" OPERATION'

Ift"ll ,.~t.h.1lI IW~~:':h'r:::rl:'":l='la

wit..

•

E*".,_~t(.KClIIoft)'

INJURY Fractura 0 None Pulmonary 0 Other Brain (specify) Abdominal Chest ---Heart Genlto-urlnary

Aut""~ hamtll

o

o o

Chemicals Radiation Sunburn

o o o

Flame

Hot gas

Eta<;trlelty

Frostbite

o PIKe of employment o indoor 0 outdoor o Home o Indoor 0 outcloor

o o o o

o other (specify) o ____ a Unknown

1. s.ptlcemla

D

2. Pneumonia Contractu,..

,.. Orga,nllm _ _ _ _

a,

White

b. Location

Health facility

Traffic.eclaent Otherllceltle"t {a.g., alrplallel Public building Recreational facility School Other {specify}

c. Spacl'" a.let locatlori WhlN I"idant happened (1.1. garage, kltchenl'

b. ESTIMATE'" FULL-THICKNESS BURN

[J

0

HotllQulcl Hot solid

0

Other (5Pllclfy)

I.~IAE~~~MMA~~~NTOTAL BURN

COMPLICATIONS [] None ~~

o o

o o o o

o o

iiC~N';.~'R~!'T

.... ..

METHOD Agant

I.

. . . .

d. Hid 'ttctlm any prtl",lou. Icclaent rlquiring mlcllcalattantlon? _ _ __ e. Had vlcllm any self-destructlv. Iinaenelas?,

y" SO,

O

•

"::"~.'I~"kI~""~=i=!1

b. H"lb"lotlm Had victim tIN" smokln,'

c. Wit victim _Iornl

8=

,•=

=!I=!I==

No I Vas t Sus IUnk

PREDISPOSING FACTORS •.

BlaCk

rTlJ7l rLl1!J PARTIAL THICKNESS.

-_.

--'

FULL-THICKNESS

(SpeCifY)

OlINr (lltH1"')1

4. _ _ _ _ _- -

~P£O,::T~if~r~~~~o~~~~c_.~_

c.

b. Date of list lutograft

Total num ber of all operltlons with general anesthesia

d. Total generll anesthesia time (,.11 procedures) ~_hfl. _ m i n .

1 _ _ ._ •

_m. _

I

..,

a

O~~:,;YAWAT~~=;"m:=t=!nt F_

...,..,lcMrwoD"'" aQ ADnonn""," (_Ify) =,.-_ _-::-_ __

Mlf

DtMaI_i

0

. Wound ft. .~

•

o o o

Original {whltl copy)-Return to:,

~~o 1;:!~:h~~!:~11 Streit Ann ArboJ. Michlg.an 41104 Dupllcat8 (yellow)-Flte 11'1 your Burn Unit TrlpUCIte {plnkj-Flle with patient rec.ord

Wound u.,....... (lpaclfy 'Mo) _ _ _ "

EXPECTED METHOD OF PAVMENT (ChIck ,.11 thlt apply) No p~yment expected Blue CrosS/Blue Shield Medica,.

IN.ILEioTM ©

~Ift,

(~_ ' - . ,:\. t

0 0 0 •

MedlClld Crippled Children's Commission Workman', ComplnSiition

o o o

Private Insurer Salfpay Other {specifY I

SIGNATURES Perlon completing form

~w1

TM This form to be used for: SURVIVORS only, Burns less than or equal to 15% NIBM 1983 \, .... ., Ann Arbor. MI " ' _ '

~Z G"l

6. •

t:J

Surgeon In ch"rge

Figure 3. Emergent and Acute Short Form (new version). Used for survivors of bums of 15 per cent or less total

"'l t"l

~

"~ o

~

&:c::

ot"l

~

~

0' z t"l

'"

NATIONAL BURN 0"

8

~

I

•

~

; ,=

UJ

I

~ Ii ~ ~;

Male

D"':~

'I MOl

e

~ I·AQ' 0

o Widowed U MarrIed

onIY"'I_

IMOl

U Separated DATa OF'

0

DavlVO'

10

•

o

•

SIGNIFICANT MEDICAL HISTORY

(lnctude

:~:;;:~=~~~~~HICKNE:S BURN FROM

,II rllidull, functlonallnd cOlmetlc deformity)

~o V...... Straka leVA)

and/or k&IOid

Cosmetic scarring

~r~~~i:~~ln~:~~;;:~;e5 ~i:~rt;7~s~055

ADMISSION -"_..

--

Black

0

White

0

None

~o V. . . . . Alcoholism Drug ilddlctlon

Renal dlseue

PreviOUS emotIonal problem

~~a:~~:s ~~:~Slons

1_

Pr.....,.

•.. --------------- .... -------------------------------------------------------------e

-

:~I::~~c~~vhe·~!~~enc115 c~~sn~:~\I:r or perCilptuiti

0 Other (specify)

INDICATE BODY AREAS RECONSTRUCTED . . ADMISSION

nus

1. _ _ _ _ _ _ _ _ _ _ _ __

~

TOTAL NUMBER OF PROCIlDURES

•

~~~SA~OQ.;I~::'o':.\:~~~~~S~~~~~:) _ _ ~. . _____ min.

•

.:~~~:~QE

COMPLICATONS THIS ADMiSSION None

o

M·I~T··

• ~S;:::;'ON

C v.

DISCH""• • *VALUATION (CMck III

v. •

•

HOIPital

o o

b. Profu,lonal

o o

Return to:

:0 1::!~h~~!:~I'

'N.B.I.E.TM

I.

, . . ... ,

f~ -fa, ,\ f\.1ll 1

Street Ann Arbor,'Mlchlgan ..1104 Duplicate (yellow) - FHe In your Burn Unit Triplicate (pink) - 'File with Pltlent record

...

@

=.,'

TM

NI8M "13 Ann Arbor, Mlchl.an

0

I c:

5:1

......

z

6 ~

~o z

tr:I

~

:Ii

~

G"l

l:'l

No

ttt.t 1NIr)

No

gIoCI'IIC~"lty gf:'::'~,~'::~ ~ult\ed

0 IIRk to _ . , D . . towotk

E.XPECTED METHOD OF PAYMENT (Chick all ttalt apply) No payment expected 0 Medlclre 0 Crippled Children's Commission Blue CroU/Blue ShIeld 0 Medlclld 0 Workmln', Compenliltlon Orlglnl' (white copy) -

V.

No

8o 90 =~ ar.... ...ti:,~

~~~I~!~~;A~~OTNHI_S__________ a.

0 uft AMA

8 ::~~~~'--------

Au.tGPIY

Z

t:P

Pulmonary dIsease

Cardiovascular disease

RECONSTRUCTIVE PROCEOUR.S THIS ADMISSION PraMdI,,"

0

Other (specify)

V.. S... HypertrOPhic 5carri"g

PERCENT TOTAL BURN FROM FIRST

l:'l PATIENT'S HOME ZIP CODE

GRACE

:1

If lui tnan 2 v ••rs _ _ Monthl "If two YUlt or oICI.'_ _ V ....

0

Unknown

THI. ADMISSION

•

•

~

... c:ltr'lntiMllhi

First,

L.ast,

OP .UR

III • II:

EXCHANGE (FORM 79/R)

HOlPlhllNlime

PATIENT'S NAME

0

INFORMATION

"-rtlll

,.,u

o o

Private Insurer SlIf Pay

o

attaer(lPeclfy)

SIGNATURES Ptir50n completing form

I -

Surgeon In charge

This form is for: RECONSTRUCTIONS

only

Figure 4. Reconstruction Form (new verSion). Used for follow-up hospital visits for corrective surgery, observation, or therapy.

,.... ~

174

IRVING FELLER AND CLAUDELLA

A.

JONES

INQUIRE. It provides basic statistical capabilities, extensive support for subset selection and combination, and a macrofacility for storing common command sequences. Information about patients can be selected on any of the items (e.g., all men, all burns between 50 per cent and 70 per cent TBSA, or all work-related burns or those for a specific occupation) and on combinations of any number of items. INQUIRE also allows placing subsets of data values into external files; these files then may be used as input to more sophisticated statistical packages or other special-purpose programs. Although INQUIRE was designed primarily to access the acute-care and etiology database, all database-specific information (such as pointers to dictionary files for categorical items) is stored in files external to INQUIRE itself. This makes conversion to a new database relatively easy, as demonstrated by the addition of databases on reconstructive operations and rehabilitation of patients. An update system provides for the addition of new patient records to the master tapes, where all patient records are stored. The update program checks consistency in each patient's record and calls the display program to print the data for verification in a format similar to the original coding form. Correction and deletion programs allow simple editing of master tapes, with redisplay of all changed records to ensure clean data. Once the data have been verified in this manner, a fileload program copies the records from tape into disk files, where they can be accessed interactively. The file load program can load portions of the database selectively to reduce file storage space. Furthermore, each data item is stored on disk in the minimum number of bits necessary to ho,ld the maximum possible value of the item. Both capabilities are important to reduce the cost of maintaining a large database, because the Michigan Terminal System (MTS), the operating system in use at the University of Michigan Computing Center, charges partly on the basis of allocated disk space. Empirical results show a savings of about 50 per cent in disk space usage over a similar scheme that only aligns each data item by byte. The cost of unpacking bit strings to access the data is offset by fewer input/output calls to the operating system, because more data values are passed on each call. All computer work is done on the University's Amdahl 5860. The MTS operating system prOVIdes excellent facilities for iriteractive terminal access (on a typical weekday the system handles more than 300 simultaneous terminal users). Both INQUIRE and the update system are written in IBM Assembler for efficiency and speed of access. Smaller, less frequently used programs are written in higher-level language, such as FORTRAN. During 1986 to 1987, because of the interest of participants, it is planned to allow participants to report directly to the NBIE using personal computers.

QUALITY ASSURANCE To ensure that data are "clean" and useful, several steps have been taken in addition to the INQUIRE display/correction/update system. From the beginning of the NBIE, faculty of the Department of Biostatistics at

THE NATIONAL BURN INFORMATION EXCHANGE

175

the University of Michigan School of Public Health (UMSPH) have contributed to the design of the forms and the analysis of the data. Dr. Richard Remington, dean of the UMSPH, assisted with the design of the first NBIE forms in 1960. The goal was for the forms to be short (one page), unambiguous, and easy to complete. This form worked well for 15 years. There were, inevitably, sources of bias both in the form and in the procedure by which cases were reported to the NBIE. The data analysts realized the importance of identifying these sources of bias, evaluating their impact on analyses, and implementing procedures to correct the problems. To this end, NBIE analysts and the biostatisticians have engaged in several data quality studies. Among the quality studies was a Monte Carlo-type analysis of 23 schemes under which a hospital can fail to report all cases (including both systematic and random omissions), and a study of potential bias in reported bum size. Since 1980, in an effort to gain complete reporting for all members, participants are asked to maintain a bum admission log and to submit the log monthly. The NBIE then can prompt the participant to submit forms for any missed patients. After a pilot study in 1978, and a few resulting corrections, the new E & A Form (Fig. 3) was distributed in 1979. This form, designed with input from Dr. Richard G. Cornell, chairman, and other professors from the UMSPH Department of Biostatistics, includes fewer open-ended questions. A yes/no check-off system was instituted to gather data on complications and reflects changes that have occurred in bum management. The new form helps to standardize responses from different hospitals and reduces under-reporting of complications, past medical history, and concurrent injuries. The form is still one page. The NBIE coding scheme, developed in 1960, allows for the addition of new codes for medical diagnoses and accident etiologies (including locations, activities, objects, and agents involved in the bum). This flexibility kept the original form and set of codes from becoming outdated as the field of bum medicine rapidly evolved. Without sacrificing a strict adherence to confidentiality, the original completed forms are kept readily accessible to analysts. Not all information can be quantified (especially anecdotal etiology statements). Because all original forms submitted by members are kept on file after coding, they can be referenced and examined manually for special studies of bums of rare etiology.

BENEFITS Annual Report The registry was formed for the benefit of the participants, and this purpose continues to be a top priority. Participants receive an annual report, summarizing the activity at their facility in the past year and all previous reporting years and, in parallel, summarizing the combined NBIE activity. Physicians can use the report to examine survival rates by age, and by size and depth of bum. The report also summarizes days of hospitalization, treatment complications, causes of death, and etiologies of

176

IRVING FELLER AND CLAUDELLA A. JONES

the accidents. Appropriate use can increase the physicians' understanding of outcomes of bums and allow comparison of their results with a national "average." The report also can be useful in determining the prognosis of newly admitted patients, and more importantly as a milestone of progress; the better institutions set the curve and beat the averages on the majority of their patients. Annual Meeting Participants also can benefit from the NBIE annual meeting, held during the annual American Bum Association meetings. This meeting provides a forum for discussing topics of interest to members regarding feedback from the NBIE office, the data collection forms, or any other bum-care questions (such as an accurate way to determine size and depth ofa bum). Special Studies Participants may write to the NBIE office and request specific analyses, either of their cases or of the combined NBIE data. These requests may be to examine certain accident etiologies (e.g., clothing bums or chemical bums of the eye), incidences of various complications, or to compare methods of treatment. Severity Factors The NBIE has contributed to an improvement in our ability to estimate the severity of a bum. The size and depth of bum as well as the age of the patient are the most significant factors to date (Figs. 5 and 6). Logic and experience indicate that a positive past medical history (PMH) also will increase the severity of a bum injury, but previously data were not available to confirm these suspicions. Recently, data on 21,440 emergent and acute patients included in the NBIE database, from 1979 to 1983 and from 52 reporting hospitals, were reviewed. 9 Of these patients, 5600 had a total of 9538 specifically identified medical histories broken into 14 different categories. Cardiovascular problems were listed most frequently (20 per cent); these were followed by pulmonary and respiratory complications and central nervous system problems (11 per cent each), emotional disorders (10 per cent), alcoholism (10 per cent), and metabolic problems (8 per cent). Findings also showed that mortality rates are significantly impacted by a positive medical history. For instance, in those patients with a TBSA of 40 to 49 per cent, the survival rate was 88 per cent for the 15,840 patients who had no PMH reported (Fig. 7). This survival rate is better than for all ages, which is 40 per cent TBSA. For the 3415 patients who had 1 PMH reported, the survival rate fell to 76 per cent. For those with two medical histories (1234 patients), survival fell to 52 per cent, and for the 951 patients with 3 or more PMH, reported survival rate was 40 per cent. These differences in mortality were not because of age; in each age group, mortality rates are generally higher because patients have more PMH. These data on the effect of PMH on survival are highly significant statistically (chi square of 615 with 1 degree of freedom). Each past medical problem that exists in a burned patient is comparable in terms of severity to a bum injury of 11 per cent TBSA.

177

THE NATIONAL BURN INFORMATION EXCHANGE

NATIONAL BURN INFORMATION EXCHANGE Ann Arbor, lIichlaan - I. Feller II.D., Director

BURNED PATIENT SURVIVAL BY AGE 1978-1985

,--.

......

"\

'\. \.

r

,

t", ~~.. '

......

.~.~...

"'.. ' ......

"-""'- r,

~

" ,

"'I~-...

~~"

I

,

'~

"-

"0.

30.

10.

. . . r-....

"-

"

-'II"... ",

\. \.

at·

... r......,.....

'~

"

~,

o

SURVIVAL CURVES BY PROBlT ANALYSIS

I .......

...

,

-_.U~Y.S

?514 18336 5804 2101

35-4'

-'-AGIS 50-51 ---AGES 60-74 _ A C t : ! 75 AND UP

......

2&88

1318

............

...... ............. ""'~ t-.......

"-

.........

CASES

'",-,

',-

......

"' """ -. "--- r-____

,

10.

10.

50.

•

i_~""

~ 40.

CURVE '---AGES 0-4

.......... AGES 5-34

" TOTAL AREA BURNED

lOG.

APRIL 1986

CASES 1964 - 1977

n=38.446

Figure 5. Probit survival curves showing survival rates of different age groups according to the per cent of total area burned.

NATIONAL BURN INFORMATION EXCHANGE Ann Arbor, lIichiaan - I. Feller II.D., Director

BURNED PATIENT SURVIVAL BY AGE 1978-1985

...

~ ~

. '\.""',

,

~~ ...... ~-:::''''''''

.. r:::-:..... .........,

\.,

" ...

\

~

\,

"-

"

1977

-

"

I',

' ... 40.

18184

5'151 Z'III8

_5

AGllS 75 AlIJ) UP

1309

~..... '

....~~

,.

~

~,

""

~

30.

CASES 1964 -

"~~

"

Q-4

ACES 5-34-

----AGES 35-49 -AGES 50--11 '---AGES 80-74-

" "-

~

CASES

I··'·AGES .......11'1

'" \.'\ '" ... ...

-

CURVE

..., '.;:..~.....

, '\. '..., r'-.

" \

SURVIVAL CURVES BY PROBlT ANALYSIS

~'''II

..... . ~. 30.

~.....

....

.~~

"

10.

.:~::...~

...... ....:......"1:1. 'III.

" FULL THICKNESS AREA BURNED

'::-:==~"'tII

10.

10.

1OG.

APRIL 1986

n = 38.081

Figure 6. Probit survival curves showing survival rates of different age groups according to the per cent of full-thickness burn.

178

IRVING FELLER AND CLAUDELLA

A.

JONES

NATIONAL BURN INFORMATION EXCHANGE Ann Arbor. Michigan - I. Feller 11.0 .• Director

BURNED PATIENT SURVIVAL BY FREQUENCY OF PAST MEDICAL PROBLEMS

'-

f=.: •• ' ::.::'.:.~:-:: ..........

~

l-___

SURVIVAL CURVES BY PROBIT ANALYSIS

~

.~,

~~~

...

<0 .......

..................

" " ... "'0-, .... ... " - ", .... ... ... . ~

....

r-...

...

- 0 PIIH ······1 PIIH - - - 2 PWH - ' - 3 OR 1I0RE ?IIH - A L L PATIENTS

~

f', " .. ~ I'--. "

"-

10.

20.

30.

40.

.........

...

'. '. ...

50.

19197

4139 1499 1189 211004

".

....~ " '.

"

'.....

,

r--.. "

.

.......... <>0.

, CASES

CURVE

,

"

80,

~""

......~

"

... "', ~ ..... ~

·i... 70.

-...

80.

-~-:

'......... ......

~;~ 90.

100.

7. TOTAL AREA BURNED

BASED ON PATIENTS FROM 1979 -

1984

MARCH 1985

Figure 7. Probit survival curves showing better survival rates for groups with fewer past medical histories.

SURVIVAL AND LENGTH OF STAY Survival rates by age and size and depth of burns have been published, as have lengths of hospitalization for survivors. 7, 8, 10, 13 These serve as baseline standards of care. Causes of death, as a function of size of burn, age of patient, and time after injury, have been published recently so that physicians can anticipate pmblems: 6 The data have been used to document improvements in care, both in increased survival rates and decreased hospitalization times, Figure 8 shows how survival rates (as modeled by probit analysis) have improved over three time periods. These improvements exist after correcting for many other severity factors simultaneously (size of full- and partial-thickness burn, part of body burned, age, sex, and time from burn to admission). Figure 9 shows the decline in hospitalization times for survivors of severe burns. RECONSTRUCTION Another NBIE project has been to document the reconstruction phase of recovery, detailing the length of time, number of readmissions, and types of operations required before treatment is completed. Figure 10 shows the length of time necessary for completion of medical care as a function of size of full-thickness burn and age of the victim. Note that a

179

THE NATIONAL BURN INFORMATION EXCHANGE

young child may require as long as 10 to 20 years to complete treatment of a serious burn. This documentation is possible only because of the length of time the registry has existed and the comprehensiveness of the database.

INSTITUTIONAL DIFFERENCES Other contributions of the NBIE data are a result of special studies using the registry. In 1975, analyses showed statistically significant differences in survival in different hospitals. 7 The differences among three different groups of hospitals are shown in Figure n. It was postulated that these differences were not related entirely to treatment alternatives but to variables of professional performance, nursing skills, organization, physical design, and resource intensity, which in aggregate significantly modify the outcomes of care. This finding led to a government grant to investigate the problem in more depth. The study, conducted jointly by the National Institute for Burn Medicine and the Department of Biostatistics at the University of Michigan, involved extensive analyses of data from 12 major burn centers as well as site visits to these centers by a team of burn specialists. The results of this study are too numerous to list here,3, 4 but a few NBIE-related findings can be highlighted: 1. A major result was the development of a valid method of standardization that allowed for reliable comparisons of hospitals' survival rates, called

NATIONAL BURN INFORMATION EXCHANGE Ann Arbor, lilchilan - I. Feller Ii.D., Director

BURNED PATIENT SURVIVAL BY YEAR OF BURN ......, SURVIVAL CURVES BY PROBIT' ANALYSIS

.. ..:::;,..... ~

~ ~,:,.

~

~

CASES

YEARS

'';'''''''.

if.... ' .

-Ise~

,'. .. . t

'.~ ~ ~'

......

-19'12

IDle

2''78,

" ...... '':)71 191a __ •• 1979 - 1ge4

Z677d

TOTAL CASIS

847'18

>;;;"

" "-

#to,. ".

"....,. "

....

~

.

'~..

... . ....

.... r--........

"

~

............ ' ...........

f ......

'I'.,

•

~

..

~

..........::: ::::::"" 10.

ao.

30.

40.

eo.

110.

'10.

110.

80.

100.

" TOTAL AREA BURN NOVEMBER 1985

Figure 8. Probit survival curves showing improved survival rates during the period 1964' 1984.

180

IRVING FELLER AND CLAUDELLA

A.

JONES

National Burn Information Exchange Ann Arbor, Michigan

I. FeUer, M.D .• Director

Length Of Hospitalization vs. Size of Burn By Year of Burn - Survivors Only g 0

-;;

~

0

'"

';l

g

0

."

.!:! fti

... '0.

.. 0

:I: fti

~,

0

c:

0 ....

,., 0

';l

'2

...

0

II>

...

CD

0

~

c:

cu

..J 0

C')

0 <"

'

..

..." "

.. '

--:i'~" ~.

;,;..---

" ""

.'

.'.'.'

.,'

..'

"

"

.V ,,4

..'

"

.'.'

,1.;" ~ ,,~

."

.'

.,...,~ .' .'

,

.' "~ / .' ." ..... " V "

0

~

.'.'

.

.'.'

..," .'

>

."

.......,'" , ,.,-

,'M ....

V

;'~

'/

V

./

,./

,.'~ ~

~ ........... 1964·197~

n=15,818

•• - • -1973·1978 n=23,602

~

1979·1984 n=31,410 10

20

30

40

Percenta~

50

60

70

so

90

100

of Total Body Area Burned

Figure 9. Average length of hospit!)lization for survivors of severe burns has declined in the period 1964-1984.

181

THE NATIONAL BURN INFORMATION EXCHANGE

ESTIMATED TIME FROM BURN TO LAST RECONSTRUCTION VS. SIZE OF FULL· THICKNESS BURN - BY AGE WHEN BURNED MICHIGAN BURN CENTER, 1959 - 1980 II

10

9

c

.2

.

~

8

0-4yr.,n=127

:>

'Iii c 7 0

" a:•

-

6

J;!

5

en 0

..J

E :>

III

-..

18-49 yr., n = 294

4

E 0

en

3

0

~ 2

051015202530

TOTAL

n=~64

40

50

60

70

Sixe of Full-Thicknes!!, Burn ('Yo of Body)

Figure 10. Average number of years from burn to completion of reconstrllction, by age when burned and size of full-thickness injury.

the standardized mortality rate (SMR).12 This procedure used multivariate logistic regression to correct for hospital diflferences in patient population related to size and depth of burn, age, part of the body burned, year of burn, sex, and time from burn to admission to the burn center. 2. The survival differences still were evident after application of the standardization procedure, and the magnitude of the differences could be documented. 3. Speed of wound closure (time from the burn to the last grafting operation) was shown to be highly correlated with survival differences. llospitals that routinely closed wounds quicker also had better survival rates. This finding is significant because many facilities still do not practice rapid wound closure. 4. It was also found that a number of organizational factors were consistent with better survival. For example, in those hospitals where care was more highly organized, where the medical director spent more actual hours directing care, and where the average hours of care provided by the burn team neared or exceeded 24 hours per patient per day, survival was also greater.

182

IRVING FELLER AND CLAUDELLA

A.

JONES

SURVIVAL vs TOTAL BODY AREA BURNED AT DIFFERENT BURN CARE FACILITIES, 1968-75 100~~~~-r--~----r---.----r---'------------,

~,t:-:',~".'.-'~~ r"'",:-- ~........ ...... ..... ....

90

I---

'C., . , \ ...."II

! I'·'·

!

HOSPITALS A,I,f

.·11.

I ____ ""'''AU C,t:,G,IC,L , , - _ -+---1 _HOa'ITAU H.1.J •• ..a

r\.'>.:-..-

.

.... eol---

\

.......

I---'-I---~~~~~~-~I \ \ I..~·---I---+---~--~--~ \\ 1

~

70

.~

60.j----i----·-·

I---

\

r---+----lH---~----"t:_-+_--_+--_j-__I

JJ

\. i\ \\ ~ 50+--~---+-~~-+1~~~~1\~-+-~r_--+----r-_; Q ) \\

~ ~

40 I - - -

\

\

\ -+---+----+---\\:-li\'---"I\--l.....3,;:..-\-...+---+---;

..

~+--4---~---+--~----~--·r--T+---~---r---l

\

\f\.

. . . . ..

~+----+---+--~~--+----+---~---j~~+---"~.r-~

\

~

..........

10+---+----+-~~-+---+--_+~~r---+'~--.~~

.... ~ r--.. "i". ._ ~

10

20

30

40

50

60

70

80

90

100

Percent TBA Burned Figure 11. Differences in survival rates at three different groups of National Burn Information Exchange hospitals,

ORGANIZATIONAL FACTORS The fact that significant institutional differences existed after correcting for severity was known; however, the differences existing among levels of specialized care, center, unit, program, and general care were still unknown. A federally funded feasibility or pilot study was conducted to look at this question. From a 9-state area, 262 hospitals were surveyed, with an in-depth study of 10 hospitals; 3 with burn centers, 3 with burn units, 1 with a burn program, and 3 general care facilities with no special facilities for treating burns. Participants in this multidisciplinary study included the Department of Biostatistics, Institute for Social Research, Center for Research of Utilization of Scientific Knowledge, School of Medicine and Department of Continuing Medical Education, School of Architecture, and the Department of Medical Care Organization all at the University of Michigan, as well as burn center personnel. Unfortunately, because of budget cuts, the larger study was not funded. However, the pilot study produced the following findings 2 : 1. Regionalization has progressed at a rapid rate--97 per cent of the 262

hospitals surveyed had a referral policy for major bums. This referral

I

THE NATIONAL BURN INFORMATION EXCHANGE

2.

3.

4.

5.

183

policy was effective; only 31 admissions for burn care were recorded (of the 928 admissions in the study) for general-care hospitals, and none of these patients died. There also was a preponderance of patients with small burns who were admitted to all facilities studied (64 per cent of admissions had a burn of 12.5 per cent or less TBSA). The costliness of care was found to be related to the level of technical sophistication offered by the institution (namely, burn centers are costlier than burn units, burn units are costlier than programs, and programs are costlier than general care). Injury severity is strongly related to cost of care (i. e., for burns greater than 12.5 per cent TBSA and 3 per cent full thickness, every 1 per cent increase in full thickness increases cost of care by $1000). Quality of care, as judged by performance of the physician, tends to be positively related to cost even after severity of injury is corrected for. This study indicated that perhaps over-regionalization has occurred because there are more specialized burn-care facilities than warranted by the number of moderate and major burns. However, it was again found that differences existed among specialized facilities. These differences were evidenced by judgments of quality of care by specially trained nurse observers. Debridement was found to be highly correlated with not only cleaner wounds, but with overall care. Significant differences also were found among institutions with regard to nursing care and condition of the patients after this care was provided. Care was judged better within burn centers. Nurses in burn centers also scored much higher on nursing knowledge tests (reliability, 0.90) and nurses and physicians scored higher overall than licensed practical nurses or technicians and aides. Those institutions where design of the facility was judged to be better also were found to be the facilities where care was judged to be better.

These findings on a limited pilot study of 10 institutions have implications in terms of planning and evaluating burn care. This is also one of the few studies to look at organizational methodologies and correlate them with outcomes such as survival, length of stay, and cost. Another separate study combined NBIE data with turnover statistics. It was found that when shortages and instability of nursing staff occurred, mortality rates of patients hospitalized during those times were increased. 14

BURN PREVENTION NBIE data have also been instrumental in improving targeted burn prevention programs. Of the 95,000 cases, etiology information exists on 75,767 cases. Because the NBIE has been in existence for over 20 years, two equal time periods and groups of patients can be compared. Also, NBIE data can be combined with data from the U.S. census, allowing a definition of the populations at greatest risk of burn injury (Fig. 12). For instance, we know from these analyses that black men are at greater risk overall-as are infants and toddlers of any race, men entering the work force through middle age, and Americans age 70 years and greater. Prevention schemes then can be aimed at these groups. Knowing who, where, and how accidents occur allows pin-point prevention programs.

IRVING FELLER AND CLAUDELLA A. JONES

184 3.0 2.8

•

2.6

•

0 - -0

Males Females

2.4

n·47.5OO 2.2 2.0

a:

1.8

IU

1.6

IL

1.4

0

< ~

!!! 1.2 a: 1.0 .8 .6 .4 .2

AGE Figure 12. Risk factor analysis shows that at the 1.0 level, risk is "average." Above 1.0 the risk is greater; below, it is less.

Figure 13 shows that from 1964 to 1977, newborn to 2-year-old children were the patients most frequently admitted to NBIE hospitals. This finding holds true from 1978 to 1985, but there is also a shift to the right as a slightly older population is being admitted. Figure 14 shows where the accidents occur that involve these age groups. As you can see, location changes as the child gets older and moves outside. Figure 15 shows the most common activities involved in the accidents of 1- to2-year-old children. Knowing who, where, and how is essential to successful prevention. Flammable Fabrics Soon after the NBIE was established, it became evident that the most devastating injuries occurred when the patient's clothing had ignited. These analyses were instrumental in establishing flammability standards for children's sleepwear and for other garments and fabrics. 15 In recent analysis of NBIE data (Table 1), it was found that the legislation in flammable fabrics was related to an overall decrease in reported fabric ignition of 33 per cent. The data also have been used, and are being

185

THE NATIONAL BURN INFORMATION EXCHANGE National Burn Information Exchange ....n" Arbor, MicI\ill" I. F.. I.... M.D., Oir.tot

EPIDEMIOLOGY OF BURNS-1964·1985

WHO: DISTRIBUTION BY AGE

1964·1977 n-41.326

3000

2500 CII

C

.!!! 2000

~

'0•

1500

i

1000

A National Burn Information Exchange

"11ft Arbor. Mic"it-n

I. Feller, M.D., OlreclOf

EPIDEMIOLOGY OF BURNS-1964.1985

WHO: DISTRIBUTION BY AGE

1978·1986 n-38.674

3000

2500 CII

C

.!!! 2000

iii

Q.

0

0 1500

Z

1000

500

B

~

'!,UL.W.l.U~.U.L 0

20

30

40

50

60

70

80

90

100

Age in Years Figure 13. Age distribution of National Burn Information Exchange patients. A, 1964-

1977. B, 1978-1985.

186

IRVING FELLER AND CLAUDELLA A. JONES

National Burn Information Exchange Ann Arbor. Michigan I. Fell.r. M.D., Director

EPIDEMIOLOGY OF BURNS-1964·1985

WHERE: SPECIFIC LOCATION IN THE HOME

1978-1985 ns19.698

~BEDROOM

~OTHERINDOOR

[JOUTDOOR

I-

~ 30+--t~------~~--~: :~~~--~: :~+4~--~~~~------~ o a: ~ ~U4--c~------~=r---;· ~-tj~--1: :~+4~---1: :r+~r---1:

0-4

5-15

16-49

50-74

75-100

AGE IN YEARS UMBC, April 1986

Figure 14. Locations of burn accidents. The most frequent indoor accidents occur in the kitchen, bedroom, and bathroom.

used now, to establish the need for flammability standards for upholstery and self-extinguishing cigarettes. In addition, the NBIE data have been used by outside investigators to study accidents involving such items as mattresses, tents, liquid propane and natural gas, butane lighters, hotwater heaters, tractors, boats, and other consumer products. PROSPECTIVE PAYMENT SCHEME

More recently, NBIE data have been used to assess the Health Care and Financing Administration's prospective payment scheme-specifically the equity of the diagnosis-related groups (DRGs) assigned to bum injuries. Preliminary findings on a subset of institutions 16 indicate that substantial and significant differences exist among the seven institutions reviewed with regard to DRG case mix and length of stay. The investigators found that the current DRG classifications for bum injuries explained only about 15 per cent of the variability in length of stay among the study hospitals. Better definitions of patient severity factors that affect outcome are needed to make an equitable payment system. NBIE data could be used toward this end.

187

THE NATIONAL BURN INFORMATION EXCHANGE

Table 1. Clothing Ignition by Age and Year Group: 1964-1977 (n = 26,448) and 1978-1985 (n=32,568) AGE

0-4

5-15

16-49

50-74

74-98

Overall

1964-1977

Yes 30% No 70%

Yes 71% No 29%

Yes 71% No 29%

Yes 81% No 19%

Yes 61% No 39%

1978-1985

Yes 7% No 93% 27%

Yes 27% No 63%

Yes 64% No 36% Yes 31% No 69%

Yes 23% No 77%

Yes 51% No 49%

Yes 28% No 72%

44%

33%

48%

30%

33%

Decrease in fabric ignition

ETIOLOGY OF BURN ACCIDENTS FOR CHILDREN - AGE 1 &2

PULLING ON KITCHEN APPLIANCES

PULLING ON CONTAINERS OF HOT LIQUIDS

PULLING ON POTS, PANS. SKILLETSON STOVE

BATHING

o

===+1===+1==~I===+===.+=I===+====11::::::::: ~::::::::

£:1

50

100

150

200

250

300

Number of Cases

350

450

500

625 650

n=1,967

Figure 15. Etiologies of bum accidents involving 1 and 2 year old children, showipg that nearly all accidents are preventable. .

188

IRVING FELLER AND CLAUDELLA

A.

JONES

SUMMARY The NBIE, a voluntary registry of specialized burn-care facilities that was founded in 1964, currently has 50 active participants representing 35 per cent of the nation's hospital beds for burned patients. Participating physicians submit information on the initial hospitalization of emergent and acute burn patients and, separately, on the reconstruction process for these patients. As of January 1986, a total of 94,594 patient's data are on file from 130 hospitals; 13,671 of these are reconstructive and 80,923 emergent and acute admissions. Information concerning new patients is submitted at a rate of about 6000 patients annually. The data are analyzed using INQUIRE, an original data retrieval system. Data on treatment methods and outcome have been used to establish baseline standards for the burned patient's care and survival. In addition, these data have been used to document institutional differences in mortality rates and indicate methods used by the more successful hospitals. The data also are being used to describe the long process of recovery from severe burns and to monitor changes in outcomes of burn accidents continually. The result of these analyses has been documentation of an overall improvement in survival and decline in hospitalization times at all levels of burn severity. Data also can be used with institution-specific data to look at organizational variables affecting survival. Use of this epidemiologic data allows prevention projects to be targeted at the groups at greatest risk. A newer application looks at the equity of the RCFA prospective payment system based on the ORCs assigned to burn severity. The NBIE is an example of how a voluntary, national registry, properly computerized and effectively managed, can contribute to resolving the problem it was established to study. The NBIE has been useful in increasing the understanding of health profeSSionals and government decision makers of a complicated disease process. It has had a direct effect on the quality of patient care and on the process of controlling the incidence of burn injuries.

REFERENCES 1. American Burn Association: Burn care services in North America. Committee on Organization and Delivery of Burn Care, 1985 2. Burn Care Facility Study: Final Report of the Pilot Study, May 1982. UM and NlBM, ISRlHSR Program/CRUSK 3. Cornell RG, Feller I, Roi LD, et al: Evaluation of burn care utilizing a national burn registry. Emerg Med Serv 7:107, 1978 4. Evaluation of emergency medical services with a national burn registry: Final Report to the National Center for Health Services Research, Grant Number HS-10906 submitted by the University of Michigan Department of Biostatistics, and to the National Institute for Burn Medicine, Ann Arbor, Michigan, 1979 5. Feller I, Crane, KH: National burn information exchange. Surg Clin North Am 50:14251436, 1970 6. Feller I, Flanders S: Mortality review based on profile analysis. Quality Review Bulletin. 5:30-35, October 1979 7. Feller I, Flora JD, Bawol R: Baseline results of therapy for burned patients. JAM A 236:1943-1947, 1976

THE NATIONAL BURN INFORMATION EXCHANGE

189

8. Feller I, Flora JD, Flanders S: Baseline data on the mortality of burn patients. Quality Review Bulletin. 5:4-8, July 1979 9. Feller I, Jones CA, Wolfe RA: Effect of past medical history on survival. Presented at the American Burn Association annual meeting, Orlando, Florida, March 1985 10. Feller I, Tholen D, Cornell RC: Improvements in burn care, 1965-1979. JAMA 244:20742078, 1980 11. Feller I, Tholen DW, Herteg C, et al: The use of a national burn registry to evaluate and improve patient care. In O'Neill JT (ed): Proceedings of the Fourth Annual Symposium on Computer Applications in Medical Care. New York, Institute of Electronics Engineers, 1980 12. Flora JD: A method for comparing survival of burn patients to a standard survival curve. J Trauma 18:701-705, 1978 13. Flora JD, Davis TM, Roi LD: Length of stay and survival times for burned patients. Bums 5:3~2, 1978 14. Jones CA, Tholen D, Feller I, et al: 1981. Nurse retention in burn care: A report of 14 years' experience. Heart and Lung 10(2):295-308, 1981 15. U.S. Department of Health, Education and Welfare: First, second, and third annual reports to the President and the Congress on the studies of deaths, injuries and economic losses resulting from accidental burning of products, fabrics, or related materials. Washington, D.C., 1969-1971 16. Wolfe RA, Harrison RV, Wheeler JRC: Equity and incentives in the burn DRC classifications, Unpublished data, 1984 The University of Michigan Burn Center 200 North Ingalls Ann Arbor, Michigan 48104