_JAccid Emerg Med 1999;16:189-193

Injury surveillance in a children's hospital-overcoming obstacles to data collection

N V Doraiswamy

AbstractObjective-To understand the problemsinvolved in collection of injury surveil-lance (Glasgow Children Hospital InjuryReporting and Prevention Programme,CHIRPP) forms.Methods-Glasgow CHIRPP forms wereissued by the clerical staff to all eligiblechild carers for details of the injury oringestions by the child, and the retrievalrate of forms was monitored. Reasons forthe poor collection of forms were identi-fied and rectified.Results-The collection rate of GlasgowCHIRPP forms was poor when the systemwas introduced in 1993. It improved whenthe forms were issued by nursing staff, andconsiderable improvement was noted whenthe triage nurse was made responsible.Conclusions-When a named individualwas made responsible there was an im-provement in the retrieval of GlasgowCHIRPP forms. A few other simplerproblems relating to the retrieval offormswere identified and rectified.(7 Accid Emerg Med 1999;16: 189-193)

Keywords: injury surveillance; children; CHIRPP form;data collection

Accident prevention is a governmental priorityin the UK.i 2 In a recent literature review, theimportance of high quality local data fortargeting interventions to reduce childhoodaccidents and then evaluating outcomes wereemphasised.3 Accident and emergency (A&E)departments are the "missing link" in the rangeof sources currently generating data on injuriesin the NHS.4 In planning preventive measuresit is vital to collect and analyse local data onfrequency and circumstances of injuries, andinjury surveillance in A&E departments is animportant means of achieving this. The pur-pose of the present study was to understandand overcome the problems underlying the ini-tial poor collection rate of the injury surveil-lance forms in order to maximise the potentialof the system for injury prevention.

Royal Hospital for SickChildren, Yorkhill,Glasgow G3 8SJ

Correspondence to:Mr Doraiswamy, Consultantin Accident and EmergencyMedicine.

Accepted 14 January 1999

MethodsWith the help of a senior member of the Cana-dian Hospitals Injury Reporting and Preven-tion Program (CHIRPP)' childhood injurysurveillance was initiated in 1993-the Glas-gow Children Hospital Injury Reporting andPrevention Programme (Glasgow CHIRPP).6A large explanatory notice (size A2) about

CHIRPP and the importance of injury surveil-lance was displayed in the patient's waitingroom so that carers and children would be able

to read about it. The same explanation (sizeA3) was displayed in all the examination, treat-ment, and plaster rooms. After completion oftriage, a verbal explanation of GlasgowCHIRPP was offered and an A4 size infor-mation sheet, with the same explanation, wasdistributed along with the injury surveillanceform (fig 1), kept in a pad with a pen, to theaccompanying carers of children (up to the ageof 14 years) presenting with injury or ingestion.The carers were not required to complete

the form if they chose not to. They recordedthe details about the child (name, sex, date ofbirth, address, postcode), about themselves(telephone number, relationship to the child,and time of visit), injuries in a free text form(time, location, circumstances, mechanism), orsitting position (if the patient was a passengerin a vehicle), and whether any consumer prod-uct or safety equipment was involved. On aver-age, it took fewer than three minutes to recordthe details by the carer. When necessary, anurse or member of the medical staff assistedthe carers when completing the form. Afterexamining the patient medical staff collectedthe form and completed the details about thenature of injury, anatomical part involved, anddisposal on the reverse side of the form (fig 2).On average this could be done in fewer than 20seconds. The forms were collected each morn-ing and computerised. Glasgow CHIRPP soft-ware has been written in a fourth generationcomputer language (SCULPTOR). At presentthis is a stand alone system, and with technicalimprovements it may be possible to incorpo-rate it into routine A&E department computersystems in the future.The collection rate of Glasgow CHIRPP

forms was monitored and discussed at intervalswith the clerical, nursing, and medical staff andthe carers to identify the problems andsolutions.

ResultsIn the initial stages, the Glasgow CHIRPPforms were handed over to the carer at the timeof registration by the clerical staff, but therewas an unacceptably low level of capture offorms (fig 3). Despite tackling the problemsindicated by the carers (table 1), retrieval of theforms was only minimally improved. The diffi-culties perceived by the clerical staff aboutissuing forms were established after discussion(table 2). Therefore, after consultations, thenursing staff took over the responsibility forthis fimction in 1996 and the collection rate offorms gradually improved (fig 4) and severalproblems were identified (table 3). The antici-pated improvement was still not occurring and

189 on N

ovember 20, 2021 by guest. P

rotected by copyright.http://em

j.bmj.com

/J A

ccid Em

erg Med: first published as 10.1136/em

j.16.3.189 on 1 May 1999. D

ownloaded from

Doraiswamy

after discussion with nursing and medical staffit was concluded that no one individual feltresponsible. It was therefore decided that thetriage nurse, who could be identified from theduty rota for any particular period, should bemade responsible for issuing the forms. It wasalso felt that the triage nurse was a better per-son to recognise injuries from illnesses as moredetails would be given to her by the carer thanto the clerical staff. Such a change hasproduced considerable improvement in thecollection rate of Glasgow CHIRPP formsfrom February 1997 onwards (figs 3 and 4).

DiscussionFew surveillance systems have been imple-mented in the UK, apart from the homeand leisure accident surveillance systems(HAAS and LASS) of the Department ofTrade and Industry.7 These systems arelimited in scope and operate in only a fewareas of the country with only one partici-pating hospital in Scotland. Glasgow CHIRPPwas therefore introduced to try to meet thelocal needs for injury surveillance data.6 Glas-gow CHIRPP forms were used to collect andanalyse data on circumstances, mechanisms,

SerialL I l]Royal Hospi for Sick Children, YorkhihI For all injuries and Ingetions etc

Pins. giv as much detaf - possible

Child'ssumameHomeaddress

PostcodeHome Tel I

Child's ,.Boy El El Girlfirst name

Day Month Year

11[:1 1 11 X] DateoofZbirth mZmNo Work Tel No

I~ 1'1 11111111111111Day Month Year am El

1. When did the incident occur? Date: [ ' [ T imef [ pm

2. Where did the incident occur? (Give exact place and district)(e.g.-at junction of Byres Rd and Gt George St, at side of road in Hilihead; in bathroom at home in Parick

__11IZ3. What was the child doing at the time the incident occurred?

(e.g. washing up; having tea and playing around with sister; playing soccer, crossing road on way to school)

4. What went wrong?(e.g. chased by dog and lost control of bike; fell out of tree; toy truck broke; hot cofee knocked over)

5. What actually caused the injuy?(e.g. swallowed perfume; grandma's sleeping tablets; jammed finger on hinge side of door)

6. If a specific product or article was involved, please give details;Product Brand or nnake

Type ormodel

7. What safety precautions or devices were being used at the time the injury occurred?(e.g. seat belt; infant car seat child rfeistant bottle cap; bicycle helmet; none)

8. If a motor vehicle was involved, please give details;Make and model I

(e.g. Honda Civic)

] Year [ Type of fehicle|e(e.g. estate, saloon)

9. If the child was in a motor vehicle, show the child's ting positionEnter the number showing the child's position using the diagram El

Sometimes we need to get in touch with parents foradditional infom8ation about an Injury. ff you donot vwsh to be contacted please place an X here

Please neots: tdriver

(other position which le not shown)

IMPORTANT: PLEASE HAND THIS SHEET BACK TO THE DOCTOR WHEN YOU ARE SEEN

Figure 1 Glasgow CHIRPPform for carer.

0 e ____ ,. ., r - s r w

L. ..... . . ,, , ,, . ,,, ........................ .. .. . _ . _ . ,I

190

on Novem

ber 20, 2021 by guest. Protected by copyright.

http://emj.bm

j.com/

J Accid E

merg M

ed: first published as 10.1136/emj.16.3.189 on 1 M

ay 1999. Dow

nloaded from

Injury surveillance in a children's hospital

and types of injuries in children up to the ageof 14 years who presented to the A&Edepartment for injuries or ingestions. The dataindicated the potential of the system for iden-tifying both hazardous environments andvulnerable population subgroups at whomspecific preventive measures could betargeted.The main methodological problem encoun-

tered with Glasgow CHIRPP was the low cap-ture rate of forms; this was due mainly to poorstaff compliance. Persuasion, encouragement,and repeated individual and group discussions

Doctor's name: (capitals)

with the clerical, nursing, and medical staffstimulated interest and motivation, with lim-ited results. More forms were issued andcollected when responsibility was transferredfrom clerical to nursing staff, especially whenthe triage nurse was made accountable for thesystem from February 1997 (figs 3 and 4). Theservices of a triage nurse has been utilised pre-viously to collect injury surveillance forms.8Clerical staff felt that they could not identifywhether the problem was due to injury orillness from parents who were possibly offeringa limited description. Therefore the forms were

Note: NEC means "not elsewhere classified"f

Complete only for first attendance of a particular episode

m Severest

Second

Third

SELECT UP TO THREE CODES

Sys_tmic andpeblp Injury90 multiple system trauma91 poisoning (through sidnfiungs/mouth/etc)88 asphyxiation or repiratory difficulty94 electric shock96over-exertion, heaot/cold stress* concqssion87 dental injury95 no injury detected

Soft.tissue01 cut/laceration02 puncture0 bite04 superficial abrasion05 penetrating wound06 other wound, including amputation07 heenistomalbruislng07 heemorrhage09 lnfismmadon/oedema/tenderness110 burn, fll thickness11 bum, perle thicknes1t2foreign body In soft tissue13 damage to major blood vessel14 crushing injury15 obstruction16 frostbite

Bons, tendon, or Joint20 fracture21 dislocation22 sprain/strain23 subluxation

[] SELECTONE CODE

0 accidental injury (that is, unintentional)1 intentionally sIf inflicted, or possibly so2 vIcm of asult or possibly so6 unknown Intent4 NAIVCSA

Severest IJ . Ithes boxes write the body part code

Second for each of the corresponding injuries

Third I]recorded in section 1 left

Systemi_ and spedal injury000 defined as in section 1 at left

Head101 eye102 ocular adnexum103 nose104 mouth external,

e.g. jaw, lip105 ear106ff1/heek/orehead/calp107 skull be108 skull vault109 neck, NEC198 other injury to headUpper eemity201 clavicle202 scapula203 shoulder, NEC204 humerus205 upper arm, NEC206 radius, ulna207 elow208 forearm209 wrist210 carpal bone211 etacrpl bone212 digWitphalnx213 hand, NEC298 other injury to upper

extremity

301 hip302 femur303 thigh304 kneeipatetla305 tibia/fibula306 log307 ankle308 tarsal bone309 metatrsal bone310 digit/phalanx311 foot, NEC398 other injury to lower

extremityv--

W| | SELECT ONE CODE

01 advice only02 treated sent home10 testarview elaw in A&E/GC03 treated, referred to outpatients04 treated, referred to family doctor05 treated, other referral

Trunk401 rib(s)402 sacroiliac joint403 spine (including cervical), excluding cord404 pelvis

405 cheat, NEC408 abdominal wall407 upper back, NEC408 lower back, NEC409 genitalia410 heart411 kidney(s)412 bladder498 other injury to trunk

Respraory tract501 pharynx502 arynx503 trachea504 bronchus505 lung598 other injury to respiratory tract

D0011 system801 mouth internal, e.g. gum, palate802 oesophagus603 stomach/duodenum604 small bowel606 colon806 rectum607 liver608 spleen809 pancreas898 other injury to digestive tract

Nevous system701 brain, not concussion702 brain stem703 cervical spinal cord704 thoracic spinal cord705 lumbar spinal cord706 peripheral nerve798 other iniurv to narvusl svstehm

06 short stay observation in daysurgery unit/A&E

07 admitted to hospital08 transferred to other hospital09 DOA or died in A&E

Figure 2 Glasgow CHIRPPform for medical staff; CSA = child sexual abuse;DOA = dead on arrival; GC = generalclinic; NAI = non-accidental injury.

191

I

on Novem

ber 20, 2021 by guest. Protected by copyright.

http://emj.bm

j.com/

J Accid E

merg M

ed: first published as 10.1136/emj.16.3.189 on 1 M

ay 1999. Dow

nloaded from

192 ~~~~~~~~~~~~~~~~~~~~~~~~Doraiswamy

--*- 1998

*- 1997

-0a- 1996

-A 1995

-0*- 199",4'-.-~~~~~

--I

T T TJ F M

T~A M J J

Month

Figure 3 Number of Glasgow CHIRPPforms collected during 1993-98.

Table 1 Reasons identified for poor collectGlasgow CHIRPPforms: carers

*Anxiety about child/injuries* "Cannot recall all details"* Details not known by accompanying person* Language problems* Reading glasses not available* Illiteracy* Influence of alcohol* Not willing (feeling of guilt, although this w

Table 3 Reasons identified for poor collection rate ofGlasgow CHIRPPforms: nursing and medical staff

* Additional workload* Too many patients waiting* Too long waiting time* Lack of encouragement (resistance to the idea)* Not aware of importance* Considered a short term research project* No individual accountability* Not handing over during change of shift* "Forgot to give/collect/fill in/lodge in collection tray"

in full also increased as the same medicalstaff had to fill in the form when the

-aincomplete forms were returned. The carers0 ~~~were helped when problems were recognisableI ~~(table 1).

) N D As this is a tertiary, referral hospital,Glasgow CHIRPP forms were not filled inwhen a child was transferred directly to theintensive care unit or ward from another

tion rate of hospital, bypassing the A&E department. Thisis being addressed and will enhance thecollection rate of forms when the problem issolved.

Injury surveillance has great potential' and,once the obstacles to collecting data have beenovercome, the complete data will help pave theway for planning local accident prevention in

vas rare) the future.

Table 2 Reasons identified for poor collection rate ofGlasgow CHIRPPforms: clerical staff

* Pressure of usual work* Considered as additional workload* Decrease in manpower-only skeleton staff duringweekends, public holidays, out of office hours (relatively highattendances)

* Not realising importance

not necessarily being distributed to the carersof all injured children or those who hadingested medications.The supplies department was alerted to

the need to provide pens, pads, and theforms in sufficient numbers in advance. Themedical and nursing staff were constantlyencouraged. The number of forms completed

ConclusionThe collection rate of injury surveillance formscan be enhanced by identifying and rectifyinglocal problems. A named person, like the triagenurse, who is available at all times in the A&Edepartment, will help to identify suitable carersand therefore considerably improve the collec-tion of injury surveillance forms. Theimportance of cooperation and teamwork atall stages should be reinforced as and whennecessary.

I would like to thank Mrs Claire Donati for secretarialassistance and Miss jean Hyslop and Mr Stephen Beaton for thefigures. My sincere thanks are due to Susan MacKenzie andMargaret Herbert, Chief of the Injury Division, who areinvolved in CHIRPP in Canada, for their help and support. Myspecial thanks are due to the clerical, nursing, and medical staffand clinical audit office for their continued support.

- *Children attending forinjurieslingestions

* *GlasgowCHIRPP1forms collected

1500 H- A

J FMAM J JASOND J FMA MJJASOND J FMA M JJASO N

1996 1997 1998

Figure 4 Number of children who attended for injuries or ingestions and Glasgow CHIRPP forms collected in 1996-98.

1600 r-

1400 F-

1200

1000

z 800

600

400

200

0

2000 r-

1000 F

500 F-

0L

dr-

192

on Novem

ber 20, 2021 by guest. Protected by copyright.

http://emj.bm

j.com/

J Accid E

merg M

ed: first published as 10.1136/emj.16.3.189 on 1 M

ay 1999. Dow

nloaded from

Injury surveillance in a children's hospital 193

Conflict of interest: none.Funding: none.

1 Department of Health. The health of the nation. A strategy forhealth in England (1986). London: Department of Health,1992.

2 Scottish Office. Scodand's health, a challenge to us all.Edinburgh: HMSO, 1992.

3 Towner E, Dowswell T, Jarvis S. Reducing childhoodaccidents. The effectiveness of health promotion interventions: aliterature review. London: Health Education Authority,1993.

4 Deane M. Child accident data: accessible and available? JfPublic Health Med 1993;15:226-8.

5 Canadian Hospitals Injury Reporting and PreventionProgram. CHIRPP news. Issue 2. Ottawa: Bureau ofChronic Epidemiology (Health Canada), July 1994.

6 Stone DH, Doraiswamy NV. The Canadian HospitalsInjury Reporting and Prevention Program (CHIRPP) inthe UK: a pilot study. InjPrev 1996;2:47-51.

7 Consumer Safety Unit, Department of Trade and Industry.Home and leisure accident research. London: DTI, 1995.

8 Maitra A. School accidents to children: time to act. J AccidEmerg Med 1997;14:240-2.

9 Ozanne Smith J, Nolan T, eds. Hazard (vol 1, editions 1-10,1988-92, injury data and prevention). Melbourne: Victo-rian Injury Surveillance System, 1993.

Annual Scientific Meeting of the Faculty ofAccident andEmergency Medicine

3-4 December 1999, Royal College ofPhysicians ofLondon

Call for abstractsForms are available from the Faculty of Accident and Emergency Medicine, 35-43 Lincoln'sInn Fields, London WC2A 3PN (tel: +44 (0) 171 405 7071, fax: +44 (0) 171 405 0318,e-mail: FAEM@compuserve.com)Abstract forms must be received by 6 August 1999Authors will receive notification of acceptance/rejection

on Novem

ber 20, 2021 by guest. Protected by copyright.

http://emj.bm

j.com/

J Accid E

merg M

ed: first published as 10.1136/emj.16.3.189 on 1 M

ay 1999. Dow

nloaded from