Costos SV Belize

7/25/2019 Costos SV Belize

1/11

326 Rev Panam Salud Publica 28(5), 2010

Economic impact of fatal and nonfatalroad traffic injuries in Belize in 2007

Ricardo Prez-Nez,1 Martha Hjar-Medina,2,3 Ileana Heredia-Pi,4

Sandra Jones,5 and Eugnia Maria Silveira-Rodrigues6

Objective. To estimate the economic cost of road traffic injuries in Belize in 2007.Methods. A cross-sectional study was conducted using secondary cost data, assuming thehealth system and social perspectives. Epidemiologic information was obtained from the mor-tality database, the national hospital discharge database, and administrative records from po-lice and the Ministry of Health. A health provider survey was carried out in order to estimatethe postdischarge ambulatory utilization figures. Direct cost was estimated with the World

Health Organization WHO-CHOICE (CHOosing Interventions that are Cost Effective) data-base. Prehospital costs were obtained from the Belize emergency response team. After estimat-ing years of potential life lost using the Belize life expectancy for 2008 and methodology pro-posed by the Pan American Health Organization, the indirect cost associated with prematuredeath was estimated with the human capital approach. Total estimation of road traffic injurieseconomic costs used a decision tree model approach. Multiway sensitivity analysis was usedto incorporate uncertainty in the estimations.Results. Sixty-one people died due to road traffic injuries during 2007, 338 were hospital-

ized, and 565 people were estimated to be slightly injured. A total of 2 501 years of potentiallife were lost in Belize due to premature death, with a total economic cost of US$11 062 544.This figure represents 0.9% of the Belize gross domestic product. Direct cost was estimated atUS$163 503, of which 2.4% was spent on fatalities, 46.7% on the severely injured, and 50.9%on the slightly injured.Conclusions. The economic cost estimations make clear the need to prevent road traffic in-

juries with a strategic and multisectoral approach that focuses on addressing the main prob-lems identified.

Accidents, traffic; health care costs; costs and cost analysis; Belize.

ABSTRACT

In recent years, there has been a signif-

icant increase in the number of collisionson major highways and roads in Belize.

Between 2004 and 2006, the Government

Information System of the Police Depart-ment reported approximately 6 295 colli-

sions, of which 128 were identified as

fatal, resulting in 143 deaths. Most of thedeaths occurred in 2004, with a slight de-crease in 2005 and 2006. Statistics fromthe Ministry of Health indicate that in2005, death as a result of road traffic in-

juries (RTI) was identified as the fourthleading cause of death in general, regard-less of age and sex (1). The adjusted mor-tality rate due to RTI in 2006 was 15.4 per100 000 population. The country has laws

Key words

Investigacin original / Original research

Prez-Nez R, Hjar-Medina M, Heredia-Pi I, Jones S, Silveira-Rodrigues EM. Economic impact offatal and nonfatal road traffic injuries in Belize in 2007. Rev Panam Salud Publica. 2010;28(5):32636.

Suggested citation

1 Health Systems Research Center, NationalInstitute of Public Health, Cuernavaca, Mexico.Send correspondence to: Ricardo Prez-Nez,[email protected].

2 Population Health Research Center, National Insti-tute of Public Health, Cuernavaca, Mexico.

3 Pan American Health Organization/World HealthOrganization Collaborating Centre for Research onInjuries and Violence, Cuernavaca, Mexico.

4 Survey and Evaluation Research Center, NationalInstitute of Public Health, Cuernavaca, Mexico.

5 Pan American Health Organization/World HealthOrganization, Belize Country Office, Belize City,Belize.

6 Pan American Health Organization, SustainableDevelopment and Environmental Health Area,Washington, DC, United States of America.


2/11

Rev Panam Salud Publica 28(5), 2010 327

Prez-Nez et al. Economic impact of road traffic injuries in Belize Original research

that cover speed limits, blood alcoholcontent (0.08%), helmet use, and seat beltuse (although not all seating positions arecovered). However, application of theselaws was considered very weak. There isnot a law for use of child restraints (2).

RTI can have a significant negative im-pact on a country at all levels, as they

cause a large number of sick days and anelevated amount of healthy life yearslost. These factors have serious conse-quences on society as a whole. In Mexico,for example, it has been documented thatRTI are the second leading cause of or-phaned children, which has implicationsin terms of poverty perpetuation in low-and middle-income countries such asMexico (3). Moreover, since RTI affect theyoung population in its most productiveyears, the economic cost in terms of med-ical treatment, rehabilitation, and loss ofproductivity tends to be high, with eco-

nomic repercussions in societies (3, 4).In the United States of America, the

costs of motor-vehicle-related fatal andnonfatal injuries in 2005 exceeded US$99

billion. Costs associated with motor ve-hicle occupants fatal and nonfatal in-

juries accounted for 71.0% ($70 billion) ofall motor-vehicle-related costs (5). InBrazil, the cost of road traffic accidentson highways amounted to US$10.0 bil-lion per year, which is equivalent to 1.2%of the Brazilian gross domestic product(GDP). Among the examined cost ele-ments, loss of production had the great-

est economic impact, followed by vehicledamage and medical costs. Fatal acci-dents had the highest mean valueUS$200 000 per accidentshowing thatthe more severe the accident, the higherthe cost. The total cost can be evenhigher should the costs of environmentaldamage and posttraumatic stress disor-der be included (6).

Not enough data are available to de-termine accurately the cost of RTI in Be-lize. In order to understand the need foraction, it is imperative that a comprehen-sive understanding of the cost of RTI be

developed. For that reason, there is anurgent need to calculate the economiccost of RTI in Belize and to evaluate theconsequences that current RTI rates rep-resent for Belize, which will providevaluable information in terms of injuryprevention and safety promotion. Thisinformation will serve not only to iden-tify the economic cost from a publichealth perspective but also to show the

potential economic benefits from reduc-ing RTI by undertaking road safety mea-sures such as developing and imple-menting appropriate interventions (witha multisector approach). The main objec-tive of this study was to estimate the eco-nomic costs (direct and indirect) of fataland nonfatal RTI in Belize in 2007 from

health system and social perspectives.

METHODS

Study population

All road traffic injuries in Belize in2007 were included in this study accord-ing to ICD-10 (7) criteria, which includethe following ICD-10 codes for fatal andnonfatal RTI: V02V04 (.1, .9), V09 (.2, .3,.9), V12V14 (.3.9), V19.4V19.6, V20V28 (.3.9), V29V79 (.4.9), V80.3V80.5,V81.1, V82.1, V83V86 (.0.3), V87.0

V87.8, V89.2, and V89.9. All injuries werecategorized in three mutually exclusiveseverity categories following previousproposals (8):

Fatal injury: an injury that causes thedeath of the injured person in the first30 days after the accident occurred.The 30-day definition of a road trafficfatality proposed by the World HealthOrganization (WHO) was used, al-though the official Belize definition in-cludes all people who die within a yearafter the collision (9). In order to facili-

tate international comparisons it wasimportant to make this adjustment.

Nonfatal severely injured: either aperson remains in the hospital as aninpatient or any one of the followinginjuries is sustained regardless ofwhether the individual is detained inthe hospital: fractures, concussions,internal injuries, crushing, severe cutsand lacerations, and severe generalshock requiring medical treatment.This category includes deaths thatoccur after 30 days.

Nonfatal slightly injured: a minor in-

jury such as a cut, sprain, or bruise.

Data collection and confirmation ofvariables

A secondary database analysis wasundertaken in order to estimate epi-demiologic information. Fatal injurieswere obtained from a mortality databasethat contained all deaths that occurred

during 2007. Serious injuries were ob-tained from the national hospital dis-charge database, which contained all dis-charges during 2007. This informationwas used to estimate the total numberof people hospitalized (assuming theyall had serious injuries). Administrativerecords from the Belize Police Depart-

ment were used to estimate the numberof slightly injured people as informationfrom the health sector was not availablefor the number of slightly injured peoplewho visited the emergency room. The re-liability of this information is not knownwith certainty, but it was the best infor-mation available. Finally, informationabout population figures was obtainedfrom the World Bank online database(http://ddp-ext.worldbank.org/ext/DDPQQ/member.do?method=getMembers&userid=1&queryId=135).

Estimation of direct costs included ex-

penditures for medical care and treat-ment such as ambulance and prehospitalcare services, emergency room use, hos-pital care, physician services, nursinghome care, drugs and other medicalneeds, postdischarge ambulatory med-ical consultations, and rehabilitation.Categories included for each severitylevel group depended on the availabilityof information.

Unitary cost data were obtained asfollows:

Ambulance and prehospital care ser-

vices: estimates for ambulance andprehospital care services used sec-ondary data provided by the BelizeEmergency Response Team (BERT).With a top-down approach, the totalexpenditure figures of this institutionwere prorated by all services providedin order to estimate an approximatecost per run. Total figures correspondto the sum of the following cost cate-gories: salaries, fuel, maintenance,utilities, medical supplies, insurance,and equipment. Medical supplies aresubsidized by the Belize Ministry of

Health. In the same way, maintenanceis subsidized by Cisco Construction(free labor). This cost was assumed foreach of the six districts that divide thecountry, although BERT services areprovided only in Belize district.

Hospitalization, emergency room hos-pital-based services, and ambulatorymedical care costs: secondary data fromthe WHO-CHOICE (CHOosing Inter-


3/11


Original research Prez-Nez et al. Economic impact of road traffic injuries in Belize

ventions that are Cost Effective) data-base were used to estimate direct costs(available from http://www. who.int/choice/country/blz/cost/en/print.html). Information on specific econo-metric techniques used to estimatecountry-specific costs is available else-where (10). Cost per bed-day represents

only the hotel component of hospitalcosts, including personnel, capital, andfood costs but excluding drugs and di-agnostic tests. Cost per outpatient visit

by hospital level and cost per visit at ahealth center include depreciated capi-tal items but exclude drugs and diag-nostics (10). All costs in this study arepresented in US dollars after convertingthem to 2007 Belize dollars by meansof the national consumer price index(available from http://www.statistics

belize.org.bz/dms20uc/dynamicdata/docs/20100901211434_2.pdf). Costs ac-

crued after a year or more were dis-counted at annual rates of 3% and 5%.

Utilization figures were obtained asfollows:

Emergency ambulance services: esti-mated by using a database containingall emergency ambulance services pro-vided by BERT on RTI patients during2007. Although this information wasonly for Belize district, it made it pos-sible to estimate the proportion of se-verely injured people who received

ambulance and prehospital care ser-vices for the entire country.

Ambulatory health care services: to es-timate postdischarge utilization figures,a health provider survey was carriedout. Experienced medical doctors withdifferent specialties (general surgery,maxillofacial surgery, neurosurgery,ophthalmology, urology, orthopedics,and traumatology and rehabilitation)were consulted and asked to note theproportion of RTI patients who would

be hospitalized based on ICD-10 diag-nosis, which was taken from the hospi-

tal discharge database. In addition, theywere asked to estimate the proportionof patients who would potentially useambulatory health services (includingrehabilitation) and the number of med-ical consultations for hospitalized andnonhospitalized patients. When spe-cialists did not fill out the questionnaireto provide missing information, thosefigures (both the proportion that useservices and the number of consulta-

tions) were assumed to be the same asthose for a similar injury (from ananatomic and severity perspective).When that option was not possible, themedian of the number of ambulatorymedical consultations (or rehabilita-tion) of all the other injuries was im-puted. The total number of rehabilita-tion consultations estimated for eachpatient was included in the total num-

ber of ambulatory medical consulta-tions. This survey also made it possibleto estimate utilization figures for se-verely injured nonhospitalized patientsand slightly injured non-emergency-room users (information presented inTable 1). Table 1 shows mean, median,and mode of hospital stay (obtainedfrom the hospital discharge database).

By multiplying the median figures forutilization per unitary cost, direct costswere estimated for Belize. In this sense,all emergency room medical consulta-

tions, specialized medical consultations,and rehabilitation were assumed to betertiary outpatient visits to hospital ser-vices. In the same way, given that 63.6%of hospital discharges received attentionat the Karl Heusner Memorial Hospital(KHMH) and as hospital costs do notconsider drugs and diagnostic tests, itwas decided to use the cost per bed-dayof a tertiary hospital level. Finally, hospi-tal stay information was obtained fromthe hospital discharge database for thosewho died. Both survivors and casualtieshave median lengths of stay of two days.

Indirect cost estimation was calculatedby using the human capital approach, inwhich the productivity cost was esti-mated as the future reduction in grossincome due to mortality and morbidity.In this study, the Belize average incomefor 2007, documented by the Belize So-cial Security Board, was used for indirectcost estimates. Indirect costs associatedwith premature death take as referencethe Belize life expectancy for 2008 docu-

mented by the Pan American Health Or-ganization (PAHO) (76.1 years) (11). Lifeexpectancy was used instead of age at re-tirement (55 years) to take into accountcountry preferences on how society val-ues life at different ages, something im-portant to consider (12). This decisionwas taken as recommended by the BelizeNational Road Safety Committee. In thissense, elderly people in low- and mid-

dle-income countries tend to have a verysmall income or none at all. This situa-tion does not mean that their time has novalue or should not be valued. This cir-cumstance is similar to the value of timefor housewives. Although in real termsthey might not receive a salary, theirwork and activities are highly valued(and needed) by societies (11). Calcula-tions were made as follows:

Indirect cost of fatal RTI =

Loss of output (death) = (1)

whereS = salary,r = discount rate, andn = number of years of potential life

lost due to premature death.

For morbidity figures, indirect costs in-cluded the valuation of postdischarge av-erage time injured spent at home recov-ering from injuries (72 days). In addition,the value of productivity losses due totransitory or permanent disability wasestimated by using the value of disable-ment pensions awarded to the road traf-

fic injured. This information was pro-vided by the Belize Social Security Board.

Estimating total RTIeconomic costs

Total estimation of RTI economic costsused a decision tree model approach aspresented in Figure 1. The decision treeallows one to disaggregate patients byfatal and nonfatal type of injury. The lat-

S

r n

n

( )10 +

TABLE 1. Use of health care services by patients with road traffic injuries, Belize, 2007

Hospitalized Nonhospitalized

Hospital Ambulatory Ambulatorystay medical Rehabilitation medical Rehabilitation

Variable (days) consultations consultations consultations consultations

Mean 5 4.87 19.71 4.63 2.11Median 2 3 2 3 2Mode 1 2 0 2 2


4/11


FIGURE 1. Decision tree to model economic cost of road traffic injuries (RTI) in Belize, 2007

Did not receiveattentionatscene

Not used ER

Nonuser

Nonuser

Used ambulatoryservices

Usedrehabilitation

Nonuser

Nonuser


Usedrehabilitation

Used ER

Slightly injured

Receivedmedical attention

Not used ER

Nonuser

Nonuser


Usedrehabilitation

Nonuser

Nonuser


Usedrehabilitation

Used ER



Nonhospitalized

Nonuser

Nonuser


Usedrehabilitation

Nonuser

Nonuser


Usedrehabilitation

Hospitalized

Severely injured

Nonfatal

RTI

Die >30 days

Fatal



Nonhospitalized

Nonuser

Nonuser


Usedrehabilitation

Nonuser

Nonuser


Usedrehabilitation

Hospitalized

Hospitalized

Nonhospitalized

Hospitalized

Nonhospitalized


Note:ER: emergency room.

ter category is disaggregated into se-verely injured and slightly injured. Inaddition, the decision tree follows thenatural history of disease, starting with

prehospital medical attention at thescene, continuing with hospital-basedcare (hospitalization versus emergencyroom only), utilization of ambulatory

health services, and disaggregating am-bulatory services into rehabilitation ser-vices. Each tree branch has a probabilityof occurrence and a specific associated


5/11



cost. In this way, total cost per treebranch corresponds to the multiplicationof unitary cost per total number of in-

jured estimated for that specific branch.The individual total cost of RTI for this

study was thus obtained after adding di-rect costs to indirect costs of those hospi-talized and those who received emer-

gency care only. The economic cost ofRTI in Belize was estimated in the fol-lowing manner:

(2)

whereTC = total economic cost of RTI,

AECd = average economic cost per RTIdeathf,

AECs = average economic cost of se-

verely injured s,AECsl = average economic cost of

slightly injured sl,D = total number of RTI deaths regis-

tered in 2007,S = estimation of total number of se-

verely injured, andSL = estimation of total number of

slightly injured.

Analysis

A descriptive analysis was performedfor all databases. Central tendency and

dispersion measures were calculated forthe continuous variable (mean, median,standard deviation, maximum, and min-imum values) as well as frequencies andpercentages for the categorical variablesusing Stata 9.2 (13).

Uncertainty surrounding some of themodels assumptions was explored witha multiway and probabilistic sensitivityanalysis. The analysis was performed forvariables such as prevalence and inci-dence figures (death, hospitalized, at-tended at emergency room) and for threediscounted rates (0%, 3%, and 5%), al-

though in the text only results using the3% discount rate are presented. Sensitiv-ity analysis of the number of RTI hospi-talized considered a literature review tosimulate potential scenarios consideringthe possibility of hospital underregistra-tion (1421). This was also the case forslightly injured people. This part of theanalysis was carried out with software@RISK 5.5 (22). Figures are presented inUS dollars, using an exchange rate of

US$1.00 = BZ$2.00, which has beenpegged to the US dollar since 1976 (11).

RESULTS

Epidemiology of RTI in Belize

During 2007, 63 people died as a con-

sequence of RTI, of whom 79.4% were

men (Table 2). This figure is equivalentto a total of 61 people if adjusted byWHOs 30-day definition (total numberof deaths multiplied by the adjustmentfactor of 0.97). Mean age was 37 years(median = 33) and 68.3% of all injuredwere between 15 and 49 years of age. Itwas not possible to determine the type of

road user most affected, because most

TC Belize AEC D AEC Sdi

n

si

n( )=

+

= =

1 1

+

=AEC SLsli

n

1

TABLE 2. Descriptive analysis of road traffic injury casualties, Belize, 2007

Variable Frequency %

Individual characteristicsSex

Women 13 20.6Men 50 79.4

Age group (years)014 4 6.31524 13 20.62549 30 47.65074 14 22.2

75 2 3.2Civil statusCommon law 7 11.1Divorced 0 0.0Married 12 19.0Single 43 68.3Unknown 0 0.0Widowed 1 1.6

SchoolingNone 3 4.8Primary 43 68.3Secondary 5 7.9Tertiary 3 4.8Unknown 9 14.3

Activity/occupationWage earner 41 65.1Nonremunerative activity 5 7.9

Unknown 17 27.0Death and medical attention informationPlace of death

Hospital 32 50.8Road/street 27 42.9Home/farm/workplace 2 3.2Route to hospital 0 0.0Unknown 2 3.2

District of residenceBelize 32 50.8Cayo 8 12.7Corozal 5 7.9Orange Walk 8 12.7Stann Creek 3 4.8Toledo 7 11.1

District of death occurrenceBelize 48 76.2

Cayo 0 0.0Corozal 3 4.8Orange Walk 6 9.5Stann Creek 2 3.2Toledo 4 6.3

Hospital that registered deathBelmopan Hospital 0 0.0Central Region Belize District 48 76.2Corozal Town Hospital 3 4.8Orange Walk Hospital 6 9.5Punta Gorda Hospital 4 6.3San Ignacio Town Hospital 0 0.0Southern Regional Hospital 2 3.2


6/11


deaths were coded as other and un-specified. Most deaths took place in Be-lize district (76.2%) and 50.8% of alldeaths occurred in a hospital. This valuetranslates to a mortality rate of 20.1deaths per 100 000 population. When an-alyzed with a Poisson regression model,on average, the relative risk of dying as a

consequence of RTI in Belize decreased7.1% each year during 20012007 (95%confidence interval 3.3% to 10.8%).

During 2007, 338 hospital dischargesdue to RTI were recorded in all publichospitals in Belize (hospital dischargerate of 117.36 discharges per 100 000population). Of them, 74.9% were men,with a mean age of 29 years (standarddeviation = 18, median = 26) and 67.5%were between 15 and 49 years of age(Table 3). Most road users were also clas-sified as other and unspecified. Alarge majority of hospital discharges

were reported by KHMH, the only ter-tiary level hospital in Belize. Of all RTIhospitalized, 2.7% died during hospital-ization (n = 8).

Table 4 presents all emergency ambu-lance services provided by BERT in Be-lize district in 2007. A total of 222 ambu-lance services were solicited, although 24of them did not require any service. Ofthe remainder, 47 (23.7%) were severelyinjured (42.0% of the total number hospi-talized in Belize district the same year),73.2% were slightly injured, and 3.0% ofthe injured died. Most people attended

were men (71.7%) and 74.5% were be-tween 15 and 49 years of age. Of all in-

jured, 90.4% received treatment andwere transferred to a medical institution.

No detailed information was availableon patients slightly injured after a colli-sion. Similarly, no information on emer-gency room hospital-based medical atten-tion was provided. Using BERT figures(an estimated 77.5% of the total numberof nonhospitalized slightly injured in Be-lize district the same year), a total of 565slightly injured was estimated.

Economic cost of RTI

Direct cost. A total of US$163 503 wasestimated for direct costs during 2007, ofwhich 2.4% was spent on fatalities, 46.7%on severely injured, and 50.9% on slightlyinjured. Figure 2 shows that most of thiscost was estimated for prehospital care(36.8%), hospitalization (21.5%), and am-

bulatory medical consultations (20.3%),while rehabilitation represented 11.2%.

TABLE 3. Descriptive analysis of hospital discharges due to road traffic injuries,Belize, 2007

Variable Frequency %

Individual characteristicsSex

Women 85 25.1Men 253 74.9

Age group (years)

014 69 20.41524 86 25.42549 142 42.05074 33 9.8 75 8 2.4

Civil statusCommon law 49 14.5Divorced 0 0.0Married 60 17.8Single 199 58.9Unknown 28 8.3Widowed 2 0.6

SchoolingNone 74 21.9Primary 145 42.9Secondary 53 15.7Tertiary 12 3.6Unknown 54 16.0

Activity/occupationWage earner 128 37.9Nonremunerative activity 119 35.2Unknown 91 26.9

Medical attention and occurrenceAdmittance

Emergency 281 83.1Outpatient 57 16.9

District of residenceBelize 112 33.1Cayo 93 27.5Corozal 35 10.4Orange Walk 20 5.9Stann Creek 44 13.0Toledo 34 10.1

District of dischargeBelize 215 63.6Cayo 53 15.7Corozal 12 3.6Orange Walk 23 6.8Stann Creek 14 4.1Toledo 21 6.2

HospitalBelmopan Hospital 40 11.8 (1.6)a

Corozal Town Hospital 12 3.6 (1.1)a

KHMH Consolidation 215 63.6 (2.5)a

Orange Walk Hospital 23 6.8 (0.9)a

Punta Gorda Hospital 21 6.2 (1.5)a

San Ignacio Town Hospital 13 3.8 (1.0)a

Southern Regional Hospital 14 4.1 (0.5)a

Description of serviceAccident and emergency 1 0.3

General medicine 32 9.5Gynecology 1 0.3In-patient services 64 18.9Pediatric 42 12.4Surgical services 177 52.4

Discharge conditionAlive 302 89.3Dead 8 2.4Self-discharge 9 2.7Transferred 18 5.3Missingvalues 1 0.3

Note:KHMH: Karl Heusner Memorial Hospital.a Numbers in parentheses indicate road traffic injuries as percent of total.



7/11



From this figure, the potential underesti-mation of hospitalization costs is evident,as drugs and diagnostics are not includedin unitary cost estimations used for thisanalysis.

Indirect cost. A total of 2 501 years of po-

tential life were lost in Belize during2007 due to RTI (Table 5). This numbertranslates to a social loss of US$8 116 917attributed to indirect cost due to produc-tivity loss (Table 6). Table 6 disaggre-gates indirect cost by district for bothmethods and considers different scenar-ios in terms of discount rates. Belize dis-trict is where more indirect cost was lost:53.3% of the total or US$4 328 717. In ad-dition, injured people who were hospi-

TABLE 4. Emergency ambulance services due to road traffic injuries, Belize district, 2007

Type of injury

Slight Serious Fatal Total

Variable No. % No. % No. % No. %

SexWomen 41 28.3 8 17.0 0 0.0 49 24.7Men 102 70.3 37 78.7 3 50.0 142 71.7

Missingvalue 2 1.4 2 4.3 3 50.0 7 3.5Age (years)

Mean 29.6 NA 28.4 NA (. . .) (. . .) NA NA014 9 6.2 6 12.8 (. . .) (. . .) 15 7.81524 45 31.0 12 25.5 (. . .) (. . .) 57 29.72549 66 45.5 20 42.6 (. . .) (. . .) 86 44.85074 12 8.3 3 6.4 (. . .) (. . .) 15 7.8 75 1 0.7 0 0.0 (. . .) (. . .) 1 0.5Missingvalue 12 8.3 6 12.8 (. . .) (. . .) 18 9.4

DayMonday 10 6.9 3 6.4 1 16.7 14 7.1Tuesday 8 5.5 2 4.3 1 16.7 11 5.6Wednesday 19 13.1 5 10.6 0 0.0 24 12.1Thursday 23 15.9 6 12.8 0 0.0 29 14.6Friday 20 13.8 7 14.9 2 33.3 29 14.6Saturday 32 22.1 17 36.2 2 33.3 51 25.8Sunday 33 22.8 7 14.9 0 0.0 40 20.2

MonthJanuary 6 4.1 4 8.5 1 16.7 11 5.6February 13 9.0 5 10.6 2 33.3 20 10.1March 13 9.0 1 2.1 0 0.0 14 7.1April 7 4.8 7 14.9 0 0.0 14 7.1May 28 19.3 8 17.0 2 33.3 38 19.2June 7 4.8 1 2.1 0 0.0 8 4.0July 24 16.6 5 10.6 0 0.0 29 14.6August 10 6.9 6 12.8 0 0.0 16 8.1September 9 6.2 2 4.3 0 0.0 11 5.6October 9 6.2 4 8.5 0 0.0 13 6.6November 6 4.1 1 2.1 0 0.0 7 3.5December 13 9.0 3 6.4 1 16.7 17 8.6

Type of service providedTreated on site only 8 5.5 0 0.0 0 0.0 8 4.0Treated and transport 135 93.1 44 93.6 0 0.0 179 90.4Air transportation 1 0.7 0 0.0 0 0.0 1 0.5

No treatment 1 0.7 3 6.4 6 100.0 10 5.1Total 145 73.2 47 23.7 6 3.0 198 89

No service provided NA NA NA NA NA NA 24 11Grand total NA NA NA NA NA NA 222 100

Source:Belize Emergency Response Team administrative records.Note:NA: not applicable, (. . .): information not available.

FIGURE 2. Health system costs estimated for Belize, by type of service, 2007

Prehospital care 36.8%

Rehabilitation 11.2%

Hospitalization21.5%

Ambulatory medicalconsultations20.3%

Emergency roomservices10.3%

Note:Cost of hospitalization does not include drugs and diagnostic tests, which could represent a largeproportion of total hospitalization costs in this type of patient.


8/11


talized stayed an average of 5 days in thehospital. This value represents an aver-age of US$76 of indirect cost per RTI hos-

pitalization and a total of US$25 456 lostfor this cause.

Total economic cost of RTI. Aggregatedfigures are presented in Table 7. As indi-cated, the base model estimated a totaleconomic cost of US$9 453 615 due to RTIduring 2007. The great majority of thecost is due to fatal injuries, specifically in-direct costs attributed to prematuredeath. When the distribution of total eco-nomic cost estimated in all 10 000 itera-tions of the simulated model was ana-lyzed, 90.0% of all estimates fell between

US$10 005 971 and US$12 273 030 (aver-age of US$11 062 544 using a discountrate of 3%). This estimation represents0.9% of the Belize GDP during 2007 and2.8% of the total government budget forthe same year (Table 8). Table 9 shows the

base model results compared with the

sensitivity analysis figures (simulation).According to the simulation, during 2007the total nonfatal RTI rate was 1 468.7 per

100 000 population. Finally, informationprovided by the Social Security Boardshowed that at least two people were dis-abled in 2007 as a consequence of RTI (0.7per 100 000 population).

DISCUSSION

The usefulness of cost-of-illness stud-ies for health systems lies in justifyingthe design and implementation of speci-fied intervention programs and makingevident the economic loss that the illnesspresents, with its specific and current

mortality and morbidity rates. In thesame way, these studies help the re-source allocation process by contributingto determination of the importance ofeach illness from an economic perspec-tive and highlighting medical care andspecific research needs. In addition, they

provide a basis for planning and estab-lishing political and public health initia-tives for prevention and control and aneconomic-referenced framework for thenext evaluation of programs and imple-mented interventions (23).

Information generated in this studywill allow all districts and national au-

thorities in Belize to have valuable infor-mation for decision making. In general,this study will aid decision making, jus-tifying the need for implementing ameaningful road traffic safety program.It also allows for understanding the realmagnitude of a problem that so farseems reflected only in causes of mor-tality and social security payments.Therefore, the economic assessment of aproblem of this extent will allow visual-ization of the potential resource savingsthat can be applied to other social andhealth problems. In this sense, the results

make evident the large problem that RTIcause to the health system in Belize andsociety as a whole. From the results ofeconomic cost estimations, there is anurgent need to prevent RTI through astrategic and multisectoral approachthat focuses on the principal problemsdetected.

This study faced methodologic chal-lenges due to the lack of specific infor-mation, such as costs and number ofnonhospitalized injured people. Al-though secondary cost information usedmight underestimate the real problem, it

gives a general idea of the magnitude ofthe problem in economic terms. Futureefforts should consider the use of pri-mary data to evaluate to what extentthese estimates change. In the same way,problems in current information systemswere evident. That was the case in the

TABLE 5. Deaths and years of potential life lost by district, Belize, 2007

Characteristic District of residence Number

Total deaths Belize (district) 32Cayo 8Corozal 5Orange Walk 8Stann Creek 3Toledo 7

Belize (country) 63Years of potential life losta Belize (district) 1 380

Cayo 371Corozal 214Orange Walk 238Stann Creek 105Toledo 194Belize (country) 2 501b

a PAHO life expectancy in Belize for 2008age at death for people with less than life expectancy.b May not sum to total due to rounding.

TABLE 6. Indirect costs of premature death (US dollars) by district, Belize, 2007

Indirect cost District of residence Discount rate 0% Discount rate 3% Discount rate 5%

Minimum wage Belize (district) 4 844 151 2 637 812 1 828 566Cayo 1 301 508 704 967 487 627

Corozal 749 385 403 516 281 063Orange Walk 834 678 570 191 425 693Stann Creek 369 603 231 689 168 315Toledo 679 536 398 073 281 701Belize (country) 8 778 861 4 946 247 3 472 963

Average wage Belize (district) 7 949 376 4 328 717 3 178 414Cayo 2 135 808 1 156 869 846 064Corozal 1 229 760 662 179 489 891Orange Walk 1 369 728 935 697 744 428Stann Creek 606 528 380 208 293 404Toledo 1 115 136 653 247 496 671Belize (country) 14 406 336 8 116 917 6 048 871

Note:Exchange rate is US$1.00 = BZ$2.00.



9/11



codification of external causes of death,as most deaths and hospitalizations werecoded as other and unspecified. Thiscoding problem did not allow estimationof cost per type of road user or analysisof the medical attention characteristics ofdifferent road users. Improved codingefforts should be attempted in the future

in order to better appreciate the type ofproblem Belize faces in terms of roadsafety. This information would be in-valuable for policy making, as most pre-vention strategies differ depending onthe type of road user who is most af-fected. In addition, no information wasavailable on the number of emergencyroom hospital-based services providedas well as the number of ambulatory ser-vices provided (including rehabilita-tion). In this sense, it is important to con-sider that better information systemswould translate to better estimates of the

economic cost of this important publichealth problem.

The methodology used to estimate in-direct costs could overestimate actualfigures because it gives more weight todeaths that occurred at an early stage oflife, according to some perspectives suchas the investmentproducerconsumermodel perspective. This conceptualframework, originally proposed byGardner and Sanborn (24), argues thatduring the investment period (less than18 years of age), people only receivefrom society (negative value) while dur-

TABLE 7. Economic impact of road traffic injuries by severity and cost type,Belize, 2007

Direct cost Indirect costa

Severity (US$) (US$) Total (US$)

Fatal 3 886 8 795 959 8 799 845NonfatalSevere 76 400 488 083 564 483Slight 83 218 6 070 89 288

Total 163 503 9 290 112 9 453 615

a Indirect costs with discount rate of 3% under base model assumptions.

TABLE 8. Economic cost of road traffic injuries as percentage of gross domesticproduct and government budget, Belize, 2007

Characteristic Amount

Total GDP (US$) 1 267 000 000Total government budget (US$) 396 657 177Total MOH budget (US$) 43 213 108MOH budget as percentage of government budget 10.9MOH budget as percentage of GDP 3.4Total cost of RTI (US$)

0% discount rate 17 995 4153% discount rate 11 062 5445% discount rate 8 599 470

Cost of RTI as percentage of GDP0% discount rate 1.43% discount rate 0.95% discount rate 0.7

Cost of RTI as percentage of government budget0% discount rate 4.53% discount rate 2.85% discount rate 2.2

Cost of RTI as percentage of MOH budget0% discount rate 41.63% discount rate 25.65% discount rate 19.9

Note:US$1.00 = BZ$2.00. GDP: gross domestic product, MOH: Ministry of Health, RTI: road traffic injuries.

TABLE 9. Economic cost of road traffic injuries estimated for Belize, 2007

Total cost (USdollars)

Basea Average of simulationa with Total injured

Injury DR 0% DR 3% DR 5% DR 0% DR 3% DR 5% Base Simulation

FatalHospitalized 1 988 122 1 153 106 856 446 2 160 922 1 253 330 930 885 8Not hospitalized 13 190 201 7 646 739 5 677 287 14 336 643 8 311 364 6 170 735 53Total 15 178 323 8 799 845 6 533 733 16 497 565 9 564 694 7 101 620 61 66

NonfatalDied > 30 days 3 970 3 970 3 970 4 320 4 320 4 320 2Hospitalized

Hospitalization 517 685 517 685 812 772 812 772 812 772 338 338 531Ambulatory users 16 003 16 003 25 125 25 125 25 125 261 261Rehabilitation users 8 672 8 672 13 615 13 615 13 615 212 212

ER users 17 161 17 161 117 918 117 918 117 918 471 471 3 239ER servicesAmbulatory users 14 506 14 506 99 674 99 674 99 674 237 237Rehabilitation users 8 255 8 255 56 719 56 719 56 719 202 202

Total (hospital + ER) 642 908 642 908 1 446 170 1 446 170 1 446 170 854Nonusers of health services NA NA NA NA NA 83

Total (fatal + nonfatal) 15 832 094 9 453 615 7 187 504 17 995 415 11 062 544 8 599 470 998 4 531

Note:Both models (base and simulation) use average income reported by Social Security Board to estimate indirect costs (BZ$960, US$480 per month). DR: discount rate, ER: emergencyroom, NA: not applicable.a Simulation figures consist on average of the 10 000 iterations for different distributions of uncertain parameters (carried out with @RISK software).


10/11


ing the production period they give backto society (positive value). However, thisframework should also take into accountcountry preferences on how society val-ues life at different age periods. In addi-tion, some authors believe there are costshouseholds would have incurred with-out suffering RTI that could be evaluated

only by using counterfactuals (controlgroup scenario), which was not consid-ered in this study (25).

WHO estimates that RTI cost repre-sents 1.0% of the GDP in low-incomecountries, while in medium- and high-income countries the cost can reach 1.5%and 2.0% of GDP, respectively (19). Thisstudy documents that the total economiccost of RTI in Belize accounts for 0.9% ofthe GDP in the study year. In this regard,Mohan argued that the method usedmay influence the final estimation (20).When he analyzed the differences be-

tween low- and high-income countries,he found that estimates in high-incomecountries tended to be more detailed andcomprehensive, since they included will-ingness to pay, quality-adjusted lifeyears, and healthy life years (20). Ac-cording to this author, if the willingness-to-pay method had been used in India,the total cost of RTI would have in-creased from 0.8% to 2.0% of GDP. Healso highlighted some problems, such aslack of access to health services and tech-nology, which, together with few workopportunities for people with disabili-

ties, contributed to underestimation ofthe RTI cost in low- and medium-incomecountries. Epidemiologic informationfrom countries with better registersshows how estimates from differentcountries would also increase to 2.0% ofGDP (20). That might be the case in theBelize estimates.

CONCLUSIONS

According to this study, it was possi-ble to identify the high cost that RTI rep-resents to Belize society in general. Dur-

ing 2007, the economic cost of RTI in Be-lize was estimated to be US$11 062 544(90% confidence interval US$10 005 971to US$12 273 030), from which the greatmajority corresponds to the indirect costof premature death. This value repre-sents almost 1.0% of the total GDP of Be-lize for the same year and is equivalent

to 2.8% of the total government and25.6% of the Ministry of Health budgets.These figures make it possible to visual-ize the potential financial resources thatcan be used for RTI prevention and othersocial and health conditions if currentepidemiologic figures decrease. Thestudy identified that the main victimsare men and youth; however, there is agap of information about the type ofroad user injured or dead. Current fatal,nonfatal, and disability rate figures (20.1,1 468.7, and 0.7 per 100 000 population,respectively) highlight the need to im-

plement effective preventive strategiesin the short term to alleviate the current

burden of RTI in the country. Informa-tion generated in this study therefore al-lows for greater understanding of thereal magnitude of RTI, including eco-nomic impact.

Recommendations

The considerable economic costs ofRTI and deaths in Belize require specialattention to apply cost-effective strate-gies and law enforcement in order to

provide road safety for all road users.Improving the health information sys-tem to capture important data on RTI isessential. In this regard, unification datafrom different sources (health, police,and other) are essential to have a betteridea of the magnitude of the problem.Belize should work on a single RTI data-

base common for the entire country,with the type of road user coded prop-erly. Epidemiologic information such asthe number of injured people attendedin emergency rooms and hospital-basedfacilities as well as ambulatory follow-

up patients as a result of RTI should beavailable and quantified. This informa-tion will allow for greater in-depthanalysis of RTI in the country.

Acknowledgments. Technical and fi-nancial support was provided by the

WHO/PAHO Country Office. This eco-nomic impact study was commissioned

by the Belize National Road Safety Com-mittee in support of activities outlinedin the National Strategic Plan to ReduceRoad Traffic Injuries in Belize. Thepreparation of this study would not have

been possible without the collaborationand support of numerous agencies andindividuals. Special thanks to PeterAllen, chief executive officer of the Min-istry of Health; Michael Pitts, director ofhealth services; and the Office of the Di-rector of Health Services for facilitating

and providing support throughout thedata collection process and finalizationof the study. Special thanks to the PAHOCountry Office for technical and ad-ministrative support in making thisstudy a reality. Thanks also go to themembers of the National Road SafetyCommittee. The following people wereexceptional in facilitating the processand contributed significantly to the col-lection of data to support the research:Yvette Burks (BERT), Jorge Polanco(Ministry of Health), Aisha Andrewin(Ministry of Health), assistant superin-

tendent of police Francis Williams (Traf-fic Department, Police), assistant su-perintendent of police Simeon Avila(Traffic Department, Police), deputycommissioner of police James Magde-lano (Traffic Department, Police), Vladi-mir Romero (Traffic Department, Po-lice), and Bruce Flowers (Belize SocialSecurity Board). In addition, the authorsacknowledge the time of personnel ofKHMH from the Records Departmentand the medical doctors who respondedto the health provider survey in order toestimate ambulatory care.

1. Ministry of Health. Belize basic indicators 2005,1st ed. Belmopan, Belize: MOH; 2006. P. 12.

2. Organizacin Panamericana de la Salud. In-forme sobre el estado de la seguridad vial enla regin de las Amricas. Washington, DC:OPS; 2009.

3. Hijar M, Vazquez-Vela E, Arreola-Risa C.Pedestrian traffic injuries in Mexico: a country

update. Inj Control Saf Promot. 2003;10(12):3743.

4. Hjar-Medina M, Vazquez-Vela E. Foro na-cional sobre accidentes de trnsito en Mxico.Enfrentando los retos a travs de una visinintersectorial, 1st ed. Cuernavaca, Morelos,Mxico: Instituto Nacional de Salud Pblica;2003.

5. Naumann RB, Dellinger AM, Zaloshnja E,Lawrence BA, Miller TR. Incidence and to-tal lifetime costs of motor vehicle-relatedfatal and nonfatal injury by road user type,United States, 2005. Traffic Inj Prev. 2010;11(4):35360.

6. Impactos sociais e econmicos dos accidentesde trnsito nas rodovias brasileiras-relatrio

REFERENCES



11/11



executivo. Brasilia: IPEA/DENATRAN/ANTP;2006.

7. Organizacin Panamericana de la Salud.Clasificacin estadstica internacional de en-fermedades y problemas relacionados con lasalud, 10th ed. Washington, DC: OPS; 1995.

8. Transport Research Laboratory. Guidelinesfor estimating the cost of road crashes in de-veloping countries. Project R7780. London:Department for International Development;

2003.9. World Health Organization. Global status re-

port on road safety: time for action. Geneva:WHO; 2009.

10. Adam T, Evans DB, Murray CJ. Econometricestimation of country-specific hospital costs.Cost Eff Resour Alloc. 2003;1(1):3.

11. Organizacin Panamericana de la Salud. In-formacin de la salud y anlisis: situacin desalud en las Amricas: indicadores bsicos2008. Washington, DC: OPS; 2008.

12. Organizacin Panamericana de la Salud. Tc-nicas para la medicin del impacto de la mor-

talidad: aos potenciales de vida perdidos.Bol Epidemiol. 2003;24(2):14.

13. Stata Corporation. Stata 9.2. College Station,TX: Stata Press; 2003.

14. Albert T, Eden C. The economic burden of un-intentional injury in Alberta. Edmonton, Al-

berta, Canada: Alberta Centre for Injury Con-trol and Research; 2002.

15. Al-Masaeid HR, Al-Mashakbeh AA, QudahAM. Economic costs of traffic accidents in Jor-

dan. Accid Anal Prev. 1999;31(4):34757.16. National Institute of Mental Health and

Neuro Sciences. Road traffic injury (factsheet). Bengaluru, India: National Institute ofMental Health and Neuro Sciences; 2008.

17. Institute for Road Safety Research. Road crashcost. Leidschendam, Netherlands: Institutefor Road Safety Research; 2007.

18. Ministry of Transport. The social cost of roadcrashes and injuries. Wellington, New Zea-land: Ministry of Transport; 2006.

19. Peden MM. World report on road traffic in-jury prevention. Geneva: WHO; 2004.

20. Mohan D. Social cost of road traffic crashes inIndia. New Delhi: Indian Institute of Technol-ogy; 2002.

21. National Safety Council. Injury facts. Itasca,IL: National Safety Council; 2007.

22. Palisade C. @RISK, version 5.0. Ithaca, NY:Palisade; 2009.

23. Rice DP. Cost of illness studies: what is goodabout them? Inj Prev. 2000;6(3):1779.

24. Gardner JW, Sanborn JS. Years of potential

life lost (YPLL)what does it measure?Epidemiology. 1990;1(4):3229.

25. World Health Organization. WHO guide toidentifying the economic consequences of dis-ease and injury, 1st ed. Geneva: WHO; 2009.

Manuscript received on 12 July 2010. Revised version ac-cepted for publication on 19 October 2010.

Objetivo. Calcular el costo econmico de los traumatismos por accidentes de trn-sito registrados en Belice durante el ao 2007.Mtodos. Se realiz un estudio transversal a partir de datos secundarios sobre loscostos, tanto desde la perspectiva social como desde la del sistema de salud. La infor-macin epidemiolgica se obtuvo a partir de la base de datos de mortalidad, la basede datos nacional de egresos hospitalarios y los expedientes administrativos de la po-lica y el Ministerio de Salud. Se llev a cabo una encuesta a los prestadores de servi-cios de salud para calcular las cifras correspondientes a la atencin ambulatoria pos-terior al egreso. Para calcular los costos directos, se utiliz la base de datos delproyecto WHO-CHOICE (eleccin de intervenciones eficaces en funcin de los costos)de la Organizacin Mundial de la Salud. El equipo de respuesta a las urgencias m-dicas de Belice aport los datos sobre los costos prehospitalarios. Despus de calcularlos aos de vida potencial perdidos tomando como parmetro la esperanza de vidade Belice correspondiente al ao 2008 y empleando el mtodo propuesto por la Orga-nizacin Panamericana de la Salud, se calcul el costo indirecto asociado a la muerteprematura desde el enfoque del capital humano. Se utiliz un modelo de rbol de de-cisiones para calcular el costo econmico total derivado de los traumatismos causadospor el trnsito y se hizo un anlisis de sensibilidad multivariado y probabilstico paraincorporar los parmetros de incertidumbre en las estimaciones.Resultados. En Belice, durante el ao 2007, los traumatismos causados por el trn-sito provocaron la muerte de 61 personas, la hospitalizacin de 338 y, segn se cal-cula, lesiones menores a 565. Se perdieron 2 501 aos de vida potencial a causa de lasmuertes prematuras, lo que se tradujo en un costo econmico total de US$11 062 544.Esta cifra representa 0,9% del producto interno bruto de Belice. Se calcul que el costodirecto fue de US$ 163 503, del cual 2,4% fue ocasionado por las muertes, 46,7% por

la atencin de las personas que sufrieron traumatismos graves y 50,9% por la atencinde quienes presentaron lesiones menores.Conclusiones. El costo econmico calculado en este estudio pone de manifiesto lanecesidad de prevenir los traumatismos causados por el trnsito adoptando un m-todo estratgico y multisectorial que se centre en abordar los principales problemasdetectados.

Accidentes de trnsito; costos de la atencin en salud; costos y anlisis de costo; Belice.

RESUMEN

Repercusiones econmicasde los traumatismos mortalesy no mortales por accidentesde trnsito en Belice en 2007

Palabras clave

Documents

Costos SV Belize