Table 2

 Effectiveness of the programs and scientific rigour of the evaluations

ProgramKey findings regarding the program effectivenessStudy design of the program evaluation
Category A: High degree of effectiveness
LidköpingAverage annual decline in injury rates from 1983 to 1991 of 2.4% for boys and 2.1% for girlsComparison with: (1) four neighbouring municipalities (total population of intervention and comparison areas roughly similar); (2) the county in which Lidköping is situated
MotalaDecline in injury rates by 13% for healthcare treated injuries, decline by 15% for hospital treated injuries, decline by 41% for non-trivial healthcare treated injuriesComparison with: (1) neighbouring municipality (slightly smaller than Motala), matched on demographic and socioeconomic variables; (2) the nation
FalunDecline in injury rates by 24% for outpatient injuries, decline by 46% for workplace injuries, and decline by 43% for school injuries. There was still a reduction in injuries after seven years, but it was possible to surmise a waning effect during the last two years of the program while still in progressComparison with: (1) neighbouring municipalities (roughly twice the size of Falun); (2) the county in which Falun is situated; (3) the nation. Also comparison of data on non-targeted and targeted injuries for intervention and comparison areas
HarstadDecline in injury rates by 53% for burns in children, decline by 27% for traffic injuries, decline by 26% for fall fractures among elderly, decline by 15% for skiing injuriesComparison with considerably larger and geographically distant city (Trondheim, 6.2 times larger than Harstad). Also comparison of data on non-targeted and targeted injuries for intervention area
Category B: Modest degree of effectiveness
Falköping1979–82: Decline in injury rates by 27% for home injuries, by 28% for occupational injuries, by 28% for traffic injuries. 1983–91: Outpatient injury rate levelled off and the inpatient rate showed an average annual increase of 8.7% for females and 4.9% for males. For injuries seen in either ambulatory or emergency department settings, it seems that the effect of the early phase of the program was a lasting one, but for injuries admitted to hospital the effect was temporaryComparison with: (1) similar sized neighbouring municipality (1979–82 years); (2) the county in which Falköping is situated (entire period); (3) the nation (entire period). Intervention and comparison areas matched on many variables, including demographic and socioeconomic variables, and injury incidence
SCIPPDistinct reduction in motor vehicle occupant injuries among children ages 0–5 years. No evidence that the program reduced other target injury categories. Intervention households that reported exposure to the program had higher safety knowledge and behaviour scores than comparison householdsComparison with five cities and towns of the state (total population of intervention and comparison areas roughly similar), matched on many variables, including age composition, education level, family income, housing characteristics
Safe BlockDistinct difference between intervention and control homes with respect to safety knowledge and home hazards that required minimal to moderate effort to correctComparison with geographically separate census tracts selected from the same area of the intervention area. Intervention and comparison areas intended to be alike
Safe KidsDecline in injury rates among school aged children. The decline was specified to the targeted age group and targeted causes. A non-specific decline also occurred in the comparison areaComparison with a neighbouring health district. Did not strive for matched comparison area
WaitakereSignificant reductions in child injury rates. Significantly more Waitakere residents were aware of injury prevention safety messages and had acquired appropriate child safety items post-intervention than pre-interventionComparison with similar sized area, matched on demographic variables, new housing developments, road safety, and crime prevention coordinator positions in both areas
Category C: Minor or no degree of effectiveness
NaxosThe program had “only modest success”. The results “cannot be considered as entirely satisfactory”Comparison with neighbouring island (Naxos is an island), matched on demographic variables and considered to be equally prosperous
Safe LivingNo significant changes were found in rates of injury deaths, hospitalisations, or emergency department presentations. Increase in program awareness was moderate and “similar to other community based programs”. The program was “unable to replicate the significant reductions in injuries reported in other community based interventions”Comparison with similar sized area, matched on demographic and socioeconomic variables
Five-cityStatistically significant decline in injury rates for 4 injury categories, non-significant decline for 13 categories, unchanged for 3 categories, statistically significant increase for 1 category, non-significant increase for 5 categories (a total of 27 injury categories were targeted)Comparison among the five cities of the project
Category D: Evaluated without comparison area
VaeröyDecline in injury rates by 29%No comparison made
CICPIHSDecline in injury rates observed primarily for fall, motor vehicle, and assault injuries. Reductions for falls were associated with specific activities, which was not the case with motor vehicle injuries or assaultsNo comparison area selected. Comparison of data on non-targeted and targeted injuries for intervention area
Latrobe ValleyDecline in injury rates for home injuries. Increases in home safety knowledgeNo comparison area selected. Comparison of data on non-targeted and targeted injuries for intervention area
IllawaraDecline in injury rates by 17% for child injuries and by 14% for accident related hospital admissions of childrenNo comparison made