3.1. Results
3.1.1. Collisions, Fatalities, and Injuries
Local roads make up about 52% (18,945) of Massachusetts’ 36,247 miles of roads (MassDOT 2012).
Table 1 shows crash data for Massachusetts. From 2006-2009, there were 477,103 total collisions reported by the RMV. Twenty-nine percent of these collisions (136,262) occurred on “functionally classified local roads” that would have been impacted by the Speed Limit Bill. Of total collisions, 23% (314) of fatal injuries and 26% (44,580) of nonfatal injuries occurred on these local roads from 2006–2009. Approximately 37% of crashes involving pedestrians and crashes involving cyclists occurred on local roads (
Table 2 and
Table 3).
Table 1.
Motor vehicle crashes from 2006–2009 reported by the registry of motor vehicles.
Table 1.
Motor vehicle crashes from 2006–2009 reported by the registry of motor vehicles.
Crash Type | 2006 | 2007 | 2008 | 2009 | Total | Annual Average |
---|
Crashes on All Roads | 124,274 | 120,667 | 120,970 | 111,192 | 477,103 | 119,276 |
Crashes on Local Roads | 34,832 | 34,953 | 34,319 | 32,158 | 136,262 | 34,066 |
Fatal Crashes on All Roads | 384 | 400 | 321 | 292 | 1397 | 349 |
Fatal Crashes on Local Roads | 97 | 93 | 69 | 55 | 314 | 79 |
Fatalities on All Roads | 410 | 426 | 345 | 316 | 1,497 | 374 |
Fatalities on Local Roads | 101 | 100 | 71 | 59 | 331 | 83 |
Injury Crashes on All Roads | 33,038 | 31,289 | 31,017 | 28,933 | 124,277 | 31,069 |
Injury Crashes on Local Roads | 8543 | 8196 | 7845 | 7560 | 32,144 | 8,036 |
Injuries on All Roads | 45,934 | 42,947 | 42,321 | 40,077 | 171,279 | 42,820 |
Injuries on Local Roads | 11,991 | 11,334 | 10,741 | 10,514 | 44,580 | 11,145 |
Table 2.
Pedestrian crashes from 2006–2009 reported by the registry of motor vehicles.
Table 2.
Pedestrian crashes from 2006–2009 reported by the registry of motor vehicles.
Crash Type | 2006 | 2007 | 2008 | 2009 | Total | Annual Average |
---|
Crashes on All Roads | 124,274 | 120,667 | 120,970 | 111,192 | 477,103 | 119,276 |
Crashes on Local Roads | 34,832 | 34,953 | 34,319 | 32,158 | 136,262 | 34,066 |
Crashes Involving Pedestrians on All Roads | 1577 | 1584 | 1765 | 1671 | 6597 | 1649 |
Crashes Involving Pedestrians on Local Roads | 576 | 585 | 672 | 615 | 2,448 | 612 |
Pedestrian Fatalities on All Roads | 58 | 65 | 72 | 46 | 241 | 60 |
Pedestrian Fatalities on Local Roads | 19 | 15 | 28 | 13 | 75 | 19 |
Pedestrian Injuries on All Roads | 1116 | 1192 | 1316 | 1331 | 4955 | 1239 |
Pedestrian Injuries on Local Roads | 408 | 437 | 478 | 479 | 1802 | 451 |
Table 3.
Cyclist crashes from 2006-2009 reported by the registry of motor vehicles.
Table 3.
Cyclist crashes from 2006-2009 reported by the registry of motor vehicles.
Crash Type | 2006 | 2007 | 2008 | 2009 | Total | Annual Average |
---|
Crashes on All Roads | 124,274 | 120,667 | 120,970 | 111,192 | 477,103 | 119,276 |
Crashes on Local Roads | 34,832 | 34,953 | 34,319 | 32,158 | 136,262 | 34,066 |
Crashes Involving Cyclists on All Roads | 1069 | 1069 | 1227 | 1248 | 4613 | 1153 |
Crashes Involving Cyclists on Local Roads | 398 | 393 | 458 | 455 | 1704 | 426 |
Cyclist Fatalities on All Roads | 6 | 11 | 10 | 6 | 33 | 8 |
Cyclist Fatalities on Local Roads | 1 | 8 | 4 | 1 | 14 | 4 |
Cyclist Injuries on All Roads | 753 | 744 | 866 | 882 | 3245 | 811 |
Cyclist Injuries on Local Roads | 277 | 281 | 309 | 313 | 1180 | 295 |
The second scenario conservatively assumed that serious and fatal crashes are just as likely to take place in congestion as they are in free flow traffic conditions and therefore modeled changes due to the 24 h average speed reduction, which accounted for congestion. Estimates that assume traffic on local roads would slow by 1.8 mph, on average, in response to a 5 mph lower speed limit, show that the Speed Limit Bill could have prevented roughly 2200 crashes, 18 fatalities, and 1200 injuries per year across the Commonwealth.
Table 4 shows expected annual decreases in crashes, injuries, and fatalities under the two speed reduction modeling assumptions. The first modeling assumption scenario assumed that crashes serious enough to cause injury or fatality, or damage to any one vehicle or other personal property in excess of $1000 were unlikely to take place in congested conditions, and therefore modeled the impact of the bill without accounting for congestion.
Table 4.
Estimated annual reductions in crashes and fatalities (power model results).
Table 4.
Estimated annual reductions in crashes and fatalities (power model results).
Estimated Annual Decrease in | 1.8 mph Speed Reduction Estimate (95% Confidence Interval) | 0.67 mph Speed Reduction Estimate (95% Confidence Interval) |
---|
Total Crashes | 2219 (286, 4042) | 811 (102, 1505) |
Fatal Crashes | 15 (2, 27) | 6 (1, 11) |
Injury Crashes | 772 (460, 1072) | 285 (168, 401) |
Fatalities | 18 (−4, 35) | 7 (−1, 15) |
Injured Road Users | 1239 (369, 2039) | 460 (133, 77) |
Pedestrian Fatalities | 4 (−1, 8) | 2 (0, 3) |
Cyclist Fatalities | 1 (0, 1) | 0.3 (−0.1, 0.6) |
Injured Pedestrians | 50 (15, 82) | 19 (5, 31) |
Injured Cyclists | 33 (10, 54) | 12 (4, 21) |
Estimates that consider congestion would slow traffic by 0.67 mph on average and could prevent roughly 810 crashes, seven fatalities, and 460 injuries per year across the Commonwealth. It should be noted that the figures for fatalities were not statistically significant, although point estimates suggest a protective effect of lower speeds. There is consistent evidence that reducing traffic speeds decreases the frequency and severity of crashes. Despite the predicted magnitude of these benefits, there are a number of limitations to this analysis. The Power Model, while parsimonious, is fairly crude. Like any model, it makes assumptions and is not completely precise and accurate. Additionally, while speeds decrease and safety increases on local roads, the CTPS model demonstrates that traffic volume may increase on highways and arterials. It should be noted, however, that average speeds on highways and arterials will decrease slightly due to increased congestion, potentially increasing safety on these roads as well. Finally, our estimates are based on imperfect data. As stated earlier, crash data is consistently underreported. As such, the estimates presented here are conservative, and more accurate data would reveal greater decreases in crashes, injuries, and fatalities.
3.1.2. Cost of Collisions
Recent CDC estimates show that the cost of death from motor vehicle crashes in Massachusetts was $394 million in 2005 (Centers for Disease Control and Prevention 2011). Work loss costs made up $388 million of these costs, while medical costs made up $6 million.
Table 5 and
Table 6 show the estimates of how much preventing injuries and fatalities based on a 1.8 mph and 0.67 mph decrease in traffic speeds (from the Power Model) could save in terms of both work loss and medical costs. Under both sets of assumptions, a speed limit reduction of 5 mph would result in savings. Estimated savings ranged from $11–$30 million in prevented fatalities and between $67–$180 million in prevented injuries. These savings would affect those involved in collisions and their families, as well as employers, property owners, and taxpayers across the state. Limitations to this analysis include the lack of data on costs for collisions that did not result in an injury or fatality. Including these personal damage costs would increase cost savings estimates. This analysis assumes that all injuries prevented by the modeled reduced speeds would have otherwise resulted in a hospital visit. This assumption is based on the fact that our baseline data came from the RMV CDS, which only registers serious crashes.
Table 5.
Estimated costs savings for a 1.8 mph decrease in traffic speeds based on CDC’s WISQARS in 2012 dollars.
Table 5.
Estimated costs savings for a 1.8 mph decrease in traffic speeds based on CDC’s WISQARS in 2012 dollars.
Outcome | Fatalities | Pedestrian Fatalities | Cyclist Fatalities |
Annual Decrease in Deaths | 18 | 4 | 1 |
Medical Cost Avoided | $346,721 | $76,699 | $18,912 |
Work Loss Cost Avoided | $29,347,334 | $6,521,513 | $1,630,641 |
Combined Cost Savings | $29,694,055 | $6,598,212 | $1,649,553 |
Outcome | Injured Road Users | Injured Pedestrians | Injured Cyclists |
Annual Decrease in Number Hospitalized | 1239 | 50 | 33 |
Medical Cost Avoided | $63,872,373 | $2,703,376 | $1,652,705 |
Work Loss Cost Avoided | $116,610,789 | $5,164,047 | $3,766,654 |
Combined Cost Savings | $180,483,163 | $7,867,423 | $5,419,359 |
Table 6.
Estimated costs savings for a 0.67 mph decrease in traffic speeds based on CDC’s WISQARS in 2012 dollars.
Table 6.
Estimated costs savings for a 0.67 mph decrease in traffic speeds based on CDC’s WISQARS in 2012 dollars.
Outcome | Fatalities | Pedestrian Fatalities | Cyclist Fatalities |
Annual Decrease in Deaths | 7 | 2 | 0 |
Medical Cost Avoided | $133,435 | $37,824 | $0 |
Work Loss Cost Avoided | $10,990,016 | $3,140,455 | $0 |
Combined Cost Savings | $11,123,451 | $3,178,279 | $0 |
Outcome | Injured Road Users | Injured Pedestrians | Injured Cyclists |
Annual Decrease in Number Hospitalized | 460 | 19 | 12 |
Medical Cost Avoided | $23,713,638 | $1,027,556 | $600,984 |
Work Loss Cost Avoided | $43,293,937 | $1,962,653 | $1,370,075 |
Combined Cost Savings | $67,007,575 | $2,990,209 | $1,971,058 |
3.1.3. Time Spent and Fuel Consumed in Traffic
Base year conditions from the CTPS model indicated that the total daily VMT for the state were 155.1 million, of which 26.6 million were on local roads, making up 17% of total daily VMT. The total daily VHT were 4.6 million. The CTPS model showed that under the 1.8 mph reduction in traffic speeds, daily VMT on total roads would increase by 184,000, while daily VHT on total roads would increase by 19,000. Daily VMT overall would increase as drivers choose new, less direct routes to avoid slower traffic on local roads. Daily VMT on local roads would decrease by 355,676, while daily VHT on local roads would increase by 2860 as a result of the bill.
Applying the TTI equations to these data, we found that fuel costs would increase by $21 million per year and the increased time spent in traffic would cost Massachusetts drivers $127 million in lost time. CTPS also modeled whether participants would shift from commuting by automobile to biking, walking, or public transit as a result of the speed reductions, and found that there would be no appreciable mode shifts.
While speed limit reductions would reduce crashes and prevent injuries and fatalities, they would also prompt drivers to reduce cut-through traffic by seeking faster, though often longer distance, routes on higher capacity roads, resulting in an additional 55.3 million vehicle miles travelled per year. At the same time, slower travel speeds on local roads and higher traffic volumes on newly preferred, higher capacity roads would result in 5.8 million additional VHT per year. These increases in time spent in traffic would cost approximately $127 million per year, while increases in fuel consumed in traffic would cost $21 million per year.
3.1.4. Air Pollution
In general, most monitored air pollutants in the state of Massachusetts are at levels below health-based standards, and levels have been declining over time [
51]. Estimates of changes in air pollutants under both speed reduction scenarios would be minor, but concentrations would lead to slight increases in risk for the citizens of Massachusetts. A speed limit reduction was expected to increase traffic congestion and related air pollution emissions, prompting us to investigate the potentially harmful health effects of additional air pollution associated with the Speed Limit Bill. Although higher concentrations of air pollutants would contribute to additional deaths and hospitalizations due to asthma, chronic lung disease, heart attacks, ischemic heart disease, and major cardiovascular disease, models showed that modeled increases associated with the bill would be negligible. The total annual expected number of increased illnesses due to the increase in air pollution was lower than 0.001 cases per year for all outcomes. Air pollution-related health costs would be approximately $474 per year.
Final estimates do not include the effects of exposure to other pollutants that may change as an impact of the bill, including SO2, CO, ozone, and ultrafine particles. We relied upon air pollution estimates from CTPS that use the EPA’s MOBILE6.2 model, which does not incorporate additional emissions that would occur due to stop-and-go traffic. Additionally, we were not able to calculate effects of air pollution on stroke, premature birth, infant mortality, and childhood asthma. These factors would contribute additional mortality and hospitalizations not calculated here. These aggregated numbers do not demonstrate the distribution of risk among different populations. Finally, our estimates also do not include increased exposures specific to commuters, who may spend more time in traffic in close proximity to elevated concentrations.
3.1.5. Pedestrian and Bicyclist Perceptions of Safety
According to the 2010 Census, nearly 200,000 (6%) Massachusetts workers bike or walk to work, however about 20% of Massachusetts residents report engaging in no leisure time physical activity [
35]. Although Massachusetts is considered one of the healthiest states in the country, 60% of adults are overweight and 24% of adults are obese, highlighting the significance of interventions that help residents become more active [
36].
While traffic calming strategies are primarily promoted as a way to reduce crashes, injuries and deaths, they may also be a feasible method to promote physical activity by helping create an environment that encourages active transportation. The World Health Organization has suggested that traffic may have a strong negative impact on health by reducing the ability to engage in active transportation [
52]. One pathway that the negative impact of traffic may have on physical activity is through the perception of safety. Studies that consider traffic and perceptions of safety generally agree that pedestrians and bicyclists have negative perceptions of traffic and that real and/or perceived danger and discomfort in traffic discourages walking and bicycling [
41,
42,
53,
54,
55,
56,
57,
58]. Safety concerns appear to be strongest in children, the elderly and women, thus contributing to health inequalities for these groups [
56,
59].
Evidence generally supports the positive impact of traffic calming overall on perceptions of safety and active transportation. Speed limit reductions would likely create more conducive conditions for active transportation by improving the objective safety of roads for all users.
3.1.6. Parental Safety Perceptions and Children’s Levels of Physical Activity
In the United States, almost half of elementary and middle school students walked or biked to school in 1969, whereas less than 15% walk or bike to school today [
42]. Almost 17% of children aged 2 to 5 and 11% of middle school students are overweight in Massachusetts [
60].
The weight of the evidence reviewed indicates that higher traffic speed/density is associated with lower levels of physical activity among youth [
61]. Conversely, classic traffic calming measures, such as controlled intersections, and supportive infrastructure, such as sidewalks, were associated with higher levels of physical activity. Additionally, Morrison and colleagues [
41] report an increase in parents’ willingness to allow children to walk and ride bicycles after the implementation of a traffic calming scheme. Objective measures of traffic indicated that safer pedestrian environments (e.g., slower speeds and lower traffic volume) predicted higher levels of physical activity. Carver and colleagues note that traffic-calming measures, quiet local streets with a speed limit of 50 km/h or less (about 31 mph), and higher street connectivity had the most positive impact on physical activity behaviors and active transportation [
62]. It is likely that the Speed Limit Bill would have supported more physical activity among children across the Commonwealth; however, accompanying bicycle/pedestrian facilities and other self-enforcing engineering interventions would maximize perceived safety.
3.1.7. Property Values
American Community Survey Data from the US Census Bureau show that the average Census tract median home value for the state of Massachusetts for 2006–2010 was $374,499 [
50]. According to a recent analysis of tax data conducted by the Boston Globe, statewide home values have more than doubled since 2000; however, home values hit an all-time high in 2007 and have dropped since the 2008 recession.
We find a consistent relationship between lower traffic speeds and higher property values [
45,
46,
47,
48,
49]. The literature, however, is sparse and could not be reliably extrapolated to assess the likely impact of the Speed Limit Bill on property values in Massachusetts, especially because all properties on local roads would be affected simultaneously. However, the literature indicates general preferences for the safety and quiet associated with slower speeds, suggesting that residents would likely enjoy quality of life benefits even if they were not monetized into higher home values.
3.1.8. Results Summary
Based on a literature review, case studies, and statistical models, the HIA predicted that lowering speed limits on local roads would have had a positive public health impact across Massachusetts, particularly by preventing traffic fatalities and injuries. Potential co-benefits included enhanced walking and biking environments that would encourage physical activity, as well as increased desirability of properties on local roads due to quieter and safer streets. The HIA also concluded that the bill is economical. The Speed Limit Bill would have prevented 2219 crashes per year, 18 fatalities per year, and 1239 injuries per year, which translates into a savings of up to $210 million annually in prevented medical payments and work lost. These economic benefits outweighed the costs of increased time spent in traffic and fuel burned estimated in our models, as well as the health impacts of the small increase in air pollution, associated with the proposed change. In addition, lower speed limits would have likely improved children’s and adults’ perceptions of road safety, which could have led to increased pedestrian and bicyclist physical activity and a resulting reduction in chronic disease risk. Although the evidence was limited, we predicted that Massachusetts residents living on local roads would have experienced increased satisfaction with their neighborhoods as a result of the proposed bill even if these benefits were not monetized.
3.2. Recommendations
Our primary recommendation was that the legislature should adopt a bill to lower speed limits on local roads. To maximize the effectiveness of a speed limit reduction, we suggest combining a change in the law with the following evidence-based recommendations.
3.2.1. Further reductions in Motor Vehicle Speeds
Further measures to decrease traffic speeds in conjunction with lower speed limits include traffic calming, enforcement, and education. To explore the potential impact of further reductions in traffic speeds to the posted 25 mph limit, we modeled the impact of a full 5 mph decrease on crashes and fatalities on functionally classified local roads.
Our analysis suggests that a 5 mph reduction in traffic speeds would confer three times the reduction in crashes and twice the cost savings of a 1.8 mph reduction. Most importantly, the 5 mph decrease would save more than twice the number of lives compared to a 1.8 mph reduction (
Table 7).
Table 7.
Crashes and Cost of Crashes in 2012 dollars.
Table 7.
Crashes and Cost of Crashes in 2012 dollars.
Estimated Annual Decrease in: | 1.8 mph Speed Reduction | 5 mph Speed Reduction |
---|
Total Crashes | 2219 (95% CI 286, 4042) | 6265 (95% CI 855, 10,794) |
Fatalities | 18 (95% CI −4, 35) | 44 (95% CI −11, 67) |
Injured Road Users | 1239 (95% CI 369, 2039) | 3336 (95% CI 1077, 5088) |
Medical and Work Lost Cost of Fatalities | $29,694,055 | $72,586,636 |
Medical and Work Lost Cost of Hospitalizations | $180,483,163 | $485,949,909 |
Given these findings, measures to help align true traffic speeds with the regulated speed would have significant health and economic benefits. In conjunction with a speed limit reduction, traffic-calming design solutions, plus enforcement and education, may help maximize health benefits associated with reduced traffic speeds.
3.2.2. Implementation—Dissemination
We recommended that the Speed Limit Bill should be accompanied by a public information campaign to help the public understand the bill’s costs and benefits. The campaign could include a media component, inclusion in the driver’s education curriculum, inclusion in RMV mailings or other documents regularly distributed to drivers.
3.2.3. Implementation—Enforcement
Enforcement policies and policing would help reduce actual traffic speeds closer to the 25 mph limit. Speed cameras may prevent speeding and crashes [
63], and enforcement approaches that remind drivers that roads are patrolled for speeding may also help raise compliance rates with lower speeds.
3.2.4. Implementation—Traffic Calming
To maximize the health benefits of speed reductions, passive and self-enforcing engineering interventions are most effective [
64]. New local roads designed to support a lower speed limit would be most beneficial to health. If the road design speed differs from the speed limit on existing roads, traffic calming measures could help reduce travel speeds without intensive enforcement. Traffic calming is an engineering strategy that slows traffic and reduces traffic volume [
65]. Studies show that traffic calming measures can reduce road traffic injuries by roughly 15% in the areas that received design and engineering interventions [
66].
Measures that change the height of the road surface appear to be among the most effective in reducing speeds and preventing injuries and fatalities [
64]. “Vertical deflections”, such as raised pedestrian crossings, speed humps, and cushions, alter the road surface height and force drivers to slow. Less effective approaches narrow roadways and/or create “horizontal deflections” (e.g., pinch-points, bump-outs, roundabouts, islands, chicanes,) that force vehicles to veer, and therefore slow.
Municipalities should also consider implementing traffic calming interventions that serve pedestrians and cyclists, including raised crosswalks, reducing motor vehicle lane width to serve bicycles, and signalization to accommodate active road users [
57,
67,
68].
3.3. Dissemination and Impact Evaluation
In January 2013, MAPC, in collaboration with WalkAmerica, WalkBoston, MDPH, and MassDOT, convened stakeholders and traffic safety experts for a “Slow Down Summit”. The Summit allowed attendees to participate in discussions about strategies to slow traffic down and improve bicyclist and pedestrian safety. The theme was safe speeds for vibrant communities, and was standing room only. The summit allowed us to disseminate our preliminary findings, receive feedback, and engage policymakers on the topic of speed limits and their health impacts.
The HIA was released via press release on 12 September 2013. There was a delay in the release of the HIA because the Speed Limit Bill did not pass during the 2011–2012 legislative session. The bill was re-filed in the next session, however the new bill did not define roads by functional classification. Therefore, the HIA did not apply to the new bill as filed. However, the HIA aided in providing a balanced quantitative and qualitative analysis to show that reducing vehicle speeds on local roads has a net benefit for health and costs to society (
Figure 2), and may inform future legislative efforts to improve traffic safety in Massachusetts.
Although the bill did not pass, the HIA process brought health more prominently into what had traditionally been viewed as a transportation discussion. Its quantitative and qualitative approach may be a useful model for transportation and public health professionals seeking to analyze how driving speeds impact road safety and other health risk factors. The HIA demonstrates an approach for assessing the health impacts of speed-related policy changes that can be easily borrowed by other HIA practitioners and public agencies. Finally, the HIA serves to increase awareness of the multiple benefits of slower traffic speeds.
Figure 2.
Infographic of estimated total savings and costs under the Speed Limit Bill.
Figure 2.
Infographic of estimated total savings and costs under the Speed Limit Bill.