In the United States, the allowable limit for sound in the vicinity of an airport is 65 decibels (dB) averaged over a 24-hour ‘day and night’ period (DNL) [1
]. This level is roughly 10 times the sound intensity (measured in power) of a 55 dB day–evening–night level (Lden
), the threshold in the European Union. The actual difference in noise in the U.S. can be higher because dBs are measured on a logarithmic scale, and DNL levels are weighted to increase the impact of nighttime noise. This high level of sound exposure can affect sleep, well-being, school performance, and economic productivity as well as mental and physical health [2
]. It has also been shown in observational and quasi-experimental analyses to produce an increase in risk of cardiovascular disease (CVD) and mental illness [7
Airports tend to have sound contour patterns that correspond to aircraft approach, departure, size, and geographic characteristics [2
]. Under Part 150 of the Federal Aviation Regulations, 65 dB DNL is considered “significant”, and houses that fall within noise contours that exceed allowable limits must receive remediation by the government.
A very large study of aircraft noise corridors in the United Kingdom showed that 55 dB is the threshold value at which mental health problems emerge [12
]. Similar findings in Europe led to the EU’s Environmental Noise Directive [8
], which limits aircraft noise corridors to 55 dB Lden
Aircraft noise abatement measures are in effect around the world [15
], and can reduce the noise impact on population [16
]. The International Civil Aviation Organization (ICAO) established a “balanced approach” to aircraft noise management, including four elements: “reduction of noise at its source; land use planning and management; noise abatement operational procedures; and operating restrictions on aircraft” [17
]. The characteristics of a given airport, such as the number of runways, the number of aircraft movements, the type of aircraft, the population of the city it serves, the per capita gross domestic product of the country an airport is located in, can inform the ideal abatement strategy [18
]. Options for addressing noise at its source (e.g., altering the runway characteristics, changing flight patterns, reducing aircraft size, requiring aircraft to glide, and altering climbs on departure) are both more effective and generally less expensive than land use planning and management (e.g., land acquisition and sound insulation); however, the latter are likely inevitable in most localities in the United States if more stringent standards are to be adopted [17
]. This is because many of the traditional means of reducing noise at its source have already been implemented, and more recent innovations that consider both airport and aircraft characteristics are challenging to implement due to conflicting air traffic at adjacent urban airports [21
Still, land use changes carry their own challenges. While aircraft noise reduces the price of property by approximately 20% on average [22
], buying properties via eminent domain is prohibitively expensive in most urban areas in the United States. Because further source mitigation is unlikely to reduce 65 dB DNL contours to 55 dB DNL, most localities would likely resort to home insulation against sound as a strategy for reducing noise exposure among individuals.
The much weaker regulatory standard in the United States means that authorities or municipalities can maintain or expand airports in the United States more cheaply than can European governments. However, this weaker standard may also harm the health of those who live near airports. This paper examines the trade-off between the direct cost of changing the regulatory DNL level from 65 dB to 55 dB versus the medical costs, loss of health, and loss of lives associated with failing to do so. We examine sound insulation as a noise abatement strategy because it is more expensive and less effective than most noise source abatement strategies—if sound insulation is cost-effective, then it is not necessary to evaluate noise source abatement strategies because they are intuitively cost-effective.
The main results of the cost-effectiveness analysis are presented in Table 2
. Over the course of one’s life, installing sound insulation is associated with approximately $
6793 in increased costs to society and an increase in 0.61 QALYs gained. The resulting incremental cost effectiveness ratio (ICER) of the intervention relative to status quo was, therefore, $
lists the effects of a series of one-way sensitivity analyses on ICERs. The most sensitive input parameter in the model was the relative risk of anxiety for noise exposure above 55 dB DNL. If it was assumed that there was no increased risk of anxiety due to aircraft noise, the ICER would increase to $
195,717/QALY gained. Among costs parameters, the intervention cost was the most influential. Increasing the intervention cost by 25% increased the ICER to $
24,448/QALY gained from its base case value (i.e., $
6793/QALY gained). Additionally, when the probability of developing CVD was increased by 25%, the ICER dropped to $
7064/QALY. Other model parameters did not show any significant effect.
Results indicate a 95% credible interval of the ICER ranging from dominance to as high as $
93,054/QALY gained. The incremental cost-effectiveness plane is shown in Figure 3
. Based on this figure, installing sound insulation would be cost-effective in 91% of simulations at a WTP value of $
The dots represent incremental cost-effectiveness pairs of Monte Carlo simulations for changing the regulatory DNL level from 65 dB to 55 dB versus the status quo for 10,000 Monte Carlo simulations. The diagonal line represents a WTP of $50,000 per QALY gained. The dots to the right of the diagonal line, of which the proportion is 91%, represent the simulations with an incremental cost effectiveness ratio less than $50,000 per QALY gained.
In the United States, the Environmental Protection Agency’s Office of Noise Abatement and Control was shuttered in the 1980s [42
]. This office had successfully restricted noise at one airport in the name of public health [42
], but failed to have a national impact on standards. As a result of generally weaker regulatory standards for noise, the burden of regulation has fallen on the Federal Aviation Administration (FAA), an agency focused on regulating lives in the air rather than on the ground.
However, while between 600 and 1300 people die in aviation accidents worldwide [43
], most deaths probably occur on the ground from day-to-day airport operations. A risk ratio of 1.12 for CVD amounts to 45 excess deaths per 100,000 people exposed to DNL > 55 dB [29
]. It is likely that many millions are exposed to airplane noise that exceeds these levels worldwide, mostly in developing nations and in the United States.
This study illustrates the risks associated with a failure to rationally allocate resources to maximize lives saved costs the lives of tens of thousands of Americans each year [44
]. Changing the standard for DNL from 65 dB to 55 dB in the United States also comes at a cost that is one to two orders of magnitude lower than the cost of most FAA regulations geared toward saving lives in the air [45
]. The air marshal program, for example, costs $
180 million per life saved [47
]. It is also more cost effective than the vast majority of medical and health interventions that experts agree are worthwhile investments. For instance, screening and treating high-risk populations for HIV comes in at over four times the cost of the lowering the DNL standard in the United States [48
]. The Monte Carlo simulations indicate that 91% of the permutations of the model would cost under $
50,000/QALY gained, which is similar in cost to HIV screening and treatment in high-risk populations.
The current study was subject to a number of limitations. The relative risk estimates of CVD and anxiety for aircraft noise exposure were only based upon observational studies. This is because existing experimental and quasi-experimental studies generally include heart rate and a psychological rating scale as outcome measures, both of which are not suitable with the current model [10
]. However, these risk estimates are in line with the observational data.
A second potential weakness of the study is the use of a case study neighborhood. While necessary to generate estimates for the purposes of illustration, the use of a case study limits the generalizability of our findings. However, the East Elmhurst neighborhood consists of relatively densely packed private houses with poor insulation and relatively low DNL exposures within sound corridors [24
]. Areas with high-rise buildings or neighborhoods near airports that support larger, long-distance aircraft could prove challenging to insulate to ≤55 dB DNL. The United States also regulates housing more stringently and has far higher health costs than less wealthy nations, further reducing the generalizability of our findings to other development contexts.
Third, the CVD noted in this study is likely to arise from anxiety associated with airplane nose. Still, we treated anxiety and CVD as independent events. We chose to do so because the nature of each condition is quite different, and would likely lead to separate, independent diagnoses in very different settings. Anxiety can influence work productivity and incur outpatient costs. CVD likely influences inpatient costs to a much greater extent. However, it may be that patients with co-morbid anxiety and CVD patients incur higher than average CVD costs (e.g., by consuming more hospital time).
Finally, it should be noted that sound insulation is an imperfect remedy. While it reduces noise in the home, outdoor activities, such as gardening, are made considerably less pleasant. Moreover, people are likely less willing to open the windows of their homes, also influencing the quality of their lives. These costs were not considered in the analysis.
Many strategies exist for abating aircraft noise, including reducing noise at the source (e.g., changes to flight patterns) and reducing noise exposure for those on the ground (e.g., insulation of homes). Of these strategies, insulating homes is likely to be both more expensive and less effective than source strategies. Nevertheless, we find that lowering the allowable limit of aircraft noise from 65 dB to 55 dB DNL would be cost-effective, even if it requires insulating homes near airports. This is because the cost of inaction is very high, taking a toll on the mental and physical health of those living near airports. Indeed, even when this strategy is adopted, changing the regulatory standard for noise exposure around airports comes at a very reasonable cost given the significant benefits for health.
It is important for agencies like the FAA to set more rigorous standards that take into account the significant health impacts for those on the ground.