Table 6.

Other Physical Intervention (Type D).

Authors	Study		N, Response Rate & Method	Exposure Levels	Outcome Measure(s)	Did Outcome Change with ‘Intervention’? Yes/No Strength of Effect	Comments	Confounders Adjusted for in Analyses
	Nature	Design
de Kluizenaar et al. (2013) [35]	Questionnaire survey	Cross- sectional Stratified sample on age and district	1967 50% RR	For each dwelling, exposure levels were calculated at the most and least exposed façade (Lden,most and Lden,least, respectively). 40–70 dB	Annoyance ISO scale (0–10 point verbal)	Yes Stronger association between noise and annoyance for those: who have relative quiet available (>10 dB difference between most and least exposed façades). Beta = 0.099 SE = 0.012, p ≤ 0.0001, who have higher noise level at the least exposed façade. Beta = 0.035, SE = 0.016, p ≤ 0.05) No interaction was confirmed		Age, gender, education, and annoyance from neighbour noise and ‘humming’ noise
Babisch et al. (2012) [36]	(HYENA) study is a large-scale multi-centered study carried out simultaneously in 6 European countries Prevalence of (designed as a hypertension study with air and road traffic sources) Study examined many modifiers. Here only the result wrt quiet side and living room facing the street are reported	Cross-sectional in stratified random samples around 6 airports (but response to road traffic noise examined here)	4861 (45–70 years old) 30–78% RR	LAeq24 h 45–65 road traffic noise	Annoyance ISO scale (0–10 point verbal)	Yes Location of the bedroom resulted in decreased annoyance at night (Beta = 1.25, CI = 1.12–1.38 vs. Beta = 0.81, CI = 0.65–0.97; interaction p < 0.001). per 10 dB Those with location of the living room facing the street were more annoyed during daytime with increasing road traffic noise level (Beta = 1.63, CI = 1.50–1.76) than those whose living room was located on the back side (Beta = 1.44, CI = 1.18–1.69); interaction p = 0.007	Samples based on air traffic noise but models adjusted for this	Full models, both continuous noise levels (Air and Road), type of housing, location of rooms shielding due to obstacles, visibility of the postal street, window opening habits, type of windows length of residence, time spent in the living room on workdays, time spent in the bedroom on workdays noise reducing remedies, building modifications to reduce noise, self reported hearing problems, rooms per occupant
van Renteghem & Botteldooren (2012) [37]	Belgium Effect of presence of a quiet façade on annoyance in high noise exposure dwellings	Comparison: of responses in dwellings with and without a quiet side All dwellings had noisy side: half also had a quiet side	100 Response rate 70% Interviews	Lden. 65–75 at most exposed façade—all dwellings. Half of dwellings also had quiet side Both levels sourced from END maps	ISO scale (5 point verbal) Analysis used mid category cut-off ‘at least moderately annoyed’	Yes Absence of quiet façade results in increased ‘at least moderately annoyed’ respondents: Odds ratio 3.3 when adjusted for noise sensitivity (95% CI 1.35–8.01) When people actually used the bedroom at the quiet side OR = 10.6.(95% CI 2.0–56)	Quiet side defined as a front/back façade level difference >10 dB	Noise sensitivity, window closing, bedroom on a quiet side,, front-façade Lden
de Kluizenaar et al. (2011) [38]	Questionnaire survey	Data drawn from a prospective cohort study For a postal questionnaire survey	18 973 (15–74 years) 70% RR	For each dwelling, exposure levels were calculated at the most and least exposed façade (Lden,most and Lden,least, respectively). 40–70 dBA (Estimates available for N = 17,650)	Total Annoyance Dichotomous scale	Yes Stronger association between noise and annoyance for those who have relative quiet available (>10 dB difference between most and least exposed façade) for all levels >45 dB Ors range: 1.33–6.54 (all significant) Interaction term significant for two noise categories: OR = 3.177 for Lden interval 57.5–62.5; OR = 5.584 for Lden >60		Age, sex, body mass index, exercise, marital status, work situation, financial difficulties, alcohol use, education
Gidlöf-Gunnarson & Öhrström (2010) [30,39]	Sweden Effect of appearance of quiet side courtyard on annoyance in dwelling with high noise exposure	Comparison: of responses in dwellings with and without an attractive courtyard All dwellings had noisy side and a quiet side	385 Response rate 59% Mail survey	LAeq,24 h Calculated levels Noisy façade in two categories: 58–62 dB (n = 241) and 63–68 dB (n = 144). All had access to a ‘quiet side’ 239 Ps had low quality courtyard, 146 had high quality courtyard	ISO (5-point verbal) scale Analysis used mid category cut-off at least moderately annoyed Percentage of noise annoyed residents was significantly lower across the two sound level categories among those who had high (16% and 29%) than low-quality quiet courtyards (27% and 42%)	Yes Percentage annoyed depended on noisy façade exposure level, but was less when quality of courtyard was high, rather than low Odds Ratio for courtyard quality was 0.59 (95% CI: 0.36–0.96)	Quiet side defined as LAeq,24 h < 48 including façade reflection Quality of courtyard was assessed objectively.	Type of housing; Lay out and population characteristics: were comparable in the two study groups
Gidlöf-Gunnarsson, & Öhrström (2007) [40]	Sweden Nearby green area	Green versus non green Quiet site available versus not available All areas above 60 dB Most aspects kept constant at similar noise exposures, road traffic dominating source	500 Response Rate 59% Interviews	>60 dB	ISO scale (0–10)	Yes Significant associations emerged for availability to green areas (p < 0.001) and for access to a quiet side (p = 0.001), However, the effect sizes were low (partial η2 = 0.029 and 0.023, respectively)	Interaction quiet side and green space not tested	Age