Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Sep 9.
Published in final edited form as: Sleep Health. 2026 Aug 24;12(5):800–810. doi: 10.1016/j.sleh.2026.07.010

Daily Associations between the Physical Environment and Sleep Among a Diverse, National Sample of Young Adults

Adwoa Dadzie a, David A Reichenberger b, Yuqi Shen a, Lindsay Master a, Angel IM Collie f, Gina Marie Mathew c, Timothy R Brick d,e, Lauren Hale c, Anne-Marie Chang a, Orfeu M Buxton a
PMCID: PMC13552506  NIHMSID: NIHMS2199857  PMID: 42637624

Abstract

Objectives

We examined daily between- and within-person associations of the physical sleep environment with actigraphic and self-reported sleep.

Methods

Young adults in the Future of Families “Young Adult Sleep Study” (n=372; 60%F) self-reported daily for 2 weeks whether three sleep environment factors (i.e., temperature, light, and noise) disrupted the previous night’s sleep while wearing wrist-actigraphs. Multilevel models examined between- and within-person associations between environmental factors and actigraphic sleep timing, duration, wake after sleep onset (WASO), sleep maintenance efficiency (SMEff), and self-reported sleep quality.

Results

Environmental factors were associated with sleep health in between-person and/or within-person findings. Between-person: A “too hot” environment was associated with greater WASO (b=9.93 mins, 95%CI [3.64,16.21], p=.006) and lower SMEff (b=−1.78%, 95%CI [−3.01, −0.54], p=.010). “Too noisy inside” was associated with lower SMEff (b=−2.97%, 95%CI [−5.24, −0.70]), p=.03) and 88% decreased odds of good sleep quality (OR=0.12, 95%CI [0.02, 0.54], p=.006). A “too bright” environment was associated with 90% decreased odds of good sleep quality (OR=0.10, 95%CI [0.01, 0.61], p = .010) and later sleep onset (b=111.05 mins, 95%CI [26.67, 195.42], p=.010), midpoint (b=122.95 mins, 95%CI [41.22, 204.69], p=.010), and offset (b=131.54 mins, 95%CI [39.7, 223.40], p=.010).

Within-person: A “too hot” environment was associated with lower SMEff (b=−0.79%, 95%CI [−1.13, −0.45], p<.001) and 45% decreased odds of good sleep quality (OR=0.55, 95%CI [0.39, 0.79], p=.001) that same night.

Conclusions

Characteristics of the physical sleeping environment (temperature, brightness, and noisiness) were associated with later sleep timing and worse sleep quality but not sleep duration. Future research should examine mechanisms by which environmental factors may affect sleep health.

Keywords: environment, sleep timing, sleep duration, sleep quality, young adults, actigraphy

I. Introduction

Environmental factors, including temperature, light, and noise, are widely believed to affect multiple dimensions of sleep health. However, few studies use actigraphic sleep data within the same individual in their real-world context.

Environmental temperature is a predictor of sleep disturbance.1 An increase in environmental temperature at the beginning of the night impedes the decline in core body temperature, making it difficult to fall asleep.2 Furthermore, temperature fluctuations in either direction can influence thermoregulatory homeostasis during the night (especially during rapid eye movement sleep) when the body is less able to regulate temperature),3 potentially impairing sleep quality. Uncomfortable temperature was found to be the most common sleep disruption in a study that assessed the daily home sleeping environment in adolescents (15–17 yrs) using ecological momentary assessment (EMA) measures.4 Laboratory and home studies have consistently found that uncomfortable temperatures are associated with poor sleep among adults. Specifically, higher temperatures were associated with lower actigraphic sleep efficiency5 and increased wake after sleep onset (WASO)6 among adults with an average age of 35 years. Hot residential indoor temperatures are also associated with delayed self-reported sleep onset in mid-life adults.7 In a large global dataset, rising temperatures were associated with earlier objectively measured sleep onset, midpoint, and offset, and shorter sleep duration.8 Among young adults, higher overnight indoor temperatures were associated with decreased actigraphic total sleep time.9 Cooler temperatures have been associated with poorer self-reported sleep quality among middle aged adults10 and delayed polysomnography-measured sleep onset among young adults.11 These findings indicate that warmer or cooler temperatures might contribute to shorter sleep duration and poorer sleep quality.

Light is the strongest synchronizer of the circadian clock (i.e., zeitgeber) in humans12,13. Mistimed light exposure, such as throughout the evening or during the night, can delay circadian timing12,14, increase sleep onset latency among adults15, and increase problems with sleep.16 For example, light at night has been associated with shorter self-reported sleep duration in middle- to older-aged adults17. Among young adults, increased light within the bedroom (measured via light sensor) has been associated with lower actigraphic sleep efficiency and increased awakenings after sleep onset.18 Generally, increased exposure to outdoor light at night was associated with poorer perceived sleep quality among adults19. Taken together, exposure to light at night might contribute to delayed onset timing, shorter sleep duration, and poorer sleep quality.

A noisy environment may also negatively impact sleep quality20 by fragmenting sleep. Laboratory and naturalistic experiments have found that noise negatively affects sleep. Loud sounds were associated with greater polysomnography-measured number of arousals in a laboratory study.21 Another laboratory study found that the louder the presented hospital noise, the greater the probability of polysomnography-measured sleep disruption.22 Increased daily residential environmental noise was also associated with lower actigraphic sleep efficiency and more actigraphic WASO among adults23, and nighttime traffic noises were associated with poorer self-reported sleep quality.24 However, few studies examine daily, within-person associations between noise and actigraphic sleep quality, particularly among young adults.

Young adulthood is a period of development characterized by changes in living environments. Developmental transitions25,26, like attending college, moving back to parent’s home after college, and moving in with romantic partners, among others, may cause young adults to move27 multiple times during this period. These changes in young adults’ sleeping environments may be associated with poor sleep quality as their sleep habituates to these new environments.28 Young adults enrolled in college are likely to live in dorms and shared living spaces where factors such as lights and noise are more variable.29,30 Moreover, young adulthood is a critical developmental period for learning needed to facilitate the developmental tasks for transition to adulthood.31,32 Hence, sleep is crucial for facilitating this continued neurobehavioral functioning in young adulthood.33 Further, young adulthood is associated with an increase in the onset of mental illnesses34, which are also associated with poor sleep health35,36. Gaps exist in the examination of associations between the physical environment and sleep among young adults. Moreover, few studies consider the between- and within-person variation in associations between the physical environment and sleep health; hence, motivating our study to examine factors that lead to sleep disruption in young adults at the between- and within-person levels. Lastly, identifying the elements of the physical environment most associated with sleep health outcomes in young adults may aid in the design of future interventions that target improvement of specific elements of the sleeping environment (e.g., temperature or light) as an avenue to improve the sleep of young adults.

The objective of this study was to examine the between- and within-person associations of perceptions of factors related to the sleeping environment on actigraphic and self-reported sleep among young adults. For between-person associations, we hypothesized that young adults reporting their environments were too hot, too cold, too bright, and/or too noisy more often would, on average, have later actigraphic sleep timing (onset and midpoint), shorter actigraphic night sleep duration, and worse actigraphic (WASO, sleep maintenance efficiency [SMEff]) and worse self-reported sleep quality than young adults not reporting those environmental factors. For within-person associations, we hypothesized that on nights when young adults reported their environment was too hot or cold, too bright, and/or too noisy inside or outside, they would exhibit later actigraphic sleep timing (onset and midpoint), shorter sleep duration, and worse actigraphic (WASO, SMEff) and self-reported sleep quality that same night compared to nights when their environment was not too hot, too cold, too bright, or too noisy.

II. Methods

Participants

The Future of Families (previously called Fragile Families) and Child Wellbeing Study (FFCWS) was approved by the Institutional Review Boards at Princeton University, Stony Brook University (CORIHS B FWA #00000125), and Pennsylvania State University (STUDY00013519). The study design and initial procedures have been previously published.37 The FFCWS includes a longitudinal birth cohort from 20 large U.S. cities (https://ffcws.princeton.edu). The study oversampled nonmarried mothers, resulting in a higher percentage of racial and ethnic minoritized groups and those with low socioeconomic status compared to the general U.S. population. A total of 4,898 mothers gave initial consent for the study and were recruited from local hospitals at the time of the child’s birth.37 Data were collected when the youth were ages 0, 1, 3, 5, 9, 15, and 22 years. At age 15, a random sample (n=1,090) of the 3,444 adolescents who participated in the core FFCWS study at that wave were invited to participate in the sleep sub-study, and 1,049 participants assented. When participants were 22 years old, they were recruited to participate in the Young Adult Sleep Study (YASS). Of the 614 young adults who participated in YASS between January 2021 and May 2023, a total of 520 young adults provided diary data; 148 were excluded for not providing ≥3 days of valid actigraphic data and accompanying next-day diary, leading to the final analytic sample (n=372) (Figure 1).

Figure 1:

Figure 1:

Flowchart of percentages for study sample size.

Measures

Environmental predictors of sleep

Young adults completed a diary every morning for 14 days on the Wear-IT mobile app.38 Participants were asked whether it had been too hot, too cold, too bright, too noisy inside, or too noisy outside the previous night (no, 0 or yes,1).

Actigraphic sleep outcomes

Participants wore an actigraphy device (Actiwatch Spectrum Plus; Philips-Respironics, Murrysville, PA) on their non-dominant wrist for two weeks. Data were downloaded using the Philips Actiware software (versions 6.0.4–6.3). Sleep intervals were initially set using a validated scoring algorithm and scored using the physical activity and light monitors of the device to approximate sleep onset and offset of each night’s sleep period.39 At least two independent scorers determined the main sleep intervals, cut points, and validity of days by using a standardized operating procedure. Scorers adjudicated differences greater than 15 minutes.

Actigraphic sleep outcomes included the following: Sleep onset: scored start of the main sleep interval, defined as the longest sleep interval in a cutpoint day (in minutes from midnight). Sleep midpoint: the halfway point between onset and offset of the main sleep interval (in minutes from midnight). Sleep offset: the end of the main sleep interval (in minutes from midnight). Total sleep time: number of minutes of sleep in the main sleep interval (sleep onset to offset, excluding WASO; in minutes) was our measure of sleep duration. WASO: the number of scored minutes of wake during the sleep duration interval, with a higher number indicating worse sleep quality. SMEff: the percentage of time during the sleep duration interval that a person was asleep, calculated by subtracting WASO from sleep duration, dividing this number by sleep duration, then multiplying by 100, creating a percentage (%); higher numbers indicate better sleep quality.

Self-reported sleep quality

Each morning, participants were asked how they would rate their sleep quality the night before, with responses of 1 (Very good), 2 (Fairly Good), 3 (Fairly Bad), and 4 (Very Bad). We dichotomized sleep quality into two groups: Lower sleep quality (fairly or very bad; 15% of nights) and higher sleep quality (fairly or very good; 85% of nights).

Covariates

All models adjusted for sex, race, ethnicity, education, employment status (between-person level), and whether it was the weekend (within-person level). Sex (female or male) was parent-reported at birth. Young adults self-reported race (exclusive categories of Black, White, Asian, Native, or Other), and we created a multiracial category of individuals who selected more than one race. We combined individuals who were Asian, Native, Multiracial, and Other into the “Other” race category due to small sample sizes. Young adults self-reported ethnicity (Hispanic and/or Latino or non-Hispanic) and education level (categorized into less than high school, high school or equivalent, some college or technical school, and college or graduated). Employment status was categorized using the number of working hours per week reported by the young adult as zero-work hours reported (0 hours/week), part-time (<35 hours/week), or full-time (≥35 hours/week). Weekend night was coded as 0 if sleep offset occurred on Monday through Friday, or 1 if sleep offset occurred on Saturday or Sunday.

Statistical analysis

Each between-person environmental predictor was calculated as the mean of the predictor for each person. Because these variables were dichotomous (0 or 1), average between-person predictors ranged from 0–1, indicating the proportion of nights per person that a given environmental factor affected one’s sleep.

Each daily, within-person environmental predictor was calculated by subtracting the person-average predictor from each person’s daily value. This indicated how much the environmental factor varied from one’s own average on a given day (a positive number indicating the environmental factor was experienced that night, and a negative number indicating the environmental factor was not experienced that night).

Multilevel linear and logistic models (R studio [2026.03.12, 12, version 4.5.3]) were conducted for the current analyses with R packages for Linear Mixed Effects40 and Generalized Linear Mixed41 models. Models were first conducted with random intercepts with both between- and within-person predictors and covariates in the same model. A second model was then specified that included a random slope for the within-person predictor (allowing associations between daily predictors and outcomes to vary across participants). All models were specified with covariance structure of AR (1) due to the assumption that dates closer to each other would be more correlated than dates further apart. If the model with both random intercepts and random slopes converged, analysis of variance (ANOVA) was performed to determine which model fit best. If the model fits were not significantly different (i.e., p>0.05), the more parsimonious model with a random intercept and no random slope was reported. If the p-value of the model comparisons was significant, the model with both a random intercept and slope was reported. All models included the between- and within-person effects of environmental predictor variables in addition to fixed effects of covariates.

Multiple testing correction

We applied the Holm-Bonferroni correction42 to limit the family-wise Type-I error rate (FWER)43. More information on correction is reported in Supplementary pg.1. We report p values in non-adjusted models as praw and p values after Holm-Bonferroni correction as pHB.

III. Results

Demographics and descriptive statistics

The sample (n=372) was primarily female (60%), Black (51%), and non-Hispanic (74%). Other descriptive statistics and sociodemographic characteristics are reported in Table 1. There were minimal demographic differences between our analytic sample (n=372) and the excluded sample (n=505) (Supplementary Table 1). There was an average of (mean ± SD) 10.0±3.5 actigraphy nights per person (range 3–17).

Table 1:

Descriptive statistics of participant sociodemographic characteristics, environmental factors, and sleep measures (n=372)

n %
Birth sex
Female 225 60.5
Male 147 39.5
Race
Black 189 50.8
White 97 26.1
Othera 86 23.1
Ethnicity
Non-Hispanic 277 74.5
Hispanic 95 25.5
Education level
Less than high school 34 9.1
High school or equivalent 133 35.8
Some college or technical school 161 43.3
College or graduated 44 11.8
Employment status
Zero-work hours reported 128 34.4
Part-time (<35 hours /week) 103 27.7
Full-time (≥35 hours /week) 141 37.9
Environmental predictors (proportion of nights) M SD
Too hot 0.21 0.28
Too cold 0.11 0.22
Too bright 0.04 0.12
Too noisy inside 0.06 0.15
Too noisy outside 0.03 0.11
Sleep outcomes
Actigraphic sleep onset (clock time) 1:02 AM 1:45
Actigraphic sleep midpoint (clock time) 4:44 AM 1:40
Actigraphic sleep offset (clock time) 8:33 AM 1:50
Sleep duration (hours) 6.63 1.15
Actigraphic wake after sleep onset (minutes) 45.06 16.91
Actigraphic sleep maintenance efficiency (%) 89.99 3.34
Self-reported sleep quality (proportion of nights)
Good sleep quality 0.85 0.23
a

“Other” includes young adults who identified as Native, Asian, Multiracial and “Other race” (i.e., did not select any of the available race options). The sleep measures and environmental predictors are the average of the person means. Self-reported sleep quality is the average of person means of nights with good sleep quality.

The physical environment and sleep timing

Environment too hot

There were no significant associations between reporting the environment was too hot and sleep timing (Table 2).

Table 2:

The physical environment and sleep timing (n=372)

Nighttime Sleep Measure (Minutes)
Environment Predictor Sleep Onset Sleep Midpoint Sleep Offset
b 95% CI praw pHB b 95% CI praw pHB b 95% CI praw pHB
Too Hot
Between Person 17.10 −20.97, 55.18 .38 .48 26.40 −10.52, 63.33 .16 .48 27.38 −14.11, 68.86 .19 .48
Within Person 6.94 −4.66,18.54 .24 .72 2.69 −8.30, 13.68 .63 1.00 0.95 −12.12, 14.03 .89 1.00
Too Cold
Between Person −5.59 −54.29, 43.11 .82 1.00 −8.73 −56.04, 38.59 .72 1.00 −32.24 −85.29, −20.81 .23 .70
Within Person −16.87 −31.52, −2.22 .02* .07 −13.23 −27.08, 0.63 .06 .12 2.48 −13.98, −18.94 .77 .77
Too Bright
Between Person 111.05 26.67–195.42 .010* .010* 122.95 41.22, 204.69 .003* .010* 131.54 39.68, 223.40 .005* .010*
Within Person 6.03 −17.14, 29.20 .61 1.00 10.93 −10.89, 32.75 .3 .98 4.40 −21.46, 30.26 .74 1.00
Too Noisy Inside
Between Person 58.06 −11.11, 127.22 .10 .30 48.33 −18.89, 115.56 .16 .32 50.28 −25.21, 125.78 .19 .32
Within Person −4.98 −22.96, 13.00 .59 1.00 3.74 −13.28, 20.76 .70 1.00 15.93 −4.30, 36.15 .12 .37
Too Noisy Outside
Between Person −11.66 −103.86, 80.54 .80 1.00 −0.69 −90.30, 88.93 .10 1.00 −11.83 −112.44, 88.77 .82 1.00
Within Person −7.5 −31.34, 16.34 .54 1.00 2.06 −20.43, 24.55 .90 1.00 11.63 −15.05, 38.30 .39 1.00

b = non-standardized regression coefficient. CI = 95% Confidence intervals (lower, upper). Alpha-corrected p-value with Holm-Bonferroni42 adjustment denoted as pHB. Alpha level of significance was p<.05, two-tailed. Significant results are denoted with asterisk for both adjusted and unadjusted models. All models show associations with a random intercept but not a random slope. Adjusted for sex, race, ethnicity, education, employment status, and whether it was the weekend.

Environment too cold

Within-person associations revealed that on nights when young adults reported their environment was too cold, sleep onset was about 16 minutes earlier than their own usual. (b = −16.87, 95%CI [−31.52, −2.22], praw = .02). However, these results were not statistically significant after multiple testing correction (pHB = .07; Table 2). No other associations between being too cold and sleep timing reached statistical significance at the praw < .05 level.

Environment too bright

Between-person associations revealed that young adults who reported their environment was too bright on all nights had 111 minutes later sleep onset (b = 111.05, 95%CI [26.67, 195.42], praw = .010), 122 minutes later sleep midpoint (b = 122.95, 95%CI [41.22, 204.69], praw = .003), and 131 minutes later sleep offset (b = 131.54, 95%CI [39.68, 223.40], p raw = .005) compared to those who did not report their environment was too bright on any night (Figure 2). These associations remained statistically significant at the p<.05 level after multiple testing correction (pHB ≤ .010 for all, Table 2). There were no within-person associations between bright environment and sleep timing.

Figure 2:

Figure 2:

Between-person associations between environment “too bright” and A. sleep onset; B. sleep midpoint; C. sleep offset. Adjusted for sex, race, ethnicity, education, employment status, and if it was the weekend.

Environment too noisy

There were no statistically significant associations between reporting the environment was too noisy (inside or outside) and sleep timing (all praw-values >.05, Table 2).

The physical environment and sleep duration

There were no significant associations between the physical environment and sleep duration (all praw-values >=.05, Table 3).

Table 3:

The physical environment and sleep duration (n=372)

Sleep Duration (Minutes)
b 95% CI praw pHB
Too Hot
Between Person 9.68 −13.49, 32.85 .41 .41
Within Person −10.31 −20.74, 0.12 .05 .07
Too Cold
Between Person −11.08 −40.69, 18.52 .46 .50
Within Person 7.16 −6.02, 20.35 .29 .86
Too Bright
Between Person 8.65 −43.06, 60.37 .74 .74
Within Person 10.83 −10.03, 31.68 .31 .93
Too Noisy Inside
Between Person −27.70 −69.74, 14.35 .20 .20
Within Person 12.78 −3.89, 28.94 .12 .24
Too Noisy Outside
Between Person 17.32 −38.61, 73.24 .54 1.00
Within Person 14.70 −6.74, 36.15 .18 .54

b = non-standardized regression coefficient. CI = 95% Confidence intervals (lower, upper). Alpha-corrected p value with Holm-Bonferroni42 denoted as pHB. Alpha level of significance was p<.05, two-tailed. No significant associations were found between ambient environment and sleep duration. All models show associations with a random intercept but not a random slope. Adjusted for sex, race, ethnicity, education, employment status, and whether it was the weekend.

The physical environment and sleep quality (actigraphic and self-reported)

Environment too hot

Young adults who reported the environment was too hot more often had higher average WASO (b = 9.93, 95%CI [3.64, 16.21], praw = .002; Table 4) in between-person findings. Specifically, young adults who reported their environment was too hot on all nights had about 10 minutes more WASO compared to those who did not report their environments were too hot on any night (Figure 3).

Table 4:

The physical environment and sleep quality (n=372)

Wake After Sleep Onset (WASO) (Minutes) Sleep Maintenance Efficiency (SMEff) (%) Odds of Good Self-Reported Sleep Quality (Dich)
b 95% CI praw pHB b 95% CI praw pHB OR 95% CI P
Too Hot
Between Person 9.93 3.64, 16.21 .002* .006* −1.78 −3.01, −0.54 .005* .01* 0.51 0.21, 1.23 .13
 Within Person 2.25 0.17, 4.33 .03* .07 −0.79 −1.13, −0.45 <.001* <.001* 0.55 0.39, 0.79 .001*
Too Cold
Between Person 4.74 −3.36, 12.83 .25 .50 −1.21 −2.82, 0.39 .14 .41 0.21 0.07, 0.62 .004*
 Within Person 0.99 −2.20, 4.17 .54 1.00 −0.08 −0.49, 0.33 .71 1.00 1.03 0.66, 1.59 .90
Too Bright
Between Person 15.15 1.03, 29.28 .04* .11 −2.66 −5.46, 0.14 .06 .13 0.10 0.01, 0.61 .010*
 Within Person 1.19 −2.93, 5.31 .57 1.00 −0.17 −0.81, 0.47 .61 1.00 0.55 0.29,1.05 .070
Too Noisy Inside
Between Person 10.21 −1.32, 21.73 .08 .17 −2.97 −5.24, −0.70 .010* .03* 0.12 0.02, 0.54 .006*
 Within Person 3.92 0.70, 7.14 .02* .05 −0.35 −0.85, 0.16 .18 .24 0.73 0.44,1.20 .21
Too Noisy Outside
Between Person 0.02 −15.38, 15.42 .10 1.00 0.42 −2.63, 3.46 .79 1.00 0.18 0.02,1.47 .11
 Within Person 2.49 −1.76, 6.74 .25 .54 −0.30 −0.97, 0.36 .37 .54 0.87 0.44,1.73 .70

b = non-standardized regression coefficient. OR = odds ratio. CI = 95% Confidence intervals. Alpha-corrected p-value with Holm-Bonferroni42 denoted as pHB. Alpha level of significance was p<.05, two-tailed. Significant results are denoted with asterisk for both adjusted and unadjusted p-values. Too hot and SMEff and too cold and WASO associations are from models with both random intercept and slope specified. Remaining models include only a random intercept. Dich = dichotomous. Adjusted for sex, race, ethnicity, education, employment status, and whether it was the weekend. Self-reported sleep quality models did not adjust the p-values as only one model was run per predictor.

Figure 3:

Figure 3:

Between-person associations between environment “too hot” with A. wake after sleep onset (WASO) and b. sleep maintenance efficiency (SMEff); and C. within-person association between environment “too hot” and sleep maintenance efficiency (SMEff). Adjusted for sex, race, ethnicity, education, employment status, and whether it was the weekend.

There were between- (b = −1.78, 95% CI [−3.01, −0.54], praw <.010) and within- (b = −0.79, 95% CI [−1.13, −0.45], praw <.001) person associations between the environment being hot and lower SMEff (Table 4). Young adults who reported their environment was too hot on all nights had about 2% lower average SMEff compared to those who did not report their environment was too hot on any night in between-person findings, and young adults exhibited about 1% lower SMEff than their own usual on nights when they reported their environment was too hot in within-person findings (Figure 3).

All results with hot environment and actigraphic sleep quality remained statistically significant after multiple testing correction (pHB ≤. 010 for all).

There were within-person associations between reporting the environment was too hot and worse self-reported sleep quality (OR=0.55, 95%CI [0.39, 0.79], p=.001), such that on nights when young adults reported their environments were too hot, there was a 45% decreased odds of having good sleep quality compared to nights that were not too hot (Table 4).

Environment too cold

There were no statistically significant associations at the praw<.05 level between reporting the environment was too cold and actigraphic sleep quality.

There were between-person associations between the environment being reported as too cold and lower self-reported sleep quality (OR = 0.21, 95% CI [0.07, 0.62], p = .004). Specifically, young adults who reported their environment was too cold on all nights had 79% lower odds of having good average sleep quality than young adults who did not report their environment was too cold on any night (Table 4).

Environment too bright

There were between-person associations between reporting the environment was too bright and more WASO (b = 15.15, 95%CI [1.03, 29.28], praw = .04; Table 4), such that young adults reporting their environment was too bright on all nights had about 15 minutes more WASO compared to young adults reporting their environment was not too bright on any night. However, these results did not remain statistically significant after multiple testing correction (pHB = .11). There were no other significant associations between the environment being too bright and actigraphic sleep quality.

There were between-person associations between higher frequency of reporting the environment was too bright and lower self-reported sleep quality (OR = 0.10, 95% CI [0.01, 0.61], p =. 010). Specifically, young adults who reported their environment was too bright on all nights had 90% lower odds of having good sleep quality than young adults who did not report their environment was too bright on any night (Table 4).

Environment too noisy

Between-person associations revealed that young adults reporting their environment was too noisy inside on all nights had about 3% less SMEff compared to young adults not reporting their environment was too noisy on any nights (b= −2.97, 95% CI [−5.24, −0.70], praw = .010; Figure 4). This remained statistically significant after multiple testing correction (pHB = .030). There were within-person associations between reporting the environment was too noisy inside with WASO (b = 3.92, 95%CI [0.70, 7.14], praw = .02), such that on nights when young adults reported their environments were too noisy inside, they exhibited 4 minutes increased WASO compared to their own usual. However, these findings did not remain statistically significant after multiple testing correction (pHB =. 05; Table 4).

Figure 4:

Figure 4:

Between-person association between environment “too noisy inside” and sleep maintenance efficiency (SMEff). Adjusted for sex, race, ethnicity, education, employment status, and whether it was the weekend.

There were between person-associations between higher frequency of reporting the environment was too noisy inside and lower self-reported sleep quality (OR= 0.12, 95% CI [0.02, 0.54], p=.006). Specifically, young adults who reported their environment was too noisy inside on all nights had 88% lower odds of having good sleep quality than young adults who did not report their environment was too noisy inside on any night (Table 4).

No other associations between the environment being too noisy (inside or outside) and actigraphic or self-reported sleep quality were found.

Supplementary analyses on seasonality in the odds of reporting hot and cold (Supplementary Table 3), and whether including seasons in the model attenuated results reported in the manuscript (Supplementary Tables 4–9) are reported in supplementary materials.

Lastly, we report the prevalence of chronic conditions in our study (Supplementary Tables 10–12), whether diagnosis of a chronic condition predicted inclusion into the study (Supplementary Table 13), and added chronic condition diagnosis as a covariate into models to assess if the results reported in this manuscript were attenuated (Supplementary Tables 14–16).

IV. Discussion

We assessed the associations between factors related to the physical sleeping environment (being too hot, cold, bright, or noisy) and young adults’ actigraphic sleep and perceived sleep quality. We found both between- and within-person associations of the physical environment with sleep timing and both actigraphic and self-reported quality but not with sleep duration. The sleeping environment being perceived as too hot or too cold was associated with worse sleep quality. Sleeping environments that were too bright were associated with later sleep timing and worse sleep quality. Lastly, sleeping environments that were too noisy inside were associated with worse sleep quality, but we found no associations between sleeping environments that were too noisy outside and any sleep outcome. Our analyses fill a gap in the current literature by demonstrating that suboptimal environmental factors acutely predict later sleep timing and worse quality but not sleep duration among young adults.

We did not find any associations between the sleeping environment being too hot and sleep timing. Results of other studies that examine this association are inconsistent. In middle-aged adults, higher daytime and warmer evening temperatures (measured no later than 7pm) were associated with lower odds of later bedtimes.7 In another study, rising temperatures were associated with later sleep onset and midpoint and earlier sleep offset.8 Sleep onset latency was longer in an environment with a hotter (30°C) compared to moderate (26°C) temperature among young adults.44 Conversely, another study showed no association between hot temperature and self-reported sleep onset latency.6 The inconsistency in studies examining the associations between a hot sleeping environment and sleep timing motivates future experimental research to determine the extent to which sleep timing is impacted by changes in temperature.

Our results revealed associations between the sleeping environment being too hot and lower sleep quality, consistent with our hypotheses and previous research. Few studies examine the associations between hot temperatures and sleep quality specifically among young adults and outside the laboratory. Previous research among adults demonstrated that a 10°C F higher daily average temperature (assessed by weather stations) was associated with about 0.88 minutes increased actigraphic WASO and 0.14% lower actigraphic sleep efficiency that night6. In another study among adults, increasing objective bedroom temperature was associated with a dose-dependent decrease in sleep efficiency.45 Additionally, in an experimental study among young adults, increased core body temperature was associated with significantly decreased sleep efficiency specifically during sleep occurring between 2:30am and 7am.46 Though we cannot make explicit comparisons on effect sizes between our study and previous literature discussed due to differences in study design, the trend of our findings on hot temperature and sleep quality are largely consistent with previous literature and demonstrate that higher temperatures are associated with worse sleep quality. Future interventions should consider reducing high ambient temperatures as a strategy to improve sleep quality.

We found the sleep environment being “too cold” was associated with earlier sleep onset; however, these results did not remain statistically significant after multiple testing correction. There are few studies that focus specifically on the associations of sleep with colder environments, with most temperature studies emphasizing the association with warmer temperatures. Moreover, few studies utilize multiple measures of sleep timing (onset, midpoint, and offset) and typically only study sleep onset in their examination of the effects of cold temperature on sleep timing. For example, among young adults, cooler bedroom temperature (25°C) was associated with delayed sleep onset.11 In an experimental study, a cooler pre-sleep environment (23°C) was associated with increased sleep onset latency among young adult females.47 Our hypothesis that the sleeping environment being “too cold” would be associated with more WASO and lower SMEff was not supported. The lack of research on associations between cold temperature and sleep preclude us from making direct comparison between our null results and past work. Our findings motivate future research to examine the causal relationships between cooler sleeping environments and sleep timing.

Young adults who reported their sleeping environment was too bright exhibited later sleep timing, consistent with our hypothesis and previous research.15,48,49 These associations could be explained by light exposure at night being associated with a delayed circadian rhythm.18 We also found that the sleeping environment being “too bright” was associated with increased WASO; however, this result did not remain statistically significant after multiple testing correction. The environment being too bright was not associated with SMEff as hypothesized, contrasting with previous literature among young adults that found that greater bedroom light exposure was associated with a decrease in actigraphic sleep efficiency and more WASO.50 Among US adults, living in areas with greater outdoor light at night was associated with increased discontent with sleep quality19, consistent with our findings of self-reported sleep quality. Overall, our findings are consistent with previous literature and demonstrate the importance of limiting exposure to artificial light at night.51,52

We found that the environment being too noisy inside was associated with significantly lower SMEff, consistent with previous literature on noise exposure. In one study, participants reported worse sleep quality on nights when recordings of traffic noises were played.24 Moreover, daily objectively measured residential environmental noise during the night was associated with lower actigraphic sleep efficiency and increased WASO.23 To our knowledge, there are no other studies that specifically ask participants daily whether the environment being too noisy inside disrupted their sleep. Therefore, we were motivated to separately examine noisiness inside and outside to understand the difference between sources of noise. Furthermore, we found no significant associations between the environment being too noisy outside and sleep outcomes, which may be due to fewer individuals stating this as a reason for sleep disruption, hence reducing our power to detect an effect (proportion of nights noisy inside/noisy outside average frequency [SD] = 0.06 [0.15]/ 0.03 [0.11]). Previous research has suggested that noise’s impact on sleep may not necessarily be due to the source or intensity of the noise but the disruptiveness and meaning attributed to the noise.20 Additionally, people can habituate to noise exposure during sleep, returning to baseline after some period of exposure.53,54 Therefore, studies that use objective measures of noise could also examine whether and how perceptions of the noise’s disruptiveness may differ across people. Lastly, our study did not find associations between the physical environment and sleep duration, contrary to our hypothesis and some previous literature.17,55,56

Overall, most of the variance in the predictors and outcomes across the study period was explained at the within-person level (Supplementary Figure 1), motivating our decision to examine both the between- and within-person associations between the environment and sleep. The only within-person associations that remained statistically significant after multiple testing correction were the associations of the environment being too hot with lower SMEff and self-reported sleep quality. For the association between the environment being too hot with SMEff, the between- and within-person findings were both significant, and independent of one another (i.e., uncorrelated),57 meaning that the association between the environment being too hot and SMEff was explained at both the between- and within-person levels. Conversely, the association between the environment being too hot and self-reported sleep quality was only significant at the within-person level. Perceptions of temperature may be variable across individuals.58 The bedroom environments of participants may also differ in temperature and influences from humidity that contribute to perceptions of hot temperature.45 Due to these differences, it is important to examine within-individual variation in addition to between-individual variation when possible.

The present work showed that reports of disruptive physical environments were associated with poorer sleep quality in young adults enrolled in the Future of Families’ Young Adult Sleep Study. These results have implications for public health because sleep is essential for physical and mental wellbeing,59–61 and understanding factors that may negatively impact sleep could identify avenues to improve health. Future interventions of the sleeping environments of young adults may improve sleep health. Additionally, future studies that examine the associations between the environment and sleep should examine other factors of the physical sleeping environment that are also associated with sleep, such as mattress firmness and comfortability62–64 and indoor ventilation and humidity.5,45,65 Examining these predictors of sleep health could add to the understanding of how sleep is disrupted by the sleeping environment. These results also have implications for building managers of dormitories and apartments, who should prioritize the installation and maintenance of modern heating, ventilation, and air conditioning (HVAC) systems to improve temperature conditions, installation of light-blocking blinds or blackout curtains, and minimization of nighttime noise through sound-attenuating building materials and strict reinforcement of noise limits. Lastly, associations between environmental disturbances may be bidirectional, as poor sleep could make individuals more sensitive to environmental stimuli; for example, by increasing irritability.66

The present study has notable strengths and several limitations. Strengths include the large, diverse, national sample of young adults. We also examined multiple dimensions of sleep, including objective measures with wrist actigraphy and perceived, self-reported sleep quality. We were also able to examine both between- and within-person associations between the sleeping environment and sleep health, allowing for examination of directionality. One limitation is the self-reported nature of our environmental measures. These interpretations of environmental factors may differ across people and may introduce measurement error, compared to studies that use objective measures. Particularly, these self-reports did not provide information regarding the source(s) of the environmental factors (e.g., ambulance sirens; source of light), which could have aided in more specific interpretation of the sources of environmental disruptions (temperature, light and noise) and how they differentially impact sleep health. For example, our analyses may be confounded by between-individual differences in seasonal exposure. Thus, future research should use both objective and subjective measures as well as information on season to have a more comprehensive view of the associations between the environment and sleep. Second, although our study captures temporal precedence, it is observational and cannot establish causation. Third, we relied upon a single item to measure self-reported sleep quality, possibly not capturing the multidimensional nature of sleep quality. Fourth, given our moderate analytical sample size and the application of multiple testing corrections, our study was likely powered to detect only large effects; thus, null findings may be an indication of insufficient power. Lastly, our analyses were conducted among young adults from primarily urban settings in a study that oversampled nonmarried mothers, and some individuals were excluded due to missing data (n=677). Due to the analytic sample being significantly smaller than the full sample, there may be concern of non-random participation into the study, which may lead to results being an indication of effects within the analytic sample that may not generalize to the main study population. However, we conducted demographic difference analyses and found minimal differences between the included and exclusion sample (Supplementary Table 2), hence reducing concerns of selection bias. Nonetheless, these results may not generalize to more rural areas, other age groups, and individuals from other family structures.

V. Conclusions

The present findings indicate that multiple aspects of the physical sleeping environment not conducive to sleep health (e.g., too bright; too hot) predict later sleep timing and worse sleep quality that same night within young adults, and that young adults who experience certain factors more often in their sleeping environments exhibit worse sleep on average. Our findings motivate future research that examines the causal relationships and underlying mechanisms between the physical sleeping environment and sleep.

Supplementary Material

1
2

Impact and relevance statement.

What was known

Observational and experimental studies have shown that suboptimal physical environmental factors (e.g., extreme temperatures, bright light, and loud noise) are associated with actigraphic and self-reported short and poor sleep quality in between-person analyses. Few studies that examine the associations between the physical environment and sleep use actigraphic data on repeated measures within the same individual in young adults. Gaps exist in studies that examine the within-person associations between the physical environment and sleep.

What this study adds

This study contributes to the literature by demonstrating both between- and within-person associations between perceived disturbances in temperature, light, and noise, and actigraphic sleep outcomes collected that same night in young adults. These findings have implications for future sleep interventions aimed at improving sleep outcomes by modifying the physical sleep environment.

Acknowledgments

We thank the families who participated in the study and the FFCWS team. The parent study was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) of the National Institutes of Health under parent grants R01HD036916, R01HD039135, and R01HD040421, as well as a consortium of private foundations.

Funding statement:

This work was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) of the National Institutes of Health under award number R01HD073352-10 (to LH).

Since this manuscript is the result of funding in whole or in part by the National Institutes of Health (NIH), it is subject to the NIH Public Access Policy. Through acceptance of this federal funding, NIH has been given a right to make this manuscript publicly available in PubMed Central upon the Official Date of Publication, as defined by NIH.

Project funding information for R01HD073352-10: <https://reporter.nih.gov/search/t9yesOJ_Y0GfpnU4tYF7Rg/project-details/10634756> included $622,473 total federal funding for the budget period. Non-federal funding included institutional funding of students and faculty time for research that included time and support for this work but was not specifically evaluated for costs, as well as the open access fee ($4,040) to meet NIH policy requirements and publisher requirements. In the absence of current guidance for these calculations and reporting, it is estimated that federal funding comprised ~ 90% and institutional funding ~10%.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Statement on the use of generative AI:

We did not use generative AI services.

Published/under review elsewhere? (required)

This work has not been published elsewhere.

Figures-color printing?

No

Declaration of conflicts of interest:

None of the authors have conflicts of interests related to the material presented.

Outside of the present work, GMM, DAR, AMC, and LH received a small honorarium from the National Sleep Foundation for participation in a consensus panel. YS is supported by the Prevention and Methodology Training Program (T32-DA017629). OMB discloses that, outside the current work, he received two subcontract grants to Penn State from Mobile Sleep Technologies (NSF/STTR #1622766, NIH/NIA SBIR R43-AG056250, R44-AG056250), received an honorarium for his role as the Editor in Chief of Sleep Health (2019-2025) sleephealthjournal.org, and received honoraria/travel support for lectures from New York University, Tufts School of Dental Medicine, Spencer Study Club, University of Utah, University of Arizona, and University of Miami; consulting fees for grant review from Georgia State University and Harvard Chan School of Public Health.

A-MC has received a grant to the Pennsylvania State University grants from Kunasan, Inc., personal fees from University of Miami, personal fees from National Sleep Foundation, outside the submitted work. LH reports consulting fees from Idorsia Pharmaceuticals, Children and Screens Institute, and the National Sleep Foundation, and for serving as an expert witness from social media litigation, outside the submitted work. All other authors have no conflicts of interest to disclose.

Given their roles as Sleep Health editorial board members, Drs Buxton, Hale and Chang were not involved in the peer review of this article and have no access to information regarding its peer-review. Full responsibility for the editorial process for this article was delegated to other journal editors.

Data availability statement:

Survey data from the Future of Families and Child Wellbeing study (https://ffcws.princeton.edu/documentation) are publicly available from Princeton University’s Office of Population Research (OPR) data archive: https://ffcws.princeton.edu/restricted. The daily diary and sleep actigraphy datasets analyzed in the current work are not publicly available yet but will be available through an application process at the above link.

References

  • 1.Okamoto-Mizuno K, Mizuno K. Effects of thermal environment on sleep and circadian rhythm. J Physiol Anthropol. 2012;31(1):14. doi: 10.1186/1880-6805-31-14 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Krauchi Kurt, de Boer Tom. Body Temperature, Sleep, and Hibernation. In: Principles and Practice of Sleep Medicine. 6th Edition. 2017:323–328. [Google Scholar]
  • 3.Bach V, Telliez F, Libert JP. The interaction between sleep and thermoregulation in adults and neonates. Sleep Med Rev. 2002;6(6):481–492. doi: 10.1053/smrv.2001.0177 [DOI] [PubMed] [Google Scholar]
  • 4.Mayne SL, DiFiore G, Hannan C, et al. Feasibility and acceptability of mobile methods to assess home and neighborhood environments related to adolescent sleep. Sleep Health J Natl Sleep Found. 2023;9(3):331–338. doi: 10.1016/j.sleh.2023.01.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Xiong J, Lan Li, Lian Zhiwei, and De dear R. Associations of bedroom temperature and ventilation with sleep quality. Sci Technol Built Environ. 2020;26(9):1274–1284. doi: 10.1080/23744731.2020.1756664 [DOI] [Google Scholar]
  • 6.Li W, Bertisch SM, Mostofsky E, Vgontzas A, Mittleman MA. Associations of daily weather and ambient air pollution with objectively assessed sleep duration and fragmentation: a prospective cohort study. Sleep Med. 2020;75:181–187. doi: 10.1016/j.sleep.2020.06.029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Deng R, Victoria G, Ucci M. Associations between residential daytime indoor temperature and self-reported sleep disturbances in UK adults: A cross-sectional study. Environ Res. 2024;257:119281. doi: 10.1016/j.envres.2024.119281 [DOI] [PubMed] [Google Scholar]
  • 8.Minor K, Bjerre-Nielsen A, Jonasdottir SS, Lehmann S, Obradovich N. Rising temperatures erode human sleep globally. One Earth. 2022;5(5):534–549. doi: 10.1016/j.oneear.2022.04.008 [DOI] [Google Scholar]
  • 9.Cedeño Laurent JG, Williams A, Oulhote Y, Zanobetti A, Allen JG, Spengler JD. Reduced cognitive function during a heat wave among residents of non-air-conditioned buildings: An observational study of young adults in the summer of 2016. PLoS Med. 2018;15(7). doi: 10.1371/journal.pmed.1002605 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Chimed-Ochir O, Ando S, Murakami S, et al. Perception of feeling cold in the bedroom and sleep quality. Nagoya J Med Sci. 2021;83(4):705–714. doi: 10.18999/nagjms.83.4.705 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Lan L, Lian ZW, Lin YB. Comfortably cool bedroom environment during the initial phase of the sleeping period delays the onset of sleep in summer. Build Environ. 2016;103:36–43. doi: 10.1016/j.buildenv.2016.03.030 [DOI] [Google Scholar]
  • 12.Czeisler Charles A., Buxton Orfeu M.. Human Circadian Timing System and Sleep-Wake Regulation. In: Principles and Practice of Sleep Medicine. 7th Edition. 2022:375–389. [Google Scholar]
  • 13.Roenneberg T, Kantermann T, Juda M, Vetter C, Allebrandt KV. Light and the Human Circadian Clock. In: Kramer A, Merrow M, eds. Circadian Clocks. Springer; 2013:311–331. doi: 10.1007/978-3-642-25950-0_13 [DOI] [PubMed] [Google Scholar]
  • 14.Bjorvatn B, Pallesen S. A practical approach to circadian rhythm sleep disorders. Sleep Med Rev. 2009;13(1):47–60. doi: 10.1016/j.smrv.2008.04.009 [DOI] [PubMed] [Google Scholar]
  • 15.Didikoglu A, Mohammadian N, Johnson S, et al. Associations between light exposure and sleep timing and sleepiness while awake in a sample of UK adults in everyday life. Proc Natl Acad Sci. 2023;120(42):e2301608120. doi: 10.1073/pnas.2301608120 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Xu Y xiang, Zhang J hui, Tao F biao, Sun Y. Association between exposure to light at night (LAN) and sleep problems: A systematic review and meta-analysis of observational studies. Sci Total Environ. 2023;857:159303. doi: 10.1016/j.scitotenv.2022.159303 [DOI] [PubMed] [Google Scholar]
  • 17.Xiao Q, Gee G, Jones RR, Jia P, James P, Hale L. Cross-sectional association between outdoor artificial light at night and sleep duration in middle-to-older aged adults: The NIH-AARP Diet and Health Study. Environ Res. 2020;180:108823. doi: 10.1016/j.envres.2019.108823 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Xu Y xiang, Huang Y, Ding W qin, et al. Exposure to real-ambient bedroom light at night delayed circadian rhythm in healthy Chinese young adults: A cross-sectional study. Environ Res. 2024;251:118657. doi: 10.1016/j.envres.2024.118657 [DOI] [PubMed] [Google Scholar]
  • 19.Ohayon MM, Milesi C. Artificial Outdoor Nighttime Lights Associate with Altered Sleep Behavior in the American General Population. Sleep. 2016;39(6):1311–1320. doi: 10.5665/sleep.5860 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Basner M, Babisch W, Davis A, et al. Auditory and non-auditory effects of noise on health. The Lancet. 2014;383(9925):1325–1332. doi: 10.1016/S0140-6736(13)61613-X [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Schade MM, Mathew GM, Roberts DM, Gartenberg D, Buxton OM. Enhancing Slow Oscillations and Increasing N3 Sleep Proportion with Supervised, Non-Phase-Locked Pink Noise and Other Non-Standard Auditory Stimulation During NREM Sleep. Nat Sci Sleep. 2020;12:411–429. doi: 10.2147/NSS.S243204 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Buxton OM, Ellenbogen JM, Wang W, et al. Sleep Disruption due to Hospital Noises. Ann Intern Med. 2012;157(3):170–179. doi: 10.7326/0003-4819-156-12-201208070-00472 [DOI] [PubMed] [Google Scholar]
  • 23.Chung S, Youn S, Lee C, et al. Environmental Noise and Sleep Disturbance: Night-to-Night Variability of Sleep/Wake Pattern. Sleep Med Res. 2016;7(2):78–81. doi: 10.17241/smr.2016.00122 [DOI] [Google Scholar]
  • 24.Popp RF, Maier S, Rothe S, et al. Impact of overnight traffic noise on sleep quality, sleepiness, and vigilant attention in long-haul truck drivers: Results of a pilot study. Noise Health. 2015;17(79):387. doi: 10.4103/1463-1741.169698 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Lenz B. The Transition From Adolescence to Young Adulthood: A Theoretical Perspective. J Sch Nurs. 2001;17(6):300–306. doi: 10.1177/10598405010170060401 [DOI] [PubMed] [Google Scholar]
  • 26.Raymore LA, Barber BL, Eccles JS. Leaving Home, Attending College, Partnership and Parenthood: The Role of Life Transition Events in Leisure Pattern Stability From Adolescence to Young Adulthood. J Youth Adolesc. 2001;30(2):197–223. doi: 10.1023/A:1010345825065 [DOI] [Google Scholar]
  • 27.Coulter R, Scott J. What Motivates Residential Mobility? Re-examining Self-Reported Reasons for Desiring and Making Residential Moves. Popul Space Place. 2015;21(4):354–371. doi: 10.1002/psp.1863 [DOI] [Google Scholar]
  • 28.Wick AZ, Combertaldi SL, Rasch B. The first-night effect of sleep occurs over nonconsecutive nights in unfamiliar and familiar environments. Sleep. 2024;47(10):zsae179. doi: 10.1093/sleep/zsae179 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Sexton-Radek K, Hartley A. College Residential Sleep Environment. Psychol Rep. 2013;113(3):903–907. doi: 10.2466/06.10.PR0.113×27z2 [DOI] [PubMed] [Google Scholar]
  • 30.Peltz JS. The day-to-day impact of nighttime noise disturbances on college students’ psychological functioning. J Am Coll Health. 2022;70(7):2061–2069. doi: 10.1080/07448481.2020.1842422 [DOI] [PubMed] [Google Scholar]
  • 31.Roisman GI, Masten AS, Coatsworth JD, Tellegen A. Salient and Emerging Developmental Tasks in the Transition to Adulthood. Child Dev. 2004;75(1):123–133. doi: 10.1111/j.1467-8624.2004.00658.x [DOI] [PubMed] [Google Scholar]
  • 32.Yau PS, Shane J, Heckhausen J. Developmental goals during the transition to young adulthood. Int J Behav Dev. 2021;45(6):475–483. doi: 10.1177/01650254211007564 [DOI] [Google Scholar]
  • 33.Griggs S, Harper A, Hickman RL. A systematic review of sleep deprivation and neurobehavioral function in young adults. Appl Nurs Res. 2022;63:151552. doi: 10.1016/j.apnr.2021.151552 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Gustavson K, Knudsen AK, Nesvåg R, Knudsen GP, Vollset SE, Reichborn-Kjennerud T. Prevalence and stability of mental disorders among young adults: findings from a longitudinal study. BMC Psychiatry. 2018;18(1):65. doi: 10.1186/s12888-018-1647-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Vestergaard CL, Skogen JC, Hysing M, Harvey AG, Vedaa Ø, Sivertsen B. Sleep duration and mental health in young adults. Sleep Med. 2024;115:30–38. doi: 10.1016/j.sleep.2024.01.021 [DOI] [PubMed] [Google Scholar]
  • 36.Nyer M, Farabaugh A, Fehling K, et al. Relationship Between Sleep Disturbance and Depression, Anxiety, and Functioning in College Students. Depress Anxiety. 2013;30(9):873–880. doi: 10.1002/da.22064 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Reichman NE, Teitler JO, Garfinkel I, McLanahan SS. Fragile Families: sample and design. Child Youth Serv Rev. 2001;23(4):303–326. doi: 10.1016/S0190-7409(01)00141-4 [DOI] [Google Scholar]
  • 38.Brick TR, Mundie J, Weaver J, Fraleigh R, Oravecz Z. Low-Burden Mobile Monitoring, Intervention, and Real-Time Analysis Using the Wear-IT Framework: Example and Usability Study. JMIR Form Res. 2020;4(6):e16072. doi: 10.2196/16072 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Master L, Nye RT, Lee S, et al. Bidirectional, Daily Temporal Associations between Sleep and Physical Activity in Adolescents. Sci Rep. 2019;9(1):7732. doi: 10.1038/s41598-019-44059-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Pinheiro J, Bates D, DebRoy S, et al. nlme: Linear and Nonlinear Mixed Effects Models. Published online March 31, 2025. Accessed June 30, 2025. https://cran.r-project.org/web/packages/nlme/index.html
  • 41.Brooks M, Bolker B, Kristensen K, et al. glmmTMB: Generalized Linear Mixed Models using Template Model Builder. Published online October 9, 2025. Accessed October 29, 2025. https://cran.rproject.org/web/packages/glmmTMB/index.html
  • 42.Holm S. A Simple Sequentially Rejective Multiple Test Procedure. Scand J Stat. 1979;6(2):65–70. [Google Scholar]
  • 43.Bender R, Lange S. Adjusting for multiple testing—when and how? J Clin Epidemiol. 2001;54(4):343–349. doi: 10.1016/S0895-4356(00)00314-0 [DOI] [PubMed] [Google Scholar]
  • 44.Lan L, Pan L, Lian Z, Huang H, Lin Y. Experimental study on thermal comfort of sleeping people at different air temperatures. Build Environ. 2014;73:24–31. doi: 10.1016/j.buildenv.2013.11.024 [DOI] [Google Scholar]
  • 45.Basner M, Smith MG, Jones CW, et al. Associations of bedroom PM2.5, CO2, temperature, humidity, and noise with sleep: An observational actigraphy study. Sleep Health. 2023;9(3):253–263. doi: 10.1016/j.sleh.2023.02.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Fletcher A, van den Heuvel C, Dawson D. Sleeping with an Electric Blanket: Effects on Core Temperature, Sleep, and Melatonin in Young Adults. Sleep. 1999;22(3):313–318. doi: 10.1093/sleep/22.3.313 [DOI] [PubMed] [Google Scholar]
  • 47.Zhang N, Cao B, Zhu Y. Effects of pre-sleep thermal environment on human thermal state and sleep quality. Build Environ. 2019;148:600–608. doi: 10.1016/j.buildenv.2018.11.035 [DOI] [Google Scholar]
  • 48.Caddick ZA, Gregory K, Arsintescu L, Flynn-Evans EE. A review of the environmental parameters necessary for an optimal sleep environment. Build Environ. 2018;132:11–20. doi: 10.1016/j.buildenv.2018.01.020 [DOI] [Google Scholar]
  • 49.Dautovich ND, Schreiber DR, Imel JL, et al. A systematic review of the amount and timing of light in association with objective and subjective sleep outcomes in community-dwelling adults. Sleep Health J Natl Sleep Found. 2019;5(1):31–48. doi: 10.1016/j.sleh.2018.09.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Xu Y xiang, Huang Y, Zhou Y, et al. Association Between Bedroom Light Pollution With Subjectively and Objectively Measured Sleep Parameters Among Chinese Young Adults. J Adolesc Health. 2024;74(1):169–176. doi: 10.1016/j.jadohealth.2023.08.010 [DOI] [PubMed] [Google Scholar]
  • 51.Peña-García A, Sędziwy A. Optimizing Lighting of Rural Roads and Protected Areas with White Light: A Compromise among Light Pollution, Energy Savings, and Visibility. LEUKOS. 2020;16(2):147–156. doi: 10.1080/15502724.2019.1574138 [DOI] [Google Scholar]
  • 52.Svechkina A, Portnov BA, Trop T. The impact of artificial light at night on human and ecosystem health: a systematic literature review. Landsc Ecol. 2020;35(8):1725–1742. doi: 10.1007/s10980-020-01053-1 [DOI] [Google Scholar]
  • 53.Kawada T, Xin P, Kuroiwa M, Sasazawa Y, Suzuki S, Tamura Y. Habituation of sleep to road traffic noise as determined ny polysomnography and an accelerometer. J Sound Vib. 2001;242(1):169–178. doi: 10.1006/jsvi.2000.3344 [DOI] [Google Scholar]
  • 54.Tamura Y, Horiyasu T, Sano Y, et al. Habituation of sleep to a ship’s noise as determined by actigraphy and a sleep questionnaire. J Sound Vib. 2002;250(1):107–113. doi: 10.1006/jsvi.2001.3900 [DOI] [Google Scholar]
  • 55.Bigler P, Janzen B. Too hot to sleep. J Environ Econ Manag. 2024;128:103063. doi: 10.1016/j.jeem.2024.103063 [DOI] [Google Scholar]
  • 56.Yan Y, Zhang H, Kang M, Lan L, Wang Z, Lin Y. Experimental study of the negative effects of raised bedroom temperature and reduced ventilation on the sleep quality of elderly subjects. Indoor Air. 2022;32(11):e13159. doi: 10.1111/ina.13159 [DOI] [PubMed] [Google Scholar]
  • 57.Hoffman L, Stawski RS. Persons as Contexts: Evaluating Between-Person and Within-Person Effects in Longitudinal Analysis. Res Hum Dev. 2009;6(2–3):97–120. doi: 10.1080/15427600902911189 [DOI] [Google Scholar]
  • 58.Van Someren EJW, Dekker K, Te Lindert BHW, et al. The experienced temperature sensitivity and regulation survey. Temperature. 2016;3(1):59–76. doi: 10.1080/23328940.2015.1130519 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Besedovsky L, Lange T, Born J. Sleep and immune function. Pflüg Arch - Eur J Physiol. 2012;463(1):121–137. doi: 10.1007/s00424-011-1044-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Tomaso CC, Johnson AB, Nelson TD. The effect of sleep deprivation and restriction on mood, emotion, and emotion regulation: three meta-analyses in one. Sleep. 2021;44(6):zsaa289. doi: 10.1093/sleep/zsaa289 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Stickgold R. Sleep-dependent memory consolidation. Nature. 2005;437(7063):1272–1278. doi: 10.1038/nature04286 [DOI] [PubMed] [Google Scholar]
  • 62.Radwan A, Fess P, James D, et al. Effect of different mattress designs on promoting sleep quality, pain reduction, and spinal alignment in adults with or without back pain; systematic review of controlled trials. Sleep Health. 2015;1(4):257–267. doi: 10.1016/j.sleh.2015.08.001 [DOI] [PubMed] [Google Scholar]
  • 63.Hu X, Gao Y, Song Y, et al. The Effect of Mattress Firmness on Sleep Architecture and PSG Characteristics. Nat Sci Sleep. 2025;17:865–878. doi: 10.2147/NSS.S503222 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64.Lee H, Park S. Quantitative effects of mattress types (comfortable vs. uncomfortable) on sleep quality through polysomnography and skin temperature. Int J Ind Ergon. 2006;36(11):943–949. doi: 10.1016/j.ergon.2006.07.007 [DOI] [Google Scholar]
  • 65.Zhang X, Luo G, Xie J, Liu J. Associations of bedroom air temperature and CO2 concentration with subjective perceptions and sleep quality during transition seasons. Indoor Air. 2021;31(4):1004–1017. doi: 10.1111/ina.12809 [DOI] [PubMed] [Google Scholar]
  • 66.Whiting C, Bellaert N, Deveney C, Tseng WL. Associations between sleep quality and irritability: Testing the mediating role of emotion regulation. Personal Individ Differ. 2023;213:112322. doi: 10.1016/j.paid.2023.112322 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
2

Data Availability Statement

Survey data from the Future of Families and Child Wellbeing study (https://ffcws.princeton.edu/documentation) are publicly available from Princeton University’s Office of Population Research (OPR) data archive: https://ffcws.princeton.edu/restricted. The daily diary and sleep actigraphy datasets analyzed in the current work are not publicly available yet but will be available through an application process at the above link.

RESOURCES