IntroductionBackground

Alzheimer disease and related dementias (ADRD) are a major public health challenge. The number of Americans aged ≥65 years with Alzheimer disease was estimated to be 6.7 million in 2023 []. In 2023, total payments for health care, long-term care, and hospice services for people aged ≥65 years with dementia were estimated to be US $345 billion in the United States. The number of people with Alzheimer disease in the United States is projected to grow to 13.8 million by 2060 [].

A major component of the high psychological and financial costs of ADRD is related to addressing the needs of people living with dementia who have behavioral and psychological symptoms of dementia (BPSD), also called neuropsychiatric symptoms. BPSD encompass the challenging behaviors exhibited by people living with dementia that comprise a variety of symptoms that commonly include, but are not limited to, apathy, anxiety, depression, agitation, delusions, hallucinations, motor disturbances, and sleep changes. These symptoms are associated with faster disease progression [], increased caregiving burden [], and earlier placement into long-term care settings []. Pharmacological intervention is often not effective and is associated with undesirable side effects [], with nonpharmacological therapy being a first-line approach to management [].

The causes of BPSD are multiple, ranging from intrinsic neuropathologic factors to human factors such as caregiver interactions and the social-environmental milieu []. There is considerable evidence suggesting that environmental factors can affect the progression of different diseases as well as human behaviors [,]. The environment as a risk factor for ADRD has been reviewed, suggesting links between increased risks of ADRD and factors such as poor air quality, environmental toxins, and occupation-related exposures [,]. However, what has been less clearly examined is the potential direct relationship between person- or residence-level experienced environmental factors and contemporaneous behavioral changes associated with ADRD. This is important because most older people and people living with dementia spend the majority of their time indoors [].

There has long been debate regarding the definition of the environment []. For the purpose of this review, we are focusing on the ambient environment as defined by Harris et al []. Factors pertaining to the ambient environment, such as light, noise, temperature, and humidity, contribute to the visual, acoustic, and thermal comforts of the occupants and may be associated with their BPSD []; for example, inadequate light has been suggested to be a risk factor for sundowning syndrome (people with late-stage dementia exhibiting more agitated behaviors in the late afternoon and evening) because of the role of light in vision and its role in circadian rhythm modulation [,]. Therefore, designing an indoor environment that would address the needs of people living with dementia is paramount and may reduce their BPSD. A first step to creating an optimal indoor environment for people living with dementia with BPSD is to understand which environmental factors are related to BPSD and how they impact the symptoms.

Previous papers have reviewed the effect of the long-term care environment on physical activity levels [], the achievement of therapeutic goals such as safety and socialization [], and neuropsychiatric symptoms [,]. However, these reviews have primarily focused on the parameters of the built environment (eg, size and layout of spaces), interior design, and occupancy and staffing ratios, rather than aspects of the environmental quality that impact comfort, such as temperature and humidity, noise levels, and lighting. Comfort is especially important to consider for people living with dementia because BPSD represent an expression of discomfort, stress, and unmet needs when insight and communication become more difficult []. Nevertheless, anosognosia and aphasia also prevent the subjective measurement of comfort. Thus, it is necessary to understand the impact of various objective aspects of environmental quality that may improve comfort and, in turn, BPSD.

Objectives

Accordingly, we conducted this scoping review to (1) examine how indoor environmental quality parameters pertaining to light, noise, temperature, and humidity are associated with BPSD and how controlling these parameters can help manage these symptoms and (2) identify the current state of knowledge on the relationship between indoor environmental quality parameters and BPSD, current gaps in the research, and potential future directions.

MethodsOverview

We followed the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) reporting guidelines for this scoping review. We conducted a scoping review because we aimed to (1) provide an overview of the broad literature examining the relationship between indoor environmental quality parameters and BPSD and (2) identify gaps and future directions.

The databases used were CINAHL, Embase, MEDLINE, and PsycINFO. A literature search strategy was crafted for each database that would retrieve records containing a combination of appropriate index terms and text words pertaining to the following concepts: dementia (dementia, Alzheimer disease, etc), environmental conditions (heat, temperature, light, humidity, etc), and BPSD (anxiety, depression, apathy, agitation, pacing, etc). The complete search strategies can be found in -. The original search was performed on October 7, 2020, and updated on April 21, 2022, and February 22, 2024. The inclusion and exclusion criteria are presented in .

Six reviewers (WMA-Y, CW, MN, LM, RH, and KW) completed the screening process, which consisted of two stages: (1) title and abstract screening and (2) full-text screening. First, the title and abstract of each paper were reviewed by 2 reviewers independently to examine whether the studies met the inclusion criteria. The reviewers then met iteratively to reconcile any differences in their decisions regarding whether a particular paper should be included. If the 2 reviewers could not come to an agreement, a third reviewer would help reach a decision. After title and abstract screening was completed, the included papers at this stage were retrieved for full-text screening, and the process was repeated after reading the papers in their entirety.

Textbox 1. Inclusion and exclusion criteria.

Inclusion criteria

Written in EnglishPeer reviewedStudies of older adults with Alzheimer disease and related dementias (ADRD), not including mild cognitive impairmentRelationships between indoor environmental quality parameters, including light, noise, temperature, and humidity and the behaviors of people living with dementiaExposure to environmental conditions leading to behavioral changeObservational or controlledAll residential settingsStudies using measurements of people’s behaviors and psychological symptoms as 1 of the outcomesStudies that include measurements or descriptions of the indoor environmental quality parameters (light, noise, temperature, and humidity)

Exclusion criteria

Review articles, commentaries, opinions, theses, and letters to editorStudies on genetics as risk factors for ADRDStudies that examine the effect of decoration, furniture arrangement, and home-like feeling on participants’ behaviorsStudies on the environment as a risk factor for developing ADRDStudies on horticultural therapy, music therapy, multisensory stimulation therapy, virtual or augmented reality, or outdoor activitiesStudies published before 2006 (the 15-year cutoff was chosen to include articles that best represent the current state of knowledge)

Six reviewers (WMA-Y, MN, ZB, LM, RH, and SG) extracted data from the included papers after the screening process was completed. A survey built in Qualtrics (Qualtrics International Inc) was created to assist this process. The survey listed questions covering study design, year of publication, the region and residential settings in which the research was conducted, sample size, demographics of the sample, which environmental conditions were assessed and how they were examined, and which BPSD were evaluated and how they were measured. Next, the papers were categorized by the examined environmental factor or factors: (1) light level, (2) noise level, (3) temperature or humidity, and (4) multiple indoor environmental quality parameters. If there were unified themes of BPSD examined in the articles (eg, agitation and mood-related symptoms), the study results of each theme would be synthesized by summarizing the counts of different study designs and providing a narrative summary []. All study data were also summarized into tables. In doing so, the current state of knowledge, gaps in research, and future directions were identified.

Ethical Considerations

Institutional review board approval was not required for this scoping review because the research did not involve human participants.

ResultsOverview

A total of 3123 articles were identified in the search initiated on October 7, 2020. After excluding papers published before 2006 as well as duplicates, the titles and abstracts of 1887 articles were screened, after which 1707 (90.46%) articles were excluded, and 180 (9.54%) were retrieved for full-text screening. Ultimately, of these 180 articles, 26 (14.4%) were included. The search was updated on April 21, 2022, and on February 22, 2024, and 12 articles were added to the final pool; hence, 38 papers were included in this scoping review (). Of the 38 papers, 8 (21%) were randomized controlled trials (RCTs), 14 (37%) were quasi-experiments, 5 (13%) were pretest-posttest studies, 6 (16%) were cross-sectional studies, 4 (11%) were case series, and 1 (3%) was a cohort study (refer to for more information).

‎Figure 1. PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram of the search and screening results. Table 1. Summary of residential settings, region, study design, intervention, sample size, participant cognitive status, and participant age for all included studies.Study, yearResidential settingsRegionStudy designIntervention (if any)Sample sizeParticipant cognitive statusParticipant age (y)Algase et al [], 2010Assisted living units, nursing homesNorth AmericaCross-sectional study—aMb: 28, Fc: 94MMSEd score: mean 7.4 (SD 7.2)Mean 83.7 (SD 6.48; range 68-102)Bankole et al [], 2020Personal homesNorth AmericaCase report or series—M: 6, F: 6 (the 12 participants lived with primary caregivers)All participants had dementiaMean 79.67 (SD 7.5; range 68-92)Barrick et al [], 2010Memory care units, psychiatric hospitalNorth AmericaQuasi-experimentFour lighting conditions were presented during multiple 3-week intervention periods: bright light in the morning (7-11 Am), bright light in the evening (4-8 PM), bright light all day (7 AM-8 PM), and standard lighting (ie, the baseline condition); bright light was administered at 2000 to 3000 lux, while standard lighting was administered at 500 to 600 luxM: 35, F: 31Cognitive status was measured using the MDS-COGSe and MMSE; the distribution of cognitive impairment severity among participants was as follows: mild=3, moderate=18, severe=31, and very severe=14North Carolina, United States: <65=6 (13%), 65-79=24 (52%), and ≥80=16 (35%); Oregon, United States: <65=0 (0%), 65-79=4 (20%), and ≥80=16 (80%)Bautrant et al [], 2019Nursing homesEuropePretest-posttest studySkylike ceiling tiles in part of the shared premises, gradual decrease of the illuminance at night, soothing streaming music, reinforcement of the illuminance during the day, walls painted in light beige, oversized clocks in corridors, and night team clothes color (dark blue) different from that of the day team (sky blue)M: 2, F: 17Severe dementia; MMSE scores were <15 for all patients≥65; mean 86.3 (SD 5.4)Bicket et al [], 2010Assisted living unitsNorth AmericaCross-sectional study—M: 80, F: 246Demented: 194, nondemented: 131Demented: mean 86.1 (SD 8.1), nondemented: mean 84.3 (SD 9.6)Bromundt et al [], 2019Nursing homesEuropeQuasi-experiment (balanced crossover, within-participant study)This was a 17-week trial with a balanced crossover, within- participant study design; the intervention involved exposure to individually timed dawn-dusk simulator over participants’ beds for 7-8 weeksM: 3, F: 17S-MMSEf score: mean 13.15 (SD 10.30); S-MMSE score at baseline: mean 14.35 (SD 10.51), S-MMSE score at the end of the study: mean 11.95 (SD 11.56)Mean 85.6 (SD 5.8)Burns et al [], 2009Nursing homesEuropeRandomized controlled trialFull-spectrum bright light therapy (10,000 lux) for 2 hours from 10 AM to noon for 2 weeks; the control group was administered standard fluorescent tube light (100 lux) for 2 hours from 10 AM to noon for 2 weeksM: 16, F: 32MMSE score at baseline: 5.1 for the standard lighting therapy group and 6.9 for the bright light therapy groupStandard light therapy group: mean 82.5 (SD 1.5), bright light therapy group: mean 84.5 (SD 1.7)Cohen-Mansfield et al [], 2010Nursing homesNorth AmericaQuasi-experiment (indoor environmental quality parameters were not manipulated)Presentation of 9 categories of stimuli: live human social, real pet, simulated social, self-identity, inanimate social, music, manipulative, reading, and task- or work-relatedM: 42, F: 151MMSE score: mean 7.2 (SD 6.3; range 0-23)Mean 86 (range 60-101)Cohen-Mansfield et al [], 2011Nursing homesNorth AmericaQuasi-experiment (indoor environmental quality parameters were not manipulated)Presentation of 9 categories of stimuli: live human social, real pet, simulated social, self-identity, inanimate social, music, manipulative, reading and task-or work-relatedM: 42, F: 151MMSE score: mean 7.2 (SD 6.3; range 0-23)Mean 86 (range 60-101)Cohen-Mansfield et al [], 2012Nursing homesNorth AmericaQuasi-experiment (indoor environmental quality parameters were not manipulated)Presentation of 9 categories of stimuli: live human social, real pet, simulated social, self-identity, inanimate social, music, manipulative, reading and task-or work-relatedM: 42, F: 151MMSE score: mean 7.2 (SD 6.3; range 0-23)Mean 86 (range 60-101)Cohen-Mansfield [], 2020Nursing homesNorth AmericaCross-sectional study—M: 26, F: 43CPSg score: mean 4.19Mean 86.58 (SD 7.93)Dowling et al [], 2007Nursing homesNorth AmericaRandomized controlled trialDuring either morning (9:30-10:30 AM) or afternoon (3:30-4:30 PM), bright light (>2500 lux in gaze direction) was administered Monday through Friday for 10 weeks; Brite Lite IV (Apollo Light Systems, Inc) light boxes were used when necessary to supplement the ambient lightM: 13, F: 57MMSE score: mean 7 (SD 7; range 0-23)Mean 84 (SD 10; range 58-98)Figueiro et al [], 2014Nursing homesNorth AmericaPretest-posttest studyLow-level bluish-white lighting designed to deliver high circadian stimulation during the daytime was installed in 14 nursing home resident rooms for 4 weeksM: 5, F: 9BIMSh score: mean 7.7 (SD 2.3)Mean 86.9 (SD 4.4)Figueiro et al [], 2019Assisted living units, long-term care unitsNorth AmericaRandomized controlled trialCrossover design clinical trial administered an all-day active or control lighting intervention for two 4-week periods (separated by a 4-week washout); the active lighting intervention provided high circadian stimulus, and the control intervention provided low circadian stimulus that was below the threshold for activation of the circadian systemM: 16, F: 30MMSE score between 4 and 24 or BIMS score between 3 and 12Mean 85.1 (SD 7.1)Figueiro et al [], 2020Assisted living units, long-term care unitsNorth AmericaPretest-posttest studyParticipants were exposed to a single daytime (from approximately 6 AM-8 AM to 6 PM) tailored lighting intervention that provided high levels of circadian-effective light; the light-delivery method (ie, custom-built floor lamps, light boxes, and light tables) was dependent on where individual participants spent most of their dayM: 20, F: 27MMSE scores: F=mean 15.2 (SD 3.4) and M=mean 13.6 (SD 2.9); BIMS scores: F=mean 4.0 (SD 1.7) and M=mean 5.6 (SD 2.9)Mean 85.3 (SD 7.1)Figueiro et al [], 2023Assisted living units, memory care unitsNorth AmericaQuasi-experimentWith a crossover, placebo-controlled design, 3 different lighting modes (light tables, light trays, and ambient room lighting) were used to deliver high levels of circadian stimulus to the participants’ eyes for two 8-week intervention periods in a counter balanced order with a 4-week washout between the study’s 2 conditions (dim light control vs active intervention)M: 3, F: 11Moderate to severe ADRDi indicated by a GDSj score of 4.0 to 6.0; the mean BCRSk score was 4.83 (SD 1.01), and the mean GDS score was 4.89 (SD 1.1)Mean 84.1 (SD 8.9)Garre-Olmo et al [], 2012Nursing homesEuropeCross-sectional study—M: 37, F 123MMSE score: mean 4.1 (SD 6.3); Barthel Index score: mean 10.5 (SD 19.3)Mean 82.6 (SD 11.6)Hickman et al [], 2007Memory care units, specialized older adult centerNorth AmericaQuasi-experimentAmbient bright light therapy was delivered through a high-intensity, low-glare lighting system installed in the public areas that involved 4 lighting conditions: morning bright light, evening bright light, all-day bright light, and standard lighting. The bright light conditions were delivered at 2000 to 2500 lux, while the standard lighting condition was delivered at 500 to 600 luxM: 35, F: 31Mild to moderate dementia: M=12 (34.3%), F=9 (29%); severe dementia: M=15 (42.9%), F=16 (51.6%); very severe dementia: M=8 (22.9%), F=6 (19.4%)<65: M=5 (14.3%), F=1 (3.2%); 65-79: M=18 (51.4%), F=10 (32.3%); ≥80: M=12 (34.3%), F=20 (64.5%)Hjetland et al [], 2021Nursing homesEuropeRandomized controlled trialA 24-week cluster-randomized, placebo-controlled trial with an intervention involving ambient light administered at 1000 vertical lux and 6000 K from 10 AM to 3 PM, with gradually increasing and decreasing light levels before and after this interval, delivered by Glamox ASM: 22, F: 47MMSE score: mean 6.4 (SD 6.7)Mean 83.5 (SD 7.1)Jao et al [], 2015Assisted living units, nursing homesNorth AmericaCase report or series—M; 9, F: 29MMSE score: mean 12.9 (SD 6.5; range 2-23)Mean 82.7 (SD 6.3; range 68-94)Joosse [], 2011Nursing homesNorth AmericaCross-sectional study—M: 11, F: 42MMSE score: mean 6.83 (SD 6.67; range 0 to 19); specifically, 37 (70%) scored <13 (severe dementia), and 16 (30%) scored 13 to 20 (moderate dementia)Mean 86.53 (SD 9.32; range 61 to 103)Kim et al [], 2021Dementia clinicAsiaQuasi-experimentThe treatment group (14 participants) sat approximately 60 cm away from a small (136×73×16 mm) light-emitting diode light box for 1 hour each in the morning for 2 weeks, and the control group (11 participants) wore dark, blue-attenuating sunglasses during the 1-hour exposures; the morning light exposure started 9 to 10 hours after each individual’s dim light melatonin onsetM: 7, F: 18MMSE-KCl: treatment group (M: 2, F: 12)=mean 16.36 (SD 5.09), control group (M: 2, F: 12)=mean 16.90 (SD 4.91)Treatment group: mean 77.36 (SD 5.79), control group: mean 78.55 (SD 7.71)Kolberg et al [], 2021Nursing home dementia unitsEuropeRandomized controlled trialA 24-week cluster-randomized, placebo-controlled trial with an intervention involving ambient light administered at 1000 lux and 6000 K from 10 AM to 3 PM, with gradually increasing and decreasing light levels before and after this intervalM: 22, F: 47MMSE score: median 4.0 (IQR 1.0-9.2)Median 85.0 (IQR 79.0-88.0)Konis et al [], 2018Dementia care communitiesNorth AmericaQuasi-experimentThe daylight exposure of participants was increased by taking them to the perimeter zone of a daylit room from 8 AM to 10 AM for socialization over a period of 12 weeks; the perimeter zone was the region of the room within 3 m from the windows; it was administered every day throughout the studyM: 21, F: 56Alzheimer disease: daylight group=17 (37%) patients, control group=8 (25.8%) patients; frontotemporal dementia: daylight group=1 (2.2%) patient, control group=1 (3.2%) patient; Lewy body dementia: daylight group=0 (0%) patients, control group=1 (3.2%) patient; vascular dementia: daylight group=2 (4.3%) patients, control group=1 (3.2%) patient; dementia not otherwise specified: daylight group=19 (41.3%) patients, control group=17 (54.8%) patients; mild cognitive impairment: daylight group: 6 (13%) patients, control group=2 (6.5%) patients; not specified: daylight group=1 (2.2%) patient, control group=1 (3.2%) patientMean 85.3 (SD 7.0)Lee et al [], 2016Long-term care units (Richmond Manor and Maple Manor)North AmericaCohort study—M: 6, F: 6Early or middle stage of ADRDRichmond Manor: mean 82.9 (SD 8.9), Maple Manor: mean 77.6 (SD 9.8)Lin et al [], 2018Dementia care centersAsiaQuasi-experimentExperimental group participants received 20 minutes of white noise consisting of ocean, rain, wind, and running water sounds between 4 PM and 5 PM daily over a period of 4 weeks, with volume maintained at 55-70 dB; the comparison group received routine careM: 23, F: 40NRmExperimental group: mean 81.57 (SD 10.52), comparison group: mean 79.29 (SD 8.41)Liu et al [], 2021Nursing homes, communitiesAsiaQuasi-experimentParticipants in the experimental group were exposed to ambient light at 2500 lux for at least 60 minutes per day from 9 AM to 10 AM, Monday through Friday, over 8 weeksM: 7, F: 28Mild dementia: 10, moderate dementia: 16, severe dementia: 960-95Liu et al [], 2022Nursing homesAsiaQuasi-experimentParticipants in the experimental group were exposed to ambient light at 2500 lux for at least 60 minutes per day from 9 AM to 10 AM, Monday through Friday, over 8 weeksM: 7, F: 28Mild dementia: 10, moderate dementia: 16, severe dementia: 960-95 experimental group: mean 83.9 (SD 7.1), comparison group: mean 80.2 (SD 7.2)McCurry et al [], 2011Personal homesNorth AmericaRandomized controlled trialIntervention 1: participants attempted to reach the goal of walking for 30 minutes continuously per day; intervention 2: participants sat in front of a light box (SunRay; The SunBox Company) for 1 hour per day, timed to be within 2-hour window before the participants’ habitual bedtime; intervention 3: combination treatment comprising walking, light exposure, and guided sleep education (NITE-ADn)M: 59, F: 73MMSE scores: walking group=mean 19.2 (SD 7.7); light group=mean 17.9 (SD 7.0), NITE-AD group=mean 19.1 (SD 5.8), control group=mean 18.7 (SD 6.9)Walking group: mean 82.2 (SD 8.5), light group: mean 80.6 years (SD 7.3), NITE-AD group: mean 80.0 (SD 8.2), control group: mean 81.2 (SD 8.0)Olsen et al [], 2016Personal homes, nursing homesEuropeCross-sectional study—M: 67, F: 126Nursing home residents (Clinical Dementia Rating Scale): mild dementia=9%, moderate dementia=43.6%, severe dementia=47.4%; home-dwelling people with dementia (Clinical Dementia Rating Scale): mild dementia=43.5%, moderate dementia=47%, severe dementia=5.2%Nursing home residents: mean 84.6 years (SD 6.50), home-dwelling people with dementia: mean 82.6 (SD 6.84)Onega et al [], 2016Long-term care unitsNorth AmericaRandomized controlled trialIndividuals were seated approximately 27 inches away from a bright light that delivered 10,000 lux of light (treatment arm) or a low-level light that delivered 250 lux of light (control) for 30 minutes twice a day 5 times a week for 8 weeksM: 17, F: 43MMSE scores ranged from 0 to 22; participants were classified into 3 categories based on the MMSE scores: 7 (11.7%) had mild dementia, 11 (18.3%) had moderate dementia, 42 (70%) had severe dementiaMean 82.6 (SD 9.6)Riemersma-van der Lek et al [], 2008Group care facilitiesEuropeRandomized controlled trialRandom assignment by facility to long-term daily treatment with whole-day bright (approximately 1000 lux) or dim (approximately 300 lux) light and by participant to evening melatonin (2.5 mg) or placebo for a mean of 15 (SD 12) months (maximum period of 3.5 years)M: 19, F: 170Treatment group, placebo: MMSE score=mean 14.3 (SD 7.0); treatment group, light: MMSE score=mean 14.5 (SD 6.2); treatment group, melatonin: MMSE score=mean 15.3 (SD 5.3); treatment group, light+melatonin: MMSE score=mean 14.7 (SD 6.8)Mean 85.8 (SD 5.5)Saidane et al [], 2023Nursing homesEuropePretest-posttest studyNaturalistic light systems that replicated the spectrum distribution of natural light from dusk to dawn5NRNRSloane et al [], 2007Long-term care unitsNorth AmericaQuasi-experimentThe study used a cluster-unit crossover experimental design to evaluate 4 ambient lighting conditions: morning (7-11)=high-intensity light, evening (4-8)=high-intensity light, all day (7 AM-8 PM)=high-intensity light, and industry minimum standard lightingM: 35, F: 31Mild to moderate cognitive impairment: 21 (31.8%), severe cognitive impairment: 31 (47%), very severe cognitive impairment: 14 (21.2%)Mean 79Son and Kwag [], 2020Center for dementiaAsiaQuasi-experimentIn the experimental group, a walking program with white noise was applied 3 times a week for 4 weeks; white noise was provided by a white noise generator, with volume maintained at 40 to 50 dB; in the control group, only the walking program was appliedM: 16, F: 16NRRange 71-80; experimental group: mean 75.06 (2.82), control group: mean 75.31 (SD 3.07)Tartarini et al [], 2017Nursing homesOceaniaCase report or series—M: 14, F: 7Psychogeriatric Assessment Scale scores of ≥10Range 61-92Wahnschaffe et al [], 2017Nursing homesEuropePretest-posttest studyFrom midwinter on, a ceiling-mounted dynamic lighting system was installed in the common room of a nursing home and programmed to produce high illuminance with higher blue light proportions during the day and lower illuminance without blue light in the eveningM: 5, F: 7MMSE score: mean 12.1 (SD 9.2)Mean 79.1 (SD 11)Wahnschaffe et al [], 2017Nursing homesEuropeCase report or series—M: 1, F: 19ICD-10o criteria: Alzheimer disease=9, vascular dementia=2, frontotemporal dementia=1, Korsakoff syndrome=1, and dementia subtypes (which were not further specified)=7 Mean 83.8 years (SD 8.8)

aNot applicable.

bM: male.

cF: female.

dMMSE: Mini–Mental State Examination.

eMDS-COGS: Minimum Data Set Cognition Scale.

fS-MMSE: Standardized Mini–Mental State Examination.

gCPS: Cognitive Performance Scale.

hBIMS: Brief Interview for Mental Status.

iADRD: Alzheimer disease and related dementias.

jGDS: Global Deterioration Scale.

kBCRS: Brief Cognitive Rating Scale.

lMMSE-KC: Korean version of the MMSE developed as part of the Korean version of the Consortium to Establish a Registry for Alzheimer’s Disease assessment packet.

mNR: not reported.

nNITE-AD: Nighttime Insomnia Treatment and Education in Alzheimer’s Disease.

oICD-10: International Classification of Diseases, Tenth Revision.

The following sections synthesize the findings of the included studies, categorized by the indoor environmental quality factor or factors examined ().

Table 2. Summary of environmental conditions and behavioral and psychological symptoms of dementia (BPSD) outcomes examined, the subjective and objective methods of assessment, and the main results for all included studies.Study, yearEnvironmental condition or conditions measured or manipulatedSubjective indoor environmental quality parameter assessment (if any)Objective indoor environmental quality parameter assessment (if any)Behavioral or psychological outcomesSubjective BPSD assessment (if any)Objective BPSD assessment (if any)Main resultsAlgase et al [], 2010Light level, noise level, temperature, humidity, environmental ambience, crowdingAmbience scale, presence or absence of people within 8 feet of the participantGossen Color-Pro 3F light meter (Gossen Color-Pro 3F light meter, Bogen Photo Corp), Quest Technologies Sound Level Meter Model 2400, and RadioShack Thermometer with Indoor Humidity Gauge 63-1013WanderingObservations from videotapes—aBrighter light, more variation in sound levels, and a higher engaging quality of the environment were associated with wandering, and a higher soothing quality of the environment was associated with periods when wandering did not occurBankole et al [], 2020Light level, noise level, temperature, humidity, atmospheric pressure—Off-the-shelf sensorsAgitation, depression, sleep, quality of life, overall level of BPSDNPI-Qb, CMAI-Cc, PSQIdShimmer3 sensor for people with dementia and Pebble sensor for caregivers in phase 1; Pebble for both people with dementia and caregivers in phase 2Temperature, atmospheric pressure, and time showed strong correlations with agitation, which was self-reported by the primary caregiversBarrick et al [], 2010Light level——AgitationCMAIe, observational agitation using the presence or absence of 8 agitated behaviors—No therapeutic conditions improved agitation in comparison to standard lightingBautrant et al [], 2019Light level, skylike ceiling tiles, soothing streaming music, walls painted in light beige, oversized clocks in corridors——Agitation, wandering, screaming, episodes of BPSDNumber and duration of disruptive BPSD episodes were systematically collected—BPSD prevalence reduced after plain environmental rearrangements aimed at improving spatial and temporal orientation were put in placeBicket et al [], 2010Light levelTESS-NH/RCf—Quality of life, overall level of BPSD, fall riskNPIg—The Assisted Living Environmental Quality Score was positively associated with the Alzheimer’s Disease–Related Quality of Life score (P=.01), while it was strongly negatively associated with NPI total score (P<.001) and negatively correlated with fall risk (P=.04)Bromundt et al [], 2019Light level——Agitation, mood, circadian rhythm (rest-activity pattern), activities of daily living, quality of life, alertness, verbal interaction, well-being, cheerfulness, memory, disturbing behaviorCMAI, VAShWrist actimetry (MotionWatch 8; CamNtech)Dawn-dusk simulator exposure led to significantly better mood in the morning hours (P=.002), while it did not significantly influence circadian parameters and sleep parameters (P>.05)Burns et al [], 2009Light level——Agitation, depression, sleepCMAI, CSDDi, MOUSEPADj, CRBRSk, sleep chartsActiwatch (supplied by CamNtech)Results suggested that the bright light intervention had a limited effect in reducing agitation, but it improved sleep in older adults with dementiaCohen-Mansfield et al [] 2010Light level, noise level, number of persons in proximityEnvironment portion of ABMIl—EngagementObservational measurement of engagement—Attention and engagement were the best when light level was moderate; engagement was the best when there was a moderate level of noise; attention to the engagement stimulus was significantly higher when there were 4 to 9 people in proximity in comparison to more or fewer people (P<.01)Cohen-Mansfield et al [], 2011Light level, noise level, number of persons in proximityEnvironment portion of ABMI—AgitationABMI, CMAI—Lighting and background noise did not affect agitation significantly, which may be due to little variation in these variables and their reliance on subjective perception scalesCohen-Mansfield et al [], 2012Light level, noise level, number of persons in proximityEnvironment portion of ABMI—PleasureLawton Modified Behavior Stream—Pleasure was most likely to occur in environments with moderate noise levelsCohen-Mansfield [], 2020Light level, noise level, temperature, location, time of day, total group sizeReported by therapeutic recreation staff—Mood, engagement, sleepinessGOMEl—Background noise and time of day significantly affected outcome variables after controlling for participants’ cognitive functioning and group topic; background noise was related with decreased engagement and increased sleepiness; there was little variation concerning temperature and lightDowling et al [], 2007Light level—Cal LIGHT 400 calibrated precision light meter for monitoring light levels in gaze direction for each participantAgitation, depression, overall level of BPSD, appetite or eating disorders, aberrant motor behavior, dysphoriaNPI-NHn—Bright light therapy was found not to clinically affect neuropsychiatric behaviorsFigueiro et al [], 2014Light level—Daysimeter worn on the wrist to estimate light exposure of each participantAgitation, depression, circadian rhythm (rest-activity pattern), activities of daily living, sleepCMAI, CSDD, PSQIDaysimeter300 to 400 lux of a bluish-white light significantly improved sleep efficiency, total sleep time, and global PSQI scores (P<.05) and decreased depression (CSDD) and agitation (CMAI) scores (P<.05)Figueiro et al [], 2019Light level—Daysimeter as a pendant to estimate light exposure of each participantAgitation, depression, circadian rhythm (rest-activity pattern), sleep, quality of lifeCMAI, CSDD, PSQIActiwatch 2 (Philips Respironics)The experimental group had significantly improved PSQI scores compared to the control group (P<.001); regarding secondary outcomes, the experimental group had significantly greater improvements in depression and agitation compared to the control group (P<.05)Figueiro et al [], 2020Light level—Daysimeter as a pendant to estimate light exposure of each participantAgitation, depression, mood, sleepCMAI, CSDD, PSQIActiwatch 2 (Philips Respironics)Long-term daily light exposure showed a positive correlation with reduced PSQI scores with increasing efficacy as the study went on; there were significant improvements in depression starting at around week 3; agitation also improved around week 9Figueiro et al [], 2023Light level——Depression, circadian rhythm (rest-activity pattern), sleepCSDD, PSQI, SDIoActiwatch 2 (Philips Respironics)Under the active condition, both objective and subjective sleep improved significantly (P=.02)Garre-Olmo et al [], 2012Light level, noise level, temperature—DT-8820 environment meter (Shenzhen Everbest Machinery Industry Co Ltd) in each participant’s bedroom and in the dining room and living room of each nursing homeAffect, mood, pain, quality of lifeQUALIDp, NPI-NH, PAIN-ADq—High temperature in the bedroom was associated with lower quality of life, high noise levels in the living room were associated with low social interactions, and low lighting levels in the bedroom were associated with more severe negative affective moodHickman et al [], 2007Light level——DepressionCSDD—The treatment had different effects on men and women; study results did not support the use of ambient bright light therapy as a treatment for depressive symptoms in persons with dementia; however, a subpopulation of persons with dementia may benefit from this interventionHjetland et al [], 2021Light level——Mood, circadian rhythm (rest-activity pattern), activities of daily living, cognition, sleep, overall level of BPSDNPI-NHActiwatch 2 (Philips Respironics)There was discrepancy in the subjective and objective sleep measures; actigraphically measured sleep outcomes showed no statistically significant differences between patients in the intervention and control groups; however, better sleep was reported using the SDI in the intervention groupJao et al [], 2015Light level, noise level, environmental ambience, crowding, staff familiarity, environmental stimulationPEARr-Environment subscaleGossen Color-Pro 3F light meter (Bogen Photo Corp) and the Quest Technologies Sound Level MeterApathyPEAR-Apathy subscale—Among the 6 characteristics of environmental stimulation, stimulation clarity and stimulation strength were the only 2 significant factors affecting apathy scores; ambience, crowding, staff familiarity, light, and sounds did not show significant effects on apathyJoosse [], 2011Noise level, number of persons in proximity, spatial environment—SoundPro DL 2 (Quest Technologies) sound level meterAgitationWisconsin Agitation Inventory—After controlling for potential confounding variables of mental status, hearing impairment, and visual impairment, sound was found to contribute to agitationKim et al [], 2021Light level——Depression, sleep, overall level of BPSDKNPI-Qs, CSDD-Kt, PSQI, VAS-GVu, VAS-GAvActiwatch 2 (Philips Respironics)The study findings suggest that morning blue-enriched light therapy has a benefit in improving sleep and cognitive function in patients with Alzheimer diseaseKolberg et al [], 2021Light level——Agitation, affect, anxiety, delusion, depression, hallucination, mood, circadian rhythm (rest-activity pattern), sleep, overall level of BPSDCSDD, NPI-NH—Compared to the control group, the intervention group had a larger reduction on the composite scores of both the CSDD and the NPI-NH, as well as on the NPI-NH Affect subsyndrome and the CSDD mood-related signs subscale at follow-up after 16 weeksKonis et al [], 2018Light level—Digital charge-coupled device spectrometerDepression, overall level of BPSDCSDD, NPI-NH—Participants in the daylight intervention experienced an average decrease over the trial in the NPI-NH and CSDD scores, while the control participants showed average but nonsignificant increases in both NPI-NH and CSDD scores; difference in outcome changes of the intervention group achieved statistical significance for the CSDD but not for the NPI-NHLee et al [], 2016Number of persons in proximity, spatial environment, social environment, crowding, light levelTESS-NHw—Mood, social engagementMOSESx—Residents who stayed in the traditional, large-scale unit showed significantly worse decline in irritable behaviors compared to those who stayed in the small-scale, home-like unit across time; the small-scale, home-like unit had significantly better ratings in stimulation (lighting, visual and tactile stimulation, and noise) as well as personalization according to the TESS-NHLin et al [], 2018Noise level—TES-1350A sound level meterAgitationCMAI—The research results showed that agitated behaviors decreased significantly in the experimental group; in the comparison group, agitated behaviors decreased insignificantlyLiu et al [], 2021Light level——Overall level of BPSDNPIAccelerometer (XA-5)The NPI scores, which were derived using generalized estimating equation with medication (benzodiazepines) as a covariate, were significantly reduced by the fifth and ninth weeks; a lower NPI score indicates less severe BPSDLiu et al [], 2022Light level——Circadian rhythm (rest-activity pattern), sleepSleep chartsAccelerometer (XA-5)The experimental group showed significantly improved sleep efficiency, sleep duration, and awakening time from baseline to the fifth and ninth week, which was higher than the improvement in the comparison group; the number of nighttime awakenings decreased in the experimental and comparison groups; for circadian rhythm, the experimental group showed significant improvement in sleep onset and sleep offset, which were higher than the improvements in the comparison groupMcCurry et al [], 2011Light level——SleepSDIMicro Mini Motionlogger actigraph (Ambulatory Monitoring, Inc)This study found that exposure to light or increase in walking can have a benefit regarding nighttime sleep; this type of treatment is implementable by caregivers to increase the well-being of those in long-term care facilities; patients who adhered to treatment best saw the best resultsOlsen et al [], 2016Light level—ActiSleep+ (ActiGraph)Sleep, quality of life, social contact, activity—ActiSleep+ (ActiGraph)People living with dementia in nursing homes had lower quality of life in comparison to those at home, while nursing homes were associated with lower light level; however, in a regression model, residency was the only significant predictor for predicting quality of life in people with moderate dementia after controlling for confounders, including exposure to light levelOnega et al [], 2016Light level——Agitation,