sleep in hospitalized elders: a pilot study

FEATURE ARTICLE

Sleep in Hospitalized Elders:A Pilot Study

Kathy Missildine, PhD, RNNancy Bergstrom, PhD, RN, FAAN

Janet Meininger, PhD, RN, FAANKathy Richards, PhD, RN, FAAN

Marquis D. Foreman, PhD, RN, FAAN

Hospitalized elders frequently experience dis-

turbed sleep related to environmental factors.

To determine relationships between sleep

and environmental noise and light, a descrip-

tive exploratory study was conducted with 48

hospitalized older adults. Participants aged 70

years or older were monitored for sleep via

wrist actigraphy, and noise and light levels

were measured the first night of hospitaliza-

tion. Sleep time was brief (mean, 3.75 hours)

and fragmented (mean, 13 awakenings per

night). The sleep environment was noisy

with a median sound level of 49.65 dB(A).

There was an average of 3 periods of elevated

light levels (mean, 64 lux) lasting an average

of 1.75 hours each night. No significant corre-

lation was found among sleep and age, light,

and sound. Recommendations include light

and sound reduction measures and dedicated

‘‘do not disturb’’ times to allow for a full 90-

minute sleep cycle. (Geriatr Nurs 2010;31:

263-271)

Hospitalizations of persons aged $65 yearshas escalated in the United States over thepast 4 decades. In 1970, 20% of all inpa-

tients were aged $65 years, with the number ris-ing to 38% percent by 2006.1 Sleep complaints arecommon among elders at home and in thehospital, and these include insomnia, difficultyfalling sleep, and difficulty maintaining sleep,2,3

yet adequate sleep is commonly considereda necessary prerequisite to healing and recoveryfrom illness. Sleep in the hospital has beenlargely limited to studies in the critical careunits rather than on general medical surgicalunits, and most studies have been limited tosubjective measures of sleep.2,3 The purpose ofthis study was to describe objectively andsubjectively the sleep of elders hospitalized ona general medical unit and to determine the link

Geriatric Nursing, Volume 31, Number 4

between sleep and night-time noise and lightlevels.

Sleep of Elders in the Hospital

Hospitalized patients often complain of an in-ability to sleep,3,4 but objective data regardingsleep characteristics are limited. Sleep, oftenstudied in intensive care units, is characterizedby sleep deprivation and fragmentation. Lighterstages of sleep predominate, with reduced rapideye movement (REM), delta sleep, and frequentawakenings. Daytime napping is often used tocompensate for night-time sleep loss.4 Even ifsleep in the hospital is sufficient in totality, sleepfragmentation can impair cognitive function sim-ilar to the effects of total sleep deprivation.5

Sleep in the hospital may also reflect age-related sleep changes. Usual sleep patterns ofhealthy older adults differ from that of youngeradults,6 with decreased total sleep time (average5.5–6.5 hours of sleep nightly, compared with 7.5–8.5 hours nightly for younger adults) and lesstime in the deeper stages of sleep. Sleep effi-ciency, defined as the percent of time in bed actu-ally spent sleeping, is also less than in youngeradults. Chronic sleep difficulties in elders includ-ing insomnia and sleep-disordered breathing areassociated with cognitive impairment,7 increasedfall risk,8 impaired respiratory functioning,9 nurs-ing home placement,10 cardiovascular events,and increased mortality.11

Sleep Environment

The sleep environment, especially levels of am-bient noise and light, is important to good sleephabits. By nature of their 24-hour-care provisionroles, hospitals are particularly prone to sleepdisruption because of noise and light. Busch-Vishniac12 and colleagues measured noise levelson 5 units of Johns Hopkins Hospital in

263

Baltimore, Maryland, at various times of the dayand found that noise levels exceeded the currentWorld Health Organization (WHO) guidelines.13

Noise levels on these units did not appreciablydecrease at night. Light levels may also interferewith sleep by modifying the circadian system.14

No recent studies were located that investigatedthe effect of noise or light on the sleep of adultsin the hospital setting.

An understanding of interrelationships amongsleep patterns, ambient light, and sound mightcontribute to the development of nursing proto-cols that enable sleep in hospitalized elders.This descriptive pilot study was designed todescribe sleep time, nocturnal sleep efficiency,subjective sleep quality, and mean number andduration of nocturnal awakenings of elders thefirst night of hospitalization. The study wasfurther designed to explore the relationshipsbetween sleep, age, gender, room type, andnight-time noise and light levels.

Methods

Design

A descriptive exploratory pilot study was con-ducted on a medical unit in each of 2 acute carehospitals in the south-central United States, insummer 2008. The units were similar in size andnurse-to-staff ratios, and patients had similardiagnoses and length of stay. The study wasapproved by the institutional review boards ofthe hospitals and the university.

Recruitment and Sample

A convenience sample of 48 participants meet-ing inclusion criteria was recruited on the day ofadmission to the hospital. The inclusion criteriawere medical patients, aged $70 years, and an an-ticipated length of stay of 72 hours or longer onthe basis of admitting diagnosis, as determinedfrom unit statistics for the previous year. Patientswere excluded on the basis of pre-existing delir-ium (as determined by the Confusion AssessmentMethod [CAM]),15 dementia (determined via phy-sician records or caregiver statement), tremors orparalysis, poor vision or hearing (inability to hearverbal explanations, inability to read consentform), pre-existing sleep disorders, and inabilityto speak English. Subjects meeting eligibility re-quirements were recruited by the admitting nurse

264

during the unit admission process, who asked forpermission for the researcher to approach thesubject. The researcher then provided further ex-planation and obtained informed consent. Theability of these elders to give informed consentwas determined by their ability to correctly an-swer 3 questions:16 1) What is the purpose of thestudy? 2) What is required of you during the study?3) How can you remove yourself from the study?All subjects correctly answered these questionsand provided their own consent. Seventy-threepatients were approached, 58 met inclusion crite-ria, and 10 chose not to participate, citing longwait times in the emergency department andfatigue.

Procedures

Demographic data were obtained via patient orfamily interview and hospital record review. Anactigraph, the MicroMini Motionlogger (Ambula-tory Monitoring, Ardsley, NY), was worn on thedominant wrist continuously for 24 hours to mea-sure the sleep variables, beginning at 7 PM the dayof admission. Although an actigraph is usuallyplaced on the nondominant wrist, an adjustmentwas made for this study because vascular accesswas typically initiated in the nondominant upperextremity. The actigraph was moved to the otherextremity if the limb was immobilized. The micro-processor of the actigraph stored the data, whichwere downloaded into a personal computer. Alight sensor (Extech Datalogging Light Meter,Waltham, MA) and a noise dosimeter (Extech Per-sonal Noise Dosimeter) were secured behind thehead of the patient’s bed. The instruments col-lected data overnight (11 PM–7 AM), which weredownloaded onto a computer daily. These timeswere chosen on the basis of a preliminary study(unpublished data). On the evening of the secondday, the Richards Campbell Sleep Questionnaire(RCSQ)17 was administered to all participants.

Measures

The RCSQ17 is a visual analogue scale of 5items to measure subjective sleep quality. Theitems assess sleep depth, falling asleep, awaken-ing, return to sleep, and sleep quality the previousnight. The individual places a mark along a 100-mm line, indicating the score for each category,for a total of 5 marks. The instrument was scoredby measuring the length of each line to the


participant’s mark in millimeters, determiningthe total of the marks, and dividing by 5 for a totalsleep score. Scores ranged from 0 (lowest possi-ble score) to 100 (best possible score). Reliabilityand validity for the RCSQ was established in 70critically ill male patients; reliability for the scalewas .90 (Cronbach’s alpha) and principle compo-nents factor analysis revealed a single factor (Ei-genvalue 5 3.61, percent variance 5 72.2. Validitywas .58 (P \.001) as a correlation among thepolysomnography (PSG) sleep characteristic,sleep efficiency index, and the total RCSQ score.

The wrist actigraph, a movement sensing de-vice that indirectly measures sleep by detectingabsence of movement, is useful in environmentsin which more sophisticated methods of sleepmeasurement would be difficult. Movement wassampled several times per second and stored in1-min epochs. The digitized data were stored nu-merically and downloaded to a computer. Thezero-crossing mode was used to count the num-ber of times per epoch that the motion signalcrossed zero.18 The validity of actigraphy, com-pared with the gold standard PSG, differs accord-ing to population, presence of sleep disorders,and availability of a sleep diary. In a study in nurs-ing home populations without available sleepdiaries, the correlation between actigraphy andpolysomnography varied between 0.81 and 0.91for total sleep time and 0.61 and 0.78 for sleepefficiency.19 Reliability studies20 between the 2devices for nocturnal data collection were satis-factory (0.60–0.96), and sleep–wake agreementswere between 93% and 99%. Percent agreementin scoring night-time sleep among the actigraphsused for this study ranged between 92% and97%, obtained by simultaneous 24-hour recordingof reference and test actigraphs on the samewrist.

Datalogging instruments (Extech Instruments)were used to record nocturnal light and soundlevels from 11 PM–7 AM, downloaded to a computerfor analysis. Both instruments had �3% accu-racy.21,22 Percent agreement for this studyranged between 92% and 95% (sound meters)and 91% and 96% (light meters) obtained by 8-hour simultaneous recording of reference andtest meters. Level of night-time light wasdefined as the mean level of light (in lux, or 1lumen per square meter). The level of night-timesound was defined as the median level of soundin decibels (dB[A]), a measure of sound asheard by the human ear.


Data Analysis

Sleep characteristics were calculated from theactigraph data, using Action-W, Version 2software (Ambulatory Monitoring) and theCole–Kripke scoring algorithm, which comparesactivity levels of the current minute with thelevels of the preceding 4 min and the following2 min. Mean nocturnal sleep time, nocturnal sleepefficiency, number and duration of nocturnalawakenings, length of daytime inactivity (definedas those periods scored as sleep during the hoursof 7 AM–11 PM), light levels, and quality of sleep viathe RCSQ were analyzed using descriptive statis-tics. A light level .10 lux was used as the cutofffor elevated light levels, based on a preliminarystudy (unpublished data). The Chi-square statis-tic was performed to determine the associationbetween light and sound levels of good sleepersversus poor sleepers, using a 300-min cutoff.Three-hundred minutes was used as a cutoff forgood and poor sleepers, based on the effect ofpartial sleep deprivation on cognitive function.23

The Wilcoxon rank-sum test for independentgroups was used to compare light and soundlevels with sleep characteristics of participantsin private and semiprivate rooms and to comparenight-time sleep by gender. Pearson correlationcoefficients were used to compare sleep vari-ables with demographic variables and soundand light levels. Significance levels were set a pri-ori at P \.05. Data were analyzed using SPSS forWindows, Version 11 (SPSS, Chicago, IL).

Multiple regression was conducted to deter-mine whether the level of nocturnal noise or noc-turnal light was associated with the quantity andquality of sleep. Age, median noise levels indB(A), and mean light levels (lux) were enteredas continuous independent variables and totalnight-time sleep as a continuous dependent vari-able. Median noise levels were used rather thanmean, because sound levels have been convertedto the A decibel scale and are logarithmic num-bers, so measures of central tendency requiringcalculations are problematic.24

Results

The sample of 48 participants consisted of 24men and 24 women, average age 79 years.Forty-two participants (92%) wore glasses; fifteen(34%) had hearing impairment, and 42 were Cau-casian (92%). An average of 2.8 (SD 5 2.03; range

265

0–6) newly prescribed medications were admin-istered to participants during hospitalization.The most common primary diagnosis (n 5 10,20%) was heart failure, followed by syncope(n 5 6, 12%). Average comorbidities were 3.Two participants reported pain, each scoringthe pain 1–2 on a 0- to 10-point scale (0 5 no

pain, 10 5 worst pain imaginable, per unitprocedure).

Sleep Characteristics

Sleep time was brief and characterized by mul-tiple awakenings. Mean nocturnal sleep time was3.75 hours (225 min, SD 137 min; range, 0–446min), at 46% (n 5 48, SD 28%; range, 0%–93%)sleep efficiency, with considerable individual var-iability. Participants awakened an average of 13times per night and remained awake for an aver-age of 45 min (SD 80; range, 5–480 min). Individ-ual sleep episodes averaged 19 min. Participantsexperienced 20 daytime periods of inactivity, last-ing a mean of 24 min. No significant correlationwas evident between total night-time sleep min-utes and age, comorbidities, or number of newlyprescribed medications. No significant associa-tion was found between total night-time sleepand night-time light or sound. Participants scoredsleep quality on the RCSQ as 50.7, about midpointof the scale, with a moderate correlation with

Table 1.Night-time Sleep by Gender

Female (n 5 24)

Variable Median

Sleep efficiency 58

Total sleep minutes/24 hours 551

Total sleep minutes at night 297

Duration longest sleep episode

at night (minutes)

68

Number of wake episodes at

night

13

Total wake minutes/24 hours 890

Total wake minutes at night 183

Longest wake episode at night

(minutes)

59

RCSQ 4.06

RCSQ 5 Richards Campbell Sleep Questionnaire.

*P \ .05.

266

total night-time sleep minutes (r 5 .577, P \.01)and with number (r 5 .355, P 5 .011) and durationof wake episodes (r 5 –.289, P 5 .042).

There was a significant trend toward bettersleep in women (P 5.04; Table 1). Womenobtained longer periods of undisturbed sleepminutes (P 5 .05). When awakened, womenreturned to sleep more quickly, as reflected infewer total awake minutes and shorter individualwake periods. No significant correlation wasfound between sleep and age.

Differences between sleep and light and soundlevels in private and semiprivate rooms, althoughnot statistically significant, did show a trendtoward better sleep in private rooms. Lights-ontimes in private and semiprivate rooms wereequal during the night, but light levels in privaterooms were brighter (Table 2). Participants inprivate rooms were subjected to higher maxi-mum sound levels and a trend toward highermean light levels but had longer periods of unin-terrupted sleep and slightly better sleep overall.Good sleepers, those who slept more than 5hours per night, were more likely to be in a privatethan a semiprivate room (Table 3).

Night-time Light and Sound

Both private and semiprivate hospital roomswere noisy (Table 2). Night-time light levels

Male (n 5 24)

Median

Percentile

25th–75th

Wilcoxon

Rank Sum P

39.5 22–71 525.5 .14

318 214–455 494 .03*

168.5 123–223 497.5 .04*

47.5 47–127 503.5 .05*

14 8.5–15 612 .79

1093.5 784–1225 537 .08

286 139–357 530 .06

84.5 34–131 555 .04*

6.1 2.6–6.8 526 .07


Table 2.Night-time Light, Sound, and Sleep Characteristics: Private or SemiprivateRoom

Variable

Private

(n 5 34)

Semiprivate

(n 5 14)

Percentile

25th–75th

Wilcoxon

Rank-Sum P

Total minutes of night-time

light .10 lux

38 38 16–175.8 863 .79

Mean intense light, lux 48 25 14–87.8 279 .12

Maximum light, lux 86 49.5 20.8–130.5 281 .13

Mean overall light (lux) 11.5 5 2.6–24.8 311 .47

Median sound, dB(A) 49.6 52.2 48.7–54.3 820 .22

Maximum sound, dB(A) 73 70.2 67–72 273 .09

Total sleep minutes 243 221 29.8–360.5 344 .89

Sleep efficiency (%) 39.5 51 6.3–67.3 314 .43

Mean duration of sleep

episode (minutes)

17 17 4.8–22.8 332 .11

Duration of longest sleep

episode (minutes)

61 45.5 7–79.8 278 .11

RCSQ 50.02 40.2 4.2–9 324 .67

RCSQ 5 Richards Campbell Sleep Questionnaire.

were low with an average of 3 periods of elevatedlight levels (mean, 64 lux) lasting an average of1.75 hours. With the exception of median soundlevels (51 vs. 44 dB[A], P\.01], no association be-tween light and sound levels and sleep by hospi-tal site was identified. Participants werestratified according to total nocturnal sleep min-utes, mean light levels, and median sound levelsto determine whether light and sound levelsaffected ability to sleep. Peak sound levels of 40dB(A), as recommended by the WHO13 and themean overall light level for this study (13.86

Table 3.Sound Associated with Good and Poor S

Low Noise

Poor sleepers

Count 2

Percent of total sleepers 4%

Good Sleepers

Count 2


Totals

Count 4


Pearson Chi-square, 1.14, df 5 1, P 5 .29.

Poor sleeper: 300 minutes or less; good sleeper: more than 300 m


lux), were used as cutoff levels. The results(Tables 3 and 4) demonstrate a nonsignificanttrend toward a brighter and noisier environmentfor poor sleepers.

Standard multiple regression did not indicatean associate among age, light or sound, and sleepefficiency (R2 5 .047, R

2 adjusted 5 –.015, F

5.756, P 5 .524; Table 5). Power analysis (nQueryAdvisor, Statistical Solutions, Saugus MA)revealed that the study (n 5 48) was underpow-ered, having only 26% power to detect an R

2 of0.058 with a medium effect size (0.6), a 5 0.05.

leepers

High Noise Totals

28 30

59% 63%

16 18

33% 37%

44 48

90% 100%

inutes; high noise: $40 dB(A); Low noise: \40 dB(A).

267

Table 4.Light Levels Associated with Good and Poor Sleepers

Low Light High Light Totals

Poor sleepers

Count 15 15 30

Percent of total sleepers 33% 31% 64%

Good Sleepers

Count 12 6 18


Totals

Count 27 21 48


Pearson Chi-square, 1.37, df 5 1, P 5 .24.

Poor sleeper: 300 min or less; good sleeper: more than 300 min; high light: $13.86 lux; low light: \13.86 lux.

Discussion

Sleep of these hospitalized elders was oftendisturbed on the first night of hospitalization.Sleep efficiency of the sample (46%, n 5 48)was less than the usual 79% efficiency for a 70year old,25 possibly related to the ‘‘first night ef-fect’’ of sleeping in a different environment or toeffect of acute illness. Arbitrary sleep times (11PM–7 AM, based on a pilot study) might have hadan effect, because many participants fell asleeplater and slept later in the morning. Accordingto Bliwise,25 older adults have an earlier bedtimeand wake-up time than younger adults, but thisusual routine may have been disrupted by illnessand hospitalization.

The average longest sleep episode of an hourwas less than the needed 90 min to completea full cycle of sleep and was perhaps age-related. Redline and colleauges6 demonstratedan age-related increase in arousals via PSG (notnecessarily resulting in awakenings) from 18per hour for ages #54 and to 21 for ages $70.Sleep fragmentation may have been linked tosemiprivate accommodations and ‘‘first night’’

Table 5.Summary of Regression for Variables inEfficiency

B

Age –3.46

Mean light, lux –.805

Median sound, dB(A) –.671

268

effect of sleeping in a strange environment. Noassociation was found among sleep, pain, andthe number of comorbidities.

A nonsignificant trend toward better sleep effi-ciency in women may be explained, in part, bywomen’s shorter wake periods, implyinga quicker return to sleep after awakening anda longer duration of individual sleep episodes.This was consistent with other studies, in thatwomen sleep longer than men but spend lesstime in slow-wave delta sleep, the stage of sleepthought to be the most vulnerable to environmen-tal disturbances.26

Light and Sound

Median sound levels were consistently elevatedand displayed intermittent spikes to even higherlevels. Although not statistically significant, poorsleepers were more likely to have experienceda noisier environment than good sleepers (Table3). Overall median sound was equal to the levelof an urban residence in the daytime, with peaklevels of 97.3 dB(A) equal to that of city traffic.27

This exceeds the recommendations by WHO13 of

the Prediction of Night-time Sleep

b t P

–.185 –1.28 .207

–.099 –.679 .500

–.031 –.209 .836


35 dB(A) at night for patient rooms, with peaklevels no higher than 40 dB(A). The findingswere consistent with a study at Johns HopkinsHospital12 in which sound levels varied from 35to .70 dB(A). Historically, sound levels in hospi-tals have been increasing an average of 0.42 dB(night-time) per year over the past 40–50 years,and Busch-Vishniac and colleagues12 postulatedthat part of the explanation for the increase ishigh velocity air flow of modern heating/coolingsystems through older ductal systems. It is impor-tant to note that as the dB(A) scale is logarithmic,a sound pressure level increase of 10 dB(A) is per-ceived by the human ear as twice as loud.24 Soundlevels did not significantly correlate with totalsleep minutes or with frequency and length ofawakenings. Subjects may have felt less restric-tion in television viewing and so on because ofthe lack of a roommate, explaining the highersound and light levels in private rooms. Consis-tently elevated baseline sound levels may havecreated a ‘‘white noise’’ effect in the study popula-tion, muffling the disruptive influence of other fac-tors on sleep.28 Care activities may have beenmore disruptive of sleep than sound levels, and re-sults may have been affected by the 33% (n 5 16)of participants who reported hearing difficulties.

Light levels were similar to those obtained ina study by Higgins and colleagues,29 in whichmean overnight light levels ranged from 13.3 to15.1 lux in a critical care unit. Differences maybe related to architectural design of the intensivecare units and the private and semiprivate roomsin this study, placement of the meters, differinglight fixtures, and variability in patient acuity, re-quiring more nursing care procedures.

This pilot study was well tolerated by partici-pants; only 1 participant reported sleep disrup-tion due to study equipment. Actigraph, light,and sound meters were well tolerated byparticipants and were deemed to be appropriateinstruments for this type of study. Sleep was ab-breviated and marked by frequent and prolongedawakenings but only weakly correlated with thelevel of ambient light and sound. Participantswho slept longer were more likely to be satisfiedwith their sleep.

Limitations

This study had several limitations. Recruit-ment was sometimes difficult because many po-tential subjects had spent several hours in the


emergency department before admission to theunit and were quite fatigued, creating a possibleselection bias. Ten subjects approached de-clined to participate. Actigraph readings withdaytime periods of inactivity may reflect nap-ping but may also reflect removal of the acti-graph for diagnostic procedures. For thisreason, daytime napping was not measured inthis study. Participant logs chronicling activity,subject-selected bedtimes, daytime napping, re-moval of the actigraph, and consideration ofpsychological factors affecting sleep would con-tribute to a stronger contextual base for inter-pretation of findings. A diary or log was notused for this study because of inconsistent useby participants in a preliminary study (unpub-lished data). The staff of the units were notblinded to the study, so the presence of the acti-graph and data logging instruments may have af-fected light and noise levels, as well as thefrequency and duration of nursing care proce-dures during the night. The study was limitedto determination of sleep on only the first nightof hospitalization. Extended monitoring overa number of nights would compensate for the‘‘first night effect’’ and reveal patterns of sleepover time. Researchers should determine thetypes of medications, diagnoses and severity ofillness, and presence of pain, all of which canaffect sleep. Sleep in the hospital could be com-pared with usual sleep at home, because poorsleep may be an established pattern. An exami-nation of sources of light and sound, octaveband sound pressure levels, and the establish-ment of controls for acoustical elements wouldresult in data helpful in planning noise and lightreduction measures.

This study was underpowered. To determinethe association of total minutes of night-timesleep with age, environmental noise and lightlevels, a sample size of 182 is needed to have80% power to detect an R

2 of .0580 using the mul-tiple linear regression test of R

2 5 0 (a 5 0.050)for 3 normally distributed covariates.

Implications for Research and Practice

It is important to study sleep in elders in theacute care setting, especially in areas of thehospital not previously studied, because disrup-tions to sleep might vary by unit. Usual sleeppatterns should be assessed on admission andpreemptive nursing measures taken to simulate

269

or improve usual bedtime routines. Adequacy ofsleep during hospitalization should be assesseddaily and measures taken to remedy sleep dis-turbances. Because a full sleep cycle of 90min can have a positive influence on sleep effec-tiveness, nurses might design unit activities totest a specific ‘‘do not disturb’’ period sufficientto allow for the patient to experience a fullsleep cycle.30 Involvement of staff in measuresto reduce light and sound levels at night willcreate an environment more conducive to sleep.Researchers could then determine the effect ofthese measures on overall sleep, mood, cogni-tive functioning, and illness-related outcomes.

In conclusion, the importance of a good night’ssleep does not disappear just because a person isadmitted to the hospital. The benefits of sleep canfacilitate positive patient outcomes, and sleeppromotion should be studied in hospitalized pa-tients. The vulnerability of the elderly and theirpredisposed sleep disturbance issues makethem a perfect target population. This studywas a beginning effort to ensure sleep of eldersas they seek to regain optimal health in a hospitalsetting.

References

1. DeFrances CJ, Lucas CA, Buie VC, et al. 2006 National

Hospital Discharge Survey. 2006. Available at http://

www.cdc.gov/nchs/data/nhsr/nhsr005.pdf. Cited

September 3, 2009.

2. Birath JB, Martin JL. Common sleep problems affecting

older adults. Ann Long Term Care 2007;12:20-6.

3. Humphries JD. Sleep disruption in hospitalized adults.

Medsurg Nurs 2008;17:391-5.

4. Redeker N. Sleep in acute care settings: An integrative

review. J Nurs Sch 2000;32:31-8.

5. Durmer J, Dinges D. Neurocognitive consequences of

sleep deprivation. Sem Neuro 2005;25:117-29.

6. Redline S, Kirchner H, Quan S, et al. The effects of age,

sex, ethnicity, and sleep-disordered breathing on sleep

architecture. Arch Int Med 2004;164:406-18.

7. Cole C, Richards K. Sleep and cognition in people with

Alzheimer’s disease. Issues Ment Health Nurs 2005;26:

156-64.

8. Stone K, Ensrud K, Ancoli-Israel S. Sleep, insomnia and

falls in elderly patients. Sleep Med 2008;9:S18-22.

9. Stege G, Vos PJ, van den Elshout FJ, Dekhuijzen R, et al.

Sleep, hypnotics and respiratory difficulties. Respir Med

2008;102:801-14.

10. Rowe MA, Kelly A, Horne C, et al. Reducing dangerous

night-time events in persons with dementia by using

a nighttime monitoring system. Alzheimers Dement 2009;

5:419-26.

270

11. Martino TA, Sole MJ. Molecular time: an often

overlooked dimension to cardiovascular disease. Circ

Res 2009;105:1047-61.

12. Busch-Vishniac I, West J, Barnhill C, et al. Noise levels in

Johns Hopkins Hospital. J Acoust Soc Amer 2005;118:

3629-45.

13. Berglund B, Lindvall T, Schwela D, eds. Guidelines for

community noise. Geneva: World Health Organization;

1999.

14. Czeisler C, Buxton O, Khalsa S. The human circadian

timing system and sleep–wake regulation. In: Kryger M,

Roth T, Dement W, eds. Principles and practice of

sleep medicine. Philadelphia: Elsevier Saunders; 2005.

p. 375-94.

15. Inouye S, van Dyek C, Alessi C, et al. Clarifying

confusion: the confusion assessment method. Ann Intern

Med 1990;113:941-8.

16. Resnick B, Gruber-Baldini AL, Pretzer-Aboff I, et al.

Reliability and validity of the evaluation to sign consent

measure. Gerontologist 2007;47:69-77.

17. Richards K, O’Sullivan, Phillips R. Measurement of

sleep in critically ill patients. J Nurs Meas 2000;8:

131-44.

18. Ancoli-Israel S. Actigraphy. In: Kryger M, Roth T,

Dement W, eds. Principles and practice of sleep

medicine. Philadelphia: Elsevier Saunders; 2005.

p. 1459-67.

19. Ancoli-Israel S, Clopton P, Klauber M, et al. Use of wrist

activity for monitoring sleep/wake activity in demented

nursing home patients. Sleep 1997;20:24-7.

20. Sadeh A, Sharkey K, Carskadon M. Activity-based sleep-

wake identification: An empirical test of methodological

issues. Sleep 1994;17:201-7.

21. Extech Instruments. Datalogging light meter with PC

interface. 2004. Available at www.extech.com. Cited

March 18, 2006.

22. Extech Instruments Product Datasheet. Personal Noise

Dosimeters. 2004. Available at www.extech.com. Cited

March 18, 2006.

23. Dinges D, Pack F, Williams K, et al. Cumulative

sleepiness, mood disturbance, and psychomotor

vigilance performance decrements during a week of

sleep restricted to 4-5 hours per night. Sleep 1997;20:

267-77.

24. Gray L. Background science: properties of sound.

J Perinat 2000;20:S5-10.

25. Bliwise D. Normal aging. In: Kryger M, Roth T, Dement W,

eds. Principles and practice of sleep medicine.

Philadelphia: Elsevier Saunders; 2005. p. 25-38.

26. Latta F, Leproult R, Tasali E, et al. Sex differences in

delta and alpha EEG activities in healthy older adults.

Sleep 2005;28:1525-34.

27. Noise Pollution Clearinghouse sound levels and human

response. Available at www.nonoise.org. Cited August

31, 2009.

28. Stanchinca ML, Abu-Hijleh M, Chaudhry BK, et al. The

influence of white noise on sleep in subjects exposed to

ICU noise. Sleep Med 2005;6:423-8.

29. Higgins P, Winkelman C, Lipson A, et al. Light

measurement in the hospital: a comparison. Res Nurs

Heal 2007;30:120-8.

30. Gardner G, Collins C, Osborne S, et al. Creating

a therapeutic environment: a non-randomized controlled


http://www.cdc.gov/nchs/data/nhsr/nhsr005.pdf

http://www.cdc.gov/nchs/data/nhsr/nhsr005.pdf

http://www.extech.com

http://www.extech.com

http://www.nonoise.org

trial of a quiet time intervention for patients in acute care.

Int J Nurs Stud 2009;46:778-86.

KATHY MISSILDINE, PhD, RN, is an assistant professor at

University of Texas at Tyler, Tyler, TX. NANCY BERG-

STROM, PhD, RN, FAAN, is a professor and director for

Aging Research, at the Center on Aging, University of Texas

Health Science Center, Houston, TX. JANET MEININGER,

PhD, RN, FAAN, is a professor at University of Texas Health

Science Center, Houston, TX. KATHY RICHARDS, PhD, RN,

FAAN, is a director of Polisher Research Institute, and pro-

fessor at University of Pennsylvania, Philadelphia, PA.

MARQUIS D. FOREMAN, PhD, RN, FAAN, is chairman of

Adult Health and Gerontological Nursing, University of

Illinois, Chicago, IL.


ACKNOWLEDGMENTS

Kathy Missildine is the recipient of the Hogstel Gerontolog-

ical Nursing Research Award from the Beta Alpha Chapter

of Sigma Theta Tau International awarded by Texas

Christian University, a research award from the Iota Nu

Chapter of Sigma Theta Tau International awarded by

the University of Texas at Tyler, and an award from the

Speros Martel Foundation, awarded by the Center on

Aging, University of Texas Health Science Center at

Houston.

0197-4572/$ - see front matter

� 2010 Mosby, Inc. All rights reserved.

doi:10.1016/j.gerinurse.2010.02.013

271

sleep in hospitalized elders: a pilot study

Documents