Sleep quality in patients admitted to an intermediate care unit

Abstract

Introduction

Patients admitted to critical care units (ICU) are subject to develop sleep disturbances, caused by multiple factors that are present in this type of units.

Objetive

Describe patients' perceptions of night rest, to relate them to sociodemographic variables, and determine factors that interfere with their sleep quality.

Methodology

Descriptive, prospective, observational study in an ICU of a tertiary level university hospital. Sleep perception was assessed with the five-item Richards-Campbell Sleep Questionnaire (RCSQ). Sociodemographic and clinical variables were collected. A self-developed questionnaire with 9 questions addressing the main factors mentioned in the literature that influence sleep was utilized. Descriptive and inferential statistics were performed, being considered statistically significant p < ,05.

Results

The sleep of 75 patients was studied, for 146 nights.

The sleep perception was rated as fair, with moderate ease in falling asleep, average awakenings, and moderate ease in returning to sleep. The mean RCSQ score was 60.22 (SD: 24.81).

It was observed that concern (p < .001), noise (p = .016), pain (p = .008), discomfort (p = .001), ambient light (p = .026), and the presence of nearby patients (p = .027) significantly influenced in the sleep.

Conclusions

Patients’ perception of night sleep was fair. Keeping patients informed, minimizing ambient light and noise, optimizing analgesic guidelines, and promoting a comfortable position could facilitate night rest.

Resumen

Introducción

Los pacientes ingresados en unidades de cuidados críticos (UCIM) son susceptibles de desarrollar alteraciones del sueño, causadas por múltiples factores presentes en este tipo de unidades.

Objetivo

Describir las percepciones de los pacientes sobre el descanso nocturno, relacionarlas con variables sociodemográficas; y determinar factores que interfieren en la calidad del sueño.

Metodología

Estudio descriptivo, prospectivo, observacional en una UCIM de un hospital universitario de nivel terciario. Se evaluó la percepción del sueño con el Cuestionario del Sueño de Richards-Campbell (RCSQ) de cinco ítems. Se recogieron variables sociodemográficas y clínicas. Se utilizó un cuestionario autoelaborado de 9 preguntas con los principales factores mencionados en la bibliografía influyentes en el sueño. Se realizó estadística descriptiva e inferencial, considerándose estadísticamente significativo: P < ,05.

Resultados

Se estudió el sueño a 75 pacientes que generaron 146 noches.

La percepción del sueño fue regular, con facilidad media para conciliarlo, despertares medios y con facilidad media para volverse a dormir. El valor medio de la RCSQ fue de 60,22 (DE:24,81).

Se observó que la preocupación (p < ,001), el ruido (p = ,016), el dolor (p = ,008), el disconfort (p = ,001), la luz ambiental (p = ,026) y la presencia de otros pacientes cercanos (p = ,027), influyeron significativamente en el sueño.

Conclusiones

La percepción del sueño nocturno de los pacientes fue regular. Mantener informado al paciente, minimizar la luz y el ruido ambiental, optimizar la pauta analgésica, promover una posición cómoda, podría facilitar el descanso nocturno.

What is known

Many physiological processes, a sense of well-being, and recovery from illness involve sleep. In critically ill patients, sleep patterns are significantly altered by their condition and the care unit environment.

What it contributes

This study provides original results on a prevalent and understudied problem in intermediate care units: sleep quality and quantity, and the factors that most influence it. Understanding these aspects facilitates the design of interventions to improve sleep quality.

Introduction

Sleep is essential for generating the energy needed for recovery from illness.^1–3 In critical care units, sleep patterns are highly disrupted, characterised by severe fragmentation, frequent awakenings, an increase in light sleep, and a decrease in deep sleep.⁴ The loss of sleep leads to alterations which have negative impacts on the body.^1,3–7 These negative effects may be observed within 24–48 h² and are associated with an increased risk of morbidity and mortality,^1,6,7 and delays in the recovery process from illness.^1,8,9

The factors responsible for this alteration are: a) altered circadian rhythm^9–11 and an altered homeostatic response^5,10 to the need to sleep or be awake; b) the patient’s critical condition,^5,7,10,11 which can cause a range of symptoms such as nausea, pain, etc., that are responsible for sleep insecurity; c) fear⁷; anxiety,^5,10,11 and pain^7,10,11; d) medication side effects^5,10,11; e) immobility^7,10; f) monitoring devices^5,10,11 and nursing interactions^5,7,10,11 and g) ambient noise and light.^5,7,10,11

Critical care nurses provide holistic care, with sleep being an important area to address. They play a fundamental role in advocating, promoting, and preserving patients' sleep since they provide constant support throughout the patient’s stay.¹²

While sleep difficulties in the intensive care unit are widely documented, the literature on this intermediate care unit (IMCU) topic is limited. This can largely be attributed to their relatively recent establishment and the limited number of IMCUs worldwide. These units generally provide care to patients whose health status is too compromised to be admitted to a general hospital ward, but who do not require the intensive care required in an ICU.¹³

We therefore undertook this study to describe patients' perceptions of night-time rest in an IMCU, analyse their relationship with sociodemographic variables, and determine influencing factors that interfere with sleep quality.

Methodology

This prospective, observational, descriptive study was conducted in the IMCU of a tertiary-level teaching hospital with more than 200 beds. This 12-bed unit includes the Coronary Care Unit, the Stroke Unit, and a complex Internal Medicine area. Admitted patients do not undergo invasive techniques such as artificial ventilation or continuous renal replacement therapy. They also do not receive short-term mechanical circulatory support. The staffing ratio on the night shift is one nurse to two or three patients.

The study was conducted from March to July 2022.

The inclusion and exclusion criteria for selecting the sample for each night are presented in Table 1. All nights that patients were in the unit were studied. If a patient could not complete the survey because they did not meet the inclusion criteria that day, they were not excluded from the study, only from that night's assessment. This means that some nights of their stay may not have been assessed. The sample was selected by convenience.

Table 1.

Inclusion and exclusion criteria.

Inclusion criteria	Exclusion criteria
Patients who are able to talk and understand the questionnaire	Prior psychiatric treatment
Conscious and aware patients*	Impossibility of communication
	Patients who presented some mental disorder, cognitive deficit, and/or sensory deficit that makes correct administration of the questionnaire difficult
	Patients who presented confusional state during the night of admission, according to the results obtained on the CAM-ICU delirium scale
	Patients whose clinical condition prevented them from completing the questionnaire
	Patients who were admitted to the unit after 20.00 h.

CAM-ICU: Confusion Assessment Method for the Intensive Care Unit.

*

Information on patients' level of consciousness and awareness was obtained from the computerised medical record recorded at the end of the night shift and in the early morning.

As with Nicolás et al.¹⁴ and Carrera et al.,¹⁵ the night-time sleep phase was considered from 11:00 PM to 7:00 AM the following morning, since all patients are scheduled for care at 10:00 PM.

The rooms were independently arranged, with doors that completely enclosed the room and glass blinds. They had exterior windows that provide natural light during the day, with blinds and curtains to control its entry. The room light could be controlled by the patient via a remote control, as could the window blinds. All rooms had a clock on the wall to facilitate patient orientation. Open visiting hours facilitated family presence (11:00 AM to 2:00 PM and 5:00 PM to 9:00 PM). In addition, a 24-h telephone service was available. If necessary, a companion was allowed to stay with them on a regular basis.

Instruments used for data collection

The research team prepared a document consisting of three sections:

• 1

  Data from the patient's medical history and computerised Nursing Care Plan.

  • o

    Age, sex, length of stay in the unit, night of assessment, reason for admission.

  • o

    Care provided during the night, whether scheduled or requested by the patient (postural changes, treatment of ulcers, wounds, and drainage, non-invasive blood pressure monitoring, etc.) and the frequency of these.

  • o

    Night-time pharmacological treatment (sleep aids, muscle relaxants, anxiolytics, all types of analgesia (intravenous, oral, etc.), etc.)

• 2

  Richards-Campbell Sleep Rating Scale (RCSQ), translated into Spanish,¹⁴ to assess patients' subjective impression of sleep (Appendix I). It consists of five items developed on a visual analogue scale. These items analyse the sleep areas reflected in polysomnography studies. The patient evaluates each item on a ruler graduated from 0 to 100 mm. A value of 0 mm corresponds to the worst sleep, and a value of 100 mm corresponds to optimal sleep. The total score of the scale is the average of the scores of the five items. This scale was chosen because it is a validated test that presents high internal validity and reliability, with a Cronbach's alpha of .90 for both estimates.¹⁶ Furthermore, its application has been evaluated in the critically ill population, thus demonstrating its validity in this group.^14,16 Finally, it is a short and easy-to-complete test, as Ritmala-Castren et al. point out.¹⁷

Sleep was considered poor if patients scored between 0 and 33 mm, average between 34 and 66 mm, and good from 67 to 100 mm, as has been suggested by other authors.¹⁴

• 3

  The study authors developed an ad hoc questionnaire on the intrinsic and environmental factors previously identified in the literature as the main causes of sleep disruption. The patient assigned a numerical value to each item, with 0 being the lowest (no impact on sleep) and 100 being the highest (major impact on sleep) (Appendix II).

Data collection

The research team collected data over a four-month period. Patients were informed of the study's objective and importance, and their participation was requested. Patients who agreed to participate were given informed consent and a questionnaire to complete, ensuring that each item/question was understood. Upon request, patients received assistance with completion. A member of the research team collected the data from the computerised nursing care plan. The entire process was conducted during the early hours of the morning shift.

Data analysis

Statistical descriptions of quantitative variables were performed using mean and standard deviation (SD), and of categorical variables using frequency and percentage. Linear mixed-effects models were used to account for data dependencies. Statistical significance was established at p values ​​< .05. Statistical analyses were performed using Stata 14 (StataCorp. 2015. Stata Statistical Software: Release 14. College Station, TX: StataCorp LP).

Ethical considerations

Anonymity, complete data confidentiality, and destruction of questionnaires at the end of the study were guaranteed for all patients (Organic Law 3/2018, of December 5, on the Protection of Personal Data and Guarantee of Digital Rights). Approval from the Clinical Research Ethics Committee (Project 2022.056) and written consent from the participants were obtained.

Results

The sample consisted of 75 patients, of whom 24 (32%) were women and 51 (68%) were men. The mean age was 70.65 years (SD 15.01, min 64, max 93). The reason for admission was cardiac intervention in 14.67% (11 patients); cardiac involvement in 52% (39 patients); Neurological patients accounted for 26.67% (20 patients) and internal medicine patients for 6.67% (5 patients). These patients accounted for 146 nights. The average length of stay was 3.48 days (SD: 1.91, min: 1, max: 11).

The mean sleep score on the RCSQ questionnaire was 60.22 (SD: 24.81). The analysis of each item on the scale is presented in Table 2.

Table 2.

Scores obtained on the Richards-Campbell questionnaire (n = 146).

Item	Mean
Your sleep last night was: deep sleep/light sleep (sleep depth)	57.00
Last night the first time you fell asleep: fell asleep almost immediately/could not fall asleep (speed of falling asleep)	63.20
Last night: you barely woke up/were awake all night (number of awakenings)	57.70
Last night when you woke up: went back to sleep immediately/could not get back to sleep (percentage time awake)	60.00
Last night: you slept well/slept poorly (total sleep quality and perception)	63.00
Total score	60.22

Worry-nervousness had a score out of 100 for sleep: 34.62 (SD: 33.79); noise: 19.79 (SD: 26.81); staff voices: 10.03 (SD: 20.18); nighttime nursing interventions: 33.08 (SD: 20.54); pain: 15.17 (SD: 25.41); discomfort: 30.73 (SD: 29.34); ambient light: 9.11 (SD: 19.85); and the presence of other patients: 4.59 (SD: 16.79).

When analysing whether age and sex influence sleep quality, no statistically significant differences were found (p = .354 and p = .993, respectively). The same is true for the reason for admission (Table 3) and length of stay (p = .913).

Table 3.

Relationship of the reason for admission with the score obtained on the Richards-Campbell questionnaire.

Reason for admission	B-Coefficient	95% confidence interval	p	Richards Campbell Scale mean score
Interventional Cardiology (n = 11)	Reference	Reference	Reference	55.09 (SD 21.10)
Cardiac involvement (n = 80)	4.42	–11.26 to 20.12	.580	60.08 (SD 25.98)
Neurological (n = 42)	7.51	–9.22 to 24.24	.379	62.571 (SD 22.34)
Internal medicine (n = 13)	1.34	–19.94 to 22.63	.901	57.73 (SD 29.40)

There was no association with medication use (p = .467). The score for patients taking medications that could influence sleep was 58.05 mm (SD 25.11) versus 63.58 mm (SD 24.15) for those who did not.

The analysis of the items collected in the second questionnaire showed that patient worry/nervousness; the noise of monitor and device alarms; patient-reported pain; discomfort with drains, monitoring cables, or a strange bed; ambient light at night, as well as the presence of other patients nearby, significantly influenced sleep (Table 4). No statistically significant differences were found with staff voices and nursing interventions on the night shift (p = .226 and p = .376 respectively).

Table 4.

Relationship of factors with scores obtained on the Richards-Campbell questionnaire.

Variable	B-Coefficient	Confidence interval	p
Worry/nervousness	–.23	–.35 a –.12	<.001
Noise of alarms and other devices	–.17	–.32 a –.03	.016
Staff voices	–.12	–.33 a –.07	.226
Nursing interventions	–.09	–.29 a .11	.376
Pain	–.21	–.36 a –.05	.008
Discomfort	–.23	–.37 a –.09	.001
Ambient lighting	–.23	–.43 a –.02	.026
Presence of other patients	–.26	–.49 a –.02	.027

Discussion

The mean score obtained from the RCSQ scale reveals that patients in the intermediate care unit have a regular perception of sleep, with moderate ease in falling asleep, moderate awakenings, and moderate ease in returning to sleep. The literature reports that in critically ill patients, sleep disruption is severe, and sleep is characterised by fragmentation, abnormal circadian rhythms, increased light sleep, and decreased deep sleep.¹⁸ In the reviewed literature, RCSQ scale values ​​ranging from 25.10 to 59.66 mm were found.^{4,6,8,15,17,19–24} Only two studies reported a higher score than the one in this study.^6,22 These lower scores in the literature may be due to the fact that the studies described were conducted in more technologically advanced settings and with different architectural structures. The higher score is justified by the fact that it was the score of the intervention group in a study to improve night-time rest in critically ill patients.

Regarding the demographic characteristics of the patients (age and sex), the resulting data must be analysed with caution, as discrepancies were found in the reviewed literature. Regarding sex, some articles report no significant differences in perceptions of sleep quality,^2,8,9,15,25 as in this study. Other studies, however, describe women^19,26 or men¹⁷ as better sleepers. Men and women have different sleep patterns. Although there is considerable knowledge about the mechanisms that regulate sleep, the reasons behind these gender differences in sleep behaviour remain unknown and poorly researched.²⁷

Some authors point out that age negatively influences both the quality and quantity of sleep. The circadian rhythm is altered with age, resulting in increased awakenings and delayed sleep onset.²⁸ Two studies were found in which older patients slept better in the ICU than younger patients.^24,26 This study found no statistical association with age, as other studies did.^{9,14,15,22,28}

This study found no statistically significant differences in relation to the reason for admission, similar to the study by Czempik et al.²⁵ and Carrera et al.¹⁵ and unlike the majority of the literature reviewed. ^5,7,10,18,20 It has been reported that the severity and intensity of the disease itself can also contribute to sleep disorders.^{5,10,12,20,22,29} Other authors, however, point out that, regardless of the reason for admission, the disease influences sleep because it generates a feeling of threat to life due to the fear and uncertainty experienced by the patient.²⁰

When analysing whether taking medication that facilitate sleep conditions influenced sleep, no association was found. This finding was similar to the studies by Navarro et al.⁸ and Carrera et al.,¹⁵ but unlike that of van der Hoeven et al.²⁴ who observed that patients who took these medications slept worse. Giusti et al. stated that analgesics, sedative therapy, and the side effects of the drugs can worsen sleep conditions.¹⁰ Nonsteroidal anti-inflammatory drugs can negatively affect sleep, increasing night-time awakenings and decreasing sleep efficiency.³⁰ Furthermore, although benzodiazepines have been shown to increase total sleep time, they result in abnormal sleep structure, increasing the stages of light sleep and reducing deep sleep.⁵

Regarding whether worry-nervousness influenced sleep perception, statistically significant differences were found, as in the literature reviewed.^{4,5,11,15,19,29} The literature describes that concern about health loss is related to the patient's lack of knowledge about their illness, diagnostic methods, treatment, and technical language. Inadequate understanding by the patient and their family³¹ can be confusing or cause anxiety. Informing the patient at all times about their concerns and explaining any technique before performing it would help minimise their stress and could facilitate a better night's rest.^28,31

The noise from monitor alarms and other devices significantly influenced sleep, as reported in the literature.^{3,5,8,10–12,15,19–21,23,32,33} The voices of healthcare professionals did not, as reported in other publications.^5,8,15,21,26 Sound levels in critical care areas are reported to continue exceeding WHO noise recommendations.³⁴ A study by Simons et al. on the association between various acoustic parameters and subjectively perceived sleep quality in ICU patients found that noise was responsible for 11%–24% of the total number of awakenings, and that patients themselves considered noise in the ICU to disturb their sleep.¹⁹ Although the surroundings of this unit were conducive to rest and the influence of noise on sleep was low, it is still necessary to eliminate harmful acoustic stimuli. Turning off suction sources when not in use, minimising artifact alarms, lowering the alarm volume at night, etc. are simple and low-cost interventions that could improve patients' sleep.^18,32,33 Earplugs have also been shown to promote better night-time rest.^4,5,9,21,22

The literature reports that pain is a barrier to sleep,^{3,5,8,11,15,18} a finding that is consistent with this research. Grossman et al., in their study to understand how patients sleep and identify potential disruptions, found that pain was one of the most commonly reported disruptions and led to the greatest reduction in sleep duration.³ Although patients' perception of the influence of pain was low, its significance was impactful and required certain interventions. Optimising analgesia regimens and using non-pharmacological methods have been described as effective interventions in pain management.^1,5 ​​Due to their continuous presence nurses play an essential role in this regard.²⁹

Another item evaluated in this study that influenced sleep perceptions was discomfort with drains, monitoring cables, strange beds, etc., a finding that is consistent with the literature.^{2,8,10,11,18,19,28} Although these factors are difficult to modify, nurses can promote a comfortable sleeping position, reduce and prioritise the number of care activities, provide extension cords to allow patient mobility, and secure devices to avoid interruptions in the patient's sleep.^28,31

Light has been found to be another factor influencing sleep perception, as is the case in most of the literature reviewed.^{2,4,5,8,11,19,21,35} This result is striking because, in this unit, patients have the ability to control the light in their room and the window blinds. Furthermore, the room doors have a metal curtain that prevents light from the hallway from entering the patient's room. This finding may possibly be explained by the fact that, as Engwall et al. stated in their study on lighting, sleep, and circadian rhythm, lighting levels have a comforting effect on patients at night, as they can see and interact with staff.³⁵ The literature describes measures such as exposure to natural light, maintaining the circadian rhythm by avoiding excessive naps, and minimising night-time exposure to artificial light as part of a package to promote sleep, with satisfactory results,^29,32 as well as the use of an eye mask.²¹

Against this prognosis, nursing interventions such as blood pressure monitoring are not perceived as negative factors or influencers of sleep/rest, contrary to the literature reviewed.^5,11 This may be due to the fact that patients perceive peace and security with the delivery of the various care options.³³

This research has methodological limitations. The study was conducted at a single centre and the sample was small. Furthermore, assessments of all the nights spent there by the patients were unavailable, and length of stay varied greatly. This may have hindered the influence of length of stay on sleep. Given the variety of medications administered to promote sleep and their combined use, a causal relationship between sleep and the use of a specific medication could not be determined. Regarding the questionnaires, the RCSQ is a tool that has not been validated in Spanish. The nine-question questionnaire has not undergone a validation process. Both questionnaires were administered by a peer. This meant that patients may not have felt free enough to express certain opinions and may have generated patient bias.

Conclusion

Patients in the intermediate care unit have a mediocre perception of sleep. Patient worry/nervousness; the noise of monitor and device alarms; patient-reported pain; discomfort with drains, monitoring cables, unfamiliar beds, and ambient light at night are all factors that influenced sleep. Informing the patient at all times regarding their concerns and before performing any procedure; minimising ambient light and noise; optimising analgesia, and promoting a comfortable position could facilitate a good night's rest.

Funding

This study did not receive any funding.