Transforming Intensive Care Unit Environments Through Human-Centred Design: A Mixed-Methods Retrofit Study of Noise and Lighting Interventions

Heightened public and professional attention to critical care in recent years, driven largely by the COVID-19 pandemic [1, 2], has coincided with continued improvements in survival among patients admitted to intensive care units (ICUs) [3, 4]. Nevertheless, accumulating evidence suggests that survival alone is an insufficient outcome, as many patients experience lasting physical, psychological, and cognitive impairments following discharge, resulting in a diminished quality of survival [5, 6]. The connection between environmental conditions and human health has been acknowledged since ancient times [7], and contemporary research has further demonstrated that the built environment influences mood, behaviour, cognition, learning capacity, overall health, and sleep quality [8]. Despite this growing body of knowledge, the role of the intensive care unit (ICU) environment in shaping patient morbidity and mortality has, until relatively recently, received limited attention and insufficient research investment.

The challenges associated with the ICU environment are multifaceted and interrelated, encompassing excessive noise, inadequate lighting, social isolation, limited opportunities for personalisation, absence of views or access to nature, and a lack of meaningful cognitive stimulation or distraction. Collectively, these factors contribute to sleep disruption, delirium, and psychological distress, which are associated with increased mortality and persistent physical, cognitive, and mental health impairments following discharge [9-11]. Family members similarly report the ICU setting as intimidating and highly stressful, with adverse psychological effects that may extend beyond the period of admission [12]. Healthcare staffs are also affected, as high noise levels and insufficient exposure to natural light and views can impair physical and mental wellbeing, reduce concentration and decision-making capacity, and contribute to headaches and alarm fatigue [13]. Moreover, by exacerbating patient confusion and delirium, the environment may increase risks to staff safety through heightened exposure to verbal or physical aggression [14, 15]. These issues are likely to intensify as advances in medical technology enable the admission and survival of increasingly unwell patients, resulting in longer ICU stays and prolonged exposure to an environment that may inadvertently compound harm.

The evidence base guiding ICU bedspace design has historically been limited, resulting in considerable variation in layouts and design approaches both within and across countries [16]. Early ICUs, much like general hospital wards, were largely derived from open-plan Nightingale ward models and, under the influence of anaesthetists who initially led critical care services [17], were conceived as extensions of the operating theatre to accommodate deeply sedated patients. Although ICU technology, clinical practices, and models of care have advanced substantially most notably with a shift toward lighter sedation and greater patient engagement the physical environment has changed little, failing to reflect these fundamental developments in patient management and rendering many traditional designs no longer appropriate for contemporary care. Several interrelated factors have contributed to this lack of innovation, including limited historical awareness of environmental harm, the perceived financial burden of design elements considered non-essential, and the practical challenges of undertaking major upgrades within existing ICUs. In addition, there has been a tendency to accept the harms associated with established environments as unavoidable, while viewing the risks of change as unacceptable or preventable [18, 19]. Current ICU planning processes also tend to prioritise clinical efficiency, with decisions often driven by administrators, architects, and builders, leaving minimal scope for meaningful input from those who use the space daily namely clinicians, patients, and families, despite increasing recognition of the value of consumer involvement in health policy and organisational change [20].

Improving the physical environment of the ICU offers considerable potential to enhance patient outcomes as well as staff health, performance, and overall efficiency of care delivery [21]. Growing attention has been directed toward the concept of biophilia, with calls to strengthen connections to nature and green environments within ICU settings [22]; however, such strategies are not always feasible, particularly when working within the constraints of existing facilities and retrofit projects. As a result, there is increasing recognition that more fundamental environmental redesign is needed to address longstanding problems, despite the absence of a robust evidence base to define a single “gold-standard” ICU or consensus on what an ideal critical care environment should achieve. At a minimum, an effective ICU design should aim to reduce preventable harms, including delirium, poor sleep quality and duration, disruption of circadian rhythms, and the persistence of physical and psychological problems after discharge. Against this background, this paper outlines the process undertaken to reconceptualise, design, and retrofit two ICU bedspaces, and presents early findings on the impact of the implemented environmental improvements. Despite increasing recognition of the ICU environment as a determinant of patient and staff outcomes, existing research remains fragmented, often addressing individual environmental factors such as noise or lighting in isolation. Moreover, there is limited empirical evidence evaluating integrated, human-centred environmental interventions implemented within operational ICU settings, particularly under retrofit constraints. As a result, there remains a lack of clarity regarding how combined design strategies can be practically translated into measurable environmental improvements in real-world clinical environments. The objectives of this study are:

• To reconceptualise ICU bedspace design using a patient-centred and evidence-informed framework.
• To identify key environmental problems within existing ICU bedspaces that adversely affects patients, families, and staff.
• To design and implement a retrofit solution addressing noise, lighting, sleep disruption, and sensory deprivation.
• To integrate innovative and feasible design and technological interventions within a live ICU setting.

To evaluate early environmental outcomes following implementation, with particular focus on acoustics and lighting.

To address this gap, the present study is guided by two core research questions: (1) To what extent can targeted environmental design interventions in ICU bedspaces measurably reduce noise exposure and improve acoustic performance? and (2) Can the implementation of a bespoke circadian lighting system enhance lighting conditions in alignment with natural biological rhythms compared to conventional ICU environments? By focusing on these specific and measurable aspects of the ICU environment, the study aims to move beyond general improvement strategies and provide empirical evidence on the effectiveness of human-centred, evidence-informed design interventions within real clinical settings.

3.1 Study design

This study employed a mixed-methods, multi-stage participatory design framework integrating qualitative inquiry, objective environmental assessment, evidence synthesis, and iterative spatial prototyping. The approach was guided by principles of human-centred design and implementation science to ensure that environmental interventions were both evidence-informed and feasible within a live ICU setting. The project progressed through interconnected phases including stakeholder engagement, baseline environmental evaluation, and development of design concepts, full-scale prototyping, and post-implementation environmental evaluation of retrofitted ICU bedspaces.

3.2 Participants and setting

The study was conducted within a tertiary hospital ICU comprising 30 beds organised across three clinical pods. Two windowless internal bedspaces of approximately 24 m² were selected for retrofit, having previously been used for storage and simulation purposes.

Participants included ICU patients, family members, and multidisciplinary clinical staff. Thirty patients, twelve family members, and thirty-five clinicians representing nursing, medical, and allied health disciplines participated in qualitative interviews and focus groups. Stakeholders were further engaged throughout the design and implementation process, including hospital managers, biomedical engineers, infection control personnel, architects, and technical contractors, to ensure clinical safety, operational feasibility, and alignment with institutional requirements.

3.3 Data collection

Data collection occurred across several complementary phases. Qualitative data were obtained through semi-structured interviews with patients and family members and focus groups with ICU clinicians to explore experiences of the existing ICU environment, perceived stressors, and priorities for improvement.

In parallel, baseline observational and technical assessments were conducted to characterise environmental conditions within the ICU, including acoustics, noise exposure, lighting levels, and alarm burden. An extensive review of relevant healthcare design and clinical literature was undertaken to support evidence-informed decision-making. During the design phase, iterative stakeholder consultations and simulation-based walkthroughs of a full-scale prototype space were used to gather practical feedback and refine architectural and technological solutions prior to in-situ implementation.

To ensure integration between qualitative and quantitative components, the study adopted a sequential and iterative mixed-methods approach. Findings from qualitative interviews and stakeholder engagement were first used to identify and prioritise key environmental stressors, including noise, lighting, and spatial constraints. These insights directly informed the development of targeted design interventions and the configuration of the retrofitted ICU bedspaces. Subsequently, quantitative environmental measurements were employed to objectively assess the effectiveness of these interventions, particularly in relation to acoustic performance and lighting conditions. This integration enabled the study to move from user-identified needs to evidence-based design solutions, and finally to measurable evaluation of environmental outcomes.

3.4 Environmental measurements

Objective environmental measurements were used to quantify physical conditions before and after the retrofit. Acoustic performance was assessed using sound level monitoring and reverberation time (RT) measurements to evaluate noise exposure and sound absorption within bedspaces. Alarm sound pressure levels were recorded to assess perceived noise at the patient head position.

Lighting assessments included measurement of illumination intensity and spectral composition across day and night periods to examine alignment with circadian rhythms, with comparisons between conventional ICU lighting and the bespoke circadian lighting system introduced in the redesigned rooms. These measurements provided empirical evidence of environmental change following implementation.

Environmental noise measurements were conducted using a calibrated Class 1 sound level meter compliant with IEC 61672 standards. Reverberation time (RT60) assessments followed ISO 3382 acoustic measurement procedures. Measurements were obtained at standardised positions, including the patient head location and the geometric centre of the room. Lighting intensity and spectral power distribution were measured using a calibrated spectroradiometer positioned at patient eye level to reflect typical visual exposure conditions during clinical care.

To improve comparability between pre- and post-intervention conditions, environmental measurements were conducted under standardised protocols. Measurements were obtained at consistent locations (including patient head position and room centre), during comparable time periods, and under typical ICU operating conditions to reflect real-world use. While factors such as patient activity, staff movement, and equipment operation cannot be fully controlled in a live clinical environment, efforts were made to minimise variability by avoiding atypical events (e.g., emergency procedures) and ensuring similar occupancy and workflow patterns during measurement periods. These procedures were implemented to enhance the reliability of comparisons and reduce the influence of external confounding variables.

Where only range data were available, approximate statistical measures were derived to support comparative analysis. Mean values were estimated using the midpoint of the reported ranges, while standard deviations were approximated as one quarter of the range (SD ≈ [max-min]/4), consistent with common estimation approaches for small environmental datasets. Assuming repeated measurements under stable conditions (n ≈ 10), 95% confidence intervals were calculated using the expression mean ± (1.96 × SD / √n).

3.5 Data analysis

Qualitative data obtained from interviews and focus groups were analysed using a framework analysis approach, enabling systematic coding, categorisation, and identification of key themes related to environmental stressors, patient experience, and staff workflow. Quantitative environmental data, including noise levels, reverberation time, and lighting measurements, were analysed descriptively to compare pre- and post-intervention conditions. Comparative analysis focused on changes in acoustic performance and lighting characteristics at standardised measurement points. Findings from qualitative and quantitative analyses were triangulated to support interpretation and to evaluate the early impact of the implemented environmental interventions.

To ensure the project genuinely reflected the needs of its end-users, the work began with qualitative interviews involving patients, family members, and ICU staff to explore how the existing environment shaped patient experience, recovery, and day-to-day care, as well as to identify practical ways to address the issues they perceived. The full methodological details and findings of these interviews are available elsewhere [36], with a brief summary presented here. Thirty patients, twelve family members, and thirty-five clinicians from medicine, allied health, and nursing participated through individual interviews and, for staff, focus groups, with data analysed using a framework approach [37]. Patients frequently described the ICU as an intimidating and overwhelming place, drawing attention to sleep disruption caused by noise and bright lighting at night, the cramped and cluttered bedspaces that hindered care activities, the lack of opportunities to personalise their surroundings, and limited access to daylight, meaningful stimulation, family connection, and external views [20]. Staff accounts largely echoed these concerns, noting that the current bedspaces did not support healing and that environmental constraints including persistent noise and inadequate or poorly designed lighting and views contributed both to staff fatigue and to challenges in delivering optimal care [29]. Following this qualitative phase, a series of objective assessments was undertaken in the study ICU to measure physical and sensory conditions such as light levels, noise, acoustics, and alarm load [17], accompanied by an extensive literature review to deepen understanding of the environmental problems observed, the feasibility of potential solutions, and any documented outcomes where improvements had been trialled. The combined evidence from these interconnected studies was then synthesised into an initial list of patient-centred environmental problems requiring action for the project Figure 1.

Figure 1. The project's list of patient-centered issues

After the project requirements were confirmed, the team identified a suitable site for the upgrade, acknowledging that every ICU has its own layout, patient profile, and operational demands, and therefore any redesign must be adapted to the realities of that particular unit. Working with the hospital’s clinical leaders and management, a plan was developed to refurbish two ICU bedspaces through a retrofit carried out within the functioning unit, following the innovation and process elements of the CFIR framework [38]. The ICU contained 30 beds arranged across three pods; the two chosen bedspaces windowless, internal areas of about 24 m² were located in one corner and had previously served as storage and simulation spaces in Figure 2. After detailed discussions to understand local needs and constraints, these spaces were reviewed thoroughly, and a set of improvement recommendations was compiled and ranked according to their likely impact on patient outcomes such as delirium, sleep, and overall experience, as well as the balance between cost and expected benefit. Initial architectural and technological concepts were drafted and then refined through five months of consultations with a wide group of stakeholders, including former patients and families, ICU staff, project managers, IT and biomedical services, engineering teams, infection control personnel, external consultants, and contractors supplying or installing equipment. Plans were revised repeatedly as new feedback emerged. Once the scheme had progressed sufficiently, a full-scale prototype in Figure 3 was built in another building on the hospital campus, allowing staff to walk through the space, test equipment placement, and provide practical comments during simulation sessions; former patients and family members were also invited to offer their perspectives. Revisions were made where feasible, taking into account the limitations of retrofitting within an active ICU, such as fixed walls, structural conditions, and spatial restrictions. The plans were then finalised to prepare for construction within the unit.

Figure 2. Patient rooms before (A) and after upgrades (B, C)

As shown in Figure 3, the full-scale prototype enabled simulation-based evaluation of spatial layout, equipment positioning, and workflow efficiency, thereby informing design refinements that were later reflected in the final environmental performance outcomes.

Figure 3. A prototype area designed for staff consultation and training through simulations

This study addressed two primary research questions concerning the effectiveness of environmental design interventions in reducing noise exposure and improving lighting conditions in ICU bedspaces. The findings demonstrate that targeted acoustic and lighting strategies can produce measurable improvements in environmental performance, supporting the role of human-centred design as a practical approach within real-world clinical settings.

Over the past two decades, improvements in intensive care practice have translated into higher survival rates, but this success has also highlighted the need to focus more carefully on what survival means for patients once the immediate crisis has passed. While there is now broad recognition that the ICU environment can influence patient outcomes in meaningful ways, comparatively little attention and funding have been directed toward environmental redesign. This paper documents the practical steps taken to enhance two ICU bedspaces, using a combination of established evidence, emerging design approaches, and new technologies to respond to patient-centred issues that are widely encountered in critical care settings. Early results suggest that these changes produced clear benefits, particularly in relation to sound and light, with reverberation times reduced by more than half indicating substantially improved acoustic absorption alongside better attenuation of noise originating outside the rooms and the introduction of a lighting system designed to follow natural circadian patterns rather than the static output of conventional electrical lighting. Detailed measurements of lighting conditions prior to the refurbishment, including comparisons between windowed and windowless bedspaces under different scenarios, have been reported separately [41] and showed that the presence of a window only supported natural light exposure when ceiling lights were turned off; once artificial lighting was activated, the quality of light resembled that of a windowless space. In contrast, the circadian lighting system implemented in this project, as shown in Figures 6 and 7, closely approximated natural daylight in both spectral composition and timing, reflecting the lighting conditions to which human circadian rhythms are biologically attuned. In addition, repositioning monitor alarms away from the patient’s head resulted in a reduction of perceived alarm noise by 8–11 dBA, a change equivalent to roughly halving the subjective loudness of the sound and comparable to the average noise reduction achieved through the use of earplugs [42]. Figure 6 provides a detailed representation of the spectral power distributions across different lighting scenarios, demonstrating how the implemented system more closely replicates natural daylight patterns compared to conventional ICU lighting.

Figure 6. Circadian lighting spectral power distributions across day-night scenes

Figure 7. Natural light spectral power distribution

Bringing together a wide range of partners through a participatory design process, and giving genuine weight to the perspectives of patients, families, and frontline staff, proved to be critical to the success of this work. This approach enabled the team not only to address specific shortcomings of the existing ICU bedspaces but also to rethink how these environments could better support current and future needs, based on real experiences rather than assumptions. While consumer involvement is increasingly acknowledged as important within healthcare improvement and change management, it has seldom been meaningfully embedded in the design of ICU bedspaces, where decisions are often made through hierarchical or technically driven processes. By contrast, the collaborative model adopted here fostered shared ownership and more grounded innovation, aligning with broader evidence that inclusive, non–top-down approaches are more likely to produce effective and sustainable change [43]. Although the final solutions were shaped by the local context and priorities of the study ICU, the challenges identified along with the process used to address them are common across critical care settings, suggesting that many of the principles and interventions described may be relevant to ICUs and other inpatient environments internationally.

As ICU care continues to evolve, the effects of the physical environment on patients are likely to become more pronounced, particularly as increasingly complex technologies are introduced to support the care of sicker patients. Noise levels within ICUs have risen over time, accompanied by a growing number of monitors, alarms, and light-emitting devices that are especially disruptive during night-time hours [44]. At the same time, patients are now more likely to survive critical illness but often require longer ICU stays, increasing the duration of their exposure to these environmental stressors [45]. Reduced use of deep sedation has further changed the patient experience, leaving individuals more alert and aware of their surroundings, yet the typical ICU setting remains stark, functional, and largely devoid of meaningful sensory engagement. In this context, narrowly focused measures that attempt to block out discomfort, such as earplugs or eye masks, are unlikely to address the complexity of the problem, underscoring the need for broader, multi-layered interventions that target the sources of noise, light, and sensory deprivation directly. These developments also raise questions about the adequacy of existing ICU design standards, which may no longer align with contemporary models of care and should be reconsidered to better support patient involvement, staff workflows, and healing-oriented environments. Although access to nature and green spaces has been increasingly advocated as beneficial for ICU patients [7], the retrofit nature of this project and the absence of windows meant that direct provision was not possible. To respond to this limitation, technological alternatives were introduced, including virtual windows and skylights displaying natural scenes with associated soundscapes that could be selected according to patient preference. Nature-based videos and sounds were also incorporated into the entertainment system and delivered through speakers integrated into the beds. While there is strong evidence supporting the health benefits of exposure to natural environments [7], further research is needed to determine how effectively digital representations can replicate these effects in critical care settings. As shown in Figure 7, the spectral profile of natural light serves as a reference benchmark, highlighting the extent to which the proposed circadian lighting system approximates biologically relevant light conditions.

Reducing noise and optimising lighting are key objectives in intensive care environments, and this project has demonstrated that such improvements are achievable even within the constraints of a live ICU retrofit; however, the findings also highlight the importance of environmental personalisation, as patients have differing needs and preferences, and the ability to individualise elements such as lighting can shift the environment from a passive contributor to negative outcomes into an active component of care, where appropriate environmental conditions at the right time may support recovery while enhancing patient autonomy, which is often limited in traditional ICU designs. Despite the technical and organisational challenges encountered during implementation including patient-related factors, staff adaptation, technological limitations, and the unexpected onset of the COVID-19 pandemic ongoing communication and stakeholder collaboration enabled progress throughout the project. While previous studies indicate that interventions such as circadian lighting solutions and physical noise-reduction measures can decrease delirium incidence and ICU length of stay [46], that sound masking can improve sleep quality [47, 48], and that earplugs are associated with reduced delirium [49, 50, 51, 52], there remains a lack of evidence assessing the impact of large-scale environmental upgrades on patient outcomes. Consequently, the next phase of this project will involve a comprehensive evaluation of environmental effects on patients, families, and staff, using qualitative interviews alongside quantitative assessments of sleep patterns, circadian rhythms, delirium occurrence, ICU outcomes, and physical, cognitive, and psychological recovery six months after ICU discharge.

From a safety perspective, the environmental improvements observed in this study have important implications for both patients and healthcare staff. Excessive noise levels and poor lighting conditions are well-established contributors to cognitive overload, reduced concentration, and impaired decision-making among clinicians, all of which may increase the likelihood of clinical errors. Similarly, environmental stressors such as sleep disruption and circadian misalignment are strongly associated with delirium, which is a known risk factor for adverse events, prolonged hospitalisation, and increased mortality. By reducing noise exposure and aligning lighting conditions with natural biological rhythms, the redesigned ICU environment has the potential to mitigate these risks and contribute to safer care delivery. These findings reinforce the importance of considering environmental design not only as a matter of comfort or experience but as a critical component of patient safety and risk reduction strategies within intensive care settings.