The Deterioration Early Warning System for aged residential care: a feasibility study with qualitative analysis

The timely identification of acute deterioration in people living in residential aged care (RAC; residents) is critical to ensure they receive the right treatment, in the right place at the right time.^1,2 The Deterioration Early Warning System (DEWS) is a unique evidence-based process, co-designed and developed in New Zealand, to support RAC staff to achieve that goal. It was developed in collaboration with the New Zealand Health Quality & Safety Commission (HQSC) and the RAC sector using mixed methodologies.³ Up to 75% of residents live with advanced frailty.⁴ Frailty is associated with physiological changes that may manifest as atypical or non-specific clinical signs during acute deterioration.^2,5 Disease presentation via non-specific signs, such as falls, weakness or fatigue, somnolence, dyspnoea, loss of mobility, functional decline and failure to eat and drink, can easily be interpreted as a normal variation of ageing and frailty and contribute to health professionals missing or under-estimating clinical acuity.^5–7 Such situations can precipitate serious adverse events, including functional decline, increased short-term mortality and death.^7–9

Background

The New Zealand RAC sector is composed of multiple non-affiliated providers (48% operated by small groups or individual owners, 52% large-group providers) and provides homes for approximately 35,000 people.¹⁰ The sector employs about 2,000 registered nurses (RNs; 40% are internationally qualified) and approximately 8,000 healthcare assistants (HCAs).¹⁰ In this context, RNs are responsible for recognising and responding to acute deterioration;^11,12 however, few effective tools exist to support them in this endeavour.^1,2,13

There is a body of literature supporting the use of early warning scores (EWS) to detect patient deterioration and support nursing critical thinking in acute hospital settings.^14,15 However, these tools lack validity for people living with frailty in hospital¹⁶ and RAC environments;^17,18 albeit, they are reported to positively impact on clinical communication and confidence in RAC.^17,18 DEWS is a comprehensive system that applies the “track and trigger” philosophy of EWS. It requires HCAs to regularly and systematically observe residents for clinical indicators correlated with acute deterioration, and subsequently alert the RN.^19,20 The system then supports RNs to conduct further clinical assessments and case escalation for definitive diagnosis or treatment.

Iterative testing of DEWS occurred with participants from the RAC sector during the co-design and development process. However, the system required testing on a larger scale to determine its efficacy and appropriateness for the sector. Feasibility studies are essential for understanding the potential of an intervention to translate into real-world settings. They offer proof of concept without requiring immediate commitment to changing clinical practice.²¹ Within this collaborative framework, any subsequent translation to practice is formed by implementation experience and RAC recommendations.²¹ A type 2 hybrid effectiveness-implementation study was conducted to evaluate clinical benefit and contextual fit within the RAC sector.²² Specifically, the goals of the study were to identify the potential of DEWS to:

1. effectively support RAC staff with the timely identification and response to acute deterioration; and
2. be implemented in the RAC sector.

Method

Design and grounding framework

The Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) framework was used to support project planning and evaluation.²³ Each of the RE-AIM domains relates to an important component of a translation project and asks key questions to ensure comprehensive planning and evaluation (see Table 1). RE-AIM has been applied in multiple countries and cultures since its development two decades ago.²⁴

View Table 1–3, Figure 1–4.

Eligibility

Eligible RAC providers were required to have an established primary care service and a minimum of 50 beds (to ensure sufficient RN participant numbers). Facilities where members of the research team had an existing contract were excluded from the study.

Recruitment and sample

Research information was distributed electronically via HQSC networks. Interested RAC providers were screened for eligibility, and purposive maximum variation sampling was applied to achieve as much sector diversity as possible. The recruitment target was three to five facilities, a number that balanced variation sampling requirements and researcher capacity.²⁵

Ethics

Ethical approval was granted by the Auckland Health Research Ethics Committee (approval AH26938) and the trial was registered with the Australian New Zealand Clinical Trials Registry (ANZCTR number 12624000244505). Locality approval was provided by the chief executive (or equivalent) of each facility. Staff working in participating RAC facilities could not opt out of the day-to-day use of DEWS, as the identification of acute deterioration is core practice. However, participation in the evaluation was voluntary and written consent was collected from each participant.

Feasibility study structure

The study was supported by two key advisory groups: future DEWS users (Aged Residential Care Quality Leads Forum) and stakeholders potentially impacted by implementation (expert advisory group).²¹ The study was led by research investigators from The University of Auckland (JD, MB, JR) and HQSC (RP). Each participating facility was required to establish a locality project leads (PLs) group to drive the day-to-day implementation of DEWS (Figure 1).

Intervention

The study was organised into four phases: plan, prepare, test and evaluate (Figure 2). In brief, the planning phase focussed on the establishment of locality PLs. During the preparation phase, a workshop was provided for PLs to facilitate their understanding of the study’s quality improvement activities and the DEWS. They used this information to support completion of project documentation (e.g., project charter and current state assessment) and deliver pre-formatted DEWS education to their staff. The test phase comprised DEWS implementation, while data collection and analysis occurred during the evaluation phase.

Data collection

Semi-structured focus group interviews were conducted separately with PLs, RN tool users and HCA tool user groups (see Appendix for the interview guide). Interviews were undertaken in-person in the RAC facilities by JD and RP at a time that met the operational needs of the facility. This was a pragmatic decision to maximise participation. Interviews were audio recorded, transcribed and loaded into NVivo software (Lumivero, version 14.23.3)

Data analysis

Data from each group type were analysed separately using template thematic analysis, a method chosen for its unique ability to balance the structure of a priori theory with the flexibility of inductive discovery.²⁶ This approach does not necessitate a rigid distinction between descriptive and interpretive themes, which enabled a more nuanced exploration of the participant experiences within a defined framework. The analysis started with an a priori template based on the RE-AIM framework;²⁶ combining template analysis with the RE-AIM framework is a recognised approach in implementation science.²⁷ This provided a theory-based approach to ensure all key dimensions of implementation were systematically addressed.²⁸ To test the template’s utility and ensure interpretive rigour, two authors (JD and RP) independently read, coded and themed a subset of the data (six transcripts; two from each interview category). Where the data exceeded the initial framework themes, new themes were inductively developed to ensure the final template reflected the depth of the transcripts rather than fitting information into predefined categories. This iterative refinement continued through the entire dataset. Critical review was provided by MB and JR during the writing phase, which challenged and refined the final thematic structure.

Reflexivity and rigour

The research team consisted of three female nurse practitioner researchers (Associate Professors MB and JR; PhD candidate JD) and a female RN quality improvement facilitator (RP). As experienced clinicians within the RAC and palliative care sectors, the researchers held “insider” status. This shared professional background facilitated rapport and a deep contextual understanding of the clinical environment. In accordance with feasibility study design, all interview data were collected during the evaluation phase, with the sample size anticipated to be sufficient for achieving informational power.²⁵ Separate analyses of RN, HC, and PL groups enabled thematic triangulation across the dataset. This approach enabled an understanding of both the practical, day-to-day use of the tool and the broader organisational and sector-wide implications of DEWS. All PLs and participants who requested results received draft findings for review; no changes were suggested.

Results

The study was conducted from December 2023 to 30 November 2024. Ten RAC facility applicants were assessed for eligibility, of which six were enrolled. To ensure maximum variation across the sample, the research team limited participation from RAC facilities with shared management structures. Specifically, three applications were received from a single major provider group; two were excluded because facilities within that group utilise a standard approach to detecting acute deterioration. A further two did not meet inclusion requirements. One participant facility withdrew in the planning phase due to staffing shortages and provided no data. The remaining five facilities completed all phases of the study (see Figure 3).

Participant characteristics

Participant facilities varied by location, size and business model and included all levels of care; one facility self-identified as Māori-led (Table 2). Most focus group participants were female (n=78, 87%) and of Asian ethnicity (n=51, 57%), with 47% (n=42) having between 1 and 5 years of experience in the RAC sector (Table 3). Focus groups were comprised of three to eight participants and were 45–60 minutes in duration.

Findings

The RE-AIM dimension questions help structure the findings (Figure 4). DEWS was found to be effective in supporting clinical proficiency and responding to the RAC environment. Project structure and organisational leadership supported its adoption, while education, responding to practice challenges and measuring for success supported implementation. Participants recommended DEWS for the RAC sector, albeit with some future improvements. 

Effectiveness domain

The effectiveness domain consisted of two themes: 1) supporting clinical proficiency, and 2) responding to the RAC practice environment.

Supporting clinical proficiency

Clinical proficiency is defined as having a skilled approach to evolving situations.²⁹ Three subordinate themes were identified from the focus group analysis: 1) timely identification of acute deterioration, 2) supporting critical thinking and effective clinical communication, and 3) accountability for practice.

Timely identification of acute deterioration

Participants across all groups (PLs, RNs and HCAs) reported that DEWS improved the timely identification of acute deterioration. One participant expressed surprise at the tool’s effectiveness and impact on getting timely help: “The surprising thing with this tool is that the residents actually were triaged and sent [for help] quicker than they probably would have been before” (PL participant). An RN agreed with this assessment, explaining, “You really couldn’t miss a deteriorating patient from that form at all.” Similarly, an HCA reported, “We became very alert in our shift to identify those residents who need to be assessed by the RN right away.” Beyond acute deterioration detection, the tool was credited with secondary benefits, with one participant commenting that DEWS had a “huge impact in reducing our falls” (PL participant).

Supporting critical thinking and effective clinical communication

Critical thinking is a deliberate process of analysing clinical information to determine the next appropriate step.¹⁵ The promotion of sound critical thinking was considered a key outcome of DEWS, with one participant commenting, “Now … it’s really more critical thinking … rather than [the RN] just asking the family, ‘what do you want to do?’” (PL participant). RNs felt the tool enabled them to “easily identify [acute deterioration] ... and it will lead us to a diagnosis; that uses a lot of our critical thinking skills.” HCA participants also reported thinking in different ways when using DEWS, noting it “sort of tricks you to do a bit of thinking.” This practice change was observed by RNs, who noted that “The caregivers are more focussing on the clinical side of things.”

Across all participant groups, the shift in critical thinking promoted by DEWS contributed to more effective clinical communication. One nurse explained: “Because we are assessing [with DEWS] during the deteriorating condition of the resident, it makes it really easy to communicate with the GP [general practitioner]” (RN participant). The DEWS structured assessment framework also provided a basis for effective communication with families, thereby increasing their confidence in the care provided:

“Sometimes we have trouble communicating with the family ... So, with the tool, we can say that all this was done, and, as per our observation, this was not there yesterday … so the family, they are more confident that their loved one is well looked after.” – RN participant

Accountability for practice

The mandatory nature of the DEWS escalation process clarified role expectations.
Consequently, most HCAs felt that “the issues that I brought up got taken further” (HCA participant). RNs, who previously relied on individual clinical judgements for every episode of acute deterioration, reported feeling less vulnerable with a formalised, organisationally mandated escalation process. One RN explained that the resident “gets really good care and early intervention, if required. It safeguards the nurse ... she can see she’s done absolutely everything she could do, and it safeguards the facility [showing] that we are taking care of the residents properly” (RN participant).

Responding to the RAC practice environment

Responding to the RAC environment consisted of three subordinate themes: 1) recognition of the unique population, 2) maximisation of human resources, and 3) workload and time effectiveness.

Recognition of the unique population

Participants recognised the need for a tool that was specifically developed for their population and environment, valuing that DEWS “didn’t take something directly from hospitals and just plop it in. You can actually see it’s something that’s for aged care, not for the average adult” (PL participant).

Maximisation of human resources

HCAs represent a significant proportion of the aged care workforce. Although primarily engaged in delivering care, they maintain an informal role in monitoring the residents’ condition. The introduction of DEWS formalised this component of their role and highlighted the sector’s reliance on their contribution: “I think that teaches us ... we don’t give our caregivers enough training. We need to really empower them because we’re so reliant on them” (PL participant). All participants identified the educational potential of DEWS, describing it as “incredibly useful to orientate new staff to aged care [and] to give them that framework and tools so that they can pin their knowledge and learning onto that” (PL participant).

Workload and time effectiveness

The resident-to-staff ratio is considered high globally.³⁰ Initially, many participants expressed concern regarding the impact of implementing DEWS on their workload: “I was kind of thinking, where’s the time to do this?” (HCA participant). However, by project completion the participants observed that DEWS had a minimal impact on their usual workload demands. One participant rationalised that “when they [residents] are sick, we take ages with them anyway” (RN participant). In contrast, one group of PLs reported that nurses were no longer “ringing the clinical manager for [after-hours] support”, representing a significant productivity gain for them.

Adoption domain

The RE-AIM adoption domain examines the organisational support required to deliver an intervention; in this study, “leadership and project structure” emerged as a critical theme of implementation.

Leadership and project structure

RAC PLs recognised the opportunity this study presented to improve clinical practice: “Picking up that recognition of the deterioration is a common theme for us; it’s something we’ve grappled with … you do a piece of training, and it just falls short” (PL participant). This motivation to improve practice ensured they remained invested in the study. This commitment was bolstered by a project structure that included an introductory workshop, as well as planning and preparation phases that allowed PLs “enough time to really plan ahead and discuss things” (PL participant). Furthermore, fortnightly teleconferences with the research team provided a valuable prompt “to just remember to do the things that you haven’t done” and provided some collegial reassurance: “If we were having issues, some others were too” (PL participant).

Implementation domain

The implementation domain considers practical strategies to ensure the intervention is delivered correctly. This domain consisted of three themes: 1) education, 2) responding to practice challenges, and 3) measuring success.

Education

The DEWS educational materials were provided by the research team and delivered by PLs. However, all PLs adapted these materials due to operational constraints. One participant explained the need for “bite-sized pieces of education” (PL participant); even if this meant DEWS education was incomplete, they felt they could deliver information that was “relevant, then you can go back and explain if there are problems” (PL participant). This approach resulted in some knowledge gaps, with one HCA asking during the evaluation phase, “Do we just need to give [residents] a couple of days of them eating less before reporting?”

Responding to practice challenges

One PL group described adapting workflows to support DEWS implementation, such as doing some additional, unrelated training so that HCAs could “jump in and do the medication” if the RN was occupied with DEWS assessments (PL participant). They also delegated specific DEWS data collection tasks to HCAs, reasoning that “interpretation and the action as a result of whatever that [DEWS] shows is RN scope of practice” (PL participant).

Measuring for success

Weekly audits were a key strategy for ensuring the intervention was delivered with fidelity. PLs used audit results to monitor implementation progress and provide corrective feedback where necessary. As one participant explained: “It’s constant reminding everyone until they embrace it” (PL participant).

Maintenance domain

The RE-AIM maintenance domain evaluates whether an intervention is recommended for long-term operational use and how it can be integrated into sustained practice. These considerations comprise the themes for this section: 1) recommendation to the sector, and 2) improvements for future implementation.

Recommendation to the sector

Most participants were supportive of the implementation of DEWS across the RAC sector. One participant noted that “Good practice, good evidence and tools that make it easy to do would just transform our capacity to deliver good care; I hope that most aged care facilities in New Zealand embrace it with open arms” (PL participant). However, one facility felt their existing processes for recognising acute deterioration were sufficient, rendering DEWS unnecessary. Participants also identified the potential for DEWS to improve intersectoral collaboration: “If we all go to the same standard nationalised tool, we can see that we are not unnecessarily transferring residents to hospital” (PL participant). Furthermore, DEWS was also seen as a vital support for internationally qualified nurses transitioning to the New Zealand RAC environment: “Aged care is a unique setting and to have that support, it’s huge” (PL participant).

Improvements for future implementation

Two subordinate themes were identified regarding improvements required for future implementation: 1) digitalisation, and 2) education and support.

Digitalisation

Four of the five participating RAC facilities used a digital patient management system (PMS). To use DEWS effectively in the future, most participants felt it must be integrated into the existing PMS platforms that support clinical workflows. One participant noted that integration would avoid having to “input data two or three times” (PL interview). The ability of a PMS to generate automated alerts that “actually automatically go to the nurses” (HCA interview) was viewed as a seamless melding of current systems with DEWS.

Education and support

The initial time investment required to train staff was a significant consideration in implementation success. Participants agreed DEWS education needed to be delivered in modular, bite-sized formats; ideally “it is a 10-minute read” (PL participant). All leadership groups expressed concern that future implementation might fail without the formalised project structure and support provided during the feasibility study. At a minimum, they suggested a “help desk, a person, or a process, or a portal … or a bit of a drop-in session” (PL participant) would be necessary for sustained uptake across the RAC sector.

Discussion

This study used the RE-AIM framework to evaluate the feasibility of DEWS in the RAC sector. Overall, findings demonstrated the value of DEWS and its “track and trigger” approach in supporting the timely identification of, and response to, acute deterioration. DEWS was found to not only support clinical proficiency but also align with the unique logistical and human resource constraints of the RAC environment. Crucially, participants strongly recommended sector-wide adoption of DEWS in New Zealand.

Reach

This study engaged a diverse cross-section of the RAC sector, ranging from a small, independent rural provider to large urban corporate and privately owned entities. Notably, one provider self-identified as a Māori-led environment where Te Tiriti o Waitangi obligations were embedded and enacted within care delivery. Overall, this ensured participants were representative of the broader industry and that due consideration was given to health equity implications of a tool with the potential for sector-wide transferability.

Effectiveness

The primary goal of DEWS was to address the challenge of timely identification of acute deterioration in the context of frailty-mediated clinical presentations. This study found that DEWS not only effectively identified unwell residents but also facilitated a faster response to deterioration compared to standard care. This is a significant finding, likely underpinned by the use of non-specific signs to identify acute deterioration, which often manifest before measurable changes in vital signs occur in people living with frailty.^2,5 Consequently, DEWS has the potential to reduce the risk of health professionals under-estimating clinical acuity.^5–8

Consistent with prior early warning system research, this study found that DEWS positively impacted critical thinking and clinical communication among both RN and HCA cohorts.^15,17–19 Furthermore, RNs valued the “safeguard” provided by a structured assessment and escalation process. This shifting of the decision-making burden from individual clinical judgement to an organisationally supported and mandated process is an important finding that warrants further investigation. In this study, 88% of RN participants were Asian, and 65% had 5 years or fewer of aged care experience. DEWS was found to provide them with essential clinical scaffolding, supporting the transition of internationally qualified and early-career nurses to the RAC environment. This is particularly salient given that 40% of the RAC nursing workforce is internationally qualified and they often enter the sector with limited aged care experience.^10–12 For HCAs, DEWS formalised their role in monitoring resident wellbeing; this builds on existing practice norms and aligns with research demonstrating the efficacy of such developments.^19,20 In an environment where HCAs outnumber RNs by a four-to-one ratio, this is a critical finding.¹⁰

Adoption and implementation

Consistent with implementation science principles, where “buy-in” is most effective when an intervention addresses a recognised clinical need,^21,24 this study found that DEWS adoption was highly influenced by locality PLs motivated to close long-standing gaps in deterioration detection processes.^8,9 Regular implementation support meetings provided by the research team were also critical to maintaining study momentum.²¹ Pragmatic adaptations of materials and processes occurred throughout the study; such modifications are a recognised phenomenon in feasibility testing.²⁴ For example, PLs navigated operational constraints by delivering education in bite-sized modules and delegating routine tasks to HCAs, thereby protecting RNs’ time for DEWS-associated clinical decision making. Importantly, given the workload pressures in RAC, DEWS had minimal impact on staff time. Furthermore, this study found a reduction in after-hours calls to clinical managers, suggesting DEWS may offer productivity gains; this potential warrants investigation in future trials. Similarly, the marked reduction in falls observed in one facility highlights the need to collect objective outcome measures in future research.

Maintenance

The maintenance dimension of the RE-AIM framework evaluates the operationalisation and long-term sustainability of an intervention. This study yielded overwhelmingly positive feedback regarding the suitability of DEWS for the RAC population and environment, indicating high sustainability potential. Participants identified digitalisation as the next logical step for DEWS development: specifically, the transition of paper-based tools and the creation of modular, online education. Ultimately, integrating DEWS into existing electronic PMS and clinical workflows was identified as the most viable strategy for achieving sector-wide, long-term institutionalisation.

Strengths and limitations

Key strengths of this study included the multi-site sampling and sustained engagement of key stakeholders. By involving potential future DEWS users and health system collaborators, the study provides confidence regarding the transferability of findings across the RAC sector. Additionally, the use of the RE-AIM framework allowed for a rigorous multidimensional evaluation beyond clinical efficacy alone. However, as a qualitative feasibility evaluation, this study lacked matched control or comparison groups, which limits the ability to draw definitive causal conclusions. Furthermore, feasibility studies inherently involve “the willing”, introducing a potential for selection bias. Nevertheless, participants were free to provide a “do not implement” recommendation, and one facility explicitly reported that DEWS was not required, suggesting a degree of participant candour. Finally, the study was time limited; a more protracted testing period may have resulted in participant fatigue and less favourable findings.

Conclusion

Participants in this study recommended the adoption of DEWS across the RAC sector. They valued the system’s capacity to effectively support the multidisciplinary team in the timely identification of, and response to, acute resident deterioration. While implementation was feasible through building on current ways of working, future uptake will be contingent upon digitalisation and integration within existing information technology infrastructure.

View Appendix.