FORWARD (Feeding via the Oral Route With Acknowledged Risk of Deterioration): evaluation of a novel tool to support patients eating and drinking at risk of aspiration

Abstract

Background

care of patients with a permanently unsafe swallow who are inappropriate for tube feeding is challenging. Eating and drinking with acknowledged risk (EDAR) may be an appropriate strategy but without clear decision making and communication patients may spend unnecessarily long ‘nil by mouth’ (NBM), they or their family may experience significant anxieties and advance care plans may not be made.

Methods

the FORWARD (Feeding via the Oral Route With Acknowledged Risk of Deterioration) care bundle was sequentially co-designed and embedded across different in-patient clinical services using ‘plan-do-study-act’ methodology to systematise best practice. Care before and after FORWARD’s implementation was evaluated using a time-series analysis of 305 ‘EDAR patients’ (19 in 6 months pre-FORWARD; 42 in a 12-month ‘pilot’; 244 patients in the subsequent 27 months).

Results

median (IQR) days patients were NBM without an alternative feeding route reduced significantly from 2 (1–4) pre-FORWARD to 0 (0–2) in the ‘pilot’ and 0 (0) post-‘pilot’ (P < 0.05). G-chart analysis demonstrated sustained performance across time and different clinical settings. Implementation of FORWARD significantly improved documentation of capacity assessment (42%→98%), discussions with next of kin (47%→98%) and onward communication of feeding plans (67%→81%). In wards where FORWARD was introduced, rate of aspiration pneumonia (a ‘balancing measure’ sensitive to harm associated with EDAR) increased at half the rate of dysphagia (0.8%/year versus 1.6%/year).

Conclusion

the FORWARD care bundle supported complex decision-making around EDAR in patients with persistent dysphagia. The benefits of FORWARD were shown to be sustained over time and in a wide in-patient context.

dysphagia, risk feeding, care bundle, older people

Key points

• Management of patients with an unsafe swallow unlikely to improve is very challenging and patients may experience significant periods nil by mouth (NBM) without alternative feeding arrangements

• Eating and drinking with acknowledged risk is a worthwhile strategy in selected patients with dysphagia

• The FORWARD care bundle produces sustained improvements in the management of feeding with acknowledged risk including a significant reduction in the time patients are NBM and increase in multidisciplinary adherence to quality markers of best practice

• The FORWARD care bundle supports a high standard, patient-centred approach across a range of in-patient settings

Introduction

Approximately 12–13% of hospital in-patients have dysphagia [1], but this proportion has been found to be up to 29% in hospitalised older people [2], 13–57% in people with dementia [3] and 37–78% after stroke [4]. A subset of dysphagic patients have an unsafe swallow that is unlikely to improve because of neurological disease such as stroke, dementia or Parkinson’s disease, anatomical abnormalities of the oropharyngeal tract, or oropharyngeal weakness in the terminal stages of an illness. Despite these impairments modifications to diet and fluid may allow for safe swallowing but sometimes suggested modifications will be declined by the patient or there will be no safe method of oral feeding. Feeding in this way may put these patients at risk of aspiration pneumonia and ‘nil by mouth’ (NBM) orders are often made. Nasogastric and gastrostomy tube feeding may mitigate this problem without completely eliminating aspiration risk, but in a significant number of cases tube feeding is declined, not tolerated, or considered not to be in the patient’s best interests by the multidisciplinary team (MDT). Furthermore, tube feeding is not recommended for patients with advanced dementia [5–7] or those receiving end of life care [8]. Eating and drinking with acknowledged risk (EDAR) is seen as an approach that affords comfort, dignity and autonomy for such patients [6, 9].

Decision making and management for patients EDAR (henceforth ‘EDAR patients’) is complex and assessments must be made of both prognosis and capacity to consent to safest feeding recommendations [6, 10]. High quality communication and co-ordination with patients, carers and MDTs is mandatory. If these do not occur clinical indecision may lead to patients being kept NBM inappropriately long without an alternative means of feeding and to the detriment of their comfort and nutrition. Individualised EDAR plans may be created for patients whose dysphagia is unlikely to improve and these need to be effectively communicated to onward care providers to avoid patients being made NBM after showing signs of aspiration which may cause distress to patients and their families. Given the risk of pneumonia, it is essential that clear ‘escalation plans’ are created so that the appropriateness of interventions such as antibiotics, chest physiotherapy, transfer to a critical care unit or cardiopulmonary resuscitation are considered, particularly if there is anticipated involvement of health care professionals unfamiliar with patients’ cases.

FORWARD (Feeding via the Oral Route With Acknowledged Risk of Deterioration) is a novel care bundle, devised by a multi-professional project team, to support EDAR patients. Care providers are taken through case selection, assessment of capacity, decision making, creating and communicating oral feeding plans and advance care planning. In a pilot study, use of FORWARD reduced the time that EDAR patients remained NBM and improved rates of documentation of capacity assessments and discussions with next of kin [11]. The aim of this study was to evaluate implementation of the FORWARD care bundle for patients with an unsafe swallow unlikely to improve in whom tube feeding had been ruled out, focussing on the time that patients remained NBM in advance of an EDAR decision and MDT adherence to key processes involved in the management of EDAR patients. In addition to assessing these ‘outcome’ and ‘process’ metrics, the study aimed to evaluate the sustainability and spread of the use of the FORWARD bundle in a hospital setting.

Methods

The methodology of the inception of the FORWARD bundle has been described previously [11]. In brief, the bundle was developed over 12 months in accordance with a plan-do-study-act (PDSA) mode of quality improvement that has been shown to be an effective model for healthcare systems improvement [12, 13]. In this project, ‘planning’ occurred in meetings with relevant stakeholders (patient groups; carers; healthcare professionals involved in EDAR), ‘doing’ involved sequentially using the FORWARD bundle in different clinical areas, ‘studying’ was achieved with data collection and face-to-face stakeholder feedback, and ‘acting’ required changes to be made to the bundle according to the findings. The FORWARD bundle underwent 18 iterations to reach the final pathway in the form of a simple flow chart for the MDT to complete (Appendix 1). Alongside this, information leaflets on EDAR were created for staff and for patients or their carers and a FORWARD-branded swallow advice poster for display above patients’ beds was created. The physical paperwork was kept in the speech and language therapy (SLT) office initially as this small and engaged group would be involved in almost all cases. Following its introduction over a 12-month period as a pilot on four wards comprising the older persons’ unit and the stroke unit, the scope of FORWARD was sequentially increased to other medical and surgical wards in the same hospital over 27 months. Clinical engagement was achieved through involving different teams in the design of the bundle at early stages, the use of leadership networks and facilitated teaching sessions.

A quasi-experimental time-series approach was used to look for evidence of the effect of FORWARD on patients’ care. Examination of data was undertaken on the number of days patients were NBM without an alternative feeding route in advance of an EDAR decision (primary outcome) as well as care provider compliance with capacity assessment, discussions with patients or relatives about EDAR and clear documentation of feeding care plans in electronic patient discharge letters, rate of completion of the Proactive Elderly Advanced CarE (PEACE) advance care planning tool [14] and rate of hospital readmission where ongoing EDAR management was required (secondary outcomes). Data were collected from the medical notes using a standardised template. These variables were recorded prospectively for all EDAR patients identified from the SLT hospital in-patient caseload database in the 6-month period prior to FORWARD’s initiation and for all in-patients subsequently supported by FORWARD after its launch. The 6-month pre-FORWARD group was compared with the 12-month FORWARD pilot group and these data were then compared with data collected on patients supported by FORWARD in the 27-month post-pilot period. Quantitative data were tested for normality using the Kolmogorov-Smirnov test. Continuous non-parametric data were compared using Mann–Whitney tests and categorical data were compared using Chi-squared tests.

In order to monitor performance over time it was necessary to map data about the time patients spent NBM on to a control chart [15]. Pilot data demonstrated EDAR patients to spend a markedly reduced time NBM when supported by FORWARD [11]. It was therefore predicted that a period of more than 1 day NBM for any patient awaiting an EDAR decision was likely to become a rare event. G-charts have been demonstrated to be used effectively as a type of control chart in healthcare to monitor low event rates [16] and were used to analyse the ‘event-rate’ of patients being NBM without an alternative means of nutrition for a day or more while awaiting an EDAR decision. Periods of 1 day or more NBM while awaiting an EDAR decision were defined as ‘lapses’ while periods of less than 1 day NBM were defined as ‘successes’. It was noted that not all ‘lapses’ so defined were clinically inappropriate, such as when a patient was found to have incident dysphagia during a weekend but the EDAR decision was only made in the week when relevant staff were available. G-charts were plotted in which the number of consecutive successes between each lapse was plotted on the y-axis against time. Statistical process control methodology was used in G-chart interpretation [17] and a 3-sigma upper control limit was derived from the pre-FORWARD data to compare it with the pilot data and, separately from the pilot data, to compare it with the subsequent post-pilot data.

It was felt theoretically possible that increasing awareness of a risk feeding bundle could negatively affect culture such that premature or inappropriate EDAR decisions were made. Rates of diagnosis of aspiration pneumonia and hospital acquired pneumonia compared with rates of dysphagia and multi-morbidity (defined as > 10 co-morbidities) were therefore chosen as balancing measures that would be sensitive to possible harm associated with increasing EDAR in the in-patient arenas that FORWARD supported. Estimates for these were derived by hospital coding which used systemic clinical record examination to determine diagnoses and co-morbidities. Development and evaluation of the FORWARD bundle formed part of a quality improvement project registered with and adhering to the standards of the Clinical Governance Department at our institution (Service Evaluation 6607) and did not require submission to a research ethics committee.

Results

Nineteen EDAR patients were identified in the 6 months prior to the introduction of the FORWARD bundle and 286 were subsequently supported by the bundle in the ensuing 3 years and 3 months across an increasing variety of clinical settings. Figure 1 demonstrates both the increasing number of patients supported by FORWARD and the increasing spread across different specialty wards. Table 1 shows a comparison between the pre-FORWARD group, the patients supported by FORWARD in its pilot phase, and those supported by FORWARD as an established tool in the post-pilot phase. Median (IQR) time NBM for EDAR patients decreased from 2 (1–4) days in the 6 months before FORWARD was initiated to 0 (0–2) days during the 12-month pilot of FORWARD (P < 0.01). This improvement was sustained in the post-pilot phase with the median time NBM remaining at 0 days but IQR decreasing in magnitude from 2 to 0 days (P = 0.04). There were significant improvements between the pre-FORWARD group and the pilot group in rates of documentation of capacity assessment, documentation of discussion with next-of-kin where they were available, and communication of the feeding plan in the electronic discharge summary. In every case of improvement, performance was sustained into the post-pilot period. PEACE documents were neither completed for EDAR patients in the pre-FORWARD group nor in the pilot phase but were subsequently completed in 19% of cases.

Figure 1

Number of patients fed with acknowledged risk, by ward type. Dark shading = pre-FORWARD, lighter shading = FORWARD pilot. OPU = older persons’ unit.

Table 1.

Comparison of EDAR management before FORWARD, in the 12-month pilot and post-pilot.

	6 months before FORWARD was initiated (Group 1) n = 19	12-month pilot (Group 2) n = 42	Post-pilot (Group 3) n = 244	Comparison between Groups 1 and 2	Comparison between Groups 2 and 3
Median (IQR) patient age	80 (64-88)	84 (76-88)	85 (76-90)	P = 0.33	P = 0.37
Median (IQR) days NBM	2 (1-4)	0 (0-2)	0 (0-0)	P < 0.01	P = 0.04
Rate of documentation of capacity assessment	42%	95%	98%	P < 0.01	P = 0.21
Rate of documentation of best interests discussion	90%	100%	100%	P = 0.08	P = 1.00
Rate of documentation of discussion with patient’s next-of-kin where available	47%	100%	98%	P < 0.01	P = 0.32
Rate of documentation of feeding plan in electronic discharge letter	67%	96%	81%	P = 0.04	P = 0.06
Rate of completion of PEACE plan	0%	0%	19%	P = 1	P = 0.02
Rate of readmission where ongoing EDAR management was required	8%	4%	4%	P = 0.25	P = 0.18

Figure 2a shows the G-chart analysis of performance over time for EDAR patients which indicates the frequency of the occurrence of a patient spending a day or more NBM. Compared with the pre-FORWARD data, the pilot data included four instances where points were above the 3-sigma upper control limit, each time signalling special cause variation; a data pattern unlikely to be attributable to chance alone. Once a new 3-sigma control limit had been calculated from the pilot data, there were no further signals to indicate special cause variation in the post-pilot phase. These findings indicated that there was a significant change after the FORWARD pilot, but not between the pilot and post-pilot periods. In statistical process control methodology ‘rational subgrouping’ describes separating data into meaningful subgroups to detect patterns. In Figure 2b, rational subgrouping by ward demonstrated post-FORWARD improvements independently across surgical, medical, older persons’ unit and stroke unit wards.

Figure 2.

a) G-chart showing number of consecutive ‘successes’ (i.e. NBM < 1 day) between ‘lapses’ (i.e. NBM ≥ 1 day) before and after FORWARD. Dark shading = pre-FORWARD; lighter shading = FORWARD pilot. Means are shown by solid lines and 3-sigma upper control limits by dotted lines derived from pre-bundle (lower 2 lines) and pilot data (upper 2 lines). Lower control limits have negative values and are not shown. (b) G-chart analysis showing number of ‘successes’ (i.e. NBM < 1 day) between ‘lapses’ (i.e. NBM≥1 day) before and after FORWARD with rational subgrouping by ward. Dark shading = pre-FORWARD; lighter shading = FORWARD pilot. OPU = older persons’ unit.

During the study period, in wards where FORWARD was introduced, hospital coding data demonstrated increases in rates of aspiration and hospital acquired pneumonia by 0.8% and 1.0% per year respectively but these were less than the increase in rate of dysphagia (1.6% per year) in an increasingly multi-morbid patient population (with a 10% per year increase in patients with ≥10 co-morbidities).

Discussion

EDAR can help support comfort, dignity and autonomy to patients with permanent dysphagia but is associated with particular challenges. Care bundles have been shown to improve outcomes when timely and comprehensive task completion by different professionals is required [18, 19]. The FORWARD care bundle met the challenges of EDAR by giving clinicians a tool to ensure reliability in timely decision making, a multidisciplinary approach and advance care planning. The primary outcome measure, days NBM prior to EDAR, was significantly reduced between the pre-FORWARD group and the FORWARD pilot group and there was a further reduction as the care bundle became embedded as part of routine practice. G-chart analyses provided confirmatory evidence that this effect could be classified as special cause variation. The analysis also showed that performance was sustained at 3 years after the bundle’s introduction, even as its use spread to clinical areas with less experience managing patients with dysphagia. Rational subgrouping confirmed that the effect held independently across ward types, refuting the idea that the results could have occurred through selection bias by ward and supporting the notion that the care bundle could be used to support clinicians for whom EDAR may be a less familiar strategy.

Process measures also showed significant and sustained improvements, exemplified by very high rates of documentation of capacity assessment and discussions with next of kin being maintained. Completion of PEACE documents improved, but only after the pilot period, perhaps as a result of feedback to clinicians. Our balancing measure supported the idea that FORWARD did not alter the culture of EDAR such that pneumonia became unexpectedly common; if anything the trend was in the opposite direction with pneumonia rates climbing slower than the corresponding dysphagia rate, even in an increasingly multi-morbid patient population.

Introducing new processes of care is challenging and sustaining those processes and healthcare professionals’ interest in them may be more so [20]. This study demonstrated sustainability of FORWARD across time and as its use expanded in different clinical areas. Clinical engagement was supported by a PDSA approach where alterations to processes of care in direct response to stakeholder feedback were made while maintaining a structured experimental approach [21]. Further strengths of this study were the stability over time of the positive FORWARD outcomes demonstrated by G-chart analysis as well as FORWARD’s safety demonstrated by our balancing measures. The generalisability of the results, however, needs to be tempered with the single centre nature of the study and the fact that the hospital benefits from a culture already supportive of innovative care bundles [14, 22]. Limitations also include the small number of EDAR patients pre-FORWARD and the fact that our balancing measure depended on the accuracy of clinical coding. A single group time-series study design may additionally introduce undetected confounding factors [23]. Since the study period the Royal College of Physicians (RCP) and British Medical Association (BMA) have published guidelines on clinically assisted nutrition and hydration which highlight a General Medical Council mandated requirement for a second opinion in cases where these interventions are not offered or withdrawn, in a non-dying patient’s best interests [24]. Further work is now required to incorporate these requirements into the FORWARD bundle and ensure this translates into a change in practice.

Conclusion

The use of the FORWARD bundle in this setting enshrined the principles of good quality and consistent management of EDAR, where interdisciplinary communication and logical process are of key importance. It resulted in less time NBM in advance of an EDAR decision and improvements in a number of process measures, with no globally increased aspiration risk compared with usual ward-based EDAR management. These changes were seen in clinical areas further afield than elderly care and stroke wards where EDAR is less common. It is therefore likely that FORWARD safely contributed to the comfort and dignity of patients who were supported by it, wherever that was. Further work is required to incorporate RCP/BMA guidance, obtain systematic patient/carer feedback and to assess the use of FORWARD in other clinical contexts including people living with an unsafe swallow in the community.

Supplementary Data:Supplementary data mentioned in the text are available to subscribers in Age and Ageing online.

Declaration of Conflict of Interest: None.

Declaration of Sources of Funding: The views expressed in this publication are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health.