The perils of ‘risk feeding’

Abstract

Key points

• Some people judged at risk for aspiration and choking continue to eat and drink.

• ‘Risk feeding’ policies are becoming more popular.

• Such policies may validate and perpetuate myths about aspiration.

• There is also the potential for delays in providing food, fluid and medications.

• Further debate is needed before risk feeding becomes an established standard of care.

Introduction

‘Risk feeding’ (and variants such as ‘feeding at acknowledged risk of deterioration’ and ‘feeding with risk’) has entered the healthcare lexicon in recent years and policies and pathways promoting this approach have become common [1–11]. Risk feeding is generally used when people with an ‘unsafe swallow’ that is unlikely to improve continue to eat and drink despite a perceived risk of choking or aspiration. In some cases, this is the person’s choice; in others, it may be because tube feeding is not clinically indicated or because eating and drinking is judged in the person’s best interests.

We argue in this paper both that the term is inappropriate and that the concepts behind risk feeding are often poorly defined and may be counterproductive.

Words matter

‘Feeding’

Animals feed, and there is feeding time at the zoo. Babies also feed and may need to be burped after feeding. Other people eat and drink even if they require some assistance to do so. ‘Feeding’ is unintentionally demeaning language if referring to the person who is eating and drinking rather than to a healthcare technique, such as ‘comfort’ hand-feeding people with advanced dementia or tube feeding [12, 13].

‘Risk-anything’

Risk can be a neutral term meaning probability but has now become synonymous with danger. The pervasiveness of the language of risk in healthcare now may encourage a risk-averse attitude among staff and disregard of what matters to patients [14, 15]. If risk feeding, why not ‘risk walking’ or ‘anticoagulation at acknowledged risk of bleeding’? ‘Risk-anything’ is not how healthcare (or other) decisions are reached: instead, people need to be informed of the potential benefits and risks of any proposed interventions and can then weigh up this information in the context of their preferences and goals and decide how to proceed.

Acknowledging the problems

Patients with dysphagia are often unnecessarily designated nil-by-mouth (NBM) due to ‘clinical indecision’ [1], and this can exacerbate dehydration and malnutrition. Staff anxiety about aspiration of food or fluid is common, and people may fear being criticised if there is an adverse outcome. It is important that dietary plans for dysphagic patients are documented and communicated to nursing and catering staff among others. Some decisions regarding the best route for providing hydration and nutrition are difficult [4, 16], and decision-making guidelines may be helpful in such cases.

Specific issue with ‘risk feeding’

‘[T]he fear of the fluid ‘going the wrong way’ was sometimes being used as an excuse by staff to withhold oral fluids, rather than being a practical decision based on the evidence of choking symptoms in the patient. Although patients, their relatives and carers may fear the prospect of choking, it occurs very rarely and is easily managed. Such a risk seems minor in comparison with the far greater distress caused by the person’s thirst’ [17].

Lack of clarity about the nature, frequency and severity of different risks

‘Risk-anything’ is meaningless without reference to the nature, frequency and severity of the risk of that activity for an individual. Risk-feeding policies aim to maintain oral intake while accepting risks of, variously, choking, coughing, aspiration, aspiration pneumonia and of an ‘unsafe swallow’. While overlapping, these risks are not identical. It is necessary to distinguish between the effects of oral intake on patients’ symptoms and experiences and on risks to life.

It is important to minimise patient distress and maximise pleasure. While distress and pleasure can often be regarded as being at opposite ends of a single symptom continuum, this is not always the case: sometimes people will both be distressed by food or fluid going down the wrong way but be desperate to eat and drink. Most patients will only want to eat and drink if the pleasure and relief afforded outweighs any distress (although some people with severe short-term memory impairment may be unable to recall their prior distress). Choking and coughing are distressing symptoms, and if they are frequent and severe enough, for a given dietary consistency and at a particular time, people will not want to consume that diet.

‘Risk feeding’ implies that eating and drinking for relevant people is particularly hazardous and that the hazards could be reduced or eliminated if feeding was avoided. Is this true? Most people will accept that asphyxiation is a particularly unpleasant prospect, and it seems sensible that a healthcare professional faced with a patient who, for example, despite severe dysphagia and a history of near-asphyxiation, insists on eating an unmodified diet will be particularly careful about documenting their advice that food should at least be provided in bite-sized chunks. The emphasis on aspiration, however, is more problematic.

Perpetuating misperceptions regarding aspiration and aspiration pneumonia

‘We are obsessed with aspiration. Aspiration itself does not cause infection’ (Paula Leslie quoted in Ref [18]).

A common belief is that aspiration may lead to potentially fatal aspiration pneumonia and that interventions will reduce the risk of pneumonia. However, although confirmed aspiration on an instrumental study is associated with a substantial increase in pneumonia, there are several reasons for questioning a direct relationship between aspiration of food and fluid and pneumonia (Table 1) [19–26]. The fact that a dysphagic patient develops pneumonia, even repeatedly, does not mean they necessarily have ‘aspiration pneumonia’. NBM status or feeding tubes (or modified diets) cannot prevent oropharyngeal secretions colonised by pathogenic bacteria being inhaled into the lungs, and it seems unsurprising that these interventions have not been shown to reduce pneumonia [21].

Need to challenge rather than validate unnecessary clinical indecision

People need food and fluid, and the only long-term alternative to eating and drinking is tube feeding, usually by gastrostomy. While this may be a reasonable option in some situations, in others, such as in advanced dementia, it is not usually clinically indicated [23]. Some decisions regarding oral intake in dysphagic patients are difficult, [16] and delay during which a patient is NBM may be inevitable. Once, however, as many risk feeding policies specify [1, 2, 8], tube feeding has been judged inappropriate, there is no alternative to oral intake, irrespective of whether there may be associated risks. If people are being left NBM in these circumstances, this should be challenged, and it may be preferable to ask those wishing to restrict oral intake sign a ‘risk of not-feeding’ declaration.

Need for individualised and flexible decision making

Labels such as NBM and for (or against) risk- or tube-feeding suggest that a fixed decision has been made. However, many people’s swallowing abilities and preferences fluctuate, sometimes from hour to hour, and staff need to have, and be encouraged to use, common sense, flexibility and judgement in these circumstances. Even if someone is ‘for’ risk feeding, it is often a bad idea to give them oral food and fluid at a time, for example, where their level of consciousness is reduced. Similarly, people who are tube fed to ensure an adequate intake may still enjoy some food and fluid.

Potential for delayed decision making

One study found that introduction of the FORWARD (Feeding via the Oral Route With Acknowledged Risk of Deterioration) bundle resulted in a reduction in the median time patients spent NBM (without tube feeding) from 2 to 0 days [1]. Although this would be a clinically significant reduction in time NBM, this was a small and single institution study, and further evaluation is required before this or similar approaches become widely used.

‘Another aspect of Mother’s case… was the absolute refusal by the hospital to allow her to have her seizure medication on the evening before this [fatal] seizure… The hospital was adamant she was not allowed to swallow anything because she might aspirate’ [27].

There is also the potential for delays in providing food, fluid and medications if meetings must be held and risk-feeding paperwork completed and signed by someone with the necessary seniority and confidence, even after any judgement about the inappropriateness of tube feeding has been made. The acknowledgement that for some patients ‘the risks [associated with oral intake] will be small or putative…’ would raise concern that staff anxiety might drive extensive requests for use of such policies [2]. Also, if it is accepted that hospital patients with an unsafe swallow who eat and drink need a special pathway, why should this not apply to patients in settings such as nursing homes? This may lead to increased periods when people are NBM unnecessarily in such settings where there may not be the same access to speech and language therapists to ‘sign off’ on eating and drinking.

Conclusions

If tube feeding is not appropriate for those with dysphagia, oral food and fluid must be provided. The widespread misperception of a clear and preventable link between aspiration while eating and drinking and pneumonia in those with dysphagia may lead to such patients being kept NBM unnecessarily. This requires education of staff. Although risk-feeding policies are well intentioned in seeking to avoid unnecessary restriction of eating and drinking, they might serve to validate and perpetuate unwarranted professional anxieties. Further debate and discussion is required before risk-feeding policies become an established standard of care.

Declaration of Conflicts of Interest: None.

Declaration of Sources of Funding: None.

References