Frailty and the patient journey

Frailty of old age and deprivation are two of the most common determinants influencing a patient’s likelihood to be admitted to A&E^​1​. There is a recognised geographical correlation between advanced age and social deprivation^​2​. It is known that both of these factors increase the risk of polypharmacy^​3​ and that complications relating to polypharmacy are among the most common causes of admission^​4​.

However, unlike other risk factors for morbidity — for example, hypertension, diabetes, bone health or even dementia — no single discipline or profession specifically attends to polypharmacy, and frailty is often overlooked as a pathology in its own right^​5​.

Frailty is under-recognised by healthcare professionals^​6,7​; however, as trusted custodians of medicines — whether in primary care or in performing medicines reconciliations in secondary care — pharmacists are ideally placed to spot common patterns of polypharmacy and use these patterns to identify and highlight frailty syndromes.

Some of these patterns — known as “prescribing cascades”, involve an unrecognised adverse reaction of one drug which leads to the prescription of a second drug and in turn results in further adverse effects — will be readily apparent to experienced pharmacy professionals. Identifying and addressing these cascades can significantly reduce the burden of polypharmacy and improve patient wellbeing^​8,9​.

This article will track the course of the patient journey following a frailty-related hospital admission. The opportunities within that journey for medicines optimisation and rationalisation will be considered, including how these might be safely achieved and how the pharmacy profession can lead in the area of problematic overprescribing and take ownership of active deprescribing in older people.

Case presentation

The patient is an 81-year-old male, Ray, who was referred directly from his GP to A&E with dysuria, dysphoria, intermittent confusion and hyponatraemia (128mmol/L). 

Ray had previously received three days of trimethoprim for a urinary tract infection (UTI), but the symptoms did not respond. After five days, he was prescribed a second course of trimethoprim. The trimethoprim was unfortunately replicated as the prescribing doctor in A&E did not have access to Ray’s primary care e-records and was unaware that this had already been tried unsuccessfully. 

He has also had two courses of nitrofurantoin, one course of cefalexin and one course of co-amoxiclav in the past three months, which were all used to treat the UTI. 

A dipstick urine test in A&E indicated nitrates, so a urine sample was sent to microbiology, and cefalexin was started empirically in the interim. Although diagnosis of UTIs by urine dipstick test is no longer recommended in patients aged over 65 years^​10,11​, owing to the high risk of false positives in this population^​12​, the practice persists and will need focused attention to rescind.

It is noted that Ray has a lying/standing drop in his blood pressure of -18mmHg. This is not associated with any symptoms. While this drop in blood pressure is not quite enough for a diagnosis of postural hypotension^​13​, it is close enough to be of concern.

Ray lives alone. His wife has advanced dementia and has lived in separate residential care for the past nine years, and his children live overseas. His cousin’s family live nearby, and they visit Ray each week or two. Recently, they have become more concerned about his overall health and vitality.

Ray has a history of: 

• Hypertension;
• Transient ischaemic attacks; 
• Gastro-esophageal reflux disease; 
• Benign prostatic hyperplasia;
• Post-traumatic stress disorder with panic disorder from having served in the British Army in Northern Ireland and the Falklands War. 

In this instance, there is not a direct pharmacy presence in A&E, so Ray’s medicines reconciliation does not occur until he is admitted to a ward. This incurs a delay of about 30 hours. The pharmacy technician on the admissions unit performed the medicines reconciliation using three sources to verify the history: GP records, pharmacy records and Ray’s own account. This is then assessed and verified by the ward pharmacist. Ray’s medicine history can be seen in Box 1.

Frailty syndromes and risk factors

Incontinence

On questioning, Ray does not consider himself to be incontinent. He is aware of when he needs to use the toilet and is capable of getting himself there on time. However, Ray does acknowledge that he sometimes experiences a little urinary leakage and often drinks less liquids in the second half of the day to avoid having to get up in the night to urinate.

Some pharmacy staff may feel uncomfortable discussing incontinence, believing that they lack the proper vocabulary to discuss the issues surrounding incontinence and have little to offer outside of pharmacological therapies. Experienced practitioners in frailty generally prefer to avoid such treatments, which can more easily manifest adverse effects in the frail and elderly population (e.g. oxybutynin or trospium), both of which carry a high anticholinergic burden^​15​. 

Conversations around urinary continence are potentially sensitive conversations. Empathy, rapport and trust-building are essential tools when discussing this and resources to help build confidence in these domains are available^​16​. However, pharmacists are well placed to identify medications that may contribute to incontinence (e.g. diuretics, narcotics, antihistamines), as well as facilitating onward referral to the medical and nursing teams. 

If a medicine is suspended or discontinued for urinary incontinence, pharmacists have a duty to ensure that there are measures in place to help maintain hygiene before the switch is made. In practical terms, this means making sure that ward staff are aware that there may be more frequent and more urgent requests for toileting from patients who are able to spontaneously vocalise their needs, while greater attention should be paid to those who cannot. It may also involve ensuring access to appropriate and comfortable incontinence aids and employing regular checks for skin conditions and integrity, as frequent incontinence can quickly cause soreness and excoriation.

Delirium

Ray’s acute admission with confusion is a good moment to screen for delirium using tools and resources, such as the British Geriatric Society’s delirium hub^​17​. Ray’s 4AT delirium screen scored 8/12 on admission, which is strongly suggestive of delirium but not diagnostic in itself. See ‘Recognition and management of delirium’ for more information.

Other possible causes must be sought for Ray’s excessive sleepiness and cognitive impairment. His family describe him as normally being anxious but otherwise bright and chatty, without evidence of cognitive impairment.

A hospital presentation and its associated factors (i.e. acute illness, uncontrolled pain, change of environment and disruption to normal routines and aids such as eye spectacles) could precipitate delirium^​18​. 

Non-specialised staff may be unaware that frequent changes of ward can increase the risk of delirium, which can prolong hospital admissions. Advocating early for admission to a dedicated bed may help reduce the incidence of delirium and improve overall patient flow^​18​.

Falls

Ray hasn’t been admitted with a fall this time; however, he does have risk factors for falling — for example, the combination of postural hypotension and urinary frequency. Postural drop in frailty is not dependent upon the presence of antihypertensives^​13​. Drowsiness from continued over-consumption of co-codamol would also be a risk factor.

Falls are one of the most preventable frailty syndromes and pose a considerable risk to morbidity and mortality. Ray’s admission to a ward should trigger a falls risk assessment, ideally through a comprehensive geriatric assessment^​19​.

Ray’s inpatient journey

Ray is admitted to a general medical ward, under the care of the ageing and complex medicine (ACM) team. This is a multidisciplinary team of doctors, nurses and pharmacists who specialise in the problems that attend frailty, polypharmacy and deprescribing. 

Another option may have been to admit Ray to a virtual ward. There is an argument that this may have been more suitable for him. Certainly, care in his home environment would have reduced Ray’s risk of disorientation, and his blood pressure could have been monitored remotely^​20,21​. However, there was a chance that Ray may also have required more invasive treatments alongside the imaging work required, which makes a short inpatient admission more suitable.

NHS England’s ‘2023/2024 priorities and operational planning guidance and the delivery plan for recovering urgent and emergency care’, published in 2022, lays out a plan to develop virtual ward capacity nationally to 50 beds per 100,000 of population^​22​. However, how this is currently achieved, staffed and monitored nationally is still quite divergent, and national standards of practice have yet to emerge.

Ray’s urine sample fails to confirm a diagnosis of UTI, while his acute inflammatory and biochemical markers (e.g. temperature, white blood cells and C-reactive protein) remain stable, as do his observations. Over two days of observations on the admissions unit, Ray is not pyrexic (i.e. raised temperature) at any point.

He is diagnosed with constipation, presumably secondary to his excessive use of co-codamol, and elevated liver enzymes, which are attributed to the paracetamol and/or his statin. Other investigations showed nothing abnormal. 

The ACM team decide that constipation, withdrawal from fluoxetine and adverse reaction to propranolol are contributing factors to Ray’s excessive sleepiness and cognitive impairment. Although Ray’s serum sodium remains low during admission (i.e. never going above 130mmol/L), his records show that this has been the case for the past 12 years. 

Other investigations for low sodium are normal. In addition, there is no evidence of syndrome of inappropriate antidiuretic hormone secretion, cortisol response is normal and there are no concerns over malignancy identified on chest or head imaging. Malignancies of these organ systems can cause low sodium^​23​.

The frailty pharmacist and consultant agree that Ray’s antihypertensive medications should be suspended for five days. During this time, the nursing team identifies a 35mm/Hg systolic deficit in lying and standing blood pressure. However, after five days without medications, Ray’s systolic blood pressure is peaking at 170mmHg in the morning, around 150mmHg from early afternoon onwards and around 130mmHg at bedtime.

Medication interventions

The following interventions are made by the ACM team and recommended for discharge.

The decision is made to withhold atorvastatin. The statin is a possible cause of Ray’s myalgia and deranged LFTs. There are no recent records of a lipid screen for Ray in the past ten years, and the only one available does not have his low-density lipoprotein measured, with all parameters being in range. Reducing total cholesterol to <4mmol/L and LDL to below 2mmol/l are important for the secondary prevention of stroke, which should be enacted because of his history of transient ischemic attack (TIA). It is not clear what statin therapy Ray was on, if any, when these readings were taken. The decision was made to measure creatine kinase (CK) levels for benchmarking, which is in accordance with NHS England’s ‘National guidance for lipid management for primary and secondary prevention of CVD’, published in 2023^​24​.

The decision is made to pause the statin, remeasure lipids and re-titrate as necessary, while encouraging Ray and his family to make a record of the severity of his aches and pains over this period. This will allow the pharmacist to reassess possible symptoms of statin intolerance.

In cases such as Ray’s, the guidance recommends measuring CK, holding the statin for four to six weeks, then — once symptom free — re-titrating at eight-week intervals. This can be done with the cooperation of his GP practice team, his primary care pharmacist and the secondary prevention of stroke team. If Ray does not tolerate statins on re-trial, he may be considered for other lipid-lowering therapies. These will be chosen in collaboration with Ray and will be influenced by which of his lipids, if any, can be modified most helpfully to reduce his risk of further cardiovascular events.

The other decisions regarding Ray’s medication and their rationale are summarised in Table 1^{​14,25–29​}.

Discharge

Ray makes a good recovery, and his 4AT score has come down to 2 (i.e. he’s still struggling to do the months of the year backwards, although he maintains that he would be unlikely to get this right at the best of times). 
 
There is an impetus to discharge him early to avoid hospital-associated infections but also to prevent decompensation or loss of physical capacity resulting from inpatient hospital stays, see here for more information.
 
Ray will need support at home. The next few weeks will involve a few follow-up appointments to monitor his recovery, as well as how he’s getting along with the changes to his medications.
 
This puts him on a discharge 2 assess pathway (D2A), which is recommended by NHS England^​30​ as someone with “new or additional health and/or social care needs on discharge”. There are four categories of discharge under this banner, see Box 4.
 

Ray would like to return home. This also makes most sense in trying to establish a useful baseline from which to make any assessments of his cognitive health.

Ideally, at discharge, Ray would be referred to the Discharge Medicines Service (DMS)^​31​. DMS was established in 2021 to allow hospital clinicians to refer patients, at discharge, to a local community pharmacy for extra guidance and support. In addition, the DMS, while not available at all pharmacies, is categorised as an essential service within the ‘Community pharmacy contractual framework’. However, access to these services is dependent upon uptake from discharging hospital teams, which is far from uniform across the UK^​32​.

In Ray’s case, the trust to which he is admitted has a pharmacist-led frailty clinic. The role of the pharmacist in this clinic is to assess patient progress after deprescribing, order and monitor any further investigations and follow-up tests that may be required, and to act as a reference point for colleagues in primary care to direct their queries regarding management of frailty and polypharmacy. Ray is scheduled a telephone appointment for post-discharge follow-up and is happy to know that the pharmacist in clinic will be able to liaise directly with the medical consultant and multidisciplinary team who attended to him as an inpatient so that his goals and wishes will be understood, and there will be continuity of care for him as he rehabilitates at home.

Conclusion

A sticking point for pharmacists and pharmacy services is who ‘owns’ the flow of information about patients in these dynamic environments and how safe dispensing and monitoring practices can be enacted. Decisions documented in either care sector may not be visible to the other. The NHS in the UK faces significant challenges, owing to the use of disparate software systems across different institutions^​33​. This lack of standardisation hinders integration, leads to duplication of efforts and increases clinical risks^​34​. NHS England acknowledges that multiple clinical systems currently do not interact with each other, resulting in repeated tests and a reliance on patients to provide accurate medical histories. Where these are successfully integrated, meaningful advances in patient care can be made.