"Care home residents five times more likely to be left thirsty," The Independent reports after an analysis of some London hospital admission records found people admitted from care homes were five times more likely to be dehydrated than people coming from their own homes.
Equally serious was the discovery that dehydration at admission was associated with a higher risk of dying while in hospital.
Much of the media seized on anecdotal reports that dehydration was the result of staff restricting access to fluids so residents were less likely to wet themselves during the night or ask to go to the toilet.
But anecdotal reports cannot be proved and, in terms of evidence-based medicine, don't hold high value.
The study did not explore, or provide any hard evidence of, why care home residents are more likely to be dehydrated.
While it would be complacent to discount suspected poor standards of care in certain homes, other factors may also be involved. For example, many people with dementia have reduced thirst and are reluctant to drink.
The truth is we do not yet know what is behind the higher dehydration levels in patients coming from care homes. Finding an explanation is the crucial next step.
Where did the story come from?
The study was carried out by researchers from Barnet and Chase Farm NHS Trust (London), the University of Oxford, and the London School of Hygiene and Tropical Medicine.
It was funded by a Wellcome Trust Investigator Award.
The study was published in The Journal of the Royal Society of Medicine, a peer-reviewed medical journal.
The media generally reported the findings of the story accurately, but many fell into the trap of reporting the study authors' speculation as if it was proven fact.
For example, the Daily Mail had the headline, "Lives of care home patients put at risk through lack of water: Staff 'don't want them going to the toilet at night'." This accusation is unproven.
The reasons why patients were dehydrated were not investigated as part of this study. Plausible explanations were put forward by the study’s authors to explain their observations.
They also raised concerns about potential poor care standards, but none of this speculation is based on new evidence. Additional work is needed to find out the reasons behind this worrying statistic.
What kind of research was this?
This was a cross-sectional study looking at the risk of dehydration on admission to hospital in older people living in care homes, compared with those who were living in their own home.
The researchers state older people are at a higher risk of dehydration, and dehydration is associated with worse outcomes while in hospital.
They also say mild to moderate dehydration in older people is easily missed, and is often only detected once individuals are admitted to hospital and have their electrolytes measured, revealing sodium imbalances. Abnormally high sodium levels can be a sign of dehydration.
A study like this can tell us whether a person is likely to have been dehydrated on admission to hospital, but it cannot tell us why this was, as there are many possible reasons.
What did the research involve?
The study team were granted permission to analyse information already collected on 21,610 people over the age of 65 who were admitted to an NHS hospital in London over a two-year period in January 2011 to December 2013.
The team obtained data on patients' ages, the type of admission (emergency or planned), and whether they lived in a care home or their own home.
They also had information on whether the person was dehydrated when they were admitted to hospital and whether they subsequently died in hospital.
The main analysis looked for links between whether a person was admitted from a care home and dehydration and death.
The team used hypernatraemia (plasma sodium levels of more than 145 mmol/L) to measure dehydration. This measure of sodium levels in the blood is a pretty accurate indicator of whether a person has had enough water or not.
Certain conditions make hypernatraemia more likely, such as prolonged vomiting or diarrhoea, sweating, and high fevers with inadequate replacement of the fluid lost. Some drugs and hormonal conditions can also increase the level of sodium in the blood.
What were the basic results?
The results came in two parts. The crude results presented did not take into account any influencing factors (confounders), while the adjusted results did.
But these did not include the reason for the admission, only whether it was planned or an emergency.
Initial crude findings showed patients admitted from care homes had a 10 times higher prevalence of hypernatraemia than those who lived in their own home (12.0% versus 1.3%, respectively; odds ratio [OR] 10.5, 95% confidence interval [CI] 8.43-13.0).
From this, the research team worked out around one in three cases of dehydration on admission would be avoided if people who lived in care homes were properly hydrated (population attributable fraction 36.0%).
Of note, 61.9% of people in nursing homes suffered from dementia, which can make it challenging for carers to ensure residents are properly hydrated, compared with 14.7% of people in their own homes.
After accounting for age, gender, mode of admission and dementia, the adjusted results found care home residents were around five times more likely to be admitted with hypernatraemia than people who lived in their own homes (adjusted OR 5.32, 95% CI: 3.85-7.37).
Care home residents were also about twice as likely to die while in hospital (adjusted OR: 1.97, 95% CI: 1.59-2.45).
How did the researchers interpret the results?
The researchers' interpretation was simple and stark: "Patients admitted to hospital from care homes are commonly dehydrated on admission and, as a result, appear to experience significantly greater risks of in-hospital mortality [death while at hospital]."
This research showed older people living in care homes were five times more likely to be admitted to hospital with dehydration than patients who lived in their own homes.
The research team and media expressed great concern this might be a result of poor-quality care in care homes.
While the study was able to show there is a worrying variation in dehydration levels linked to care homes, it was not able to provide evidence to explain these statistics.
There are many possible explanations for these results, many of which are highlighted by the study authors and the media. This study does not provide any direct evidence supporting any of these explanations, which are speculative at this stage.
The analysis attempted to correct for the finding that people in care homes were slightly older, more likely to be admitted as emergency cases, and far more had dementia. This made a large difference to the relative risk, taking it from 10 times more likely to five times more likely.
This hints at the possibility that people in care homes may be more unwell or have more complex illness and care issues than people who live in their own homes, which may influence their ability to remain hydrated. This is an alternative explanation to the conclusion that the care provided by care homes is inadequate.
The analyses also did not take into account the reason why patients were admitted to hospital, which would have clarified this issue. It is possible these factors (residual confounding) and other unmeasured factors (bias) may still be influencing the results to some degree.
This type of study is useful in flagging up potential care issues for further investigation by care regulators. In the UK this job falls to the Care Quality Commission (CQC).
The Independent informs us that, "The CQC said ensuring residents get enough food and drink was central to their inspections of care homes," reassuring readers that, "Deputy chief inspector of adult social care in London, Sally Warren, said information on dehydration supplied by Dr Wolff [the author of this study] had been shared with local inspectors."
Analysis by Bazian
Edited by NHS Website
Links to the headlines
The Times, 16 January 2015
Daily Mail, 16 January 2015
The Daily Telegraph, 16 January 2015
The Independent, 16 January 2015
Links to the science
The Journal of the Royal Society of Medicine. Published online January 15 2015