Hospital Quality Alarms: How deficits, patient flow and access are affecting quality

I’m certain that there are countless components that underlie the running of a good hospital. However, as I had the opportunity to discuss in a very spirited conversation with several NHS executives this week, I believe that there are three fundamental elements which provide the foundation for good hospitals the world over. In a manner not unlike going into space, it seems to me that there are a few vital ingredients for success. I have had the unfortunate opportunity to see some excellent hospitals go into free fall as a result of not getting the golden triangle of healthcare right.

So, what are the three things which form this golden triangle? Well, over the years I have consistently experienced this, and as such believe it boils down to these three factors: hospital finances, patient access, a.k.a. good flow, and workforce, linked to quality. These components must always be aligned, and at times they seem almost impossible to balance. I have seen this happen over and over again. When hospitals struggle with their finances and poor patient access, quality suffers in turn and ultimately leads to harm. We currently have a majority of hospitals in the country running at a deficit and struggling to find ways to deliver against their cost improvement programmes. In parallel, we see an enormous number of organisations finding it impossible to deliver against all of the core access and flow standards, from A&E performance, which continues to decline, to cancer access and care, which is down at an all-time low. Most have stopped talking about Referral to Treatment waiting times (RTT) or are just not reporting it, with more and more patients stuck in beds for longer lengths of hospital stay. This normalisation of ‘failure’ is worrying but, when we then compound this with the unprecedented clinical workforce issues seen globally – “Houston, we really do have a problem”.

I fear that, without implementing some significant changes, we will start to see more and more patients receiving substandard care and coming to harm. In some places, we are already seeing quality slip, with greater numbers of patient falls, sepsis, delayed discharges, readmissions, and more. And for those of you who recall the Mid Staffordshire mortality issues back in our not-so-distant past, you will also recall that the ingredients were not dissimilar to those found in the Keogh review. Targeting 14 trusts that displayed mortality rates too unusual to be ignored, this report examined the critical components of quality care and treatment, eventually determining that 11 of these trusts required submission to ‘special measures’ in an attempt to curb the worrying trends in their performance. Unfortunately, these issues seen back in 2013 do not appear to have been curbed in our present-day trusts.

Just this week, more details of the shocking 40 year-long and still ongoing failures at a maternity department of an NHS trust have come to light. Described as “the worst ever NHS maternity scandal”, the trust has been found to have cultivated a culture of “toxic” treatment towards mothers and their babies. Most worrying of all, however, has been the number of avoidable deaths and children suffering from permanent disabilities that have stemmed directly from the insufficient level of care provided to them. Unfortunately, these cases of patient harm are indicative of failings in one of the core components of the golden triangle, workforce. Whether this arises from understaffing or the resulting pressures, it is evident something needs to be done to address and correct the way staff are operating at the trust.

As horrifying as the findings at this trust’s maternity ward are, it is unfortunately not an isolated case when it comes to avoidable deaths within the NHS. November also saw the release of trends in weekend death rates at another well-known NHS foundation trust. While 2018/2019 figures are not yet available, the 2017/2018 period saw a deficit of £62 million against a planned £38 million deficit, showing a staggering £24 million variance against plan, as well as a significant increase from the £43 million seen in 2016/2017. Moreover, staff retention is below the national average at 82.9% compared to a national median of 85.7%, justified by the trust as resulting from a focus on recruitment rather than retention due to high vacancy and agency usage. The trust’s deficit is one of the largest across the entire non-specialist acute trust sector, and alongside staffing inadequacies, subsequent patient outcome measurements appear to reflect this. The 2018/2019 reporting period saw 1,427 deaths, leading to 293 case record reviews, 24 investigations, and 3 patient deaths ruled most likely due to failures in the care provided to the patient. In the same period, 9.9% of patients aged 16 and over were readmitted within 28 days, 20 patient safety incidents lead to severe harm or death (0.6% higher than the national average), and the percentage of patients treated or admitted within 4-hours at the emergency department dropped by 5.51% from the previous year, performing significantly below national averages.

Of specific focus recently at this trust has been an in-depth examination into the specific HSMRs seen across the week, which found that weekend figures have risen to 117.7, in sharp contrast to the 101.9 seen on weekdays. This discrepancy is even further compounded in the case of frail patients who show a HSMR of 130 when admitted at the weekend, drastically higher than the 105 seen when admitted on a weekday. These HSMR scores reflect a significantly greater proportion of patients coming to harm at the trust, with weekend mortality rates sitting at 17.7% and 30.0% higher than the NHS average overall and for frail patients, respectively. As such, the number of patients ultimately coming to harm over this period is unacceptably high and inconsistent with the average performance of NHS trusts as a whole. While no avoidable patient death should be tolerated, these numbers are particularly distressing and require immediate intervention. By and large, the figures have been attributed to the lack of an acute frailty service at the weekend, leaving patients extremely vulnerable and at risk of death just by being unlucky enough to require medical attention on a Saturday or Sunday. While this lack of “wrap around” services in the community at the weekend has been posited as a significant contributor to the less than favourable outcomes, it is likely not the full story. NHSI itself has also initiated a review into the trust’s mortality outcomes, both to add to the current exploration of weekend death rates and to examine palliative coding practices. They believe a lack of palliative care services may mean that patients who may be older, have greater severity of illness, and therefore require such services are instead defaulting to a hospital rather than potentially being better treated at home. Again, whilst a possible contributor to the story, it cannot account for all of the trust’s failings. Indeed, this was one identified and put under ‘special measures’ by the original Keogh report. While some improvements have been seen following the Keogh review, 6 years have passed and the remaining need for drastic improvement to protect these patients is clear.

The issue of data confidence, clinical coding and its impact on HSMR levels is certainly not a unique one. Accurate data, insight and clinical engagement in the healthcare setting is crucial, and underpins delivery of payment, performance, and patient outcome measures. Without it, the ability of a trust to improve its services and cost effectiveness, to compete in the market, and to operate within guidelines is severely hampered. Much of this relies on the coding department extracting data from medical records, often produced by junior clinicians or those under extreme time pressure. This has an additional downstream effect on mortality rates, as the severity of a patient’s condition prior to death is not always fully taken into account. Further to this is its influence on patient outcomes, due to the consequence of incorrect complexity and co-morbidity reporting on the accuracy of diagnoses.

Preventing future cases such as those mentioned in this article is a multi-faceted journey. Correct clinical coding – although incredibly important both to the trust and to patients – comprises only one of several changes that hospitals must begin to make in order to drive quality and safety. With over 9,000 avoidable deaths across the NHS, rising yearly, it is clear that new practices need to be introduced to our healthcare trusts. The current financial, patient flow and workforce challenges trusts face are having devastating consequences on patients and their families, we mustn’t lose sight of the quality of care under such demands.

Alongside the golden triangle, working as a critical metric and smoke alarm in identifying organisational issues, and often going ignored or under-utilised at the board or executive level, is feedback. Whether this is patient feedback, broader staff feedback, or junior doctor survey feedback, the key to ensuring that potential issues are discovered before they become truly problematic is listening to what is being said by the people at the “frontlines”. Experience shows that those doing the job daily are the ones who know best how it can be done better. At the moment, much of a trust’s awareness of the situation at their organisation is based upon a single, annual survey. As you might expect, people are often reluctant to complete such a task, and often the resulting feedback is a “one-off dip” that likely is not even particularly accurate. This leaves a wealth of untapped knowledge and experience within healthcare centres that could otherwise be the transforming factor for a hospital’s functionality. At the heart of this is therefore an organisation’s ability to listen to those most qualified to assess, in order to drive quality and patient experience. Currently, there are few if any that aren’t doing this at a suboptimal level, and this will be an important driving force for future improvements.

Clinical teams are under constant pressure to deliver more productivity and can therefore be disconnected from data that is essential to optimal patient diagnosis, treatment, and outcome. There is a glaring need to engage these teams in the data discussion, and to provide them with greater visibility and understanding of the vast amounts being generated on a daily basis in our hospitals. The way things currently stand, clinicians simply do not have the time or capacity to really stop and look at this data, and the crucial things it could be trying to tell them. With thousands of patient interactions at healthcare centres, the clinical team sees only a fraction of the resulting information, and this is having substantial downstream effects on the patients themselves. What is needed are ways in which this data can be more efficiently captured, processed, analysed and subsequently facilitated, presented and engage both one to one and in groups with care teams. I am confident that, through this, we can begin to triangulate datasets and create modules for clinicians to monitor in real time how their patients are doing, what next steps are needed, and ultimately how to predict and prevent harm.