The puzzle of why excess weekend deaths are difficult to research?

It has been observed across many healthcare systems that there are excess deaths in hospitals for patients admitted at weekends. In the United Kingdom, BBC Radio 4 ran “The Report: 7-Day NHS” last Thursday looking at recent research into “excess deaths” at the weekend and the link being made by Government Ministers to “avoidable deaths” and consultant and junior doctors weekend cover arrangements.  That any link is hypothetical and not proven seems clear … and then it becomes political theatre.

The interesting and significant question, unless you are a politician, is why it’s so hard to identify common causes for the excess deaths ?

Freemantle et al have shown convincingly that excess deaths are occurring, it is a significant effect, and that they cannot be explained away through differences in emergency vs. elective mix. They used nationally collected data for the analysis. It looks like a real effect and the search is on to find the cause.

The HiSLAC (High-Intensity Specialist-Led Acute Care) study has been funded to:
evaluate a key component of NHS England’s policy drive for 7-day services: the intensity of specialist-led care of emergency medical admissions, with a particular focus on weekend provision. This research is important for patients and for NHS strategy because it offers a unique opportunity to evaluate the impact of the transition to seven-day working, and to understand factors likely to impede or enhance the effectiveness of this change in practice.

It is important research and should enable future policy to be more evidence-based than is currently the case. Interestingly their website says that the initial research is being carried out by reviewing paper case notes from a number of case study hospitals.If their approach is going to be similar to that of the Global Trigger Tool, developed by IHI to detect adverse events, then it is very labour intensive process and applies practical limits to the size of research population that can be studied.

The downside of applying 19th Century technologies in the 21st Century!

The limitations of having to work with paper case notes means working with small samples (100s or 1000s rather than millions) and makes it that more difficult to identify any common causes unless the association is very strong – and the arguments over possible causes suggest that this isn’t the case. This is a stark contrast to the situation in other industries and the consumer sector where “big data” means that such questions are much more easily answered using purely digital data.

Increasingly there are hospital digital data sets being captured that should be relevant to this question. A lot of hospitals have used money from the nursing technology fund to implement systems to collect patient’s vital signs and record NEWS scores and other assessments for VTE, Sepsis etc. There are a number of hospitals with electronic patient record systems with clinical notes on their patients. The data from these systems could be analysed more quickly than paper case notes and covers the full patient population. Most of these systems are being used for clinical management and secondary use, for research for example, is not a priority in the typical hard pressed NHS trust.

There would be great value in establishing standards for the collection of clinical data and an “open data” policy which would enable important policy questions to be addressed through “big data”.

Planning for the future
We shouldn’t be making policy based upon opinion, rather than evidence, in the 21st Century. We should apply the data-driven approaches and supporting technologies being used successfully in many policy areas to inform decision making in healthcare.