In our latest report covering CQC’s troublesome Single Assessment Framework, we invited former inspector and registration assessor for CQC, Simon Cavadino to give his thoughts on why the new inspection regime is not fit for purpose, and why calls for inspectors to get out from behind their laptops and back on site are growing louder. The report also contains Neil Grant’s own thoughts on the subject in response to Simon’s comments.
You can view and download the full report of the framework HERE.
Why is CQC’s New Inspection Regime Not Fit For Purpose?
Running social care services is difficult. People who do so deserve much more respect than they often get. But care services can go wrong, occasionally seriously. Sometimes teams lack resources, manpower, skills and/or knowledge. They become overwhelmed. They cut corners. They make mistakes. That is why it is necessary to have strong, effective and supportive regulation. Inspection is a force for good, but only when it is done well. At the moment CQC is doing it badly.
To understand why, it is important to consider two questions in depth.
1) What is ‘care?’ 2) What is a CQC inspection rating?
Care is abstract. Care is felt and experienced, mostly in the eye of the receiver, but also in the eye of the caregiver and witnesses. ‘Feeling cared-for’ or ‘caring for another’ is an emotional state, involving love, respect, attentiveness, thoughtfulness, warmth, kindness, resilience and fortitude. It is a desire to alleviate both physical and emotional discomfort or anxiety. It is an attempt to plot the best course of promoting wellbeing through meaningful engagement and activity, while reducing risk. I could go on. The concept of care is infinitely complex and largely ineffable, inexpressible in numbers and not an idea that can be ‘audited.’ A CQC inspection rating is the human judgement (or opinion) of an inspector about the quality of care they witness. That is all.
There are inspection reports with poorly evidenced judgements, for which the criticism, “That’s just the inspector’s opinion,” is often made. There is an assumption in saying this that better evidenced judgements become ‘facts’ rather than opinions. But they do not. They are merely better evidenced judgements. There is a further assumption that if you have enough ‘facts’ (or data) then the ‘factually correct’ rating can be reached without the need for any human judgement.
In the world of social care this assumption is the most inaccurate and pernicious idea of all. Yet it is the assumption upon which CQC is basing its current dream of ratings based upon data, without the need for site inspections.
A CQC rating of ‘Good’ is not a fact (like 2+2=4 is a fact). A ‘Good’ rating is a description, based on a professional judgement, not a numerical score. For years, under CQC’s inspection system there were five ratings (one for each key question) and those ratings (descriptions) were combined into one overall rating (overall description). Those descriptions were based upon a multitude of conversations, records, observations and events; they weren’t just made up or invented on a whim. But, at their core, those ratings were human judgements, not numerical calculations. This actually worked remarkably well. Granted there were sensible debates to be had about the many edge cases and there were occasional judgements that were way off. But, for the most part, I have found on my travels that the ratings seem broadly correct in most cases.
In its desire to come up with a utopian desktop system that has a valid, up-to-the-minute rating for each home (a hubristic and impossible aim), CQC has made the incorrect assumption that the human judgement descriptors map exactly onto their equivalent numerical scores. They have assumed that every judgement (score of 1-4) reached for every one of the quality statements and evidence categories are valid numerical calculations that will aggregate to ‘correct’ overall ratings. They won’t and they don’t. CQC has not even considered the relative weighting and importance of each score (for example, are safe staffing levels more important than environmental sustainability?). CQC’s underlying assumption is untested, statistically unproven and I would argue it is nonsense.
In practice, it means CQC inspectors and assessors are now expected to complete hundreds of boxes of evidence and give 1-4 scores for each.
Inevitably, we are starting to see examples where CQC is unable to award the rating it wants to, due to the constraints of this invalid scoring system. The process is highly time-consuming and, I would say, pointless and counterproductive. CQC is falling miles behind with conducting necessary inspections.
While not perfect, the tried and tested method of an experienced inspector observing care, speaking with people, reading relevant live records on site and awarding descriptive ratings is the best way of reaching judgements that make sense to lay-people. Inspection reports should be written freely as a narrative of what was witnessed and contemplated, not a series of meaningless scores. CQC’s new approach is not fit for purpose because it fundamentally misunderstands what a rating actually is and how it is useful. Data is only useful in support of fundamental human judgements, not instead of them. CQC needs to dramatically simplify its aims and processes, appreciate what a rating actually is and then get inspectors out from behind their laptops and back on site.
Simon Cavadino
Director
The Woodberry Partnership
Neil Grant’s response:
Simon Cavadino’s article highlights the importance of the human element of regulation which I agree is missing in CQC’s new regulatory model – the Single Assessment Framework or SAF – with its focus on technology, data collection, remote working and the scoring of evidence.
In March 2018, The King’s Fund and Alliance Manchester Business School published a set of papers evaluating the performance of CQC over the period 2013-2017 which coincided with the introduction of ratings. The paper stated:
“Our work has highlighted the social dimension of regulation and the importance of relationships between CQC staff and those working in health and social care provision, and suggests a more transactional approach to regulatory intervention risks undervaluing and discounting the value of the soft, informal, influencing power of regulation. It suggests that however well conceptualized, designed and planned the regulatory model may be, its impact is fundamentally shaped by how it is experienced, how it is implemented, and the skills and values of those who are involved in regulation.”
Moving forward to the present day, I am struck by how inspectors now play a secondary role in CQC regulation. The importance of the inspector has been diminished by the creation of the role of the assessor. It is the assessor who reviews data on an ongoing basis about registered services. Inspectors are sent out on site visits, when required, with no ongoing relationship with providers in their area.
On 26 March 2024, Ian Trenholm, the Chief Executive of CQC published a blog entitled, “Reflecting on your feedback supports our ongoing improvement” in which he did not mention inspectors once. Instead, his focus was on the “use of data,” “the latest generation of technology,” “the provider portal” and, more broadly, “the transition.” Similarly, in his article in Caring Times in June 2024 about the implementation of the SAF entitled “How can it be better!?” he makes no mention of inspectors. It is all about making the tech-driven SAF work more effectively. Ideologically, it is apparent he is wedded to technology as the future of regulation.
So can inspectors and assessors still have a significant impact in this new world of regulation?
CQC will no doubt argue that human judgement still has a key role to play in regulation. Assessors and inspectors are supposed to make decisions about the focus or scope of an assessment. In reality, CQC has imposed rules on them in this regard. When it started to roll out the SAF from the end of November last year, CQC told assessors and inspectors to focus on priority quality statements as part of site visits. However, under the threat of legal challenge, on 2G March 2024, Ian Trenholm committed CQC to look at all quality statements under key questions rated Inadequate or Requires Improvement. Therefore, the autonomy of assessors and inspectors is severely limited by the CQC framework and the reactive decisions of its senior management.
Where human judgement may have some impact is in relation to the “opening out” of assessments beyond the original selected quality statements when wider concerns are identified. There is evidence that this has happened to a significant degree in the early phase of the SAF roll out which calls into question the predictive capabilities of the CQC model.
However, the main example of an element of human judgement coming into play is when assessors and inspectors award scores to the evidence collected as part of an assessment. The question is whether this is a true exercise of human judgement when what are often complex and nuanced matters are reduced to numerical scores. Some things cannot be scored, they need to be described and explained, as Simon articulates so impressively.
What is clear is the way the SAF has been implemented so far has led to some bizarre outcomes. Services that would have been rated Good under the old system have come out Requires Improvement, while others which would have been rated Inadequate have been rated Requires Improvement. Furthermore, services with several breaches of regulation have been rated Good. So whatever the validity of the “judgement” at the evidence category level, the overall scoring mechanism is the dominant force. The human element is subordinate.
The solution? I would revert back to Sir David Behan’s model of regulation based on the KLOES, ratings characteristics and scheduled inspections supported by enhanced monitoring between inspections. In particular, I would abolish the assessor and inspector split and place the inspector back at the centre of regulation. Sounds simple but when CQC has spent millions and millions on the SAF, I cannot see the retreat happening anytime soon or in any way voluntarily. Sadly, it may take a scandal like Winterbourne View to shift the dial back to an inspection-based model of regulation where human judgements matter, not arithmetical scores.
Neil Grant
Partner
Gordons Partnership Solicitors