Improvement action plans that actually move the rating

Regulation 17 of the Health and Social Care Act 2008 (Regulated Activities) Regulations 2014 requires providers to have an effective system for assessing, monitoring and improving the quality and safety of the services provided. The improvement loop is the improving half of that clause. The regulation does not prescribe a format for action plans; it expects the evidence to be there when an inspector samples.

For services rated Requires Improvement or Inadequate after a CQC inspection, the action plan response carries specific weight. The provider is expected to publish an action plan against the rating findings, and the inspector returns to sample whether the actions have moved practice in the time elapsed. The route from Requires Improvement back to Good runs through actions that close with evidence, not actions that close with sign-off.

For mock inspections (the consultant-led pre-inspection review), the deliverable action plan inherits the same shape: named action, named owner, specific due date, evidence of closure at completion. Verivius runs the same lifecycle whether the action originated from an incident, a complaint, a safeguarding outcome, a mock inspection, or a real inspection report.

Across the inspection portfolio Klaudiusz worked over thirteen years inside CQC, three improvement-action patterns surfaced more than any others when sampling a service moving from Requires Improvement back towards Good.

One: the action is too general to be verifiable.An action reads “improve documentation of consent” or “strengthen governance arrangements”. When the inspector returns six months later and asks how the team knows the action is complete, the response is a paragraph describing intent rather than an evidence trail. The action needed to be specific enough that closure is unambiguous: “by 31 March, all elective surgical consent forms include the financial-consent section, audited monthly by the registered manager, target 100 per cent compliance”.

Two: actions get listed but ownership is implicit.The plan shows the action; the owner column says “clinical team” or “all staff”. When ownership is collective, accountability is absent. The action that moves practice is the action with a single named owner who can answer the question “is this done?” without consulting anyone else.

Three: the closure paragraph describes completion, not change.“Training delivered to all staff on 15 February” is completion. It is not change. The closure paragraph that actually moves the rating describes what is now different: “Since the training, the three-monthly consent audit shows compliance moved from 78 per cent to 96 per cent; the falls incident rate is unchanged but the documentation rate on the falls assessments is up”. Completion without change is paperwork. Change is the inspectable evidence.

On a return inspection after a Requires Improvement rating, the inspector reads the action plan against the original report findings. The expected pattern is one action (or a small set) per finding, each specific, owned, dated, and either completed-with-evidence or in-progress-with-trajectory. Actions marked completed need to carry the evidence of completion: the policy with the revised date, the audit result post-training, the incident trend showing the new control is working, the named change to practice.

Actions marked in-progress need to carry a credible trajectory: what is the next milestone, when is it due, what is the evidence the work is genuinely underway versus a placeholder. An action plan with every line marked in-progress at the return inspection is itself a finding; the inspector reads it as evidence the service is going through the motions rather than making the change.

For sampled actions, the inspector typically goes from the action plan back to the source: the original incident, complaint, audit finding, or report quote that prompted the action. The reading test is whether the action targets the actual cause. An action that addresses a symptom rather than the cause looks plausible on paper but is the pattern most often associated with a return rating that does not move from Requires Improvement.

At the aggregate level, the inspector also looks for momentum. A service that closed twelve actions in the last quarter shows a different rhythm to one that closed two. Both can be defensible depending on the scale of the underlying findings; the inspector reads the rhythm against the report findings.

Verivius runs improvement actions as a closed lifecycle: create (linked to the source record, with a single named owner, a specific due date, a priority grade), update (with progress notes), complete (with named change-to- practice entry, not just “done”), close. Actions spawned from incidents, complaints, safeguarding outcomes, mock inspection findings, or inspection reports all use the same lifecycle so the audit trail is consistent. Linked actions surface on the source record so the original incident or complaint evidences whether the closure loop completed. The aggregate dashboard shows the open-action backlog and the closure rhythm over the trailing quarter. For the full feature walk-through see what Verivius actually does.

See also the Day-to-day use section on the FAQ for the short answers across every lifecycle.