Capacity, Consent and Risk: How to Support “Unwise Decisions” Safely and Lawfully
In adult social care, “unwise decisions” are where autonomy, risk management and safeguarding collide. The Mental Capacity Act makes it clear that people are allowed to make decisions others disagree with, as long as they have capacity for that specific decision. The operational challenge is proving you recognised a capacitous choice while still managing foreseeable harm. Strong services anchor this work in mental capacity, consent and best interests decision-making and ensure it reflects the wider core principles and values of person-centred care, dignity and least restrictive practice.
Commissioners and inspectors often see poor practice at both extremes: either services become risk-averse and restrict choice unnecessarily, or they “stand back” in a way that leaves unmanaged risk and weak safeguarding. The defensible middle ground is showing your working: how capacity was assessed, how risk was explored with the person, what proportionate safeguards were offered, and what you did when risk crossed into harm or safeguarding thresholds. This also links closely with wider positive risk-taking practice, where autonomy and protection must be balanced carefully.
What “unwise decisions” means in day-to-day practice
An unwise decision is not evidence of incapacity. Services must demonstrate they understand this distinction and can evidence it consistently. For wider context, providers may also find this guide to making lawful, person-centred decisions in safeguarding practice useful when reviewing staff understanding.
In practice, this means:
- capacity assessments are decision-specific and time-specific
- the person is supported to understand relevant information and consequences
- the decision is recorded clearly as a choice, not “non-compliance”
- risk is managed through proportionate safeguards, not blanket restriction
The key operational question is not “is this a good decision?” but “does the person understand, weigh and communicate their decision, and have we done what is reasonable to reduce foreseeable harm without removing choice?”
Operational example 1: refusing support with medication
Context: A person receiving domiciliary care chooses to skip medication on some days, stating they “don’t like how it makes me feel”. Staff are concerned about health risks and potential deterioration.
Support approach: The provider treats this as a decision point rather than a compliance issue. Staff explore capacity specifically in relation to understanding the purpose of medication and consequences of refusal.
Day-to-day delivery detail: Staff use simple explanations and check-back questions (“can you tell me what the tablets are for?”). They record the person’s stated reasoning and preferences, and offer alternatives: contacting the GP, requesting a medication review, and discussing timing or food intake. Staff agree a monitoring plan: observe for changes, record refusals, and escalate if health indicators worsen.
How effectiveness is evidenced: Care records show the person consistently demonstrates understanding and repeats their reasons coherently. The service evidences actions taken to reduce harm (GP contact, medication review request) while respecting the decision. Where refusals correlate with deterioration, escalation is documented as a health response, not a punitive measure.
Where the person’s decision-making appears uncertain, staff should also understand how mental capacity assessments are tested under scrutiny, particularly where health risks and safeguarding duties overlap.
Operational example 2: choosing to go out alone despite falls risk
Context: A supported living tenant with a history of falls wants to walk independently to local shops. Family members request that staff prevent this.
Support approach: The service separates family anxiety from the person’s rights. Capacity is assessed for the specific decision about independent community access and falls risk.
Day-to-day delivery detail: Staff support the person to weigh options: accompanied walks, mobility aids, agreed routes, and time-of-day planning. A risk enablement plan is co-produced: the person carries a phone, uses a walking aid, agrees a check-in protocol, and staff record any near misses. Family are involved in discussion but the plan remains person-led.
How effectiveness is evidenced: Records show progressive reduction in staff support over time as confidence and safety improve. Incidents and near misses are reviewed to refine the plan. This demonstrates proportionality and learning rather than either banning outings or ignoring risk.
If the person lacks capacity for a specific decision, services must shift from risk enablement to a lawful best interests process. This means recording best interest decisions with evidence, proportionality and accountability, rather than relying on informal restriction.
Operational example 3: relationships, visitors and safeguarding thresholds
Context: A person with capacity chooses to continue a relationship that staff believe is emotionally harmful or financially exploitative. The person refuses to end contact.
Support approach: The provider recognises that “unwise” does not automatically mean “safeguarding”. The service must evidence support, advice and monitoring while remaining alert to threshold changes.
Day-to-day delivery detail: Staff record conversations about risks and offer advocacy support. They check for coercion indicators, document what the person understands, and create a plan that reduces harm without banning contact (for example, support with financial boundaries, safe visiting arrangements, and agreed check-ins). If indicators escalate (coercion, threats, theft, inability to access money, fear), the service triggers safeguarding procedures and records the rationale.
How effectiveness is evidenced: The provider can show a timeline: initial capacitous choice supported with proportionate safeguards, then escalation to safeguarding when harm indicators changed. This demonstrates lawful respect for autonomy alongside effective protection.
Relationship-based risk requires particular care because apparent consent may not always be valid. Providers should understand how coercion and undue influence affect defensible safeguarding decisions before concluding that a person is freely choosing the situation.
How to document “capacitous choice” without creating safeguarding gaps
High-scoring tender responses and strong inspection evidence usually include a consistent recording approach. Services typically evidence:
- the decision being made and why it matters
- how the person was supported to understand information
- how capacity was assessed (and by whom)
- the person’s stated reasons and preferences
- agreed safeguards and what the person accepted or declined
- review triggers and escalation routes
This protects people’s rights while ensuring the service can demonstrate it acted proportionately if outcomes deteriorate.
Where capacity changes over time, records must show review rather than assumption. This is especially important in cases involving fluctuating capacity and changing safeguarding risk, where a person may be able to decide on one day but not another.
Commissioner expectation: proportionate risk management that does not default to restriction
Commissioner expectation: Commissioners expect providers to evidence how they enable choice while managing risk in a structured way. They look for clarity on capacity assessment practice, documented risk enablement, and escalation when risks become unmanaged harm.
Regulator / inspector expectation: clear distinction between incapacity, safeguarding and “unwise decisions”
Regulator / inspector expectation: Inspectors assess whether staff understand that an unwise decision does not equal incapacity. They also expect services to evidence safeguarding awareness: where a capacitous choice becomes coercion, abuse or neglect, the service must recognise threshold changes and respond appropriately.
Where a person appears to agree to something harmful, staff should test whether that agreement is genuine. The distinction between free consent and pressure is explored further in this article on when consent is not valid because of coercion or undue influence.
Governance and assurance mechanisms that stand up to scrutiny
Services strengthen defensibility through governance that tests real practice, not just policy. Effective mechanisms include:
- supervision prompts that review recent “unwise decision” scenarios and recording quality
- monthly audit sampling of capacity and risk enablement documentation
- incident review that checks whether capacity and consent were revisited after harm events
- escalation logs showing when cases were discussed with senior leaders or safeguarding leads
These mechanisms help prove consistency across teams and reduce reliance on individual judgement alone.
Outcomes and impact
Well-governed support for unwise decisions produces measurable outcomes: fewer blanket restrictions, improved trust and engagement, clearer safeguarding thresholds, and stronger defensibility when incidents occur. Most importantly, it protects the person’s right to live their life while ensuring the service can evidence reasonable steps taken to reduce avoidable harm.