When Domestic-Abuse Risk Assessment Becomes Institutional Authority
SAFECHAIN™ GOVERNANCE RESEARCH PAPER
The Institutional Risk-Assessment Assurance Problem™
When Domestic-Abuse Risk Assessment Becomes Institutional Authority
A SAFECHAIN™ Governance Analysis of DARA, Coercive Control, Professional Judgement and System Assurance
Publication Type: SAFECHAIN™ Governance Research & Systems Analysis
Research Domain: Domestic Abuse · Coercive Control · Policing · Risk Assessment · Safeguarding · Institutional Governance
Year: 2026
1. Executive Summary
Risk-assessment tools have become deeply embedded within institutional decision-making.
In domestic-abuse policing, those tools may influence:
Recognition → Risk Classification → Referral → Prioritisation → Resource Allocation → Safeguarding Intervention → Professional Perception → Outcome
Their governance significance is therefore substantial.
A paper published on 14 July 2026 in Policing: A Journal of Policy and Practice by Jacqueline Sebire and Matthew Bland critically examines the evidence base underpinning the Domestic Abuse Risk Assessment (DARA) tool used by police in England and Wales. The study examines DARA outside the original pilot environment and considers recording of coercive and controlling behaviour, patterns of risk grading, consistency between officers and associations with repeat victimisation. The authors expressly describe the research as exploratory and do not claim to establish predictive accuracy or causality. (OUP Academic)
The governance significance extends beyond DARA itself.
Once a risk-assessment instrument becomes embedded within institutional practice, its output can acquire authority far beyond the individual form on which the assessment is recorded.
The principal SAFECHAIN™ proposition is therefore:
A risk-assessment tool should never become a substitute for professional pattern recognition.
For coercive control in particular, danger may emerge through combinations of behaviour rather than isolated indicators. Surveillance, dependency, financial restriction, intimidation, isolation, escalation, restriction of autonomy and separation may acquire significance through their cumulative relationship.
A structured assessment may assist professional judgement.
It should not replace it.
SAFECHAIN™ therefore identifies a wider governance challenge:
The Institutional Risk-Assessment Assurance Problem™
The risk that an institution progressively treats the output of a structured assessment instrument as authoritative without maintaining sufficient governance over the evidence base, operational performance, professional interpretation, downstream reliance, limitations and continuing validity of that instrument.
This paper establishes a governance architecture for addressing that problem.
2. Current Evidence Context
DARA was endorsed by the College of Policing in 2022 as a replacement for the Domestic Abuse, Stalking, Harassment and Honour-Based Violence assessment, with improved sensitivity to coercive control forming part of the rationale for its introduction.
Sebire and Bland's 2026 study uses administrative data from a police force that did not participate in the original pilot evaluation. Their analysis examines changes in recorded coercive and controlling behaviour, patterns of risk grading, consistency in officer assessments and associations with repeat victimisation. They emphasise that the findings are descriptive and exploratory rather than evidence of predictive accuracy. (OUP Academic)
The study reports an increase in recorded coercive and controlling behaviour following DARA's introduction, while also examining differences in risk grading and officer assessment patterns. Its broader conclusion is that evaluating risk-assessment instruments at scale remains difficult and that national implementation should be accompanied by stronger evidence and continuing evaluation. (OUP Academic)
This concern sits within a broader contemporary debate. A separate 2026 paper examining ethical challenges associated with policing risk-assessment instruments identifies issues including bias, transparency, accuracy, fairness and governance, and argues for rigorous validation and adaptive oversight. (OUP Academic)
3. The Governance Significance
The presence of a structured assessment instrument can create institutional confidence.
The form looks systematic.
The questions appear standardised.
The output appears comparable.
The risk classification appears authoritative.
But standardisation is not equivalent to certainty.
A tool may have been:
developed under particular conditions;
piloted within a limited environment;
implemented differently across forces;
interpreted inconsistently;
influenced by recording practices;
affected by professional experience;
dependent upon information available at one moment in time.
The existence of a structured tool therefore does not remove the need for governance.
It increases it.
4. The Institutional Risk-Assessment Assurance Problem™
SAFECHAIN™ defines the Institutional Risk-Assessment Assurance Problem™ as:
The governance risk created when an institutional risk-assessment instrument acquires operational authority without equivalent assurance over its validity, implementation, interpretation, limitations, downstream consequences and continuing effectiveness.
The risk develops through the following pathway:
Tool Adoption → Operational Normalisation → Institutional Reliance → Downstream Decision-Making → Reduced Challenge → Apparent Certainty
At the final stage, a risk score may cease to be treated as one evidential input and begin to function as institutional truth.
5. SAFECHAIN™ Risk Assessment Integrity Principle™
No structured risk-assessment instrument should be treated as an autonomous determination of danger. Institutional risk assessment should combine structured evidence, professional judgement, pattern recognition, contextual analysis, affected-person evidence and an accessible route for escalation where the tool output appears inconsistent with the wider evidence.
6. The Tool–Judgement Distinction™
SAFECHAIN™ establishes three different levels of institutional assessment:
Level 1 — Structured Tool Output
The answers recorded and classification generated through the instrument.
Level 2 — Professional Risk Assessment
The professional's interpretation of the structured assessment together with wider information.
Level 3 — Institutional Risk Judgement
The final institutional understanding of risk informed by:
the tool;
professional judgement;
pattern evidence;
contextual evidence;
previous history;
escalation information;
affected-person evidence.
These levels must not be collapsed.
7. Tool Authority Inflation Alert™
Triggered where a structured assessment score progressively acquires more authority than the underlying evidence justifies.
Examples include:
downstream professionals relying solely upon the risk category;
risk grades being copied without reassessment;
historical classifications continuing despite changed circumstances;
professional concerns being overridden because the score is lower.
8. Risk Grade Certainty Alert™
Triggered where classifications such as:
standard,
medium,
high
are treated as objective descriptions of danger rather than professional or institutional risk judgements carrying evidential limitations.
9. Coercive Control Pattern Integrity™
Coercive control presents a particular challenge because individual behaviours may appear relatively ordinary when examined separately.
Risk may emerge through:
Frequency + Pattern + Dependency + Surveillance + Isolation + Escalation + Restriction of Autonomy + Consequence
The governance requirement must therefore include pattern recognition.
10. SAFECHAIN™ Coercive Control Pattern Test™
Ask:
Are behaviours being considered collectively?
Has the chronology been reconstructed?
Is escalation visible?
Is dependency present?
Is autonomy being restricted?
Is monitoring or surveillance present?
Has access to money, transport, communication or relationships been affected?
Has separation altered risk?
Do apparently minor incidents form a larger pattern?
Is the tool capable of representing that pattern?
11. Incident Fragmentation Alert™
Triggered where coercive or controlling behaviour is broken into isolated incidents in a manner that obscures cumulative meaning.
12. Pattern-to-Score Compression Risk™
SAFECHAIN™ defines Pattern-to-Score Compression Risk™ as:
The risk that a complex, evolving pattern of coercive behaviour becomes compressed into a simplified institutional classification that does not adequately preserve the pattern from which the risk arose.
13. Professional Pattern Recognition Standard™
Professionals using structured risk tools should retain the authority and responsibility to recognise risk that may not be adequately captured by the tool.
A professional should be able to record:
The structured output does not adequately represent the risk demonstrated by the wider pattern.
That judgement should trigger review rather than procedural resistance.
14. Professional Override Integrity Standard™
Where professional judgement differs materially from the tool classification:
the difference should be recorded;
reasoning should be documented;
escalation should be available;
supervisory review should occur where appropriate;
the tool should not automatically prevail.
15. Tool Deference Alert™
Triggered where practitioners believe they cannot depart from a structured assessment despite credible evidence of greater or different risk.
16. Institutional Reliance Chain™
Risk-assessment outputs do not remain at the point of assessment.
They travel.
SAFECHAIN™ identifies the following reliance pathway:
Initial Assessment
↓
Risk Classification
↓
Police Decision
↓
Safeguarding Referral
↓
Multi-Agency Review
↓
Resource Allocation
↓
Professional Perception
↓
Subsequent Institutional Decision
The farther the classification travels, the greater the possibility that the underlying evidence becomes detached from the score.
17. Risk Classification Inheritance Alert™
Triggered where later decision-makers inherit a risk classification without reviewing:
its date;
evidential basis;
limitations;
changed circumstances;
contradictory evidence.
18. Risk Assessment Freshness Standard™
Risk classifications should be understood as time-sensitive.
Institutions should establish triggers for reassessment following:
escalation;
separation;
new threats;
stalking;
technology-enabled abuse;
financial restriction;
breach of protective measures;
significant life changes;
new evidence.
19. Stale Risk Classification Alert™
Triggered where historic assessment continues to influence decisions despite material changes in circumstances.
20. False Negative Governance Risk™
A false-negative risk assessment may cause an institution to underestimate danger.
Potential consequences include:
lower safeguarding priority;
reduced intervention;
reduced monitoring;
missed escalation;
professional reassurance unsupported by reality.
21. False Positive Governance Risk™
A false-positive assessment may also have consequences through:
disproportionate intervention;
resource diversion;
inaccurate institutional records;
downstream professional assumptions.
The governance objective is therefore not maximum classification.
It is defensible, evidence-based risk recognition.
22. Risk Assessment Evidence Standard™
A risk classification should remain traceable to:
questions asked;
answers recorded;
professional observations;
contextual evidence;
previous incidents;
contradictions;
affected-person account;
professional judgement;
reasons for final classification.
23. Risk Evidence Traceability Test™
Ask:
Can an independent reviewer reconstruct why this risk classification was reached?
24. Score-without-Evidence Alert™
Triggered where a classification is visible but its evidential basis cannot be reconstructed.
25. Affected-Person Evidence Standard™
The affected person's account should not be treated merely as another data input.
Their evidence may reveal:
context;
meaning;
escalation;
fear;
dependency;
surveillance;
control;
changes in perpetrator behaviour.
26. Affected-Person Evidence Dilution Alert™
Triggered where lived evidence becomes fragmented into discrete form responses and its wider meaning disappears.
27. Fear Context Standard™
Where an affected person expresses fear, institutions should understand:
what they fear;
why;
what has changed;
what the perpetrator has previously done;
what they believe may happen next.
28. Institutional Pattern Recognition Standard™
Pattern recognition should not depend solely upon the individual officer conducting one assessment.
Institutional systems should enable comparison with:
previous incidents;
police intelligence;
previous assessments;
safeguarding referrals;
relevant orders;
recorded breaches;
other available evidence.
29. Historical Fragmentation Alert™
Triggered where relevant previous information exists within institutional systems but is not visible to the current assessor.
30. Tool Implementation Integrity™
A risk tool cannot be evaluated solely by examining its design.
Institutions must also examine implementation.
Implementation variables include:
training;
practitioner understanding;
time available;
recording culture;
supervision;
workload;
professional confidence;
technology;
local operational procedures.
31. SAFECHAIN™ Tool Implementation Test™
Ask:
Are practitioners trained?
Do they understand the purpose and limitations of the tool?
Is sufficient time available?
Are answers recorded consistently?
Is professional judgement encouraged?
Is supervisory review effective?
Is outcome data monitored?
Are implementation differences identified?
32. Tool Fidelity Alert™
Triggered where real-world use materially differs from the intended methodology.
33. Form Completion Fallacy™
Completion of a risk-assessment form demonstrates process completion. It does not, by itself, demonstrate that risk has been adequately understood.
34. Operational Validity Standard™
Institutions should test whether assessment instruments continue to perform effectively outside pilot or validation environments.
35. Operational Reality Test™
Ask:
Does the tool perform in routine practice as expected from its development or pilot environment?
36. Pilot-to-Practice Gap Alert™
Triggered where implementation conditions differ materially from those under which the tool was initially evaluated.
37. Risk Tool Assurance Architecture™
SAFECHAIN™ establishes:
Development → Validation → Implementation → Practitioner Use → Classification → Institutional Reliance → Outcome → Evaluation → Revalidation
38. Validation-Is-Not-Permanent Principle™
A tool should not be treated as permanently validated merely because evidence supported its adoption at an earlier point.
Operational environments change.
Patterns of abuse change.
Technology changes.
Professional practice changes.
Evidence should therefore be continually reviewed.
39. Revalidation Standard™
Material risk-assessment instruments should have defined revalidation or reassessment mechanisms considering:
operational performance;
new research;
changing abuse patterns;
demographic performance;
practitioner consistency;
outcome evidence;
implementation changes.
40. Evidence Base Currency Alert™
Triggered where institutional reliance continues despite the supporting evidence base becoming materially outdated or contested.
41. Risk Tool Governance Register™
Institutions should maintain a register identifying:
tool;
purpose;
owner;
validation evidence;
implementation date;
training;
limitations;
review cycle;
revalidation status;
known concerns.
42. Risk Assessment Assurance Register™
Record:
assessment tool;
assurance activity;
sample;
findings;
limitations;
corrective action;
reviewer;
date;
follow-up.
43. Professional Override Register™
Record material instances where professional judgement differs from structured classification.
Analysis should examine whether overrides reveal:
tool limitation;
training issue;
emerging risk factor;
inconsistent application.
44. False-Negative Review Standard™
Where serious harm occurs following a lower-risk classification, institutions should review whether:
relevant evidence was available;
the tool captured it;
professional judgement identified concern;
escalation was available;
implementation contributed.
45. False-Positive Review Standard™
Institutions should also examine materially disproportionate classifications to identify:
data quality issues;
bias;
inconsistent application;
threshold problems.
46. Risk Assessment Outcome Dashboard™
Potential indicators include:
risk classifications;
professional overrides;
repeat victimisation;
escalation following assessment;
reclassification;
serious outcomes following lower classifications;
practitioner variation;
tool completion quality;
reassessment frequency.
47. Officer Consistency Standard™
Institutions should assess whether materially similar circumstances produce broadly comparable assessments.
48. Inter-Assessor Variation Alert™
Triggered where different practitioners classify materially similar evidence very differently without defensible reasons.
49. Professional Calibration Test™
Present comparable scenarios to multiple practitioners and evaluate:
pattern recognition;
classification;
professional override;
safeguarding action.
50. Risk Assessment Stress Test™
Institutional risk tools should be tested using difficult cases involving:
coercive control without obvious physical violence;
technology-facilitated abuse;
financial abuse;
stalking;
separation;
dependency;
conflicting evidence;
minimisation;
repeat low-level incidents.
51. SAFECHAIN™ Coercive Control Stress Test™
Ask:
Can the risk system recognise serious danger where no single incident appears sufficient on its own?
52. Pattern Blindness Alert™
Triggered where structured assessment systematically under-recognises cumulative or patterned behaviour.
53. Digital Coercion Integration Standard™
Risk assessment should be capable of incorporating:
location tracking;
account monitoring;
device surveillance;
password control;
impersonation;
smart-home abuse;
digital financial control;
persistent digital contact.
54. Digital Risk Blindness Alert™
Triggered where technology-facilitated abuse is treated as peripheral rather than as part of the coercive-control pattern.
55. Cross-System Risk Continuity™
Where risk information transfers between organisations, the receiving body should understand:
classification;
evidence;
pattern;
limitations;
professional concern;
reassessment triggers.
56. Score-Only Transfer Alert™
Triggered where another organisation receives the risk category without sufficient supporting context.
57. Safeguarding Intervention Integrity Standard™
The institution should examine whether risk classification actually produces proportionate safeguarding action.
58. Classification-to-Intervention Test™
Ask:
What institutional action followed the classification, and was that action proportionate to the evidence?
59. Assessment-without-Action Alert™
Triggered where a risk assessment identifies serious concern but does not produce meaningful safeguarding intervention.
60. Risk Assessment Closure Standard™
Completion of assessment should never equal completion of safeguarding responsibility.
61. Assessment Closure Fallacy™
Triggered where the administrative completion of the tool becomes the endpoint rather than the beginning of risk management.
62. Risk Escalation Integrity Standard™
Professionals should be able to escalate where:
tool classification appears inconsistent;
professional concern remains high;
pattern evidence is significant;
risk is changing rapidly;
information is incomplete.
63. Risk Escalation Gate™
Verify:
✓ Tool output recorded
✓ Professional judgement recorded
✓ Pattern evidence considered
✓ Contradictory evidence considered
✓ Affected-person evidence considered
✓ Escalation threshold reviewed
✓ Decision owner identified
64. Tool Challenge Standard™
Institutions should maintain mechanisms for practitioners, supervisors, affected persons and assurance reviewers to question tool outputs or application.
65. Tool Immunity Alert™
Triggered where organisational culture treats questioning the assessment instrument as questioning institutional policy itself.
66. Risk Assessment Assurance Standard™
Independent assurance should assess:
evidence base;
operational implementation;
practitioner consistency;
limitations;
pattern recognition;
downstream reliance;
outcomes;
revalidation.
67. Self-Assured Risk Tool Alert™
Triggered where the organisation adopting and operating a tool is the only body assessing whether it remains effective.
68. SAFECHAIN™ Risk Tool Assurance Test™
An institution should demonstrate:
Why was this tool selected?
What evidence supported adoption?
What limitations were identified?
Has implementation been independently evaluated?
Are professional overrides permitted?
Is pattern recognition preserved?
Are affected-person accounts preserved contextually?
Are false-negative outcomes reviewed?
Is practitioner consistency tested?
Are downstream consequences monitored?
Is revalidation scheduled?
Can reliance be reduced if evidence changes?
69. Risk Tool Integrity Classification™
RTI1 — Strong Risk Tool Integrity
Robust evidence, implementation, professional judgement and continuing assurance.
RTI2 — Effective With Improvement
Generally reliable with identifiable development requirements.
RTI3 — Material Assurance Gap
Evidence or implementation limitations create material uncertainty.
RTI4 — Serious Risk Assessment Integrity Failure
Tool reliance materially weakens safeguarding or accountability.
RTI5 — Systemic Risk Assessment Governance Breakdown
Institution continues to rely upon a tool despite serious unresolved concerns about validity, implementation or safety.
70. Pattern Recognition Integrity Classification™
PRI1 — Strong Pattern Recognition
PRI2 — Effective With Improvement
PRI3 — Material Pattern Recognition Gap
PRI4 — Serious Pattern Blindness
PRI5 — Systemic Pattern Recognition Failure
71. Tool Reliance Classification™
TRL1 — Appropriate Decision Support
Tool informs but does not control professional judgement.
TRL2 — Elevated Reliance
Tool carries significant institutional weight.
TRL3 — Material Tool Dependency
Professional reasoning becomes substantially structured around output.
TRL4 — Excessive Tool Deference
Output routinely overrides wider contextual evidence.
TRL5 — Institutional Tool Substitution
Tool effectively replaces substantive professional risk assessment.
72. SAFECHAIN™ Risk Assessment Verification Gate™
Before reliance upon a risk classification, verify:
✓ Tool appropriately applied
✓ Evidence complete enough for purpose
✓ Professional judgement recorded
✓ Pattern considered
✓ History considered
✓ Digital risk considered where relevant
✓ Affected-person evidence considered
✓ Limitations acknowledged
✓ Escalation available
✓ Classification current
✓ Safeguarding response identified
73. Risk Assessment Integrity Closure Gate™
A risk-assessment governance concern should not close until:
implementation has been assessed;
identified limitations have owners;
affected decisions are reviewed where required;
corrective actions are implemented;
revalidation or retesting occurs where appropriate;
downstream reliance is addressed.
74. Institutional Risk Assessment Reality Test™
Ask:
If the structured form disappeared tomorrow, could the institution still explain the risk through evidence, professional reasoning, pattern recognition and contextual understanding?
If not, the institution may have become excessively dependent upon the instrument.
75. SAFECHAIN™ Research Proposition
The quality of institutional risk assessment cannot be measured solely by whether the correct tool was completed. Governance integrity requires examination of whether the institution recognised the complete pattern of risk, exercised professional judgement, preserved contextual evidence, acted proportionately and continued to test whether the assessment instrument itself remained fit for purpose.
76. Governance Implications for Coercive Control
For coercive control, the institutional challenge is particularly acute.
Coercive behaviour may be:
cumulative;
adaptive;
hidden;
digitally mediated;
financially embedded;
dependent upon relationship context;
intensified during separation.
Risk systems therefore require the ability to interpret relationships between facts, not merely record individual indicators.
77. SAFECHAIN™ Pattern-over-Form Principle™
Where the evidence demonstrates a coherent pattern of coercion, the pattern should not be displaced merely because individual components produce a lower structured risk score.
78. Integration with SAFECHAIN™ Governance Architecture
This research proposition integrates directly with:
SYSTEMS-001™ — Institutional Systems Architecture & Governance Framework™
Risk assessment must be understood as part of a broader decision architecture.
FLOW-001™ — Institutional Process Flow, Decision Pathway & Governance Handoff Framework™
Risk classifications travel through institutional pathways and influence downstream decisions.
INTERFACE-001™ — Cross-System Interface, Boundary & Institutional Coordination Framework™
Risk information must retain context when moving between police, safeguarding partners and other institutions.
DESIGN-001™ — Institutional Governance Design & Safeguard-by-Design Framework™
Risk-assessment governance should be designed before implementation.
SYSTEMCHECK-001™ — Institutional Systems Testing, Stress-Test & Failure Simulation Framework™
Assessment systems should be stress-tested before weaknesses affect real people.
Digital Safeguarding Maturity Model™
Technology-facilitated coercion should form part of risk-assessment maturity.
AIASSURANCE-001™
Material governance claims concerning effectiveness should be independently verified.
AIRELIANCE-001™
Institutions must understand downstream reliance upon risk classifications.
AIPRIORITY-001™
Risk classifications influence institutional prioritisation.
AIESCALATIONPATH-001™
Professional concern must remain capable of escalation beyond the tool output.
AIRECORD-001™
Assessment evidence, context and amendments must remain traceable.
AICAUSAL-001™
Where safeguarding failure occurs, the contribution of risk-assessment systems must be examined.
79. Governance Statement
A structured risk-assessment tool can support professional judgement, consistency and safeguarding decision-making. It should never become a substitute for understanding the person, the pattern or the context. The governance obligation is not satisfied because an approved form was completed. It is satisfied only where the institution can demonstrate that risk was recognised, contextualised, challenged where necessary, translated into proportionate safeguarding action and continually reassessed as evidence and circumstances changed.
80. Research Conclusion
The 2026 independent examination of DARA raises a broader question that extends well beyond one instrument or one police force.
The question is not whether structured risk assessment should exist.
It is whether institutions possess sufficient governance around the tools they increasingly depend upon.
The evidence base supporting DARA's national implementation has been described by Sebire and Bland as limited, and their study deliberately avoids claiming predictive validation. That does not establish that DARA is ineffective. It establishes a legitimate need for continuing evaluation of how such tools function beyond pilot conditions. (OUP Academic)
Contemporary research into policing risk instruments more broadly similarly emphasises the importance of validation, transparency and adaptive governance. (OUP Academic)
For SAFECHAIN™, the governance proposition is therefore clear:
Risk assessment must itself be governed as a high-impact institutional system.
And for coercive control:
The pattern must remain visible even when the individual data points appear ordinary.
81. Comprehensive Copyright & Intellectual Property Notice
© 2026 Samantha Avril-Andreassen. All Rights Reserved.
The Institutional Risk-Assessment Assurance Problem™ — When Domestic-Abuse Risk Assessment Becomes Institutional Authority: A SAFECHAIN™ Governance Analysis of DARA, Coercive Control, Professional Judgement and System Assurance is an original governance analysis, research architecture and systems-reform publication developed and authored by Samantha Avril-Andreassen, LLB (Hons), LLM, LPC, FRSA, Founder of SAFECHAIN™.
The publication forms part of the wider SAFECHAIN™ Governance Architecture™, SAFECHAIN™ Research Programme and SAFECHAIN™ institutional systems and safeguarding governance methodology.
The original expression, selection, arrangement, governance architecture, terminology, analytical methodologies, tests, principles, alerts, classifications, registers, assurance mechanisms and framework integrations contained within this publication constitute proprietary intellectual property.
This includes, where original to this SAFECHAIN™ publication, the Institutional Risk-Assessment Assurance Problem™, SAFECHAIN™ Risk Assessment Integrity Principle™, Tool–Judgement Distinction™, Tool Authority Inflation Alert™, Risk Grade Certainty Alert™, Coercive Control Pattern Integrity™, SAFECHAIN™ Coercive Control Pattern Test™, Incident Fragmentation Alert™, Pattern-to-Score Compression Risk™, Professional Pattern Recognition Standard™, Professional Override Integrity Standard™, Tool Deference Alert™, Institutional Reliance Chain™, Risk Classification Inheritance Alert™, Risk Assessment Freshness Standard™, Stale Risk Classification Alert™, Risk Evidence Traceability Test™, Score-without-Evidence Alert™, Affected-Person Evidence Dilution Alert™, Institutional Pattern Recognition Standard™, Historical Fragmentation Alert™, SAFECHAIN™ Tool Implementation Test™, Form Completion Fallacy™, Operational Reality Test™, Pilot-to-Practice Gap Alert™, Risk Tool Assurance Architecture™, Validation-Is-Not-Permanent Principle™, Risk Tool Governance Register™, Risk Assessment Assurance Register™, Professional Override Register™, Professional Calibration Test™, SAFECHAIN™ Coercive Control Stress Test™, Pattern Blindness Alert™, Digital Risk Blindness Alert™, Score-Only Transfer Alert™, Classification-to-Intervention Test™, Assessment-without-Action Alert™, Assessment Closure Fallacy™, Risk Escalation Gate™, Tool Immunity Alert™, SAFECHAIN™ Risk Tool Assurance Test™, RTI1™–RTI5™ Risk Tool Integrity Classification, PRI1™–PRI5™ Pattern Recognition Integrity Classification, TRL1™–TRL5™ Tool Reliance Classification, SAFECHAIN™ Risk Assessment Verification Gate™, Risk Assessment Integrity Closure Gate™, Institutional Risk Assessment Reality Test™ and SAFECHAIN™ Pattern-over-Form Principle™, together with the associated original governance analysis and framework materials.
No part of this publication may be reproduced, copied, republished, substantially adapted, translated, distributed, licensed, sublicensed, commercially exploited or incorporated into another proprietary governance framework, assessment methodology, assurance system, consultancy methodology, training programme, certification scheme, accreditation programme, artificial-intelligence system, software platform, analytics product or derivative commercial offering without prior written permission from the applicable rights holder, except to the extent otherwise permitted by applicable law.
Publication, citation, academic discussion or public accessibility of this work does not transfer ownership of the original SAFECHAIN™ architecture and does not grant any licence, assessment authority, certification right, accreditation right or authority to represent any implementation, assessment or classification as officially SAFECHAIN™ authorised.
References within this publication to DARA, DASH, the College of Policing, Oxford University Press, Policing: A Journal of Policy and Practice, academic research, policing practices, domestic abuse, coercive control, risk assessment, professional judgement, validation and related established concepts or third-party materials remain the property of their respective rights holders where applicable.
SAFECHAIN™ does not claim ownership over DARA, DASH, third-party research findings, established academic concepts or other externally developed risk-assessment methodologies.
The proprietary claim relates specifically to the original SAFECHAIN™ governance interpretation, terminology, arrangement, analytical architecture, tests, alerts, classifications, principles, assurance structures and systems-reform methodology developed within this publication.
Nothing within this publication constitutes legal advice, clinical advice, policing guidance or an independent validation or invalidation of DARA or any other specific risk-assessment instrument.
The publication does not conclude that DARA is ineffective, unsafe or incapable of identifying domestic-abuse risk. The cited research is exploratory and expressly does not establish predictive accuracy or causality. (OUP Academic)
SAFECHAIN™ uses that research as the evidential basis for a wider institutional-governance proposition concerning how high-impact risk-assessment instruments should be implemented, relied upon, challenged, evaluated and independently assured.
Author and Governance Framework Developer:
Samantha Avril-Andreassen, LLB (Hons), LLM, LPC, FRSA
Founder — SAFECHAIN™
Publication: The Institutional Risk-Assessment Assurance Problem™
Research Domain: Domestic Abuse · Coercive Control · Policing · Safeguarding · Institutional Risk Assessment · Governance Assurance
Year: 2026
© 2026 Samantha Avril-Andreassen. All Rights Reserved.