Overview#
Most serious officer misconduct incidents are preceded by warning signs that were visible in the data but not acted upon. A rising complaint rate, a shift in use of force frequency, or an unusual pattern of vehicle accidents are the kinds of signals that a supervisor handling a full caseload of operational responsibilities can easily miss in manual reviews. An Early Intervention System does not replace supervisory judgment. It ensures supervisors have the information to exercise that judgment before a situation escalates.
Argus Early Intervention System (EIS) uses configurable risk factors and peer comparison algorithms to identify officers who may benefit from proactive support, training, or intervention. It is designed to support officer wellness and professional development, not as a punitive mechanism. Data protections and due process controls are built into the architecture.
Open Standards#
- GraphQL (June 2018 Specification): All EIS queries, mutations, and type definitions are exposed through a Strawberry GraphQL schema, enabling strongly-typed, introspectable access to risk scores, alerts, and intervention records.
- OAuth 2.0 (RFC 6749) and JSON Web Tokens (RFC 7519): Every EIS endpoint enforces authentication via RS256-signed JWTs validated against a JWKS endpoint, with access decisions following the OAuth 2.0 bearer-token model.
- OASIS XACML 3.0: Attribute-based access control decisions on EIS officer data are evaluated against OASIS eXtensible Access Control Markup Language 3.0 policies, limiting visibility to authorised roles and organisational scopes.
- FBI CJIS Security Policy: Officer performance and misconduct-indicator data handled by the EIS is classified as CJIS-sensitive; the platform enforces CJIS audit, encryption-in-transit, and encryption-at-rest controls (sections 5.4, 5.10, 5.11) on all such records.
- HIPAA (45 CFR Parts 160 and 164): Wellness referral and health-indicator data surfaced alongside EIS assessments is treated as protected health information; access is opt-in, individually consented, and subject to HIPAA privacy and security rule requirements.
- JSON (RFC 8259): Risk factor scores, contributing-factor payloads, peer-comparison results, and intervention recommendations are serialised and stored as RFC 8259-compliant JSON documents within the PostgreSQL data store.
- RFC 3161 (Internet X.509 PKI Time-Stamp Protocol): The platform's audit-chain infrastructure anchors EIS audit-log entries with RFC 3161 timestamp tokens, providing cryptographically verifiable proof of when each access or action was recorded.
Last Reviewed: 2026-02-05 Last Updated: 2026-04-14
Key Features#
Risk Assessment and Scoring#
- Multi-factor risk scoring with configurable risk factor weights
- Threshold-based alert triggers customisable by agency policy and collective bargaining requirements
- Trend analysis tracking risk scores over time to identify escalating patterns
- Peer percentile comparison placing individual metrics in context of officers with similar assignments, tenure, and workload
- Automatic recalculation as new data enters the system
- Weighted scoring models that account for assignment risk levels and call volume differences
Risk Factors Tracked#
- Use of force incidents and force severity trends
- Citizen complaints by type, frequency, and disposition
- Pursuit incidents and outcomes
- Sick leave patterns and usage anomalies
- Disciplinary actions and policy violations
- Training deficiencies and overdue certifications
- Vehicle accidents and equipment damage
- Officer-involved shooting incidents and post-incident indicators
- Additional factors configurable by agency policy and labour agreements
Alert Management#
- Severity-tiered alerts across Low, Medium, High, and Critical levels
- Full status workflow from new through acknowledged, in progress, and closed
- Supervisor assignment with due date tracking and escalation notifications when alerts remain unaddressed
- Resolution documentation with structured outcome recording
- Alert history maintained for longitudinal analysis and pattern identification
Intervention Tracking#
- Intervention type categorisation covering counselling, training, mentoring, schedule adjustment, administrative review, and wellness referral
- Outcome measurement for each intervention with success indicators
- Follow-up scheduling with automated reminders for supervisors and programme coordinators
- Progress documentation tracking officer development over time
- Confidential notes and observations restricted to authorised personnel
- Multi-intervention coordination when officers are enrolled in concurrent support programmes
Privacy and Compliance#
- Role-based access control limiting data visibility to authorised supervisors and administrators
- Comprehensive audit logging of every access, view, and action taken within the system
- Configurable data retention policies meeting legal and labour agreement requirements
- Officer notification rights ensuring transparency about EIS assessments
- Appeal process support with documentation and review workflows
- Data anonymisation capabilities for aggregate reporting and programme evaluation
Use Cases#
Proactive Wellness Support. Identify officers showing stress indicators by monitoring sick leave patterns, tracking complaint trends, comparing to peer norms, and recommending wellness resources. Document all support provided and track engagement with recommended resources to measure whether interventions are effective.
Training Needs Assessment. Analyse incident patterns across use of force, complaints, and pursuit involvement to identify skill gaps. Recommend targeted training programmes, track completion, and measure post-training improvement in relevant metrics.
Supervisory Review. Provide supervisors with a dashboard of assigned officers showing risk trend visualisation, intervention history, outcome tracking, and performance documentation. Enable proactive management and early engagement with officers who may benefit from support.
Algorithm Fairness Review. Support administrators in regularly reviewing risk scoring algorithms for fairness, ensuring thresholds are appropriate, bias is minimised, and the system operates consistently with agency values and legal requirements.
Integration#
- Connects with officer wellness resources and employee assistance programmes for referral coordination
- Integrates with use of force analysis and incident tracking systems for automated data collection
- Links to training management and certification records for deficiency identification
- Works alongside internal affairs and complaint handling workflows for comprehensive oversight
- Feeds into supervisory dashboards and command staff reporting for organisational awareness
- Complete audit trail of all access and actions maintained for compliance and legal discovery
- Compatible with personnel records systems for officer profile context and assignment history
- Supports export of anonymised aggregate data for programme evaluation and accreditation reporting