Completion of early response actions accounted for most elapsed time across the 7-1-7 pathway in Sierra Leone, 2023–2025

Abstract

Objectives

To assess the timeliness of detection, formal notification and completion of applicable early response actions for priority public health events in Sierra Leone, and to identify documented operational constraints in delayed events and enabling process features in timely events.

Study design

Retrospective observational event-level study.

Methods

We analysed all 33 verified priority public health events recorded in the national Situation Room register from 1 February 2023 to 31 January 2025. Using a prespecified protocol adapted from World Health Organization Early Action Review and 7-1-7 Alliance tools, two reviewers reconstructed four milestone dates from dated surveillance, response, laboratory and specimen-transport records. We summarised target attainment, interval distributions and the post-notification share of total elapsed time. Reviewers coded documented constraints in 22 delayed events and conducted a positive-case review of affirmative enabling features in 11 timely events.

Results

Detection, notification and early-response targets were met in 22/33 (66.7%), 25/33 (75.8%) and 11/33 (33.3%) events, respectively; 11/33 (33.3%) met all three targets. Median emergence-to-detection, detection-to-notification and notification-to-completion intervals were 5 days (interquartile range [IQR] 3–9; range 1–21), 1 day (IQR 0–1; range 0–4) and 17 days (IQR 7–35; range 3–74). The median event-level proportion of total elapsed time occurring after notification was 73.5%. Delayed events commonly documented financing, logistics, specimen, laboratory and deployment constraints. Timely events showed prompt escalation, district-led initiation, ready transport and diagnostic pathways, available initial supplies, early coordination and familiar procedures.

Conclusions

Only one-third of registered events met all three 7-1-7 targets. Most elapsed time occurred after formal notification, when recognised threats needed to be converted into completed early response actions. Improvement should remove mobilisation constraints while preserving and standardising the enabling practices observed in timely events.

Graphical abstract

The graphic summarises timeliness across detection, formal notification and completion of applicable early response actions in Sierra Leone, the median 17-day post-notification interval, the most common documented operational constraints, recurring enabling features in timely events and the proposed management actions.

What this study adds.

• •The study operationalises the third 7 as completion of the full set of applicable early-action milestones, rather than the first response activity, which prevents early initiation from being mistaken for completion of the early-response package.
• •It shows how World Health Organization Early Action Review and 7-1-7 Alliance tools can be adapted for an auditable national multi-event synthesis that examines both constraints in delayed events and affirmative enabling features in timely events.
• •Pairing delayed-event review with positive-case review shows what slowed completion and what supported timely action, providing a stronger bridge from retrospective measurement to prospective Early Action Reviews.

Implications for policy and practice.

• •At formal notification, release a pre-approved initial response-funding ceiling and complete financial reconciliation after mobilisation.
• •Preserve and standardise the enablers observed in timely events: prompt Situation Room escalation, district-led initiation, ready transport and specimen referral, responsive laboratory pathways, minimum response stocks and early coordination.
• •Open an event-specific action register at notification, assign an owner and deadline to every applicable early action, initiate an Early Action Review when any requirement remains incomplete on day 7, and track remedial actions to closure in the national 7-1-7 database.

1. Introduction

Timely outbreak control requires rapid progression from threat emergence to detection, notification and early action. The 7-1-7 target sets limits of 7 days from emergence to detection, 1 day from detection to notification of a public health authority responsible for action, and 7 days from notification until all applicable early-response milestones occur. [[1], [2], [3], [4]] The third interval ends on the latest qualifying date across applicable domains. Several domains require initiation, not completion of the entire response; the milestone therefore does not indicate outbreak closure.

We selected 7-1-7 because it provides a common stage-specific benchmark across heterogeneous infectious threats and links measurement to performance improvement. Guidance combines milestone narratives and timeliness metrics with identification of bottlenecks, enablers and corrective actions. [2,5] The target draws on prior timeliness studies, Integrated Disease Surveillance and Response standards and multi-country implementation. [[6], [7], [8]] The World Health Organization includes 7-1-7 in its Fourteenth General Programme of Work and Early Action Reviews. [5,9] It complements the International Health Regulations, Joint External Evaluation and disease-specific protocols, which assess capacities, obligations and technical requirements, but does not measure response quality or outcomes. [[10], [11], [12]]

Published applications report variation in target attainment across stages and settings. In five countries, 54% of events met detection, 71% notification, 49% early response and 27% all three targets. [13] All-target attainment was 24% in Uganda, 43% in Liberia, 48% in Cambodia and 44% in South Sudan. [[14], [15], [16], [17]] Recurrent constraints involved financing, transport, laboratories, supplies, workforce and coordination. [[13], [14], [15], [16], [17]] A Sierra Leone measles study reported timely case detection and notification but identified gaps in supplies and referral pathways. [18]

Sierra Leone is an important setting for event-level performance analysis. The 2014–2016 Ebola epidemic exposed fragmentation in surveillance, laboratory services and emergency coordination. [19,20] Subsequent reforms expanded electronic surveillance. [21] COVID-19 tested these reforms; Sierra Leonean studies described strong use of Ebola-era systems and field epidemiology capacity alongside resource, communication, service-continuity and workforce pressures. [[22], [23], [24]] The Public Health Act established the National Public Health Agency, consolidating core public health emergency functions. [25,26]

Accordingly, our primary objective was to assess attainment of each 7-1-7 target and the complete 7-1-7 target among all eligible registered events during a fixed 24-month period. Secondary objectives were to describe interval distributions and the proportion of total time occurring after notification, examine event-type patterns and the robustness of the principal result, characterise operational constraints among events missing the early-response target, and identify affirmative enabling process features among events meeting it. Sierra Leone's 2025 mpox epidemic, most of which occurred after the data freeze, underscores the importance of this stage-specific approach but is not evaluated as an outcome of this study. [27,28]

2. Methods

2.1. Study design, setting, period and reporting

We conducted a retrospective observational event-level study using routine records from Sierra Leone's national public health emergency management system. The registered event was the unit of analysis. Reporting followed the Strengthening the Reporting of Observational Studies in Epidemiology statement. [29] The prespecified period, 1 February 2023 to 31 January 2025, covered two annual cycles under the same Situation Room register and ended at the parent assessment data freeze. This was a national retrospective multi-event synthesis, not a real-time Early Action Review or introductory small-event workshop. The World Health Organization positions Early Action Reviews within ongoing response coordination, while retrospective review guidance recommends a few recent, varied events for sensitisation and corrective-action consensus. [5,30] We included every eligible registered event to estimate performance across the fixed period.

2.2. 7-1-7 framework, tool adaptation and study outcomes

The protocol adapted the World Health Organization Early Action Review assessment approach and the 7-1-7 Alliance Digital Toolkit, Assessment Tool and Milestone Dates Reference Guide to Sierra Leone's records and disease-specific protocols. [[2], [3], [4], [5]] We retained the four milestones, seven early-response domains, date narratives and sources, distinction between not applicable and missing, and the latest-applicable-milestone rule for early-response completion.

The event abstraction form captured event type, candidate and assigned milestone dates, date narratives and sources, applicability and dates for each early-response domain, documented operational constraints and affirmative enabling features. We adapted the 7-1-7 Data Consolidation Spreadsheet and Synthesis Report template to calculate intervals and target indicators and synthesise events. [31,32] The participatory workshop, stakeholder voting and prospective action tracking were not data-collection procedures. Instead, the retrospective protocol included a structured positive-case review of timely events using affirmative evidence in milestone narratives and response records.

Primary outcomes were attainment of each target and all three targets. Secondary outcomes were interval distributions, total emergence-to-completion time, successive target attainment and the event-level post-notification share. Documentary outcomes were category frequency, burden and co-occurrence of constraints among delayed events, and recurring affirmative enabling features and lessons among timely events. The study assessed timeliness, not response quality or population impact.

2.3. Event population and eligibility

The source population comprised every event entered in the national Public Health Emergency Operations Centre Situation Room register during the study period. A priority event was a discrete occurrence linked by time, place and epidemiological evidence that involved an Integrated Disease Surveillance and Response-listed condition, unusual cluster or other reportable threat; passed formal verification; entered the register; and triggered district or national response. [33] A high-consequence case could constitute one event; epidemiologically linked cases shared one identifier.

The register contained 33 events; all met eligibility criteria and contained evidence for the four milestone dates. Signals closed before registration were outside the source population. Laboratory confirmation and declaration of a national emergency were not required because 7-1-7 starts with a verified suspected threat and early action may precede confirmation. [[1], [2], [3], [4]] The 11 unconfirmed events were registered and response-triggering, not signals awaiting verification. Event-level confirmation required at least one linked case or specimen meeting the disease-specific criterion and was used in sensitivity analysis.

2.4. Data sources and milestone reconstruction

Sources included electronic Integrated Disease Surveillance and Response alert and verification records, situation reports, Situation Room and response-action records, laboratory records and specimen-transport records. Event-period records supported milestones; intra-action and after-action reviews informed only constraint coding. The abstraction form required a narrative, source and adjudication note for each milestone. [[2], [3], [4], [5]] Two reviewers independently examined each event file and reconciled dates by consensus using Supplementary File 3.

Each event contributed one emergence, detection, notification and early-response completion date; linked-case dates were not averaged. Emergence was the earliest linked-case onset for a non-endemic disease, the date an alert threshold was first met for an endemic condition, or the date another threat met national reporting criteria. [[2], [3], [4]] Detection was the first clinical or public health record. Notification was the first report to an authority responsible for action, not notification to the World Health Organization. [3,4] Reviewers selected the earliest supported emergence and detection dates and the first documented notification date. Same-day milestones contributed zero days and thresholds were inclusive.

2.5. Applicable early response actions

Reviewers determined applicability from the risk assessment and disease-specific protocol. The seven domains were investigation or deployment; epidemiological analysis and risk assessment; laboratory confirmation or diagnostic resolution; case management and infection prevention and control; countermeasures or public health and social measures; risk communication and community engagement; and coordination (Supplementary Table S4; Supplementary File 3). [[2], [3], [4]] Reviewers recorded the earliest qualifying milestone in each domain. Most counted when action began; epidemiological analysis required a completed assessment, laboratory action a communicated result or diagnostic resolution, and coordination an established mechanism or incident action process. A plan without implementation did not count.

Early-response completion was the latest milestone across applicable domains. Non-applicable actions were distinguished from missing information. Verification did not count as early action. Because selected threats used diagnostic resolution rather than positive confirmation, we repeated the analysis among laboratory-confirmed events.

2.6. Documentary review of operational constraints and enabling features

We reviewed 22 delayed and 11 timely events. In delayed events, a constraint required explicit documentation linking an operational problem to late completion. In timely events, an enabler required affirmative evidence linking a process or resource to prompt action. This follows the 7-1-7 distinction between bottlenecks and enablers. [2,5,34]

We cross-walked constraints to the 7-1-7 Alliance taxonomy and retained seven local categories: fund release, district logistics, specimen transport, laboratory processes, field deployment, supply availability and coordination or decision clearance (Supplementary Table S10). [34] Enablers were grouped as prompt escalation, district-led initiation, ready transport and referral, responsive diagnostic pathways, available initial supplies, early coordination and familiar procedures (Supplementary Table S11). Two reviewers applied written rules and resolved differences by consensus. Enablers were not inferred from absent constraints and are reported qualitatively because documentation was not standardised.

2.7. Analysis

We reported counts and percentages meeting each target and all three targets as the primary multi-event summary, consistent with 7-1-7 synthesis guidance. [2,13,32] Medians, interquartile ranges (IQRs) and ranges described right-skewed intervals. For each event, notification-to-completion time was divided by total emergence-to-completion time; we report the median proportion. Among delayed events, we counted constraint frequencies, categories per event and pairwise co-occurrence. Among timely events, we synthesised recurring affirmative enabling features and operational lessons without estimating prevalence. Event-type results were descriptive because groups contained 2–9 events. Analyses were descriptive and included all eligible registered events. No significance tests were performed. Sensitivity analyses excluded the longest-delay event, included only laboratory-confirmed events and excluded foodborne events. Supplementary File 2 contains the de-identified data and live formulas; Supplementary Table S5 presents all event-level intervals.

2.8. Ethics

The Sierra Leone Ethics and Scientific Review Committee approved the parent capacity assessment (reference SLESRC/0017/2025/02). This secondary analysis used de-identified routine event records; no individual was identifiable.

3. Results

3.1. Event population

The 33 events occurred in 14 of 16 districts. Twenty-two had laboratory confirmation; 11 passed verification, entered the register and triggered response without event-level confirmation. Event types were mpox-related events (n = 9), measles (n = 7), acute watery diarrhoea or cholera (n = 5), suspected viral haemorrhagic fever (n = 4), foodborne illness (n = 3), meningitis or severe febrile illness (n = 3), and anthrax or other zoonotic events (n = 2). Supplementary Tables S1 and S2 describe record completeness and emergence-date sources.

3.2. Overall 7-1-7 performance

The detection, notification and early-response targets were met in 22/33 (66.7%), 25/33 (75.8%) and 11/33 (33.3%) events, respectively (Table 1; Fig. 1, Fig. 3; Supplementary Table S5). All 11 events meeting the early-response target also met detection and notification, so 11/33 (33.3%) met the complete 7-1-7 target. All 22 events meeting detection also met notification; three events that missed detection were still notified within 1 day (Fig. 4A).

Table 1.

Timeliness and target attainment across the 7-1-7 pathway for 33 registered priority public health events.

Interval	Median (days)	IQR (days)	Range (days)	Target met, n/N (%)
Emergence to detection (target: ≤7 days)	5	3–9	1–21	22/33 (66.7)
Detection to formal notification (target: ≤1 day)	1	0–1	0–4	25/33 (75.8)
Formal notification to completion of applicable early response actions (target: ≤7 days)	17	7–35	3–74	11/33 (33.3)
Total time from emergence to completion	23	11–45	4–99	Not applicable
All three targets met	Not applicable	Not applicable	Not applicable	11/33 (33.3)

IQR, interquartile range. Early-response completion was the latest date on which every applicable action met its defined milestone; it did not indicate event closure. All 11 events meeting the early-response target also met the detection and notification targets.

Fig. 1.

The 7-1-7 pathway from event emergence to detection, formal notification and completion of applicable early response actions. For each interval, the figure shows the median, interquartile range, target and the number and percentage of events meeting the target. The boxed summary shows the median proportion of each event's total elapsed time that occurred after notification.

Fig. 3.

Time intervals for 33 registered priority public health events. Each row represents one event, which could include one high-consequence case or several epidemiologically linked cases; dates from linked cases were not averaged. Markers show emergence (circle, day 0), detection (square), formal notification (triangle) and completion of applicable early response actions (diamond). Dashed vertical lines at days 7, 8 and 15 show the ideal cumulative pathway when each preceding interval reaches its maximum permitted duration. They are reference points, not event-specific deadlines. Events are ordered by total time from emergence to completion.

Fig. 4.

Successive 7-1-7 target attainment and operational constraints documented in the same events. (A) Of 33 events, 22 met the detection target. All 22 also met the notification target and 11 subsequently met the early-response target. Overall notification attainment was 25/33 because three events that missed detection were notified within 1 day. (B) Constraint categories among the 22 events that missed early response. Diagonal cells show category frequencies and off-diagonal cells show how often two categories were documented in the same event. Categories documented together do not establish interaction or causation.

Intervals were right-skewed. Median emergence-to-detection time was 5 days (IQR 3–9; range 1–21), detection-to-notification time was 1 day (IQR 0–1; range 0–4), and notification-to-completion time was 17 days (IQR 7–35; range 3–74). Median total time was 23 days (IQR 11–45; range 4–99). The median proportion of each event’s total time occurring after notification was 73.5% (IQR 62.5–76.3; range 33.3–89.7).

3.3. Event-type and sensitivity analyses

Suspected viral haemorrhagic fever events had the longest median notification-to-completion interval at 42.5 days, and none met the target. Medians ranged from 7 days for foodborne illness to 22 days for meningitis or severe febrile illness across the other event groups (Supplementary Table S8). The groups were too small for comparative inference.

Low early-response target attainment persisted in the sensitivity analyses. After exclusion of the longest-delay event, median notification-to-completion time was 15.5 days and 11/32 (34.4%) met the target. Among 22 laboratory-confirmed events, the median was 11 days and 8/22 (36.4%) met the early-response and complete 7-1-7 targets. After exclusion of foodborne events, the median was 17.5 days and 9/30 (30.0%) met the target (Supplementary Table S3).

3.4. Documented operational constraints

Among the 22 events that missed the early-response target, fund-release delays were documented in 20 (90.9%), district logistics in 18 (81.8%), specimen transport in 15 (68.2%), laboratory processes in 14 (63.6%), field deployment in 14 (63.6%), supply availability in 11 (50.0%) and coordination or decision clearance in 10 (45.5%) (Fig. 2; Supplementary Tables S6 and S7). Delayed events contained a median of 4.5 documented categories (IQR 4–5; range 3–7); 18/22 contained at least four and 11/22 at least five categories.

Fig. 2.

Documented operational constraints among the 22 events that missed the early-response target and the number of constraints recorded per event. More than one constraint could be recorded for an event. Eighteen events (81.8%) had at least four categories and 11 (50.0%) had at least five.

Records described recurring sequences. Funding required several approval, release or transfer steps before mobilisation. Districts sometimes lacked vehicles, fuel or drivers; specimens waited for scheduled referral; and laboratory records cited queues, batch testing, reagent shortages, repeat testing, validation or delayed result release. Deployment and supply records described waits for staff, activation, transport, protective equipment, medicines or investigation materials. Coordination records documented pending incident-management or interagency decisions. The most common pairs were fund release with district logistics (16 events), specimen transport (14), laboratory processes (14) and field deployment (12), and specimen transport with laboratory processes (13) (Fig. 4B; Supplementary Table S9). Co-occurrence did not establish interaction or causation.

3.5. Documented enabling features in timely events

Across the 11 timely events, records described prompt escalation, district-led initiation, ready transport and specimen referral, responsive diagnostic pathways, available initial supplies, early coordination with assigned responsibilities and familiar procedures. These features often occurred together and are presented as qualitative lessons rather than prevalence estimates (Supplementary Table S11).

4. Discussion

This national event series showed a marked imbalance across the 7-1-7 pathway. Detection and notification met their targets in 66.7% and 75.8% of events, but only 33.3% completed all applicable early response milestones within 7 days. The median event-level proportion of total elapsed time occurring after notification was 73.5%. Low early-response target attainment persisted in the sensitivity analyses. Delayed events usually contained several documented constraints, while timely events showed combinations of prompt escalation, local initiation, operational readiness and early coordination.

Detection performance was stronger than early response, but 11 events took more than 7 days to detect. Attainment exceeded the five-country and Uganda estimates but remained below Liberia and some recent applications. [[13], [14], [15], [16], [17]] Event mix and definitions limit comparison. Community surveillance, clinical suspicion, threshold recognition and verification still require attention. In Uganda, longer detection intervals were associated with larger, longer outbreaks, but comparable outcome data were unavailable here. [35]

Notification was the strongest stage. Three late-detected events were still notified within 1 day, indicating prompt reporting after recognition. Attainment was close to five-country and Uganda estimates but below Liberia, Cambodia and South Sudan. [[13], [14], [15], [16], [17]] Same-day channels and prompt Situation Room registration remain important, while date-only records may conceal sub-day delays.

The principal deficit occurred after notification. The 17-day median more than doubled the target, and nearly two-thirds of laboratory-confirmed events missed the third 7. Restriction to confirmed events produced a similar result, supporting action at verified suspicion and recognising diagnostic resolution as an early-response requirement. [[1], [2], [3], [4]] Responses requiring several linked functions, such as suspected viral haemorrhagic fever, may be especially vulnerable to one late domain, although groups were small.

Most of Sierra Leone's 2025 mpox epidemic occurred after the data freeze and is not an outcome of this study. [27,28] It illustrates the stakes: nationwide response required laboratory confirmation, investigation, case management, vaccination, risk communication, logistics and coordination. [28] Prospective milestone narratives and action dates could distinguish detection delay from mobilisation delay while response remains active.

The documentary analysis describes post-notification delay without claiming causation. Funding appeared most often and commonly accompanied logistics, specimen transport, laboratory processes or deployment. Following establishment of a revolving outbreak investigation fund in Nigeria, median response time decreased from 6 to 2 days. [36] Similar constraints recur in other 7-1-7 studies. [[13], [14], [15], [16], [17], [18]] Multiple categories per event argue against single-bottleneck solutions.

Timely events provide complementary lessons. Records from timely events described prompt escalation, district-led initiation, operational readiness, responsive diagnostic pathways, available initial commodities, early coordination and familiar procedures. These features should be standardised through job aids, readiness checks and Early Action Reviews. [2,5]

Three linked improvements follow. Formal notification should trigger a pre-approved initial funding ceiling, with reconciliation after mobilisation. [37] Districts need standing transport and specimen-referral arrangements, laboratory escalation procedures and minimum response stocks. Timely-event enablers should be formalised through standing authority for district initiation, clear escalation routes, event-specific job aids and early coordination. The Situation Room should assign an owner and deadline to each applicable domain, initiate an Early Action Review when any remains incomplete after 7 days and track corrective and sustaining actions in the national database. [2,5,31,32,37]

The transferable contribution is an adapted toolchain: event-level milestone reconstruction, disease-specific applicability rules, a consolidation workbook, a recognised constraint taxonomy, a positive-case enabler review and synthesis for planning. [[2], [3], [4], [5],31,32,34] Retrospective research cannot reproduce a real-time participatory Early Action Review, and capacity, quality and outcomes still require other frameworks. [[10], [11], [12]] The estimates apply to Sierra Leone's registered events during this period.

4.1. Strengths and limitations

Strengths include complete enumeration, prespecified rules, source triangulation, two-reviewer reconciliation, explicit completion criteria, sensitivity analyses, review of delayed and timely events and reproducible data. Limitations include exclusion of undetected threats and unregistered signals, uncertainty from date-only and delayed records, variation in applicable actions and use of diagnostic resolution for selected verified threats. Constraint frequencies apply only to delayed events. Enabler documentation was non-standardised, so the qualitative lessons may be incomplete. The study did not include real-time stakeholder validation, participatory root-cause analysis or prospective action tracking, and lacked case burden, severity, deaths, spread and response-quality data. The period ended during the opening phase of the 2025 mpox epidemic. Target awareness, small event groups and one country also limit interpretation.

4.2. Conclusions

Only one-third of registered events met all three 7-1-7 targets, and most elapsed time accumulated after notification. The actionable challenge was delayed completion of applicable early actions amid financing, logistics, specimen-transport, laboratory, deployment, supply and coordination constraints. Timely events also showed what should be preserved: prompt escalation, district-led initiation, operational readiness, responsive diagnostic pathways and early coordination. Sierra Leone should remove recurrent constraints, standardise these enabling practices and use prospective 7-1-7 monitoring and Early Action Reviews to test improvement.

Author contributions

Eric Nzirakaindi Ikoona: Conceptualisation, methodology, investigation, data curation, formal analysis, visualisation, writing: original draft and project administration. Lucy Namulemo: Investigation, data curation, interpretation and writing: review and editing. Mary Magdalene Sinnah: Validation, investigation, documentation review and writing: review and editing. Mohamed Alex Vandi: Validation, surveillance-system interpretation and writing: review and editing. Foday Sahr: Supervision, institutional oversight, interpretation and writing: review and editing. All authors approved the revised version and agree with its resubmission.

Ethical statement

Ethical approval was obtained from the Sierra Leone Ethics and Scientific Review Committee under the parent capacity assessment (reference SLESRC/0017/2025/02). The focused secondary analysis used de-identified routine event records.

Data availability

Supplementary File 2 contains the de-identified event-level dataset and formulas used to reproduce target attainment, interval distributions, the proportion of total elapsed time occurring after notification and sensitivity analyses. It also contains de-identified summaries of documented constraint categories and a qualitative enabler synthesis. Supplementary File 1 contains Table S1–S11, operational definitions and the completed STROBE checklist. Supplementary Table S11 presents recurring enabling features and operational lessons from timely events. Supplementary File 3 documents the adapted event-abstraction and adjudication rules, including event boundaries, milestone dates in events with several linked cases, date assignment, case classification, event-level laboratory confirmation, early response actions, applicability, source reconciliation and the complementary reviews of constraints and enablers. The underlying administrative, clinical, intra-action-review and after-action-review records are not public because they contain sensitive operational or potentially identifiable information. Access may be considered by the corresponding author, subject to institutional approval.

Declaration of writing assistance

During manuscript preparation, the authors used Grammarly to support grammar editing and clarity. The authors reviewed and revised all text, verified the evidence and references and take full responsibility for the content of the article.

Funding

This research received no specific grant from funding agencies in the public, commercial or not-for-profit sectors.

Declaration of competing interest

We, the undersigned authors of the above-titled manuscript, declare that we have no known competing financial interests, personal relationships, or other affiliations that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Supplementary File 1

Table S1–S11, including record completeness, emergence-date sources, sensitivity analyses, event-type response packages, full event-level intervals, compound-delay results, operational-constraint definitions and explanations, event-type summaries, the taxonomy cross-walk, recurring enabling features and the completed STROBE checklist.

Supplementary File 2

De-identified event-level 7-1-7 dataset with live formulas, aggregate performance summaries, sensitivity analyses, documented constraint summaries and a qualitative enabler synthesis.

Supplementary File 3

Adapted event-abstraction and adjudication rules for event boundaries, linked-case milestone dates, date assignment, case classification, event-level laboratory confirmation, early response actions, applicability, source reconciliation, complementary constraint and positive-case enabler reviews, and the audit trail.