U.S. National DNA Index System (NDIS)

• NDIS is a national DNA repository that aggregates profiles from convicted offenders, arrestees, and forensic samples to enable cross-jurisdiction investigations.
• It operates as the apex of the CODIS hierarchy by integrating data from local, state, and federal labs, thereby generating actionable investigative leads.
• Advanced methods in data acquisition, cleaning, and harmonization provide a reliable foundation for longitudinal analysis and informed forensic policy debates.

The National DNA Index System (NDIS) is the FBI-administered, top-level repository of DNA profiles in the United States, constituting the national component of the Combined DNA Index System (CODIS). NDIS centralizes DNA profile data collected from convicted offenders, arrestees (where permitted by state law), and forensic (crime-scene) samples. It is designed to support interstate and federal DNA matching, thus enabling the identification of serial crimes, aiding investigations, and informing forensic policy analysis. NDIS aggregates records from all state and local CODIS-participating laboratories and is the locus at which profile comparisons generate investigative leads, with regular reporting of population statistics and system activity (Pryor et al., 15 Nov 2025).

1. System Architecture and Role

NDIS is structurally embedded as the apex tier of the CODIS hierarchy, with the two lower tiers being the State DNA Index Systems (SDIS) and Local DNA Index Systems (LDIS). Profiles originate at the LDIS or SDIS levels and are eligible to flow upward into NDIS if they meet specified statutory and technical criteria. NDIS’s principal operations include:

• Aggregation of profiles from state/local CODIS labs.
• Cross-jurisdictional searches to link cases and persons of interest at a national scale.
• Publication of monthly statistics on profile counts (by type), the number of participating laboratories, and “investigations aided”—an enumeration of forensic cases in which database matches generated actionable leads.

NDIS participation encompasses all 50 states and several federal or territorial reporting entities, providing a comprehensive platform for U.S. forensic DNA investigations (Pryor et al., 15 Nov 2025).

2. DNA Profile Indices

NDIS data are organized into three primary categories:

• Offender Profiles: Cumulative records of individuals convicted of qualifying offenses. Between 2001 and 2012, only offender profiles were tracked; after January 1, 2012, tracking continued but arrestee profiles were separated due to statutory changes in some states.
• Arrestee Profiles: Collected from individuals arrested (but not yet convicted) for qualifying crimes, in jurisdictions where state law permits such collection. The arrestee index entered published NDIS statistics as states began to adopt arrestee-collection statutes starting January 2012.
• Forensic Profiles: DNA records derived from crime scene or evidentiary samples associated with unknown perpetrators. Each sample is represented once, regardless of future investigative outcomes or partial matches.

This systematization enables longitudinal and cross-sectional analyses of national DNA database growth and function (Pryor et al., 15 Nov 2025).

3. Data Acquisition, Processing, and Harmonization

Reconstructing and harmonizing the NDIS time series required an extensive digital archaeology approach:

• Snapshot Acquisition: 11,359 distinct captures from the Wayback Machine, spanning 255 URL patterns, were queried to obtain historical statistics.
• Parsing and Extraction: Four specialized HTML parsers were deployed, each tailored to different web presentation “eras,” to extract jurisdiction, date, and relevant metric counts.
• Standardization: 169 jurisdiction name variants were collapsed to 54 canonical categories, including all states and several non-state entities (e.g., District of Columbia, U.S. Army).
• Anomaly Detection and Cleaning: A composite framework flagged and removed reporting artifacts through sequential steps—
  • Spike–Dip Detection: $N^{(x)}_{j,t} > 2N^{(x)}_{j,t-1}$ or $N^{(x)}_{j,t} < 0.5N^{(x)}_{j,t-1}$;
  • Zero-Error Identification: zero value after a positive cumulative count;
  • Update-Lag and Repetition Filtering: removal of oscillatory or repeated anomalies;
  • Metric-Specific Thresholds: tailored rules for each metric (e.g., a 2.5× spike for forensic, 10× for investigations aided, empirical rules for laboratory counts).
• External Calibration: National aggregates were cross-validated with FBI CODIS brochures (2000–2015) and other publications (Ge et al. 2012, Wickenheiser 2022, Link et al. 2023, Greenwald & Phiri 2024).

Growth rates followed the standard discrete-time formula:

$r_t = \frac{N_t - N_{t-1}}{N_{t-1}} \times 100\%,$

and average annual growth over $T$ periods as

$\bar r = \frac{1}{T}\sum_{t=1}^{T} r_t.$

Only anomalies violating cumulative monotonicity were excluded to ensure a conservative, reliable time series (Pryor et al., 15 Nov 2025).

4. Quantitative Evolution: 2001–2025

National aggregates extracted at quinquennial intervals demonstrate the scale and trajectory of NDIS expansion:

Year	Offender (M)	Arrestee (M)	Forensic (k)	Investigations Aided (k)
2001	0.44	NA	21.6	1.6
2006	3.98	0.054	160.6	45.4
2011	10.1	—	351.9	141.3
2016	11.8	2.03	638.2	274.6
2021	14.8	4.51	1,144.3	587.8
2025 (Aug)	17.5†	6.1†	1,375.0†	710.0†

† Projected August 2025 from the harmonized series.

• Offender profiles grew by approximately 17.1 million from 2001 to 2025, corresponding to an average annual increase of $\bar r_{\rm off}\approx 0.70$ million profiles.
• Forensic profiles averaged ≈57,000 new entries per year.
• Investigations aided rose at ≈29,000 per year.

Notable step changes in growth coincided with policy events: the introduction of arrestee uploads in early 2012 and the expansion of CODIS core loci to 20 STRs in July 2017. Monthly uploads for offenders at times surpassed 200,000 profiles, particularly in late 2021 (Pryor et al., 15 Nov 2025).

5. Laboratory Participation and Investigative Outcomes

The number of state, local, and federal laboratories submitting data to NDIS increased from fewer than 50 in 2001 to over 200 by 2025, indicating both the proliferation of CODIS-accredited labs and consolidation or expansion of existing facilities.

The “investigations aided” metric aggregates all instances where a database match provided an investigative lead. This cumulative count reached approximately 710,000 as of August 2025, typically exceeding the number of forensic profiles because a single match can contribute to multiple investigations (Pryor et al., 15 Nov 2025).

6. Policy, Practice, and Analytical Implications

Broadening statutory collection criteria (notably, inclusion of arrestees) and technical advances (e.g., expansion to 20 core CODIS loci) demonstrably increased both the scale and investigative impact of NDIS. Crossing thresholds such as 10 million offender profiles (2012) and 1 million forensic profiles (2021) correlates with increased match rates and aided investigations.

Despite technological and operational gains, reporting gaps persist: as of 2025, six states withhold raw arrestee profile totals, and demographic detail on racial or gender composition is only available for seven jurisdictions responding to FOIA requests. This suggests ongoing challenges in monitoring demographic representativeness and privacy impacts.

The existence of a harmonized, longitudinal dataset supports:

• Quantitative modeling of the relationship between database size and hit-rates.
• Empirical analysis of the impact of state-level policy variation on national growth trends.
• Assessment of demographic disparities in sampling and outcomes as a function of statutory structures.
• Informing policy debates around universal versus targeted DNA collection, considering both forensic efficiency and civil liberties (Pryor et al., 15 Nov 2025).

A plausible implication is that transparent, well-curated longitudinal data can inform evidence-based adjustments in forensic practice, policy design, and debates over privacy versus public safety.

7. Data Transparency and Research Foundations

By synthesizing nearly twenty-five years of monthly NDIS statistics, including rigorous cleaning and cross-jurisdictional harmonization, the underlying dataset enables reproducible research into the structure, composition, and societal impact of U.S. forensic DNA databases. The dataset, methodology, and crosswalks form a platform for both retrospective and predictive analytics within forensic science, legal policy, and bioethics (Pryor et al., 15 Nov 2025).