skinvision logo
The Healthcare Claims Study
Real-World Healthcare Utilization Study
Smak Gregoor AM, Sangers TE, Bakker LJ, et al. An artificial intelligence based app for skin cancer detection evaluated in a population based setting. NPJ Digit Med. 2023;6(1):90.
Objective:To assess the impact of the SkinVision app on dermatological healthcare consumption in a population-based setting.
Study Design
  • Retrospective, population-based analysis.

  • Dutch health insurer CZ offered free access to the SkinVision app for one year (2019) to 2.2 million adults.

  • App users were matched to non-users to evaluate differences in dermatological healthcare utilization.

Study Population
  • 47,879 individuals created an account (2.2% uptake)

  • 20,777 became active users

  • 18,960 had complete 12-month data available for analysis

    • Mean age: 48.4 years

    • Gender: 53.2% women

    • SkinVision risk assessment:

      • • at least 1 high: 37%
      • • Low: 62.9%
  • Control group: 56,880 matched non-users (matched on age, sex, socioeconomic status, location, prior skin cancer, and co-morbidities)

Primary Outcome:
  • Comparison of the frequency and type of dermatology-related claims between SkinVision users and matched controls.

Key Results

Increased detection of (pre)malignant skin lesions:

SkinVision Users:
6.0% dermatology claims
Matched controls:
4.6% dermatology claims

Represents a 32% higher likelihood of claims for pre(malignant) lesions (odds ratio: 1.32-1.31)

Higher claims for nevi and benign lesions were also observed among SkinVision users compared with controls:

Premalignant lesions claims
1.6% controls [896] vs 2.1% SV users [391]
Malignant lesions claims
3.1% controls [1785] vs 4.1% SV users [773]
Nevi lesions claims
1.2% controls [667] vs 4.5% SV users [850]
Benign lesions claims
0.5% controls [279] vs 1.4% SV users [267]
Unrelated dermatological claims
4.9% controls [2800] vs 5.3% SV users [1001]
Odds ratio for dermatological claims (95% CI)
Key Takeaways
  • Higher detection rates: Users of the SkinVision app demonstrate higher detection rates of malignant and premalignant lesions compared to matched controls.
  • Earlier Healthcare Engagement: App usage is linked to greater awareness of skin health and earlier medical consultations.
  • Proactive skin health behaviour: An increase in benign consultations suggests a rise in user engagement and vigilance regarding skin health.