Exploring Perceptions of Cross-Sectoral Data Sharing with People with Parkinson’s

Document Title

Exploring Perceptions of Cross-Sectoral Data Sharing with People with Parkinson’s

Document Information

• Thematic Area: Digital Health, Smart Homes, Data Sharing and Privacy
• Keywords: Parkinson’s Disease, Smart Homes, Data Sharing, Internet of Things (IoT), Privacy and Security
• Publication Details:
  • Conference: CHI Conference on Human Factors in Computing Systems (CHI ’22)
  • Date and Location: April 29–May 5, 2022, New Orleans, USA
  • DOI: 10.1145/3491102.3501984

Research Background and Issues

• Identified Problems and Challenges:
  1. Traditional assessment methods for Parkinson’s disease (e.g., Movement Disorders Society Unified Parkinson’s Disease Rating Scale, MDS-UPDRS) have several limitations, such as reliance on patients’ condition at specific times and their recollection of past symptoms.
  2. High-quality, multimodal health datasets have the potential to improve healthcare services and research but face challenges related to privacy control and data security. Moreover, existing data-sharing practices have often neglected the perspectives of data subjects (e.g., Parkinson’s patients).
• Significance of the Research:
  • Data sharing can facilitate cross-sectoral collaboration (research, healthcare, industry), accelerating innovation and disease management.
  • It enables more objective and long-term monitoring of symptom changes in Parkinson’s patients while offering opportunities for self-understanding and treatment optimization.
• Motivation and Related Work:
  • Smart homes and wearable devices have been used in health monitoring research, but users’ concerns about data privacy may hinder widespread adoption of these technologies.
  • Existing studies largely focus on single-sector data usage, with limited exploration of the complexities of cross-sectoral health data sharing.

Solution

• Methodology:
  • Study Design:
    1. Use a fully sensor-equipped smart home to simulate the collection of multimodal sensor data as Parkinson’s patients and their partners (control group) perform daily living tasks.
    2. Conduct the study in two phases: the first phase focuses on sensor acceptance and user experience; the second phase engages participants in discussions about potential use cases, details, and concerns regarding data sharing.
  • Innovative Aspects:
    • Combine data visualization to directly present health data to patients, allowing them to explore data-sharing recipients (e.g., healthcare, government, companies) and privacy boundaries.
    • Use context-based design workshops to enable participants to discuss broader ethical and legal issues with a clear understanding.
  • Implementation Steps and Techniques:
    1. Sensors in the smart home record movement and activity patterns, capturing sample data from scenarios such as food preparation and stair climbing.
    2. Provide data visualization tools (e.g., electricity usage curves, activity silhouette videos) to intuitively display Parkinson’s symptom-related data.
    3. Facilitate workshops guiding patients to consider the potential risks and benefits of cross-sectoral data sharing.

Research Findings

• Specific Findings:

  1. Participant Attitude Analysis:
     • Overall acceptance of smart home sensors, with initial concerns about camera-based sensing being alleviated after explanations.
     • Patients are inclined to share data for the benefit of the Parkinson’s community or scientific progress, provided there is clear informed consent and transparent data management.
  2. Trust and Transparency Issues:
     • Healthcare institutions and universities are perceived as trustworthy “data gatekeepers,” while commercial companies (especially small tech firms) face greater challenges.
     • Participants expressed mixed attitudes toward secondary data use and commercial applications (e.g., by pharmaceutical companies or advertisers), recognizing their importance but harboring reservations.
  3. Ethical Exploration:
     • Discussed how contextualized informed consent could enhance the data-sharing process, recommending against a one-size-fits-all “blanket authorization” approach.

• Strengths and Contributions:

  • Provided a patient feedback-based guideline for data-sharing methods, emphasizing the necessity of transparency and long-term control.
  • Used interactive data formats (visualizations, silhouette videos) instead of traditional surveys to enhance understanding.
  • Highlighted patients’ active call for involvement in the consent process, which could influence the ethical framework of future health technology research.

• Experimental Results:

  • Participants generally expressed a willingness to share data with trusted medical and research organizations. However, they remained cautious about potential misuse of data (e.g., by insurance companies for pricing or unreliable medical recommendations).

• Limitations and Future Directions:

  • Limitations:
    • Small and relatively narrow sample size (12 participants; mild to moderate Parkinson’s), limited to the UK context.
    • Participants’ responses may be biased by the unique European data protection regulations (GDPR).
  • Future Directions:
    • Expand the study to patient communities in other countries and regulatory environments.
    • Explore opinions on cross-disease data sharing with long-term, large-scale collection to develop a more universally applicable framework.

Conclusion

• By analyzing Parkinson’s patients’ perceptions of “smart home data sharing” solutions, this study reveals how to balance the trade-offs between health data openness and privacy protection, offering valuable insights for future ethical data-sharing practices.
• Data sharing is not only about the technology itself but also relies on building trust and transparency within the health community.