Charting the COVID Long Haul Experience - A Longitudinal Exploration of Symptoms, Activity, and Clinical Adherence

Title of the Paper

Charting the COVID Long Haul Experience: A Longitudinal Exploration of Symptoms, Activity, and Clinical Adherence

Paper Information

• Subject Area: Health Informatics and Human-Computer Interaction (HCI), multifaceted study on the impact of long-term COVID-19 symptoms on patients' lives
• Keywords: COVID Long Haul (CLH), Long COVID, PASC, Post-COVID, COVID-19, survey, Fitbit, Electronic Health Records (EHR), interviews, qualitative methods

Research Background and Problem Statement

• Problems and Challenges:

  • COVID Long Haul (CLH) is an emerging chronic condition characterized by diverse and highly uncertain symptoms, with no standardized diagnostic criteria currently available.
  • The medical field's understanding of CLH is often limited to data from Electronic Health Records (EHR), such as diagnoses or problem lists, which fail to fully capture symptom variability, severity, and the impact on patients' daily lives.
  • There is a lack of research on how CLH symptoms manifest in natural settings such as home and workplace environments.

• Significance of the Study:

  • CLH has long-term health impacts on millions of people globally, and its complexity increases the need for designing supportive technologies for patients.
  • The Health Informatics and HCI communities are uniquely positioned to leverage technology to study patients' lived experiences and translate these insights into designs for supportive technologies.

• Research Motivation:

  • To design a comprehensive research framework that integrates multiple data streams (EHR, Fitbit records, survey data, and interviews) to explore the life trajectories and impacts on CLH patients.
  • To provide preliminary recommendations for designing technologies to support CLH patients and advance a more holistic conceptualization of the condition.

Solution

• Research Methods:

  • A 3-month longitudinal study was designed to track the life trajectories of 14 CLH patients.
  • Data sources included:
    • Objective Data: EHR records and daily Fitbit data (activity and sleep).
    • Subjective Data: Weekly surveys and final interviews.
  • Cross-validation and analysis of data were conducted to provide a holistic picture of the interaction between patients' lives and their condition.

• Innovative Aspects:

  • A method combining subjective and objective data streams to address potential information gaps caused by reliance on a single data source.
  • In-depth exploration of how lived experiences influence the use of health-tracking technologies, providing design implications.

• Implementation Steps:

  1. Patient Recruitment and Informed Consent: Patients were recruited through post-COVID clinics, ensuring data privacy and ethical protections.
  2. Data Collection:
     • Regular surveys: Recorded symptom changes, severity, health status, and adherence to medical advice.
     • Fitbit records: Daily tracking of patients' activity and sleep data.
     • EHR data: Monitored compliance with medical recommendations.
     • Interviews: Gained deeper insights into patients' perceptions and life impacts.
  3. Data Analysis: Qualitative methods (thematic analysis) and quantitative analysis (descriptive statistics and visualization) were used to uncover the interaction between patients' lives and health data.

Research Findings

• Specific Discoveries:

  • CLH symptoms are complex and highly variable, with patients reporting a total of 39 symptoms. The most severe included memory loss, fatigue, coughing, shortness of breath, and loss of taste/smell.
  • Patients commonly faced barriers to adherence to medical advice (on average, 23% of recommendations were not followed), with non-adherence primarily attributed to financial burdens, time constraints, and psychological challenges.
  • Fitbit data revealed that patients' activity levels and deep sleep durations were significantly below recommended levels, highlighting the impact on their health.
  • Regular interviews uncovered the challenges patients faced in adapting to a new "normal," including symptom uncertainty, lack of family support, and difficulties navigating the healthcare system.

• Comparative Advantages:

  • Cross-validation of data addressed potential biases from single data streams.
  • Provided rich contextual information to guide the design of future technologies to support CLH patients.

• Experimental/Evaluation Results:

  • Data visualization and thematic analysis revealed the dynamic nature of symptoms and their profound impact on patients' lives.
  • The integration of survey and interview data played a crucial role in understanding patients' subjective health states, particularly regarding self-management and mental health.

• Limitations and Future Directions:

  • Sample Limitations: The sample was primarily composed of white, middle-aged women, necessitating broader demographic representation to improve external validity.
  • Data Incompleteness: Missing data due to memory loss or hospitalizations among patients.
  • Future Research Directions:
    • Explore design solutions to improve patient-technology interaction, such as tools to support memory challenges.
    • Deepen research on data integration techniques and the relationships between multidimensional health metrics.

Conclusion

This study highlights the complexity of CLH through longitudinal data and cross-validation of multiple data streams, emphasizing the potential of technology and design to support patient self-management, social connection, and adaptation to new life circumstances. By systematically integrating patients' subjective and objective data, the research provides valuable insights into understanding this emerging condition and offers guidance for the development of related technologies by the Health Informatics and HCI communities.