Multidisciplinary ecosystem to study lifecourse determinants and prevention of early-onset burdensome multimorbidity (MELD-B) – protocol for a research collaboration

Simon Fraser, Sebastian Stannard, Emilia Holland, Michael Boniface, Rebecca B. Hoyle, Rebecca Wilkinson, Ashley Akbari, Mark Ashworth, Ann Berrington, Roberta Chiovoloni, Jessica Enright, Nick Francis, Gareth Giles, Martin Gulliford, Sara Macdonald, Frances S Mair, Rhiannon K Owen, Shantini Paranjothy, Heather A. Parsons, Rubén J. Sánchez-García, Mozhdeh Shiranirad, Zlatko Zlatev, Nisreen A AlwanJournal of Multimorbidity and Comorbidity 2023 · DOI

TL;DR

This paper outlines the MELD-B research collaboration, which aims to understand and prevent early-onset, burdensome multimorbidity (MLTC-M) by analyzing lifecourse determinants and using AI to identify critical intervention points. By integrating diverse datasets and employing advanced analytical techniques, MELD-B seeks to uncover how early life factors and the sequence of long-term conditions contribute to MLTC-M, ultimately informing targeted prevention strategies.

What Problem Does This Paper Address?

The paper addresses the growing challenge of early-onset, burdensome multimorbidity (MLTC-M), defined as multiple long-term conditions (LTCs) occurring before age 65. While the prevalence of MLTC-M is increasing, current understanding of its causes, particularly the influence of early-life factors and the sequence in which conditions develop, is limited. Existing research often focuses on shorter timeframes and doesn't fully capture the complex interplay of social determinants, sentinel conditions (the first LTC), and subsequent LTC accrual. This lack of understanding hinders the development of effective prevention strategies, especially for individuals from disadvantaged backgrounds who are disproportionately affected. MELD-B aims to bridge this gap by providing a comprehensive, lifecourse perspective on MLTC-M.

Methods

MELD-B employs a mixed-methods approach, integrating qualitative evidence synthesis with quantitative analysis. Key methodological components include:

  • Qualitative Evidence Synthesis (QES): To understand the lived experiences of "burdensomeness" and "complexity" from the perspectives of patients, carers, and healthcare professionals.
  • Delphi Technique: A consensus-building method to refine the definition of burdensomeness and complexity and identify indicators for routine healthcare data, as well as potentially preventable aspects.
  • Artificial Intelligence (AI) and Machine Learning: AI methods will be used for clustering MLTC-M patterns, identifying sentinel conditions, and developing semi-supervised learning models to match individuals across birth cohort and routine healthcare datasets.
  • Causal Inference Modeling: To identify "preventable moments" in the lifecourse where interventions could be most effective in preventing MLTC-M.
  • Data Linkage and Harmonization: A significant effort will be made to link and harmonize data from various sources, ensuring consistency for AI and analytical applications.
  • "Safe People, Settings, and Data" Framework: This approach ensures ethical and secure handling of sensitive data, with rigorous training for researchers and robust data governance.

Datasets

The MELD-B collaboration will utilize the following datasets:

  • Longitudinal Birth Cohort Studies:
    • Aberdeen Children of the 1950s (ACONF)
    • National Child Development Study (NCDS)
    • 1970 British Cohort Study (BCS70)
  • Routine Health Datasets:
    • Secure Anonymised Information Linkage (SAIL) Databank (Wales)
    • Clinical Practice Research Database (CPRD) GOLD and Aurum (UK-wide)

These datasets provide rich, longitudinal information from birth through adulthood, encompassing socioeconomic factors, health behaviors, clinical events, and administrative data.

Key Insights and Ideas

  • The Lifecourse Matters: The project emphasizes that understanding MLTC-M requires a lifecourse perspective, recognizing that early-life exposures and developmental trajectories significantly influence later health outcomes.
  • Beyond Simple Counts: Moving beyond simply counting long-term conditions, MELD-B aims to define and quantify "burdensomeness" and "complexity," acknowledging the subjective and multifaceted nature of living with multiple illnesses.
  • AI for Complex Patterns: The collaboration leverages AI to uncover complex, non-obvious patterns and clusters of MLTC-M that might be missed by traditional statistical methods.
  • Identifying "Preventable Moments": A key innovation is the focus on identifying specific time periods in life where interventions could be most impactful in preventing or delaying the onset of burdensome MLTC-M.
  • Patient and Public Involvement (PPI): Integrating PPI throughout the research process is crucial for ensuring that the research is relevant, meaningful, and addresses the lived experiences of those affected by MLTC-M.

Key Results

As this is a protocol paper, specific quantitative results are not yet available. However, the expected outputs include:

  • A deeper understanding of what "burdensomeness" and "complexity" mean to individuals with MLTC-M.
  • A suite of indicators for identifying burdensome MLTC-M in routine healthcare data.
  • Identification of novel, early-onset MLTC-M clusters.
  • Characterization of early-life risk factors and determinants of these clusters.
  • Models of lifecourse trajectories of LTC accrual and burden.
  • Identification of critical time points and targeted prevention scenarios for early-onset MLTC-M.

Limitations and Future Work

The paper acknowledges several challenges:

  • Data Access and Harmonization: Gaining access to and harmonizing variables across five distinct datasets presents a significant operational challenge.
  • Methodological Complexity: Ensuring the use of the most appropriate and evolving analytical methods for complex questions will require ongoing adaptation.
  • Interdisciplinary Collaboration: Managing a large, multidisciplinary team across multiple institutions requires effective coordination and communication.
  • Defining Novel Concepts: Developing and agreeing upon new definitions and terminology for "burdensomeness" and "complexity" across different disciplines is a key undertaking.

Future work will focus on implementing the outlined work packages, generating the identified outputs, and translating these findings into actionable policy and practice recommendations through extensive stakeholder engagement. The project aims to establish pathways for impact, ensuring that the research contributes to real-world improvements in preventing and managing MLTC-M.

Summary updated

Abstract

We will develop deeper understanding of 'burdensomeness' and 'complexity' through a qualitative evidence synthesis and a consensus study. Using safe data environments for analyses across large, representative routine healthcare datasets and birth cohorts, we will apply AI methods to identify early-onset, burdensome MLTC-M clusters and sentinel conditions, develop semi-supervised learning to match individuals across datasets, identify determinants of burdensome clusters, and model trajectories of LTC and burden accrual. We will characterise early-life (under 18 years) risk factors for early-onset, burdensome MLTC-M and sentinel conditions. Finally, using AI and causal inference modelling, we will model potential 'preventable moments', defined as time periods in the life course where there is an opportunity for intervention on risk factors and early determinants to prevent the development of MLTC-M. Patient and public involvement is integrated throughout.
BibTeX
@article{fraser2023multidisciplinary,
  author = {Simon Fraser and Sebastian Stannard and Emilia Holland and Michael Boniface and Rebecca B. Hoyle and Rebecca Wilkinson and Ashley Akbari and Mark Ashworth and Ann Berrington and Roberta Chiovoloni and Jessica Enright and Nick Francis and Gareth Giles and Martin Gulliford and Sara Macdonald and Frances S Mair and Rhiannon K Owen and Shantini Paranjothy and Heather A. Parsons and Rubén J. Sánchez-García and Mozhdeh Shiranirad and Zlatko Zlatev and Nisreen A Alwan},
  title = {{Multidisciplinary ecosystem to study lifecourse determinants and prevention of early-onset burdensome multimorbidity (MELD-B) – protocol for a research collaboration}},
  journal = {Journal of Multimorbidity and Comorbidity},
  year = {2023},
  doi = {10.1177/26335565231204544},
}
Figure