Identifying households with children who have complex needs: a segmentation model for integrated care systems

Area of Science:

• Public Health
• Health Services Research
• Social Policy

Background:

• Childhood adversity is rising in the UK, with complex health and social issues clustering in high-need families.
• Existing services often fail to support the whole family unit effectively.
• Household-level segmentation is proposed as a method to identify families requiring integrated support.

Purpose of the Study:

• To develop a segmentation model to identify households with children exhibiting complex, cross-sectoral needs.
• To utilize this model as a case-finding tool for health and social care services.
• To characterize identified households to inform service integration strategies.

Main Methods:

• Collaborated with stakeholders to define complex needs for service integration, focusing on high-intensity use and co-occurring physical/mental health issues and social risks.
• Utilized linked individual and household data across multiple health and social care services.
• Developed an algorithm for a case-finding interface and employed descriptive statistics and Poisson regression for profiling and comparison.

Main Results:

• Identified 21,527 households (8% of regional population) with complex needs, encompassing 89,631 individuals, including 41,382 children.
• These households accounted for 34% of total health and social care costs for families with children (£362 million).
• Characterized by high prevalence of mental health problems, often comorbid with respiratory issues, and high emergency mental health service use, particularly among teenage girls.

Conclusions:

• The developed model serves as a valuable tool for identifying households eligible for integrated services and targeted family support.
• This approach can potentially disrupt the intergenerational transmission of adversity.
• Highlights the significant cost burden associated with these complex needs within the health and social care system.