What Helps or Hinders Therapeutic Relationships in Inpatient Mental Healthcare: a Systematic Review to Inform Policy and Practice

Background: Therapeutic relationships between service users and staff on inpatient mental health wards are central to the delivery of high-quality care and have been prioritised by policy makers. Yet these relationships are frequently found to be poor across inpatient settings, and more evidence is needed on how to foster good therapeutic relationships. In collaboration with lived experience researchers, we conducted a large systematic review, which was requested and carried out in partnership with NHS England, to identify facilitators and barriers to effective therapeutic relationships on inpatient wards.

Methods: We searched MEDLINE, EMBASE, PsycINFO, CINAHL, HMIC (between January 1999-September 2025) for quantitative, qualitative or mixed methods primary research studies. Study quality was assessed using the Mixed Method Appraisal Tool (MMAT). Data were synthesised using an iterative analytic process, informed by framework analysis and narrative synthesis, across five interconnected levels: 1) individual (service user and staff experiences), 2) microsystem (daily interactions), 3) mesosystem (ward environment), 4) macrosystem (the wider organisation), 5) chronosystem (time and change across levels).

Results: 184 studies with 19,984 participants (n=16,138 service users, n=3,846 staff) were included. Across study designs, four main categories emerged: 1) reducing power imbalances helped facilitate positive relationships and, conversely, restrictive practices damaged them; 2) staff relational skills including empathy, non-judgemental approaches and treating service users as individuals, rather than simply ‘patients’, helped build positive relationships; 3) staff continuity and time for staff to connect with service users was critical to building positive relationships; ward contexts that prioritised task completion and/or were risk-focused weakened relationships; 4) ward environments that created opportunities to build interpersonal connections between staff and service users (e.g. shared spaces) strengthened relationships whereas those that provided limited opportunities for interaction weakened relationships.

Conclusion: We identified four key factors that can help or hinder therapeutic relationships on mental health inpatient wards: power dynamics between staff and service users; restrictive practices; relational skills of staff; staff availability and consistency, and ward environments that facilitated interpersonal connections. We found evidence gaps particularly for minoritised ethnic communities and organisational influences. We suggest strategies which could help relationship building, including support with repairing relational ruptures, greater emphasis on relationship building in staff roles, and ward environments that support connection.

Registration: The protocol was registered with PROSPERO (registration number: CRD420251047929).