Do you still trust me? Exploring trust in research partnerships over time

Context: Trust is essential to the participatory health research (PHR) process and strongly influenced by context. A recent study explored trust in a PHR partnership comprehensively and in a multidimensional way but was limited in that they explored trust cross-sectionally without attention to specific attributes. This longitudinal study builds on these findings.

Objective: To explore the evolution of trust longitudinally as a multidimensional concept using social network analysis.

Study Design and Analysis: Participants (both local and national partners) [n=57 (T1); n=56 (T2)] were invited to complete a network survey at two timepoints (May 2021; May 2022) asking 7 statements representing 7 dimensions of trust. Separate trust networks were constructed for each trust dimension at both timepoints. If participants ‘agreed’ or ‘strongly agreed’ with trust statements, a connection was present in the network map.

Setting or Dataset: Public and Patient Involvement (PPI) Ignite Network in Ireland.

Population Studied: Individuals representing academic, service and community organizations from PPI Ignite Network.

Intervention/Instrument: Social network analysis survey Outcome Measures: Network measures were calculated to identify changes in the networks at T1 and T2 and between partner type. Individual level measures: in-degree, weighted in-degree, betweenness centrality, reciprocity; Network level measures: average in-degree, clustering coefficient, Freeman centralization about the in-degree. Kolmogorov-Smirnov (KS) test and Hamming and Ipsen-Mikhailov (HIM) distance explored changes in trust over time

Results: Subtle changes were detected across the 7 dimensions of trust over time. When comparing all 7 trust dimensions networks over time stratified by local or national partnership type, both saw a slight reduction in the number of incoming trust connections over time. Some networks were more similar (e.g., vulnerability and integrity) and strikingly different (e.g., integrity and shared values, visions and goals) to each other, with a higher number of incoming connections for national partners vs. local partners.

Conclusion: Our empirical findings support three key considerations for researchers including: 1) SNA is beneficial to explore the evolution of trust in co-production; 2) trust is dynamic and should be explored over time; and 3) trust needs to be explored as a multidimensional concept to support appropriate interventions to improve partnerships.