Skip to main content
. 2026 Sep 18;13:1910725. doi: 10.3389/fnut.2026.1910725

Table 2.

Results of stakeholder network analysis.

Network Type Structural characteristics Cohesion Centrality
Nodes Edges Components Isolates Largest component nodes Average path length Density Highest in-degrees† Highest out-degrees‡ Highest betweenness§
Policy/regulatory Directed 32 39 2 0 30 3.04 0.04 MHC(3) NHC(9) PHC(10.5)
Technical advisory Directed 33 36 3 0 27 2.19 0.03 CNLIC(3) CNSIC Ltd. (3) HCPs(22) 0
Collaboration Undirected 57 156 2 0 55 3.06 0.10 N/A N/A HCPs(501)
Advocacy/promotion Directed 25 23 3 0 21 2.15 0.04 Public(17) OLSC(2)
MHC(2)
PHC(2)
OLSC(2)

†In-degree measures the number of links directed toward a stakeholder, indicating how frequently the stakeholder is referenced, targeted, or receives actions from others within a given network.

‡Out-degree measures the number of links initiated by a stakeholder and directed toward others, indicating how frequently the stakeholder provides guidance, initiates actions, or communicates with other actors.

§Betweenness centrality values are unnormalized scores and therefore may exceed 1; they represent centrality scores rather than numbers of relationships.

NHC, National Health Commission; PHC, Provincial Health Commission; MHC, Municipal Health Commission; CNLIC, China National Light Industry Council; HCPs, Healthcare providers; OLSC, Other Local Salt Companies; CNSIC Ltd., China National Salt Industry Corporation.

N/A indicates that the measure was not applicable because collaboration relationships were treated as undirected networks and therefore in-degree and out-degree were not calculated. A value of 0 indicates that no stakeholder had a measurable bridging role based on betweenness centrality in the corresponding network. Values in parentheses represent the corresponding centrality scores.