Social networks are associated with individual’s health and well-being. Working life offers opportunities to create and maintain social networks, while retirement may change these networks. This study examined how the number of ties in social network changes across the retirement transition. The study population consisted of 2319 participants (84% women, mean age 63.2 years) from the Finnish Retirement and Aging study. Information about social network ties, including the number of ties in the inner, middle and outer circles of the social convoy model, was gathered using annual postal surveys before and after retirement. Three repeat surveys per participant covered the retirement transition and the post-retirement periods. Mean number of network ties was 21.6 before retirement, of which 5.6 were situated in the inner, 6.9 in the middle and 9.1 in the outer circle. The number of ties in the outer circle decreased by 0.67 (95% CI − 0.92, − 0.42) during the retirement transition period, but not during the post-retirement period (0.11, 95% CI − 0.33, 0.12) (interaction period * time, p = 0.006). The pattern of change in these ties did not differ by gender, occupational status, marital status, number of chronic diseases and mental health during the retirement transition period. The number of ties in the inner and middle circles overall did not decrease during these periods. The number of peripheral relationships decreased during the retirement transition but not after that, suggesting that the observed reduction is more likely to be associated with retirement rather than aging.

社交网络与个人的健康和福祉相关。工作生活提供了创建和维护社交网络的机会，而退休可能会改变这些网络。这项研究研究了社交网络中的联系数量在退休过渡过程中如何变化。研究对象包括来自芬兰退休与老龄化研究的 2319 名参与者（84% 为女性，平均年龄 63.2 岁）。有关社交网络联系的信息，包括社会护航模型内圈、中圈和外圈的联系数量，是通过退休前后的年度邮政调查收集的。每个参与者进行三项重复调查，涵盖退休过渡期和退休后时期。退休前的平均网络联系数为21.6，其中5.6个位于内圈，6.9个位于中圈，9.1个位于外圈。在退休过渡期，外圈的联系数量减少了 0.67 (95% CI − 0.92, − 0.42)，但在退休后时期则没有减少 (0.11, 95% CI − 0.33, 0.12)（交互作用期 *时间， p = 0.006）。退休过渡期间，这些关系的变化模式并不因性别、职业状况、婚姻状况、慢性病数量和心理健康状况而有所不同。在此期间，内圈和中圈的领带数量总体上并没有减少。外围关系的数量在退休过渡期间减少，但之后没有减少，这表明观察到的减少更有可能与退休有关，而不是与衰老有关。