Background: The incidence of hospital-treated concussion is 100–300/100,000 person years. Reporting of long-lasting post-concussion symptoms (PCS) is estimated at 5–15%. Attachment insecurity is a potential vulnerability factor for physical illness and poorer disease outcomes in general. This study aimed to explore associations between attachment insecurity and PCS in young people sustaining a concussion. Methods: This cross-sectional study was embedded in a cohort of 15–30-year-old patients (n = 3080) 3 months after sustaining a concussion. Data were obtained from a database and questionnaires. PCS were measured by the Rivermead Post-Concussion Symptoms Questionnaire and attachment dimensions (anxiety and avoidance) by the Experiences in Close Relationships-Relationship Structures Questionnaire. Multiple linear regression models were performed to investigate the association between the attachment dimensions and PCS with adjustment for demographic, injury-related and psychological factors and with additional testing for interaction between the attachment dimensions. Results: In the final study sample, comprising 973 patients (31.6%), we found an interaction between the attachment dimensions. Hence, the effect of attachment anxiety on PCS was statistically insignificant at low avoidance (25th percentile) but significant at high avoidance (75th percentile, β = 0.64 (95%CI: 0.02; 1.26)), whereas the effect of attachment avoidance was significant regardless of level of attachment anxiety (25th percentile, β = 1.09 (95%CI: 0.18; 2.01); 75th percentile, β = 2.71 (95%CI: 1.80; 3.61)). Conclusion: Attachment insecurity, especially characterised by high avoidance in combination with high anxiety, also called fearful attachment, is associated with PCS. Considering the attachment perspective can potentially improve health care for this patient group.