Health service utilization after terrorism: a longitudinal study of survivors of the 2011 Utøya attack in Norway

BACKGROUND For effective organization of health services after terror attacks, it is vital to gain insight into survivors' health service utilization. Following the 2011 Utøya mass shooting in Norway, a proactive outreach programme was launched to prevent unmet help needs. All survivors received health services during the first five months, yet an important minority were not proactively followed-up. This study assessed the prevalence of health service utilization and factors associated with mental health service utilization among the survivors 5-15 months after the attack. METHODS The study comprised data from interviews using standardised questionnaires performed 4-5 (T1) and 14-15 (T2) months after the attack. Altogether 281 of 490 (57.3%) survivors answered questions on health service utilization at T2 and were included in this study. Users and non-users of mental health services were compared using Pearson Chi Square tests (categorical variables) and independent t-tests (continuous variables). Multivariate logistic regression analyses were conducted to examine the relationship between mental health service utilization at T2 and early (model 1) and concurrent (model 2) posttraumatic stress reactions, mental distress and somatic symptoms. Both models were adjusted for age, gender and predisaster utilization of mental health services. RESULTS Altogether 267 (95.0%) of 281 survivors reported contact with health services at T2, including 254 (90.4%) with ≥1 types of primary care services; and 192 (68.3%) with mental health services. In bivariate analyses, mental health service utilization was associated with female gender, injuries, PTSD, mental distress, somatic symptoms, and sleep problems. After multivariate adjustments for early symptom levels (model 1), only mental distress remained significantly associated with mental health service utilization at T2 (OR 2.8, 95% CI 1.2-6.8). In the analysis adjusting for concurrent symptom levels (model 2), only somatic symptoms were associated with mental health service utilization (OR 4.4, 95% CI 1.8-10.8). CONCLUSIONS The high utilization of both primary and secondary health services among young survivors 5-15 months after the attack underscores the importance of allocating resources to meet the increased demand for services over a longer time period. The results further highlight the need to address somatic symptoms in disaster survivors who receive mental health services.