This study examined factors associated with post-disaster mental health service utilization in survivors of the Enschede fireworks explosion in The Netherlands. Our results provided evidence that demographic- and disaster related variables, self-reported symptoms and physician diagnosed health problems predicted MHS utilization after the disaster. Younger age, unemployment, immigrant status, low SES, forced relocation and personal loss/injuries as a result of the disaster were among the demographic- and disaster related variables predisposing for post-disaster MHS-utilization. Survivors who reported higher levels of emotional problems and problems on social functioning directly after the disaster were more likely to seek post-disaster MHS utilization. Regarding physician diagnosed health problems, pre-disaster psychological and musculoskeletal problems predicted post-disaster MHS use. When al variables were taken into account, disaster intrusions and avoidance reactions, symptoms of hostility and chronic diseases prior to the disaster were found to be the most important factors to predict post-disaster MHS utilization.
Our findings regarding the influence of demographic variables and disaster related variables on service utilization are to a large extent in line with what is found in earlier studies [1
]. However, in our study we found an opposite effect for immigrant status (more likely to use MHS) in comparison to other studies [10
]. Our finding that immigrants had a higher chance of using MHS was also found in other studies after this disaster where evidence was found that affected immigrants reported more psychological problems (before the disaster) and use MHS more often than native survivors [32
]. A plausible explanation for the higher rates of MHS utilization among this group is that their higher rates of pre-disaster psychological health problems may be indirectly related to increased post-disaster MHS use. Possibly immigrants also displayed higher MHS use prior to the disaster. However, in our study pre-disaster MHS data were not available so we could not test this. Another explanation for the higher MHS utilization among immigrants in our sample could be found in the strategy of the MHS-unit which acted pro-actively with regard to minority groups. However, our definition of immigrants (first and second generation) is different from studies who use ethnicity and therefore can not be compared with [9
Furthermore, we found in our study is that sex did not predispose for MHS use although former studies showed a positive relation for females [9
]. Also, Dutch women use mental health services more often than their counterparts [38
]. Another study among survivors of the same disaster found no differences in post-disaster psychological problems between men and women in general practice [39
]. The absence of sex differences in help seeking behaviour may therefore be characteristic for this disaster.
We found that disaster intrusions and avoidance reactions and symptoms of hostility were important factors to predict post-disaster MHS utilization. The symptoms can be seen as the main clusters (re-experiencing, avoidance, arousal) of PTSD (according to the DSM-IV-criteria) which is found to be positively related to MHS use [10
It is acknowledged that survivors presenting pre-disaster psychological problems are more at risk for psychological and physical health problems after the disaster [1
]. Never reported before is our finding that survivors presenting physical health problems before the disaster are more likely to make use of MHS, even after controlling for confounders (table ). Pre-disaster chronic diseases remained also an important independent predictor for post-disaster MHS-utilization after controlling for other variables (table ). Our finding that somatic symptoms predict mental health seeking can be explained by the understanding that physical health problems are positively related to psychopathology and disability, and as a result of that, also to the need for mental health treatment. Another explanation for higher MHS utilization among individuals presenting chronic diseases is that they might be more likely to visit their general practitioner who diagnoses mental health problems and initiate treatment [11
]. This finding implies that patients with chronic diseases are vulnerable during a disaster and therefore may be in need for mental support afterwards.
Several limitations should be addressed. The sample included a limited number of people as information from only one MHS unit was used. Persons visiting other health care providers (private psychologists and psychiatrists, inpatient institutions) were not included. This indicates that our results may be typical for the population who sought help in this unit. Although the study-population did not differ on demographic and disaster-related variables from MHS-patients who did not participate, it is possible that a selection has occurred, limiting the generalizability of the results. However, a study investigating selective participation in the health surveys of affected residents found that even though there was selective participation in the surveys, this did not affect the results [42
]. Also our MHS data did not differentiate between the type of contact given (visits for preventive reasons, treatment) and treatment-time. A drawback of the study is that we did not measure pre-disaster mental health care utilization as the registration system was not running before the disaster. It is acknowledged that former treatment predicts post-disaster treatment [43
]. The registration system in the MHS did not attain information on diagnoses (and therefore the diagnoses of PTSD could not be made), our finding underscores the importance of diagnosing in mental health services and in general practice (also for comparison purposes). Our 'non-MHS-group' was defined as disaster survivors who did not visit the specific MHS unit. However, it is possible they received post-disaster treatment of trauma in the private circuit. Further examination by self-reports 18 months post-disaster showed that around 5.6% of the disaster survivors in this group contacted a private psychologist/psychiatrist in the past 12 months for their disaster-related health problems.
The strength of our study is that we had a unique opportunity to combine survey data with medical records from general practitioners, allowing the collection of both subjective and objective information respectively. Besides, actual pre-disaster information on health status was available from the medical records. Having these pre-disaster data is rather unique in disaster research as most studies lack these data or are measured retrospectively which is more prone to recall bias. By using electronic records (EMRs of GPs and MHS) we excluded the possibility of respondents' recall bias and patients were not burdened in the data collections, which is an important issue after a catastrophic event. The present study fills a gap in disaster literature as most studies rely on self-reports only while in the present study both self-reported data and data from electronic medical records (containing pre-disaster data) were combined. Another strength of our study is that we studied a MHS-unit which was specially implemented for disaster-related treatment and all patients registered in this unit suffered disaster-related problems. GPs and health care professionals in the city were motivated to refer patients with disaster-related problems to this service.