We aimed to explore the impact of suicide bereavement on GP consultations for mental health reasons in the years before and after loss. The results show that losing someone to suicide leads to a significant increase in GP consultations for mental health reasons. The size of this effect depends on the kinship to the deceased and the gender of the bereaved, with women exhibiting a larger increase than men and bereaved partners having the largest increase followed by those losing an offspring, parent and sibling.

Our finding that those bereaved by suicide had a higher contact rate with their GP for mental health reasons compared to both the general population and those bereaved by other types of deaths is in line with previous studies finding pre-bereavement differences [5]. Reasons for this might be both genetic and/or environmental risk factors for mental health problems shared with the deceased and assortative mating for partners. The strain of having a suicide threatened relative might also affect mental health, thus one could expect an increasing contact rate as the time of death approaches. We did not find such a pattern, except for partners. One plausible explanation for this is that partners are often the ones closest to the person dying and might experience more of the strain of taking care of a person in crisis.

Compared with those bereaved by other causes, the impact of losing a relative to suicide on GP consultations for mental health reasons is much stronger. This is also reflected in the proportion having had at least one consultation after the death being highest in the suicide group at all time points. Caution is warranted when comparing these groups as they may differ in important aspects. For instance, the age of the deceased is higher in the other groups, and the distribution of kinship types varies. These factors probably contribute to the differences in impact. In any case, our findings do not indicate that those bereaved by suicide are less in contact with their GP for mental health reasons than other bereaved. Regarding the time at which help is received, one study has shown bereaved by suicide were more likely to receive formal help at a delayed time compared to other bereaved [16]. Such a tendency was not found in the current study as the peak consultation rate was in the 0–1 months following the death and that around one third had seen their GP for mental health reasons at least once at this point of time. As time passed after the death, we observed a decrease in consultation rate and a stabilization at a somewhat higher level than pre-bereavement. It is natural that the contact rate is highest right after the death and might reflect that many need assistance to deal with the initial shock such as sick leave, prescriptions and information. Many might also need a referral to specialized health care, and one would expect contact with the GP to decline when patients receive treatment elsewhere.

Two previous studies assessing the needs of those bereaved by suicide showed that a large majority (around 90%) indicated a need for professional help, while around half the sample indicated that they had received formal help [6, 7]. The latter is comparable to the proportion of those bereaved by suicide that have been in contact with their GP for mental health reasons in the current study. This might point to an under-usage of primary health care after suicide bereavement. On the other hand, the mentioned studies are prone to selection bias and might overrepresent bereaved keenly interested in receiving treatment and services. The GP is only one of several sources of formal support available after suicide bereavement. As one has to go through the GP to access specialized psychiatric health care, most of those receiving this type of help will first have seen a GP. Private health care providers are an option for those who can afford it, but the number of private health services is relatively low in Norway [22]. Another important source of support is from NGOs offering support groups or counselling [8], although evidence from a Danish study indicates that a minority of those bereaved by suicide sought this kind of help [20]. We did not have information on alternative sources of support, and receiving help elsewhere might be a reason for not contacting a GP after bereavement. However, rather than seeing different sources of support as alternatives, previous research points out that those bereaved by suicide desire a range of assistance forms [7, 8]. It could also be the case that those accessing one type of support are more likely to also get support elsewhere. One recent report found this to be the case with an association between informal and formal support, open-ended qualitative questions pointed towards informal support from friends and family being an important reason for seeking professional help [23].

In this study, men had a lower rate of GP consultations for mental health reasons than women in the years leading up to the death, as would be expected [19]. This gender difference was amplified when the death occurred and remained high in the first year. These findings support the notion that gender is an important determinant of how those bereaved by suicide will respond to the loss [24]. It might be that men engage in different coping styles than women and, therefore, to a lesser degree rely on help from the health care system. On the other hand, men and women are shown to have a similar increase in risk of mental disease and suicide after suicide bereavement [12, 24]. Considering this, it seems likely that at least some of the large gender differences can be ascribed to an under-usage of primary health care among suicide bereaved men.

Regarding the kinship to the deceased, we found considerable differences between the four groups of suicide bereaved. When the death occurred, all groups exhibited a clear increase in the rate of GP consultations due to mental health reasons. The increase was highest for partners, maybe reflecting that this group undergo the largest changes in their life situation because of the death. Although we controlled for age, the difference seen between kinship groups might still be influenced by age differences between the groups that can lead to differences in occupational status and the need for sick leave. Siblings exhibited the lowest increase. Losing a sibling to suicide has received less attention compared to other relationships and suicide bereaved siblings have been referred to as “the forgotten bereaved” [25]. The experiences of suicide bereaved siblings vary, with some having difficulties reaching out for help because of overwhelming grief, while others do not experience a need for professional help [26].

The use of registry data entails some important limitations. Using health care use as an outcome, we only gain information on actual help seeking behavior, and we lack information on the mental health of those not seeking help. It is likely that this group consists of a mix of those not in need of help and people who are struggling, but unable to obtain adequate help. Findings using survey data support this [23]. Furthermore, we do not know the exact reasons for contacting the GP, nor do we have information on whether or what kinds of interventions were used in the consultations. To increase the chance of consultations being related to the bereavement we limited the outcome to those receiving a diagnostic code for psychological symptoms or disease. The validity of Norwegian GPs ICPC-2 coding has been deemed as good [27], but the fact that the patient presented with some kind of psychopathology, does not equate that the suicide loss was related to the patients’ problems. The statistical analyses we used take into account previous GP use, this allows us to claim with more certainty that the increase we see at the time of the death is related to the suicide loss. For the descriptive results there is more uncertainty regarding how much can be ascribed to the loss, and these numbers should be regarded as a maximum estimate of how many receive help from their GP in relation to the loss. Even if we assume that a majority of these were consultations related to the loss, we do not know if the contact was perceived as helpful. In fact, some qualitative and survey studies show that not all help from GPs in relation to suicide loss is experienced as helpful [28] and that many GPs feel unsure of how to meet the needs of bereaved patients [8, 29]. It is also important to note that other factors than degree of health issues influence help-seeking behavior. Those who are working might need to see their GP to obtain certified sick leave, while welfare beneficiaries and those who are retired who are similarly affected by the suicide might not contact their GP because they don’t have this need. Finally, we could only include the formal relations present in the official registers. Some bereaved that might be just as close to the deceased, such as close friends and romantic relations [30], are not included. We also lacked information on cohabitants that do not share their officially registered address.

Although the nature of the data used restricts the richness of our conclusions, the use of registry data covering the entire population is also the major strength of this study. By including virtually all those bereaved by suicide that can be identified in registries, the study minimizes selection bias, which is a common methodological weakness in bereavement studies [5, 31]. While there are several studies examining health care use among suicide bereaved, including specific studies on support from the GP [8], there are very few studies offering objective estimates on actual use that are not prone to selection bias. In this regard, our findings are an important contribution to the field of suicide postvention as a much-needed supplement to studies using survey and qualitative methods. This is especially important with regard to gender as the participation of men in bereavement research is difficult to obtain [24], which has led to a lack of gender balanced studies within the field [8]. The use of registries also entails a large sample size, allowing for examination of smaller subgroups and focus on different types of kinship, something that has been invited within the field [5].

Future studies should try to uncover the needs of those who have not been in contact with the health care system after suicide bereavement. Many studies have uncovered barriers to seeking help [23, 26, 32] and point out that the GP should actively reach out to those bereaved by suicide [8]. Studies examining the effect of such active outreach would be a promising avenue to pursue. The current study focused on contact with the GP, but there is also a need for more studies on total health care use to better understand the different ways those bereaved by suicide acquire help from the health care system. Preferably such studies should employ large, population representative samples to avoid selection bias. As mentioned, obtaining sick leave is probably one important need the bereaved have when they meet their GP. One study showed a tenfold increase in the risk of psychiatric sickness absence among parents who had lost an adolescent or young adult offspring to suicide [33]. Effects on occupational outcomes can have a large impact on both the individual and society, hence the need for sick leave among those bereaved by suicide should be further examined.

Our study shows that around half of those bereaved by suicide are in contact with their GP for mental health reasons following the death, and most of these are within the first months. This supports the notion that the GP is a natural place to seek help and GPs should be aware of common needs and challenges of those bereaved by suicide. At the same time, we show that there are also many that are not in contact with their GP, and this is especially true for men and bereaved siblings. Clinicians should be aware of this and consider actively reaching out to patients they know have suffered a suicide loss. Our finding that the contact rate is heightened for a relatively long period of time after the loss, and for many groups does not returns to pre-bereavement level during the study period, is useful knowledge for GPs that can be used in planning care and communicated to those bereaved to help destigmatize the need of long-term professional support after suicide bereavement. Better care for those bereaved by suicide is a goal in the Norwegian action plan for suicide prevention [34]. This paper provides important background information on one aspect of the health care use of those bereaved by suicide that is useful for policy makers going forward. The findings indicate a need to think about how to make systemic changes so that those with an under-usage can be reached, and the special needs of men and bereaved siblings should inform this work.