The problem of suicide among Amerindians in Camopi-Trois Sauts, French Guiana 2008–2015

Although there are marked variations, suicide kills more persons worldwide than wars or natural disasters [1]. In 2012, the global age-standardized suicide rate was 11.4 per 100,000 inhabitants (15 among men and 8 among women). Among industrialized countries, France has a relatively high suicide rate with about 195,000 suicide attempts and 12,000 suicides [2]. Although completed suicide mostly affects men, suicide attempts mostly affect women in the 15–19-year-old age group, followed by women in the 40–50 -year-old age group. From a geographical point of view, there are differences between French regions, the suicide rate ranging from 9 per 100,000 in Île-de-France to 28 per 100,000 in Bretagne.

The national strategy of suicide prevention is articulated around four priority interventions which are: prevention, reduction of access to lethal means and improvement of care and of epidemiological knowledge. In France, INSERM –CépiDC (French National Medical Research Institute) releases essential indicators on suicide that are based on medical death certificates. However, in French Guiana death certificates are not always performed because of the isolation of certain villages and the cultural barriers in some populations who bury their dead before any investigation.

For 2013, the CépiDC tallied 16 suicides in French Guiana for a total population of 246,507 inhabitants, thus corresponding to a suicide rate of 6.5 per 100,000 inhabitants, a rate that is probably under-estimated. Moreover, it was not possible to disaggregate this data per municipality. In contrast to Mainland France, there is no regional observatory for suicide in French Guiana. The population in French Guiana is very diverse in terms of culture, language, socioeconomic level. Most people live along the coast in one of the three main cities Saint Laurent du Maroni, Kourou, and Cayenne. About 20% of the population, mostly Maroons and Amerindians lives along the Maroni and Oyapock rivers bordering Suriname and Brazil, respectively. In 2006, a study on suicides among Wayana/Emerillon Amerindians living on the French bank of the Maroni river revealed a suicide rate of 280 per 100,000 inhabitants [3]. The study also showed that persons under the age of 25 years, and notably those <15 years, were most affected (70%). Suicide affected both sexes.

In contrast with mainland France, suicide among Wayana Amerindians starts at a very early age and declines thereafter.

On the Oyapock river, which delineates the border with Brazil, the municipality of Camopi-Trois Sauts has also been affected by suicides. Wayãpi and Teko Amerindians have resided there since the 1950’s and the villages have rapidly grown with 1741 persons in 2014 (https://​www.​insee.​fr/​fr/​statistiques/​2011101?​geo=​DEP-973). Hunting, fishing, gathering and slash and burn agriculture are the traditional means of subsistence. Wayãpi and Teko populations have long remained in isolation, with strict regulations for accessing their territories, notably for the protection against “western” communicable diseases. However, for the past two decades they have been confronted with accelerated modernization, the consequences of welfare benefits, the environmental, health, and social consequences of illegal gold mining. Although there are primary schools in the remote villages, there are no high schools and the children thus need to go to other towns to continue their education. All these changes have led to psychological suffering, notably addictions and suicidal behavior. The main objective was thus to tally the suicide attempts and the suicides between 2008 and 2015. The secondary objectives were to identify risk factors and to grasp some of the representations of suicide in this population.

The age of the youngest individual who committed suicide identified by the health center was 13 years. The oldest individual was 41 years old. The median age was 17 years. Overall 17/24 of all deaths were in the 10–20-year-old age group. The suicide method was mostly hanging (18 hangings, five suicides by firearm, one using medication).

Figure 2 shows the frequency of suicides by age between 2008 and 2015, in the Camopi-Trois Sauts district.

The problem of suicide in autochthonous populations is not new [4‐16], but it is neglected [11]. This is the first epidemiological study on suicide in the population of the Camopi-Trois Sauts district. It shows a gradual increase of suicide events between 2008 and 2015. It is difficult to say whether this trend is stable or whether it reflects an transient phenomenon [17], a Werther effect [18, 19], suggested by the frequency of suicidal events in the close family [20]. The present study had several limitations: case ascertainment, questionnaire development and administration and outcome ascertainment were potential limitations; even though at the scale of a small village suicide was a major problem, the absolute overall number of events per year was “small”, which may lead to wide fluctuations and more extreme rates; some of the data was declarative, which may have biased responses towards more socially desirable answers; and finally the study period may have been too short to witness long term trends. However, despite broad confidence intervals in the annual estimates, even the lower boundary of the 95% confidence interval in Camopi-Trois Sauts was still higher than estimates in France (36 vs. 16 per 100,000, respectively).

Elders in the village reported that, in the past, it was not exceptional for persons to want to end their life, but that it was not as frequent as in recent years [21]. Thus there seemed to be a real increase of suicide. This trend has also been observed in other indigenous populations. The observed suicide rate observed in Camopi-Trois Sauts is close to what was reported on the upper Maroni in 2006 (280 per 100,000) [3] with similar characteristics in terms of age and gender. Although age categorization differences made it impossible to standardize suicide rates by age, the crude comparison with mainland France is still quite striking: 172/100000 versus 16/100000 (2015), hence over 10 times more [22, 23]. In practice, this means that for the 2008–2015 period 1.4% (95% CI = 0.9–2.1) of the population of Camopi-Trois Sauts committed suicide, more specifically 3% of the 15–29 year-old age group; it also means that 3.2% (95% CI = 2.4–4.2) of the same population, and 6.5% (1 in 15 persons) of the 15–29 year-old age group, had attempted suicide. The small size of the villages implies that nearly everyone has seen or felt the impact of this trauma. The epidemiologic studies of different autochthonous populations usually observe high suicide rates, with a similar epidemiological profile in terms of age but they often report a male predominance, which was not the case in Camopi-Trois Sauts. Despite similarities, the incidence observed in French Guiana seems particularly high relative to these studies (51.7/100000 in young men in Alaska [16, 24, 25]). It is possible that this very high rate is a peak value measured on a small sample at the apex of a transient phenomenon, which may subsequently drop to pre-episode levels. However, it underscores the considerable intensity of the phenomenon.

Apart from the sole quantitative aspect, there were also specific differences with mainland France, with a very young age at the time of the suicidal act in Camopi-Trois Sauts (71% were less than 25 years) whereas, in France the incidence of suicide was 6.4/100000 for the 15–24 year-old age group and 40.3/100000 for the 85–94 year-old age group.

The average yearly suicide attempt rate was 396 per 100,000 inhabitants for Camopi-Trois Sauts, a rate over twice greater than the crude rate observed in mainland France (177 per 100,000 inhabitants).

In addition, the lethality rate of 30.4%, which is about four times higher than in mainland France (8.2%), shows the radical nature of suicidal acts in Camopi-Trois Sauts, where most suicide attempts were by hanging.

The principal life events preceding suicide were the death of someone close, interpersonal conflicts, professional or financial problems, an interruption of school or unemployment, and finally health problems. This is relatively similar to the literature. It is of note that persons often cumulated several difficult life events, and nearly all reported at least one.

It is not easy to intervene on the occurrence of these first life events. However, the prevention and support provided aim at reducing suicidal events.

Concerning the prevention and management of interpersonal conflicts, traditional leaders and traditional modes of regulation must be rehabilitated [26] in partnership with the municipality (https://​www.​interieur.​gouv.​fr/​SG-CIPDR/​Outils-et-initiatives/​Les-outils-du-Maire/​Le-guide-d-installation-du-CDDF). The chronic problem of illegal gold mining, which is very real in these remote areas, adds stress and division in the community. The acknowledgement of psychological suffering and mental health and addiction issues by psychiatry professionals assisted by local mediators seems important [5, 6, 27, 28]. Childhood abuse is a known risk factor for suicide [29], it was reported in a non-negligible number of cases. The creation of the regional cell for the well-being of populations of the interior (CeRMEPI) has an important coordinating role for actions that aim at preventing the life events triggering suicide.

The frequency of conflicts, the fear of gossip, and the resonance of life events may be linked to the small size of the villages and the geographic and family proximity of suicides and suicide attempts. Thus 61% of those surveyed revealed a suicidal event of a family member, 45% of whom were first degree relatives. These percentages are the proof a serial phenomenon that must be contained.

As elsewhere suicide was closely associated with alcohol [30, 31]. Most persons who were drunk during their suicide attempt were addicted to alcohol. The repetition of a suicide attempt was significantly linked to the consumption of marijuana, a potentially interesting information for prevention [27]. Recent studies of a different design seemed to suggest that alcohol and marijuana consumption were greater than in the general population [32, 33]. Psychiatric risk factors have a clear influence on suicidal acts. Untreated depression is considered as the first cause of suicide [27, 28].

The unusually high lethality rate could raise the question of ascertainment bias and the underestimation of suicide attempts relative to suicides. However, given the size of the villages, underestimation seemed less likely than at the scale of larger territories. Therefore, the high lethality rate, and the lethal methods used in suicide attempts seem more indicative of the acuity of lethal intent in attempted suicides. The balanced sex ratio, which is unusual, could suggest particularities regarding gender relations with particular psychological distress among young women. We did not have any data to further explore this hypothesis but previous studies showed 11% of women experiencing forced first intercourse [33]. The absence of secondary schools, leads to separation between youths and their family as they move to other towns to continue studying. This presumably leads them to grow apart from traditional culture and fuels intergenerational conflicts, situations that may exacerbate psychological suffering when they come back to their village, isolated, torn between traditional and modern cultures in a context where suicide is omnipresent [34]. The regional cell for the well-being of populations of the interior (CeRMEPI), trained local mediators and educators to assist those with mental health problems, and prevent suicide, rapid response psychiatric teams may have an important role to relieve psychological suffering among youths in these areas. The above interventions (CeRMEPI, on site schools, trained educators and mediators, psychiatric teams to implement an emergency medico-psychologic response in less than 24 h) may help alleviate psychological suffering and suicidality. The interviewed persons were favorable to the idea of psychiatric help. However, these interventions should be prospectively evaluated and the present efforts to quantify the problem of suicide should be pursued in order to see if there is any progress following the interventions. A recent parliamentary mission published a report on the subject of suicide in this area [35]. The report detailed 37 propositions to improve well being and reduce suicidality in these territories, notably the presence of a permanent psychiatric facility in Camopi, and emergency medico-psychological task force able to respond within 24 h, and the creation of an observatory of suicide, which definitely would be useful to follow future trends.

Despite limitations linked to case and outcome ascertainment, small population size and limited time frame, the study of the data from the health centers allowed quantifying and characterizing the phenomenon of suicide in Camopi-Trois Sauts. There was a high suicide rate with an increasing trend, (1 person out of 15 in the 15–29 year-old age group had attempted suicide). Suicides essentially concerned young persons of both sexes, most occurring between 10 and 20 years old. These suicides seemed to follow one or more precipitating factors, notably interpersonal conflicts, often in a context of alcohol addiction and isolation. The consumption of marijuana seemed associated with recurrent suicidal acts. Persons who had attempted suicide were willing to accept psychiatric help or addiction treatment. The suspected determinants were linked to the individual (mental illness), to social aspects (absence of high schools) and to culture (conflict management) [36]. Thus suicide prevention will require the therapeutic alliance between the community, traditional healers [5], health professionals, and educators.