The strict relationship between health and environment is well known, and there is increasing evidence that climate and weather affect multiple areas of human life, including mental health (1). For example, “pleasant weather” (e.g., elevated temperature or barometric pressure) seems to be related to higher mood levels, better memory, and broadened cognitive style (2), whereas high levels of humidity have been associated with impaired concentration and increased sleepiness (3). Though generally experienced only to a mild degree, some individuals display extreme sensitivity to weather and climatic changes (4, 5).

Psychiatric patients have been described to be particularly vulnerable to climate and weather changes in terms of clinical severity, admission rates, duration of hospitalization, recurrences, and outcome (6–9). Accordingly, a recent study found correlations between different weather conditions and the number of patients accessing the emergency room for psychiatric symptoms (10).

Among psychiatric diseases, bipolar disorder (BD) is the fifth highest as to self-reported disability, represents a significant public health problem, and is saddled with a markedly high mortality rate from suicide (11, 12).

There is a consensus on considering the changing of the seasons as one important environmental trigger for mood switches in BD, and emerging evidence suggests that climatic variables may contribute to recurrences (13, 14). Some studies reported a relationship between hospitalizations for mania and hours of sunshine, or temperature (15–17). Christensen and colleagues (18) found that several meteorological factors, including changes in minimum and maximum temperature, rainfall, atmospheric pressure, and cloudiness, might contribute to triggering new episodes in bipolar patients. Additional findings entailing climate variables include the observed association between hospital admissions for bipolar depression and temperature (19), as well as the prevalence of depressive episodes in relation to latitude (20), although other researchers reported different results (21). Interestingly, a recent study found a significant inverse association between the maximum monthly increase in solar insolation and the age of onset of type I BD (22).

Several evidences described seasonal and climatic-related variations in suicidality, with different patterns in men and women (23–28). Of note, climate and weather seem to independently influence suicidal behavior, both in psychiatric patients and in the general population. Fountoulakis et al. (29) described a correlation between the number of suicides and environmental temperature in different European countries, and climate changes appeared to influence suicidality even more than socio-economic factors. These observations have been further supported by a recent study that found a significant impact of temperature fluctuations on suicide rates in a large sample of subjects across the USA and Mexico (30). Seasonal effects have been reported to play a role in the likelihood of suicidal behaviors in affective disorder patients (31, 32). In manic BD subjects, higher suicide risk has been associated to rapidly increasing temperatures (33).

Following these considerations, we hypothesized that bipolar subjects would show a higher rate of sensitivity to climatic changes and that this feature could be related to suicide risk. Our group devised and validated a self-administered instrument, the METEO-Questionnaire (METEO-Q), to detect sensitivity to climate and weather changes in the general population and to assess their impact on symptom variations in psychiatric patients (34).

The aims of the study were: (i) to investigate sensitivity to weather and climate variation in euthymic bipolar subjects as compared to homogeneous healthy controls, using the METEO-Q; (ii) to evaluate the relationship of susceptibility to weather/climate changes with lifetime suicide attempts (SAs) in BD.

BD is characterized by a high degree of psychopathological burden, with significant rates of relapses and recurrences. Many different factors, both individual and environment-related, have been recognized as triggers for the development of new episodes and the worsening of residual inter-episode symptoms. Among these factors, seasonality (41), the dysregulation of circadian rhythms (42), as well as abnormalities in the biological response to sunlight (43, 44) and moon cycle (45), have been demonstrated to play a significant role. Therefore, climate and weather sensitivity would be expected to be related to the course of the illness as well.

In this study, we found a higher susceptibility to weather and climatic changes in BD patients compared to homogeneous HC. Since patients were all euthymic and evaluated at different times of the year, results are likely not influenced by current mood state or specific seasonality, possibly representing an enduring trait (46).

The hypothesized neurobiological mechanisms influencing sensitivity to climate changes are complex and involve many factors, including different neurotransmission pathways, the immune and neuroendocrine systems, and the activity of genes regulating circadian rhythms (14, 47).

Recently, genetic variations in the circadian gene neuronal PAS domain protein 2 (NPAS2) have been proposed as a biomarker for a seasonal pattern in BD (41). Of note, both circadian rhythms and the response to sunlight and to climatic variations are regulated by the hypothalamus, whose role in the pathophysiology of BD has been discussed (48).

BD is associated with a high risk of SAs and deaths. It is estimated that the risk of suicide in BD patients is 20–30 times greater than in the general population (49). Epidemiological studies report that about 23%–26% of BD patients attempt suicide and that BD accounts for up to 14% of all deaths by suicide (50).

Intriguingly, in our clinical sample the sensitivity to weather changes was associated with the presence and number of lifetime SAs. This finding is relevant, since a history of a prior SA has a documented predictive role for completed suicide, and repeated attempts further increase the risk of death by suicide (51). There is evidence supporting the influence of seasonal changes on suicidal behavior that seems to be related to climatic variance and light exposure, possibly through an interaction with serotonergic and noradrenergic circuits (29, 47). Meteorological variables, such as temperature, atmospheric pressure, and sunlight, are linked to changes in the concentration of cerebral neurotransmitters and alterations in the serotonin turnover in the brain (47). A study by Makris et al. (52) found a correlation between sunshine and SAs in patients treated with serotonergic medications (i.e., SSRI). Therefore, environmental, weather-related neurobiological alterations of the serotonergic system might mediate suicidal behavior in predisposed individuals (47). Of note, dysfunctions in the serotonergic and dopaminergic systems have been also shown to play a role in the pathophysiological mechanisms underlying emotion dysregulation, which, in turn, is strongly linked to mood recurrences, impulsive behaviors, and substance abuse, ultimately leading to an increased risk of suicide in BD patients (53–55). Similarly, meteorological factors, including low barometric pressure, high temperatures and precipitations, have been associated with decreased sleep quality and abnormal circadian patterns (56, 57), which, in turn, may represent a potential marker of SA in BD (58).

Seasonal and climatic factors might interact with each other, influencing the manifestation of psychiatric symptoms mostly affecting suicide completers (59). Balestrieri et al. (60) suggested that the assessment of lifetime rhythmicity and manic-hypomanic features may be clinically useful to identify potential suicide attempters in high-risk groups. In a recent study, Bauer et al. (61) found an association between latitude-related changes in solar insolation and increased SAs in patients with BD. Finally, the weather has a significant impact on social activities and interactions, especially in vulnerable populations. Interpersonal stress, low social connectedness, and poor social support have been recognized as major risk factors for suicide (62–64).

In our study, women displayed a greater tendency to climatic sensitivity than men both in the clinical and in the control group, a finding in line with earlier observations (65). The greater influence exerted by seasonality and climatic changes on women could be determined by the different action of bioclimatic factors on neurohormonal systems.

Increased sensitivity to climate and weather variables, especially among fragile populations such as psychiatric patients, represents an issue to be considered in the light of the severe climatic modifications currently taking place. In fact, global climate change is now regarded as one of the biggest crisis facing humanity, and the impact of its consequences on mental health, both in adults and children, has been recognized as a major concern among mental health specialists (66–72). Accordingly, in 2017 the American Psychiatry Association (APA) released a position statement that climate change “poses a threat to public health, including mental health”, with people suffering from psychiatric disorders expected to be more dramatically impacted (73). This statement has been recently followed by an editorial of the European College of Neuropsychopharmacology (ECNP) President highlighting the need for more research on this field, and for further measures to mitigate the consequences of climate changes on mental health (74).

Increasing evidence indicates that climate change may affect mental health via multiple mechanisms, both direct and indirect, leading to outcomes ranging from psychological distress, fear, anxiety, or depression, to sleep disorders, substance use, and suicide (67, 70, 75). In our sample the symptomatic load, both physical and psychological, related to climate and weather variations was associated with lifetime SAs. Currently, suicide is one of the leading causes of death worldwide, and suicide rates in the U.S. have risen dramatically over the last 15 years, especially in adolescents and young adults aged 15 to 34 (76). Burke et al. described that temperature rising could increase the suicide rate by 1.4% in the U.S. and by 2.3% in Mexico, leading to an estimated number of about 21.000 additional suicides by 2050 in these countries (30). Therefore, a better understanding of the multiple causes influencing suicidal behaviors is indeed a public health priority.

Our results must be considered in the light of some issues referred to the design of the study, such as: the absence of objective measures of climatic parameters (i.e., latitude, barometric pressure, temperature, brightness, and precipitation); the reliability of a self-administered questionnaire (the METEO-Q), that may be partially biased as compared with semistructured interviews; the cross-sectional nature of the study, that does not follow individuals over time and does not help determine cause and effect. Therefore, our results in this sample may not be generalizable for other groups of subjects.

Our findings support the relevance of sensitivity to weather and climatic variations as a clinical characteristic in a large sample of BD patients. The association between sensitivity to weather changes and SAs could be useful in the assessment of suicide risk in BD patients.

Further studies that replicate our findings and investigate clinical and research implications are certainly needed.

The datasets analyzed for this study are available from the corresponding authors on reasonable request.

The studies involving human participants were reviewed and approved by the Ethics Committee of the Fondazione Policlinico Universitario “A. Gemelli” IRCCS, Università Cattolica del Sacro Cuore, Rome, Italy. The patients/participants provided their written informed consent to participate in this study.

MN and LJ were primarily responsible for study design and contributed to data interpretation and article writing. MM, GC, and PG performed data collection and contributed to data interpretation. GG and GM were involved in data entry and database management. MN performed statistical analysis. MN, MM, MP, and DDB were involved in data interpretation. IP, LM, and GG contributed to data interpretation and article writing. All authors personally revised and approved the final version of the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.