Introduction

Depression is one of the most common mental disorders in elderly people. It has a negative impact on the quality of life (Unützer et al. 2002), and the comorbidity of physical diseases increases the seriousness of the disorder (Penninx et al. 2000). Moreover, depression in the elderly has been reported to be a predictor of suicide and death (Baldwin et al. 2002), and determining its prevalence is a priority. The prevalence of depression in the elderly varies widely within a range of 1–16 % (Bland et al. 1988; Blazer and William 1980; Copeland et al. 1987; Livingston et al. 1990; Weissman et al. 1988; Smit et al. 2006; Forsell and Winblad 1999). Ideally, studies on the prevalence of depression would be conducted over fairly long periods, with participants selected from across an entire region. However, this is extremely difficult to achieve; hence, in practice, regional surveys are mostly conducted with a limited number of participants, to estimate the prevalence over short periods. Diagnosis of depression can involve written questionnaires, semi-structured interviews, and general psychiatric interviews, but use of the different methods to make the diagnosis poses problems. Compared with individuals in the West, those in Asia tend to complain more about physical symptoms than about psychiatric symptoms, and numerous studies have suggested a lower prevalence of depression in Asia (Simon et al. 2002; Hibbeln 1998; Weissman et al. 1996). It is unclear whether these differences are due to actual differences in the prevalence, differences in the research methods used, or regional differences in the concept of depression.

Successive epidemiological studies in the US have reported a strong positive correlation between high altitudes and suicide rates. Suicide rates have been reported to rise as altitudes increase, even after the data are adjusted for various factors associated with suicide risk, such as age, gender, race, income, and gun ownership, suggesting that the high elevation is a risk factor for suicide (Brenner et al. 2011; Cheng 2010; Haws et al. 2009; DelMastro et al. 2011). Because depression is closely connected with suicide, a relationship between depression and high altitude has also been suggested (Haws et al. 2009; DelMastro et al. 2011) By living at high altitudes, the resultant hypoxemia, as well as changes in atmospheric pressure, melatonin, pharmacokinetics, and metabolism could also affect a person’s behavior, mental state, and suicide risk (Maldonado et al. 2009; Arancibia et al. 2003; Jürgens et al. 2002; Schory et al. 2003). However, other sociological factors, which have not been considered in previous studies, may be involved in the relationship between high altitude and suicide rates. The literature contains no reports discussing the prevalence of depression among residents in the Himalayas or the Tibetan plateau. In this study, we surveyed depression in the elderly living in the Himalayas and the Tibetan plateau, to investigate the relationship between high altitude and depression, as well as the impact of cultural factors on differences in depressive symptoms.

Methods

Samples and Setting

The village of Domkhar, which is in the outlying areas of the Ladakh district of India, is located in the foothills of the Himalayas. It is a small village with a population of approximately 1500, and it is 3000–3800 m above sea level. Domkhar is a developing region with little electricity or running water, and many people lead old-fashioned, traditional lives. In Ladakh, the Ladakhis are predominant group, and they are all Buddhists. Yushu in the Qinghai Province of China is an ancient trading city with a population of 80,000, and it is located at an altitude of 3700 m., near a boundary with the Tibetan Autonomous Region. Tibetans constitute more than 97 % of the population.

We visited these areas in July–August 2009, and recruited 114 Ladakhis and 173 Tibetans in Yushu, Qinghai Province from among local residents who were 60 years of age or older. According to our investigation, the population of people aged 60 or over was 156 in Domkahr, and the participation rate was 73.1 %. The exact number of population in Qinghai is unknown. This study used purposive sampling in order to maximize participation instead of using random sampling. All participants received an explanation of the procedures and gave written informed consent. This project was reviewed and approved by the Ethics Committee at the Research Institute for Humanity and Nature.

The screening examination was conducted by a local physician and a Japanese neurologist. The interview concerned demographic and social data, personal and family medical history, and medical treatments. This interview was supplemented by examination of blood test, blood pressure and radial pulse, oxygen saturation, and evaluation of weight. Diabetes was determined using oral glucose tolerance test (OGTT) according to the criteria of the World Health Organization.

Mental Health

The Patient Health Questionnaire-2 (The PHQ-2) (Spitzer et al. 1999) has been reported to have a sensitivity of 100 % and a specificity of 77 %, and this tool is considered appropriate for use in an older adult population (Li et al. 2007). We scored a response as positive if a resident answered “yes” to having a depressed mood or anhedonia, regardless of the severity. This method is similar to the valid and reliable two-item PRIME MD (Whooley et al. 1997), but it maintains a 2-week frame of reference (versus 1 month), which is similar to standard diagnostic procedures. The most popular scale for late-life depression is the geriatric depression scale (GDS) (Sheikh and Yesavage 1986). The GDS is a 15-item questionnaire with a simple yes/no response format. It was purposely designed to assess the few somatic symptoms that might complicate the diagnoses in older people. In other studies of the GDS, the sensitivity and specificity ranged from 79–100 to 67–80 %, respectively, in the elderly (Watson and Pignone 2003).

A local member of the research staff administered the PHQ-2 and the GDS to the subjects enrolled in the study. In Ladakh and Qinghai, people do not usually consult doctors and are not familiar with medical vocabulary. Therefore, the interviewers had to adapt their questions to the education level of the people and had to complete a written questionnaire. Subjects who replied “yes” to either or both of the PHQ-2 items were categorized as depression-suspected. A psychiatrist blind to their screening scores then used the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR) (APA 2000) to conduct a semi-structured interview of the subjects who were suspected of having depression and to decide on a diagnosis of major depressive disorder. Interviews were conducted in English with the help of local translator.

Results

The mean age of the participants was 69.2 ± 6.7 years in Ladakh and 66.5 ± 6.1 years in Qinghai. The proportion of subjects who had a positive response to at least one item on the PHQ-2 was 7.0 % in Ladakh and 36.9 % in Qinghai, and the proportion of subjects with a GDS score of five points or more was 84.2 % in Ladakh and 66.5 % in Qinghai. The numbers of participants who had a positive response to at least one item on the PHQ-2 and who had a GDS score of five points or more were nine (7.9 %) in Ladakh and 43 (24.9 %) in Qinghai. Their characteristics are shown in Table 1.

Table 1 General characteristics of the subjects

When the psychiatrist interviewed the depression-suspected subjects in the screening examination, two people (1.8 %) in Ladakh and four people (2.3 %) in Qinghai met the diagnostic criteria for major depressive disorder according to the DSM-IV. All of them were women. None of the people in either place was receiving drug therapy with agents such as antidepressants. Bereavement over the loss of close relatives was a common cause for depression onset.

The results of each item on the GDS are shown in Table 2. The mean GDS score was 4.8 points in Ladakh and 5.4 points in Qinghai, and there were large differences in positive rates for some of the items.

Table 2 Positive rates (%) for each GDS items

Most of the depressed persons thought that the mental symptoms of depression were attributable to physical disease. A characteristic common to both Ladakh and Qinghai was that depression was seldom described as despair or dejection and was often described as “thinking too much.” None of the participants had any knowledge of depression. The depressed people showed only physical symptoms and hardly ever had mental symptoms, and many people did not think that depression was a curable disease. Life events, such as the death of a close relative or health problems, were suggested as possible causes for the onset of depression. Few of the residents in these areas satisfied the diagnostic criteria for depression, although there were many participants who experienced depressed mood when their environment changed. Below, we report on one case where environmental change may have triggered the onset of depression.

Case 1: 60-Year-Old Tibetan Female Living in Yushu

This Tibetan female was originally a nomad living in a nearby village. A series of family members became ill and passed away over the course of a few years: her mother 5 years previously, her husband and daughter 2 years previously, and her father and son 1 year previously. It became difficult to continue living as a nomad, so she moved into the house of the son who had lived in Yushu. At that time, she was living with her three grandchildren. Her daughter-in-law had left home after the death of the son to become a nun. The son had kidney failure, while he was alive and had borrowed large sums of money to pay for treatment. They sold sheep to repay half of the debt, but all of the remaining sheep died in major snowstorms 1 year previously. The only income the participant received was a small amount of financial assistance from the government for giving up nomadic life and settling in a permanent residency under the ecological migration policy. All of her income went to schooling for the three grandchildren, so they could only afford to eat tsampa (barley flour) at most meals. The debt had to be repaid by the following year, so they needed to move out of their accommodation. She had no means of repaying the debt. After her son died, she had continuous lethargy and was easily fatigued; she started experiencing insomnia a few months previously. She cried every day because of her uncertain future. The physical checkup showed she was obese and had hypertension and diabetes (height 161 cm, weight 99 kg, BMI 38.2, blood pressure 194/116 mmHg, fasting blood sugar 141 mg/dl). She was receiving pharmacotherapy for the hypertension and diabetes. She had a gloomy expression and moved sluggishly. Both her PHQ-2 scores were positive, and she scored 10 points on the GDS. She was diagnosed with depression because she had mostly lost interest in daily life; had depressed mood, insomnia, and appetite loss; was easily fatigued; and lacked the ability to concentrate.

Discussion

This study shows that a high frequency of depression is not universal at high altitudes. Previously, others have suggested that the correlation between suicide rates and high altitude might stem from biologic factors related to the effects of hypoxia. Indeed, several previous studies have demonstrated links between brief exposure to the effects of high altitude in a hypobaric chamber and vigor, fatigue, and the results of neuropsychological testing (Gerard et al. 2000; Nicolas et al. 2000). Chronic hypoxia is associated with changes in monoamine systems, including dopamine and serotonin (Arregui et al. 1994), both of which play central roles in the pathophysiology and treatment of depression (Dunlop and Nemeroff 2007). It has recently been suggested that mitochondrial dysfunction may also play a role in depression (Lucca et al. 2009). It is possible that neurotransmitter alterations might be related to changes in neuronal energetic metabolism, specifically reduction of mitochondrial respiration. However, it is not clear whether these prior studies can be generalized to people living at high altitudes in other parts of the world, and none of them demonstrated clinically significant mood changes or depression. If the mechanism underlying the depression-altitude relationship involves hypoxia, we would predict a higher prevalence of depression in the high-altitude Himalayas than in the lower-altitude United States. However, the results of this study revealed that the frequency of depression in community-dwelling elderly subjects was lower than that reported in other regions. A number of risk factors for depression have been identified, including old age, poverty, female gender, chronic physical illness or dementia, unemployment, and social isolation (Samuelsson et al. 2005; Østbye et al. 2005; Tsai et al. 2005). Many demographic and sociocultural factors are associated with depression, so any association between high altitude and depression must be considered speculative.

Assessments of the prevalence of depression are sensitive to the detailed methods used for investigation. Some investigators have reported the prevalence of depressive symptoms based on questionnaires alone, whereas others have conducted accurate surveys based on structured interviews by specialists. Differences in survey methods may be related to the differences in observed prevalence. In this study, there were large differences in the depression-positive rates between subjects assessed with the PHQ-2 and the GDS. Thus, differences in the survey method had a strong impact on the range of observed prevalence. The PHQ-2 is useful for screening, but it is thought to have a high false-positive rate. It has been suggested that it should be used as part of a comprehensive diagnostic process rather than by itself (Spitzer et al. 1999), and our findings are consistent with this suggestion.

Although high proportions of the residents in this survey, 84.2 % in Ladakh and 66.5 % in Qinghai Province, had GDS scores of five or more, the numbers of residents in both areas that were actually diagnosed with depression were low, revealing that the GDS also yielded high false-positive rates. In order to determine why the GDS values were so high, we examined the positive rate for each survey item in more detail. The results suggest that answers to some questions were influenced by the culture. For example, there were high rates of positive answers to the questions “Do you often get bored?” and “Do you prefer to stay at home, rather than going out and doing things?” on the GDS in Ladakh. This may have been because most of the day in Ladakh is spent farming and in prayer, and there are not many opportunities to go out. The GDS was originally prepared in English. Even though some reports have shown that these tests are useful in a variety of languages, neither the PHQ-2 nor the GDS have been validated among Ladakhis or Tibetans. When depression-screening tools have been used in non–English-speaking areas, the results have been affected by cultural factors (Beekman et al. 1999). Thus, caution is necessary when using these tools alone in surveying the prevalence of depression in Ladakhis or Tibetans.

Another potential explanation for the low prevalence of depression is that the social or cultural backgrounds of these populations may suppress the development of depression. Social support and social interaction are also known to be higher in rural areas, with denser networks that are often kin-based and associated with religious institutions. We observed that depressive elderly received care not only from their family but also from the whole community. Ladakhis and Tibetans have very strong family ties, and a strong interpersonal network may prevent the development of depression. It has often been said that ethnicity influences the association between depressive symptoms and the way in which depressive behavior is expressed (Blazer et al. 1998).

The subjects of this study were all devout Tibetan Buddhists. When asked, “What are the times when you feel happy?” the majority replied, “When I am praying.” Ladakhis and Tibetans believe in karma. According to the laws of karma, both happiness and misery in this world are the results of previous lives. It is possible that such strong beliefs inhibit the development of depression. We previously reported that PTSD and depression after natural disasters are less common in the Tibetan cultural areas than in other areas (Ishikawa et al. 2013). Our finding seems to indicate that social or cultural background factors might play important roles in suppressing the development of depression.

This study has several limitations. First, it is too limited in sample size and in recruitment methods to be representative. Second, the PHQ-2 and the GDS have not been validated in a Ladakhi or Tibetan context. Third, we did not examine the influence of risk factors for depression, such as chronic physical disorders, family structure, poverty, and activities of daily living. Fourth, considering that cultural factors play an important role in the development of depression, a test of the effect of hypoxia would require a comparison of depression prevalence between these samples and comparable lowland Ladakhi or Tibetan populations. Finally, language translation is always an issue in the utilization of the instruments in cross-cultural settings.

Despite these limitations, we used several instruments in the present study, and the interviews were conducted by a professional, who was accompanied by a translator. Our finding that only females were diagnosed with depression and that most of them had experienced bereavement, is consistent with findings revealed in studies globally. This supports the accuracy of the diagnosis. Further, having two independent highland samples from the opposite edges of the Tibetan plateau also strengthens our findings.

This study has highlighted the possibility that strong religious beliefs and strong interpersonal networks in the community might prevent the onset of depression in the elderly. The participants do not necessarily trust modern medicine and many prefer to rely on traditional medicine practiced by shamans and monks. In the course of our research, we asked individuals with depression to describe their life stories, including where they were born, how they were brought up, how many siblings they had, what type of life they led, when they became sick, and how they managed illness. We did not presume that modern medicine had all the answers, and tried to take a step back and gain an accurate and objective understanding of the culture and values of the patient. We emphasized on the medical anthropology perspective of fully reflecting the many value systems possible. We believe that these findings will aid in providing clinical care to the people of Ladakh and Qinghai, and help understand the hardiness and resilience that characterizes this under-studied population, which might have important implications for the prevention of depression. Such knowledge would also lead to new prevention and intervention strategies. Indeed, further studies will be required to establish the relationship between depression and altitude.

Conclusion

A few of the elderly residents of high-altitude communities in Ladakh and Qinghai had depression, despite the harsh environment at high elevations. Our finding seems to indicate that cultural factors such as religious outlook and social/family relationship inhibit the development of depression.