Some community surveys have specifically looked for depression among elderly population (Rajkumar et al. 2009; Barua and Kar 2010; Deshpande et al. 2011; Reddy et al. 2012). These are summarized in Table 2. Between 21.7 and 47 % of elderly screened positive for depression, while a more definitive diagnosis of depression was found in 12.7–19 % of elderly.

Depression is also common among elderly users visiting the general health settings and those residing in old age homes (Dey et al. 2001; Tiwari et al. 2012 Ganguli et al. 1999). For use in illiterate elderly, Ganguli et al. (1999) developed and validated a Hindi version of the geriatric depression scale (GDS-H).

The prevalence rates of depression in elderly are considerably higher than the general population rates in India (very low to 15 %) (Reddy and Chandrashekhar 1998; Poongothai et al. 2009). Further, the prevalence is also higher than the elderly residing in other countries of Asia, America and Europe. A meta-analysis of 74 community-based surveys of elderly population across the world (including 6 Indian surveys) found that the proportion of depressed elderly in India was higher compared to rest of the world (Barua et al. 2010). Further, there appears to be an upward trend in prevalence of elderly depression in India.

Female gender and an advanced age are two non-modifiable risk factors for elderly depression. Several other factors have been found to increase the likelihood of depression, which include illiteracy, financial difficulties, living alone, low social/emotional support, widowhood, economic or physical dependency and presence of chronic physical illnesses or disability of any kind (Barua et al. 2010).

Studies also found that life events, including a perceived family crisis, are an important precipitating event for depression (Agrawal and Jhingan 2002). Venkoba Rao et al. observed that the stressors were related to ‘bereavement’, ‘occupation’ and ‘family and social relationships’ in the 2 years preceding onset. Living in joint families does not, by itself, guarantee a good social integration, as old ones in family can be ‘lonely islands’ (Venkoba Rao 1981). Having a good social support and having more confidants emerged as protective factors for elderly depression.

Not much work is available on biological aspects of elderly depression. A recent study reported that persons with ApoE4 allele have 4.7 times more risk of developing depression in old age (Sureshkumar et al. 2012).

Some phenomenological differences have been observed in elderly depression, which include an increased agitation, restlessness rather than retardation, rarity of ideas of sin and guilt and a higher somatic presentation. The commonest symptoms of depression in elderly using the 40-item WHO assessment schedule were sadness (93.3 %), somatic symptoms (71 %), suicidal ideas (66.6 %), and among others (Venkoba Rao 1981, 1983). The other manifestations of adult depression, viz. depressed mood, lack of energy and fatigue, decline in work and interest and anorexia have poor discriminatory power for the diagnosis of elderly depression (Khandelwal 1995, 2001). The depressed elderly are more likely to have undiagnosed physical disorders and multiple illnesses compared to controls (Sagar et al. 1992). The use of a substance or abuse of a prescription medication should be asked specifically from elderly patients. The medications used for physical conditions need an enquiry for their potential to cause or aggravate depression (Sagar et al. 1990).

The course and outcome of depression remain under-researched in Indian elderly. In a one-year follow-up of 50 elderly patients with depression, 28 % had recovered, 30 % partially recovered, 23 % had intermittent relapses, 6 % had been continuously ill, 11 % had died, and 6 % had dementia (Jhingan et al. 2001). Factors predicting a good outcome were shorter duration of episode and those living in joint families.

Mania appears to be uncommon in elderly surveys. In a recent descriptive study of 30 elderly with mania, 50 % had delusions and/or hallucinations (Prakash et al. 2009). The psychotic and cognitive symptoms were higher in elderly mania compared to depression. A significant proportion had neurological disorders (10 %), substance abuse (53.3 %) and medical comorbidity (66.6 %). Venkoba Rao (1983), in his series of elderly patients with affective disorders, found that depression occurred thrice more common after 60 years than mania (3:1) (Venkoba Rao 1983). Few case reports are available which have described an unusually late age of onset or an association with hyperthyroidism (Aggarwal et al. 2010; Nath and Sagar 2001). A thorough examination and investigations are required in late-onset cases of mania to rule out secondary causes. Safety profile appeared to be main consideration in choosing a medication.

Overall, mania in elderly is an uncommon clinical presentation and needs more research attention.

Anxiety disorders were common in community surveys of elderly, point prevalence ranging between 6.4 and 10.4 % (refer Table 1). The clinic-based studies have reported neurotic disorders in 5–19.8 % of elderly, a significant proportion of which are likely to be anxiety disorders (Prasad et al. 1996; Singh et al. 2004; Prakash et al. 2007). Not much original research from India is available on other aspects of anxiety disorders in the elderly.

The data derived from various sources e.g. Government of India statistics, suicide prevention centres, suicide autopsies, geropsychiatric clinics and surveys indicate an upward trend in suicide among the elderly. The average annual suicide rate among elderly (≥55 years) is 189 per 100,000 in a study from South India, which is considerably higher than the national average. The male:female ratio was 1.5:1. Further, the age-specific suicide rates increased with advanced age (Abraham et al. 2005).

Among the autopsy cases of elder suicide over a 10-year period (n = 43) at a tertiary care centre, the age range was 60–87 years, commonest being 60–69 years age group. Males were over-represented (60 %) in this sample. Hanging, followed by poisoning, appear to be common methods of suicide by elderly (Abraham et al. 2005; Behera et al. 2007). Depressive illness is the leading cause for suicide, and nearly two-thirds of depressed elderly had suicide ideation (Venkoba Rao and Madhavan 1983). Lack of family and social integration is more important risk factor rather than ‘living alone’. Certain ethical, religious and familial factors [described as suicide counters by Venkoba Rao (1985)] may hold back the person from attempting suicide. Suicides in older people are often associated with high intent, long planning and involve highly lethal methods. Further, elderly are medically fragile and may live alone, thereby increase the probability of a fatal outcome (Suicidal Behaviour in Special Population 2007).

Very few Indian studies have described psychotic disorders among the elderly population. Venkoba Rao reported a series of patients with late paraphrenia (onset after 60 years), which formed about 4 % of geropsychiatric clinic patients. Most had accompanying visual or hearing impairment (Venkoba Rao and Madhavan 1981). The late-onset cases (defined as onset ≥45 years) formed nearly 2 % of all schizophrenia patients presenting to a tertiary care centre in North India over a two-year period. Further, female preponderance was seen in the late-onset cases, with female to male ratio being 1.67:1 among those with onset after 40 years and 10: 1 among those with an onset after 45 years of age (Jayaswal et al. 1987; Jayaswal et al. 1988; Adityanjee et al. 1989; Khandelwal 1999). Another study described the profile of late-onset schizophrenia (mean age: 53.5 years), majority being treatment-naive. Delusion of persecution was universal, followed by delusions of reference in 85 % and, less commonly, hallucinations (Bhuyan et al. 2009; Kulhara et al. 1999; Shaji et al. 2009). Follow-up study of late paraphrenia series revealed a sustained remission for 4–5 years in 75 % of patients with treatment (Khandelwal 1999). There is a need for more hospital-based studies for late-onset psychosis.

Studies on elderly substance use are relatively infrequent. Among rural elderly, the overall prevalence of smoking is 51–71.8 % among males and 5–41 % among females (Goswami et al. 2005; Jotheeswaran et al. 2010; Shaji 2002). In a large-scale population study, the current use of alcohol varied between 10 and 28 % among 50 years plus age groups, majority being current heavy users (Mohan et al. 2002; Gupta et al. 2003). In a clinic-based review, alcohol (60 %) and opioid (35 %) were commonly misused substances (Grover et al. 2005). Individuals over 60 years should be routinely screened for substance use, especially in the presence of depressive symptoms.

In the oldest-old group, substance use is likely to decline. Isolated reports have described continued use of substances e.g. poppy tea dependence in octogenarian age group (Subodh et al. 2010).

Only a few Indian studies have separately focused on delirium occurring in the geriatric patients (Grover et al. 2009; Pinto 2010). Available studies on delirium in elderly had sample sizes ranging between 100 and 400, from hospital settings. Nearly two-thirds had hospital-emergent delirium (Grover et al. 2012). Hypoactive delirium was relatively common than other forms. Most patients had sleep–wake cycle disturbance, disturbance in orientation, attention and short-term memory impairments, fluctuation of symptoms, temporal onset of symptoms and a physical disorder. Most common aetiologies were sepsis followed by metabolic. The mortality rate was nearly 15–16.5 % (Grover et al. 2012; Chrispal et al. 2010).

There are no specific Indian studies on treatment aspects or long-term course and outcome of elderly post-discharge.

Nearly two-thirds of the world’s elderly with dementia reside in developing countries (Alzheimer’s & Related Disorders Society of India 2010). Several community surveys from India, with samples ranging from 740 to 24,000, have found the prevalence to range between 0.8 and 4.9 % among those aged 65 years and above (Shaji et al. 1996; Rajkumar et al. 1997; Chandra et al. 1998; Vas et al. 2001; Shaji et al. 2005; Mathuranath et al. 2009; Saldanha et al. 2010). Concerns have been raised that milder cases may be missed out, pointing to a need to operationalize the criteria to diagnose dementia in community (Prince et al. 2003). Using the 10/66 dementia research group algorithm, a higher prevalence rate for dementia was found (urban Chennai: 7.5 % and rural Vellore: 10.6 %) (Llibre Rodriguez et al. 2008). Vellore screening instrument was developed after employing the ‘activities of daily living’, which are not influenced by education and culture and can reduce the false-positive rates in community studies (Stanley et al. 2009).

Few studies are available on the incidence of dementia in Indian settings. It varies between 1.74 and 11.67 per 1,000 person years, which is lower than the Western figures (Chandra et al. 2001; Mathuranath et al. 2009; Raina et al. 2009).

Alzheimer’s disease appears to be more common than vascular dementia in some studies, though results are conflicting in Indian studies. The risk factors found in Indian studies include an advanced age, female gender, low socio-economic status, low education, family history, urban living, lack of exercise, depression, presence of diabetes, hypertension and hyperlipidemia. Living in joint families, being married and increased intake of polyunsaturated fats were protective factors (Shaji et al. 1996; Rajkumar et al. 1997; Mathuranath et al. 2009; Saldanha et al. 2010).

The behavioural and psychological symptoms of dementia (BPSD) were described by Khandelwal et al. (1992) in one of the earliest studies from India. The delusion that ‘people are stealing thing’ was most common (27 %) followed by believing that ‘one’s house is not one’s own’ (10 %). Depressive and anxiety, features, agitation, wandering and paranoid ideation were observed to be common features in patients with dementias (Khandelwal et al. 1992; 10/66 dementia research group 2004). BPSD appear to be more common in Alzheimer’s disease compared to vascular dementia and have been associated with an increase in caregiver burden (Shaji et al. 2009).

Presence of dementia in elderly is associated with twice the risk of mortality after adjusting for age and gender. The mortality risk is linearly correlated to the severity of cognitive impairment (Jotheeswaran et al. 2010).

Given the huge deficits of mental health resources in country, there is a need to develop locally based, low-cost models of dementia identification, care and support. The local community health workers in rural Kerala could identify the possible dementia cases in community after being trained for 2.5 h only (Shaji 2002). Another study from Goa described a randomized trial of a low-cost home-based intervention, which was feasible, acceptable and led to significant improvements in caregiver mental health and burden (Dias et al. 2008).