Prevalence of psychiatric morbidity among urban elderlies: Lucknow elderly study

Abstract

Background:

Paucity of systematic studies in elderly mental health in an aging population is an urgent need, which is required to address services and planning issues for health.

Aim:

The present study aims to investigate the distribution of physical, neuropsychiatric, and cognitive disorders of a community sample of elderlies with certain socioeconomic data.

Materials and Methods:

A door-to-door household survey was conducted to identify houses with elderlies (≥55 years) in two urban localities of Lucknow. Mini mental state examination (MMSE), Survey Psychiatric Assessment Schedule (SPAS)/Mood Disorder Questionnaire (MDQ) and physical and neurological examination were used for screening all consenting elderlies. MMSE positive participants were assessed on Cambridge Mental Disorders of the Elderly Examination-Revised for diagnosis of cognitive disorders; SPAS/MDQ positives were assessed on Schedule for Clinical Assessment in Neuropsychiatry based clinical interview for diagnosis of neuropsychiatric disorders other than cognitive disorders (using ICD-10 criteria). Routine and indicated laboratory/radiological investigations on all and on MMSE/SPAS (organic section) positive/physically ill participants respectively were done to confirm organic and/or physical illness. Only percentages were calculated to find the distribution of morbidity.

Results:

The sample had proportionate age structure as that of the surveyed population and had balanced gender representation in each age deciles. Prevalence of neuropsychiatric disorders (with/without comorbidities) was 11.8% in the elderlies (60 years and above) highest being in the 60-69 years age group. Being women and of lower socioeconomic status was more commonly associated with a neuropsychiatric diagnosis. 7.6% of the elderlies had cognitive impairment. Overall findings suggest a prevalence rate of 17.34% of total psychiatric morbidity among elderlies. A significant number had comorbid physical illness diagnoses.

Conclusion:

More than half the elderlies had some diagnosable physical or mental ailment. The study familiarizes us to the significant amount of physical and psychiatric comorbidity in the particular age group. About one-fifth was found to suffer from psychiatric morbidity, which any health services for the elderly should be oriented towards.

INTRODUCTION

An estimate of the world mental health report states that the number of elderly in the world would reach more than 1 billion by the year 2025, of which the proportion of India's geriatric population is expected to reach 18.4%.[1] The population of older adults aged 60 years and above increased from about 20 million in 1951[2] to 77 million in 2001[3] to 83.6 million in 2006, and is expected to increase to 173 million in 2026.[4] Thus, a doubling of this population is anticipated between 2006 and 2026. This expectantly will add to the already existing enormous mental health service demands in this segment of the population.[5] A recent review reported a wide range of estimates for mental health morbidities in the elderly, ranging from 2.2% to 33.3% for age specific populations. In addition, it is also reported that mental health morbidity is seldom an isolated event in the elderly and a minimum of two to three other clinical diagnoses are a rule.[6]

Mental health surveys have reported variable psychiatric morbidity in the elderly viz., 2.23%;[7] 33.3%;[8] 35%;[9] 8.9%;[10] 17.39%;[11] 8.1% (urban) and 4.9% (rural);[12] 42.0%,[13] and 23.6% (rural and urban).[14] A study from Northern India in the last decade that targeted only rural elderly population, conducted by the lead author of the present study found psychiatric morbidity to be 43.32%.[15] Recently, published an epidemiological study sponsored by Indian Council of Medical Research (ICMR) conducted in a rural area of Lucknow found the prevalence of psychiatric morbidity to be 23.7%.[16]

To summarize epidemiological studies have found prevalence ranging from 2.2% to 33.3% in the general population surveys;[7,8,11,12] however, prevalence has been found to be higher among the elderly population ranging from 8.9% to 43.3%[9,10,13,14,15,16] in our country.

The discrepancy commonly found in the reported prevalence of mental health morbidity in general as well as in the elderly population is due to the differing methodologies viz., the choice of criteria for caseness, choice of sample (general vis-à-vis elderly) and locations (urban vis-à-vis rural), etc., To summarize, there is a paucity of systematically conducted epidemiological studies on mental health problems among the elderly drawn from representative cross-section of the community. Therefore for a realistic and need based planning of mental health care services for the elderly, an assessment of prevalence of psychiatric disorders on the elderly population living in the community is required. In addition, the high prevalence of comorbidity makes it imperative to investigate into the associated problems of physical ailment.

The courtesy and kind gesture of ICMR, New Delhi made it possible to plan and undertake a large scale, well-designed and comprehensive community-based epidemiological study on elderly from urban and rural communities.[16,17] The study was conducted with the objective of screening elderly people (55 years and above) to find the prevalence of neuropsychiatric disorders among them with special reference to cognitive disorders. The intention was to classify the studied population into following categories: (1) Normal aging, (2) organic including symptomatic mental disorders as per ICD-10[18] criteria, (3) other neuropsychiatric disorders as per ICD-10 criteria, and the (4) physically ill group. The studies are known as “Lucknow Elderly Study.” The current paper reports, the prevalence of psychiatric and physical morbidity among urban elderly only.

MATERIALS AND METHODS

Study area

The study area consists of an old and a new locality of urban Lucknow selected purposely. Different wards of these localities were listed of which two wards (Musahebganj ward of old Lucknow locality and Jankipuram ward of new Lucknow locality) were selected randomly.

Characteristics of study universe

Population characteristics of the study universe are given in Table 1. Data relates to the two wards selected for the purpose of the study.

Table 1.

Characteristics of the study universe in Musahebganj (M), Jankipuram (J) and the total study area (M+J)

Of the 13,000 households present in the study area, 88.9% households were screened. The population that was surveyed from these households was 33,746. Of this population, 5.80% were of age 60 years and above (of which young-old were the largest group).

Study sample

Table 2 gives age-wise break-up of the sample of population included in the study and informs about the number of elderlies excluded due to uncooperativeness, lack of consent, etc.

Table 2.

Age wise break-up of population included for the study in Musahebganj (M), Jankipuram (J) and total study area (M+J)

Table 2 shows that in the elderly population of the study area, a total 2283 elderlies consented and completed the detailed assessment. Elderlies belonging to young-old (60-69 years) age group were significantly more than “old-old” and “oldest-old” age group in the study sample.

Tools for investigation

Hindi translated versions of the standardized tools were used in the study. The tools consisted of:

• Household screening form for identification of houses where individuals aged 55 years and above were permanently residing

• Assessment schedules for included families

  1. Semi structured proforma for identification of bio-sociocultural information and other relevant family details (sociodemographic proforma, i.e., SDP)
  2. Socioeconomic status (SES) scale for assessment of the SES of the family.[19]

• Screening schedules for included individuals aged 55 years and above in the family

  1. Informed consent form for index subjects and their caregivers
  2. Mini mental state examination (MMSE):[20] To control education bias in the screening of cognitive status, education, and age based cut-off criteria were used[21] and is given in Table 3
  3. Survey Psychiatric Assessment Schedule (SPAS)[22] /Mood Disorder Questionnaire (MDQ):[23] SPAS consists of 51 items divided into three sections viz., organic disorders, affective disorders/psychoneurosis and schizophrenia/paranoid disorders. For identification of “cases” each section of SPAS is scored independently.
     Section 1: A simple additive score of the twelve responses classifies the subjects as suspects in different categories. The following cut points are used:
     • No organic disorder: 9-12
     • Mild organic disorder: 7-8
     • Severe organic disorder: 0-6.
     Section 2: The 44 responses are summed and the following cut points are used to identify “suspected cases” and “noncases:”
     • Noncase: 0-10
     • Case: 11-65.
     Section 3: In this section, any positive answer indicated possible case.
     However, in this study, SPAS was used as the screening instrument to identify suspects for neuropsychiatric disorders. Subjects found positive on one or more than one sections using above cut-off score and/or MDQ positive subjects (cut-off = 7) were considered as SPAS positive. MDQ was used as a safeguard against “false negative” classification of possible mood disorder cases on SPAS.
  4. Physical and neurological examination (PNE)
  5. Routine and specific laboratory/radiological investigations
  6. Problem Checklist and Strain Scale (PCLS).[24]

• Assessment schedules for individuals positive on MMSE, SPAS/MDQ:

  1. Schedule for Clinical Assessment in Neuropsychiatry (SCAN)[25] (Hindi version)
  2. Cambridge Mental Disorders of the Elderly Examination-Revised (CAMDEX-R)[26]
  3. Functional Assessment Staging Technique (FAST)[27]
  4. Global Deterioration Scale (GDS)[28]
  5. Brief Cognitive Rating Scale (BCRS).[29]

Table 3.

Age and education specific cut-off criteria for the MMSE

Procedure of translation of the tools into Hindi

Three translators, well-versed in English and Hindi, translated the original English versions of MMSE,[20] SPAS[22] /MDQ,[23] CAMDEX-R,[26] BCRS,[29] FAST,[27] GDS,[28] and PCLS[24] into Hindi independently. The translators then discussed and compared the translation item by item to agree upon a prefinal translated Hindi version (PFHV) of the tools. These versions were administered to ten literate and ten illiterate persons aged 60 years and above, drawn from another community to know the comprehensibility of the items. These people were also asked, whether the items are clear, culture fair, simple, and comprehensive. Most of the items of PFHV were found to be comprehensible to literate as well as illiterate persons in respect to the nature and content of the items. Some items that were not found to be culture and education fair, modifications were made according to the suggestions, given by the participants who had taken the test. During this process, the originality of the assessed domains was conserved as much as possible. Therefore, these versions were taken as final translated Hindi versions (FHV) of the tools. Two bilingual experts back translated the FHV of the tools into English to establish meaning equivalence. The original English, the final Hindi versions of the tools and the back translated English versions of the final Hindi versions of the tools were referred to three bilingual mental health professionals to assess logical validity of the instruments. These mental health professionals unanimously agreed that the three instruments of each tests had very high balancing meaning and lingual equivalence. These final Hindi versions of the tools were administered in the study.

Study procedures

Surveyors visited and numbered all the houses in selected localities starting from one corner of the locality. Information on age structure of family members residing permanently (nonfloating) in these houses was obtained on household screening form. Houses where at least one family member aged 55 years and above was residing were identified and listed. Adequate precautions were taken in ascertaining the age of the subjects employing documents/method/anecdotal events, etc., like:

• Date of birth or age mentioned in ration card, high school certificate, driving license, identity card, passport, etc

• Ascertaining age by other sources such as age at the time of significant historical events like independence, floods in Lucknow in 1960, etc

• Information given by the family members, neighborers.

• Children's age or presence of grand and great grandchildren

• Physical appearances, skin, teeth, cataract, and arcus senilis

• Biological milestones like menopause in women

• Age and date at marriage, duration of marriage

• Retirement date.

Family details of included families were obtained on SDP and SES assessment was done on SES scale. Informed consent of included subjects as well as their caregivers was obtained on proper proforma. The caregivers of all included elderlies were assessed on PCLS. Included subjects were screened on MMSE, SPAS/MDQ, and PNE to identify suspected subjects for probable cognitive, neuropsychiatric or medical problems. MMSE positive subjects were assessed in detail on CAMDEX-R for diagnosis of cognitive disorders whereas, SCAN based clinical interview was conducted on SPAS/MDQ positive subjects for diagnosis of neuropsychiatric disorders other than cognitive disorders using ICD-10 criteria. MMSE and SPAS/MDQ positive subjects were also administered FAST, BCRS, and GDS for the assessment of functional, cognitive, and global performance of the subjects, respectively. Routine pathological investigations on all subjects and special laboratory/radiological investigations on MMSE/SPAS (organic section) positive/physically ill subjects were done to confirm organic and/or physical diagnoses based on PNE, available prescription, laboratory, and radiological investigations and other records. Diagnostic formulation was done on the basis of all available information. ICD-10 criteria were followed in assigning the diagnostic label.

On the basis of all the above information, the subjects were finally categorized into following groups:

• Group A: Normal group (subjects without discernable abnormality on physical, neuropsychiatric or cognitive status)

• Group B: Neuropsychiatric group (subjects having diagnosable neuropsychiatric disorders only other than cognitive disorders)

• Group C: Cognitive disorder group (subjects having diagnosable organic disorders only)

• Group D: Physically ill group (subjects having diagnosable physical illnesses only)

• Group E: Neuropsychiatric group with comorbid cognitive disorders

• Group F: Neuropsychiatric group with comorbid physical illnesses

• Group G: Cognitive disorder with comorbid physical illnesses

• Group H: Neuropsychiatric group with comorbid cognitive disorders as well as physical illnesses.

RESULTS

A total of 2462 elderlies (1274 from Musahebganj and 1188 from Jankipuram area) drawn from 6622 households (3841 from Musahebganj and 2781 from Jankipuram area) were recruited in the study. In spite of best possible efforts, 179 subjects (58 from Musahebganj and 121 from Jankipuram area) had to be subsequently excluded from the study sample after inclusion. Administration of the battery of tools except household screening form was not possible on excluded elderlies for various reasons. 179 (7.8%) initially included subjects were dropped for reasons viz., refusal, avoidance, wrong inclusion, due to misinterpretation of age, etc. Among the dropped subjects, maximum (45.8%) refused to participate in the study later after inclusion.

A total of 23.5% subjects did not refuse directly, but they willfully avoided meeting with the research team for assessment despite several visits (at least five). Efforts to establish contact with such elderlies were subsequently abandoned. These subjects were more in new Lucknow (10.18%) than old Lucknow.

Table 4 shows age, sex, and SES wise distribution of assessed study probands (N = 2283) from both the areas of Musahebganj and Jankipuram.

Table 4.

Age, sex and socioeconomic status wise distribution of assessed study probands (N=2283)

Out of a total 2283 elderlies included in the study, 447 (19.6%) belonged to “preelderly” age group, 1263 (55.3%) to “young-old” age group, 398 (17.4%) to “old-old” age group and 175 (7.7%) to “oldest-old” age group. Maximum percentage of elderlies (58.1%) belonged to middle SES followed by lower (28.56%) and upper SES (13.4%). There was balanced gender representation at all levels of SES.

Table 5 shows out of total 2283 subjects, 447 subjects (19.58%) belonged to “preelderly” age group and 1836 subjects (80.42%) to “elderly” age group. Among “preelderly” age group, 47.0% subjects were normal while 37.3% in “elderly” age group were normal. Maximum subjects in the “preelderly” age group were found to be suffering from medical illness (41.8% and an additional 7.8% with psychiatric comorbidity). Out of the total of 1836 subjects aged 60 years and above, 37.3% were normal. Maximum elderly were suffering from physical illness (45.4% and an additional 12% with psychiatric comorbidity) followed by neuropsychiatric disorder with comorbid physical illness (6.6%). Out of total 1263 subjects in “young-old” age group, 38.1% were normal. Among “young-old” group, maximum (45.7%) subjects were suffering from physical illness, followed by 6.7% subjects suffering from neuropsychiatric disorder with comorbid physical illness. Out of the total of 398 subjects in “old-old” group, 36.4% subjects were normal. 43.7% “old-old” subjects were suffering from physical illness followed by 7.3% physical illness with comorbid cognitive disorders/impairment. Out of the total of 175 “oldest-old” age group, 33.1% were normal. Maximum (47.4%) oldest-olds were found to be suffering from physical illness followed by 6.3% cognitive disorders/impairment with physical illness.

Table 5.

Health status of assessed probands by various age group categories

Prevalence of neuropsychiatric disorder with physical illness was almost equal in “old-old” and “young-old” (6.8% and 6.7% respectively) closely followed by 6.3% in “preelderly” and 5.7% in “oldest-old” age group. However, 3.4% prevalence of multiple morbidity (physical, neuropsychiatric, and cognitive disorders) was highest among “oldest-olds” than other age groups.

Table 6 shows age wise prevalence of neuropsychiatric disorders, which shows the maximum prevalence among the “young-old” (60-69 years) (12.2%) followed by the “old-old” (10.8%), “oldest-old” (10.8%) and “preelderlies” (9.4%). The findings suggest that the prevalence of neuropsychiatric disorder is lesser in “elderlies” aged 70 years and above in comparison to the “young-old.”

Table 6.

Prevalence of neuropsychiatric disorder (B + E + F + H) by age groups

Table 7 shows higher prevalence of neuropsychiatric disorders other than cognitive disorders among “elderlies” aged 60 years and above (11.8%) than “preelderlies” (9.4%). In addition, the overall prevalence of neuropsychiatric disorder was more in preelderly and elderly females than males (11.1% and 6.8%, respectively and 13.6% and 10.1%, respectively). In both age groups, the prevalence of neuropsychiatric disorders was the maximum in the sample belonging to lower SES.

Table 7.

Prevalence of neuropsychiatric disorders (B + E + F + H) overall and by sex and SES among preelderly and elderly age groups

Table 8 below shows overall prevalence of cognitive disorders (C + E + G + H) among elderly in the community.

Table 8.

Overall prevalence of cognitive disorders (C + E + G + H) among elderly in the community

Table 8 above shows overall prevalence of cognitive disorders among “preelderly” and “elderly” age groups. Prevalence of cognitive disorders/impairment among “elderly” age group (60 years and above) is 3.8 times more than the prevalence among “preelderly” age group.

DISCUSSION

The aim of this paper was to present the psychiatric and physical morbidity among urban elderly based on “Lucknow urban elderly study.” There was balanced representation of either gender in each age deciles. The age structure of the sample was proportionate to that of the surveyed population.

The strength of the study is in its methodology in terms of allowing evaluation of physical illness comorbidity along with total psychiatric morbidity, which in this study has been found to be high. The salience of this finding can be speculated to be that physical ailments and psychiatric morbidity can function as an etiological determinant of either condition or that they may be an outcome of a common etiological pathway.

Another strength of the present study was that it paid keen attention to evaluate the SES of the participating sample. Although, the lower SES group was the second largest group in the sample, it was the largest group with participants having neuropsychiatric disorders. This finding denotes underlying pathological effects of poverty afflicting the elderly in terms of mental health morbidity. In a similar manner, female gender too was found to be associated with more (physical and) mental health morbidity in the elderly (female to male ratio was 1.35). Though many previous population-based studies have noted the higher prevalence of psychiatric morbidity in women and in lower SES[30,31,32,33] few have tried to replicate this finding in the elderly population. Gender as a determinant of inequality in mental health achievement might influence such an outcome in an urban setting due to the unique gender roles of the elderly in such setting. Future studies might help to find this influence interacting with other social characteristics (e.g. poverty).

Another important finding was the increased prevalence of neuropsychiatric disorders in the “young-old” (60-69 years) compared with the older age deciles. This can be assumed to be partly contributed by the demographic distribution of the population, attributed to the mortality patterns. Assuming that healthier elderlies are self-selected into the subsequent age deciles, such a pattern of neuropsychiatric morbidity is understandable.

On the whole preeldrlies were healthier than the elderly age group, suggesting deterioration of physical and mental health status with advancing age. The overall prevalence of total psychiatric morbidity among elderlies aged 60 years and above was found to be 17.34% (it includes Groups B [3.1%], C [1.8%], E [0.44%], F [6.6%], G [3.8%], and H [1.6%]).

Certain limitations of the study need to be mentioned which could have enhanced the quality of the study manifolds. It did not try to investigate the pattern of individual physical illness that commonly coexists with neuropsychiatric disorders and cognitive impairment. Since, it was a cross-sectional study the temporality of the evolution of each disorder/s was not investigated which could inform about the etiological pattern and the stability of the diagnosis. Limited data relating to social and behavioral factor allowed inadequate analysis of factors determining distribution patterns of illness found in the study.

CONCLUSIONS

This study is one of the rare ones that have tried to investigate comorbidity of physical and mental ailments of a community sample of elderly. All screening positive participants were evaluated in detail with reliable tools to identify and pinpoint diagnosis. The study found more than half the elderlies to have some diagnosable physical or mental ailment. The study familiarized us to the significant amount of physical and psychiatric comorbidity in the particular age group. About one-fifth was found to suffer from psychiatric morbidity. Certain social factors were also identified to be associated with elderly neuropsychiatric problems. Thus, any health services for the elderly should be oriented toward these facts while planning, budgeting, and training personnel.