Changes in activities of daily living during treatment of late-life depression*
A portion of this paper was presented at the 21th Annual Meeting of the Japanese Psychogeriatric Society.
Dr Haruna Saito MD, Department of Psychiatry, Shinshu University School of Medicine, 3-1-1 Asahi, Matsumoto, Nagano 390-8621, Japan. Email: email@example.com
Background: Patients with late-life depression often exhibit cognitive deficits or disability. The relationship between late-life depression and disability is an area of significance for geriatric psychiatry. The purpose of the present study was to investigate the association between changes in the severity of depression and activities of daily living (ADL) during treatment of late-life depression.
Methods: We examined the severity of depression, cognitive function and ADL among 70 subjects aged 65 years or older with a history or a current episode of major depression. Examination with the Hamilton Depression Rating Scale (Ham-D), the Mini-Mental State Examination (MMSE), the Barthel Index (BI) and the Hyogo Activities of Daily Living Scale (HADLS) were performed twice at an interval of approximately 1 month. All patients were receiving drug therapy for their depression.
Results: Participants presented with varied depressive severity on the Ham-D and a slight cognitive deficit on the MMSE; approximately 60% had at least one problem on the HADLS. Between baseline and the follow-up examination, significant improvements were noted in Ham-D, MMSE and HADLS scores, but not the BI score. The correlations between the Ham-D and ADL scores were significant at baseline and at the follow up examinations. In a multiple regression analysis, the difference in follow-up and baseline scores on the Ham-D contributed significantly to that of HADLS.
Conclusions: The present study demonstrates that many patients with late-life depression have some problems with ADL and that a substantial improvement in ADL occurs in association with temporal improvement of depressive severity among subjects with late-life depression.
Depression is common in the elderly population and is one of the leading causes of disability. Depression can worsen the outcome of other medical illnesses and even increase mortality.1–3 The comorbid status of late-life depression has been investigated and a significant percentage of patients with late-life depression also have cognitive deficits, multiple medical burdens or disability,4,5 defined as problems that restrict an individual's ability to engage in one or more major life activities. Disability is considered a distinct dimension of health status.6 A strong association between the severity of depression and disability has been repeatedly reported in patients with late-life depression, even when controlling for possible confounding variables, such as age, medical burden, social support and cognitive status.7 Disability is typically categorized as either impairment in self-maintenance or basic activities of daily living (BADL) or the impairment of instrumental ADL (IADL), which comprises complex activities. Previous studies that focused on the relationship between depression and disability indicated that both BADL and IADL were often impaired in subjects with late-life depression8,9 and that the impairment of IADL was more frequent and strongly associated with specific symptoms or the severity of depression.6
In the present study, we assessed the temporal association between the severity of depression and impaired ADL in patients with late-life depression, examining the severity of depression, disability (including IADL) and cognitive status during a short interval of treatment.
The present study was conducted from August 2003 to October 2005 in the Department of Psychiatry at Shinshu University Hospital (Matsumoto, Japan). The study population included outpatients and inpatients aged 65 years or older with a history or presence of major depression according to the diagnostic criteria of DSM-IV.10 To prevent the inclusion of subjects with significant physical disabilities or dementia, subjects were excluded based on the following criteria: (i) history of another psychiatric disorder before the onset of depression; (ii) severe or acute medical illness, such as metastatic cancer, brain tumor, myocardial infarction or stroke, within 3 months of the start of the study; (iii) primary neurological disorders, such as delirium, Parkinson's disease or dementia; (iv) severely disabled communication, such as being blind, deaf or mute; and (v) a Mini-Mental State Examination (MMSE)11 score below 24. The present study was approved by the Ethics Committee and written informed consent was obtained from all patients before the start of the study.
Subjects were evaluated using the following scales: the Hamilton Depression Rating Scale (Ham-D)12 for the severity of depression; the MMSE for cognition; and the Barthel Index (BI)13 and the Hyogo Activities of Daily Living Scale (HADLS)14,15 for ADL. In the present study, 21 items of the Ham-D were classified into 10 subscales based on previous studies.16,17 The subscales were: (i) core depressive symptoms, including sad mood, loss of motivation for work and interests and psychomotor retardation; (ii) feelings of guilt; (iii) suicidal thoughts; (iv) sleep disturbances, including initial insomnia, middle insomnia and early awakening; (v) lack of energy, including fatigability, decreased libido and diurnal variation; (vi) somatic anxiety/hypochondriasis; (vii) agitated irritability, including agitation and psychological anxiety; (viii) loss of appetite/weight loss; (ix) insight into or awareness of depressive symptoms; and (x) other psychotic symptoms, including depersonalization, paranoid symptoms and obsessive–compulsive symptoms. The BI included the 10 item ADL scale and was scored from 0 to 100, with higher scores indicating better performance of ADL.13 The HADLS included seven items of BADL (toileting, eating, dressing, grooming, face washing, brushing teeth and bathing) and 11 items of IADL (going out, telephoning, shopping, making meals, cleaning rooms, making beds, washing dishes, laundering, managing fire, using electric light and television switches and paying bills) and was scored from 0 to 100, with higher scores indicating poorer performance of ADL.14 The HADLS was originally developed to measure deterioration of ADL among the elderly with dementia and shows a satisfying inter-rater reliability (κ–indices = 0.14–1.00) and a validity for the measurement of global deterioration of ADL among elderly subjects with dementia.14,15
All scales were evaluated by a trained psychiatrist or psychologist by observing patients and interviewing patients and their families. The rater examined all patients twice at an interval of approximately 30 days.
Correlations between ADL scores (BI and HADLS) and individual characteristics (age, education, Ham-D score and MMSE score) were analyzed using Spearman's correlation coefficients and the changes between baseline and the follow-up scores were analyzed with paired t-test. After calculating the difference between the two scores by subtracting each subject's baseline score from his or her follow-up score, multiple regression analyses were used to evaluate the determinant of the difference in Ham-D and Ham-D subscales on the difference in HADLS. All data are expressed as the mean ± SD. Statistical analyses were performed using spss version 12.0 for Windows (SPSS, Chicago, IL, USA). P < 0.05 was considered significant.
A total of 70 patients (22 men and 48 women) participated in the present study. Fifty-nine were outpatients and 11 were inpatients. The mean age was 72.3 ± 5.3 years (range 65–87 years). All subjects were receiving treatment for a current depressive episode or a maintenance therapy for depression with antidepressants and/or benzodiazepines. The mean interval of examinations was 29.7 ± 15.7 days. No significant differences were observed between men and women regarding mean age and scores on the Ham-D, MMSE, BI and HADLS. At baseline, the mean Ham-D score was 13.7 ± 7.9 (range 2–33), the mean MMSE score was 26.9 ± 1.9 (range 24–30), the mean BI score was 97.5 ± 4.9 (range 80–100) and the mean HADLS score was 7.7 ± 10.7 (range 0–43.5). Forty-three patients (61.4%) had one or more positive scores on HADLS (Table 1).
Table 1. Clinical characteristics of the subjects
|Total no. participants||70|
|No. women (%)||48 (68.6)|
|No. outpatients (%)||59 (84.3)|
|Age (years)||72.3 ± 5.3 (range 65–87)|
|Education (years)||10.6 ± 2.2 (range 6–16)|
|Baseline clinical assessment scores|
| Ham-D||13.7 ± 7.9 (range 2–33)|
| MMSE||26.9 ± 1.9 (range 24–30)|
| BI||97.5 ± 4.9 (range 80–100)|
| HADLS||7.73 ± 10.71 (range 0–43.5)|
| Basic ADL||0.99 ± 3.53 (range 0–26.3)|
| Instrumental ADL||6.68 ± 9.04 (range 0–36.5)|
|Using light/television switches||5||7.1|
Table 2 shows the correlation between ADL scores and selected variables. The baseline BI and HADLS scores were significantly correlated with the baseline Ham-D score (r =−0.265 and r = 0.454, respectively). At the follow-up examination, significant correlations with scores on both the BI and HADLS with the Ham-D were also noted.
Table 2. Correlations between baseline and follow-up scores on clinical activities of daily living assessments and selected variables
|Age||0.087 (0.476)||−0.047 (0.699)||0.138 (0.256)||0.056 (0.644)|
|Years of education||0.190 (0.116)||−0.138 (0.256)||0.061 (0.616)||−0.078 (0.520)|
|MMSE score||−0.253 (0.034)||0.132 (0.278)||−0.173 (0.151)||−0.045 (0.710)|
|Ham-D score||−0.265 (0.027)||0.454 (<0.001)||−0.387 (0.001)||0.445 (<0.001)|
Changes between baseline and the follow-up scores on clinical assessments
The results of a paired t-test for changes between baseline and the follow-up scores on total and individual items of the Ham-D, MMSE and HADLS are summarized in Table 3. Significant changes were observed in total scores for the Ham-D (P < 0.001), MMSE (P = 0.003) and HADLS (P = 0.017), but not on the BI (P = 0. 211). For the HADLS subscales, significant changes were observed on the BADL score (P = 0.009), but not on the IADL score (P = 0.078). The subscales that changed significantly were core depressive symptoms, suicidal thoughts, sleep disturbances, lack of energy, somatic anxiety/hypochondriasis, agitated irritability, loss of appetite/weight loss and insight on the Ham-D, orientation and attention/calculation on the MMSE and telephoning on the HADLS.
Table 3. Paired t-test for changes between baseline and follow-up scores on total and individual items of the clinical assessments
|Ham-D (66)||13.7 ± 7.9||9.1 ± 6.4||<0.001|
| Core depressive symptoms (12) (sad mood, work and interests, retardation)||2.8 ± 2.3||1.8 ± 2.0||<0.001|
| Feelings of guilt (4)||0.8 ± 1.2||0.6 ± 0.8||0.198|
| Suicidal thoughts (4)||0.9 ± 1.5||0.2 ± 0.4||<0.001|
| Sleep disturbances (8) (initial and middle insomnia, early awakening||2.2 ± 2.4||1.4 ± 2.1||0.002|
| Lack of energy (6) (fatigability, decreased libido, diurnal variation)||2.0 ± 1.4||1.4 ± 1.3||<0.001|
| Somatic anxiety/hypochondriasis (8)||2.3 ± 1.8||1.4 ± 1.1||0.008|
| Agitated irritability (8) (agitation and psychological anxiety)||2.2 ± 1.9||1.6 ± 1.8||0.013|
| Loss of appetite/weight loss(4)||0.9 ± 1.1||0.5 ± 1.0||0.008|
| Insight (2)||0.2 ± 0.5||0.1 ± 0.3||0.006|
| Other psychotic symptoms (10) (depersonalization, paranoid and OCS)||0.2 ± 0.6||0.1 ± 0.5||0.497|
|MMSE (30)||26.9 ± 1.9||27.7 ± 2.0||0.003|
| Orientation of date and place (10)||9.7 ± 0.6||9.9 ± 0.3||0.013|
| Registration (3)||3.1 ± 0.3||3.0 ± 0.0||0.159|
| Attention/calculation (5)||2.7 ± 1.7||3.2 ± 1.6||0.017|
| Recall (3)||2.6 ± 0.7||2.6 ± 0.7||0.784|
| Language† (9)||8.9 ± 0.3||8.8 ± 0.6||0.052|
|Bl (100)||97.5 ± 4.9||98.0 ± 4.1||0.211|
| Transfers (bed to chair and back) (15)||14.9 ± 0.8||14.9 ± 0.8||1.000|
| Mobility (on level surfaces) (15)||14.6 ± 1.3||14.6 ± 1.3||1.000|
| Stairs (10)||9.0 ± 2.9||9.3 ± 2.3||0.288|
| Feeding (10)||10.0 ± 0.0||10.0 ± 0.0||1.000|
| Dressing (10)||9.6 ± 1.9||9.6 ± 1.9||1.000|
| Grooming (5)||4.7 ± 1.1||4.8 ± 1.0||0.567|
| Bathing (5)||4.8 ± 1.0||4.9 ± 0.6||0.159|
| Toilet use (10)||10.0 ± 0.0||9.9 ± 0.6||0.321|
| Bowels (10)||10.0 ± 0.0||10.0 ± 0.0||1.000|
| Bladder (10)||9.9 ± 0.8||9.9 ± 0.6||0.321|
|HADLS (100)||7.73 ± 10.71||6.03 ± 9.36||0.017|
| Basic ADL (53.3)||0.99 ± 3.53||0.98 ± 2.50||0.009|
| Toileting (3.2)||0.05 ± 0.29||0.00 ± 0.00||0.182|
| Eating (5.5)||0.02 ± 0.17||0.00 ± 0.00||0.321|
| Dressing (14)||0.32 ± 1.40||0.16 ± 0.94||0.102|
| Grooming (6.5)||0.22 ± 0.88||0.06 ± 0.47||0.072|
| Washing face (10.6)||0.12 ± 0.80||0.09 ± 0.77||0.321|
| Brushing teeth (7.3)||0.08 ± 0.55||0.00 ± 0.00||0.207|
| Bathing (6.2)||0.18 ± 0.74||0.13 ± 0.58||0.467|
| Instrumental ADL (46.7)||6.68 ± 9.04||5.55 ± 8.51||0.078|
| Going Out (9.4)||0.86 ± 1.68||0.96 ± 1.77||0.680|
| Telephoning (4)||0.49 ± 1.22||0.08 ± 0.50||0.012|
| Shopping (5.2)||0.89 ± 1.81||0.68 ± 1.52||0.261|
| Making meals (4.7)||1.33 ± 1.99||1.10 ± 1.85||0.219|
| Cleaning rooms (2.4)||0.62 ± 0.95||0.47 ± 0.86||0.234|
| Making beds (2.5)||0.56 ± 0.96||0.51 ± 0.96||0.606|
| Washing dishes (2.7)||0.60 ± 1.04||0.53 ± 0.98||0.535|
| Laundering (3.7)||0.72 ± 1.35||0.46 ± 1.13||0.080|
| Managing fire (4.9)||0.42 ± 1.32||0.37 ± 1.28||0.780|
| Using light/TV switches (3)||0.19 ± 0.72||0.13 ± 0.61||0.517|
| Paying bills (4.2)||0.20 ± 0.87||0.18 ± 0.86||0.871|
Difference in the Ham-D (dHam-D), MMSE (dMMSE), and HADLS (dHADLS) scores were entered into a multiple regression analysis. Result showed the dHam-D was related to the dHADLS (P = 0.017), but not the dMMSE (P = 0.996). We then examined whether the differences on the Ham-D subscore influenced dHADLS. The differences of subscore for suicidal thoughts (P = 0.038), sleep disturbances (P = 0.020) and loss of appetite/weight loss (P = 0.004) were related to dHADLS. The difference in the subscore for agitated irritability was negatively correlated with the dHADLS (P = 0.009; Table 4).
Table 4. Multiple regression analyses of determinants on Hyogo Activities of Daily Living Score score 2-1†
|Ham-D score 2-1†||0.312||0.295||0.017|
|MMSE score 2-1†||−0.002||−0.001||0.996|
|r2 = 0.087|
|Ham-D subscore 2-1†|
| Core depressive symptoms (sad moods, work and interests, retardation)||−1.21||−0.27||0.224|
| Feelings of guilt||0.47||0.09||0.664|
| Suicidal thoughts||6.74||0.49||0.038|
| Sleep disturbance (initial and middle insomnia, early awakening)||2.02||0.62||0.020|
| Lack of energy (fatigability, decreased libido, diurnal variation)||0.7||0.13||0.447|
| Somatic anxiety/hypochondriasis||−1.53||−0.46||0.069|
| Agitated irritability (agitation and psychological anxiety)||−1.61||−0.57||0.009|
| Loss of appetite/weight loss||4.77||0.78||0.004|
| Other psychotic symptoms (depersonalization, paranoid and OCS)||−3.97||−0.38||0.064|
|r2 = 0.705|
The principal finding of the present study was that substantial improvement in ADL occurs in association with temporal improvement of the severity of depression among subjects with late-life depression. Despite the fact that most subjects who participated in the present study were free of physical impairments due to medical illness, approximately 20% of subjects were impaired in at least one BADL subscale of the HADLS and 60% were impaired in at least one IADL subscale of the HADLS. Most patients showed slight cognitive impairments on the MMSE. The examinations were conducted twice at short intervals to exclude the influence of possible confounding variables, such as age, medical illness or social support. Significant improvement in the severity of depression was observed in these subjects. Significant improvements in HADLS score and cognitive function were also observed. Regardless of individual changes in the severity of depression or ADL function, the association between the severity of depression and impaired ADL persisted in the present group. Multiple regression analysis showed that the changes in the severity of depression contributed significantly to changes in HADLS; however, changes in cognitive status did not contribute significantly to changes in HADLS. These results indicate that the beneficial changes in ADL observed in the present study were closely associated with improvement in depressive symptoms and appeared to be relatively independent of cognitive improvement.
In the present study, patients with greater improvement in depressive symptoms also experienced substantial functional improvement. To date, several studies have suggested that treatment for depression improves functional status in late-life depression.18–20 A recent large-scale clinical trial with a treatment program reported the efficacy of collaborative care management on both depression and functional status in patients with late-life depression.21 Most patients remained depressed in the 1 year follow-up period. Callahan et al. also conducted a secondary analysis to estimate the magnitude of improvement in depression (at least a 50% reduction in symptoms) on disability. Patients with significant reduction in depressive symptoms experienced improvement in physical function and preservation of IADL at 12 months, although the difference in IADL between the improved and unimproved groups was not significant at baseline.20
Previous studies have suggested a possible link between impaired IADL and cognitive dysfunction in late-life depression.22 Cognitive dysfunction, especially in executive domains, is common among patients with late-life depression and may be considered part of the clinical manifestation of depression.23 Recently, Kiosses and Alexopoulos examined the relationship between the severity of depression and cognitive domains in depressed elderly subjects with impaired IADL and reported that the severity of depression alone, or along with impairment in executive function, influenced some IADLs.24 The association between depressive symptoms and impaired IADL in late-life depression may be complex, because depressive symptoms and cognitive dysfunction are reciprocally interrelated.23 In the present study, changes in the severity of depression, cognitive status and ADL were all significant. However, improvement in cognition was modest and a number of patients remained impaired in most IADL domains, such as making meals, cleaning rooms, making beds, washing dishes, going out and shopping. Therefore, a supportive approach for treatment that focuses on impaired IADL may be necessary for individuals with late-life depression.
An exploratory analysis in the present study suggested possible associations between an improvement in ADL and changes in specific depressive symptoms, such as suicidal thoughts, sleep disturbances and loss of appetite/weight loss. Indeed, these symptoms are common in late-life depression and may represent early signs of global improvement of depression during treatment.25 There is a possibility that these symptoms are closely linked with disability. Previous studies have also suggested that sleep disturbances and weight loss are possibly linked to disability.26,27 It is not certain why the change in agitated irritability was negatively related with changes in ADL. The restlessness accompanied by agitated irritability would counterfeit an active status. At present, the causal relationship between the severity of specific depressive symptom and improvement in ADL remains to be elucidated.
From the viewpoint of the vascular depression hypothesis, cerebrovascular disease with silent stroke and white matter lesions is closely linked to a subset of late-life depression. Patients with vascular depression have distinct clinical features with poor response to treatment, cognitive impairment and disability. The underlying etiology for disability and executive dysfunction in late-life depression is of concern.28 Further investigations, including neuroimaging, are needed to elucidate the precise links between late-life depression and vascular conditions.
The present study has demonstrated that a large number of patients with late-life depression have problems in ADL in the hospital setting and that a substantial improvement in ADL occurred in association with temporal improvement of depressive severity among subjects with late-life depression. Because depression is considered to be a reversible condition, and a common cause of disability in older adults, cognizance of a patient's ADL during treatment for late-life depression should be emphasized.