Dr Yvonne Skarbek* Psychiatrist, St John of God Medical Centre Burwood

Dr Nalin A Singh* Geriatrician, Department of Geriatrics, Balmain Hospital, Central Sydney Area Health Service.

Dr Gary Galambos Psychiatrist, Rozelle Hospital, Central Sydney AHS & St John of God Medical Centre Burwood

Theodora M Stavrinos Occupational Therapist (Research Assistant)

MA Fiatarone-Singh U.S. Physician

*principal authors

BACKGROUND:

This study explored PRT exercise effectiveness in cases of clinically significant depressive disorder and the intensity required to elicit an antidepressant response, which was compared with general practitioner care. A  randomised-controlled trial with blinded outcome assessment of 8-weeks duration was conducted. Primary depression outcomes included the Geriatric Depression Scale (GDS), the Hamilton Rating Scale of Depression (HRSD) & DSM-IV criteria.

OBJECTIVES

The principal objectives of this research were:

1. To assess whether physical exercise (PRT) was an effective form of treatment for mild-to-moderate forms of depressive illness (DSM-IV unipolar non-psychotic major or minor depression or dysthymia) in the elderly.
2. To determine the appropriate exercise intensity or 'dose' needed to result in mood improvement.
3. To ascertain if any significant personality traits may be predictive of a subject's response to exercise being used as a form of treatment for their depressive illness.
4. To determine whether the psychological constructs of self-efficacy and locus-of-control are possible mechanisms of the antidepressant effect of exercise.

1. PRT exercise would result in a decrease in both self-rated and therapist-rated depressive symptoms.
2. High-intensity PRT would result in a greater improvement in depressive symptoms as compared with low-intensity PRT.
3. Subjects with a high neuroticism personality trait score would be predicted to have a poorer response to treatment with PRT as compared with those subjects with a lower neuroticism trait score.
4. Subjects with high extroversion personality trait scores would be predicted to have a better response to treatment with PRT as compared with those subjects who were more introverted.
5. That the psychological mechanisms by which PRT caused an improvement in mood was by both an increase in self-efficacy and an internalisation of one's locus-of-control.

METHOD

Recruitment: A mailing list of community dwelling elderly over the age of 60 was created from general practitioners working within the Hospital catchment area. All on the list were sent a letter with a Geriatric Depression Scale-30 item (GDS). Telephone screening & a repeat GDS was conducted on subjects scoring >14. Subjects meeting inclusion criteria were recruited to attend a formal assessment. On the day of baseline screening, subjects were seen by a physician who obtained a medical history & conducted a physical examination. Then a psychiatrist conducted a psychiatric clinical interview, where a number of instruments to measure mood and secondary outcome psychological variables of interest were administered.

Criteria: Subjects had to be aged >60, be community dwelling, score >15 on the GDS and fulfil DSM-IV criteria for either major or minor depression or dysthymia. Exclusions included subjects taking antidepressants, those currently seeing a psychiatrist or with medical contraindications. If they met all inclusion criteria, they saw an occupational therapist & underwent baseline strength testing of their major muscle groups.

Randomisation: Subjects were computer randomised to 1 of 3 groups:

Treatment program: supervised by a trained occupational therapist. Six pneumatic resistance exercise machines (Keiser Sports Health equipment) were chosen for their importance in strengthening large muscle groups of functional significance. The exercises performed included knee flexion, knee extension, chest press, upright row, shoulder press and leg press. The subjects in the high intensity PRT group had each machine set at 80% of their one-repetition-maximum (1RM) (the maximum load that could be lifted for one repetition only) for each exercise, whilst the subjects in the low intensity group had each machine set at 20% of their 1RM. The exercise intensity was maintained throughout the 8-week program by progressively increasing the load from session to session as tolerated. Subjects were blinded to which exercise group they were enrolled in. In both groups, each subject performed 3 sets of 8 repetitions on each machine. Every exercise session began with 5-10 minutes of warm-up stretching exercises and concluded with 5-10 minutes of stretching cool-down exercises. One to 8 subjects were trained simultaneously within a group. Subjects trained 3 times a week. For subjects in the general practitioner control group, a letter was sent to the GP informing them of the subject’s diagnosis of depression. Further care was determined by their GP.

Outcome measures:

Geriatric Depression Scale (GDS, score 0-30)

This was the self-rated screening measure for depression. It focuses upon cognitive & emotional aspects of depression and does not address physical complaints, which are common in the elderly. It is a valid and sensitive measure of depression in the elderly (Yesavage, 1983). The GDS score >14 chosen as the cut-off for inclusion into the study was based on the finding of Brink et al (1982) that this yields an 80% sensitivity rate & 100% specificity rate for the diagnosis of depression.

The 17-item Hamilton Rating Scale of Depression (HRSD, score 0-52)

This was the therapist-rated measure of depression used. It is probably the most widely used scale in research studies of treatments for depression & therefore allows for comparison with other studies. It is a reliable and valid measure of depression (Carroll, 1973) & the structured Interview guide (Williams 1988) for the HRSD was used to further increase its reliability.

DSM-IV symptoms for psychiatric diagnoses

The General Self-efficacy scale (Sherer 1982)

This is a 12 item scale that is scored on a 5 point Likert scale (score 0- 60). The higher the score, the greater a person's self-efficacy. Factor analysis has shown 3 aspects underlying the scale: willingness to initiate behaviour 'initiative', willingness to expend effort in completing the behaviour 'effort' and persistence in the face of adversity. Its use has been validated in elderly populations (Bosscher 1998).

The Multidimensional Health Locus of Control (MHLC) internal subscale

This instrument was developed by Wallston et al (1978) to identify individuals’ sources of reinforcement for health-related behaviours. It measures the degree to which individuals believe that health-related outcomes are under the control of the self (Internal HLC), powerful others (PELC) & chance (CHLC). A higher score indicates that the dimension measured is a stronger source of reinforcement for health-related behaviours for the respondent. Thus those scoring highly on the HLC-internal believe that their health related outcomes are a consequence of their own actions, and potentially under their control. The MHLC has been shown to be a valid and reliable measure of locus of control amongst both adult and elderly populations.

The Eysenck Personality Questionnaire- short scale (EPQ-R) - Neuroticism/Stability and Extroversion/Introversion subscales

12 items are present on each subscale, and are answered with a yes/no response. It is a quick and efficient method of assessing personality in a dimensional manner and has been used widely in personality research. The manual contains normative scores for age groups up to the age of 70. To reduce the confounding effect of depression on personality measurement the EPQ-R was administered in accordance with the findings made by Kendell et al (1968), who found that spurious high neuroticism scores and spuriously low extroversion scores were prevented by asking the subject to try and disregard their current mood state when answering the questions on personality, and "answer according to how you feel or behave when you are your usual self".

PRELIMINARY FINDINGS

Results for the first 34 subjects  are as follows:

For the first 38 subjects:

There were no significant differences in any baseline characteristics between groups.  

Changes in outcomes over the 8 weeks for the first 38 subjects were:

In post hoc testing, high-intensity PRT was significantly better than control group in both self (p=0.0006) and therapist-rated scales (p=0.002).

Changes in outcomes over the 8 weeks for the first 34 subjects were:

CONCLUSION

High-dose PRT exercise may be an effective treatment or adjunct to orthodox treatments of mild-to-moderate forms of depressive illness in community-dwelling elderly people who have comorbid physical health problems & a tendency towards internal local of control (rather than external LOC, which has been found to be more representative of community samples of depressed elderly). The fact that this may have been a somewhat biased sample was reflected in the lower than expected prevalance rate (as measured by the GDS) of 3.8%. Other limitations of the study were its short follow-up (8 weeks) & positive effects secondary to social interaction within the exercise group.

[ Home ] [ Up ] [ Computers in Psych ] [ Metaphorical models ] [ Doctors' Mental Health ] [ Dissertation - Exercise for Elderly Depression ] [ Publication - Exercise for Elderly Depression ] [ Inpatient Dynamics ] [ Sensory Deprivation ] [ Medication use in first onset psychosis ] [ Surviving Trauma ] [ Teen Exam Stress ] [ Anxiety Disorder Lecture ] [ Noradrenergic antidepressants lecture ] [ PTSD Lecture ] [ DMH Lecture ] [ DMH Lecture ] [ EP lecture 1 ] [ Psychiatric Emergencies ] [ Low self-esteem ] [ Being Perfect ] [ Anger Interview ] [ Living in a wealthy safe city... ] [ Stress Put to the test ] [ War of Words ] [ Over 55s ] [ EP lec ] [ EP radio IV ]

The information on this Web page have been obtained from:
Skarbek Y, Exercise as a Treatment for Depression in the Elderly, NSW, 1999, Dissertation, Royal Australian and New Zealand College of Psychiatrists
& from a draft Abstract prepared by Dr N Singh on behalf of the research group:
Singh NA, Skarbek Y, Galambos G, Stavrinos TM & Fiatarone-Singh

The materials provided on this website may be freely cited but reposting on other websites, publishing or other reproductions, whole or in part, are subject to the written permission of Gary Galambos. Images may be reproduced provided the source is properly acknowledged.  

Page last updated: 23 August 2005