Design paper
The DOSE study: a clinical trial to examine efficacy and dose response of exercise as treatment for depression

https://doi.org/10.1016/S0197-2456(02)00226-XGet rights and content

Abstract

The Depression Outcomes Study of Exercise (DOSE) was a randomized clinical trial to determine whether exercise is an efficacious treatment for mild to moderate major depressive disorder (MDD) in adults ages 20 to 45 years. The specific hypotheses under investigation were (1) active exercise is an efficacious monotherapy for mild to moderate levels of MDD, and (2) there is a dose-response relation between the exercise amount and reduction in depressive symptoms. The primary outcome measure was the Hamilton Rating Scale for Depression (HRSD) collected weekly over 12 weeks. Secondary outcome measures were the Inventory of Depressive Symptoms (clinician and self-report), HRSD scores at 24 weeks, cardiorespiratory fitness, self-efficacy, and quality of life. Eighty men and women who were diagnosed with a Structured Clinical Interview for Depression and who had mild (HRSD 12–16) to moderate (HRSD 17–25) MDD were randomized to one of five doses of exercise: 7.0 kcal/kg/week in 3 days/week; 7.0 kcal/kg/week in 5 days/week; 17.5 kcal/kg/week in 3 days/week; 17.5 kcal/kg/week in 5 days/week; or 3 days/week of stretching and flexibility exercises for 15 to 20 min/session. Participants exercised under supervision in our laboratory over the course of 12 weeks. Symptoms of depression were measured weekly by trained clinical raters blinded to the participant's treatment assignment. The design of the study restricted participant characteristics to mild to moderate MDD and controlled exercise features to permit the evaluation of exercise as a sole treatment for depression. This study is the first to examine dose-response effects of exercise in participants diagnosed with MDD.

Section snippets

Background and aims

Depression is a prevalent illness (17% lifetime prevalence) [1] with a high cost to society (estimated $44 billion/year) [2] and to the quality of life of the individual 3, 4. The Global Burden of Disease study [5] found that in developed nations unipolar major depression ranks second behind ischemic heart disease in lost years of healthy life due to premature death or disability. Current estimates indicate that only 25% of adults suffering from depression seek treatment despite major advances

Specific objectives

There were two major objectives of the DOSE study. The first was to determine the efficacy of aerobic exercise as a sole treatment of mild to moderate MDD in participants randomized to exercise compared with an “equal contact” exercise placebo control group over a 12-week acute phase treatment. The second objective was to determine the dose-response relation between different amounts and frequencies of aerobic exercise with the reduction of depressive symptoms. We hypothesized that the active

Participant selection criteria

We recruited men and women between the ages of 20 to 45 years with mild to moderate MDD. The HRSD score for mild depression is 12 to 16 and for moderate depression 17 to 25 [17]. The choice of age group was made for both scientific and practical reasons. The scientific literature suggests that MDD usually begins when individuals are 20 to 30 years of age, although it can begin at any age [18]. Also, visit rates to psychiatrists are the highest for the 20- to 40-year age group, and the

Primary outcomes

The HRSD-17 (17 items) is a clinician-rated measure designed to assess the severity of depression by assessing both the intensity and frequency of depressive symptoms. This is the most widely used measure of depression severity for clinical trials research. One review published in 1996 reported it was used in 500 published studies over a 10-year period [21], using the Cronbach α=0.88 for internal consistency. Others have found the HRSD to be correlated with psychiatrists' global rating (r

Trial design

The design of this study was a 2×2 factorial design with an exercise placebo as the control group. The two exercise factors manipulated in this design were total weekly energy expenditure per kilogram of body weight (7 kcal/kg/week and 17.5 kcal/kg/week) and frequency of exercise (3 days/week or 5 days/week). Therefore, for a 70-kg person, the total weekly energy expenditure would be 490 kcal/week if he or she received the 7 kcal/kg/week dose or 1225 kcal if he or she received the 17.5

Recruitment and screening procedures

We used a combination of recruitment techniques based, in part, on recommendations from The Cooper Institute Community Advisory Board. This included media stories on television and in the newspaper, mass mailings, letters to churches and physicians, community health fairs, and flyers distributed to businesses such as grocery stores, libraries, and apartment buildings.

Size of the trial and statistical power

We anticipated that HRSD-17 scores at baseline would have a mean of 20 points and standard deviation of seven points 49, 50, 51, 52, 53. Clinically meaningful and important treatment effects would reduce HRSD scores ⩾50%, a score of ten or less. Without preliminary data for within-individual correlation of HRSD scores pre- and postintervention, statistical power was conservatively based on comparing postintervention scores. Power for comparing active treatments with control was based on the

First 12 weeks: supervised exercise

Following randomization, all participants exercised under supervision at our laboratory for the first 12 weeks of the treatment study. Weight (kg) was collected weekly to calculate the weekly exercise dose. Participants were allowed to use a treadmill (Technogym RUNRACE) or stationary bicycle (Technogym BIKERACE) and wore a Polar Vantage XL heart- rate monitor during each exercise session. Data from the heart-rate monitor were downloaded into the exercise database. If participants used the

Data collection

Table 2 shows a summary of all data collection and time points for the DOSE study beginning with the telephone prescreen.

Data management and quality control

The data management group at The Cooper Institute oversees all data collection. Manually entered data was double-keyed and verified for accuracy, using on-screen database forms that match the format of paper forms, programmed with appropriate skip patterns and range and logic checks. Scannable data forms were reviewed prior to processing to ensure marks were clear and dark

Plans for statistical analyses

To assess the effectiveness of randomization, we will tabulate mean baseline levels of the HRSD and IDS scores among randomized participants by treatment assignment, along with age, gender, body mass index (BMI), energy expenditure, resting heart rate, blood pressures, predicted exercise capacity, medical history, and psychosocial measures. Characteristics of randomized participants who refuse treatment, drop out before completing acute-phase treatment, or complete treatment will be tabulated

Discussion

The DOSE study is the first study to our knowledge to manipulate the dose of exercise in a group of participants stringently diagnosed with mild to moderate MDD. The high internal validity of the study will allow us to assess whether exercise could be an efficacious treatment for these individuals and to identify what dose of exercise is necessary for treatment. Randomization of participants started in August 1998, and treatments were completed in October 2001. The recruitment to randomization

Acknowledgements

Participating institutions: The Cooper Institute, Dallas, Texas, USA; The University of Texas Southwestern Medical Center Depression and Anxiety Disorders Program, Dallas, Texas, USA. Steering Committee: Andrea L. Dunn, Madhukar H. Trivedi, Heather O. Chambliss.

We wish to thank participants in the DOSE study; The Cooper Institute Scientific Advisory Board: Claude Bouchard, William L. Haskell, Norman M. Kaplan, I-Min Lee, Kiang Liu, Guy S. Parcel; The Cooper Institute Community Advisory Board:

References (58)

  • R.L Spitzer et al.

    Health-related quality of life in primary care patients with mental disorders. Results from the PRIME-MD 1000 Study

    JAMA

    (1995)
  • National Center for Health Statistics. Healthy people 2000. Hyattsville, Maryland: Public Health Service;...
  • U.S. Department of Health and Human Services. Tracking healthy people 2010. Washington, D.C.: U.S. Government Printing...
  • U.S. Department of Health and Human Services. Goldman HH, Rye P, Sirovatka P, editors. Mental health: a report of the...
  • J.P Koplan et al.

    The risks of exercisea public health view of injuries and hazards

    Public Health Rep

    (1985)
  • Rush AJ, Golden WE, Hall GW, et al. Agency for Health Care Policy and Research, eds. Depression in primary care, Vol....
  • E.W Martinsen et al.

    Effects of aerobic exercise on depressiona controlled study

    Br Med J (Clin Res Ed)

    (1985)
  • J Fremont et al.

    Aerobic exercise and cognitive therapy in the treatment of dysphoric moods

    Cognitive Therapy Res

    (1987)
  • N.A Singh et al.

    Exercise and depression in the older adult

    Nutrition in Clinical Care

    (2000)
  • L.L Craft et al.

    The effect of exercise on clinical depression and depression resulting from mental illnessa meta-analysis

    J Sport Exerc Psych

    (1998)
  • J.A Blumenthal et al.

    Effects of exercise training on older patients with major depression

    Arch Intern Med

    (1999)
  • E Frank et al.

    Conceptualization and rationale for consensus definitions of terms in major depressive disorder. Remission, recovery, relapse, and recurrence

    Arch Gen Psychiatry

    (1991)
  • American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington, D.C.:...
  • S Schappert

    Office visits to psychiatristsUnited States 1989–90

    Advance Data (from Vital and Health Statistics of the CDC)

    (1993)
  • American College of Sports Medicine. Franklin BA, Whaley MH, Howley ET, editors. ACSM's guidelines for exercise testing...
  • I McDowell et al.

    The Hamilton Rating Scale for Depression

  • A.T Beck et al.

    BDI-II manual

    (1996)
  • A.J Rush et al.

    The Inventory of Depressive Symptomatology (IDS)psychometric properties

    Psychol Med

    (1996)
  • D.R Bassett et al.

    Accuracy of five electronic pedometers for measuring distance walked

    Med Sci Sports Exerc

    (1996)
  • Cited by (0)

    View full text