Women & Health, Vol. 34(3) 2001
"The Effects of a Multi-Modal Intervention Trial of Light, Exercise, and Vitamins on Women's Mood"

Marie-Annette Brown, PhD, FNP, FAAN
Jamie Goldstein-Shirley, MSN, RN
Jo Robinson
Susan Casey, PhC, MN, RN

ABSTRACT. The purpose of this study was to test the efficacy of a tri-modal intervention (LEVITY) to improve women's mood. This eight week randomized experiment with a placebo-control group targeted women with symptoms of mild to moderate depression. Women in the intervention group were instructed to take a brisk 20-minute outdoor walk at target heart rate of 60% of maximum heart rate, to increase light exposure throughout the day and to take a specific vitamin regimen. Women in the control group took a daily placebo vitamin. The sample consisted of 112 women ages 19-78 who reported mild to moderate depressive symptoms. They were in otherwise good health and were not currently taking any mood-altering medication. Pre-and post-intervention assessment utilized five measures of mood: Center for Epidemiology Studies Depression Scale, Profile of Mood States, Depression-Happiness Scale, Rosenberg Self-Esteem Scale, and the General Well-Being Schedule. Analysis of covariance indicated that the intervention was effective in improving women's overall mood, self-esteem, and general sense of well-being and in decreasing symptoms on two measures of depression. Of particular note, the data from all five outcome measures supported the efficacy of the intervention. In addition, a high level of adherence to the intervention protocol was observed with two-thirds of the women reporting 100% adherence. Study implications suggest that this type of intervention may provide an effective, clinically manageable therapy for mild-to-moderately depressed women who prefer a self-directed approach or who have difficulties with the cost or side-effects of medication or psychotherapy.

KEYWORDS: Depression, subsyndromal depression, depression-atypical symptoms, vitamins, exercise, light, women, lifestyle change, alternative therapies, CES-D, POMS.

INTRODUCTION

Epidemiological surveys consistently show that after adolescence, girls and women are twice as likely as are boys and men to be diagnosed with depression (Williams et al. 1995). This is true both for major depression and milder forms such as dysthymic disorder, minor depression, seasonal affective disorder (SAD), and subsyndromal depression (SSD) (Olfson et al. 1996). Gender differences are also apparent in the manifestation of depression (Kornstein 1997). For example, women are more likely to present with vegetative or atypical symptoms such as fatigue, increased appetite, sleepiness, and weight gain, while men are more likely to present with "more typical" symptoms such as decreased appetite, insomnia, and weight loss (Frank, Carpenter, and Kupfer 1988; Young et al. 1990). Women are more likely than men to have a seasonal pattern to their depression (Lee and Chan 1998). Comorbidity is also seen more often in depressed women than in depressed men (Blazer et al. 1994; Regier, Burke, and Burke 1990). In particular, women are more likely to experience anxiety and eating disorders along with a depressed mood. Finally, several studies have shown that women may have longer episodes of depression and may be at greater risk of having a "chronic and recurrent course of illness" (Kornstein 1997, 13).

Gender may also be relevant when considering the mode of therapy, with non-pharmacological intervention especially well-suited for women. For example, drugs may have longer half-lives in women and result in more side effects and toxicity (Kornstein 1997). Also, the effectiveness of the medications can fluctuate with changes in hormone levels (e.g., pregnancy, menstrual cycle, oral contraceptives, and hormone replacement therapy). Women are more likely than men to employ alternatives to traditional medication, such as vitamins, homeopathy, and yoga (Beal 1998). In one study, women expressed a clear preference for "general psycho-educational classes about health and stress" and "less interest in group therapy and medication" (Alvidrez and Azocar 1999, 340).

Despite this, female gender is a positive predictor of antidepressant use with depressed women as much as 55 percent more likely than depressed men to be prescribed antidepressant medications (Simoni- Wastila 1998). Lifestyle interventions offer possible alternative or complementary strategies for relieving depression. Three of the most promising–exercise, increased light exposure, and specific vitamins–were selected for this study. The extensive literature on exercise suggests that exercise can have a significant antidepressant effect in mild to moderate forms of depressive illness in both men and women (Fox 1999). Blumenthal et al. (1999) determined that supervised aerobic exercise was just as effective in treating older patients with major depression as the antidepressant Sertraline. This medication is one of the newer selective serotonin reuptake inhibitors widely prescribed for the treatment of depression, which further underscores the antidepressant potential of exercise. Exercise may have added value for women because the relationship between mood and exercise appears stronger in women than in men (Stephens 1988). Exercise may also relieve physical and emotional symptoms associated with late luteal phase dysphoria (Hightower 1997) and menopause (Slaven and Lee 1997), two conditions unique to women.

Moderate exercise has been shown to be more effective than vigorous exercise in decreasing the anxiety and stress components of depression. In a 1989 randomized, controlled trial, walking at a moderate pace for 20 minutes a day at 60% maximum heart rate was more effective in enhancing overall mood and reducing stress than more rigorous and sustained exercise at a higher heart rate. In fact, the participants who were assigned to the more vigorous program registered an increase in overall stress (Moses et al. 1989).

High-intensity artificial light was first identified as a treatment for seasonal depression in 1985 by Rosenthal et al. Since then, it has also proven effective in treating non-seasonal depression (Kripke et al. 1992; Beauchemin and Hays 1997; Kripke 1998), premenstrual tension (Parry et al. 1993; Lam et al. 1999), and eating disorders (Lam et al. 1994). Light therapy also has the potential to reduce common symptoms of atypical depression, like increased appetite and carbohydrate cravings (Krauchi, Wirz-Justice, and Graw 1990).

New research indicates that increased exposure to natural light may have an antidepressant effect similar to high-intensity artificial light, especially for symptoms of atypical depression. A San Diego survey determined that those adults who spent the most time out-of-doors had the fewest number of symptoms of atypical depression (Espiritu et al. 1994). In a study conducted in Finland and Sweden, "the amount of light and the length of the day as measured by monthly global radiation totals" was found to be the best single predictor of general well-being (Molin et al. 1996). A 1997 London survey of healthy women without SAD found a strong correlation between mood and outdoor light levels. Indeed, mood was more highly correlated with outdoor light levels than with the phase of a woman's menstrual cycle (Einon 1997). In a controlled trial of women with SAD, one hour of natural outdoor light had a significant antidepressant response and was as effective in treating SAD as a 2500 lux high intensity light (Wirz-Justice et al. 1996).

Certain vitamins have proven effective to enhance mood in women. In 1990, a group of female and male volunteers in a double blind cross-over trial were treated with 100 mcg of selenium and showed a significant improvement on the Profile of Mood States (POMS) in 2.5 weeks (Benton and Cook 1990). In a placebo-controlled, double-blind study, women, but not men, who took 50 milligrams of thiamine for 2 months became more clear-headed, composed and energetic (Benton, Griffiths, and Haller 1997). Vitamin D supplements have shown a rapid antidepressant effect during the winter months when sun exposure, and therefore body stores of vitamin D, are most likely to be low (Lansdowne and Provost 1998). Other vitamins with demonstrated potential for mood-enhancing effects in women are folic acid (Alpert and Fava 1997), pyridoxine and riboflavin (Benton, Fordy, and Haller 1995).

The purpose of this study was to test the efficacy of a tri-modal intervention of moderate intensity exercise, increased exposure to natural light and a specific vitamin regimen using a randomized experiment with a placebo control group. The following three rationale supported our use of a tri-modal intervention. Combining the three modalities may result in greater relief of depressive symptoms in a given individual. In the Partonen et al. study (1998), the people who took part in fitness training under bright lighting conditions experienced greater relief from both typical and atypical symptoms of depression than those who participated in the same exercise program under normal lighting conditions. Additionally, an intervention that combines several elements might influence a broader range of symptoms than a single element. For example, bright light has been show to deepen sleep (Campbell et al. 1993), an effect not demonstrated with vitamins. Similarly, exercise has been shown to improve self-esteem (McAuley et al. 1997), a benefit not commonly seen with bright light. Finally, the combination of three separate elements could also increase the likelihood of mood enhancement in a greater percentage of the study participants. In all the studies of the antidepressant effects of light, exercise or vitamins as single components, a substantial number of the participants have not benefited. For example, in a controlled trial of bright light by Kripke et al. (1992), one out of three people did not experience relief from their nonseasonal depression. It is possible that someone who fails to respond to one of the interventions would respond well to another, increasing the overall effectiveness of the program.

Although most studies designed to relieve depression have focused on a single modality, multi-faceted programs have been tested and shown to be effective (Fordyce 1977; Smith, Compton, and West 1995). The benefits of such an intervention, however, must be weighed against its disadvantages. Each added element increases the demands on the participants, potentially lowering adherence and making the program less enjoyable, as well as making it more time-consuming and thus costly for busy health care professionals to teach. Thus, we limited our intervention to a brisk, daily walk outdoors combined with taking vitamins, a regimen that required only twenty minutes a day, five times a week.

METHODS

Screening, Selection & Characteristics
Subjects were recruited from the greater Seattle area using a combination of media, including flyers, a radio talk show, a television news program, and newspaper announcements about the study. Special attention was given to recruiting women from communities of color by means of a focused distribution of flyers to businesses, churches and clinics. Subjects were screened by a member of the research team using the following inclusion criteria: women over age 18; generally healthy with no significant chronic illness; not currently taking any medications which alter mood; and mood scores between 11 and 29 on the Center for Epidemiology Studies Depression Scale (CES-D) (Locke and Putnam 1971). A literature review suggested that this range of CESD scores would be likely to capture women with the presence of mild to moderate depressive symptoms and eliminate women with severe symptomatology or clinical depression. This range of scores also reflects the generally accepted cutoff of 16 (in a possible range of 0-60) to indicate the presence of some depressive symptomatology and the analysis of researchers who used the CES-D concurrently with a diagnostic clinical interview for depression (Weissman et al. 1997; Schulberg et al. 1985).

Exclusion criteria included current daily use of high doses of specified vitamins; aerobic exercise three or more times per week; physical disability that does not allow daily brisk walking; and regular participation in life activities which occur outdoors and exceed one hour a day (e.g., employment as a gardener). Acutely depressed respondents were offered a crisis hotline number. Women whose depression scores were above the cut-off were referred to their primary care provider and encouraged to seek care for their distress.

Four hundred ninety-one women volunteered and 146 met the inclusion criteria. These 146 potential participants were mailed a consent form that included the following: a study description, the potential risks and benefits, and the possibility of receiving placebo vitamins. Of the 125 women who returned these forms, 112 women attended the first two-hour orientation session, yielding an accrual rate of 77%.

Upon arrival at the orientation session, the women completed the Time 1 questionnaires and were then randomized into two groups. To insure the integrity of the randomization process, a consulting statistician who was not a member of the research team designed and implemented the allocation procedures. One group received the entire tri-modal intervention; the other received only placebo vitamins. Because of the recruitment process, which included media discussions of the intervention program, many of the women were aware that we were interested in three particular aspects of lifestyle change. However, in the randomization process, investigators were purposefully vague about the number and nature of the groups. All participants understood that 50% of them would be given placebo vitamins. The placebo and active vitamins were identical in appearance and the bottles labeled only with a code number.

There were no statistical differences between the two groups in the demographic characteristics (See Table 1). They were generally white, college-educated, employed, middle-class women; 43% of the women reported a family income of less than $40,000, 10% reported trade or high school education, 11% were from communities of color, and 51% worked part-time or less. Their average age was 43, the age range from 19-78, with 14% under 30 and 27% over 50 years of age.

Procedures
All the procedures and interventions in this study were approved by the human subjects committee prior to beginning the study. The intervention group was provided with a one-hour educational session about the mood-enhancing benefits of exercise, light and vitamins. The participants were instructed to walk outside during daylight hours for 20 minutes, 5 days each week, with a target heart rate of 60% of their maximum heart rate. In addition, they were asked to utilize strategies learned in the informational session to maximize their exposure to commonly available indoor and outdoor light in all aspects of their life. No high-intensity light boxes were used in this study. Finally, they were provided with a daily vitamin tablet that contained: thiamine (B1) 50 mg, pyroxidine (B6) 50 mg, riboflavin (B2) 50 mg, folic acid 400 mcg, selenium 200 mcg and vitamin D 400 IU. This combination of supplements and these specific doses were based on randomized, controlled trials to verify their effectiveness in improving mood.

Each woman was assigned to a member of the research team who served as her "coach." The coach taught her to calculate her target heart rate (220 minus age times 0.6), take her own pulse, and determine the pace necessary to attain her target heart rate. Women were instructed to monitor their heart rate several times during their daily walk and to focus on what pace was necessary for them to achieve their target heart rate. Coaches also assisted the women to develop strategies to integrate the lifestyle changes into their daily routines. The coach contacted each woman with a brief phone call at two-week intervals to assist her to overcome barriers to adherence. This coaching was streamlined to reflect the kind of counseling currently offered by nurse practitioners in clinical settings who are helping patients implement lifestyle changes.

The control group members, by contrast, received a educational session about the mood-enhancing effects of vitamins, were given an 8-week supply of placebo vitamin tablets to take daily, and were assigned a coach. They received the same supportive coaching calls throughout the study period to assist with overcoming barriers to adherence in regular vitamin use. At the completion of the study, the control group members were notified of their group assignment, given general information about the placebo effect, and provided with active vitamins. They were also offered the opportunity to participate in the full intervention, including 8 weeks of supportive coaching.

All of the instruments below were completed at the orientation session (Time 1) and immediately post-intervention (Time 2). In addition to the standardized instruments, information about demographics, lifestyle habits, and general health status was collected. During the course of the intervention, all participants were asked to complete a brief daily diary documenting their adherence to their group's intervention. After completing the Time 2 questionnaires, women in the intervention group participated in taped interviews to obtain qualitative data about the challenges and benefits of participation in the intervention.

Instruments
The following measures were used to assess the perceived well-being of the participants. The women were asked to consider the previous two weeks as the referent time period in responding to the questions. Center for Epidemiology Studies Depression Scale (CES-D) (Locke and Putnam 1971) is a 20-item scale that measures current affective depression. It was designed to assess for depression in the general community, rather than differentiate among clinical populations, and has been used extensively in research (Cronbach's alpha = 0.83). The 25-item Depression-Happiness Scale (DHS) (McGreal and Joseph 1993) was developed to measure variation along the continuum from depression to happiness. By including items that assess positive mood, this scale is designed to compensate for the narrow range of mood variation described by other scales, such as the CES-D, which are designed to detect only depression (Cronbach's alpha = 0.92). The 10-item Rosenberg Self-Esteem Scale (Rosenberg 1965) measures self-esteem as a global favorable or unfavorable attitude (Cronbach's alpha = 0.86). The Profile of Mood States (POMS) (McNair, Lorr, and Droppleman 1971) is a 65-item survey divided into six factorally-derived subscales: vigor, tension, fatigue, confusion, anger, and depression (Cronbach's alpha = .91). The 18-item General Well-Being Schedule (GWB) (Monk 1981) is a broad indicator of psychological well-being with six subscales: anxiety, depression, general health, positive well-being, self-control, and vitality (Cronbach's alpha = 0.85). The 16-item Levity Inventory was developed by the investigators and pilot tested during this study. The scale included indicators of atypical depression, as well as other aspects of well-being, including optimism, satisfaction with physical and mental health state, and positive coping behaviors. Due to the preliminary nature of the scale, it was not analyzed as an outcome measure. However, initial performance measures were promising and correlations with the other outcome variables ranged from 0.42 to 0.57 (Cronbach's =0.78).

Statistical Analysis
Demographic characteristics of the two groups were compared using ANOVA and t-tests. Efficacy analysis was designed to evaluate the change in scores of the major dependent variables from Time 1 to Time 2 using analysis of covariance (ANCOVA).

RESULTS

Major Dependent Variables
The ANCOVA results indicate that the intervention was effective in improving women's overall mood, self-esteem and general sense of well-being and in decreasing their depression. Despite random assignment, at Time 1 the control group reported lower mood and general well-being, greater depression, and lower self-esteem than the intervention group. Both groups improved significantly (as reflected in the paired t-tests reported in Table 2) on every dependent variable, but for all outcomes, the ANCOVA revealed that the women in the intervention group improved significantly more than their control group counterparts. All but 4 of the 53 participants in the intervention group improved or had no change in their depression scores, whereas, the participants in the control group exhibited more variability in their outcomes.

Additional analysis further illuminated the effects of the intervention on the atypical symptoms of depression. Women in the intervention group reported decreased tension and anger as measured by the POMS subscales (Table 3). The subscales of the GWB revealed that the women experienced decreased depression and increased self-control, vitality, and positive well-being (Table 4). Further analyses were also conducted to determine whether the level of depressive symptoms was related to the degree of women's improvement in mood during the intervention. A MANCOVA was used to stratify the participants into two groups by their Time 1 CES-D score on a median split. The results of the analysis indicated that there was no significant interaction effect; thus the intervention was equally efficacious for women with varying levels of symptomatology.

TABLE: Total Scales Means and ANCOVA Significance (N(Control) = 51, N(Treatment) = 53)

+ At Time 1, the control group scores were significantly (p .05) worse on all scales

# Paired t-tests compare Time 2 to Time 1 scores for each group

* p-value for ANCOVA comparing post-test scores controlling for pre-test scores

Perception of Mood Improvement
At Time 2, women were asked to give a subjective overall impression of their change in mood over the 8 weeks of the project on a 7-point scale from "a lot better" to "a lot worse." In the control group, 59% reported that they had at least some improvement in mood from the beginning of the study. In contrast, 85% of the intervention group reported improved mood. Thirty-seven percent of the control group felt that their mood stayed the same over the course of the study, while only 11% of the intervention group reported this perception. Two women in each group (representing 4% of each group) reported that their mood was worse at the end of the study period. Further analysis explored the extent to which this overall subjective summary of mood change was associated with change on the five dependent variables. There was a moderate to high correlation between women's perceived change in mood and their change scores derived from the standardized instruments. (r = .54 to .68, p = .00 for all correlations).

Adherence to the Intervention Protocol
A high level of adherence to the intervention regimen was recorded in the women's logbooks. A total adherence score was calculated as the sum of the number of occurrences of each the activities (days when walking, light exposure, or vitamin taking were reported in the women's logbooks). One hundred percent adherence was considered to be 40 occurrences of walking, 40 occurrences of light, and 56 days of vitamins totaling a score of 136 in an 8-week period. Some women did more than prescribed by the intervention (for example, walked more than 5 days per week), so achieved scores higher than 136. Almost two-thirds (64%) of the women reported 100% adherence or more. Only 2 participants fell below a threshold of 73% of the total adherence score. On what might be the most challenging portion of the intervention, the walking component, 83% of the participants were compliant at 75% or better. Cited reasons for non-adherence were temporary loss of logbook, discontinuing the vitamin due to increased appetite, and persistent fatigue.

Drop Outs
Of the 112 women who began the intervention, only 8 failed to complete the eight-week intervention and all of the questionnaires. Five of these women were in the control group and three were in the intervention group. Two of these women were lost to follow-up. This small number does not allow for comparisons with those who completed the intervention. Two primary reasons for drop-out were: (1) a decision to seek other treatment or start anti-depressants due to increased depression, and (2) a lack of time or interest.

Side Effect of Vitamins
Some participants in both the control (14%, n = 7) and intervention groups (30%, n = 16) reported a single minor side effects from the vitamins, such as bright yellow urine, gastrointestinal symptoms, sleep disturbances, and skin reactions.

DISCUSSION

These study findings suggest that a program of moderate-intensity walking, increased light exposure, and selected vitamins can improve women's mood. The high level of adherence to the intervention suggests that women could comfortably incorporate this tri-modal program into their daily lives. These findings extend the work of other studies that have demonstrated the positive influence of each independent component (light, exercise, and vitamins) on mood (Kripke 1998; Wirz-Justice et al. 1996; Blumenthal et al. 1999; Moses et al. 1989; Benton, Fordy, and Haller 1995).

Women in the intervention group improved significantly compared to those in the control group on all five dependent variables that measured mood and well-being. Not only did their depression scores decrease, they also reported greater self-esteem, improved general well-being, and greater happiness.

We were particularly interested in determining whether the intervention addressed symptoms more prevalent in women than men, such as anxiety and fatigue. The subscales of the POMS showed that the women in the intervention group experienced a significant decrease in anger and tension. Meanwhile their vitality improved, as measured by the GWB subscales.

Study findings suggest that the tri-modal intervention may also relieve seasonal mood swings that are more common in women than men. The study took place from October into December, a time of increasing rainfall in the Pacific Northwest. (There were 9 days of rain in the first four weeks of the study and 17 days in the second four weeks.) Thus the women were exposed to both cloudier weather and declining hours of daylight, conditions known to trigger seasonal decreases in mood. Nonetheless, the women in the intervention group experienced a significant improvement in mood.

Support for the idea that a tri-modal lifestyle intervention might prove more broadly effective than a solitary intervention comes from a comparison of our results with the results of Moses et al. (1989), that used brisk walking at 60% of maximum heart rate for 20 minutes as the only intervention and used the POMS as one outcome measure. Moses et al. observed improvement on one subscale of the POMS (tension) while the LEVITY participants improved on two (tension and anger). Moses et al. reported one non-significant trend in the confusion subscale, whereas this study found non-significant trends in both the depression and vigor scales. LEVITY participants also improved on the total POMS. (Moses et al. presented results on the subscales only.) This broader improvement shown by combining light and vitamins with moderate-intensity exercise suggests that multi-modal interventions warrant further study.

Our ability to test our supposition that this tri-modal intervention would be effective for a larger percentage of participants than single- modality interventions was limited. To explore this supposition, we calculated the number of participants in the intervention group whose improvement in mood, as measured by the CES-D, exceeded that of the mean change score of the control group (
5.5 points). Nearly two-thirds (63%) of the intervention group had improvement in their CES-D scores greater than 5.5 points from Time 1 to Time

STUDY LIMITATIONS

Study results need to be considered in light of the limitations inherent in the design. The self-selection bias of women volunteering for the study could have yielded women in the action-oriented phase of behavior change as described by Prochaska et al. (1994), a stage characterized by increased motivation and commitment. A woman presenting symptoms of depression to her health care provider may not be as motivated.

Also, the relationship that developed between the coaches and participants might have increased the participants' motivation, or elicited socially desirable responses. These factors could influence the effectiveness of the intervention in actual clinical application.

Additionally, the study participants do not reflect the race and class diversity in the American population. They were, in general, white, middle-class, and well-educated. Because of their privileged status, they were more likely to have flexibility in working arrangements, a safe walking environment, and access to childcare. This intervention warrants further testing in a broader population, including different ages, partner status, ethnicity, and class.

Although other research documents the effectiveness of each individual component of this intervention, this study does not allow direct comparisons between the effects of light, exercise, vitamins, and the combination of all three components. The next step in this program of research will include design aspects that allow for the separation of the effects of each component. The design could also be enhanced by a larger, more diverse sample and by selecting participants using diagnostic tools (such as the HAM-D).

The results of this study offer preliminary support for the use of this tri-modal intervention that would offer primary care practitioners a low cost, effective treatment for mild depression as an alternative to the use of psychotropic medications and referrals for psychotherapy. The costs and side-effects of these standard therapies had led the women in our study to seek another option. This intervention is simple enough that the instructions can be conveyed in a standard 15-minute appointment and accomplished without direct supervision. Additionally, it would be feasible for women to independently re-initiate the program should they experience recurring or chronic depression. Indeed, our participants were able to easily incorporate all components into their daily lives and maintain a commitment to the program. They reported satisfaction in knowing that their improved mood was due to their own activities, not a prescribed medication. As one participant noted: "I found there was something I could do that was within my own control. I didn't have to go to the doctor to get a prescription. It is such a helpless feeling to not know how to change your moods."

REFERENCES

1. Alpert, J. E., and M. Fava. 1977. Nutrition and depression: The role of folate. Nutrition Reviews 55(5):145-49.
2. Alvidrez, J., and F. Azocar. 1999. Distressed women's clinic patients: Preferences for mental health treatments and perceived obstacles. Gen Hosp Psychiatry 21(5): 340-47.
3. Beal, M. W. 1998. Women's use of complementary and alternative therapies in reproductive health care. J Nurse Midwifery 43(3): 224-34.
4. Beauchemin, K.M., and P. Hays. 1997. Phototherapy is a useful adjunct in the treatment of depressed in-patients. Acta Psychiatr Scand 95:424-27.
5. Benton, D., and R. Cook. 1990. Selenium supplementation improves mood in a double-blind crossover trial. Psychopharmacology 102:549-50.
6. Benton, D., J. Fordy, and J. Haller. 1995. The impact of long-term vitamin supplementation on cognitive functioning. Psychopharmacology (Berl) 117:298-305.
7. Benton, D., R. Griffiths, and J. Haller. 1997. Thiamine supplementation, mood, and cognitive functioning. Psychopharmacology 129:66-71.
8. Blazer, D.G., R.C. Kessler, K.A. McGonagle, and M.S. Swartz. 1994. The prevalence and distribution of major depression in a national community sample: the National Comorbidity Survey. Am J Psychiatry 151:979-86.
9. Blumenthal, J.A., M.A. Babyak, K.A. Moore, W.E. Craighead, S. Herman, P. Khatri, R. Waugh, M.A. Napolitano, L.M., Forman, M. Appelbaum, P.M. Doraiswamy, and K.R. Krishnan. 1999. Effects of exercise training in older patients with major depression. Arch Intern Med 159(19):2349-56.
10. Campbell, S. S., D. Dawson, and M.W. Anderson. 1993. Alleviation of sleep maintenance insomnia with timed exposure to bright light. J AmGeriatr Soc 41(8):829-36.
11. Einon, D. 1997. The influence of ambient light and menstrual status on the moods of a nonclinical population of young women. Psychosomatic Medicine 59:616-19.
12. Espiritu, R. C., D. F. Kripke, S. Ancoli-Israel, M. A. Mowen, W. J. Mason, R. L. Fell, M. R. Klauber, and O. J. Kaplan. 1994. Low illumination experienced by San Diego adults: association with atypical depressive symptoms. Biol Psychiatry 35:403-07.
13. Fordyce, M. W. 1977. Development of a program to increase personal happiness. J Counseling Psychology 24:511-21.
14. Fox, K. R. 1999. The influence of physical activity on mental well-being. Public Health Nutr 2: 411-18.
15. Frank, E., L. L. Carpenter, and D. J. Kupfer. 1988. Sex differences in recurrent depression: are there any that are significant? Am J Psychiatry 145:41-45.
16. Hightower, M. 1997. Effects of exercise participation on menstrual pain and symptoms. Women Health 26(4):15-27.
17. Kornstein, S. G. 1997. Gender differences in depression: implications for treatment. J Clin Psychiatry 58(Suppl 15):12-18.
18. Kripke, D. F. 1998. Light treatment for nonseasonal depression: speed, efficacy, and combined treatment. J Affect Disord 49(2):109-17.
19. Kripke, D.F., D.J. Mullaney, M.R. Klauber, S.C. Risch, and J.C. Gillin. 1992. Controlled trial of bright light for nonseasonal major depressive disorders. Biol Psychiatry 31(2):119-34.
20. Krauchi, K., A. Wirz-Justice, and P. Graw. 1990. The relationship of affective state to dietary preference: winter depression and light therapy as a model. J Affect Disord 20:43-53.
21. Lam, R. W., D. Carter, S. Misri, A. J. Kuan, L. N. Yatham, and A. P. Zis. 1999. A controlled study of light therapy in women with late luteal phase dysphoric disorder. Psychiatry Res 86(3):185-92.
22. Lam, R. W., E. Goldner, L. Solyom, and R. A. Remick. 1994. A controlled study of light therapy for bulimia nervosa. Am J. Psychiatry 151:744-750.
23. Lansdowne, A. T., and S. C. Provost. 1998. Vitamin D3 enhances mood in healthy subjects during winter. Psychopharmacology (Berl) 135:319-23.
24. Lee, T. M., and C. C. Chan. 1998. Vulnerability by sex to seasonal affective disorder. Percept Mot Skills 87:1120-22.
25. Locke, B. A. and P. Putman. 1971. Center for Epidemiological Studies Depression Scale. Washington, DC: Epidemiology and Psychopathology Research Branch, Public Health Service, National Institute of Mental Health.
26. McAuley, E., S. L. Mihalko, and S. M. Bane. 1997. Exercise and self-esteem in middle-aged adults: Multidimensional relationships and physical fitness and self-efficacy influences. J Behav Med 20: 67-83.
27. McGreal, R., and S. Joseph. 1993. The Depression-Happiness Scale. Psychological Reports 73: 1279-82.
28. McNair, D. M., M. Lorr, and L. F. Droppleman. 1971. Manual for the Profile of Mood States. San Diego, CA: Educational and Industrial Testing Service.
29. Molin, J., E. Mellerup, T. Bolwig, T. Scheike, and H. Dam. 1996. The influence of climate on development of winter depression. J Affective Disorders 37:151-55.
30. Monk, M. 1981. Blood pressure awareness and psychological well-being in the Health and Nutrition Examination Survey. Clinical Investigations in Medicine 4:183-89.
31. Moses, J., A. Steptoe, A. Mathews, and S. Edwards. 1989. The effects of exercise training on mental well-being in the normal population: A controlled trial. J Psychosom Res 33:47-61.
32. Olfson, M., E. Broadhead, M. M. Weissman, A. C. Leon, L. Farber, C. Hoven, and R. Kathol. 1996. Subthreshold psychiatric symptoms in a primary care group practice. Arch Gen Psychiatry 53:880-6.
33. Parry, B.L., A. M. Mahan, N. Mostofi, M. R. Klauber, G.S. Lew, and J.C. Gillin. 1993. Light therapy of late luteal phase dysphoric disorder: an extended study. Am J Psychiatry 150(9):1417-19.
34. Partonen, T., S. Leppamaki, J. Hurme, and J. Lonnqvist. 1998. Randomized trial of physical exercise alone or combined with bright light on mood and health-related quality of life. Psychol Med 28:1359-64.
35. Prochaska, J. O., J. C. Norcross, and C. C. DiClemente. 1994. Changing for Good. NY: Avon Books.
36. Regier, D. A., J. D. Burke, and K. C. Burke. 1990. Comorbidity of affective and anxiety disorders in the NIMH Epidemiologic Catchment Area Program. In J. D. Maser and C. R. Cloninger, eds., Comorbidity of Mood and Anxiety Disorders. Washington, DC: American Psychiatric Press. pp. 113-122.
37. Rejeski, W. J., L. R. Brawley, and S. A. Shumaker. 1996. Physical activity and health-related quality of life. Exerc Sport Sci Rev 24:71-108.
38. Rosenberg, M. 1965. Society and the adolescent self-image. Princeton, NJ: Princeton University Press.
39. Rosenthal, N. E., D. A. Sack, C. J. Carpenter, B. L. Parry, W. B. Mendelson, and T. A. Wehr. 1985. Antidepressant effects of light in seasonal affective disorder. Am J Psychiatry 142(2):163-70.
40. Schulberg H. C., M. Saul, M. McClelland, M. Ganguli, W. Christy, and R. Frank. 1985. Assessing depression in primary medical and psychiatric practices. Arch Gen Psychiatry 42:1164-70.
41. Simoni-Wastila, L. 1998. Gender and psychotropic drug use. Med Care 36(1):88-94.
42. Slaven, L., and C. Lee. 1997. Mood and symptom reporting among middle-aged women: The relationship between menopausal status, hormone replacement therapy, and exercise participation. Health Psychology 16:203-08.
43. Smith, P. W., W. C. Compton, and W. B. West. 1995. Meditation as an adjunct to a happiness enhancement program. J Clinical Psychology 51:269-73.
44. Stephens, T. 1988. Physical activity and mental health in the United States and Canada: Evidence from four population surveys. Preventive Medicine 17:33-47.
45. Weissman, M. M., D. Sholomskas, M. Pottenger, B. A. Prusoff, and B. Z. Locke. 1977. Assessing depressive symptoms in five psychiatric populations: A validation study. Am J Epidemiology 106(3): 203-13.
46. Williams, J. B., R. L. Spitzer, M. Linzer, K. Kroenke, S. R. Hahn, F. V. deGruy, and A. Lazev. 1995. Gender differences in depression in primary care. Am J Obstet Gynecol 173:654-59.
47. Wirz-Justice, A., P. Graw, K. Krauchi, A. Sarrafzadeh, J. English, J. Arendt, and L. Sand. 1996. Natural light treatment of seasonal affective disorder. J Affective Disorders 37:109-20.
48. Young, M. A., W. A. Scheftner, J. Fawcett, and G. L. Klerman. 1990. Gender differences in the clinical features of unipolar major depressive disorder. J Nerv Ment Dis 178:200-03.

The LEVITY Study as published in Women & Health. All rights reserved by Hayworth press. The authors received special permission to publish the study contents on this web site.

Marie-Annette Brown, Jamie Goldstein-Shirley, and Susan Casey are affiliated with the University of Washington, Seattle, WA. Jo Robinson was study volunteer.

Address correspondence to: Marie Annette Brown, PhD, FNP, University of Washington, Box 357262, Seattle, WA 98195.

This research was supported in part by grants from The Center for Women's Research at the University of Washington (supported by National Institute for Nursing Research, NIH, grant number: P30-NR04001) and Psi Chapter of Sigma Theta Tau, Seattle, WA, a chapter of Sigma Theta Tau International, Indianapolis, IN. The vitamins and placebos used in the study were supplied by Designing Health, Valencia, CA. The authors gratefully acknowledge the important contributions of the research assistants: Crista Langston, ARNP, Kathy Pearce, ARNP, Vernetta Stewart, ARNP and Ann Frolich, ARNP.

Article copies available for a fee from The Haworth Document Delivery Service:
Phone: 1-800-342-9678.
E-mail address: <getinfo@haworthpressinc.com>
Website: <http://www.HaworthPress.com>
2001 by The Haworth Press, Inc. All rights reserved.]