The objective of this study is to evaluate the ability of a behavioral intervention, cognitive behavioral therapy for sleep quality (CBT-SQ) to reduce sleep complaints, depression recurrence, and cellular and genomic markers of inflammation in older adults with sleep complaints who have a prior history of depression. The investigators aim to: 1) evaluate the effects of CBT-SQ vs. Sleep Seminar (SS) on objective (actigraphy) and subjective (sleep diary; questionnaire) measures of sleep symptoms over a two-year follow-up; 2) determine the effects of CBT-SQ vs. SS on recurrence of depressive symptoms and depression episode(s) over a two-year follow-up. The investigators will also secondarily examine the effects of CBT-SQ vs. SS on cellular and genomic markers of inflammation over a two-year follow-up, and explore whether markers of inflammation and cytokine genes can explain variability in the risk of depression recurrence in those older adults receiving CBT-SQ vs. SS. The present study is highly significant by being the first study, to the investigators knowledge, to focus on the prevention of depression in community dwelling older adults who have a history of depression, and by targeting sleep disturbance, a modifiable risk factor to prevent depression recurrence.

Depression, one of the most common diseases in older adults, carries significant risk for morbidity, and mortality. However, many older adults with depression are not identified and even when identified, they face protracted courses of treatment, with over 60% of elderly patients failing to achieve symptomatic remission. Given the burgeoning population of older adults, as well as the enormous burden of depression, efforts to maximize depression prevention are needed.

Despite advances in understanding the behavioral pathways that contribute to depression, there has been little attention aimed at targeting behavioral risk factors such as sleep disturbance, even though such strategies have the potential to optimize efficiency (i.e., decrease number needed to treat (NNT) among vulnerable older adults with a history of depression. In this study, we hypothesize that recognition and treatment of sleep disturbance, a modifiable behavioral risk factor, will prevent depression incidence in older adults. Whereas sleep disturbance in depressed patients often lingers and its persistence can represent a residual phase of a major mood disorder, emergence of disturbed sleep in non-depressed older adults serves as an independent risk factor depression that occurs later in life. In a 2-year prospective cohort study of community-dwelling older adults aged 60 years or older (N=351), we have found that sleep disturbance is prospectively associated with depression incidence independent of other current depressive symptoms, as well as antidepressant and hypnotic medication use, and medical status. To evaluate sleep disturbance at the community level, sleep disturbance was defined by a self-report measure (i.e. scores > 5 on the Pittsburgh Sleep Quality Index (PSQI), which highly correlates with insomnia diagnosis. We have found similar prospective results between reported sleep disturbance and depression in adults (N=1716).

Increasing evidence also implicates inflammation as a biological mechanism that contributes to depression, and we further hypothesize that increases in inflammation are associated with the link between sleep disturbance and depression incidence. Whereas multiple other factors including but not limited to, psychosocial stress, medical illness, obesity, sedentary lifestyle, social isolation, low socio-economic status, female sex, and smoking can drive inflammation and are associated with depression, our preliminary data have found that sleep disturbance induces activation of inflammatory signaling,and additional naturalistic and epidemiologic preliminary findings show that sleep disturbance is associated with increases in markers of inflammation especially among those with a history of depression. In turn, we and others have found that inflammation prospectively predicts depression recurrence in community dwelling adults who have a history of depression (N=1716)16 Although there is also evidence that this association can be reciprocal,33 our experimental preliminary data show that inflammatory activation induces depressed mood. In contrast, cognitive behavioral therapy for insomnia (CBT-I)) reduces cellular markers of inflammation in older adults who show a remission of clinical sleep complaints.

The over-arching objectives of this study are to evaluate the ability of CBT-I vs. an active comparator control, Sleep Education Therapy (SET) to reduce sleep complaints, depression incidence, and cellular and genomic markers of inflammation in older adults with self-reported sleep disturbance with follow-up up to three years.

Inclusion Criteria:

• Older adults >= 60 year of age
• Evidence of self-reported sleep disturbance as indexed by PSQI scores >5.

Exclusion Criteria:

Psychiatric Disorders.

1. current major depressive disorder or other DSM-IV psychiatric disorder (e.g. substance dependence) with the exception of a anxiety disorder;

2. presence of psychotic symptoms

3. acute suicidal or violent behavior or history of suicide attempt within the last year;

   Sleep Disorders.

4. current or lifetime history of sleep disorder (sleep apnea, nocturnal myoclonus, phase-shift disorder) as identified by SCID-IV and the Duke Structured Interview for Sleep Disorders (DSISD); persons with co-morbid insomnia will be included

   Medical conditions.

5. severe or acute medical illness (e.g., major surgery, metastatic cancer, stroke, or myocardial infarction) six months prior to study entry presence of co-morbid medical conditions; 5) neurological diseases (e.g., Parkinson's diseases, multiple sclerosis; neurodegenerative dementia,);

6. severe pain disorders requiring daily pain management;

7. presence of co-morbid inflammatory disorders such as rheumatoid arthritis and other autoimmune disorders that would confound the assessment of sleep as well as inflammatory markers;

8. presence of uncontrolled medical condition that is deemed by the investigators to interfere with the proposed study procedures, or put the study participant at undue risk (e.g., an active heart failure categorized as Class III or greater according to New York Heart Association criteria; symptomatic cardiac arrhythmias; symptomatic, hemodynamically significant mitral or aortic valvular disease);

9. presence of chronic infections, which may elevate proinflammatory cytokines. While we will not exclude this individuals only behavioral outcomes will be measured on these subjects. No analysis of blood will be performed. (If a history of an acute infectious illness is identified within two weeks of a scheduled blood sampling, the session will be rescheduled to occur outside this two week period);

   Medication and substance use.

10. use of hormone containing medications including steroids;

11. immune modifying drugs that target specific immune responses agents such as TNF antagonists;

12. daily use of analgesics such as opioids. We will include those taking psychotropic medications (except for anti-psychotic medications or who have have been treated with a depot-neuroleptic within 6 months prior to study entry) In regards to psychotropic medication use, we will quantify past, current, and ongoing use of psychotropic medications (e.g., selective serotonin reuptake inhibitors, other antidepressants, anxiolytics, hypnotics, sedatives) and assess the effects of these medications on primary outcomes during the intervention, and follow-up period.