The purpose of this study is to determine whether Cymbalta (duloxetine) is effective to improve cognition in individuals with Mild Cognitive Impairment.

The goal of this study is to conduct a proof of concept clinical trial using antidepressant therapy to improve cognition. This study will utilize a randomized, double blinded, placebo - controlled trial design. The investigators will recruit up to 100 patients. Patients will be screened multiple times to determine eligibility. Patients will be randomized into one of two study groups (placebo control group and Duloxetine group). Patients in the Duloxetine group will receive up to 60 mg of an FDA approved antidepressant, Duloxetine. Patients in the control group will receive a placebo that is the exact shape and size of the study drug. The patients, primary investigators, and research personnel will be blinded to the study condition. The investigators will designate one on-site personnel to serve as data and safety monitor. This person will not be blinded and will randomize patients to condition, work with the pharmacy, and can un-blind condition if necessary.

Potential patients will be invited to a screening visit. This visit is designed to ensure that the patients meet study criteria and that it is safe for them to participate in the study. The screening visit will consist of physical examinations, medical and psychological history, cognitive and functional testing, interviews, questionnaires, research/clinical venipuncture, and a meeting with the study doctor. If a patient is accepted to the trial, he/she will be expected to stay in the study for a minimum of 6 months.

After the screening visit, the study team will meet to determine if the patient will continue to remain in the study. If a patient does remain in the study, he/she will be invited to a randomization visit. Patients will be randomized into either the Duloxetine group or the placebo control group. The data and safety monitor will use a randomization software. This program allows the researcher to enter in the number of subject, and condition, and will generate a table of randomly assigned patients to condition. The patients and study personnel will be blinded to the condition. The data safety monitor will randomly assign patients to condition and work with the pharmacy to properly label the study drugs. Only the data monitoring personnel will be un-blinded and will generate the table of random numbers. After randomization, patients will undergo vitals, interviews, testing, and the study drug will be dispensed to them. First dosage will consist of 30 mg of Duloxetine or placebo, and the participant will be asked to take the first dose at this visit.

Patients will be invited to a 2 week post randomization visit that will consist of a meeting with the study doctor to discuss concerns or side effects. Medication dosage will be raised to 60 mg of Duloxetine or placebo.

Patients will be seen monthly for the next three months. During these visits the patients will receive their study drugs, have their vital signs measured, and be questioned about adverse events or any health changes.

One month later, patients will have a follow up visit. This visit will consist of follow up interview and neuropsychological testing, clinical/research blood draw, and study doctor visit. This will be the final data collection study visit. The investigators will reduce the dosage of the study drug to 30 mg at this visit. The patients will be instructed that the investigators will be weaning off the study drug at this time.

The last study visit will be 2 weeks after the follow-up visit. This visit will consist of a study debriefing with the patient. At this visit the investigators will discontinue the study drug. The investigators will also arrange for the data monitoring personnel to un-blind the study at this time. No data will be collected at this visit. The participants will be informed of whether or not they were on the study drug or placebo. If a patient in the study drug group wishes to remain on Duloxetine, the patients will be advised to discuss it with their personal healthcare provider.

Research data will be stored and managed in a secure manner following NIH guidelines and according to state and institutional policies. Only authorized key personnel shall have access to research related documents. All personnel will be properly trained and supervised regarding the management and handling of confidential materials. The Principal Investigator assumes full responsibility for such training, supervision, and conduct.

All data will be stored in locked file cabinets behind locked doors in the PIs research laboratory until entered into the research database. Computer-based data entry will not require hard copy storage. All data collected via paper-pencil will be double entered into the research database by independent research assistants and results checked for quality control (QC). Once all hard copies have been entered, they will be scanned into PDF files for storage; all hard copies will be shredded. We will maintain the original signed consent forms for our records (these documents will not be shredded and will be kept in a locked file cabinet). Electronic scanned files will be stored in password protected files for security purposes. All discrepancies will be validated by chart review before the data are merged into the larger database. The database will contain item-level data to avoid the need for subsequent data entry processes as potential data analyses arise. Periodic QC checks will be conducted by our IT personnel and provided to the PI. All electronic records will be maintained on password protected computers behind locked doors in the PI's office space. All files will be backed up weekly on an independent external hard drive, which is also password protected.

Inclusion Criteria:

• Male or female, age 50 and up.
• Female participants must be post-menopausal for at least two consecutive years.
• Health and Aging Brain Study participant, who provided consent for re-contact
• Diagnosis of MCI (by Health and Aging Brain Study Consensus Review).
• Has an elevated DepE score (2 or more). This is calculated by summing scores for five items (Items 14, 16, 17,25 &26) on the Geriatric Depression Scale.

Exclusion Criteria:

• Inability to provide informed consent by self or by proxy.
• Pregnant or breast feeding women
• Uncontrolled narrow angle glaucoma
• Known hypersensitivity to duloxetine.
• Participation in a Clinical Trial in the last three months.
• Other psychiatric disorder like bipolar disorder, schizophrenia, or dementia.
• Use of antidepressants, anti-psychotics, and mood stabilizers.
• History of stroke.