This study is a pilot study. The investigators have designed a randomized, controlled trial of financial incentives in medication adherence, focusing primarily on poorly-controlled diabetes, and secondarily on hypertension, and high cholesterol. Prior work has shown that many patients do not take their medications as prescribed by their doctors. This contributes to increased rates of bad outcomes such as blindness, kidney failure, heart attack, and death. The investigators hypothesize that use of a financial incentive will motivate patients to improve their medication adherence and ultimately their control of their chronic diseases.

The investigators plan to identify patients who get Primary Care at Boston Medical Center who still have high blood sugars more than a year after their diabetes diagnosis, and randomize them to a control arm, or one of two intervention arms. Subjects will be approached at the time of a regularly-scheduled appointment with their Primary Care doctor and offered the opportunity to participate in the study. All subjects who agree to participate in the study will meet with a Clinical Pharmacologist to review their medications in detail, and then undergo randomization. Subjects in the first intervention group will receive a cash incentive for picking up medications for the targeted conditions from the pharmacy each month. Subjects in the second intervention group will receive a cash incentive for picking up medications for the targeted conditions from the pharmacy each month, and a one-time payment at the conclusion of the study based on the amount of hemoglobin A1c decrease. The investigators will enroll a total of 100 subjects in the study, and anticipate an observational cohort of approximately 1,000 patients.

All patients who are eligible for the study but who are not enrolled in the study and have not declined to participate in the study will become the observational cohort for the study. The observational cohort will be used to determine whether randomization to the control arm of the study has a negative, rather than neutral, effect on patients.

At the end of eight months, all subjects will meet with a Visiting Nurse in their home, to have their blood pressure checked and to have their blood drawn so that their blood sugar and cholesterol can be measured. Outcomes to be evaluated include hemoglobin A1c, lipid panel, systolic blood pressure, diastolic blood pressure, self-reported health, microvascular and macrovascular complications, and death.

Inclusion Criteria:

• Age greater than or equal to 18 years
• Speaks English, Spanish, or Haitian Creole
• Established patient in Boston Medical Center Section of General Internal Medicine Primary Care Practice
• Uses Boston Medical Center Pharmacy
• Diagnosed with diabetes for more than one year
• Prescribed medications for diabetes
• Last hemoglobin A1c > 7.9

Exclusion Criteria:

• Pregnant
• History of Brittle Diabetes
• Meets April, 2012 American Diabetes Association criteria for an increased goal hemoglobin A1c