Chronic pain, by definition, persists over long periods of time, but both sufferers of pain and their caregivers are aware that pain levels can fluctuate greatly over time. Being able to define pain variability and its underlying causes would be an important advance in our understanding of chronic pain, and would in turn lead to better approaches to the treatment of chronic pain. In this study, the investigators propose to evaluate in 60 people with chronic pain of the knee due to osteoarthritis over 3 months, collecting data regarding magnitude of pain, physical activity level, and medication status twice daily during this period of time. The investigators then will assess the variability in pain levels over time, the frequency of "flares" of osteoarthritis (OA) pain and their relationship to physical activity, medications, and clinical descriptors of the population, e.g., age, gender, body mass index (BMI). Additionally, evaluations will be made regarding temporal patterns of OA knee pain as well as assessment of subgroups of patients with the overall group being studied.

Inclusion Criteria:

1. Men and women, age 40 years and above
2. Knee pain most days of the week for the past month
3. Diagnosis of knee OA
4. Meet ACR criteria for knee OA
5. No significant limitations in physical activity
6. Own and able to use a cellular phone capable of text messaging
7. Able to return for all clinic visits
8. Able to read and understand the informed consent document

Exclusion Criteria:

1. Use of a walker to ambulate or inability to ambulate (use of cane is allowed)
2. Other forms of arthritis
3. Other major causes of pain that could be expected to interfere with assessment of pain during this trial, e.g., recurrent migraine, back pain, fibromyalgia
4. Scheduled for and likely to need joint replacement surgery in the next 3 months
5. Any medical condition that in the judgment of the investigator would make the participant not suitable for the study
6. Living in a long-term care facility

Individuals with OA