Skip to main content

Data Sets

Below are data sets housed across multiple NIAMS-Funded Core Centers for Clinical Research.

These descriptions are intended to inform researchers about data sets relevant to various rheumatic and musculoskeletal conditions. Researchers who are interested in any of these data are encouraged to contact the investigators listed below to find out more about potential opportunities for collaboration or data analyses.

UNC CCCR – Methodology Core Data Sets

#1 of 6
MUSC CCCR cohort

  • Brief description of the data set:
    • Inclusion criteria for participants
      • Patients meeting criteria for the diagnosis of lupus or scleroderma, matched healthy controls and limited numbers of first degree relatives
    • Number of participants
      • Approximately 750 lupus participants and 300 scleroderma participants.  Lupus patients are 65% Black while the Scleroderma patients are ?30% Black
    • Follow-up time points
      • Patients are followed differently depending on their disease and circumstance.  Some we have every three month data, some every six months and some once a year.  Controls and FDRs are a one time visit other than those participating in Dr. Kamens Gullah cohort some of whom do have chronologic data.  The collection began in 2006
    • General types of data available
      • Demographics, disease activity measures etc
    • Are biospecimens available?
      • Biospecimens are available on most subjects from most visits
      • We also continue to maintain the biobank of the Carolina Lupus Study which included approximately 300 newly diagnosed patients and 285 controls from North and South Carolina.  These samples were collected in the late 1990s, early 2000s
      • Dr. Tsao has her individual biobank of lupus and control patients DNA with thousands of individuals of multiple ethnicities and race
      • Dr. Carol Feghali Bostwick has a biobank of a variety of skin fibroblasts, lung tissue etc from scleroderma patients
      • Dr. Bostwicks and Dr. Tsaos samples are under their control so access to these samples will require personal approval from them
  • Contact person (name and email address): Gary Gilkeson, MD  gilkeson@musc.edu

#2 of 6
Physical Therapy vs. Internet-Based Exercise Training for Knee Osteoarthritis (PATH-IN)

  • Brief description of the data set:
    • Inclusion criteria for participants
      • Diagnosis of knee osteoarthritis (OA) & current joint symptoms
    • Number of participants
      • 350 patients with symptomatic knee OA
      • Randomized (2:2:2) to physical therapy, internet-based exercise training & usual care /wait list control
    • Follow-up time points
      • 4-months & 12-months following baseline assessment
    • General types of data available
      • Demographic characteristics
      • Self-reported pain and function
      • Physical performance test results
      • Clinical and psychosocial variables
      • Physical activity
      • Patient global assessment of change
    • Are biospecimens available?
      • No
  • Contact person (name and email address): Kelli Allen: kdallen@email.unc.edu

#3 of 6
Back pain Outcomes using Longitudinal Data (BOLD)

  • Brief description of the data set(s), including the following as appropriate: Cohort of patients aged 65 years and older who initiated a new episode of care for back pain.
    • Inclusion criteria for participants (e.g., patients with symptomatic knee OA; patients with early RA).
      • Persons 65 years or older were prospectively identified from March 2011-March 2013 when they had a primary care visit for back pain at 3 healthcare systems: Harvard Vanguard (Boston, MA); Henry Ford Health System (Detroit, MI); Kaiser-Permanente Northern California. Patients were excluded if they had a prior back pain visit in the 6 months prior to index; had a major medical comorbidity, did not speak English, had prior back surgery, or failed the cognitive assessment.
    • Number of participants.
      • The total cohort consisted of n=5239 patients. However, most analyses exclude n=126 people for not having EMR data (n=84); having history of cancer (n=37); having lumbar surgery in the year prior to index (n=5).
    • Follow-up time points.
      • Patient-reported outcomes assessed at 3, 6, 12, 24 months. EMR data obtained from 1 year prior to 2 years after index back pain visit.
    • General types of data available (e.g., demographic characteristics, joint symptoms, disease severity, EHR data…).
      • Comorbidity scores (Quan, Charlson, Elixhauser),
      • Inpatient data (day of admission relative to index; year of admission; length of stay; diagnosis related group (DRG) code; primary ICD-9 code; secondary ICD-9 codes; procedure codes (CPT and ICD-9); relative value units; cost; spine-relatedness)
      • Outpatient procedure data (day of procedure relative to index date; year of procedure; CPT code; ICD-9 code associated with procedure (NOTE this is not available for patients at Kaiser-Permanente Northern California); provider performing procedure’s specialty (NOTE this is not available for patients at Henry Ford Health System); RVU; cost; spinerelatedness)
      • Patient insurance (primary and secondary and at 3 timepoints: 12 months prior to index; at index; 12 months after index)
      • Lab tests and results
      • Prescription orders (National drug code; prescription day relative to index; drug name; days’ supply; prescription text; quantity; dose)
      • Prescription fills (National drug code; prescription day relative to index; drug name; days’ supply; prescription text; quantity; dose)
      • Vital signs (height, weight, blood pressure, date relative to index)
      • Demographics (age, race, ethnicity, sex, marital status, education, employment status, smoking status, death)
      • Patient reported outcomes (at baseline and 3, 6, 12, 24 months)
        • Brief pain inventory
        • Duration of pain
        • Euro-QOL 5D
        • Whether the patient experienced falls in the 3 weeks prior and whether they were injurious
        • Expectation of recovery from back pain (assessed at baseline only)
        • Back pain numeric rating scale (NRS)
        • Leg pain numeric rating scale (NRS)
        • Patient health questionnaire (PHQ4)
        • Roland Morris Disability Questionnaire
    • Are biospecimens available? No.
  • Contact person
    • Sandra K Johnston, PhD, RN/Jeffrey G Jarvik, MD, MPH
    • Executive Administrative Director- UW CLEAR Center (theclearcenter.org)
    • stonesk@uw.edu

#4 of 6
Lumbar Epidural steroid injections for Spinal Stenosis (LESS).

  • Brief description of the data set: Double-blind randomized control trial of epidural steroid injections (ESIs) plus local anesthetic injections versus local anesthetic injections for treatment of pain associated with lumbar spinal stenosis
    • Inclusion criteria for participants (e.g., patients with symptomatic knee OA; patients with early RA). Adults must have been age 50 or older; had pain in the lower back, buttock or lower extremity with a numeric rating scale >4; had a Roland-Morris score of at least 7; had mild/moderate/severe lumbar central canal spinal stenosis identified on MRI or CT imaging; had lower extremity symptoms consistent with neurogenic claudication; and have been proficient in English.
    • Number of participants: 400 participants. 200 randomized to each arm (epidural steroid injections (ESIs) plus local anesthetic injections versus local anesthetic injections) of the study.
    • Follow-up time points. Baseline (day of injection); 3 weeks; 6 weeks; 3 months; 6 months; 12 months; 24 months.
    • General types of data available: Comorbidity scores (Quan, Charlson, Elixhauser),
    • • Inpatient data (day of admission relative to index; year of admission; length of stay; diagnosis related group (DRG) code; primary ICD-9 code; secondary ICD-9 codes; procedure codes (CPT and ICD-9); relative value units; cost; spinerelatedness)
      • Outpatient procedure data (day of procedure relative to index date; year of procedure; CPT code; ICD-9 code associated with procedure (NOTE this is not available for patients at Kaiser-Permanente Northern California); provider performing procedure’s specialty (NOTE this is not available for patients at Henry Ford Health System); RVU; cost; spine-relatedness)
      • Patient insurance (primary and secondary and at 3 timepoints: 12 months prior to index; at index; 12 months after index)
      • Lab tests and results
      • Prescription orders (National drug code; prescription day relative to index; drug name; days’ supply; prescription text; quantity; dose)
      • Prescription fills (National drug code; prescription day relative to index; drug name; days’ supply; prescription text; quantity; dose)
      • Vital signs (height, weight, blood pressure, date relative to index)
      • Demographics (age, race, ethnicity, sex, marital status, education, employment status, smoking status, death)
    • Patient reported outcomes (at baseline and 3, 6, 12, 18, 24 months)
      • Brief pain inventory
      • Duration of pain
      • Euro-QOL 5D
      • Whether the patient experienced falls in the 3 weeks prior and whether they were injurious
      • Expectation of recovery from back pain (assessed at baseline only)
      • Fear avoidance
      • Generalized Anxiety Disorder
      • Medication diaries
      • Pain catastrophizing
      • Back pain numeric rating scale (NRS)
      • Leg pain numeric rating scale (NRS)
      • Patient health questionnaire (PHQ4)
      • Roland Morris Disability Questionnaire
      • Swiss Spinal Stenosis Questionnaire

o Are biospecimens available? No
o Contact person
•Sandra K Johnston, PhD, RN/Jeffrey G Jarvik, MD, MPH
• Executive Administrative Director- UW CLEAR Center (theclearcenter.org)
• stonesk@uw.edu

#5 of 6

The Johnston County Osteoarthritis Project

  • Brief description of the data set(s), including the following as appropriate:
    • Inclusion criteria for participants (e.g., patients with symptomatic knee OA; patients with early RA)—population-based sampled from Johnston County, NC, at least 45 years of age, self-identify as Black or White, capable of completing study protocol (no requirement or exclusions for symptoms or disease).
    •  Number of participants
      • More than 4000 unique individuals enrolled at baseline (1991-7) or cohort enrichment (2003-4)
    • Follow-up time points
      • 30 years of follow-up, approximately every 5 years
    • General types of data available
      • Pain, function, radiographs at multiple joint sites, performance-based function, medications, injury
    • Are biospecimens available?
      • Blood and Urine
  • Contact person (name and email address): Amanda Nelson aenelson@med.unc.edu

#6 of 6

The Johnston County Health Study 

  • Brief description of the data set(s), including the following as appropriate:
    • Inclusion criteria for participants (e.g., patients with symptomatic knee OA; patients with early RA)– population-based sampled from Johnston County, NC, 35-70 years of age at baseline, self-identify as Black, White or Hispanic, capable of completing study protocol (no requirement or exclusions for symptoms or disease).
    • Number of participants
      • 500 to date
    • Follow-up time points
      • Baseline only (2019-present enrollment)
    • General types of data available
      • Pain, function, radiographs at multiple joint sites; performance-based function; medications; medical conditions; injury; knee ultrasound; cognitive function; acculturation
    • Are biospecimens available?
      • Blood and Urine
  • Contact person (name and email address): Amanda Nelson aenelson@med.unc.edu