Severity of Dyspepsia Assessment (SODA)

The SODA is a 17-item primary outcome measure for dyspepsia-related health. The three dimensions which the scale measures are: satisfaction, pain intensity, and non-pain symptoms associated with dyspepsia. It was designed for use in large, randomized, controlled clinical trials. (Ref: 2)

Free (public domain)

10 Minutes (Ref: 3)

2001

Flesch-Kincaid Grade Level 7.2. Items are approximately 25 words or less and contain compound sentences. Technical terms such as "abdominal discomfort" are explained (e.g. "stomachache").

Combination of select-response and visual analog scale

Self-report

Scores are summed and converted using scale conversion table.

No other forms exist.

Linda Rabeneck, MD, MPH, Karon F. Cook, PhD, Kimberly Wristers, PhD, Julianne Souchek, PhD, Terri Menke, PhD and Nelda P. Wray, MD, MPH

Houston Center for Quality of Care and Utilization Studies
Department of Veterans Affairs (152)
2002 Holcombe Blvd Houston, Texas 77030

lrabeneck@swchsc.on.ca

public.bcm.tmc.edu

top

The SODA includes three factors: Pain Intensity, Non-Pain Symptoms, and Satisfaction with Dyspepsia-Related Health. Confirmatory factor analyses provided evidence for the uni-dimensionality of each scale. Data from the six Pain Intensity items fit the single pain factor model well; X2(8) = 6.6, P=.58, RMSEA=.00. Data from the seven Non-Pain Symptoms items fit the single nonpain factor model well, X2(12)=15.7, P=.21, RMSEA = 0.6. Data from the four Satisfaction with Dyspepsia-Related Health items also fit the single satisfaction factor model well, X2(2) =1.4, P=.51, RMSEA =.00 (Ref: 2)

In a sample of 3,907 non-VA arthritis patients with a mean age of 60.2 years (Ref: 5), the goodness of fit index (GFI) (for the six Pain Intensity items was 0.99, AGFI was 0.98, and RMSEA was 0.09. For the seven Non-Pain Symptoms items, GFI was 0.99, AGFI was 0.98, and RMSEA was 0.05. For the four Satisfaction with Dyspepsia-Related Health items, GFI was 0.97, AFGI was 0.94, and RMSEA was 0.04. (Ref: 5)

No information found.

Patients(n=3,907) with rheumatoid arthritis (RA) and osteoarthritis (OA) were randomized to celecoxib 800 mg daily or diclofenac 150 mg daily. Test-retest was assessed using data from baseline and four weeks. Reproducibility was fair to good: Pain Intensity (ICC=0.49); Non Pain Symptoms (ICC=0.61); and Satisfaction (ICC=0.45). (Ref: 5)

Not applicable; all measures are based on self-report.

Data from 98 VA patients with uninvestigated dyspepsia demonstrated excellent internal consistency estimates with median Cronbach’s alphas over 8 measurement occasions in one year of: Pain Intensity: 0.97; Non-Pain Symptoms: 0.90; Satisfaction with Dyspepsia-Related Health: 0.92. (Ref: 2)

Data from 3,907 non-VA rheumatoid arthritis (RA) and osteoarthritis patients at baseline demonstrated excellent internal consistency estimates with median Cronbach’s alphas over 8 measurement occasions in one year of: Pain Intensity: 0.93; Non-Pain Symptoms: 0.82; Satisfaction with Dyspepsia-Related Health: 0.89. (Ref: 5)

Coefficient alpha values for the 3 scales based on a sample of 126 patients with dypepsia who presented for treatment at a VA outpatient clinic and a family physician’s private office (mean age = 44 years, non-Hispanic Caucasians = 59%) were as follows: Pain Intensity: 0.8; Non-Pain Symptoms: 0.84; Satisfaction with Dyspepsia-Related Health: 0.74. (Ref: 1)

Discriminant Validity Hits* from this Sample Showed: Pain Intensity: 24/24; Non-Pain Symptoms: 84/84; Satisfaction with Dyspepsia-Related Health: 24/24 (Ref: 1)

(*Numerator refers to the number of items comprising a scale correlated more highly with their intended scale (corrected for overlap) than with one of the other scales; the denominator represents the total number of comparisons involving items comprising the scale)

None.

Adjusted mean SODA change scores were estimated according to adverse event severity and withdrawal status. Using withdrawal due to dyspepsia as the criterion, discriminant validity was evaluated by the area under the curve (AUC) from receiver operating characteristic (ROC) curves. ROC curves were generated for multiple SODA change scores (4-week score minus the baseline score). The AUC was highest for the SODA Pain Intensity scale (0.78), followed by the Satisfaction (0.75) and Non-Pain Symptoms (0.74) scales. (Ref: 5)

In 3,907 arthritic patients, an external criterion was defined as a patient with a dyspepsia adverse event based on a combined set of WHOART (World Health Organization Adverse Reaction Terminology) terms, similar to the concepts measured by SODA (such as abdominal pain, dyspepsia, nausea, etc). The ability of SODA to measure changes in dyspepsia health status was evaluated by comparing SODA change scores with dyspepsia adverse event severity and withdrawal status. SODA change scores increased with increasing severity of dyspepsia adverse events (p for trend of <0.001). Adjusted mean change scores for Pain Intensity, Non-Pain symptoms, and Satisfaction with Dyspepsia-Related Health discriminated from the others; they worsened with increasing dyspepsia adverse event severity and discriminated among adjacent levels at p<0.05), with only one exception (moderate and severe levels for Non-Pain Symptoms, p=0.11). The mean change scores were significantly worse for those who withdrew because of a dyspepsia adverse event. All comparisons, by withdrawal status, were statistically significant (p<0.001). (Ref: 5)

Data from 98 VA patients with uninvestigated dyspepsia demonstrated excellent internal consistency estimates with median Cronbach’s alphas over 8 measurement occasions in one year. The mean change scores (enrollment score minus follow-up score) for patients who stated they were “improved” at the one-week follow-up visit were as follows:

Sixteen percent of the SODA Satisfaction with Dyspepsia-Related Health scores fell below the effective measurement range of the scale, indicating that this measure does not discriminate well among patients with low levels of satisfaction.

Development of items using data from 126 patients at the Houston VA General Medicine outpatient clinic and from the nearby private practice of a board-certified family physician. Patients were at least 18 years of age, English speaking, and reported at least a week-long history of dyspepsia. (Ref: 1) Items were reviewed by a panel of 14 gastroenterologists for content validity. (Ref: 3)

Change scores for all three scales statistically different for patients who reported that their abdominal pain was improved compared those who reported it was unchanged over six weeks. (Ref: 2)

Developed using VA patients and tested with VA patients. (Ref: 1-4, 6-8)

Psychometric testing with non-VA patients who have rheumatoid and osteoarthritis. (Ref 5)