Validating the Patient‐Reported Outcomes Measurement Information System Short Form v1.0–Anxiety 8a in a sample of young adult cancer survivors: Comparison with a structured clinical diagnostic interview

Abstract

Background The objective of this study was to validate the Patient‐Reported Outcomes Measurement Information System Short Form v1.0–Anxiety 8a (PROMIS‐A‐SF) against a structured diagnostic interview in order to determine its accuracy and the most appropriate cutoff score for identifying anxiety disorders in young adult cancer survivors (YACSs). Methods Two hundred forty‐nine YACSs aged 18 to 40 years (50% male) completed the PROMIS‐A‐SF and the Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (SCID). The SCID was used to determine whether participants met the criteria for an anxiety diagnosis. Receiving operator characteristic analyses were performed to determine the concordance of the PROMIS‐A‐SF and the SCID as well as cutoff scores with sensitivity (≥0.85) and specificity (≥0.75) appropriate for an anxiety screening measure. Results The PROMIS‐A‐SF was determined to have good overall discrimination in comparison with the SCID (area under the curve, 0.84). A PROMIS‐A‐SF t‐score cutoff ≥ 53.2 (total predictive value, 67.9%) came closest to meeting the study criteria with a sensitivity of 88%, but the specificity was only 65%. In a hypothetical screening example, this cutoff led to moderate levels of missed cases (14%) and a significant proportion of clinical referrals that were unnecessary by SCID criteria (35%). Of the survivors referred for services according to these criteria, less than one‐third (29%) would have a SCID anxiety diagnosis. Conclusions The PROMIS‐A‐SF demonstrated moderately strong concordance with anxiety disorders measured by the SCID, but cutoff scores did not meet study criteria for clinical screening. Although it may not be appropriate as a standalone screener, the PROMIS‐A‐SF may be useful for assessing anxiety in YACSs when it is incorporated into clinical practice or when it is combined with other measures.

Reported Mental Health, Diet, and Physical Activity in Young Adult Cancer Survivors

“…Previous research has found increased levels of distress among YA cancer survivors compared to age-, sex-, and education-matched controls [ 45 ]. On the other hand, mean anxiety and depression scores in our study were low compared to existing thresholds that have been previously assessed in YA cancer survivors [ 46 , 47 ].…”

Screening young adult cancer survivors for depression and anxiety with the Primary Care Evaluation of Mental Disorders (PRIME‐MD): Comparisons with a structured clinical diagnostic interview

“…Future research should examine the utility of existing measures (e.g., GAD-7 33 ) to determine if any are more effective as a second step in screening than the PRIME-MD anxiety diagnosis threshold. Although a single measure would be preferable, the lack of consensus on a brief validated screening measure for anxiety in YACS 4,5,14 means that a two-step screening process using the PRIME-MD could be useful until an appropriate single measure has been established.…”

“…Participants were a convenience cohort of survivors recruited from a single cancer center to participate in E-Quest, a study examining screening methods for depression and anxiety in YACS. As previously described, [12][13][14][15] individuals were eligible if they were 18-40 years old, English-speaking, at least 3 years from their first cancer diagnosis, and at least 2 years since completion of cancer treatment. As has been previously described, 16,17 cancer types, treatments, and treatment settings vary significantly across the age span from childhood through young adulthood.…”

“…The SCID 22 is a semi-structured interview widely used to assign psychiatric diagnoses according to Diagnostic and Statistical Manual of Mental Disorders 4th edition (DSM-IV) criteria. 23 As previously described, 13,14…”

Parental Distress in Pediatric Inflammatory Bowel Diseases: Associations With Time From Diagnosis, Disease Activity, and Demographic Factors

“…Parental self-reported anxiety was measured using the Patient Reported Outcomes Measurement Information System Short Form v1.0-Anxiety 8A (PROMIS-Anx), an 8-item self-report measure of anxiety symptoms experienced over the previous 7 days. Raw scores are converted into t -scores on the basis of published PROMIS scoring guidelines, 36 where t -scores 0–55 indicated little concern for anxiety, 55–60 indicated mild symptoms of anxiety, 61–70 moderate clinical concern for anxiety, and scores 70 or above suggested clinical concern for anxiety disorder. 37 Previous research has shown that the PROMIS-Anx has strong psychometric properties including internal validity (Cronbach’s α = 0.93–0.95), 38 , 39 reliability (Intraclass Correlation Coefficient [ICC] = 0.79), 39 and relatively invariance across ethnically diverse subgroups.…”