Depression and anxiety screening in emergency department patients with recurrent abdominal pain: An evidence synthesis for a clinical practice guideline

Abstract

Background Recurrent abdominal pain in the emergency department (ED) might represent an opportunity for screening of depression and/or anxiety. Methods We systematically searched five databases for studies evaluating the effect of screening for depression and/or anxiety in ED patients with recurrent and undifferentiated abdominal pain. Given paucity of direct evidence, we also searched for indirect evidence including studies that assessed prevalence of depression and/or anxiety in EDs (not necessarily recurrent abdominal pain), diagnostic accuracy of screening tools, effectiveness of screening in other settings, and outcomes such as repeat ED visits of patients with abdominal pain who were screened in the ED. Two methodologists evaluated certainty in the evidence using the GRADE approach. Results A total of 4,337 citations were reviewed, and zero studies were found on the effect of screening in patients with recurrent and undifferentiated abdominal pain in the ED. A total of 35 studies were included as relevant indirect evidence. In studies of ED patients with abdominal pain, depression ranged from 10% to 29%, while anxiety ranged from 18% to 50%. False positives appear to be an issue given relatively low specificity of screening tools. One randomized trial including ED patients with vague symptoms evaluated the effect of depression screening on a composite outcome of depression recognition, psychiatric consultation, or referral by the emergency physician (risk ratio = 1.49, 95% confidence interval [CI] = 0.49 to 4.53, very low certainty). One study reported that patients with undifferentiated abdominal pain who screened positive for depression have had increased ED recidivism (odds ratio = 3.17, 95% CI = 1.14 to 8.85, very low certainty). Conclusions We were unable to identify any evidence that confirms that depression or anxiety screening in ED patients with recurrent and undifferentiated abdominal pain improves outcomes or changes management downstream.

Guidelines for Reasonable and Appropriate Care in the Emergency Department 2 (GRACE‐2): Low‐risk, recurrent abdominal pain in the emergency department

“…Our scoping review provides details of the search strategy and inclusion processes 142 . We found no direct evidence on the effect of depression or anxiety screening strategies compared to usual care without screening in adult patients with recurrent and undifferentiated abdominal pain in the ED.…”

“…In ED populations, depression was reported in 8%–55% of patients (median 25%) across 19 studies, while anxiety ranged from 9% to 74% (median 27%) across 14 studies 142 . The wide‐ranging prevalence estimates are partly explained by varying definitions, different diagnostic instruments, and heterogeneous ED populations.…”

“…Using the data above to estimate the pretest probability of depression and/or anxiety in our population of interest, we evaluated the diagnostic accuracy of pragmatic and short instruments for screening of depression and anxiety to assess potential misclassifications. For depression screening, we found one high‐quality systematic review and individual patient meta‐analysis evaluating the diagnostic accuracy of the Patient Health Questionnaire‐2 (PHQ‐2) and PHQ‐9 using the reference standard of semistructured interviews (44 studies, 10,627 patients, various settings including both outpatient and inpatient) 142,149 . The meta‐analysis did not indicate whether ED studies were included.…”

“…The GRADE assessment for the use of PHQ‐2 of ≥2 in our population of interest is detailed in Table 5. All other GRADE summary of findings tables, considering other possibilities of pretest probability and evaluation of certainty in the evidence, are available in the supplementary material of our evidence synthesis 142 . Our certainty in the evidence when applying these estimates to our population of interest was very low due to risk of bias, indirectness, and inconsistency.…”

“…Pictogram on the number of false positives (red) and false negatives (black) using different approaches for depression screening with an estimated pretest probability (prevalence) of 25%. From Oliveira J. e Silva et al, reproduced with permission 142 …”