The Thai version of the nursing delirium screening scale-Thai: Adaptation and validation study in postoperative patients.

BACKGROUND

The Nursing Delirium Screening Scale (Nu-DESC) is an effective instrument for assessing postoperative delirium (POD). This study translated the Nu-DESC into Thai ("Nu-DESC-Thai"), validated it, and compared its accuracy with the Diagnostic and Statistical Manual of Mental Disorders-5 (DSM-5).

METHODS

The translation process followed the International Society for Pharmacoeconomics Outcome Research guidelines. Recruited participants were ≥ 70 years old, fluent in Thai, and scheduled for surgery. The exclusion criteria were cancellation or postponement of an operation, severe visual or auditory impairment, and patients with a Richmond Agitation Sedation Scale score of -4 or less before delirium assessment. Post-anesthesia care unit (PACU) nurses and residents on wards each used the Nu-DESC to assess delirium in 70 participants (i.e., 140 assessments) after the operation and after patient arrival at wards, respectively. Geriatricians confirmed the diagnoses using video observations and direct patient contact.

RESULTS

The participants' mean age was 76.5 ± 4.6 years. The sensitivity and specificity of the Nu-DESC-Thai at a threshold of ≥ 2 were 55% (95% CI, 31.5-76.9%) and 90.8% (84.2-95.3%), respectively, with an area under a receiver operating characteristic curve (AUC) of 0.73. At a threshold of ≥ 1, the sensitivity and specificity were 85% (62.1-96.8%) and 71.7% (62.7-79.5%), respectively (AUC, 0.78). Adding 1 point for failing backward-digit counting (30-1) to the Nu-DESC-Thai and screening at a threshold of ≥ 2 increased its sensitivity to 85% (62.1-96.8%) with the same specificity of 90.8% (84.2-95.3%).

CONCLUSION

The Nu-DESC-Thai showed good validity and reliability for postoperative use. Its sensitivity was inadequate at a cutoff ≥ 2. However, the sensitivity improved when the threshold was ≥ 1 or with the addition of backward counting to Nu-DESC-Thai and screening at a threshold of ≥ 2.