Inter‐rater and intra‐rater reliability of nursing process records for patients with schizophrenia

Aims. This study explored the inter‐rater and intra‐rater reliability to evaluate the consistency of nursing process records for patients with schizophrenia. Background. By writing accurate and complete nursing process records, nurses can quickly communicate the care that has been delivered. However, little is known about the accuracy of nursing records to reflect the patients’ problems, especially in psychiatry. Design. A prospective observational study. Methods. Two nurses with similar work experience in psychiatric wards assessed patient records produced by 14 psychiatric nurses to compute inter‐rater reliability of nursing diagnoses and their defining characteristics. Collecting the records and the time spans between the first and the second data collection took one month to compute the intra‐reliability of the nursing diagnoses by the same nurse. Results. The greatest intra‐rater consistency was in identifying ‘disturbed thought processes’ (kappa = 0·77, 95% CI: 0·56–0·98). A moderate level of inter‐rater agreement among nurses was observed for the nursing diagnoses of ‘disturbed thought process’ and ‘disturbed sleep pattern’ from 0·41–0·53. Furthermore, the inter‐rater agreement of among nurses with less work experience (less than four years) showed greater higher consistency on ‘disturbed thought process’ (kappa = 0·56, 95% CI: 0·23–0·89) and ‘disturbed sleep pattern’ (kappa = 0·41, 95% CI: 0·07–0·73) than that observed among nurses with more work experience (more than four years). Conclusions. Overall, intra‐rater reliability was greater than inter‐rater reliability for psychiatric nursing process records. Furthermore, more inter‐rater and intra‐rater agreement were observed among records from less experienced nurses than among records produced by more experienced ones. To evaluate the consistency of nursing process records, both the intra‐rater reliability and the inter‐rater reliability show the importance of using standardised terms and more detailed nursing records. Relevance to clinical practice. Our results clearly indicate that using standardised terms to describe patient symptoms and more detailed descriptions of the nursing process could improve the accuracy of nursing records.