Irma Ahmad1, Matilda Sandberg1, Ole Brus2, Carl Johan Ekman3, Åsa Hammar4,5, Mikael Landén6,7, Johan Lundberg3, Pia Nordanskog8,9, Lars von Knorring10, Axel Nordenskjöld11. 1. School of Medical Sciences, Örebro University, Örebro, Sweden. 2. Clinical Epidemiology and Biostatistics, School of Medical Sciences, Örebro University, Örebro, Sweden. 3. Department of Clinical Neuroscience, Centre for Psychiatry Research, Karolinska Institutet, and Stockholm Health Care Services, Stockholm, Sweden. 4. Department of Biological and Medical Psychology, University of Bergen, Bergen, Norway. 5. Division of Psychiatry, Haukeland University Hospital, Bergen, Norway. 6. Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden. 7. Institute of Neuroscience and Physiology, The Sahlgrenska Academy at Gothenburg University, Gothenburg, Sweden. 8. Department of Clinical and Experimental Medicine, Center for Social and Affective Neuroscience, Faculty of Health Sciences, Linköping University, Linköping, Sweden. 9. Department of Psychiatry, Region Östergötland, Linköping, Sweden. 10. Department of Neuroscience, Psychiatry, Uppsala University, Uppsala, Sweden. 11. Faculty of Medicine and Health, University Health Care Research Centre, Örebro University, Örebro, Sweden.
Abstract
BACKGROUND: The Swedish national quality register for electroconvulsive therapy (Q-ECT) contains data on patients receiving treatment with electroconvulsive therapy (ECT) in Sweden. AIM: This study determined the validity of diagnoses, treatment dates, and rating scales in the Q-ECT by investigating the degree of accordance between data from the Q-ECT and patient records. MATERIALS AND METHODS: From January 2016 to December 2017, 200 treatment series were randomly selected from the Q-ECT. The corresponding patient records were requested from the treating hospitals. Data on the indicative diagnosis, dates for the first and the last ECT session, and rating scales were compared between the Q-ECT and patient records using (i) a strict and (ii) a liberal method of assessment. Using the liberal method, each variable was assessed as accordant if it belonged to the same diagnosis group, or if the dates differed by less than 1 week, or ratings differed by only 1 point on the Clinical Global Impression Scale (CGI- S), or no more than 3 points on the Montgomery Åsberg Depression Rating Scale between the Q-ECT and the patient record. RESULTS: A total of 179 patient records were received. The strict method of assessment showed an accordance of 89% or higher for all studied variables. The liberal method showed an accordance of 95% or higher. CONCLUSIONS: We conclude that data on the studied variables in the Q-ECT have high validity. However, limited use of some rating scales makes the results uncertain. Measures can be taken to further improve the data quality.
BACKGROUND: The Swedish national quality register for electroconvulsive therapy (Q-ECT) contains data on patients receiving treatment with electroconvulsive therapy (ECT) in Sweden. AIM: This study determined the validity of diagnoses, treatment dates, and rating scales in the Q-ECT by investigating the degree of accordance between data from the Q-ECT and patient records. MATERIALS AND METHODS: From January 2016 to December 2017, 200 treatment series were randomly selected from the Q-ECT. The corresponding patient records were requested from the treating hospitals. Data on the indicative diagnosis, dates for the first and the last ECT session, and rating scales were compared between the Q-ECT and patient records using (i) a strict and (ii) a liberal method of assessment. Using the liberal method, each variable was assessed as accordant if it belonged to the same diagnosis group, or if the dates differed by less than 1 week, or ratings differed by only 1 point on the Clinical Global Impression Scale (CGI- S), or no more than 3 points on the Montgomery Åsberg Depression Rating Scale between the Q-ECT and the patient record. RESULTS: A total of 179 patient records were received. The strict method of assessment showed an accordance of 89% or higher for all studied variables. The liberal method showed an accordance of 95% or higher. CONCLUSIONS: We conclude that data on the studied variables in the Q-ECT have high validity. However, limited use of some rating scales makes the results uncertain. Measures can be taken to further improve the data quality.
Entities:
Keywords:
Electroconvulsive therapy; data accuracy; registries