Diagnostic data for neurological conditions in interRAI assessments in home care, nursing home and mental health care settings: a validity study
1 School of Public Health & Health Systems, University of Waterloo, Waterloo, ON, Canada
2 Institut universitaire de gériatrie de Montréal, Montreal, QC, Canada
3 Department of Community Health Sciences, University of Calgary, Calgary, AB, Canada
4 Departments of Medicine and Community Health Sciences, University of Manitoba, Winnipeg, MB, Canada
BMC Health Services Research 2013, 13:457 doi:10.1186/1472-6963-13-457Published: 1 November 2013
The interRAI suite of assessment instruments can provide valuable information to support person-specific care planning across the continuum of care. Comprehensive clinical information is collected with these instruments, including disease diagnoses. In Canada, interRAI data holdings represent some of the largest repositories of clinical information in the country for persons with neurological conditions. This study examined the accuracy of the diagnostic information captured by interRAI instruments designed for use in the home care, long-term care and mental health care settings as compared with national administrative databases.
The interRAI assessments were matched with an inpatient hospital record and emergency department (ED) visit record in the preceding 90 days. Diagnoses captured on the interRAI instruments were compared to those recorded in either administrative record for each individual. Diagnostic validity was examined through sensitivity, specificity and positive predictive value analysis for the following conditions: multiple sclerosis, epilepsy, Alzheimer’s disease and other dementias, Parkinson’s disease, traumatic brain injury, stroke, diabetes mellitus, heart failure and reactive airway disease.
In the three large study samples (home care: n = 128,448; long-term care: n = 26,644; mental health: n = 13,812), interRAI diagnoses demonstrated high specificity when compared to administrative records, for both neurological conditions (range 0.80 – 1.00) and comparative chronic diseases (range 0.83 – 1.00). Sensitivity and positive predictive values (PPV) were more varied by specific diagnosis, with sensitivities and PPV for neurological conditions ranging from 0.23 to 0.94 and 0.14 to 0.77, respectively. The interRAI assessments routinely captured more cases of the diagnoses of interest than the administrative records.
The interRAI assessment collected accurate information about disease diagnoses when compared to administrative records within three months. Such information is likely relevant to day-to-day care in these three environments and can be used to inform care planning and resource allocation decisions.