Validity of autism diagnoses using administrative health data

Abstract

It is necessary to monitor autism prevalence in order to plan education support and health services for affected children. This study was conducted to assess the accuracy of administrative health databases for autism diagnoses. Three administrative health databases from the province of Nova Scotia were used to identify diagnoses of autism spectrum disorders (ASD): the Hospital Discharge Abstract Database, the Medical Services Insurance Physician Billings Database and the Mental Health Outpatient Information System database. Seven algorithms were derived from combinations of requirements for single or multiple ASD claims from one or more of the three administrative databases. Diagnoses made by the Autism Team of the IWK Health Centre, using state-of-the-art autism diagnostic schedules, were compared with each algorithm, and the sensitivity, specificity and C-statistic (i.e. a measure of the discrimination ability of the model) were calculated. The algorithm with the best test characteristics was based on one ASD code in any of the three databases (sensitivity=69.3%). Sensitivity based on an ASD code in either the hospital or the physician billing databases was 62.5%. Administrative health databases are potentially a cost efficient source for conducting autism surveillance, especially when compared to methods involving the collection of new data. However, additional data sources are needed to improve the sensitivity and accuracy of identifying autism in Canada.

Prevalence and Correlates of Autism Spectrum Disorders in Quebec

“…We have nevertheless included them because this longitudinal study covers the transition from the DSM-4 to DSM-5 . Our case definition fits the one validated by Dodds et al 32…”

“…39 –41 However, studies have shown that diagnoses recorded in medical records fully corresponded to entries in administrative databases for other diseases, in Quebec as in other provinces. 40,42 In addition, Dodds et al 32 reported that administrative health databases can clearly identify children with autism. Second, since QICDSS data only include information on people who use health care services, the case definition only identifies people who have received a diagnosis or treatment from a physician.…”

Validation of administrative health data for the pediatric population: a scoping review

“…Only five conditions were investigated in more than one study: diabetes (10.6%) [ 13 - 16 ], inflammatory bowel disease (5.4%) [ 17 , 18 ], asthma (5.4%) [ 19 , 20 ], rotavirus infection (5.4%) [ 21 , 22 ], and tuberculosis (5.4%) [ 23 , 24 ]. However, a diverse range of conditions were investigated in single studies, including obesity [ 25 ], vaccine-related illness [ 26 ], injuries [ 27 ], autism [ 28 ], febrile neutropenia in oncology patients [ 29 ], high risk conditions [ 30 ], dermatologic conditions [ 31 - 33 ], congenital anomalies [ 34 ], cardiac defects [ 35 ], respiratory illnesses excluding asthma [ 36 - 38 ], neurologic conditions [ 39 ], other gastrointestinal conditions [ 40 - 43 ], genitourinary conditions [ 44 , 45 ], serum sickness [ 46 ], thrombosis [ 47 ], maternal/perinatal conditions [ 48 ], and drug-related anaphylaxis [ 49 ].…”

Factors affecting the age at diagnosis of autism spectrum disorders in Nova Scotia, Canada

“…Another limitation of the administrative data usage is that the three health claim databases almost certainly did not capture all individuals who were diagnosed with ASDs during the study period. The sensitivity of the three databases has been previously reported as approximately 70% (Dodds et al, 2009). However, this estimate was not found to vary significantly based on several of the variables evaluated in the present study, including sex, birth order, birth weight, county of residence, maternal age, and the diagnosis of a major congenital anomaly.…”

“…Due to concerns regarding the reliability of diagnoses in children under the age of two years (Chawarska et al, 2007; Cox et al, 1999), those whose only ASD diagnostic code appeared before this age were excluded from the study. An earlier study reported the sensitivity and specificity of these data sources for detecting true cases of ASD as 69.3% and 77.3%, respectively (Dodds et al, 2009).…”