PURPOSE: Using a population health services perspective, this article defines and assesses an efficient criteria-based algorithm to identify treatment prevalent and incident cases of schizophrenia. We refer here "treatment" prevalence and incidence since its evaluation depends on a patient receiving a health care service with a diagnosis of schizophrenia. METHODS: A population-based cohort study was conducted among all adults having a hospital discharge or a physician claim for schizophrenia in the public health plan databases between January 1996 and December 2006. Four algorithms to characterize patients with schizophrenia were defined. To identify treatment incident cases in 2006, we removed from the treatment prevalent pool patients with a previous record of schizophrenia between 1996 and 2006 (10-year clearance period). Using this 10-year period as reference, Kappa coefficients (KC) and positive predictive values (PPV) were calculated to determine the "optimal" length of clearance period to identify incident cases. RESULTS: The lifetime treatment prevalence and incidence of schizophrenia varied from 0.59 to 1.46% and from 42 to 94 per 100,000, respectively. When compared to the 10-year clearance period, the KC is excellent in a clearance period of 6-7 years. To achieve a PPV of 90%, a clearance period of 7-8 years would be necessary. CONCLUSIONS: With an appropriate algorithm, treatment prevalence and incidence of schizophrenia can be conveniently estimated using administrative data. These estimates are a vital step toward appropriate planning of services for schizophrenia.
PURPOSE: Using a population health services perspective, this article defines and assesses an efficient criteria-based algorithm to identify treatment prevalent and incident cases of schizophrenia. We refer here "treatment" prevalence and incidence since its evaluation depends on a patient receiving a health care service with a diagnosis of schizophrenia. METHODS: A population-based cohort study was conducted among all adults having a hospital discharge or a physician claim for schizophrenia in the public health plan databases between January 1996 and December 2006. Four algorithms to characterize patients with schizophrenia were defined. To identify treatment incident cases in 2006, we removed from the treatment prevalent pool patients with a previous record of schizophrenia between 1996 and 2006 (10-year clearance period). Using this 10-year period as reference, Kappa coefficients (KC) and positive predictive values (PPV) were calculated to determine the "optimal" length of clearance period to identify incident cases. RESULTS: The lifetime treatment prevalence and incidence of schizophrenia varied from 0.59 to 1.46% and from 42 to 94 per 100,000, respectively. When compared to the 10-year clearance period, the KC is excellent in a clearance period of 6-7 years. To achieve a PPV of 90%, a clearance period of 7-8 years would be necessary. CONCLUSIONS: With an appropriate algorithm, treatment prevalence and incidence of schizophrenia can be conveniently estimated using administrative data. These estimates are a vital step toward appropriate planning of services for schizophrenia.
Authors: Baber K Khan; Josh D Woolley; Steven Chao; Tricia See; Anna M Karydas; Bruce L Miller; Katherine P Rankin Journal: Psychosomatics Date: 2012-01-28 Impact factor: 2.386
Authors: Kelly K Anderson; Rebecca Fuhrer; Willy Wynant; Michal Abrahamowicz; David L Buckeridge; Ashok Malla Journal: Soc Psychiatry Psychiatr Epidemiol Date: 2013-02-21 Impact factor: 4.328
Authors: Arjen L Sutterland; Jeanne Dieleman; Jitschak G Storosum; Bettie A C Voordouw; Jojanneke Kroon; Joris Veldhuis; Damiaan A J P Denys; Lieuwe de Haan; Miriam C J M Sturkenboom Journal: Soc Psychiatry Psychiatr Epidemiol Date: 2013-01-23 Impact factor: 4.328
Authors: Donald Addington; Elizabeth Anderson; Martina Kelly; Alain Lesage; Chris Summerville Journal: Can J Psychiatry Date: 2017-09 Impact factor: 4.356