At least a fifth of people with epilepsy, primarily children, engaged in hospital crossover over a 2-year period, a cross-sectional study of epilepsy patient data sets showed.
Health exchange information (HIE) technology has the potential to bridge the data gaps among the unaffiliated physicians who serve these patients, helping to "lower health care costs, reduce adverse events and medical errors, and improve population health," wrote Dr. Zachary M. Grinspan of Cornell University, New York, and New York Presbyterian Hospital and his coauthors. But the technology’s full value has yet to be studied.
Dr. Zachary Grinspan
To determine who among the epileptic community are most likely to use two or more hospitals (crossover), and why, Dr. Grinspan and his colleagues conducted a cross-sectional analysis of HIE patient-level data sets for 8,704 patients seen by seven New York City tertiary-care teaching hospitals. They examined demographics, encounter dates, and ICD-9 codes for outpatient, emergency department, inpatient, and radiology encounters for visits that occurred April 2009 through March 2012. All data were taken from the New York state–sponsored NYCLIX (New York Clinical Information Exchange), an "honest broker" of such information, and were deidentified prior to the analysis (Epilepsia 2014 March 5 [doi:10.1111/epi.12552]).
In all, 1,770 (22%) patients were found to engage in crossover during the study period. Compared with adults (median age, 36 years), children (between age 0 and 10 years) were more likely to be associated with crossover (adjusted odds ratio, 1.4; 95% confidence interval, 1.2-1.7), as were those who lived near the hospitals in the study (the borough of Manhattan vs. other New York boroughs (adjusted OR, 1.6; 95% CI, 1.4-1.8).
Crossover was also associated with significantly greater odds of more visits to emergency, radiology, inpatient, and outpatient settings (P less than .001 for each), and of having 6 or more days with a head CT scan (P less than .05 for 6-10 days and P less than .01 for 11 or more days). Both bivariate and multivariate analyses consistently associated an encephalopathy diagnosis with crossover (adjusted OR, 2.66; 95% CI, 2.14-3.29); any relationship between crossover and other comorbidities was uncertain.
Dr. Grinspan and his associates noted the possibility that their findings may actually reflect lower crossover than actually occurs, since there are 14 other hospitals in Manhattan that were not included in the data pool. Reasons for patients seeking care in multiple settings may include a desire for a second opinion or the need for emergency services in patients with uncontrolled epilepsy, wrote the authors. "Further research should investigate why hospital crossover occurs, how it affects care, and how HIE can most effectively mitigate the resultant fragmentation of medical records."