Background and Methods:
Although studies have demonstrated that medical rehabilitation patients have many complications that warrant attention, none has attempted to categorize complications by severity. This retrospective cohort study examined the incidence, types, and severity of problems that interrupt rehabilitation and the major risk factors for these events.
Of 1075 patients, 359 (33.4%) had acute medical complications on rehabilitation considered severe enough to interrupt treatment. Of the 359 patients, 158 (44%) required an unexpected transfer off rehabilitation. The most common reasons for unexpected transfer were surgical causes (22.8%), followed by infection or fever (17.1%) and by thromboembolic events (16.5%). Logistic regression revealed that major risk factors for complications requiring transfer were a primary diagnosis of deconditioning or nontraumatic spinal cord injury (adjusted odds ratio, 2.7; confidence interval, 1.8 to 4.2), severity of initial disability (adjusted odds ratio, 1.2; confidence interval, 1.1 to 1.3 for every 10-point drop in a Modified Barthel Index), and number of comorbid conditions (adjusted odds ratio, 1.1; confidence interval, 1.0 to 1.2). Risk factors for any complication were similar, but there was an interaction between comorbidity and the degree of functional impairment; in patients who were severely functionally impaired, the number of comorbidities was not as strongly associated with the risk of complications as it was in patients who were less functionally impaired.
There is a complex relationship among the type of underlying medical impairment, severity of functional limitation, comorbidity, and unanticipated medical or surgical complications that interrupt rehabilitation. The interruptions vary both in type and in severity.(Arch Intern Med. 1994;154:2185-2190)
Siegler EL, Stineman MG, Maislin G. Development of Complications During Rehabilitation. Arch Intern Med. 1994;154(19):2185–2190. doi:10.1001/archinte.1994.00420190085010
Customize your JAMA Network experience by selecting one or more topics from the list below.