This quasiexperimental study investigated interdisciplinary collaboration over a 16-month period on units using different collaborative practice strategies. Measures of collaboration and perceived physician involvement in collaborative practice were completed by 335 licensed staff members working on seven general adult units in an acute care hospital located in an academic medical center. Data were collected at two time points: in 1993 and 1995. A small but statistically significant decline in collaboration was found (p = 0.01) over the 16-month period. Analysis of variance revealed a significant difference (p = 0.03) in collaboration related to the method used to develop collaborative paths. Post hoc Tukey's test indicated that the presence of a case manager without collaborative paths did show higher levels of collaboration (p = 0.05). Regardless of the strategy used, perceived high physician involvement was related to greater collaboration than perceived low involvement with differences increasing over time (p = 0.02). These findings suggest the importance of perceived physician involvement in collaborative practice to interdisciplinary collaboration.