Changes to physician processing times in response to clinic congestion and patient punctuality: a retrospective study

OBJECTIVES: We examine interactions among 3 factors that affect patient waits and use of overtime in outpatient clinics: clinic congestion, patient punctuality and physician processing rates. We hypothesise that the first 2 factors affect physician processing rates, and this adaptive physician behaviour serves to reduce waiting times and the use of overtime. SETTING: 2 urban academic clinics and an affiliated suburban clinic in metropolitan Baltimore, Maryland, USA. PARTICIPANTS: Appointment times, patient arrival times, start of service and physician processing times were collected for 105 visits at a low-volume suburban clinic 1, 264 visits at a medium-volume academic clinic 2 and 22 266 visits at a high-volume academic clinic 3 over 3 distinct spans of time. INTERVENTION: Data from the first clinic were previously used to document an intervention to influence patient punctuality. This included a policy that tardy patients were rescheduled. PRIMARY AND SECONDARY OUTCOME MEASURES: Clinicians' processing times were gathered, conditioned on whether the patient or clinician was tardy to test the first hypothesis. Probability distributions of patient unpunctuality were developed preintervention and postintervention for the clinic in which the intervention took place and these data were used to seed a discrete-event simulation. RESULTS: Average physician processing times differ conditioned on tardiness at clinic 1 with p=0.03, at clinic 2 with p=10(−5) and at clinic 3 with p=10(−7). Within the simulation, the adaptive physician behaviour degrades system performance by increasing waiting times, probability of overtime and the average amount of overtime used. Each of these changes is significant at the p<0.01 level. CONCLUSIONS: Processing times differed for patients in different states in all 3 settings studied. When present, this can be verified using data commonly collected. Ignoring these behaviours leads to faulty conclusions about the efficacy of efforts to improve clinic flow.