Does real-time feedback to residents with or without attendings improve medical documentation?

Teaching resident physicians to appropriately document patient diagnoses in the medical record is critical to ensure accuracy of publicly reported data and to enable quality improvement initiatives. We sought to determine if providing real-time feedback to resident or attending physicians would lead to improved medical documentation.

We audited charts to determine if target diagnoses were present and documented. After a baseline data-gathering period, during the intervention period the first author randomly provided feedback to either residents alone or residents plus their attending physicians. Capture rates for targeted diagnoses, total diagnoses documented and coded, severity of illness (SOI), risk of mortality (ROM), and case mix index were compared pre- and postintervention.

During the intervention, the proportion of target diagnoses captured increased from 30% to 54% (P < 0.001), as did the median (interquartile [IQ] range) number of diagnoses captured (from 11 [IQ range, 8-13] to 12 [IQ range, 9-16], P < 0.05). We also achieved significant increases in median (IQ range) SOI scores (from 2 [IQ range, 2-3] to 3 [IQ range, 2-3], P = 0.04) and complication code capture (61% vs 75%, P = 0.01). However, there was no significant increase in ROM, case mix index, or hospital billing. Notifying attending physicians in addition to residents was of no benefit.

An educational program based on real-time feedback to resident physicians regarding documentation improved diagnostic capture and SOI scores, but not case mix index or billing at our public, safety net, teaching hospital. Even with enhanced education of resident physicians regarding documentation, other interventions are needed to further improve diagnostic capture.