Background: There is no formal curriculum for medical students, residents or advanced practitioner students on appropriate clinical documentation improvement (CDI). As the primary documenters for medical records, several studies have shown benefit in billing, coding and hospital metrics with trainee education [1-2]. Despite this, there remains a scarcity of education on documentation for residents and attending physicians alike. We present a novel peer education program utilizing standardized phrasing to improve accuracy and completeness in medical documentation.
Methods: Our intervention aimed to reduce the prevalence of hospital documentation queries directed towards resident internal medicine teams by 25%. All 178 internal medicine residents in our program were included. We identified the most frequently queried diagnoses and designed a curriculum with the hospital CDI team and internal medicine resident peers. The eight most queried were converted into standardized diagnoses with supporting phrases, together called smart phrases. Monthly workshops taught by peer CDI specialists were given on hospital ward rotations, which explained the need for accurate documentation and use of these smart phrases. Information was reinforced in team rooms with both posters and reminder cards placed on computers. Outcomes were measured by trending monthly queries for each of the eight diagnoses after phrase implementation and reviewed every 6 months. Results were normalized against overall hospital trends in queries to account for standard variability and diagnoses with competing documentation initiatives were excluded. Resident satisfaction with both education and smart phrases was measured qualitatively before and annually after intervention implementation. Expected length of stay, case mix index, and expected mortality were also recorded over the course of the study and normalized against non-teaching teams fluctuations.
Results: Before intervention, 40% of internal medicine residents reported receiving little or no education in documentation and 80% believed they would benefit from more education. After the initiation of workshops and smart phrases, queries reduced by 64% over 18 months. Total queries reduced from baseline 181 queries over initial 4-month period to 49 queries during final 4-month period after intervention. Workshop attendance by teams was 100% and standard phrases were used 114 times during the final study period. Hospital metrics for study teams also increased during the study period when normalized for other variables. Notably expected length of stay increased from 6.52 to 6.64, case mix index increased from 1.67 to 1.78, and expected mortality increased from 2.09 to 2.36. Post-survey showed 100% of participants were satisfied with the education provided by our workshops. Greater than fifty percent of participants utilized provided standard phrases into >50% of their documentation.
Conclusions: Our peer-to-peer education and stakeholder derived smart phrases are effective in reducing commonly mis-documented diagnoses. This increase in documentation accuracy, demonstrated through reduction in documentation queries, directly impacted teams’ expected length of stay, case mix index, and expected mortality. Future directions include expanding this program to other diagnoses and departments, and measuring reduction in time and burden of documentation.