Evolving health information technology and the timely availability of visit diagnoses from ambulatory visits: a natural experiment in an integrated delivery system

Abstract

Background: Health information technology (HIT) may improve health care quality and outcomes, in part by making information available in a timelier manner. However, there are few studies documenting the changes in timely availability of data with the use of a sophisticated electronic medical record (EMR), nor a description of how the timely availability of data might differ with different types of EMRs. We hypothesized that timely availability of data would improve with use of increasingly sophisticated forms of HIT.

Methods: We used an historical observation design (2004-2006) using electronic data from office visits in an integrated delivery system with three types of HIT: Basic, Intermediate, and Advanced. We calculated the monthly percentage of visits using the various types of HIT for entry of visit diagnoses into the delivery system's electronic database, and the time between the visit and the availability of the visit diagnoses in the database.

Results: In January 2004, when only Basic HIT was available, 10% of office visits had diagnoses entered on the same day as the visit and 90% within a week; 85% of office visits used paper forms for recording visit diagnoses, 16% used Basic at that time. By December 2006, 95% of all office visits had diagnoses available on the same day as the visit, when 98% of office visits used some form of HIT for entry of visit diagnoses (Advanced HIT for 67% of visits).

Conclusion: Use of HIT systems is associated with dramatic increases in the timely availability of diagnostic information, though the effects may vary by sophistication of HIT system. Timely clinical data are critical for real-time population surveillance, and valuable for routine clinical care.

Figure 1

Schematic of HIT implementation and use during study period. *Advanced HIT implementation was staggered by medical center, and by team within medical center (estimated 3-week lag between teams in medical centers). Once Advanced HIT was implemented, no other interface was available for charting or entering diagnostic information.

Figure 2

All Office Visits' HIT Use and Time to Data Availability. a. HIT use in all office visits. b. Percentage of all office visits with data available within 0, 1, 4, or 7 days.

Figure 3

Change in mean time to data availability with change in HIT use over time.

Figure 4

Medical Center A increased use of Advanced HIT (Fig 4a) associated with >98% visits with data available on the same day as the visit (Fig 4b). a. HIT use in office visits at Medical Center A. b. Percent of office visits at Medical Center A with data available on the same day as the visit.

Figure 5

At Medical Center B, Advanced HIT not implemented during study period (Fig 5a), and percentage of visits with data available on same day does not reach >91% (Fig 5b). a. HIT use in office visits at Medical Center B. b. Percent of office visits at Medical Center B with data available on the same day as the visit.