Am. J. Respir. Crit. Care Med.,
Volume 158, Number 2, August 1998, 418-423
Outreach Education to Improve Quality of
Rural ICU Care
Results of a Randomized Trial
MICHAEL S.
HENDRYX,
JOHN F.
FIESELMANN,
M.
JEANNE BOCK,
DOUGLAS S.
WAKEFIELD,
CHARLES M.
HELMS,
and
SUZANNE E.
BENTLER
Health Policy and Administration, Washington State University, Spokane, Washington; and Health Services Research Center and
Departments of Internal Medicine and Preventive Medicine, University of Iowa, Iowa City, Iowa
This study tests whether an outreach educational program tailored to institutional specific patient
care practices would improve the quality of care delivered to mechanically ventilated intensive care
unit (ICU) patients in rural hospitals. The study was conducted as a randomized control trial using 20 rural Iowa hospitals as the unit of analysis. Twelve randomly selected hospitals received an outreach
educational program. After review of the medical records of eligible patients, a multidisciplinary
team of intensive care unit specialists from an academic medical center delivered an educational program with content specific to the findings and capacity of the hospital. The outcome measures included patient care processes, patient morbidity and mortality outcomes, and resource use. Results
indicated that the outreach program significantly improved many patient care processes (lab work,
nursing, dietary management, ventilator management, ventilator weaning). The program marginally
reduced hospital ventilator days. Both total length of stay and ICU length of stay fell markedly in the
intervention group (by an average of 3.2 and 2.1 d, respectively), while the control group fell only 0.6 and 0.3 d, respectively. However, these effects did not reach statistical significance. Unfortunately, the program had no detectable effects on the clinical outcomes of mortality or nosocomial events.
We conclude that an outreach program of this type can effectively improve processes of care in rural
ICUs. However, improving processes of care may not always translate into improvement of specific
outcomes.
