Understanding keys to successful implementation of electronic decision support in rural hospitals: analysis of a pilot study for antimicrobial prescribing.

Electronic clinical decision support systems (CDSS) have been hailed for their potential to improve clinical outcomes. Using a pretest/posttest design, an Internet-based CDSS designed to optimize antimicrobial prescribing was pilot tested for community-acquired pneumonia in 5 rural hospitals in southwestern Idaho. An antimicrobial management team was created in each hospital to address clinicians' perception of excessive time required for direct use of the CDSS. In pooled hospital data, agreement with CDSS recommendations improved to a statistically significant level. However, inspection of data at the individual hospital level demonstrated that almost all improvement occurred in a single hospital. Failure in the other hospitals appeared to be primarily a consequence of organizational and cultural barriers. These barriers are discussed to understand keys for successful future implementation of CDSS in rural hospitals, drawing on experience with cultural barriers from other industries, specifically aviation.

Assessing the status of infection control programs in small rural hospitals in the western United States.

BACKGROUND: Organized infection control (IC) interventions have been successful in reducing the acquisition of hospital-associated infections. Rural community hospitals, although contributing significantly to the US health care system, have rarely been assessed regarding the nature and quality of their IC programs. METHODS: A sample of 77 small rural hospitals in Idaho, Nevada, Utah, and eastern Washington completed a written survey in 2000 regarding IC staffing, infrastructure support, surveillance of nosocomial infections, and IC policies and practices. RESULTS: Almost all hospitals (65 of 67, 97%) had one infection control practitioner (ICP), and 29 of 61 hospitals (47.5%) reported a designated physician with IC oversight. Most ICPs (62 of 64, 96.9%) were also employed for other activities outside of IC. The median number of ICP hours per week for IC activities was 10 (1-40), equating to a median of 1.56 (0.30-21.9) full-time ICPs per 250 hospital beds. Most hospitals performed total house surveillance for nosocomial infections (66 of 73, 90.4%) utilizing Centers for Disease Control and Prevention (CDC) definitions (69 of 74, 93.2%). Most also monitored employee bloodborne exposures (69 of 73, 94.5%). All hospitals had a written bloodborne pathogen exposure plan and isolation policies. CDC guidelines were typically followed when developing IC policies. Access to medical literature and online resources appeared to be limited for many ICPs. CONCLUSIONS: Most rural hospitals surveyed have expended reasonable resources to develop IC programs that are patterned after those seen in larger hospitals and conform to recommendations of consensus expert panels. Given these hospitals' small patient census, short length of stay, and low infection rates, further studies are needed to evaluate necessary components of effective IC programs in these settings that efficiently utilize limited resources without compromising patient care.

Pharmacist involvement in antimicrobial use at rural community hospitals in four Western states.

PURPOSE: Pharmacist involvement in antimicrobial use at small rural hospitals in four Western states was studied. METHODS: Surveys were mailed in July 2000 to hospitals with a daily patient census of <150 in Idaho, Nevada, Utah, and eastern Washington. RESULTS: Seventy-seven (77%) of 100 hospitals returned completed surveys. Only 5% of the hospitals had onsite pharmacists 24 hours per day. An onsite pharmacist was present for a median of 26 hours per week in hospitals without 24-hour pharmacist coverage (range, 0-116 hr/wk). Many hospitals (71%) had policies for monitoring or controlling antimicrobial use, but only 28% had a system capable of monitoring compliance with such policies. Few hospitals had systems for recommending changes in antimicrobial selection on the basis of susceptibility test results (27%) or for monitoring physician compliance with dosage recommendations by pharmacists (21%). Onsite pharmacist hours were significantly associated with pharmacists being involved in the initial ordering of antibiotics and providing active oversight of antimicrobial use. There was a negative correlation between onsite pharmacist hours and use of third-generation cephalosporins and carbapenems. CONCLUSION: A survey showed that rural hospital pharmacists in four Western states spent relatively little time monitoring and influencing antimicrobial prescribing.

Detection of antimicrobial resistance by small rural hospital microbiology laboratories: comparison of survey responses with current NCCLS laboratory standards.

Microbiology laboratory personnel from 77 rural hospitals in Idaho, Nevada, Utah, and eastern Washington were surveyed in July 2000 regarding their routine practices for detecting antimicrobial resistance. Their self-reported responses were compared to recommended laboratory practices. Most hospitals reported performing onsite bacterial identification and susceptibility testing. Many reported detecting targeted antimicrobial resistant organisms. While only 5/61 hospitals (8%) described using screening tests capable of detecting all 8 targeted types of resistance, most (57/61, 93%) were capable of accurately screening for at least 6 types. Conversely, most hospitals (58/61, 95%) reported confirmatory testing capable of identifying only 3 or fewer resistance types with high-level penicillin resistance among pneumococci, methicillin and vancomycin resistance among staphylococci and enterococci, and extended spectrum beta-lactamase production by Gram-negative bacilli presenting the greatest difficulties. Furthermore, only 50% of hospitals compiled annual antibiogram reports to help physicians choose initial therapy for suspected infectious illnesses. This survey suggests that the antimicrobial susceptibility testing in many rural hospitals may be unreliable.