Employing IT to improve antibiotic prescribing

teaser

Richard H Drew
PharmD MS BCPS
Clinical Pharmacist
Duke University School of Medicine
Durham, NC
USA
Associate Professor
Campbell University School of Pharmacy
Buies Creek, NC
USA
E:[email protected]

Numerous studies have demonstrated the need for appropriate and timely antibiotic therapy in the treatment of patients with serious illnesses.(1–8) Appropriate use has also been identified as an integral component in strategies to reduce antimicrobial resistance.(9) A significant factor contributing to inappropriate antibiotic use is the lack of disease-, patient- and pathogen-specific information at the time therapy is initiated.(10) Despite the availability of treatment guidelines for selected infections,(11–20) optimal utilisation may be hampered by a lack of awareness of guideline availability, a lack of accessibility at the point of care, a need to be adapted for local use and a need for frequent updating.(21)

Recent advances in computer and communication technologies could significantly contribute to more appropriate use of antimicrobial therapy by delivering vital information to prescribers. This article briefly summarises the past contributions of information technology (IT) to infectious diseases management, and offers an example of a present-day solution to the need for customised information resources.

Past applications: antibiotic order processing and feedback
Medication order processing and distribution systems used by pharmacies rely heavily on IT. Many are capable of performing screening for drug–drug interactions, allergies and selected contraindications. Computerised physician order entry (CPOE) is a recent advance that allows clinicians to order medications, diagnostic tests, laboratory studies, diet, nursing care and consultations directly through the institution’s computerised order processing system. CPOE may include order sets, decision algorithms and/or calculators to facilitate antibiotic selection and dosing.

IT has facilitated the collection of information used to provide feedback to prescribing clinicians on appropriate antibiotic use. Reports including prescribing trends, adherence rates to treatment guidelines, antibiotic purchases and pathogen resistance trends are often used by institutions to improve antibiotic use. Reports such as prolonged treatment durations, inappropriate dosing, switching from parenteral to oral therapy and discordance with treatment and pathogen susceptibility profiles may also be useful to identify opportunities for intervention by pharmacists.(22) While such systems usually improve the safe use of antibiotics, they often lack the capacity to assist in identifying the optimal selection.

IT has also provided the ability to expand opportunities for infectious disease prevention, such as the need for immunisation. Limited integration with hospital medical records can help identify patients who may require isolation due to increased risks of antibiotic-resistant pathogens (such as those with prior history of methicillin-resistant Staphylococcus aureus [MRSA] or Clostridium difficile– associated colitis).

Evolving use of information technology: decision support
Advances in computer hardware (such as notebook computers and personal digital assistants [PDAs]), software and communication technology (such as wireless networks) have allowed information resources to be available at the point of care.(23) As a result, a growing number of references are available via PDAs (see Table 1).(24) The advantages of such systems include their rapid updateability, easy accessibility, ease of use, low cost and expert-provided content.

[[HPE17_table1_86]]

The ability to provide true decision support to the clinician regarding optimal antimicrobial selection has been limited, due largely to the lack of systems capable of integrating patient-specific data. This would include the information needed to determine the likelihood of infection, risk of antibiotic-resistant pathogens, dosing requirements, risk of antibiotic toxicities, expected treatment response and treatment costs.(25,26) While integrated systems have demonstrated their ability to improve drug selection, reduce adverse drug events, reduce antibiotic-related costs and stabilise or improve the prevalence of resistance isolates,(27–31) such systems can be extremely expensive. Initial investment costs range from US$5 to 10 million, while annual maintenance costs can exceed US$1 million a year.(32,33) An additional barrier to the implementation of such systems is the need for distribution of data across multiple systems that often do not communicate with each other.(34) Thus, integrated systems are currently not attainable for most hospitals.(35)

Customising resources
Information resources for selecting appropriate antimicrobial therapy must include local and institutional considerations. For example, institution-specific data on pathogen susceptibility (usually in the form of an antibiogram) is an important consideration in determining empirical antibiotic treatment.(36,37) Infectious disease treatment practices (particularly empiric antibiotic therapy) should also consider antibiotic formulary considerations, drug availability, local clinician preferences, research protocols and local policies and/or restriction programmes.(38–40)

Paper-based antibiograms and antibiotic selection guidelines have been employed by many facilities in an attempt to address the need for current, institution-specific information. However, such sources are currently cumbersome (and often expensive) to produce and maintain. Web- and/or PDA- based guidelines provide a nonintegrated solution to the challenge of customising, maintaining and communicating institution-specific treatment recommendations. An example of such a system, CustomID (©), was developed (and is currently in use) at Duke University Medical Center (Durham, NC, USA). The use of a web authoring tool permitted the easy entry of content into Microsoft(®) SQL (an open database connectivity [ODBC]-compliant relational database). Microsoft(®) Active Server Pages present content in a web view. The content is delivered via AvantGo(®) to PDAs, which use either Palm(®) or Pocket PC(®) software. Physicians, pharmacists, nurses and microbiologists with an expertise in infectious diseases and infection control continually provide institution-specific content. Information is organised into disease-, drug- and pathogen-specific sections. Additional content is available in areas such as infection control, surgical prophylaxis, investigational protocols, antibiotic restriction programmes, information on research protocols, biological hazard response information and caremaps. Extensive cross-referencing is available. For example, clinicians choosing to review information on in-vitro susceptibilities from the institution’s antibiogram are able to choose cross-referenced sections such as treatment options, dosing guidelines, restriction programme information and investigational protocols recruiting patients with such infections.

Within the first year, CustomID was used to communicate rapidly several important infection management changes and new information, such as drug shortages (and necessary treatment recommendation changes), local influenza outbreak data, revised institutional infection control practices (such as those necessary for a suspected case of severe acute respiratory syndrome [SARS]), updates on biological hazard response procedures, updated national treatment guidelines, new investigational protocols, changes to the institution’s antibiotic formulary and update to the antibiogram.

Conclusion
IT significantly enhances the ability to provide decision support for clinicians choosing an antimicrobial therapy. Such information should be based on published guidelines and customised to the local environment. While systems which integrate decision support capabilities with patient-specific data exist, they are largely unavailable. Institutions can use “off-the-shelf” technology to provide timely information important to the selection of appropriate antibiotic use.

References

1. Pediatr Infect Dis J 2003;22:1143-51.
2. Am J Manag Care 2002;8:713-27.
3. Chest 1999;115:462-74.
4. Chest 2000;118:146-55.
5. Chest 1997;111:676-85.
6. Clin Infect Dis 1997;24:584-602.
7. JAMA 1997;278:2080-4.
8. Ann Pharmacother 2000;34:446-52.
9. Clin Infect Dis 1997;25:584-99.
10. JAMA 1998;280:1270-1.
11. Clin Infect Dis 2000;30:679-83.
12. Clin Infect Dis 2000;30:658-61.
13. Clin Infect Dis 2000;30:662-78.
14. Clin Infect Dis2000;30:710-8.
15. Clin Infect Dis2000;30:696-709.
16. Clin Infect Dis2000;30:688-95.
17. Pediatrics 2001;108:798-808.
18. Ann Intern Med 2001;134:521-9.
19. Pediatrics 2004;113:1451-65.
20. Clin Infect Dis 2003;37:1405-33.
21. Med Care 2001;39:II55-69.
22. Arch Intern Med 2004;164:1206-12.
23. J Am Med Inform Assoc 2003;10:605-7.
24. Clin Infect Dis2003;36:1018-29.
25. Stud Health Technol Inform 1999;68:690-5.
26. Med Decis Making 1989;9:5156.
27. J Hosp Infect 1991;18 Suppl A:424-31.
28. Am J Infect Control 1992;20:4-10.
29. Med Clin North Am 2001;85:1397-411.
30. Pediatrics 2001;108:E75.
31. J Chemother 1999;11:530-5.
32. Ann Intern Med 2003;139:31-9.
33. Am J Health-Syst Pharm 2004;61:190-7.
34. Crit Care Med 2001;29:N87-91.
35. JAMA 1998;279:1024-9.
36. Clin Infect Dis 2004;39:497-503.
37. Int J Clin Pract 2002;Suppl:29-36.
38. Ann Intern Med 1996;124:884-90.
39. Clin Perform Qual Health Care 1998;6:12-6.

Resource
Centers for Disease Control and Prevention campaign to prevent antimicrobial resistance in healthcare settings
W:www.cdc.gov/drugresistance/healthcare/default.htm