Judith V. Douglas

In an industry beset by cost pressures, the cost justification of information technology is a challenge. For the third straight year, the 2001 HIMSS Leadership Survey 1 ranked providing quantifiable benefits and return on investment (ROI) among the top three barriers to implementing information technology. According to David M. Barrett, MD, CEO of Lahey Clinic, Burlington, Mass, ROI is increasingly critical. “We can’t keep spending more and more money on [information technology] and not have it saving us money in other areas,” 2 Barrett notes.

The evidence that efforts in health care informatics pay off is beginning to accumulate, with much of it appearing in the literature on patient safety. The call to reduce medical errors and improve the quality of care sounded by the Institute of Medicine in To Err Is Human 3 and The Quality Chasm 4 spurred a number of activities that are now being documented and reported. In July 2001, for example, the US Agency for Healthcare Research and Quality released an extensive evidence report and technology assessment, Making Health Care Safer 5 . Portions of this document addressed information technology. Along with private-sector initiatives, these efforts are helping to define the role of information technology as an enabling entity. They are also demanding proof that technological tools are effective.

Marion J. Ball, EdD

The quality movement in health care is helping organizations to focus on information-technology investment as a means of value creation, not just overhead-cost reduction. 6 With patient care at the center of an initiative, the objective becomes not simply cost savings (although the cost of medical errors is staggering), but overall organizational performance. The 2001 HIMSS Survey reflects this concern for value. Clinical information systems (CIS) ranked first among application areas considered most important over the next two years, while financial and administrative systems placed sixth. The top business issues cited by those surveyed included improving operational efficiency, at 45% (down from 60% in 2000), and reducing medical errors, at 44% (up from 31% in 2000).

A coalition of major purchasers of heath care coverage, the Leapfrog Group, has helped make error reduction a more prominent issue by directing the coalition’s business to health care providers having computerized physician-order entry (CPOE) in place. A CPOE system, it should be noted, is not a stand-alone system; it exists within a larger CIS, along with other components such as databases. 7 Because medication errors have been well documented and are considered to be high-yield areas for improvement, information from CPOE installations constitutes some of the best available data8 showing that information technology does, in fact, provide the ROI that health care executives need.

In a major white paper, leading informaticians 8 recently concluded that decision support and linkages within and across systems should be increased in health care. They also found hard evidence that decision-support tools, such as alerts and prompts, are effective. By providing these capabilities at the point of care, information technology acts to enable the improvement of clinical processes.

Documenting Results

Early studies of CPOE documented savings in costs and/or inpatient days. For example, charges at Wishard Memorial Hospital, Indianapolis, fell 13% and length of stay (LOS) was reduced by 0.9 days after CPOE was implemented. 9 In 1999, CPOE at Sarasota Memorial Hospital, Fla, a community hospital, produced savings equaling about 10% of the hospital’s total annual budget and 25% of the variable costs that are controlled by physicians (in guiding the delivery of care). Specifically, physicians who regularly used order entry, results retrieval, and other functions realized an 8.4% decrease in resource utilization costs and a 6.8% decrease in LOS, while physicians with low CPOE usage had a 0.7% increase in both costs and LOS. 10

Brigham and Women’s Hospital, Boston, invested approximately $1.9 million to develop and implement CPOE, and its system has continuing maintenance costs of $500,000 per year. The associated ROI has been between $5 million and $10 million per year in savings. 11 Decision-support features, including prompts and alerts, contributed to these savings. 12 Prompts generated by orders for one medication resulted in dosing changes 92% of the time and saved $500,000 in charges per year. Prompts for another medication reduced ordering by 85% and related charges by $177,000 per year. Alerts implemented within the system also reduced costs; for example, the alerts decreased the time that elapsed between extreme laboratory values and clinical response from 2.1 to 0.7 hours. Alerts also provided drug-allergy warnings that were followed 70% of the time.

At Integris Health, Oklahoma City, decision-support tools for clinicians, which were made available at the point of care, are credited with helping the organization to achieve savings attributable to CIS implementation. Analysis of an area that represented 20% to 25% of the total system indicated savings of more than $3.5 million per year. Of the dollars saved, 74% were attributed to decreased labor costs, efficiency gains, and patient safety enhancements; 26%, to improved and more appropriate reimbursements. Extrapolated to the system as a whole, these savings would reach $14 million to $18 million. 13

University Community Hospital, Tampa, Fla, realized savings of $1 million to $1.3 million per year by using a CIS linked to a patient safety database and by supplying point-of-care reports. With an annual system cost of $50,000, the ROI was 20:1 for medical errors alone. These savings were sustained over 3 years. In addition, the ability to identify trends and track multiple data points allowed the hospital to eliminate other costs; in one instance, root-cause analysis pinpointed transfer delays as a cause of pressure sores (which increase LOS and costs). 14

Assessing Value

Today, there is growing skepticism about large-scale projects, accompanied by a lack of resources for these so-called big-bang implementations. CEOs want to realize the maximum ROI in the short term, “while not destroying the opportunity to move further toward an electronic record in the future.” 2

Because computer systems are dynamic, according to James, 15 it is possible, over time, to direct the growth of an information technology system in a particular (desired) direction. He suggests thinking in terms of incremental steps to create a full electronic medical record and calls a strategic plan the most important part of implementing that record. At Intermountain Health Care, Salt Lake City, James reports, “We were not able to show a return on investment for our electronic medical records systems until we combined them with our clinical improvement. Informatics builds the tools; clinical quality improvement builds the content.” 15

To ensure value, the organization’s mission must be explicitly linked to measures of performance, not just administrative functions. Such measures are essential because, as Stewart and Lockamy note, “Any effort to improve financial performance that does not consider clinical activities can offer only limited success.” 16 Traditional accounting information should be integrated with data covering patient satisfaction, unscheduled hospital readmissions, and measures of clinical quality and productivity (such as costs per care episode, LOS, infection rates, and timeliness of intervention). Any assessment of performance absolutely must include these measures, as they all, ultimately, affect value, in addition to the bottom line. There is no widely accepted model that includes them available today.

Increasingly, organizations are focusing on data-driven methods in order to improve value. An example is the quality initiative toolkit based on statistical process control and total quality management that was recently described by Hospitals and Health Networks. 17 Training is part of the cycle involved in taking this approach, which targets projects that improve performance and are linked to measurable financial results. According to John Desmarais, president and CEO of Commonwealth Health Corp, Bowling Green, Ky, an investment of $1.25 million in this program has yielded $2.9 million. 17

Whatever approach is adopted, training costs must be included. According to Joann E. Turnbull, director of the National Patient Safety Foundation, “the re-engineering of work process that’s happened in hospitals in the last few years has left people with huge skills deficits.” 18 This is compounded by the fact that health care organizations have traditionally underestimated training costs for information technology. There is increasing recognition that this underfunding and inattention to human factors, including organizational and professional development, combine to contribute to implementation failure. Today, experts 19 put the costs of training at $2 to $3 for every $1 spent on hardware and software.

With the broad dissemination of web-enabled systems and multiple access technologies, quantifying costs for information technology may become more complex. If there is a stunning ROI for using electronic data interchange for transactions, 20 information infrastructure costs may be accepted as essential to doing business. That would simplify the task of justifying the cost of information technology. For example, if physicians use handheld devices for several other purposes in addition to writing prescriptions, 16 those devices could be considered part of the infrastructure. Over the past few years, this has become the case for wireless telephones, for example.


If measuring ROI for charge capture, online services, and other administrative systems is “a tricky exercise,” 21 as Briggs puts it, then measuring ROI for clinical systems is even more demanding. Nonetheless, it is absolutely essential to controlling costs and improving quality. In health care, those two acts constitute value, and the pursuit of value is a never-ending task. In this task, information technology is only an enabler, no more-but it is an enabler that makes process redesign possible and change management mandatory. 22,23 ?

Judith V. Douglas is adjunct faculty at the Johns Hopkins University School of Nursing and co-editor of Cancer Informatics: Essential Technologies for Clinical Trials.

Marion J. Ball, EdD, is vice president of Healthlink, Inc, adjunct professor at the Johns Hopkins School of Nursing, and an elected member of the Institute of Medicine.


  1. Healthcare Information and Management Systems Society 12th Annual Leadership Survey Results for 2001. Chicago, Ill: HIMSS; 2001.
  2. Gillespie G. What lies ahead. Health Data Management. 2001;9(12):35-40.
  3. Kohn LT, Corrigan JM, Donaldson MS (eds). To Err is Human: Building a Safer Health System. Washington DC: National Academy Press; 2000.
  4. Committee on Quality Health Care in America, Institute of Medicine. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington DC: National Academy Press; 2001.
  5. Agency for Healthcare Research and Quality. Making Health Care Safer: A Critical Analysis of Patient Safety Practices. Evidence Report/Technology Assessment No 43. Rockville, MD: AHRQ; 2001 (July).
  6. Strassmann PA. The Squandered Computer: Evaluating the Business Alignment of Information Technologies. New Canaan, Conn: Information Economics Press; 1997.
  7. Metzger J, Turisco F. Computerized physician order entry: a look at the vendor marketplace and getting started. Available at: http://www.leapfroggroup.com. Accessed December 1, 2001.
  8. Bates DW, Cohen M, Leape LL, Overhage JM, Shabot MM, Sheridan T. Reducing the frequency of errors in medicine using information technology. J Am Med Inform Assoc. 2001;8:299-306.
  9. Tierney WM, Miller ME, Overhage JM, McDonald CJ. Physician inpatient order writing on microcomputer workstations: effects on resource utilization. JAMA. 1993;269:379-383.
  10. Eclipsys Corp. Sarasota Memorial Hospital documents millions in expected cost savings, reduced LOS through use of Eclipsys? Sunrise Clinical Manager. Available at: http://www.elipsnet.com. Accessed January 3, 2002.
  11. Bates DW, Leape LL, Cullen DJ, et al. Effect of computerized physician order entry and a team intervention on prevention of serious medication errors. JAMA. 1998;280:1311-1316.
  12. Teich JM, Glaser JP, Beckley RF, et al. Toward cost-effective, quality care: the Brigham integrated computing system. In: Steen EB, ed. The Second Annual Nicholas E. Davis Award: Proceedings of the CPR Recognition Symposium. New York: McGraw-Hill; 1996:19-55.
  13. Cerner Corp. INTEGRIS Health realizes substantial return on investment from Cerner solutions. Available at: http://www.prnewswire.com. Accessed December 30, 2001.
  14. Gillespie G. I.T. shows error of provider?s way. Health Data Management. Available at: http://www.healthdatamanagement.com. Accessed December 29, 2001.
  15. James BC. Information technology in support of process management. Paper presented at: Institute of Medicine Annual Meeting; Washington, DC; October 15, 2001.
  16. Stewart LJ, Lockamy A. Improving competitiveness through performance-measurement systems. Healthcare Financial Management. Available at: http://www.hfma.org/Magazine/cover.htm. Accessed December 29, 2001.
  17. Six Sigma Toolkit: the quest for quality. Hospitals and Health Networks. 2001; December (suppl).
  18. Reducing medical errors, improving patient safety: how should the healthcare industry respond? Healthcare Business Roundtable. 2000; July (suppl).
  19. Metzger J, Slye D. Inpatient e-ordering. Healthcare Informatics. 2001 (May): 63-68.
  20. Forecast 2002. Health Data Management. 2001; 9(12); (suppl).
  21. Briggs B. Many happy returns: measuring ROI. Health Data Management. Available at: http://www.healthdatamanagement.com. Accessed January 4, 2002.
  22. Ball MJ, Douglas JV, Reel SL. Measuring Benefits, Cultivating Value. In: Shapiro J, ed. Guide to Effective Healthcare Information & Management Systems and the Role of the Chief Information Officer. 3rd ed. Chicago: HIMSS, pp 83-93.
  23. Ball MJ, Douglas JV. IT, Patient Safety, and Quality of Care. JHIM. Winter 2002, in press.