Anesth Analg 2008; 107:1323-1329
© 2008 International Anesthesia Research Society
doi: 10.1213/ane.0b013e31818322d2
ECONOMICS, EDUCATION, AND POLICY
Adoption of Anesthesia Information Management Systems by Academic Departments in the United States
Christoph B. Egger Halbeis, MD, DEAA, MBA*,
Richard H. Epstein, MD, CHPIMS ,
Alex Macario, MD, MBA*,
Ronald G. Pearl, MD, PhD*, and
Zvi Grunwald, MD
From the *Department of Anesthesiology, Stanford University School of Medicine, Stanford, California; and Department of Anesthesiology, Jefferson Medical College, Philadelphia, Pennsylvania.
Address correspondence and reprint requests to Christoph B. Egger Halbeis, MD, DEAA, MBA, 300 Pasteur Drive, Stanford, CA 94305. Address e-mail to cbeh{at}stanford.edu.
Abstract
BACKGROUND: Information technology has been promoted as a way to improve patient care and outcomes. Whereas information technology systems for ancillary hospital services (e.g., radiology, pharmacy) are deployed commonly, it has been estimated that anesthesia information management systems (AIMS) are only installed in a small fraction of United States (US) operating rooms. In this study, we assessed the adoption of AIMS at academic anesthesia departments and explored the motivations for and resistance to AIMS adoption.
METHODS: Members of the Society of Academic Anesthesiology Chairs and the Association of Anesthesiology Program Directors were solicited by e-mail to participate in an online survey of AIMS adoption. Two months after closing the survey, another e-mail was sent with a single question asking for an update to their AIMS implementation status.
RESULTS: Surveys were fully completed by 48 (34%) of the 140 Society of Academic Anesthesiology Chairs and Association of Anesthesiology Program Directors departments surveyed, with 72 (51%) providing AIMS status information. Twenty of these 72 departments have an AIMS installed, 12 are currently implementing, 11 have selected but not yet installed, and 18 are planning to purchase an AIMS in 2008 or 2009. Therefore, at least 61 (44%) of all 140 US academic anesthesia departments have committed to AIMS. This estimated adoption rate is conservative because the numerator equals the affirmative responses, whereas the denominator equals the total population of academic departments. Among adopters, the top ranked anticipated benefits from installing an AIMS included improved clinical documentation, improved data collection for clinical research, enhancement of quality improvement programs, and compliance with requirements of regulatory authorities. The hospital provided funding in almost all facilities (90%), with co-funding by the anesthesia group in 35%.
CONCLUSIONS: At least 61 or 44% of the 140 US academic departments surveyed in this study have already implemented, are planning to acquire, or are currently searching for an AIMS. Adoption of AIMS technology appears to have reached sufficient momentum within academic anesthesiology departments to result in a fundamental change.
The use of information technology (IT) has been promoted widely as a way to improve patient care and outcomes.1–3 Based on the 2007 annual report by the Healthcare Information and Management Systems Society, IT systems for ancillary hospital services in general (e.g., radiology, pharmacy) have been deployed almost universally, whereas other systems less so. For example, a recent Healthcare Information and Management Systems Society survey covering an extensive menu of health care IT categories reported that computerized physician order entry systems are used in 43% of hospitals.4 However, this report did not even mention of the presence of anesthesia information management systems (AIMS).
AIMS have been estimated to be installed in <5% of operating rooms in the Unites States (Epstein R, personal communication, unpublished data).5 If one accepts the hypothesis that health care IT improves patient care, this lag in the adoption of AIMS is surprising, as anesthesiology has been touted as a model specialty in American medicine for the adoption of patient safety initiatives.6,7 There are many potential barriers to AIMS adoption that might explain this gap, including acquisition and maintenance costs, inadequate return on investment (ROI) for the hospital, complexity of implementation, lack of interoperability with other health information systems, immaturity of the software, and lack of proven benefits.*
The goals of this study were to assess the adoption of AIMS by academic anesthesia departments and to explore the motivations for and barriers to adoption. For the purpose of this study, academic anesthesia departments were defined as either those affiliated with a medical school or those departments with a residency program, but not affiliated with a medical school. Understanding motivations to adopt AIMS and barriers preventing such adoption is valuable for developing strategies to facilitate adoption, and may foster additional research that may either validate or mitigate such resistance. We hypothesized that those anesthesia departments committing to AIMS technology, compared with AIMS nonadopters, are more likely to be larger (in terms of the number of anesthetics per year), are part of the hospital's organization (as opposed to being an independent organization), and belong to a hospital that has already implemented a number of other health care IT systems. We chose to study academic departments because they are often at the forefront of new patient care and safety initiatives.
METHODS
This study was approved by the authors' IRBs.
Survey Development and Validation
A review of the literature on AIMS identified possible reasons for adoption and resistance to address in the survey. Content validity was provided by repetitive rounds of refinement of questions (and potential answers) among the authors and through feedback from clinical AIMS administrators at several institutions. Feedback also came from a pilot study with a convenience sample of six anesthesiologists.
A logic-branched survey design (available at www.anesthesia-analgesia.org) was chosen. This allowed for irrelevant questions based on a prior response to be eliminated. For example, if the respondent indicated that an AIMS was not installed, the question "Which was the year of install of the AIMS" would not be presented. In situations in which duplicate responses were submitted from a department, the most recent response was compiled in the final results.
Survey Administration
In late September 2007, an e-mail requesting completion of our web-based survey (hosted by Vovici Inc., Dulles, VA) was distributed using the combined e-mail list manager of the Society of Academic Anesthesiology Chairs (SAAC) and the Association of Anesthesiology Program Directors (AAPD) to its members. A follow-up request was emailed 1 mo later. Six weeks after the initial request, the web survey site was closed.
Eight weeks after closure of the survey site, another e-mail was sent via the SAAC/AAPD e-mail list manager offering all members a preliminary report of the findings. In fulfilling this request, respondents were asked to optionally indicate their current status of AIMS adoption. This strategy was planned in advance to generate a larger sample size, expecting that making available the preliminary report and asking a single question would increase the response rate.
Responders were verified to be from SAAC or AAPD institutions and duplicate responses were identified by examining their Internet protocol address, organization to which they belonged, and the online list of current SAAC/AAPD members. Subsequently, the responses were de-identified for analysis. The survey population comprised 112 academic anesthesia departments and 28 anesthesiology residency programs not associated with a school of medicine.
Analysis
We evaluated if academic anesthesia departments committing to AIMS were more likely than nonadopters to have a larger anesthetic case volume, to be part of the hospital's organization (as opposed to being an independent entity), and to belong to a hospital that has already implemented a number of other health IT systems.
We defined adoption of AIMS technology as a response indicating that the department had already implemented, was in the process of implementing, had selected but not yet installed, or was searching for or about to begin looking for an AIMS. We defined nonadoption as a response indicating no plans to implement or when respondents did not report their status.
To assess the penetration of IT in the two subgroups (AIMS adopters and nonadopters), a relative IT penetration score (rITPS) was computed for each respondent's hospital. The rITPS reflects the fraction of implemented health care IT systems assessed in our survey (Appendix, question 3).
Categorical differences between adopters and nonadopters were analyzed by determining the confidence interval for the difference between two independent proportions.8 Differences in the rITPS were analyzed using the unpaired two-tailed Student's t-test, with P < 0.05 required to claim significance. Descriptive statistics are reported as the mean, standard deviation, or the median.
RESULTS
Responses to the Initial Survey
During the initial online survey, responses were received from 48 of the 140 SAAC/AAPD departments: 44 responses from the 112 SAAC departments and 4 from the 28 nonoverlapping AAPD departments. Thirty-three of the initial 48 respondents were adopters of AIMS technology, whereas 14 were not engaged in the process of choosing a system, and no respondent refused to indicate their status. Eleven departments had already implemented an AIMS, 16 were currently implementing such a system, and six were in the process of selecting an AIMS. No respondents were piloting an AIMS.
Of the 15 nonadopters, funds were unavailable for 11 and nonfinancial resource limitation (e.g., lack of hospital administration support) was a barrier for another four. None of the respondents indicated resistance by the staff anesthesiologists as a reason for nonadoption. One anesthesia department deferred adopting the AIMS selected by the hospital as part of the hospitals integrated clinical information system because the AIMS was perceived by the department to be "immature."
Responses to the Follow-up Question
In the follow-up assessing current AIMS status, 72 responses were received: 44 from the original respondents along with responses from 24 additional departments (four duplicate responses were excluded). The status of several of the 44 previous respondents had changed relative to the original response and that most recent response was used for the analysis of the combined responses. For example, one department initially reporting lack of financial resources as a reason for nonadoption changed its status to "refuse to disclose," and three departments initially indicating they were implementing an AIMS in the original response changed their status to "AIMS installed" in the follow-up.
Of the 24 new responses in the follow-up, 23 were AIMS adopters, with five who had implemented an AIMS, three who were currently implementing, four who had selected but not yet implemented an AIMS, six who were in the process of selecting, and five who were planning to start a search for an AIMS within the next 2 yr.
Analysis of the Combined Responses
Therefore, information about AIMS status was received from 72 unique departments or 51% of the 140 SAAC/AAPD departments (Fig. 1). To ensure the estimated adoption rates are conservative, the numerator of the adoption rates was based on the responses, whereas the denominator was considered to be the total population of SAAC/AAPD departments. As of February 29, 2008 (the study closure date), AIMS had been implemented in at least 20 US academic departments, and an additional 12 were in the process of AIMS implementation. Including those departments searching for an AIMS (11) or planning to begin a search in 2008 or 2009 (18), at least 61 or 44% of the 140 US academic departments had committed to AIMS.
View larger version (21K):
[in this window]
[in a new window]
|
Figure 1. Anesthesia information management system (AIMS) adoption by United States (US) academic anesthesia departments. The status of AIMS adoption was surveyed from all 140 US anesthesia departments affiliated with a medical school, or with a residency program but not so affiliated. As of February 29, 2008, AIMS had been implemented in 20 of 72 responding US academic anesthesia departments, and were in the process of being implemented in an additional 12. Including those departments searching for an AIMS or planning to begin a search in 2008 or 2009, a total 61 of the responding academic anesthesia departments had decided to commit to AIMS.
|
|
Comparisons Between AIMS Adopters and Nonadopters in the Initial Survey
The unexpectedly high percentage of AIMS adoption (71%) by programs completing the initial AIMS survey meant only a small number of nonadopter facilities responded. This resulted in large confidence intervals (all of which included 0) and lack of statistical significance for the differences tested between the groups.
Of the 48 SAAC members who responded to the initial survey, 15 anesthesia departments were part of the hospital's organization and 33 belonged to a different corporate entity. Thirteen of 15 departments that are part of the hospital's organization were adopters, compared to 21 of 33 of the other departments. Twenty-one of 27 larger departments (defined as >20,000 anesthetics performed per year) were adopters, compared to 13 of 20 of smaller departments (<20,000 anesthetics per year).
IT systems deployed at respondents' institutions are shown in Table 1. For AIMS adopters and nonadopters, the mean rITPS were not statistically different (0.68 and 0.58, respectively).
Thus, we were unable to demonstrate an association between hospital size, hospital affiliation, and IT penetration and AIMS adoption.
AIMS Funding
The hospital provided funding in almost all facilities (90%), with co-funding by the anesthesia group in 35% (Table 2). The anesthesia department's School of Medicine rarely contributed funds to the purchase. The hospital chief executive officer and the anesthesia chair were most often involved with the financial decision to purchase the AIMS, with almost no participation by the University or School of Medicine.
View this table:
[in this window]
[in a new window]
|
Table 2. Parties Involved in the Funding and the Financial-Decision Making Process for the Purchase of the Anesthesia Information Management Systems (AIMS) (AIMS Adopters, N = 34)
|
|
AIMS represented at respondents' facilities were a cross-section of currently available systems, including several home-grown solutions.
Previous AIMS implementations at respondents' facilities between 1997 and 2007 demonstrated a slow but steady increase in the cumulative number of implementations (Fig. 2). If one considers the anesthesia departments currently implementing AIMS to become operational in 2008 and those who have already selected, but not yet started implementation likely to be put into production in 2009, there is a clear inflection point in 2007.
View larger version (7K):
[in this window]
[in a new window]
|
Figure 2. Implemented anesthesia information management systems (AIMS) in United States (US) academic anesthesia departments. The values for 2008 and 2009 are estimated based on the assumption that departments that are in the process of implementing their system will do so during 2008, and that departments that have purchased their AIMS but had not yet begun installation will do so during 2009. The numbers plotted are conservative in that information on AIMS status is missing from 49% of non-responding programs. The inflection point in the graph at 2007 suggests that a "tipping point" has been reached with respect to adoption of AIMS technology by US academic anesthesia departments.
|
|
Only 6 of 34 (18%) reported that the price differences among systems and vendors was a major consideration in the purchase of the system.
The distribution of the number of clinical AIMS workstations (including holding area, operating room, and postanesthesia care unit sites) installed and the ones to be installed at institutions currently implementing had a range of 8 to 200 and a median of 40.
Motivation for Implementing AIMS and Realization of Benefits
The 11 SAAC/AADP departments, who in the initial survey indicated that they had already implemented an AIMS, served as the subgroup to analyze how well AIMS functionality matched their expectations.
The top ranked (mean Likert score  4) anticipated benefits for AIMS adoption were improved clinical documentation, improved data collection for quality improvement programs, improvement of patient care and safety, improved data collection for clinical research, compliance with requirements of regulatory authorities, and the reduction of recordkeeping workload (Table 3). These benefits were rated highly as having been achieved (Table 4). In addition, the AIMS were rated highly as to user acceptance, reliability, improvement of billing processes, and reporting purposes.
View this table:
[in this window]
[in a new window]
|
Table 3. Major Motivating Factors for Installing an Anesthesia Information Management Systems (AIMS) Before Implementation
|
|
View this table:
[in this window]
[in a new window]
|
Table 4. Please Rate Each of the Following Items for the time After Anesthesia Information Management Systems (AIMS) Implementation (Early Adopters)
|
|
Barriers to Implement AIMS
The two most important financial barriers for AIMS adoption were a general limitation of funds for IT projects and other IT systems competing with the AIMS implementation for funding (Table 5). For the three departments reporting nonfinancial barriers for implementing an AIMS, lack of support from the hospital administration was the most important nonfinancial barrier for implementing an AIMS.
For both open-ended questions, "Why do you think AIMS installs lag behind installs of other hospital IT systems?" and "Why do you think some anesthesia departments decide to adopt AIMS whereas others don't?," the high AIMS purchase cost (adopters), and the small value of AIMS for the hospital (adopters and nonadopters) were mentioned most often (Table 6). Interestingly, almost half of the current AIMS users concluded that AIMS implementation lags because of the lack of a clear ROI.
DISCUSSION
At least 61 or 44% of the 140 US academic departments surveyed in this study have already implemented, are planning to acquire, or are currently searching for an AIMS. This estimated adoption rate is conservative, in that the numerator is based on the number of affirmative responses (61), whereas the denominator equals the total population of academic departments (140). AIMS implementation may increase further if financial barriers and competition for resources from other IT projects can be overcome. We were unable to demonstrate an association between AIMS adoption and hospital size, hospital affiliation, or IT penetration, perhaps in part due to the sample size.
Based on the prior 5% estimate of AIMS adoption, the percentage of current use and implementation of AIMS (23%) in academic anesthesia departments responding to the survey was unexpectedly high, as was the additional 21% of programs who are in the process of looking for an AIMS. The graph of total implementations over time (Fig. 2) suggests that 2007 represents a "tipping point" in the adoption of AIMS9: 30 yr after the first reported AIMS implementation in the 1970s, sufficient momentum has been achieved to result in a fundamental change.10
For adopters, expectations for AIMS installations were generally met and included improved clinical documentation, data collection for clinical research, enhancement of quality improvement programs, and regulatory compliance.
As indicated above, financial and other commitments by the hospital appear to be prerequisites for AIMS adoption. AIMS adoption would be even higher if institutions desiring the technology were able to overcome financial barriers or competing IT projects. O'Sullivan et al. identified a typical scenario in which the anesthesiology department would like to implement an AIMS, but the hospital refuses to provide funds.11 One explanation for this resistance is that the hospital believes such expenditures benefit the anesthesia department but not the hospital. To overcome such objections, anesthesia departments will likely need to more strongly state the case for AIMS implementation among the competing projects. The fact that our survey found that the expectations of adopters were generally met may be used to obtain hospital funding of AIMS. Arguments based on ROI alone may not be effective in obtaining hospital financing of AIMS.
Our study has several limitations: First, our sample was nonrandom, and thus the results are subject to responder bias. Therefore, we cannot extrapolate our adoption rates to the entire population of SAAC/AAPD departments. It is possible that the true adoption rate in the nonresponding departments is higher or lower than what we reported. However, the adoption of AIMS is no <44% since this figure makes the conservative assumption that all nonresponders are nonadopters. Second, because the survey was restricted to academic anesthesia departments, the AIMS adoption rate cannot be extrapolated to nonacademic settings. This is especially relevant since nonacademic centers usually lag in health care IT implementation.4 Considering that at least 44% of academic anesthesia departments have decided to adopt AIMS, it is likely that nonacademic hospital anesthesia departments will follow their lead. Finally, in order to assess the penetration of IT systems at SAAC/AAPD hospitals, we assumed that the person completing the survey would be either familiar with all IT systems used in the hospital or would ask someone able to provide the answer. However, we cannot assure this, and the results on the IT system penetration must be interpreted accordingly.
In summary, our survey demonstrated that AIMS are currently implemented in at least 14% of academic anesthesia departments, and that at least 30% of other programs are in the process of implementing or selecting a system. We conclude that a watershed has been reached for AIMS technology adoption.
ACKNOWLEDGMENTS
The authors thank Nicole Bradle and Jeff Shulz, ASA staff, for their assistance in communicating with the SAAC/AAPD members.
Footnotes
*Monk TG, Hurell M, Norton A. Toward Standardization of Terminology in Anesthesia Information Management Systems: Anesthesia Patient Safety Foundation. http://www.apsf.org/initiatives/infosys.mspx; last accessed on February 01, 2008. 
SAAC/AAPD Departmental Details. Society of Academic Anesthesiology Chair, 2008. http://www.aapd-saac.org/membersdetails. php, last accessed on February 29, 2008.
Accepted for publication June 4, 2008.
Supported by Departmental Sources.
Christoph Egger Halbeis and Richard Epstein are unpaid members of the Epic Anesthesia Advisory Council and Dräger Medical InnovianTM Advisory Board, respectively.
REFERENCES
- Bates DW, Gawande AA. Improving safety with information technology. N Engl J Med 2003;348:2526–34[Free Full Text]
- Johnston ME, Langton KB, Haynes RB, Mathieu A. Effects of computer-based clinical decision support systems on clinician performance and patient outcome. A critical appraisal of research. Ann Intern Med 1994;120:135–42[Abstract/Free Full Text]
- Hunt DL, Haynes RB, Hanna SE, Smith K. Effects of computer-based clinical decision support systems on physician performance and patient outcomes: a systematic review. JAMA 1998;280:1339–46[Abstract/Free Full Text]
- 2007 Annual Report of the U.S. Hospital IT Market. Healthcare Information and Management Systems Society (HIMSS) and HIMSS AnalyticsTM, Chicago, IL, 2007
- Epstein RH, Vigoda MM, Feinstein DM. Anesthesia information management systems: a survey of current implementation policies and practices. Anesth Analg 2007;105:405–11[Abstract/Free Full Text]
- Kohn LT, Corrigan JM, Donaldson MS. To err is human-building a safer health system. Washington, DC: National Academy Press, 2000
- Gaba DM. Anesthesiology as a model for patient safety in health care. BMJ 2000;320:785–8[Free Full Text]
- Newcombe RG. Interval estimation for the difference between independent proportions: comparison of eleven methods. Stat Med 1998;17:873–90[Web of Science][Medline]
- Gladwell M. The Tipping Point: how little things can make a big difference. Little Brown, 2000
- Zollinger RM Jr, Kreul JF, Schneider AJ. Man-made versus computer-generated anesthesia records. J Surg Res 1977;22:419–24[Web of Science][Medline]
- O'Sullivan CT, Dexter F, Lubarsky DA, Vigoda MM. Evidence-based management assessment of return on investment from anesthesia information management systems. AANA J 2007;75: 43–8[Medline]
This article has been cited by other articles:
|
|
|
|
 
J. H. Lipschutz, J. T. Mueller, T. L. Trentman, I. D. Miziara, A. K. Jha, C. DesRoches, and S. Rosenbaum
Electronic health records in hospitals.
N. Engl. J. Med.,
July 23, 2009;
361(4):
421 - 422.
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. Macario
Truth in Scheduling: Is It Possible to Accurately Predict How Long a Surgical Case Will Last?
Anesth. Analg.,
March 1, 2009;
108(3):
681 - 685.
[Full Text]
[PDF]
|
|
|
|
