DesRoches, Catherine M. DRPH; Miralles, Paola BS; Buerhaus, Peter PhD, RN; Hess, Robert PhD, RN; Donelan, Karen ScD
The modernization of the nation's health information infrastructure (HIT) has been a prominent policy priority over the past 3 years, culminating in the passage of the Health Information Technology for Economic and Clinical Health Act of 2009 (HITECH). This act provides significant financial incentives for hospitals and clinicians to use HIT to improve patient care.1,2 Clinicians can qualify for these incentives by achieving the meaningful use of electronic health records. The first stage of meaningful use, as defined by the Office of the National Coordinator for Health Information Technology and the Centers for Medicare and Medicaid Services, requires clinicians and hospitals to use HIT to (1) capture health information in a coded format, (2) track key clinical conditions and communicate that information for care coordination, (3) facilitate disease and medication management, and (4) report clinical quality measures and public health information. Clinicians and hospitals that do not achieve these goals over the next 5 years will be subject to financial penalties in the form of reduced Medicare payments. The structure of these incentives was based on the hypothesis that the meaningful use of HIT can improve quality and decrease costs. Although current research has found no effect or mixed effects on quality and cost, the impetus for adoption remains. Many policymakers predict a rapid increase in HIT adoption in the coming years.3-5
There have been numerous surveys of HIT adoption among physicians and hospitals.3,6-9 However, there are few national surveys of how nurses view and use these technologies. Given their relative size in the health care workforce and the range of services they provide, nurses' inclusion in initiatives to increase the implementation and use of HIT are likely to affect the success of these efforts.10,11
A 2008 national survey of RNs found fewer than 1 in 5 nurses employed in a setting with a basic level of HIT adoption.11 However, studies exploring RNs' perceptions of HIT and the effect on nursing efficiency and workflow report an overall positive benefits and reinforce the importance of involvement of RNs in all stages of HIT planning and implementation.12-24 Initiatives aimed at engaging nurses in HIT policymaking and the integration of informatics and nursing represent RNs as important stakeholders, with the goal of transforming nursing practice guidelines and patient care workflow into electronic, real-time modalities.25-27
This study extends a previous work11 using data from a new national survey of RNs. Specifically, the study investigates the following questions: What HIT functionalities are available to RNs in direct patient care, and does this vary by care setting? How do RNs in direct patient care view the effect of these systems on their own daily work and the quality of care they provide? Is the availability of HIT functionalities associated with less time spent on paperwork and documentation and more time spent in patient care?
Description of the Study
Data and Methods
This study was conducted as part of an evaluation of the Johnson and Johnson Campaign for Nursing's Future through a collaboration of researchers at Vanderbilt University Medical Center, Massachusetts General Hospital, and Gannett Healthcare Group. The campaign is a national initiative that began in 2002 aimed at increasing nursing recruitment in the United States, retaining nurses in clinical practice, and increasing the capacity of the nation's nursing education programs.
The survey was conducted by mail from May through August 2010 by Harris Interactive. A cover letter, the 8-page questionnaire, and an incentive for participation were mailed to a random sample of 1500 RNs drawn from a national database of RNs. Up to 4 additional mailings were sent to nonresponders to encourage participation. Response enhancement incentives included 4 months of unlimited online continuing education through Gannett Education and a $10 prepaid check sent to nurses who had not completed the survey after the third round of mailing. After exclusion of retired nurses or those not working at the time of the survey, we obtained a 56% response rate among eligible respondents. Our analysis is restricted to RNs who reported working in direct patient care (n = 532).
The 2010 RN survey contained 116 close-ended questions and 1 open-response question. Reflecting the extensive HIT policy activity at the federal level,1 the 2010 RN survey included a series of questions assessing the availability of HIT functionalities and the perceived effect of these technologies on RNs' daily work. The survey asked whether the hospital or organization had a computerized system for each of the following: electronic medical information about patients, including problem lists and key patient demographics; electronic ordering of tests, procedures, or drugs; electronic clinical and patient notes; electronic access to test results; electronic decision support, including clinical guidelines and pathways; electronic communication between providers; patient support, including patient education materials, teaching tools, and home monitoring; secure e-mail with patients; and patient internet sites (ie, "patient gateway"). Nurses reporting the availability of a patient gateway were asked if the system allowed patients to perform the following online: (1) view their medical records, (2) request a prescription refill, (3) access test results, and (4) schedule appointments. Response options were yes, no, and don't know for all questions.
To assess RNs' perceptions of the effects of HIT, the survey included the following items: "How have these computerized systems affected the quality of care in your organization?" and "How have these computerized systems affected your daily work?" Response options for the first question were "made it better," "made it worse," and "made no difference." Response options for the second question were "made it easier," "made it harder," and "made no difference."
Finally, nurses were asked to provide the average percentage of time they spent in each of the following tasks during a typical week of work: direct patient care, including hands-on care; patient/family teaching and discharge planning; educating other nurses; educating/precepting nursing students; patient care notes/documentation; patient-related telephone calls (Rx, laboratory results, referrals, etc); and others (eg, shift change, transporting patients, meetings).
The survey questions on HIT were taken from federal surveys developed to assess HIT adoption in physicians' offices. These survey items were developed in collaboration with an expert panel. For this survey, the items were revised for this project to ensure they were relevant for RNs working in hospitals. All other survey items were developed by the research team or drawn from previous surveys conducted for the Johnson and Johnson Campaign for Nursing's Future. The survey development team included several RNs who reviewed all items to ensure that they were valid for RNs.
χ2 Analysis was used to test associations between the setting of care and categorical variables-availability of functionalities and the effect of these functionalities on quality of care and daily work. We created a dichotomous care setting variable with the following categories: (1) direct patient care in a hospital and (2) direct patient care in other care settings. Other care settings included long-term or subacute care facility, home health care or community care, ambulatory (office, surgery, dialysis, or urgent care center), and school health or student health service. For brevity, these work settings are referred to as outpatient.
To examine associations between level of HIT functionality adoption and time spent in daily nursing-related tasks, we created a categorical variable that classified RNs as employed in a hospital or organization with 0 to 3 HIT functionalities (low), 4 to 6 HIT functionalities (medium), or 7 or more HIT functionalities (high). A 1-way analysis of variance was used to compare the percentage of time spent in nursing-related tasks across the 3 categories of HIT functionality adoption. Logistic regression models were used to assess RNs' perceptions of the effect of HIT on 2 outcomes of interest: quality of care (better vs worse/made no difference) and daily work of RNs (easier vs harder/made no difference). These models controlled for RN (age, education) and work setting characteristics (number of hours worked per week, work setting), and work setting location (urban, suburban, rural).
Our study has several limitations. First, as with all surveys, this RN survey is subject to sampling error, and thus, results may differ from what would be obtained if every RN in the United States had been interviewed. With a response rate of 56%, the possibility of response bias cannot be completely discounted. However, the unweighted demographic characteristics of our sample are similar to the Bureau of Labor Statistics' estimates; we are therefore confident that our respondents do not differ significantly from nurses overall. Second, because the study relies on self-reported data, we were unable to verify RNs' reports of the availability of HIT functionalities and the percentage of time spent on nursing-related tasks. In addition, some RNs many not be aware of all the HIT functionalities implemented in their organization. However, our survey design attempts to adjust for this bias by asking specifically about individual HIT functionalities. This specificity should have reduced underreporting or overreporting. Notably, all national estimates of HIT adoption are based on self-reported data.3,6-9 Finally, our measures of quality of care and the effect of HIT on the daily work of RNs are subjective and could not be verified with objective quality metrics because of the nature of the study.
Characteristics of the Sample
Similar to the characteristics of nurses overall, registered nurses eligible for inclusion in this analysis (working in direct patient care) were overwhelming white (80%) and female (95%), with an average age of 49 years. Sixty-six percent of our eligible respondents worked in hospitals, whereas 34% worked in outpatient settings. There were several statistically significant differences between nurses working in hospitals and those in outpatient settings. Nurses working in hospitals reported a mean age of 47 years, compared with 52 years among those in other care settings. Furthermore, nurses in hospitals were statistically more likely to have at least a bachelor's degree in nursing (45% vs 38%), report a higher income (32% of RNs in hospitals settings reported at least $75,000 in income, compared with 17% of RNs in outpatient settings), and work in an urban location compared with their counterparts in outpatient settings (56% vs 32%). P values for these differences were all equal to or less than .029, meaning that these results are unlikely to be owing to chance alone.
See Table, Supplemental Digital Content 1, which shows the overall characteristics of respondents, http://links.lww.com/JONA/A54.
HIT Available to Nurses in Their Work Settings
As shown in Table 1, findings from the survey items assessing the availability of HIT functionalities suggest that HIT is generally more available to nurses working in hospitals than to those working in outpatient settings. Nurses in hospitals were significantly more likely than those in outpatient settings to report that their organization had a computerized system for each of the following: electronic information about patients; ordering of tests, procedures, or drugs; clinical notes; access to test results; clinical decision support; electronic communication between providers; patient support materials; and a patient gateway. The only functionality that direct-care nurses in outpatient settings were as likely to report as hospital-based nurses was the use of secure e-mail with patients. Many RNs worked in settings with some HIT functionalities; however, comparatively fewer RNs overall worked in offices or institutions with a high level of HIT adoption.
Thirty-four percent of RNs reported 7 or more functionalities in their institution. This varied by work setting, with higher levels of HIT adoption overall reported by RNs in hospitals compared with those in other work settings. Notably 53% of RNs in outpatient settings reported fewer than 4 HIT functions in their workplace, compared with 11% of RNs in hospitals.
Clinical decision support, defined in the survey as including clinical pathways, reminders, and drug interaction alerts, was among the least commonly available functionalities across care settings. Among those RNs with access to this technology, 21% reported never or only sometimes taking action based on an alert or prompt from their computerized system. Twenty-four percent of RNs reported that they frequently or often overrode the alert and 18% frequently or often found the alert to be irrelevant. There were no differences on these measures by work setting (data not shown).
Effect of HIT on Perceived Quality of Care
Most RNs in both hospitals (64%) and outpatient settings (57%) rated the quality of nursing provided in their organization as excellent or very good, and 63% reported that the use of HIT made quality of care better. As shown in Table 2, there were few significant differences among RNs regarding their beliefs about the effect of HIT on quality of care. After controlling for RN and work setting characteristics, only RN age and average number of hours worked per week were associated with beliefs about the effect of HIT on quality of care (Table 2). Older RNs and those working a greater number of hours per week were significantly less likely than younger RNs and those working fewer hours to rate the effect of HIT on quality of care as positive.
Perceptions of quality of care were also associated with the number of HIT functionalities present in the organization (Figure 1). RNs employed in organizations with 7 or more HIT functionalities were significantly more likely than those with fewer functions to rate the overall quality of nursing care in their organization as excellent or very good. Furthermore, these RNs were significantly more likely than those working in organizations with fewer functions to report that the quality of care in their organization had improved a lot or some in the past 2 years.
Effect of HIT on the Daily Work of RNs
Nurses were asked to provide an estimate of the percentage of their overall time at work spent in direct patient care and other related tasks. The presence of HIT did not seem to be associated with significant differences in how nurses reported spending their time at work, with 2 exceptions. RNs in outpatient settings reported a significantly higher average percentage of time spent on patient care notes and documentation and less time spent on patient care when they were employed in settings with a greater number of HIT functionalities. As shown in Table 3, RNs in outpatient settings on average spent 20% of their time on documentation. In settings with fewer than 4 HIT functions, RNs reported spending 18% of their time in documentation, compared with 24% among RNs with between 4 and 6 functions, and 22% among those with more than 6 functions. In addition, findings suggest that RNs in outpatient settings with a greater number of HIT functionalities seem to spend less time in direct patient care. In both cases, the distribution is U shaped, with RNs in outpatient setting with a moderate level of HIT adoption (4-6 functionalities) reporting the most time spent in documentation and the least time spent in patient care.
Registered nurses had an overall positive impression about the effect of HIT on their daily work. Approximately one-half of RNs reported that the use of HIT made their daily work easier. As shown in Table 2, there were few significant differences among RNs regarding their beliefs of the effect of HIT on their daily work. After controlling for level of education, average number of hours worked per week, work setting, and practice location, only age was significantly associated with beliefs about HIT. Older RNs were significantly less likely than younger RNs to believe that HIT made their daily work easier.
This study examines the availability of HIT functionalities in RNs' work settings and the perceived effect of these functionalities on quality of care and the daily work of RNs and finds wide variation in the availability of specific HIT functionalities. This has important implications for policymakers as they begin to implement the meaningful use incentive program and fund the regional extension centers (established to assist clinicians to achieve meaningful use by providing ongoing technical support) mandated under the HITECH act. Many RNs are working in hospitals and organizations that have some but not all of the necessary functions. Focusing on the functionalities that have not been widely adopted, such as clinical decision support and those that increase patient satisfaction with physician and nursing care,28 including patient gateways, may be an effective and efficient way to ensure that hospitals and clinicians are able to achieve meaningful use and maximize the effect of investing in new HIT systems.
Several studies have shown an increase in documentation time during the initial HIT implementation period.12,14 Our study suggests that RNs employed in hospitals with a higher level of HIT adoption do not spend more time in documentation than RNs employed in hospitals with fewer HIT functionalities. In fact, the findings suggest that these nurses spend fewer hours in documentation; however, this finding was not statistically significant. This does not seem to be the case among RNs in outpatient settings where RNs in organizations with moderate levels of HIT availability spending more time on average in documentation tasks and less time in patient care than those with low levels of adoption. As the level of adoption increases, estimates of time spent documenting begin to trend back toward the levels report by RNs in low HIT settings. There are several possible explanations for this finding. It may be that outpatient settings with a moderate level of adoption use a "part paper, part electronic system." This can result in inefficiencies leading to more paperwork and less time in patient care. Health information technology systems used in outpatient settings may be more cumbersome than those in hospital systems, or RNs may be required to use multiple systems for billing, scheduling, and clinical care. Alternatively, hospital systems may have been in place for a longer period, giving RNs a greater opportunity to learn to use the system efficiently. We were unable to control for these factors in our analysis. Regardless of the cause, organizational leaders, including nursing leadership, must be aware of this potential increase in documentation time when planning to implement new HIT functionalities. Furthermore, they must ensure that RNs have sufficient training to learn to use the system in an optimal way.
Similar to earlier work suggesting important differences between older and younger nurses,29 our study finds older nurses significantly less positive than younger nurses about the use of HIT, both in its effect on quality and that on daily work. Current estimates place the median age of RNs at 46 years, with approximately 40% of employed RNs aged 50 years and older.10 Given the age and size of the older nursing workforce, this could present a significant problem as hospitals and outpatient settings implement these new technologies. If the older nursing workforce is resistant to the changes in practice and workflow that must occur for these technologies to be used optimally, organizations may fall short of achieving true meaningful use.
Over the next 5 years, policymakers expect to see a significant increase in HIT adoption across all sectors of health care. The recent health care reform bill requires a level of quality and safety reporting that will be difficult, if not impossible, to achieve without the widespread implementation of such systems. Furthermore, clinicians and hospitals that adopt or upgrade their systems stand to gain financially and those who do not may incur financial penalties. These policy changes foretell increasing activity around HIT adoption and implementation in hospitals and outpatient settings nationwide. It is likely that RNs will play a significant role in the success or failure of these efforts. Ensuring that HIT systems are relevant to and usable by RNs is within the purview of nursing leadership at the executive and midmanagement levels. Initiatives such as the Technology Informatics Guiding Education Reform (TIGER), which aims to "enable practicing nurses… to fully engage in the unfolding digital electronic era," can play an important role in helping to ensure that RNs are fully engaged in this transformation. RNs, in both administrative and clinical roles, will be critical to the success of these efforts and in achieving the ultimate goal of improved quality, safety, and efficiency in health care delivery.
2. Blumenthal D. Launching HITECH. N Engl J Med
3. DesRoches CM, Campbell EG, Rao SR, et al. Electronic health records in ambulatory care-a national survey of physicians. N Engl J Med
4. Amarasingham R, Plantinga L, Diener-West M, Gaskin DJ, Powe NR. Clinical information technologies and inpatient outcomes: a multiple hospital study. Arch Intern Med
5. Keyhani S, Hebert PL, Ross JS, Federman A, Zhu CW, Siu AL. Electronic health record components and the quality of care. Med Care
6. Jha AK, DesRoches CM, Campbell EG, et al. Use of electronic health records in U.S. hospitals. N Engl J Med
8. Hsiao C, Beatty PC, Hing ES, Woodwell DA, Rechtsteiner EA, and Sisk JE. Electronic medical record/electronic health record use by office-based physicians: United States, 2008 and preliminary 2009. Available at http://www.cdc.gov/nchs/data/hestat/emr_ehr/emr_ehr.pdf
. Accessed November 23, 2010.
9. Gans D, Kralewski J, Hammons T, Dowd B. Medical groups' adoption of electronic health records and information systems. Health Aff (Millwood)
10. Bureau of Labor Statistics. Occupational Outlook Handbook
. 2010-2011 Edition, Registered Nurses. Available at http://www.bls.gov/oco/ocos083.htm
. Accessed November 23, 2010.
11. DesRoches C, Donelan K, Buerhaus P, Zhonghe L. Registered nurses' use of electronic health records: findings from a national survey. Medscape J Med
12. Waneka R, Spetz J. Hospital information technology systems' impact on nurses and nursing care. J Nurs Adm
13. Asaro PV, Boxerman SB. Effects of computerized provider order entry and nursing documentation on workflow. Acad Emerg Med
14. Bolton LB, Gassert CA, Cipriano PF. Technology solutions can make nursing care safer and more efficient. J Healthcare Inf Manage
15. Dykes PC, Carroll DL, Benoit A, et al. A randomized trial of standardized nursing patient assessment using wireless devices. AMIA Annu Symp Proc
16. Gunningberg L, Fogelberg-Dahm M, Ehrenberg A. Improved quality and comprehensiveness in nursing documentation of pressure ulcers after implementing an electronic health record in hospital care. J Clin Nurs
17. Lindgren CL, Elie LG, Vidal EC, Vasserman A. Transforming to a computerized system for nursing care: organizational success within Magnet idealism. Comput Inform Nurs
18. Oroviogoicoechea C, Elliott B, Watson R. Review: evaluating information systems in nursing. J Clin Nurs
19. Poissant L, Pereira J, Tamblyn R, Kawasumi Y. The impact of electronic health records on time efficiency of physicians and nurses: a systematic review. J Am Med Inform Assoc
20. Ralston JD, Coleman K, Reid RJ, Handley MR, Larson EB. Patient experience should be part of meaningful-use criteria. Health Aff (Millwood)
21. Sassen EJ. Love, hate, or indifference: how nurses really feel about the electronic health record system. Comput Inform Nurs
22. Thompson D, Johnston P, Spurr C. The impact of electronic medical records on nursing efficiency. J Nurs Adm
23. Westra BL, Subramanian A, Hart CM, et al. Achieving "meaningful use" of electronic health records through the integration of the nursing management minimum data set. J Nurs Adm
24. Moody LE, Slocumb E, Berg B, Jackson D. Electronic health records documentation in nursing: nurses' perceptions, attitudes, and preferences. Comput Inform Nurs
28. Donelan K, Rao SR, Miralles PD, Buerhaus PI, Dutwin D, DesRoches CM. U.S. Public perceptions of the use of electronic health records in physician offices. J Am Med Inform Assoc
. In press.
29. Norman LD, Donelan K, Buerhaus PI, et al. The older nurse in the workplace: does age matter? Nurs Econ
. 2005;23(6):282-289, 279.
© 2011 Lippincott Williams & Wilkins, Inc.