Analysis of Health IT

Health IT refers to a variety of electronic tools for use in the management of health information, including the electronic medical record (EMR) and computerized physician order entry (CPOE). Policy options focus on approaches to accelerate adoption, including financial incentives, direct provision, regulatory mandates, development of standards, and enhancing the interoperability of health IT.

These are the nine performance dimensions against which we measured Health IT:

Spending

Modeling predicts that increased health IT adoption and connectivity will decrease health spending in the long run, but little empirical evidence exists to support this. Substantial new investments will be required:

  • Based on a small number of case studies and microsimulation modeling, researchers predict that increasing health IT adoption and connectivity will decrease health spending by enabling administrative simplification, process improvements, and business transformation. Read more below
  • Achieving the goals of increased health IT adoption and connectivity requires significant investments, and the savings may not be uniformly realized: Payers will see more savings than the purchasers of health IT in the current U.S. health care system. Read more below
  • To date, health IT adoption and the establishment of national connectivity have proceeded slowly, and the overall effect of this policy option on health spending is not fully understood. Read more below

Based on a small number of case studies and microsimulation modeling, researchers predict that increasing health IT adoption and connectivity will decrease health spending by enabling administrative simplification, process improvements, and business transformation.

Theoretically, health IT adoption and connectivity will facilitate improvements in the efficiency of health care delivery, as IT has done in some other industries, such as banking and retail. The means by which efficiencies are realized include administrative simplification, process improvements, and wholesale business transformation.

Few empirical studies have been conducted on the effects of health IT adoption and overall spending. A systematic review of the literature on health IT found that limited quantitative research exists in this area, that the systems studied were often very heterogeneous, and that limited financial information was available (Chaudhry et al., 2006). Although several health systems have shown that health IT can improve efficiencies, the authors of the literature review question whether the findings from these few institutions can generalize to other organizations and the health system as a whole. Given this lack of empirical evidence, researchers have used microsimulation modeling to estimate the effect on spending. Walker et al. (2005) estimated the value (net costs and benefits) of electronic health care information exchange and interoperability. They found that, with full interoperability among providers and other health care organizations over ten years, the net savings would be $77.8 billion per year.

RAND performed an analysis of the potential savings and costs of health IT (Girosi, Meili, and Scoville, 2005). The researchers postulated that the widespread adoption of interoperable health IT should reduce overall health care spending through several mechanisms. Cross-provider access to health records should reduce duplicate tests, reduce adverse drug events, and improve the utilization of medications. IT-based reminders about preventive care and support for chronic disease management should improve patient health, reducing the costs of more expensive interventions. Reductions in medical records storage, retrieval, and transcription costs, and improved scheduling are additional potential savings.

RAND's estimate of the health care efficiency savings is approximately $80 billion annually after a 15 year implementation period. This estimate is based on models that extrapolate the limited evidence to date and assume 90 percent adoption of interoperable health IT. (The estimates used in the study were not "best case," because the authors used a mean value of several observations, but neither where they "worst case," because the authors did not include results from apparently unsuccessful or badly used implementations of health IT, reasoning that ineffective technologies would not be adopted over the long run.) The RAND study estimated that savings would accrue to:

  • Outpatient settings ($20 billion/year) through improvements in drug use, laboratory and radiology use, chart administration, and patient scheduling;
  • Inpatient settings ($60 billion/year) through reductions in length of stay, time spent by nurses in administrative activities, time required for medical records administration, and use of laboratory and radiology.

The average cost of adoption during a 15 year period was estimated to be $7.6 billion per year, and the average savings estimated during that period was $41.8 billion per year. Again, these savings are based on the assumption of 90 percent adoption, with more of the savings being accrued toward the end of the 15 year period.

Another way to estimate the potential effect of health IT on health care spending is to examine the effect of IT on efficiency in other industries and to estimate aggregate efficiency gains similar to those found in other industries. During the late 1990s and continuing into this century, those industries recorded 6 to 8 percent annual productivity growth, of which economists agree that at least one-third to one-fourth could be attributed to IT. The figure superimposes a range of productivity improvements on a plot of estimated growth in national health care spending from 2002 to 2016. The smaller improvement (1.5 percent per year) is similar to the productivity gains in retail and wholesale attributed to IT; the upper end (4 percent per year) is half the IT enabled gains in telecommunications. Either level of productivity improvement could greatly reduce national health care spending, $346 billion per year at the lower end and $813 billion at the upper end.

Possible Effects of Health IT Adoption on the National Health Expenditure (NHE)

Possible Effects of Health IT Adoption on the National Health Expenditure (NHE)

SOURCE: Data are from Centers for Medicare and Medicaid Services, not dated.

Achieving the goals of increased health IT adoption and connectivity requires significant investments, and the savings may not be uniformly realized: Payers will see more savings than the purchasers of health IT in the current U.S. health care system.

The average cost of adoption during a 15 year period was estimated to be $7.6 billion per year. The study (Girosi, Meili, and Scoville, 2005) assumed relatively uniform investments across the 15 years.

The expected savings from increased health IT adoption and connectivity will accrue primarily to payers as opposed to providers, who are the purchasers of health IT. This is a serious barrier to adoption. Closed health care systems (such as health maintenance organizations [HMOs]) and physicians with a significant fraction of capitated patients (i.e., typically, patients insured by HMOs in which a flat or limited fee is paid per patient, regardless of the extent of the medical services provided to the patient) are more likely to receive the savings benefits of health IT. Smaller providers and others with limited investment capital or time are likely to be slower adopters and realize benefits much later, if at all.

To date, health IT adoption and the establishment of national connectivity have proceeded slowly, and the overall effect of this policy option on health spending is not fully understood.

Some parts of the health care system are adopting at a slower pace than necessary to get to 90 percent adoption within 15 years. We do not know how rapidly or evenly the adoption of health IT will occur, and we do not know what will motivate interoperability and connectivity in the system. The business case for community and national connectivity has not been made and is slowing the widespread interconnections of providers. Ongoing empirical work would increase our understanding of the likely effect on health care spending associated with widespread adoption of health IT. Currently, community level experiments are under way that may provide more evidence for making estimates in the future.

References

  • Chaudhry B, Wang J, Wu S, Maglione M, Mojica W, Roth E, Morton SC, Shekelle PG, "Systematic Review: Impact of Health Information Technology on Quality, Efficiency and Costs of Medical Care," Annals of Internal Medicine [Epub April 11, 2006], Vol. 144, No. 10, May 16, 2006, pp. 742-752.
  • Centers for Medicare and Medicaid Services, National Health Expenditures: National Health Expenditure Data, "Historical," and "Projected," not dated. As of November 18, 2008: http://www.cms.hhs.gov/nationalhealthexpenddata/
  • Girosi F, Meili RC, Scoville R, Extrapolating Evidence of Health Information Technology Savings and Costs, Santa Monica, Calif.: RAND Corporation, MG-410-HLTH, 2005. As of June 19, 2009: http://www.rand.org/pubs/monographs/MG410/
  • Walker J, Pan E, Johnston D, Adler-Milstein J, Bates DW, Middleton B, "The Value of Health Care Information Exchange and Interoperability," Health Affairs, Web Exclusives, January 19, 2005, pp. w5.10-w5.18.

Back to top

Consumer Financial Risk

We know of no studies or theory that has examined the effect of increased health IT adoption and connectivity on consumer financial risk:

  • In theory, if health IT improves data and decision support for providers, it could reduce financial risk for the consumer by reducing unnecessary utilization. Read more below
  • We know of no studies that evaluate changes in consumer financial risk with the adoption of health IT. Read more below

In theory, if health IT improves data and decision support for providers, it could reduce financial risk for the consumer by reducing unnecessary utilization.

Theoretically, if health IT facilitates improvements in the efficiency of health care delivery, which in turn result in reduced spending, consumer financial risk will decline through lower premiums, reductions in unnecessary or duplicate utilization, and the availability of less expensive alternatives to current methods of service delivery.

Patients seen by safety net providers may realize less of the potential savings from health IT adoption than other groups (Shields et al., 2007), partly because safety net providers may be slower to adopt health IT. One recent survey of community health centers found that only 13 percent of clinics had adopted an electronic health record that met the minimum federal standards of functionality, compared with 17 percent of physicians overall. In addition, because patients seen at such clinics have fewer cost sharing responsibilities (e.g., they are Medicaid recipients or receive free or discounted care), they may not share financially in efficiency gains achieved by health IT implementation. Small group practices tend to lag behind larger groups or hospitals in health IT adoption (Lee et al., 2005), so patients cared for in these practices also may not see any change in consumer financial risk.

We know of no studies that evaluate changes in consumer financial risk with the adoption of health IT.

No literature has been found to link health IT to consumer financial risk.

References

  • Shields AE, Shin P, Leu MG, Levy DE, Betancourt RM, Hawkins D, Proser M, "Adoption of Health Information Technology in Community Health Centers: Results of a National Survey," Health Affairs, Vol. 26, No. 5, September/October 2007, pp. 1373-1383.
  • Lee J, Cain C, Young S, Chockley N, Burstin H, "The Adoption Gap: Health Information Technology in Small Physician Practices," Health Affairs, Vol. 24, No. 5, September/October 2005, pp. 1364-1366.

Back to top

Waste

Health IT adoption and connectivity could reduce waste, but few studies examine this relationship:

  • In theory, health IT adoption and connectivity have the potential to reduce administrative, operational, and clinical waste. Read more below
  • Waste has not been measured consistently, and we found few empirical studies that examined the role of health IT in reducing waste. Read more below

In theory, health IT adoption and connectivity have the potential to reduce administrative, operational, and clinical waste.

We define three categories of waste in health care: administrative, operational, and clinical (Bentley et al., 2008). Health IT can theoretically facilitate the reduction of waste in each of these dimensions.

Administrative waste: The cost of health care administration has been estimated to be as high as 31 percent of total health care spending (Woolhandler, Campbell, and Himmelstein, 2003). Health IT could provide support for and streamline many of the health care administrative functions. For example, IT may enable faster and more accurate claims and billing processes through online access to patient specific coverage levels and limits, health network use requirements, provider payment rules and allowances, and deductible and copayment or coinsurance levels; as well as through online claims submission and adjudication. Currently, although IT has enabled more electronic billing and insurance related functions, deficiencies in interconnectivity and adoption have limited the full range of potential benefits.

Data from health IT billing, claims, and care related accounts functions may automatically interface with general administrative and business support functions to create integrated accounts receivable, manage cash flow, and support general accounting functions and other administrative and contract management functions. Health IT may include support for developing reports and other data in response to regulatory, certification, or external review organizations, and to public health agencies.

Considering that the administrative overhead of private insurance is much higher than that for public programs, health IT's potential to reduce administrative waste will be larger for privately insured patients (Davis et al., 2007; McKinsey Global Institute, 2007; Catlin et al., 2007). Health IT-enabled chronic disease management can help to avoid unnecessary hospitalizations and reduce adverse drug events, overall reducing clinical waste for patients with chronic disease.

Operational waste includes duplication of laboratory testing and imaging, inefficient patient scheduling, and use of expensive drugs when generic substitutes are available. Connected health IT systems should permit rapid access to previous test results and images, reducing one kind of incentive to repeat tests. Ready access to insurance formularies should reduce the costs of medications, and the ubiquity of the EHR should permit more efficient scheduling of patients and sequencing through physician offices and hospitals. Preventive care reminders that trigger recommended interventions based on the patient's longitudinal record could help maximize visit productivity.

Clinical waste includes avoidable adverse drug events and the provision of low-value, ineffective, or unnecessary care. Reminders and care guidelines could help ensure that recommended care guidelines are followed. Connectivity between care providers can enable a continuity of care not generally available in the current compartmentalized health care system. Home monitoring and better patient connectivity with health IT could help chronically ill patients avoid unnecessary emergency department visits and hospitalization by helping patients keep their illnesses under better control. Computerized physician order entry could help document a patient's current allergies and medications to avoid potential adverse drug events.

To achieve savings from efficiency efforts in any of the areas just noted, some resources must be removed from the health care system. For example, if nursing time is saved by more efficient scheduling, then assigning nurses to other tasks does not lead to cost savings, unless this assignment permits a reduction in staff. If the requirement for an on-site radiologist is reduced by the Picture Archiving and Communication System (PACS), then savings are achieved only when there are fewer on-site radiologists. Similarly, redundant lab tests and images generate revenue for the labs and imaging facilities, and there may be incentives to maintain that revenue stream. To remove the waste and achieve the savings, we need to assume that, eventually, a new equilibrium employing fewer health care resources will be reached.

Waste has not been measured consistently, and we found few empirical studies that examined the role of health IT in reducing waste.

There is little empirical literature that evaluates the relationship between health IT and waste in the health care system. In general, most evidence of efficiency savings from health IT is case study specific, and those cases must be extrapolated across the health care system to estimate the potential savings. Concerns about the extrapolation include the fact that we do not know how consistently health IT will achieve such savings in the various health care systems (Shekelle, Morton, and Keeler, 2006).

Health care is different from other business enterprises in that it is a highly fragmented industry; it is highly personal in the way it interacts with the consumer; the payment process is very different in the incentives it promotes; and almost any change threatens some stakeholder's well-being. These factors make the interaction of health IT and waste difficult to predict.

References

  • Bentley T, Effros R, Palar K, Keeler E, "Waste in the U.S. Health Care System: A Conceptual Framework," Milbank Quarterly, Vol. 86, No. 4, December 2008, pp. 629-659.
  • Catlin A, Cowan C, Heffler S, Washington B, National Health Expenditure Accounts Team, "National Health Spending in 2005: The Slowdown Continues," Health Affairs, Vol. 26, No. 1, January/February 2007, pp. 142-153.
  • Davis K, Schoen C, Guterman S, Shih T, Schoenbaum SC, Weinbaum I, Slowing the Growth of U.S. Health Care Expenditures: What Are the Options? New York: Commonwealth Fund, January 2007.
  • McKinsey Global Institute, "Accounting for the Cost of Health Care in the United States," January 2007. As of November 10, 2008: http://www.mckinsey.com/mgi/rp/healthcare/accounting_cost_healthcare.asp
  • Shekelle PG, Morton SC, Keeler EB, Costs and Benefits of Health Information Technology, Evidence Report/Technology Assessment No. 132 (Prepared by the Southern California Evidence-Based Practice Center), AHRQ Publication No. 06-E006, Rockville, Md.: Agency for Healthcare Research and Quality, April 2006. As of November 10, 2008: http://www.ahrq.gov/downloads/pub/evidence/pdf/hitsyscosts/hitsys.pdf
  • Woolhandler S, Campbell T, Himmelstein D, "Costs of Health Care Administration in the United States and Canada," New England Journal of Medicine, Vol. 349, No. 8, August 21, 2003, pp. 768-775.

Back to top

Reliability

A few small studies show that health IT improves reliability of care, but these results may not generalize to the entire U.S. health care system:

  • In theory, health IT should improve the reliability of health care by reducing medical errors and adverse drug events, increasing the rates of recommended care, and decreasing the duplication of tests. Read more below
  • Several small studies document a decrease in errors and adverse drug events with computerized physician order entry systems and an improvement in quality and consistency with health IT based performance measurements Read more below
  • It is unclear how generalizable these results are to the entire U.S. health care system. Read more below

In theory, health IT should improve the reliability of health care by reducing medical errors and adverse drug events, increasing the rates of recommended care, and decreasing the duplication of tests.

Theoretically, health IT should improve the reliability of health care in several ways. The ready availability of allergy and other drug information in an EHR should reduce the frequency of adverse drug events. Connected health care systems should make this information available when and wherever the patient seeks care. Decision support should permit easy reference to recommended care for particular conditions. Quality measurement through EHR systems could enable more consistent application of recommended care. Health IT that supports monitoring of any preventive measures and allows for easy notification of the patient and physician about needed interventions should improve the reliability of adherence to recommended preventive measures.

One randomized controlled trial in 14 primary care practices found that health IT increased the use of guideline adherent care for hypertension but did not demonstrate improvements over the 18 months of the study in blood pressure control (Hicks et al., 2008). Another small population based study in southwest London found that health IT improved rates of documenting smoking status and providing smoking cessation advice and was associated with a 4 percentage point rate of decline in smoking (Millett et al., 2007).

If health IT enhances reliability, it should theoretically reduce the differences across subgroups. Patients who are more vulnerable to failures in reliability may realize greater health benefits from improved reliability.

Several small studies document a decrease in errors and adverse drug events with computerized physician order entry systems and an improvement in quality and consistency with health IT based performance measurements.

The key study about avoiding errors and adverse drug events with computerized physician order entry (CPOE) systems is by Bates et al., 1998. Health care comparisons between the Veterans Administration and other institutions demonstrate health IT-enabled quality improvement and consistency as a result of performance measurement (Asch et al., 2004). Other studies report several improved protocols resulting from EHR based performance measurement (James, 1989, 1993). Alternatively, some studies report how the use of CPOE can introduce errors if not properly integrated into the medication prescribing and administration process (McDonald et al., 2004). These limited studies generally imply a positive effect on reliability, but they also suggest caution in that the positive effect will depend on careful implementation.

Extrapolation of these results to the broader health care setting indicates that health IT and CPOE result in significant reduction in adverse drug events and related costs. Assuming health IT adoption by 90 percent of hospitals, researchers estimate that CPOE could eliminate 200,000 adverse drug events, two-thirds of them in patients over the age of 65. With a similar level of adoption in the ambulatory setting, they estimate that CPOE could prevent 2 million such events (Hillestad et al., 2005).

It is unclear how generalizable these results are to the entire U.S. health care system.

Because the empirical evidence to date is limited, the key assumption made in estimating the effect of health IT on reliability is that the experience of early adopters can be generalized as the technology is implemented broadly. For example, can the limited experience reported about reductions in adverse events be replicated by other providers in other settings? (The data about health IT and adverse drug events derive primarily from results reported by David Bates and colleagues, 1998, in a single setting.) It is unclear how generalizable these results are to the entire U.S. health care system.

Another major assumption made in estimating the effect of health IT on reliability is that providers adopt effective technologies. Furthermore, it is assumed that EHR systems are connected across providers so that patient allergies and prescriptions can be known wherever a patient seeks care. Similarly, to improve adherence to preventive care and recommended care guidelines, the EHR systems must have widespread adoption and have appropriate functionality.

References

  • Asch, SM, McGlynn EA, Hogan MM, Hayward RA, Shekelle P, Rubenstein L, Keesey J, Adams J, and Kerr EA, "Comparison of Quality Care for Patients in the Veterans Health Administration and Patients in a National Sample," Annals of Internal Medicine, Vol. 141, No. 12, December 21, 2004, pp. 938-945.
  • Bates DW, Leape LL, Cullen DJ, Laird N, Petersen LA, Teich JM, Burdick E, Hickey M, Kleefield S, Shea B, Vliet MV, Seger DL, "Effect of Computerized Order Entry and a Team Intervention on Prevention of Serious Medication Errors," Journal of the American Medical Association, Vol. 280, No. 15, October 21, 1998, pp. 1311-1316.
  • Hicks LS, Sequist TD, Ayanian JZ, Shaykevich S, Fairchild DG, Orav EJ, Bates DW, "Impact of Computerized Decision Support on Blood Pressure Management and Control: A Randomized Controlled Trial," Journal of General Internal Medicine, Vol. 23, No. 4, April 2008, pp. 429-441.
  • Hillestad R, Bigelow J, Bower A, Girosi F, Meili R, Scoville R, Taylor R, "Can Electronic Medical Record Systems Transform Health? Potential Health Benefits, Savings, and Costs," Health Affairs, Vol. 24, No. 5, September/October 2005, pp. 1103-1117.
  • James BC, Quality Management for Health Care Delivery, Chicago, Ill.: AHA, Hospital Research and Educational Trust, 1989.
  • James BC, "Quality Improvement in the Hospital: Managing Clinical Processes," Internist, Vol. 34, No. 3, March 1993, pp. 11-17.
  • McDonald, CJ, Overhage JM, Mamlin BW, Dexter PD, and Tierney WM, "Physicians, Information Technology, and Health Care Systems: A Journey, Not a Destination," Journal of the American Informatics Association [Epub December 7, 2003], Vol. 11, No. 2, March/April 2004, pp. 121-124.
  • Millett C, Gray J, Saxena S, Netuveli G, Majeed A, "Impact of a Pay-for-Performance Incentive on Support for Smoking Cessation and on Smoking Prevalence Among People with Diabetes," Canadian Medical Association Journal, Vol. 176, No. 12, June 5, 2007, pp. 1705-1710.

Back to top

Patient Experience

Health IT adoption has the potential to improve patient experience, but the empirical evidence is limited and shows mixed results:

  • In theory, health IT should enhance patient experience through such mechanisms as improved care coordination, communication, decreased duplication of tests, better scheduling, access to information, and even financial decision aids. Read more below
  • There are limited empirical studies on the effects of health IT on patient experience, and those studies have produced mixed results. Read more below

In theory, health IT should enhance patient experience through such mechanisms as improved care coordination, communication, decreased duplication of tests, better scheduling, access to information, and even financial decision aids.

Theoretically, health IT should enhance the patient health care experience in a number of ways. EHR based care coordination for those patients seeing multiple providers should ease the navigation of the health care system. In conjunction with this coordination, there should be improved communication with patients, reduced duplication of services and tests, a single point entry of patient demographic and vital data, improved scheduling of visits, and more consistent reminders of needed services. Home monitoring, patient decision support, and email contacts with providers should reduce the frequency of office visits and keep patients more involved in their care. Interoperable health care networks should permit continuity of care, and information about medications, allergies, and conditions should be available wherever care is sought or needed. Given the copayments, health savings accounts, tax deductions, and high patient deductibles that make financial issues regarding care choices more complex, personalized financial decision aids should help patients make better choices from both a cost perspective and a health perspective. One concern that patients may have is the extent of control they or their physicians exert over the release of private health information into a digital network. The balance to date has favored privacy (the potential impact of unauthorized release of medical information) over health IT supported information exchange. The issue is ongoing for both Congress and the U.S. Department of Health & Human Services (HHS).

Large-scale effects on patient experience will not be evident until providers adopt EHRs more broadly and until connectivity and interoperability are increased. Some aspects of health IT adoption will require changes to the process of physician-patient interaction that may be to the detriment of patient experience. For example, if the physician must sacrifice face-to-face discussion with the patient to input or retrieve data or otherwise interact with the EHR, the patient experience may be compromised.

There are limited empirical studies on the effects of health IT on patient experience, and those studies have produced mixed results.

An observational study in four primary care offices of the effect of an EHR on the patient-doctor relationship concludes that, with some small but important process changes, the effect may be mixed (Ventres et al., 2006). The study reported that, for those physicians who were good communicators, the EHR could enhance the patient experience, but for those who were poor communicators, the EHR could hinder the communication process. A survey of over 4,000 patients in the Geisinger Health System in Pennsylvania indicated that patients were generally happy with their access to portions of their EHR and that they believed the information to be correct and up to date (Hassol et al., 2004). They preferred email communication with their providers for most information, but face-to-face discussions for some subjects, such as treatment instructions. Generally, the evidence regarding the effect of health IT on the patient experience is limited to just a few studies. We did not find enough studies with sufficient data, power, or comparability to draw any firm conclusions.

To date, there has been little reporting or documentation of care continuity and coordination of interconnected, interoperable health IT from a patient or family perspective.

References

  • Hassol A, Walker JB, Kidder D, Rokita K, Young D, Pierdon S, Diets D, Kuck S, Ortiz E, "Patient Experiences and Attitudes About Access to a Patient Electronic Health Care Record and Linked Web Messaging," Journal of the American Medical Informatics Association, Vol. 11, No. 6, August 6, 2004, pp. 505-513.
  • Ventres W, Kooienga S, Vuckovic N, Marlin R, Nygren P, Stewart V, "Physicians, Patients and the Electronic Health Record: An Ethnographic Analysis," Annals of Family Medicine, Vol. 4, No. 2, March/April 2006, pp. 124-131.

Back to top

Health

While, in theory, health IT adoption should improve health, there is little empirical evidence of this relationship:

  • Theory suggests that health IT has the potential to improve health by enabling reduced medical errors, improved chronic disease management, improved decisionmaking on the part of providers, and a number of other mechanisms. Read more below
  • Individuals with chronic health problems have the greatest potential to benefit from health IT because of their high levels of contact with the health care system. Read more below
  • Because of the relatively low rate of health IT adoption to date, little empirical evidence exists on the health effects of health IT adoption and connectivity. Read more below

Theory suggests that health IT has the potential to improve health by enabling reduced medical errors, improved chronic disease management, improved decisionmaking on the part of providers, and a number of other mechanisms.

Theoretically, health IT can facilitate improvements in care delivery that, in turn, lead to improvements in health. Some examples include reducing adverse drug events (reducing handwriting errors, providing allergy warnings and warnings of drug-drug interactions), improving adherence to preventive care guidelines (population surveillance and reminders), preventing and improving management of chronic diseases (communication, home monitoring, education, and coordination among providers), offering decision support to providers and patients (reducing variance and improving the adherence to evidence based practice), and providing access to a greater array of health resources for patients (education, online support, etc.).

Key assumptions in estimating the health benefits associated with widespread adoption of health IT relate to the likelihood that IT will facilitate process improvements that are known to affect health. The assumptions surround policy implementation and the effectiveness of the adopted technologies.

Individuals with chronic health problems have the greatest potential to benefit from health IT because of their high levels of contact with the health care system.

We would expect that the greatest health effect from IT would accrue to those with chronic conditions, who are particularly vulnerable to failures in the delivery of health care. One randomized controlled trial in 14 primary care practices found that health IT increased the use of guideline adherent care for hypertension but did not demonstrate improvements over the 18 months of the study in blood pressure control (Hicks et al., 2008).

Because of the relatively low rate of health IT adoption to date, little empirical evidence exists on the health effects of health IT adoption and connectivity.

There is little direct evidence in the literature today demonstrating a relationship between IT adoption and health. Estimates of the expected effect on health rely on secondary effects: Health IT facilitates better care delivery, which in turn is expected to improve health. However, evidence from empirical studies of disease management does not provide strong support for this relationship — probably because most disease management studies have not followed patients long enough to observe significant health effects.

A small population based study in southwest London found that health IT improved rates of documenting smoking status and providing smoking cessation advice and was associated with a 4 percentage point rate of decline in smoking (Millett et al., 2007).

EHR systems can tie patient data together with evidence based recommendations for preventive services to identify patients needing particular tests or services. Based on evidence that computerized reminders increase compliance with preventive services (Burack and Gimotty, 1997; Kaplan, 2001), a prior RAND study (Hillestad et al., 2005) estimated the potential benefits of using EHR to improve compliance with the recommendations of the U.S. Preventive Services Task Force for five diseases. These benefits are shown in the table. Assuming 100 percent compliance, we show the results at an upper bound, but the results are significant at even lower levels of compliance.

Summary of Estimated Results for Increasing Five Preventive Services

Program Description Influenza vaccination Pneumococcal vaccination Screening for breast cancer Screening for cervical cancer Screening for colorectal cancer
Population
Target age 65 and older 65 and older Women 40 and older Women 18-64 50 and older
Frequency of treatment 1 per year 1 per lifetime 0.5—1 per year 0.33—1 per year 0.1—0.2 per year
Not currently compliant 13.2 million 17.4 million backlog; 2.1 million new per year 18.9 million 13.0 million 52.0 million
Health benefits
Reduced workdays missed 180,000-325,000 per year 100,000-200,000 per year NA NA NA
Reduced bed days 1.0-1.8 million per year 1.5-3.0 million per year NA NA NA
Deaths avoided 5,200-11,700 per year 15,000-27,000 per year 2,200-6,600 per year 533 per year 17,000-38,000 per year
Life years gained NA NA NA 13,000 per year 138,000 per year

SOURCE: Excerpt from Exhibit 3 in Hillestad et al., 2005.

NOTE: NA means "not applicable."

References

  • Burack RC, Gimotty PA, "Promoting Screening Mammography in Inner-City Settings: The Sustained Effectiveness of Computerized Reminders in a Randomized Controlled Trial," Medical Care, Vol. 35, No. 9, September 1997, pp. 921-931.
  • Hicks LS, Sequist TD, Ayanian JZ, Shaykevich S, Fairchild DG, Orav EJ, Bates DW, "Impact of Computerized Decision Support on Blood Pressure Management and Control: A Randomized Controlled Trial," Journal of General Internal Medicine, Vol. 23, No. 4, April 2008, pp. 429-441.
  • Hillestad R, Bigelow J, Bower A, Girosi F, Meili R, Scoville R, Taylor R, "Can Electronic Medical Record Systems Transform Health? Potential Health Benefits, Savings, and Costs," Health Affairs, Vol. 24, No. 5, September/October 2005, pp. 1103-1117.
  • Kaplan B, "Evaluation Informatics Applications—Clinical Decision Support Systems Literature Review," International Journal of Medical Informatics, Vol. 64, No. 1, November 2001, pp. 15-37.
  • Millett C, Gray J, Saxena S, Netuveli G, Majeed A, "Impact of a Pay-for-Performance Incentive on Support for Smoking Cessation and on Smoking Prevalence Among People with Diabetes," Canadian Medical Association Journal, Vol. 176, No. 12, June 5, 2007, pp. 1705-1710.

Back to top

Coverage

Anecdotal evidence suggests that health IT may enhance enrollment into public health insurance programs, but the overall effect on coverage is uncertain:

  • The relationship between health IT adoption and coverage has not been studied. Read more below
  • Anecdotal evidence from states that use electronic enrollment in public programs suggests that adoption of health IT may increase insurance enrollment among uninsured but eligible individuals. Read more below

The relationship between health IT adoption and coverage has not been studied.

Evidence of improvements in health insurance coverage as a result of expanded adoption and interoperability of health IT is limited. No empirical studies estimate the magnitude of a change in health insurance coverage as a result of the adoption of IT to encourage enrollment. Nor are there any existing studies that estimate the changes in health insurance coverage among subgroups of the population if health insurance plans adopted health IT or improved existing health IT connectivity.

Anecdotal evidence from states that use electronic enrollment in public programs suggests that adoption of health IT may increase insurance enrollment among uninsured but eligible individuals.

Adoption of health IT could, in theory, increase insurance coverage. IT mechanisms that may encourage health insurance enrollment among the uninsured include Web sites that provide information about health coverage options, eligibility requirements, and enrollment or reenrollment procedures. Allowing applicants to submit enrollment or reenrollment applications or e-signatures online can also facilitate health insurance enrollment and retention. Public health insurance programs might also benefit from the use of IT if eligible individuals are automatically enrolled, by identifying eligibility information through data obtained electronically from other public programs or other database sources that would contain the necessary eligibility information.

One study using focus groups of low income parents who used online enrollment in Medicaid and the State Children's Health Insurance Program (SCHIP) found generally positive support for this technology. In particular, parents felt that online applications avoided enrollment barriers, such as having to miss work in order to travel and wait in welfare or other enrollment locations (Perry and Paradise, 2007). Another study, by Morrow and Horner, 2007, documented best practices that are currently being employed by states to improve and expand electronic Medicaid and SCHIP application and enrollment procedures. Although this study provides a number of examples of state level efforts to encourage enrollment, as well as estimates of potential cost savings, it does not provide empirical estimates of the magnitude at which IT expansions will advance Medicaid and/or SCHIP enrollment.

A number of studies demonstrate the insurance enrollment effects of health IT and the connectivity of different public programs. New York State has integrated its Medicaid and Food Stamp programs under one information management system, which allows the state to identify children who receive Food Stamps but are not enrolled in Medicaid. In turn, the state is able to notify parents of uninsured but eligible children that their child will automatically be enrolled in Medicaid unless the parent declines. Of those parents who were notified that their child would be automatically enrolled in Medicaid, only 2 percent opted out of automatic enrollment (Neuberger, 2006). Louisiana has connected its Medicaid program to income databases maintained by the other public social programs operating in the state, which allows the Louisiana Medicaid program access to necessary income information and other key information to determine Medicaid reenrollment eligibility. As a result, 60 percent of Medicaid automatic renewals were successful, without any need to contact families for additional information. Further, according to Dorn and Kenney (2006), the percentage of children whose coverage was terminated for procedural reasons, such as failure to provide requested documentation, dropped from 25 to less than 4.

Sharing of information across public government programs may be hindered by privacy and disclosure laws. The Health Insurance Portability and Accountability Act (HIPAA), for example, contains privacy rules that may inhibit the electronic exchange of eligibility information across public programs and prevent automatic enrollment (Rosenbaum, MacTaggart, and Borzi, 2006).

References

  • Dorn S, Kenney GM, Automatically Enrolling Eligible Children and Families into Medicaid and SCHIP: Opportunities, Obstacles, and Options for Federal Policymakers, New York: Commonwealth Fund, June 2006.
  • Morrow B, Horner D, Harnessing Technology to Improve Medicaid and SCHIP Enrollment and Retention Practices, Washington, D.C.: Children's Partnership and Kaiser Commission on Medicaid and the Uninsured, May 2007.
  • Neuberger Z, Reducing Paperwork and Connecting Low-Income Children with School Meals: Opportunities Under the New Child Nutrition Reauthorization Law, Washington, D.C.: Center for Budget and Policy Priorities (CBPP), as cited in Dorn and Kenney, 2006.
  • Perry M, Paradise J, Enrolling Children in Medicaid and SCHIP: Insights from Focus Groups with Low-Income Parents, Washington, D.C.: Kaiser Commission on Medicaid and the Uninsured, Kaiser Family Foundation, Pub. No. 7640, May 2007.
  • Rosenbaum S, MacTaggart P, Borzi PC, "Medicaid and Health Information: Current and Emerging Legal Issues," Health Care Financing Review, Vol. 28, No. 2, Winter 2006, pp. 21-29.

Back to top

Capacity

In theory, health IT could increase capacity by increasing efficiency, but studies show mixed results:

  • Theory suggests that health IT could increase capacity by enabling providers to see patients more efficiently and reducing the number of face-to-face visits for individuals with chronic disease. Read more below
  • The limited empirical evidence on this subject shows mixed results, with some providers finding that health IT enhanced their ability to see more patients and others not seeing this gain. Read more below
  • The long-term effects of health IT on capacity are unknown. Read more below

Theory suggests that health IT could increase capacity by enabling providers to see patients more efficiently and reducing the number of face-to-face visits for individuals with chronic disease.

Theoretically, several aspects of health IT can contribute to increased capacity of the health care system. Efficiencies in scheduling and reduced duplication of tests should provide additional capacity, and home monitoring and telemedicine should reduce demands on provider time and make health care resources available to populations with access issues. Over the longer term, better disease management with EHR based connectivity and guidance should reduce the frequency of emergency department visits for chronic illness. The potential to improve patient health should reduce the demand for health care resources. Safety improvements possible with health IT, such as reduced adverse drug events, should reduce the need for additional hospitalization or treatment and make those resources available for others. On the contrary, there may be some short term decrease in capacity during the health IT implementation period as productivity temporarily declines.

As with other measures of health IT, the effect depends on how uniformly providers serving different populations adopt EHR systems. If the adoption is lower for safety net providers and those institutions serving disadvantaged patients, then those patients will also have a lower EHR enhanced capacity.

The key assumptions surrounding the effect of health IT on capacity revolve around the effectiveness with which IT is adopted and the likely short term and long term responses to potential capacity increases. As with other categories of performance, capacity will increase if IT is used to redesign work processes. For example, IT could be used to replace face-to-face visits with electronic visits (email, Webcam). The second set of assumptions revolves around how extra capacity is used. We could assume that the increased capacity is maintained and results in increased utilization. We could also assume that the increased capacity is not maintained and leads to providers leaving the market.

The limited empirical evidence on this subject shows mixed results, with some providers finding that health IT enhanced their ability to see more patients and others not seeing this gain.

Generally, the adoption of EHRs is at such a low level and early stage that it is not possible to measure long term capacity increases. One study of the value of EHR for 14 small practices reported that about 8 percent of financial gains from EHR adoption resulted from the ability to see more patients, but such increases were reported in only three of the 14 practices in the study (Miller et al., 2005). The literature reports some case studies on capacity improvement. Kilo (2005) reports, for example, that IT and an innovative medical practice in one system has eliminated many office visits and improved continuity of care for patients with chronic conditions. Further, the practice now has 80 percent of the patient contacts occurring via phone and email, with only 20 percent occurring in visits. In a summary of case study results, the Healthcare Information and Management Systems Society (HIMSS) reported that, in one hospital, use of electronic medical records (EMRs) allowed registered nurses in a charge nurse role to save 2.75 hours in administering medication whereas nurses in direct patient care gained an hour per shift. The gains were redirected into direct patient care (HIMSS, not dated).

Generally, the limited evidence indicated some short term gains in capacity in specific institutions and settings; however, we did not find sufficient evidence to draw more general conclusions from the literature.

The long term effects of health IT on capacity are unknown.

Most quantitative results reported in the literature are based on case studies, and those studies do not inform us about the generalizability of the effect across the health care system or the possible long term adaptation that may take place with respect to capacity.

References

  • Healthcare Information and Management Systems Society (HIMSS), "EHR and the Return on Investment," not dated. As of November 10, 2008: http://www.himss.org/content/files/EHR-ROI.pdf
  • Kilo, Charles M, "Transforming Care: Medical Practice Design and Information Technology," Health Affairs, Vol. 24, No. 5, September/October 2005, pp. 1296-1301.
  • Miller RH, West C, Brown TM, Sim I, Ganchoff C, "The Value of Electronic Health Records in Solo or Small Group Practices," Vol. 24, No. 5, September/October 2005, pp. 1127-1137.

Back to top

Operational Feasibility

Achieving widespread health IT adoption and connectivity will be difficult because of a fragmented health care system and barriers to successful adoption:

  • The slow pace of health IT adoption in U.S. health care systems suggests that significant barriers, including cost, are inhibiting the use of this technology. Read more below
  • Large systems that have successfully implemented health IT (the VA, Kaiser, Geisinger Health System) have invested substantial resources in helping providers make the transition; such help may not be available to providers outside such systems. Read more below
  • Establishing interconnectivity of health IT systems may also be difficult, because providers have few financial incentives to share health information and there are concerns about the privacy and security of that information. Read more below

The slow pace of health IT adoption in U.S. health care systems suggests that significant barriers, including cost, are inhibiting the use of this technology.

Although adoption of health IT is increasing in hospitals and physicians' offices, the slow pace and unevenness of that adoption across the system indicate that the operational feasibility of large-scale adoption is difficult. Adoption involves large financial costs, including the fixed costs of purchasing the systems, maintenance costs, and costs to train users of the systems. In addition, difficulties often arise in choosing between the many IT systems available, adapting the workflow to the new system, and encouraging the organizational changes that will promote acceptance and use of the new system.

The operational feasibility of a regional or nationally connected, interoperable health information system is more questionable. Few regional health information organizations (RHIOs) have been able to develop a sustainable business model, and some prominent early health information exchanges, such as the Santa Barbara Care Data Exchange and the Portland Health Data Exchange, have shut down or stalled out because of a lack of such a model (Miller and Miller, 2007; Korn, 2007). Furthermore, few financial incentives are given for providers to share health information. And strong concerns have been expressed about the privacy and security of health information released into an interoperable network. Moreover, most of the EHR systems are not yet sufficiently interoperable to permit the health information exchanges to be built without significant efforts at creating interfaces between the disparate EHR systems. Efforts to develop the standards for exchanging health information continue.

Literature about the adoption of health IT is plentiful, although most reported rates of adoption are tracked through cross-sectional surveys of hospitals and physicians. A 2008 report by the Congressional Budget Office (CBO, 2008) reviewed the literature on health IT adoption rates, noting that in physicians' offices, the estimated rates of adoption varied: One study found that 24 percent of office based physicians used some sort of EHR, whereas another study found that only 12.4 percent had a comprehensive system. Physicians in solo practice were less likely to use an EHR than those in larger practices (16 percent versus 39 percent). For hospitals, the CBO reported several studies from 2005 that found adoption rates of comprehensive computerized physician order entry (CPOE) systems of between 4 and 5 percent. Finally, the CBO described a study by the American Hospital Association from 2007, which found that 11 percent of nonfederal hospitals had fully implemented EHRs, with larger urban and teaching hospitals more likely to have EHRs than smaller community hospitals (CBO, 2008).

Surveys of health IT adoption shed light on barriers to adoption (Fonkych and Taylor, 2005). In addition to cost, barriers included the difficulty in selecting an EHR appropriate for the setting, the decreased productivity of providers during the adoption period, the willingness of the physicians to both adopt EHRs and change practice patterns accordingly, familiarity with IT, concern about buying a system that is not technically mature, and the limited return on investment that has been demonstrated to date (Poon et al., 2006; Shortliffe, 2005).

Large systems that have successfully implemented health IT (the VA, Kaiser, Geisinger Health System) have invested substantial resources in helping providers make the transition; such help may not be available to providers outside such systems.

Large hospitals and integrated health care delivery systems generally have the resources to initiate and maintain the transition to EHRs. Large provider groups can similarly sustain the effort and costs related to EHR systems.

However, small physician groups and smaller or more disadvantaged hospitals are slower to transition to EHRs, indicating the possible dependency of operational feasibility on access to capital and the potential of return on investment. Smaller physician groups typically face higher costs per provider for implementation of health IT systems than do larger groups. In addition, as the benefits of health IT adoption accrue to patients in the form of improved care and to payers in decreased costs, smaller physician groups may not see a financial return on investment (Poon et al., 2006).

Establishing interconnectivity of health IT systems may also be difficult, because providers have few financial incentives to share health information and there are concerns about the privacy and security of that information.

We do not yet know the types or magnitudes of incentives that would make different kinds of providers in different settings increase their adoption rates of EHRs. That is, we do not know the elasticity of demand of EHRs (or how much demand for them is affected by their price) and the degree of their dependence on the setting or context. Nor do we know how to achieve a sustainable business model for health information exchanges, and there is no national consensus about what privacy and security protections need to be put in place to achieve significant participation of providers and patients in interoperable health information networks.

References

  • Congressional Budget Office (CBO), Evidence on the Costs and Benefits of Health Information Technology, Washington, D.C., Congress of the United States, May 2008, Pub. No. 2976. As of November 10, 2008: http://cbo.gov/ftpdocs/91xx/doc9168/05-20-HealthIT.pdf
  • Fonkych K, Taylor R, The State and Pattern of Health Information Technology Adoption, Santa Monica, Calif.: RAND Corporation, MG-409-HLTH, 2005. As of June 18, 2009: http://www.rand.org/pubs/monographs/MG409/
  • Korn P, "Record Sharing Stalls," Portland Tribune, August 10, 2007. As of November 10, 2008: http://www.portlandtribune.com/news/story.php?story_id=118670243207447600
  • Miller RH, Miller BS, "The Santa Barbara Care Data Exchange: What Happened?" Health Affairs, Web Exclusives [Epub August 1, 2007], Vol. 26, No. 5, September/October 2007, pp. w.568-w.580.
  • Poon EG, Jha AK, Cristino M, Honour M, Fernadopulle R, Middleton B, Newhouse J, Leape L, Bates DW, Blumenthal D, Kaushal R, "Assessing the Level of Health Information Technology Adoption in the United States: A Snapshot," BMC Medical Informatics and Decision Making, Vol. 6, No. 1, 2006. As of November 10, 2008: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1343543
  • Shortliffe EH, "Strategic Action in Health Information Technology: Why the Obvious Has Taken So Long," Health Affairs, Vol. 24, No. 5, September/October 2005, pp. 1222-1233.

Back to top