US: Health Information and Communication Technologies

“As of 2021, nearly 4 in 5 office-based physicians (78%) and nearly all non-federal acute care hospitals (96%) adopted a certified EHR [Electronic Health Record]. This marks substantial 10-year progress since 2011 when 28% of hospitals and 34% of physicians had adopted an EHR.”

Source: Office of the National Coordinator for Health Information Technology. ‘National Trends in Hospital and Physician Adoption of Electronic Health Records,’ Health IT Quick-Stat #61, last accessed January 30, 2025.

“As of 2021, nearly 9 in 10 (88%) of U.S. office-based physicians adopted any electronic health record (EHR)[2], and nearly 4 in 5 (78%) had adopted a certified EHR[4]. This is a consistent trend since 2015. Since 2008, office-based physician adoption of any EHRs has more than doubled, from 42% to 88%. Since ONC and the CDC began tracking it in 2014, the percent of office-based physicians that adopted a certified EHR increased from 74% to 78%.”

Source: Office of the National Coordinator for Health Information Technology. ‘Office-based Physician Electronic Health Record Adoption,’ Health IT Quick-Stat #50, last accessed January 30, 2025.

“Stakeholders and policy makers designing AI guidelines need empirical evidence on how these tools are used and evaluated as the landscape shifts. In this national study of model use and local evaluation in US hospitals in 2023, 65 percent of hospitals reported using models integrated into the EHR. It is concerning that 56 percent of these hospitals did not report evaluating their deployed models for bias and therefore were not systematically protecting patients from possibly biased or unfair AI. Although more hospitals reported locally evaluating for accuracy than for bias, more than one-third did not provide accuracy assurance, either. These results reveal the need for broader awareness of the importance of local evaluation and resources to support evaluation to reduce AI risk.

“Our analysis highlights the urgency of addressing these issues. Even at this early stage of AI adoption, models are being used in ways that present risk to patients. The most commonly reported uses of AI and predictive models (to predict health trajectories or risks for inpatients, identify high-risk outpatients for follow-up care, and facilitate scheduling) are susceptible to bias and substantial differences in validity across health care settings.¹⁰^,²⁵^,²⁶ We also found that local evaluation varied by model application, with hospitals that used models to predict inpatient risk being more likely to report local evaluation. Hospitals that used models for outpatient follow-up and to automate billing procedures were less likely to locally evaluate models, which may reflect a misperception that administrative applications of AI are lower risk than clinical tools. Yet outpatient risk models have been shown to propagate bias in access to care, and patients have expressed concern about the use of models to automate billing.¹⁰^,²⁷ These findings suggest that health care delivery system leaders, model developers, and policy makers need to design guidelines for how models are evaluated and deployed, and to determine what kinds of oversight, best practices, and training are necessary for AI that is not directly guiding diagnosis or treatment.²⁸“

Source: Paige Nong, Julia Adler-Milstein, Nate C. Apathy, A. Jay Holmgren, and Jordan Everson. Current Use And Evaluation Of Artificial Intelligence And Predictive Models In US Hospitals. Health Affairs 2025 44:1, 90-98

“Between 2016 and 2021, ransomware attacks affected 3.7% of short-term acute care hospitals in the United States and 23.5% of hospital markets. Attacked hospitals had larger operating revenue and were more likely to provide trauma, emergency, and obstetric services in urban areas. The finding that nearly one in four hospital markets contains a ransomware-attacked hospital is particularly concerning, given emerging evidence that ransomware attacks generate spillover effects for nearby hospitals that may absorb displaced patient volume.³“

Source: Claire C McGlave, Sayeh S Nikpay, Carrie Henning-Smith, Katie Rydberg, Hannah T Neprash, Characteristics of short-term acute care hospitals that experienced a ransomware attack from 2016 to 2021, Health Affairs Scholar, Volume 1, Issue 3, September 2023, qxad037, doi.org/10.1093/haschl/qxad037

“Telehealthcare use has increased, especially for behavioral health specialties (such as psychology and psychiatry) and primary care. Telehealthcare accounted for about 1% of behavioral health specialty visits and non-behavioral health primary care visits in January 2020. But rates increased during the COVID-19 pandemic such that more than 55% of behavioral health specialty visits and approximately 23% of non-behavioral health primary care visits were telehealthcare visits in April 2020.

“Although telehealthcare use waned as COVID-19 receded, it still accounted for approximately 35% of behavioral health specialty visits and 5% of non-behavioral health primary care visits in December 2021.

“Overall, 8.8% of all physician’s office visits in 2021 were telehealthcare visits. Use varied by medical practice type and by population characteristics. Rates were higher for practices that provided services that do not require in-person contact, such as ordering and interpreting diagnostic studies, prescribing medications, and counseling patients. Rarely, practices that require physical contact to examine or treat people, such as ophthalmology, provided telehealthcare services.

“In 2021, telehealthcare use rates were lowest for people age 65 years and over (3.1% vs. 15.8% for people ages 18-44 years) and people living in noncore areas and small metropolitan areas (3.7% and 4.8%, respectively, vs. 11.7% in large central metropolitan areas).

“In 2021, more practices provided audio-only services (67.0%) than video-plus-audio services (56.4%).

“In 2021, older patients and low-income people were less likely to have audio-plus-video visits: 42.2% of people age 65 and over vs. 71.7% of adults ages 18-44 years; and 50.1% of people with family incomes less than 100% of the poverty guideline vs. 71.5% of people with family incomes 400% or more of the poverty guideline.

“Telehealthcare poses identifiable barriers to healthcare delivery. In 2021, of practices that use telehealthcare technology, 70.4% of physicians reported patients’ difficulty using technology as a factor affecting use of telehealthcare and 64.3% reported limitations in patients’ access to technology as a factor.

“While healthcare providers perceived telehealthcare as a useful resource, they indicated the need for telehealthcare quality improvement. In 2021, 62.0% of practices that used telehealthcare technology were very or somewhat satisfied with using telehealthcare technology for patient visits and 70.8% planned to continue using telehealthcare visits when appropriate after the COVID-19 pandemic ended. However, only 31.0% of practices that used telehealthcare technology described being able to provide similar quality of care during telehealthcare visits as during in-person visits.”

Source: 2023 National Healthcare Quality and Disparities Report. Rockville, MD: Agency for Healthcare Research and Quality; December 2023. AHRQ Pub. No. 23(24)-0091-EF.

“Of practices that use telehealthcare technology:
“■ 67.0% of practices reported using telephone audio for telehealthcare visits,
“■ 56.4% reported using video software with audio,
“■ 44.7% reported using telehealthcare platforms that were not integrated with electronic
health records (EHRs), and
“■ 28.6% reported using telehealthcare platforms that were integrated with EHRs (Figure 6).”

Source: 2023 National Healthcare Quality and Disparities Report. Rockville, MD: Agency for Healthcare Research and Quality; December 2023. AHRQ Pub. No. 23(24)-0091-EF.

“More medical specialists (27.4%) used telemedicine for 50% of their patient visits or more than primary care physicians and surgical specialists.

“Most primary care physicians (76.7%) and medical specialists (73.1%) were able to provide a similar quality of care during telemedicine visits as in-person visits “to some extent or a great extent” compared with about one-half of surgical specialists (50.6%).

“Primary care and medical specialists were more likely to be satisfied with telemedicine technology compared with surgical specialists.

“Surgical specialists were most likely to indicate that telemedicine technology was not appropriate for their specialty or patients (49.7%), followed by medical specialists (26.7%) and primary care physicians (15.5%).”

Source: Myrick KL, Mahar M, DeFrances CJ. Telemedicine use among physicians by physician specialty: United States, 2021. NCHS Data Brief, no 493. Hyattsville, MD: National Center for Health Statistics. 2024. DOI: dx.doi.org/10.15620/cdc:141934

“CONCERNS AND BARRIERS TO ADOPTING EHR [Electronic Health Records]

“1. Shift focus away from patients. According to a Delphi study, physicians expressed concerns that EHR would reduce the number of patients they were able to see in a day. In addition, they were worried that EHR would shift their focus away from patients to screens and may cause them to miss important medical information [20]. In another study, researchers supported this claim and found that physicians’ screen gaze and keyboarding time were negatively correlated with patient-centered communications during patients’ visits. Specifically, screen gaze disrupted physicians’ communication with patients by inhibiting eye contact, psychosocial inquiry, and emotional responsiveness between providers and patients [21]. Thus, more education is needed for physicians to understand how to optimize the use of EHR and achieve optimal interpersonal exchange with patients.

“2. Distinctive interoperability of EHR and state laws and practices. Interoperability refers to the ability to exchange information among different information systems. The diversity of EHR vendors in the market makes exchanging information challenging. Moreover, different states have specific privacy laws concerning private data and various practices regarding health information technology, making interoperability across states challenging [13].

“3. High cost of installing and maintaining EHR. EHR is expensive as the cost associated with adopting EHR is not a one-time fee. Before adopting an EHR system, organizations must identify and assign appropriate administrative and medical personnel to work on implementation with a tight network with EHR vendors. The other costs include hardware (computers, laptops, all-in-one computers, tablets, mouses, keyboards, monitors, and supporting tables), EHR software which could vary a lot based on the type of EHR, IT support, training in how to use EHR, loss of productivity during installation, and ongoing network fees and maintenance [22, 23]. The associated expenses could be challenging, especially for small healthcare agencies in remote areas with limited resources.

“4. Human resistance. According to a survey, providers may not be willing to take the time to familiarize themselves with the available systems, select the suitable EHR, implement it, or get trained to use it. In addition, providers may not possess adequate computer skills to exert all the functions of EHR and deem EHR too complicated to use [24]. Last but not least, providers have concerns that EHR generates new issues on patients’ privacy and confidentiality. All these reasons hamper their desire to use EHR [25].

“5. Design issues of EHR. Poor design and use of EHR can lead to errors, adverse events, and even mortality [26]. This is further aggravated by the increasing functional complexities of EHR to meet clinicians’ demands and the pressure to adopt them within a specific time frame [3, 27]. In one study, researchers identified many EHR-related errors resulting from poorly designed EHR that could threaten patients’ safety, compromise the quality of care, and lead to abuse and fraud of health information [3]. These errors include poor system usabilities, such as confusing interfaces and workflow incompatibility, over-alerts that lead to clinicians’ fatigue and ignorance, inappropriate copy and paste of information, and mistakes in documentation capture. With EHR, there are also greater risks of breaking down information integrity with hackers, malicious insiders, and passcode leakage. In one study, researchers noticed an unexpected increase in death after implementing a computerized physician order entry system in a children’s hospital [28]. These challenges require efforts from both EHR developers and providers to overcome by designing and implementing solid usability standards to optimize system safety to achieve information integrity [27].”

Source: Ge, S., Song, Y., Hu, J., Tang, X., Li, J. and Dune, L. (2022), The development and impact of adopting electronic health records in the United States: A brief overview and implications for nursing education. Health Care Sci, 1: 186-192. doi.org/10.1002/hcs2.21

“This study used 100 percent response survey data from a national cross-sectional census of family medicine physicians in 2018 as a lens with which to examine a variety of physician practice types and their adoption of PCMH [Patient-Centered Medical Home Model] and health IT meaningful-use elements across the US. Overall, 42.1 percent of practices were PCMH certified, and adoption of the majority of individual PCMH and health IT meaningful-use elements was high (more than 70 percent).

“With regard to PCMH certification and adoption of PCMH elements, we found wide variation by practice ownership type. Compared with independent practices, federally qualified health centers stood out as leaders in terms of both PCMH certification and the adoption of specific PCMH elements. This is most likely a reflection of HRSA’s federally qualified health center policies, including the provision of technical assistance, a related national cooperative agreement with NCQA to facilitate certification, provision of funds to pay for certification, and financial incentives that included increased payments for quality for federally qualified health centers that were NCQA certified. Federal military practices demonstrated a high likelihood of adoption of PCMH elements, although the odds of these practices being PCMH certified were not as high as for federally qualified health centers. This may be explained by VA and DOD policies to promote adoption of PCMH elements throughout their health systems while placing relatively little emphasis on PCMH certification compared with federally qualified health centers.¹³ Academic health centers, where most training of the future physician workforce takes place, had higher odds of achieving PCMH certification compared with independent practices but were less likely than federally qualified health centers, federal military practices, and practices owned by managed care/health maintenance organizations and hospitals/health systems to have adopted PCMH elements.

“Independently owned practices and rural health clinics, constituting 34.8 percent of physician practices, demonstrated much lower odds of PCMH certification and much lower likelihood of adoption of PCMH elements. Rural health clinics are an interesting comparison to federally qualified health centers because they received enhanced Medicare reimbursement but not the same technical support and PCMH incentives as federally qualified health centers, potentially explaining the difference in our findings.¹⁴ Participation in the Transforming Clinical Practice Initiative, which provided technical assistance but no financial incentives, was not associated with PCMH certification, increased adoption of PCMH elements, or increased health IT meaningful-use adoption. Although PCMH certification was not an objective of the Transforming Clinical Practice Initiative, it was a focus for some of the practice transformation networks that the initiative supported, and many of the improvements that the initiative sought to achieve aligned with PCMH elements.²^,¹⁵“

Source: Diane R. Rittenhouse, James A. Wiley, Lars E. Peterson, Lawrence P. Casalino, and Robert L. Phillips Jr. Meaningful Use And Medical Home Functionality In Primary Care Practice. Health Affairs 2020 39:11, 1977-1983.

“One area in which the US has been very successful is in the adoption of EHRs: 95.5 percent of family physicians’ practices have adopted some version of an EHR. This reflects the success of federal policies such as the HITECH Act, which provided substantial financial incentives, and the subsequent funding of the regional extension centers for health IT meaningful use, which provided substantial regional technical assistance. The EHRs themselves vary in capabilities, however. We found that although 84.1 percent of physicians reported the ability to exchange secure messages with patients, only 31.2 percent reported having the capability for patients to add patient-generated health data (for example, blood pressure or blood glucose levels checked at home) through a portal into their EHR systems. The variation in EHR capabilities may be attributed to policies surrounding vendor certification rather than to physicians in practice.

“Our findings confirmed the well-established relationship between larger practice size and adoption of PCMH and health IT meaningful-use elements.⁸^,¹⁶ This is important because 47.4 percent of family medicine physicians reported practicing in small practices with one to five providers. Various demonstration programs have sought to support these small practices with transformation efforts, but most have not combined facilitation with strong financial support or incentives similar to HITECH or HRSA.¹^,¹⁷ Our findings suggest that this combination may be important. As policies are crafted to address PCMH and health IT adoption in independent practices, many lessons from these prior, more successful, efforts could be applied. The Medicare Access and CHIP Reauthorization Act (MACRA) of 2105, which went into effect in 2019, allocates $100 million to support organizations (for example, quality improvement organizations, regional extension centers) that provide technical assistance to small practices participating in new systems of payment based in part on meaningful use of health IT and on clinical practice improvements. The Affordable Care Act authorized, but did not fund, the Primary Care Extension Program, which could provide technical assistance for PCMH practice transformation.¹⁸ Our findings suggest that practices need both intensive tailored technical assistance and substantial financial incentives to successfully transform.”

“The adoption of these systems was slow in the United States until the 2009 enactment of the Health Information Technology for Economic and Clinical Health (HITECH) Act, which provided over $20 billion in grants and financial incentives to promote the adoption of EHRs [Electronic Health Records] among healthcare providers (more on HITECH below) (DesRoches et al., 2010). Prior to the Act, in 2007, only 35% of office-based physicians used at least one EHR component in their office, only 12% of physician EHRs met the criteria for having a basic system, and only 4% met the criteria for having fully functional systems (Hing & Hsiao, 2010). By 2015, a few years after the Act, 54% of office-based physicians had adopted a basic system that included information on patient history/demographics, lists of patients’ problems, medications and allergies, physician clinical notes, computerized orders for prescriptions, and the ability to view laboratory and imaging results electronically) (NEHRS, 2015).

“Likewise, in 2008 only 7.6–0.9% of hospitals had a basic EHR system in at least one clinical unit (Jha et al., 2009). In 2014, 41.4% of hospitals had basic systems while 39.1% had comprehensive systems (Adler-Milstein et al., 2017). Since 2014, small and rural hospitals have increased their adoption of basic EHR systems by 14% and critical access hospitals have increased their adoption by 18% (Henry et al., 2016). Basic EHR adoption in psychiatric hospitals doubled between 2008 and 2015, and increased five-fold for children’s hospitals (Henry et al., 2016).

“Other types of healthcare provider are also adopting EHRs. Information regarding the adoption of EHRs in nursing homes is limited but a 2012 survey reported that 18% had a fully implemented and operational EHR, while 30% had a system that was partially implemented and operational (Abramson et al., 2014). Only 11.4% of nursing homes reported no implementation plans. In 2007 about 41% of home health and hospice organizations had EHRs, and an additional 15% planned to have EHRs within the next year (Bercovitz, Sengupta & Jamison, 2010). Of the home health and hospice organizations with EHRs, 98% used components for recording patient demographics, 83% used clinical notes, and over half used clinical decision-support systems or computerized physician-order entry. EHRs are used extensively in freestanding dialysis facilities, particularly in large for-profit dialysis chains. All the five largest dialysis chains use EHRs (Kochevar et al., 2011). Even many small dialysis facilities use EHRs (around 61% in 2010).”

Source: Rice T, Rosenau P, Unruh LY, Barnes AJ, van Ginneken E. United States of America: Health system review. Health Systems in Transition, 2020; 22(4): pp. i–441.

“A few large healthcare systems have achieved EHRs that are interoperable between providers within the same healthcare system. The VHA – the largest integrated healthcare system in the United States – is an example. The VHA developed an HIT system called the Veterans Health Information Systems and Technology Architecture (VistA) that was capable of interconnectivity between all providers within the VHA system (Byrne et al., 2010). The VHA achieved close to 100% adoption of several VistA components, including inpatient and outpatient EHRs, bar code medication administration and computerized physician order entry (CPOE). In 2013 VistA incorporated a personal health record component (see next subsection) so that patients have access to their records (VA, 2018). In 2018 the VHA began the process of transitioning from VistA to a commercial system, Cerner, also used by the Department of Defense (DOD), so that records could be interoperable between DOD and VA patients (VA, 2018). The new system began operating in October 2018.

“Another example is Kaiser Permanente, the largest private non-profit integrated healthcare system in the United States (Chen et al., 2009). Kaiser Permanente provides group health insurance, outpatient care such as primary and specialty care, testing, imaging and pharmaceuticals, and inpatient hospital care. In 2004 Kaiser began implementing a system-wide EHR, HealthConnect, rolling it out to 431 medical offices and 36 hospitals by 2010 (Wheatley, 2013). The EHR provides clinical documentation and decision support across care settings, and real-time connectivity to testing, imaging, pharmacy and other ancillary systems (Wheatley, 2013).”

Source: Rice T, Rosenau P, Unruh LY, Barnes AJ, van Ginneken E. United States of America: Health system review. Health Systems in Transition, 2020; 22(4): pp. i–441.

“In the world’s most advanced health care markets, calls are growing to move away from fee-for-service care and toward value-based care. Such a transition includes a number of structural changes involving new payment models such as increased use of bundles, thoughtfully collecting, analyzing, and sharing patient-reported outcome measures (PROMs), and re-organizing health care delivery infrastructure into integrated practice units.

“Although payment models in the US have evolved since HITECH (for example, as seen in both private and public initiatives to encourage the use of bundles), EHRs are typically linked to revenue cycle management and traditional, fee-for-service billing. Consequently, some technology remains at odds with—or at least partially misaligned with—target payment models. Ideally, databases designed for the delivery of value-based care would go beyond “standard” medical data to include data on social determinants of health and other factors.

“The potential mismatches between the design of digital tools and the goals of the health care system are worth keeping in mind: Software systems designed around fee-for-service health care delivery will perpetuate existing waste and shortcomings, while design that builds in opportunities for broader data collection, user-friendly personal health records, and the evidence-based deployment of personalized digital tools will support the transition to value-based care. In this respect, both the US and Germany have a long way to go. Germany, in particular, has a great opportunity to thoughtfully roll out such tools over the years ahead.

“Furthermore, to take full advantage of digitized health care delivery data, systems must develop algorithms based on large and diverse population data to ensure that risk adjustment for individuals can be done on the basis of representative data from an appropriately comparable group. Algorithms need access to unprecedented amounts of anonymized data, which in turn need to be “cleaned”—not only for errors and incompleteness, but also for inherent biases.”

Source: “On The Brink Of A Digital Health Care Transformation: What Germany Can Learn From The United States, “ Health Affairs Blog, October 20, 2021.
DOI: 10.1377/hblog20211018.865750

“Forty-one FQHCs [Federally Qualified Health Centers] with 534 physical locations provided data. The FQHCs participating in the CHCF [California Health Care Foundation] initiative served 1.7 million patients in 2019 and were similar to FQHCs in California that were not included in the sample; however, the smallest FQHCs (serving ≤9999 patients in 2019) were underrepresented (Table).

“During the prepandemic period, there was a mean of 231.7 primary care visits per 1000 patients per month compared with 228.6 visits per 1000 patients per month during the COVID-19 pandemic period. Adjusted models showed a 6.5% decrease (95% CI, −104% to −2.3%; P = .03) in total visit volume for primary care, with the decline concentrated in March and April 2020. There was no significant change in total behavioral health visits (Figure).

“Prior to the COVID-19 pandemic, there was minimal telehealth use. During March 2020, FQHCs rapidly substituted in-person visits with telephone and video visits. During the pandemic period, there were 109.9 in-person, 111.0 telephone, and 7.8 video visits per 1000 patients per month for primary care with 48.1% occurring in person, 48.5% via telephone, and 3.4% via video. For behavioral health, there were 6.6 in-person, 18.2 telephone, and 4.0 video visits per 1000 patients per month with 22.8% occurring in-person, 63.3% via telephone, and 13.9% via video. Telephone visits peaked in April 2020, comprising 65.4% of primary care visits and 71.6% of behavioral health visits.”

Source: Uscher-Pines L, Sousa J, Jones M, et al. Telehealth Use Among Safety-Net Organizations in California During the COVID-19 Pandemic. JAMA. Published online February 02, 2021. doi:10.1001/jama.2021.0282.

US: Health Information and Communication Technologies - National Policies - World Health Systems Facts

World Health Systems Facts is a project of the Real Reporting Foundation. We provide reliable statistics and other data from authoritative sources regarding health systems and policies in the US and sixteen other nations.

Page last updated February 25, 2025 by Doug McVay, Editor.