Abstract
Introduction
Prior research has shown that provider positive attitudes about EHRs are associated with their successful adoption. There is no evidence on whether comfort with technology and more positive attitudes about EHRs affect use of EHR functions once they are adopted.
Methods
We used data from a survey of providers in the Primary Care Information Project, a bureau of the New York City Department of Health and Mental Hygiene and measures of use from their EHRs. The main predictor variables were scores on three indices: comfort with computers, positive attitudes about EHRs, and negative attitudes about EHRs. The main outcome measures were four measures of use of EHR functions. We used linear regression models to test the association between the three indices and measures of EHR use.
Results
The mean comfort with computers score was 2.37 (SD 0.53) on a scale of 1 to 3 with 3 being the most comfortable. The mean positive attitude score was 2.74 (SD 0.40) on a scale of 1 to 3 with 3 being more positive. The mean negative attitude score was 1.81 (SD 0.54) on a scale of 1 to 3 with 3 being more negative. Within the first twelve months of having the EHR, 59.5% of visits had allergy information entered into a structured field, 64.8% had medications reviewed, and 74.3% had blood pressured entered. Among visits with a prescription generated, 24.5% had prescriptions electronically prescribed. In multivariate regression analysis, we found no significant correlations between comfort with computers, positive attitudes about EHRs, or negative attitudes about EHRs and any of the measures of use.
Discussion
Comfort with computers and attitudes about EHRs did not predict future use of the EHR functions. Our findings suggest that meaningful use of the EHR may not be affected by providers’ prior attitudes about EHRs.
Introduction
Studies of the impact of electronic health records (EHRs) to improve quality of care have shown mixed results.1–7 One possible explanation for these mixed results is that clinicians use EHRs more as electronic document writers and not as tools to better manage patients and to improve efficiency.8,9 In order to improve meaningful use of EHRs, the Centers for Medicare and Medicaid Services launched the Electronic Health Record Incentive Program which paid out more than $5.7 billion to providers in the first year of the program.10,11
Prior research has shown that positive attitudes about EHRs are associated with successful implementation.12–15 However, to our knowledge, there is no evidence on whether comfort with technology and more positive attitudes about EHRs prior to implementation affect use of EHRs once they are implemented. We hypothesize that providers who are comfortable using computers and who feel optimistic about their potential effects on patient care might use more features of the EHR.
In this study, we used data from a survey of providers who enrolled in the Primary Care Information Project (PCIP). PCIP is a bureau of the New York City Department of Health and Mental Hygiene (NYC DOHMH) that subsidized EHRs for 3,200 providers (most of whom were small practice providers) serving underserved areas of New York City. PCIP, as a nationally recognized regional extension center, currently provides technical assistance to providers to help them achieve meaningful use.16
We sought to address two research questions: 1) what were provider levels of comfort with computers and attitudes about EHRs prior to implementation of an EHR and 2) did provider reports of comfort with computers and attitudes about EHRs prior to implementation predict future use of EHR functions?
Methods
Data Sources and Sample
Primary data for the study came from a pre-implementation survey administered prior to going “live” on the EHR. The survey was developed by PCIP staff, and the goal of the survey was to measure providers’ comfort with computer tasks (e.g., typing, printing) and expectations about EHRs (e.g., the EHR will improve medication safety, the EHR will disrupt workflow). The survey also solicited demographic data (e.g., how long the provider had been in practice, provider gender), their comfort level with computers, and their attitudes about EHRs. We obtained additional provider characteristics (provider work load, type of provider, provider specialty) for both survey responders and non-responders from SalesForce©, a customer relations management software used for tracking administrative data about participating practices.
The survey was sent to all providers who enrolled with PCIP. Providers were mailed an advance letter describing the survey after they enrolled with PCIP but before they implemented the EHR. Providers with email addresses were sent a web-based survey via SurveyMonkey©. Providers without an email address were mailed a paper survey. If there was no response after two weeks, providers were sent another email or paper survey. If there was no response after four weeks, PCIP staff called providers.
For this analysis, we included only data from small practices (ten or fewer providers). We excluded providers who eventually did not implement the EHR (n=54), were sent a survey after their EHR had been implemented (n=18), were a temporary employee of the practice or resident physician (n=5), were on leave at the time of the survey (n=2), or whose address was incorrect, (n=3). This resulted in an invited sample of 654 providers. Among these 654 providers, 433 (66.2%) received the survey by email and 221 (33.8%) received it by mail. Among the 433 providers who received the survey by email, 227 (52.4%) were sent another email survey after two weeks and 91 (21.0%) were called after four weeks. Among the 221 providers who received the survey by mail, 176 (79.6%) were sent another mail survey after two weeks and 57 (25.7%) were called after four weeks.
Data on measures of use were transmitted directly from the EHRs to PCIP on a monthly basis. An office visit was defined as an encounter in which the provider recorded that the patient both checked in and checked out.
The study was approved by the Institutional Review Boards of Weill Cornell Medical College and the New York City Department of Health and Mental Hygiene.
Variables
The main predictor variables were scores on three indices: comfort with computers, positive attitudes about EHRs, and negative attitudes about EHRs. We chose these three indices because they have face value as indicators of provider attitudes and because there was high internal consistency within each index but low correlation between the indices (correlation coefficients ranged from −.0.02 to 0.28).
The comfort with computers index consisted of five questions assessing providers’ comfort completing the following tasks: email, printing, restarting a computer, typing, and searching on the internet. Each question was recoded to a three point scale: uncomfortable, comfortable, and very comfortable. We computed a mean comfort score based on the answers to these five questions. The index had high internal consistency across the five questions (Cronbach’s alpha = 0.89).
The positive attitudes about EHRs index consisted of responses to the following five statements: 1) an EHR will improve my access to patient information when I need it, 2) an EHR will improve my ability to make decisions about patient care, 3) an EHR will improve my ability to provide preventative care, 4) an EHR will reduce medication errors and adverse drug events, and 5) I think the benefits of adopting an EHR will outweigh the challenges I have to overcome. Each question was recoded to a three point scale: disagree (“completely disagree” or “generally disagree”), unsure (“don’t know”), and agree (“completely agree” or “generally agree”). For this index, we again calculated a mean score based on the answers to these five questions. The scale had high internal consistency (Cronbach’s alpha = 0.72).
The negative attitudes about EHRs scale consisted of responses to the following seven statements: 1) using an EHR will decrease the amount of time I can spend talking with patients, 2) using an EHR will cause disruptions to my workflow, 3) using an EHR will cause a patient visit to last longer, 4) the use of the computer in the exam room will interfere with the patient visit, 5) an EHR will generate too many alerts and reminders during the patient visit, 6) using an EHR will limit my discretion as a primary care provider, and 7) using an EHR will make it more difficult to protect patient privacy. Each question was coded using the same three point scale as the positive attitude score, but for this scale a high score equated to strong negative attitudes about the EHR. The scale had high internal consistency (Cronbach’s alpha = 0.74).
We standardized the scores for each question in each index by taking the z-score (mean of question score was subtracted from individual question score and divided by the standard deviation of the question score). As a result, scores for each item have a mean of 0 and a standard deviation of 1.
The main outcome measures were four measures of EHR use: 1) the percentage of visits with a documented blood pressure, 2) the percentage of visits where medications were reviewed, 3) the percentage of visits with allergy information entered into a structured field, and 4) the percentage of visits with a prescription generated and the prescription was electronically prescribed. We chose the four measures of use because they were closely aligned with the Stage 1 meaningful use measures and because they were the most reliable measures available from the electronic health record.17 The use data is at the encounter-level and each encounter was credited to a provider even if staff performed a function. For the first three measures, the denominator was all visits; for the fourth measure (electronic prescription), the denominator was visits in which a prescription was generated. We calculated the outcome measures for the 12 month time period after EHR implementation. We had EHR use data for 302 of 328 respondents (92.1%). Data on EHR use was missing for a limited number of practices due to problems with transmissions.
Analysis
We used the Pearson Chi-square test to compare characteristics of responders and non-responders.
We performed an item-level analysis to evaluate the relationship between provider characteristics and comfort with computers/attitudes about EHRs. By item-level, we mean that we analyzed each individual question that each provider answered. For each index (comfort with computers index, positive attitudes index, negative attitudes index), we estimated a linear regression model in which the z-score for each item in that index was the dependent variable and provider characteristics were the independent variables. To account for variation in item response across providers, the models controlled for the specific items reported by each provider. For each index, we tested whether item z-scores varied significantly across each provider characteristic.
We then estimated linear regression models to test the association between item-level comfort with computers and attitudes about EHRs and the four physician-level measures of EHR use listed above. Separate models were estimated for each measure of EHR use. All of the items from all the indices (comfort with computers index, positive attitudes index, negative attitudes index) were entered simultaneously into the model to control for the providers’ responses to items in the other indices. We generated estimates of the relationship between items in each domain and the measures of use. As before, the models controlled for the specific items reported by each provider. These models also controlled for provider characteristics (workload, provider type, specialty, gender, and years in practice) and the quarter-year in which the practice started using the EHR.
Standard errors for all item-level models were clustered at the provider-level. The data were analyzed using Stata 12.0 (Stata Corp, College Station, TX).
Results
Of the 654 eligible providers, 328 responded (response rate 50.2%). Among providers who received an email survey, 54.0% responded; among providers who received a paper survey 42.5% responded. Even if a provider was initially sent an email survey, they were free to respond to either form of the survey. Responders were more likely to work more than 20 hours per week at the practice (85.4% of responders vs. 70.6% of non-responders, p=0.001), be in family practice (15.5% vs. 10.7%, p=0.02) and pediatrics (28.4% vs. 14.5%) than obstetrics/gynecology (3.7% vs. 5.2%) or other specialties (8.8% vs. 16.0%, p=0.02), practice in an office with only one location (85.1% vs. 77.6%, p=0.01), and be in a practice with fewer providers (mean 2.3 vs 3.3, p<0.001; Table 1).
Table 1.
Characteristics of responders and non-responders
| Responders, no. (%)(n=328) | Non-responders, no. (%)(n=326) | p-value | |
|---|---|---|---|
| Provider work load (hrs/wk) | <0.001 | ||
| <20 | 48 (14.6) | 96 (29.5) | |
| ≥ 20 | 280 (85.4) | 230 (70.6) | |
| Type of provider | 0.48 | ||
| MD or DO | 284 (86.6) | 275 (84.4) | |
| Non-MD (PA, NP) | 44 (13.4) | 51 (15.6) | |
| Provider Specialty | 0.02 | ||
| Internal Medicine | 143 (43.6) | 142 (43.6) | |
| Family Practice | 51 (15.5) | 35 (10.7) | |
| Pediatrics | 93 (28.4) | 80 (24.5) | |
| Obstetrics and Gynecology | 12 (3.7) | 17 (5.2) | |
| Other | 29 (8.8) | 52 (16.0) | |
| Type of Survey | |||
| Paper | 94 (28.7) | 127 (39.0) | 0.005 |
| Web | 234 (71.3) | 199 (61.0) | |
| Practice Size, mean | 2.3 | 3.3 | <0.001 |
| Years in Practice | - | ||
| 0–5 | 60 (19.9) | - | |
| 6–10 | 102 (33.9) | - | |
| 11–20 | 69 (22.9) | - | |
| >20 | 70 (23.3) | - | |
| Gender | - | ||
| Female | 123 (39.8) | - | |
| Male | 186 (60.2) | - | |
| Practice Size | |||
| 1 to 2 Providers | |||
| 3 or more Providers | |||
| Comfort with computers , mean score (SD) | 2.37 (0.53) | - | - |
| Positive attitudes about EHRs, mean score (SD) | 2.74 (0.40) | - | - |
| Negative attitudes about EHRs, mean score (SD) | 1.81 (0.54) | - | - |
The majority of providers were comfortable or very comfortable with email (90.5%), printing (89.9%), and searching the internet (85.1%, Table 2). A smaller percentage of providers were comfortable or very comfortable restarting a computer (76.0%) and typing (83.6%). The mean comfort with computers score was 2.37 (SD 0.53) on a scale of 1 to 3 with 3 being the most comfortable.
Table 2.
Providers’ comfort with computers and attitudes about EHRs
| Comfort with Computers (n=328) | ||||
|---|---|---|---|---|
| Uncomfortable, % | Comfortable, % | Very Comfortable, % | Missing, % | |
| Comfort with email | 3.7 | 41.1 | 49.4 | 5.5 |
| Comfort with printing | 3.4 | 43.3 | 46.6 | 6.7 |
| Comfort restarting a computer | 18.9 | 40.9 | 35.1 | 5.2 |
| Comfort typing | 11.3 | 48.2 | 35.4 | 5.2 |
| Comfort searching on the internet | 4.0 | 40.9 | 44.2 | 11.0 |
| Positive attitudes about EHRs (n=328) | ||||
|---|---|---|---|---|
| Disagree, % | Unsure, % | Agree, % | Missing, % | |
| The EHR will improve access to patient info | 3.7 | 6.1 | 86.9 | 3.4 |
| The EHR will improve ability to make decisions | 18.3 | 9.8 | 68.0 | 4.0 |
| The EHR will help me provider better preventative care | 6.7 | 8.5 | 80.8 | 4.0 |
| The EHR will lead to fewer Rx errors | 5.2 | 5.8 | 86.0 | 3.1 |
| The benefits of the EHR outweigh the challenges | 6.1 | 11.3 | 79.0 | 3.7 |
| Negative attitudes about EHRs (n=328) | ||||
|---|---|---|---|---|
| Disagree, % | Unsure, % | Agree, % | Missing, % | |
| I will have less time with patients | 39.9 | 17.7 | 38.7 | 3.7 |
| The EHR will disrupt my workflow | 35.1 | 16.2 | 45.1 | 3.7 |
| Visits will be longer | 37.2 | 17.1 | 41.5 | 4.3 |
| The EHR will interfere with the patient visit | 27.1 | 11.3 | 58.5 | 3.0 |
| There will be too many alerts | 32.0 | 26.2 | 39.0 | 2.7 |
| The EHR will limit my discretion as a provider | 18.9 | 21.0 | 55.2 | 4.9 |
| The EHR will be more difficult to protect patient privacy | 19.8 | 15.6 | 61.6 | 3.1 |
Note: Percentage may not add up exactly to 100% due to rounding.
Overall, providers had positive attitudes about EHRs. For example, 86.9% of providers felt that having an EHR would improve access to patient information and 86.0% felt that the EHR would lead to fewer medication errors. The mean positive attitude score was 2.74 (SD 0.40) on a scale of 1 to 3 with 3 being more positive.
Despite these positive attitudes, providers had concerns about EHRs. For example, 58.5% thought the EHR would interfere with patient visits and 55.2% thought the EHR would limit their discretion as providers. The mean negative attitude score was 1.81 (SD 0.54) on a scale of 1 to 3 with 3 being more negative.
Table 3 shows the bivariate associations between physician characteristics and comfort with computers, positive attitudes, and negative attitudes scores. The values show the standardized difference (i.e., the difference in units of the standard deviation) between the reference category and a given physician characteristic. A higher value indicates a larger difference. No provider characteristic was significantly associated with comfort with computers or negative attitudes about EHRs (Table 3). General internists had less positive attitudes about EHRs than family practitioners (standardized difference = −0.26); gynecologists had more positive attitudes about EHRs than family practitioners (standardized difference = 0.20).
Table 3.
Association between physician and practice characteristics and comfort with and attitudes about EHRs
| Comfort with Computers | Positive Attitude about EHRs | Negative Attitude about EHRs | ||||
|---|---|---|---|---|---|---|
| Standardized Difference | p-value | Standardized Difference | p-value | Standardized Difference | p-value | |
| Provider work load | ||||||
| <20 hours/wk | ref | 0.27 | ref | 0.87 | −0.08 | 0.41 |
| ≥20 hours/wk | 0.14 | 0.02 | ||||
| Type of provider | ||||||
| MD or DO | ref | 0.47 | ref | 0.20 | ref | 0.41 |
| Non-MD (PA, NP) | 0.20 | −0.21 | −0.11 | |||
| Provider Specialty | ||||||
| Family Practice | ref | 0.78 | ref | 0.02 | ref | 0.29 |
| Adult General Internal Medicine | −0.02 | −0.26 | −0.26 | |||
| Pediatrics | −0.08 | −0.03 | −0.04 | |||
| Obstetrics and Gynecology | −0.07 | 0.20 | 0.03 | |||
| Other | 0.17 | −0.47 | −0.04 | |||
| Years in practice | ||||||
| 0–5 | ref | 0.08 | ref | 0.21 | ref | 0.14 |
| 6–10 | −0.08 | 0.19 | −0.13 | |||
| 11–20 | −0.32 | −0.02 | −0.20 | |||
| >20 | −0.17 | 0.03 | −0.11 | |||
| Gender | ||||||
| Female | ref | 0.27 | ref | 0.15 | ref | 0.31 |
| Male | −0.11 | −0.11 | −0.07 | |||
Note: Standardized difference is difference from the reference category on scale of the standard deviation. A higher absolute value signifies a larger difference from the reference category.
Within the first twelve months of having the EHR, over half of visits had allergy information entered into a structure field (mean 59.5%, SD 32.0), medications reviewed (64.8%, SD 30.0), and blood pressured entered (74.3%, SD 27.5; Table 4). However, less than a quarter of visits in which a prescription was generated had that prescription electronically prescribed (24.5%, SD 28.3).
Table 4.
Measures of use in the first year after implementation
| Mean percentage (SD) (n=302) | |
|---|---|
| Visits with allergy information entered into a structured field | 59.5 (32.0) |
| Visits with medication reviewed | 64.8 (30.0) |
| Visits with BP entered | 74.3 (27.5) |
| Visits with a prescription generated where the prescription was e-prescribed | 24.5 (28.3) |
Table 5 shows the associations between comfort with computers, positive attitudes, and negative attitudes scores and performance on four measures of EHR use in multivariable analysis. The values show the correlation coefficients (i.e., degree of correlation) between scores and measures of use. In multivariate regression analysis, we found no significant relationship between comfort with computers, positive attitudes about EHRs, or negative attitudes about EHRs for any of the measures of use (Table 5).
Table 5.
Association between comfort with computers and attitudes about EHRs and measures of use in the first year
| Dependent variable | ||||||||
|---|---|---|---|---|---|---|---|---|
| Percentage of visits with allergy information entered into a structured field, coefficient | p- value | Percentage of visits with medication reviewed, coefficient | p- value | Percentage of visits with BP entered, coefficient | p- value | Percentage of visits with a prescription generated and prescription was e-prescribed, coefficient | p- value | |
| Comfort with Computers Score | 1.49 | 0.35 | 0.09 | 0.95 | 1.11 | 0.40 | 1.56 | 0.19 |
| Positive attitudes about EHRs Score | −1.09 | 0.44 | −0.56 | 0.70 | 0.57 | 0.64 | 0.98 | 0.39 |
| Negative about EHRs attitudes Score | −0.69 | 0.55 | −0.11 | 0.92 | 0.50 | 0.56 | 0.93 | 0.37 |
Note: Coefficient is the association between a one standard deviation increase in each scale score and the dependent variable. All models controlled for provider characteristics (workload, provider type, specialty, gender, years in practice) and the quarter-year in which the practice started using the EHR.
Discussion
In this study of providers in small practices in New York City, we found that most providers had positive expectations for how the EHR would affect their delivery of patient care. Even with positive attitudes, however, almost a third of providers had concerns about the EHR - particularly about whether it would decrease their time with patients. Contrary to our hypothesis that provider comfort with computers and attitudes (both positive and negative) prior to adoption would predict measures of EHR use after implementation, we found no significant relationship between attitudes prior to implementation and EHR use.
These findings are encouraging given prior reports of concern by physicians about EHRs and prior research showing that positive attitudes correlate with successful implementation.12,13,18,19 For example, a study by Ancker and colleagues found that in interviews with PCIP providers, positive attitudes about EHRs correlated with successful EHR implementation.12 Our findings suggest that once an EHR is in place, prior comfort with computers and attitudes about the EHR do not predict future use of that EHR. These findings differ from findings of EHR adoption studies and may signify that once providers have overcome the hurdle of adopting an EHR, their attitudes about it do not affect their future use of EHR functions.
Given the significant amount of resources that have been invested into incentive programs to promote the meaningful use of EHRs, these are optimistic findings for policymakers: providers seem to overcome barriers like discomfort with computers and negative attitudes and use the EHR just as successfully as those who feel comfortable and have positive attitudes. That being said, we did see unexplained low utilization of EHR functions among all providers.
Our study has some limitations. The providers in our sample volunteered to enroll in PCIP, so there could be selection bias in that the providers enrolled in the program have more positive attitudes about EHR adoption than a random group of practices. There could also be response bias, in that the providers who responded to the survey might have greater comfort with computers and might hold more positive attitudes about adopting an EHR than those that did not respond. This bias may be compounded by the fact that a higher percentage of providers who received the survey by email responded compared with those who received a paper survey. In addition, there was little variation in the comfort with computers and positive attitudes scores. The majority of providers were both comfortable with computers and optimistic about adopting an EHR. As a result, we may not have had enough variation to predict EHR use. However, we did not find a correlation between negative attitudes and use even though the negative attitude score had more variation. Additionally, the questions about comfort with computers were relatively basic. We did not ask questions about experience with data management like familiarity with spreadsheets or ability to create a data table using a word processing program. We also only looked at four measures of use. There may be additional use measures that are important for quality of care but that we did not examine. Finally, the providers in our study practice in underserved areas in an urban setting. Due to this practice environment and patient population, it may be difficult to extrapolate our results to small practices outside of New York City.
All of the providers used one EHR system, e-ClinicalWorks. Though it is also of interest how EHR use may vary across vendors, we do not have access to such data. In addition, the EHR use measures included in this report were derived from queries of structured fields within each practice’s EHR system. Variation in documentation workflows across practices may lead to underreporting of EHR use. Though PCIP, in consultation with eCW, designed queries to pull from the most commonly used structured fields when generating EHR use reports, practices and their staff may have chosen to document information in fields not included in the measurement query, such as a free text or history field, or customized or specialized structured fields. The consequence for these practices is that their rates may show up as ‘zero’ or lower than what is actually occurring in the practice.
In summary, we found that comfort with computers and attitudes about EHRs did not predict future use of the EHR. Despite reports of negative attitudes about EHRs and prior findings that attitudes predict successful implementation of EHRs, our findings suggest that meaningful use of the EHR may not be affected by these attitudes. If so, programs such as the meaningful use program may not need to overcome this hurdle and as a result may hold more promise to be effective.
Appendix A.
Association between physician and practice characteristics and comfort with and attitudes about EHRs (non-standardized)
| Comfort with Computers | Positive Attitude about EHRs | Negative Attitude about EHRs | ||||
|---|---|---|---|---|---|---|
| Difference | p-value | Difference | p-value | Difference | p-value | |
| Provider work load | ||||||
| <20 hours/wk | ref | 0.31 | ref | 0.86 | -0.07 | 0.40 |
| ≥20 hours/wk | 0.09 | 0.01 | ||||
| Type of provider | ||||||
| MD or DO | ref | 0.46 | ref | 0.21 | ref | 0.37 |
| Non-MD (PA, NP) | −0.06 | −0.16 | −0.10 | |||
| Provider Specialty | ||||||
| Family Practice | ref | 0.77 | ref | 0.009 | ref | 0.31 |
| Adult General Internal Medicine | −0.01 | −0.20 | −0.25 | |||
| Pediatrics | −0.05 | −0.02 | −0.04 | |||
| Obstetrics and Gynecology | −0.04 | 0.15 | 0.03 | |||
| Other | 0.11 | −0.36 | −0.04 | |||
| Years in practice | ||||||
| 0–5 | ref | 0.09 | ref | 0.23 | ref | 0.18 |
| 6–10 | −0.05 | 0.13 | −0.12 | |||
| 11–20 | −0.21 | −0.02 | −0.18 | |||
| >20 | −0.11 | 0.02 | −0.10 | |||
| Gender | ||||||
| Female | ref | 0.29 | ref | 0.14 | ref | 0.35 |
| Male | −0.07 | −0.07 | −0.07 | |||
Appendix B.
Association between comfort with computers and attitudes about EHRs and measures of use in the first year (non-standardized)
| Dependent variable | ||||||||
|---|---|---|---|---|---|---|---|---|
| Percentage of visits with allergy information entered into a structured field, coefficient | p- value | Percentage of visits with medication reviewed, coefficient | p- value | Percentage of visits with BP entered, coefficient | p- value | Percentage of visits with a prescription generated and prescription was e-prescribed, coefficient | p- value | |
| Comfort with Computers Score | 1.98 | 0.22 | 0.31 | 0.83 | 1.21 | 0.36 | 1.47 | 0.23 |
| Positive attitudes about EHRs Score | −0.73 | 0.60 | −0.31 | 0.83 | 0.85 | 0.48 | 1.05 | 0.37 |
| Negative attitudes about EHRs Score | −0.26 | 0.83 | 0.05 | 0.96 | 0.60 | 0.48 | 0.69 | 0.52 |
Note: All models controlled for provider characteristics (workload, provider type, specialty, gender, years in practice) and the quarter-year in which the practice started using the EHR.
Acknowledgments
Funding: This project was funded by the Agency for Healthcare Research and Quality (Grant No. 18275). Dr. Bishop is also supported by a National Institute On Aging Career Development Award (K23AG043499) and as a Nanette Laitman Clinical Scholar in Public Health at Weill Cornell Medical College. Dr. Ryan is supported by an Agency for Healthcare Research and Quality Career Development Award (K01 HS018546).
Footnotes
The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Chaudhry B, Wang J, Wu S, et al. Systematic review: impact of health information technology on quality, efficiency, and costs of medical care. Ann Intern Med. 2006;144:742–52. doi: 10.7326/0003-4819-144-10-200605160-00125. [DOI] [PubMed] [Google Scholar]
- 2.Goldzweig CL, Towfigh A, Maglione M, Shekelle PG. Costs and benefits of health information technology: new trends from the literature. Health Aff (Millwood) 2009;28:w282–93. doi: 10.1377/hlthaff.28.2.w282. [DOI] [PubMed] [Google Scholar]
- 3.DesRoches CM, Campbell EG, Vogeli C, et al. Electronic health records’ limited successes suggest more targeted uses. Health Aff (Millwood) 2010;29:639–46. doi: 10.1377/hlthaff.2009.1086. [DOI] [PubMed] [Google Scholar]
- 4.Keyhani S, Hebert PL, Ross JS, Federman A, Zhu CW, Siu AL. Electronic health record components and the quality of care. Med Care. 2008;46:1267–72. doi: 10.1097/MLR.0b013e31817e18ae. [DOI] [PubMed] [Google Scholar]
- 5.Kinn JW, Marek JC, O’Toole MF, Rowley SM, Bufalino VJ. Effectiveness of the electronic medical record in improving the management of hypertension. J Clin Hypertens (Greenwich) 2002;4:415–9. doi: 10.1111/j.1524-6175.2002.01248.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Linder JA, Ma J, Bates DW, Middleton B, Stafford RS. Electronic health record use and the quality of ambulatory care in the United States. Arch Intern Med. 2007;167:1400–5. doi: 10.1001/archinte.167.13.1400. [DOI] [PubMed] [Google Scholar]
- 7.Buntin MB, Burke MF, Hoaglin MC, Blumenthal D. The benefits of health information technology: a review of the recent literature shows predominantly positive results. Health Aff (Millwood) 2011;30:464–71. doi: 10.1377/hlthaff.2011.0178. [DOI] [PubMed] [Google Scholar]
- 8.Blumenthal D, Tavenner M. The “meaningful use” regulation for electronic health records. N Engl J Med. 2010;363:501–4. doi: 10.1056/NEJMp1006114. [DOI] [PubMed] [Google Scholar]
- 9.DesRoches CM, Audet AM, Painter M, Donelan K. Meeting meaningful use criteria and managing patient populations: a national survey of practicing physicians. Ann Intern Med. 2013;158:791–9. doi: 10.7326/0003-4819-158-11-201306040-00003. [DOI] [PubMed] [Google Scholar]
- 10.Centers for Medicare and Medicaid Services. [Accessed March 15, 2013.];Press Release: More than 100,000 Healthcare Providers Paid for Using Electronic Health Records. 2012 Jun 19; Available at http://www.cms.gov/apps/media/press/release.asp?Counter=4383&intNumPerPage=10&checkDate=&checkKey=&srchType=1&numDays=3500&srchOpt=0&srchData=&keywordType=All&chkNewsType=1.
- 11.Centers for Medicare and Medicaid Services. [Accessed March 15, 2013.];EHR Incentive Programs. Available at http://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/index.html.
- 12.Ancker JS, Singh MP, Thomas R, et al. Predictors of success for electronic health record implementation in small physician practices. Appl Clin Inform. 2013;4:12–24. doi: 10.4338/ACI-2012-09-RA-0033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Rao SR, Desroches CM, Donelan K, Campbell EG, Miralles PD, Jha AK. Electronic health records in small physician practices: availability, use, and perceived benefits. J Am Med Inform Assoc. 2011;18:271–5. doi: 10.1136/amiajnl-2010-000010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Morton ME, Wiedenbeck S. A framework for predicting EHR adoption attitudes: a physician survey. Perspect Health Inf Manag. 2009;6:1a. [PMC free article] [PubMed] [Google Scholar]
- 15.Simon SR, Kaushal R, Cleary PD, et al. Physicians and electronic health records: a statewide survey. Arch Intern Med. 2007;167:507–12. doi: 10.1001/archinte.167.5.507. [DOI] [PubMed] [Google Scholar]
- 16.Anderson GF, Greenberg G, Lisk CK. Academic health centers: exploring a financial paradox. Health Aff (Millwood) 1999;18:156–67. doi: 10.1377/hlthaff.18.2.156. [DOI] [PubMed] [Google Scholar]
- 17.Centers for Medicare and Medicaid Services. [Accessed March 10, 2014.];Medicare & Medicaid EHR Incentive Program Meaningful Use Stage 1 Requirements Overview. Available at http://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/Downloads/MU_Stage1_ReqOverview.pdf.
- 18. [Accessed March 15, 2011];Despite incentives, doctors are wary about switching to electronic health records. 2011 at http://www.washingtonpost.com/national/despite-incentives-doctors-are-wary-about-switching-to-electronic-health-records-/2011/02/24/ABvRCuV_story.html.)
- 19.Boonstra A, Broekhuis M. Barriers to the acceptance of electronic medical records by physicians from systematic review to taxonomy and interventions. BMC Health Serv Res. 2010;10:231. doi: 10.1186/1472-6963-10-231. [DOI] [PMC free article] [PubMed] [Google Scholar]