Table 2.
PICO characteristics of included studies.
| ID | Authors | P (population) | I (intervention) | C (comparator) | O (outcomes) | Quantitative findings | Outlook on EMR |
|---|---|---|---|---|---|---|---|
| 3 | Beresniak et al. (2016) | Oncology sector | EHR4CR | Current practice | Expected net benefit | The expected benefits were estimated at €161.5 m (clinical scenario S1, protocol feasibility assessment), €45.7 m (clinical scenario S2, patient identification for recruitment), €204.5 m (S1 + S2), €1906 m (clinical scenario S3, clinical study execution), and up to €2121.8 m (S1 + S2 + S3) | Positive |
| 2 | Li et al. (2012) | Hospital patients | EMR | Paper-based medical record | NPV, ROI | The net benefit (total) from EMR implementation for a 6-year period was $559,025 in the general hospital. The time of return on investment is 3 years; and the pessimistic time of return on investment is 5.38 years. | Positive |
| 1 | Choi et al. (2013) | Outpatients | EMR | Paper-based medical record | NPV, BCR, DPP | The estimated NPV was US$3,617,000 for an 8-year period. The estimated BCR was 1.23. The estimated DPP was about 6.18 years. | Positive |
| 11 | Nuckols et al. (2015) | Inpatients | CPOE | Paper ordering system | Cost, QALY, ICER | CPOE, on average, had 99% probability of yielding savings to society and improving health, compared to the paper ordering system. Per hospital (by size, approximated by number of beds), mean lifetime savings –in millions- were $11.6 (25-72 beds), $34.4 (72-141 beds), $71.8 (141-267 beds), and $170 (267-2,249 beds) (2012 dollars). Quality-adjusted life-years (QALYs) gained were 19.9, 53.7, 109, and 249, respectively. Nationwide, anticipated increases in CPOE implementation from 2009 through 2015 could save $133 billion and 201,000 QALYs. | Positive |
| 12 | Sevick et al. (2017) | Inpatients | EDCT | Usual care | Cost, QALY, ICER | The incremental cost effectiveness ratio (cost per QALY gained) was estimated at $C239,933 for EDCT arm compared with usual care. There was a small gain in effectiveness and approximately $C800 difference in resource utilization costs. | Positive |
| 15 | Ben-Assuli et al. (2016) | Patients in ED | EHR | Without EHR | Cost, QALY, ICER | The incremental cost effectiveness ratio (cost per QALY gained) was estimated at $1,228.52 when the EHR system was made available to physicians compared to when EHR was not available. | Positive |
| 13 | Spaulding et al. (2013) | Hospitals | CPOE | Before CPOE | Cost | Even when 100% CPOE usage was not attained in hospitals, there remain benefits. From 51- 90% usage is associated with the lowest predicted nursing cost costs per patient day. A large increase in nursing salaries was associated with 91-100% usage and the most beneficial cost outcomes accrue at under 50% usage for the pharmacy. | Positive |
| 10 | Kazley et al. (2014) | Inpatients | EHR | Without EHR | Cost | After accounting for variations in patient and hospital characteristics, it was estimated that on average patients treated in hospitals with advanced EHRs cost $731 (approximately 9.66%) less than patients admitted to hospitals without advanced EHRs. | Positive |
| 14 | Teufel et al. (2012) | Pediatric patients | EMR | Without EMR | Cost | EMR is creating a safer health care system but not always associated with inpatient cost savings. Advanced stage EMR was associated with an average 7% greater cost per case ($146 per discharge). | Neutral |
| 6 | Dranove et al. (2014) | Hospitals | EMR | Without EMR | Operating cost | EMR adoption is initially associated with a rise in cost. EMR adoption at hospitals in IT-intensive locations leads to a decrease in costs after 3 years. Hospitals in other locations experience an increase in costs even after 6 years. | Neutral |
| 5 | Atasoy et al. (2018) | Hospitals | EHR | Without EHR | Operating cost | The adoption of an additional EHR system in the focal hospital increases its own costs 1.8% in the current year and 2.3% in 4 years. If hospitals in the same HSA (neighboring hospitals). The adoption an additional HER system corresponds to 1% decrease in the costs of the focal hospital in the current year, and a cumulative effect of 1.5% decrease in four years. | Positive |
| 27 | Zhivan et al. (2012) | Hospitals | CPOE | Without CPOE | Cost | Hospitals, on average, exceeded costs at the frontier by 16%. The hospital cost-inefficiency is positively related to the EMR adoption decision, but not CPOE adoption. An 1%-point increase in inefficiency score was associated with a 3.3% increase in the odds of EMR adoption. | Neutral |
| 18 | Eastaugh et al. (2012) | Hospitals | EHR | Without EHR | Nursing productivity | It was estimated that 25% improvement in financial health of a hospital is associated with 5.1% increase in EHR. A 25% increase in school dependency on the hospital as a source of clinical rotation leads to a 2.0% increase in EHR. The implementation of EHR was associated with a 1.6% improvement in productivity. | Positive |
| 9 | Furukawa et al. (2010) | Surgical units in hospital | EMR | Without EMR | Cost | EMR stages 1 and 2 were associated with significantly higher inefficiency scores. EMR stage 3 shows no or negative association with inefficiency, depending on the estimation models used. It was concluded that EMR, overall, is associated with higher inefficiency in medical-surgical acute settings | Negative |
| 8 | Dupont et al. (2017) | Service providers | EHR4CR | Without EHR4CR | Cost, Revenue | It was estimated that a profitability ratio 1.8 or higher could be achieved at year 1. There are potential for growth for the ratio in subsequent years if the market uptake is higher. | Positive |
| 7 | Driessen et al. (2013) | Inpatients | EMR | Without EMR | Cost | It was estimated that the total cost savings US$284,395 annually (for in length of stay, transcription time, and laboratory use). There is a net financial gain by year 3, after accounting for the costs of installing and sustaining the EMR system. The estimated cost savings was US$613,681 over the 5 years. | Positive |
| 17 | Gowrisankaran et al. (2016) | Hospitals | EMR (CPOE) | Without EMR (CPOE) | Administrative practice in hospital (upcoding of medical and surgical procedures) | EMR adoption in hospital led to increases in reported severity for medical relative to surgical patients at EMR hospitals because EMRs decreases coding costs for medical patients. Medicare costs might increase by $689.6 million annually with post-reform completeness of coding with EMRs. There was a positive and significant impact from EMR adoption on the mean DRG weight following the reform. | Neutral |
| 22 | Haque et al. (2015) | Patients | EMR | Without EMR | Length of stay, thirty-day mortality, thirty-day readmission | It was found that EMRs had the largest impact for relatively less-complex patients. Admission to a hospital with an EMR is associated with a 2% reduction in length of stay and a 9% reduction in thirty-day mortality for less complex patients. In contrast, there was no evidence of statistically significant benefit for more-complex patients in hospital with EMR. | Positive |
| 24 | Hydari et al. (2019) | Patients | CPOE & EMR | Without CPOE & EMR | Patient safety events (medication errors, falls, complications) | EMRs were found to lead to a 17.5% decline in patient safety events, driven by reductions in medication errors, falls, and complication errors. There was also a decline in medium- and high-severity events with advanced EMRs. | Positive |
| 21 | Freedman et al. (2014) | Inpatients | EMR (CPOE & PD) | Without EMR (CPOE & PD) | Preventable adverse events | There was evidence that EMRs improve patient safety (reduced the likelihood of adverse events), particularly for less complex patients. Adoption of CPOE was associated with an 11% drop in the probability of experiencing at least one postoperative adverse event for cases with no more than one comorbidity and a 17% drop in probability for patients with more common DRGs. The results indicated EMR is likely to have the greatest impact on patient safety indicators when the technology has a decision support feature that is relevant and accurate for the patient’s condition. | Positive |
| 16 | DesRoches et al. (2010) | Hospitals | EHR | Without EHR | Quality of care, risk-adjusted length of stay, readmission rate, cost | The relationship between quality and efficiency were modest at best and not statistically significant. Hospitals with EHR had slightly better performance on prevention of surgical complication (93.7% for hospitals with comprehensive EHR, compared to 93.3% with basic EHR, and 92% without). Length of stay was about 0.5 days shorter for cases of pneumonia in hospitals with comprehensive EHR compared to those without. Inpatient costs are comparable across hospitals with and without EHR. | Neutral |
| 4 | Iturrate et al. (2016) | Patients requiring lab tests | EHR with modified ordering system | EHR without modification | Lab test per patient adjusted for outcome | Following introduction of the modified EHR ordering system there was a significant reduction in target lab tests per patient day. Segmented regression analysis indicated a 20.9% reduction in the utilization of target lab tests. Student’s t test analysis indicated a .5% reduction. The estimated reduction in hospital costs was $300,000 due to the EHR modification. | Positive |
| 28 | Zlabek et al. (2011) | Service records in hospital | EHR, CPOE | Without EHR and CPOE | Lab test, radiology examinations, paper consumption, transcripts, medication errors and near misses, | Laboratory tests per week per hospitalization decreased about 18% (from 13.9 to 11.4). Radiology examinations per hospitalization decreased 6.3% (from 2.06 to 1.93). Monthly transcription costs declined 74.6% (from $74 596 to $18 938). Reams of copy paper ordered per month decreased 26.6% (from 1668 to 1224). Medication errors per 1000 hospital days decreased 14.0% (from 17.9 to 15.4). Near misses per 1000 hospital days increased 38.9% (from 9.0 to 12.5). The percentage of medication events that were medication errors decreased from 66.5% to 55.2%. | Positive |
| 25 | Miller et al. (2011) | Babies | EMR | Without EMR | Neonatal mortality | EMRs increase speed and accuracy of access to patient records, leading to improved diagnosis and monitoring. It was estimated that a 10% increase in births that occur in hospitals with EMR reduces neonatal mortality by 16 deaths per 100,000 live births. The estimated cost-effectiveness suggested that EMR was associated with $531,000 per baby’s life saved. | Positive |
| 26 | Xue et al. (2012) | Inpatients | EMR | Without EMR | Length of stay, infection rate, mortality rate, and cost per inpatient | EMR was associated with reduced length of stay, infection rate, and mortality rate but had no correlation with patient costs. Length of stay grew at 0.027 bed-days per month in the pre-EMR period and declined at 0.043 bed-days per month in the post-EMR period. Infection rate rose at 0.036 infections per 100 patients per month in the pre-EMR period and declined at 0.062 infections per 100 patients per month in the post-EMR period. Mortality rate grew at 0.048 deaths per 1000 patients per month in the pre-EMR period and decreased at 0.005 deaths per 1000 patients per month in the post-EMR period. Cost per patient stay declined at 33 RMB per month in the pre-EMR period and increased at 16 RMB per month in the post-EMR period. | Neutral |
| 20 | Encinosa et al. (2012) | Adverse events in patients | EMR | Without EMR | Death, 90-day readmission for surgeries, 90-day hospital expenditure for surgeries | While EMRs did not reduce the rate of patient safety events, they reduce death by 34%, readmissions by 39%, and spending by $4,850 (16%) if a safety event occurred. This led to a cost offset of $1.75 per $1 spent on IT capital. | Positive |
| 19 | Encinosa et al. (2013) | Inpatients | EMR with meaningful use (MU) requirements | EMR without MU | Hospital acquired adverse events | It was estimated that hospital cost savings per averted adverse events were $4,790. If all hospitals in Florida had adopted all 5 functions in the EMR, 55,700 ADEs would have been averted and $267 million per year would have been saved. The cost savings was estimated to recoup only 22% of information technology costs. | Neutral |
| 23 | Himmelstein et al. (2010) | Hospitals | EHR | Without EHR | Quality of care, administrative costs as share of total cost | Hospitals on the “Most Wired” list performed equally compared to others on quality, costs, or administrative costs. Hospital computing however might modestly improve process measures of quality. | Neutral |
ED emergency department, EMR electronic medical records, EHR electronic health records, EHR4CR electronic health records for clinical research, CPOE computerized physician order entry, PD physician documentation, EDCT electronic discharge communication tool, PHR personal health record, NPV net present value, BCR benefit cost ratio, DPP discounted payback period, ROI return on investment, QALY quality adjusted life year, ICER incremental cost-effectiveness ratio.