Annie N. Simpson, PhD

 The asthma medication ratio (AMR) (number of controller medications / [number of controller medications + number of rescue medications]) can be calculated using claims data. This measure has not previously been studied longitudinally. Our objective is to conduct a longitudinal examination of the AMR in a large national cohort of children with asthma.

 Retrospective analysis of pharmacy and medical claims data.

 Using 2013-2014 TruvenHealth MarketScan data, we identified children with asthma. Beginning with the month of first controller claim, we calculated an AMR for each rolling 3-month period and each rolling 6-month period and examined the proportion who had AMRs classified as low-risk (≥0.5), high-risk (<0.5), and missing for each period. Using logistic regression, we tested how a rolling AMR predicted a child’s hospitalization or emergency department (ED) visit for asthma.

 We identified 197,316 patients aged 2 to 17 years with a claim for a controller. AMRs were relatively stable over time, with the majority of patients remaining in the same AMR category through a 12-month period. Using both the rolling 3-month and 6-month AMRs, a higher proportion of patients with high-risk AMRs (9.6% and 9.5%,

respectively) had an ED visit or hospitalization compared with patients with low-risk (5.0% and 5.7%) and missing (3.5% and 3.2%) AMRs (

<.0001). Using logistic regression, the 3-month AMR is more strongly associated with subsequent ED visit or hospitalization than the 6-month AMR.

AMR-based risk assignment is relatively stable over time. Three-month AMR calculation periods appear to provide the most accurate assessment of risk. Children with missing AMRs likely have inactive asthma and are at the lowest risk for emergent asthma visits.

The asthma medication ratio (AMR) can be calculated using pharmacy claims data and used to identify patients with asthma who are at highest risk for exacerbation in the coming months.

AMR-based risk assignment is relatively stable over time.

In any given time period, 5% to 8% of children with asthma will be at high risk.

Children with no pharmacy claims for either rescue or controller medications are at lowest risk for exacerbation.

Asthma remains a frequent cause of emergency department (ED) visits and hospitalizations among children.

 Controller medications, particularly inhaled corticosteroids (ICSs), are effective at reducing the incidence of these acute care visits for asthma.

 However, these medications continue to be underutilized.

 Using pharmacy claims data to identify patterns of poor controller medication adherence is a potential way to target medication adherence interventions to high-risk children. With this in mind, the asthma medication ratio (AMR; number of controller medication claims / [number of controller medication claims + number of rescue medication claims]) has been developed to measure adherence and assign risk for exacerbation.

 Findings from previous studies have shown that the AMR predicts risk for future exacerbation on the patient level.

 The AMR has the potential to risk stratify large populations of children with asthma in real time, thereby accurately identifying the patients at highest risk for exacerbation and allowing for intervention before exacerbation occurs. This could ultimately prevent costly ED visits and hospitalizations, driving down healthcare costs and improving quality of life attributed to this common pediatric chronic disease.

Despite its potential for risk assessment and risk communication to prevent exacerbations, the AMR has not yet been translated to a point-of-care, real-time monitoring tool. All previous AMR studies have utilized a fixed cross-sectional AMR assessment period, capturing adherence behaviors for 1 specific moment. Before designing and testing an AMR-based intervention, we must better understand the longitudinal behavior of the AMR using rolling periods. This represents the most practical way to calculate the AMR in real time and will allow risk assessment using the most recent claims data available.

Traditionally, studies have relied on the Healthcare Effectiveness Data and Information Set (HEDIS) criteria for persistent asthma to determine who is eligible for AMR measurement. HEDIS is a quality tool and was designed to measure systems, not individuals. HEDIS criteria work well for reporting on the performance of health systems through tools such as the Quality Compass. Previous studies' results have shown that patients often do not meet HEDIS criteria for persistent asthma in consecutive years and that the number of consecutive years of HEDIS qualification was strongly associated with ICS use.

 HEDIS requires up to 1 year of claims monitoring in order to classify a patient as persistent asthmatic. This would lead to missed opportunities for intervention during the measurement year and potential inappropriate interventions the following year. Instead, we propose that the AMR can be measured in all children with a pharmacy claim for an ICS. According to the National Heart, Lung, and Blood Institute (NHLBI) Guidelines for the Diagnosis and Management of Asthma, ICSs are recommended only for children with persistent asthma.

 Therefore, it is unlikely that children with intermittent asthma would be identified using this criterion.

The objective of this study was to examine the longitudinal behavior of the AMR among a large national cohort of privately insured children with asthma, including a comparison of the predictive accuracy of a 3-month rolling AMR with that of a 6-month rolling AMR, as well as determining the proportion of patients in each risk category at any given time.

In order to identify all children for whom the AMR would be a potentially valid risk assessment tool, we first identified all asthma medication claims from the Truven Health MarketScan pharmacy claims databases for 2013 and 2014 using National Drug Code numbers. Medications were then categorized as rescue medications or controller medications. Patients aged 2 to 17 years with any claim for an ICS-containing medication were eligible for inclusion in our study cohort. Because we took an alternative approach to identifying the cohort of children eligible for AMR measurement, we assessed the proportion of our cohort that met HEDIS persistent asthma criteria and determined in any given AMR measurement period how many patients would qualify as at high risk for exacerbation but not as persistent asthmatic according to HEDIS criteria.

We defined each patient’s study index date as the date of his or her first ICS-containing medication claim in the study period. Patients with fewer than 360 (12 × 30) days of continuous insurance enrollment after their index date and patients with a diagnosis of cystic fibrosis (

International Classification of Diseases, Ninth Revision

] code 277.XX) at any time during the study period were excluded. We then identified all inpatient and ED visit claims with a primary diagnosis of asthma (

 code 493.XX) for each patient in the cohort. Only those claims representing visits that occurred after the patient’s index date were retained in the final analytical data set. Covariates used in this analysis include patient age (defined as age at index date, operationalized continuously and in age categories), sex, geographic region, and season of index date (winter, January-March; spring, April-June; summer, July-September; fall, October-December). MarketScan does not include a race variable.

Using the formula of number of controller medication claims/(number of controller medication claims + number of rescue medication claims), an AMR was calculated for each patient in each month from the index date to study month 12. AMRs can range from 0 (only rescue medication claims) to 1 (only controller medication claims) or be “missing” (no rescue or controller medication claims) for any given period. Leukotriene-receptor modifiers were included as controllers in the AMR calculation if there was not also an ICS-containing medication claim that month, but they were not included as index controllers. This decision was made to ensure that our cohort represented children with persistent asthma who received first-line guideline-recommended therapy for that diagnosis. Oral albuterol was not included as a rescue medication.

Similar to previous studies, we first calculated fixed 12-month, 6-month, and 3-month AMRs for each patient for the 12 months after the index date. Therefore, each patient has one 12-month AMR, 2 fixed 6-month AMRs, and 4 fixed 3-month AMRs. Next, we calculated rolling AMRs for each patient. Rolling 3-month AMRs were calculated using months 1 to 3, 2 to 4, 3 to 5, and so on. Rolling 6-month AMRs were calculated using months 1 to 6, 2 to 7, 3 to 8, and so on.

Based on previous study results, AMRs were classified as high-risk (<0.5), low-risk (≥0.5), or missing for each calculation period. Because of our definition of index date, the AMR values for the first period (those that include month 1) are artificially inflated. Identifying this phenomenon influenced further analytical decisions.

Population-level AMR distribution through 12 study months was plotted using stacked bar charts. Patients in each category (high-risk, low-risk, and missing) for the first analyzable rolling 6-month and 3-month period were tracked at the population level over the course of the year. We also determined the proportion of patients who remained in the same risk category from one month to the next.

To identify differences by patient characteristics,

AMR classification changes over time were also assessed by age group, season of index date, and category of index controller (ICS vs ICS/long-acting β agonist [LABA]).

ED visits and hospitalizations with a primary diagnosis of asthma (

 code 493.XX) were identified. Having any ED visit or hospitalization for asthma in a given period was the primary dichotomous outcome for this study. Future studies should include more in-depth analyses of the rolling AMR’s relationship to this outcome as well as individual outcomes of ED visits, hospitalizations, and oral steroid dispensing events.

To begin to understand how patients with missing AMRs should be classified in future interventional studies, we compared proportions of patients with any ED visit or hospitalization for asthma among those classified as having high-risk, low-risk, and missing AMRs in the first analyzable rolling 3-month and 6-month periods (months 2-4 and months 2-7, respectively). Chi-square tests were used to identify any statistically significant differences.

Determining the Relative Strength of Association of 3-Month Versus 6-Month Rolling AMRs With Events

Simple logistic regression models with the outcome of ED visit or hospitalization in 3-month and 6-month outcome windows were built for rolling 3-month AMRs and rolling 6-month AMRs. Comparisons of odds ratios (ORs) between the 3-month AMRs and 6-month AMRs were used to quantify strength of association to help inform the decision of which calculation strategy to use in subsequent analyses.

Of the 9.5 million children aged 2 to 17 years present in the 2013 MarketScan data, 197,316 patients had least 1 claim for an ICS or ICS/LABA and at least 360 days of continuous enrollment after their index date. Of these patients, 60% were male and the mean age was 8.8 years; 36% were aged 2 to 6 years, 41% aged 7 to 12, and 23% aged 13 to 17. Forty-seven percent of assigned index dates were in the winter (January-March). Eighty-three percent of children had an ICS as their index controller. Ultimately, 4.5% of patients had at least 1 ED visit or hospitalization with a primary diagnosis of asthma within 18 months of their study index date (

HEDIS Categorization of Patients in Cohort

Forty-two percent of the patients in our cohort qualified as persistent asthmatic using the HEDIS criteria. The majority of the 58% who would not qualify as such under HEDIS criteria (“non-HEDIS”) are represented in the missing AMR category in any given AMR measurement period. For example, in months 2 to 4, 85,152 of the 114,320 non-HEDIS patients had a missing AMR; 26,569 had a low-risk AMR; and 2599 had a high-risk AMR (

Rolling 3-Month Versus Rolling 6-Month AMR

Using a rolling 3-month AMR calculation period and excluding the month 1 values due to the issue of AMR inflation, an average of 5% of the cohort were identified as high-risk in each calculation period, while 59% had a missing AMR (no asthma prescriptions filled during the calculation period) (

). Using a rolling 6-month AMR calculation period and excluding the month 1 values, an average of 8% of the population were identified as high-risk in each calculation period and an average of 45% had a missing AMR (

). Moving from a 3- to a 6-month calculation period significantly reduced the proportion of patients with missing AMRs in each given AMR calculation period. In regression analysis with the outcome variable of any ED visit or hospitalization for asthma, the 3-month AMRs had a stronger predictive ability than the 6-month AMRs (OR for 3-month AMR in months 2-4 with 3-month outcome window, 2.5; 95% CI, 2.1-2.9; OR for 6-month AMR in months 2-7 with 3-month outcome window, 1.8; 95% CI, 1.6-2.1). 

 presents complete results through the entire study period.

The issue of AMR inflation with our index date assignment is demonstrated graphically (

). Because time is not accounted for in the AMR, even 1 month with a controller fill can significantly inflate a patient’s AMR for the following 11 months.

There was a slightly higher proportion of children 13 years and older in the high-risk AMR category compared with younger children (

). A higher proportion of children with a winter index date maintained low-risk AMRs throughout the year (

). Children whose index controller was an ICS/LABA rather than an ICS were more likely to have a low-risk AMR throughout the year, indicating better adherence to controller medication therapy (

Do Children Stay in the Same Category Throughout the Year? 

To determine if early risk-category assignment held throughout the year or if children frequently bounced in and out of categories, we followed children in each group (low-risk, high-risk, and missing, based on their assignment in the first analyzable period) through the year. The majority of children with a missing AMR in the first analyzable period continued to have missing AMRs through month 12. Similarly, the majority who were initially low-risk remained low-risk through month 12. The biggest departures from original classification occurred in the high-risk group, with only 20% (in the 3-month approach) and 39% (in the 6-month approach) remaining high-risk through month 12 (

). Analyzed another way, of patients who had a high-risk AMR in any given calculation period, an average of 68% had a high-risk AMR in the following calculation period. Of patients with a low-risk AMR in any given period, an average of 84% had a low-risk AMR in the following calculation period. Finally, of patients with a missing AMR in any given period, an average of 90% had a missing AMR in the following calculation period.

With such a large proportion of the population having missing AMRs, it was important to begin to understand how to handle these patients. The fact that they had no albuterol claims suggests that they are at low risk for exacerbation, but could they actually be at high risk and either using albuterol from previous months or not recognizing their symptoms? To help answer this question, we used each patient’s classification from the first analyzable period and calculated the proportion of patients in each category with any emergent event in study months 1 through 18. The patients with missing AMRs had the lowest proportion of events in both approaches (3.5% based on 3-month AMR classification and 3.2% based on 6-month AMR classification). This was significantly lower than the proportion with events in both the high-risk and low-risk AMR categories (

<.0001). This finding suggests that those children with missing AMRs are likely to be children with inactive asthma at low risk for exacerbation (

This longitudinal analysis of the AMR in a large cohort of privately insured children with asthma supports the rolling AMR calculation as a practical approach to risk assessment. This approach is superior to a cross-sectional fixed-AMR calculation period approach as it allows for risk category assignment based on the most recent available claims data. When applied to this cohort of privately insured children with asthma, the rolling 3-month calculation approach identified approximately 5% of the population as being at high risk for exacerbation in any given period.

As the AMR calculation period decreases (from 12 to 6 to 3 months), a larger proportion of patients have missing AMRs in any given period. To have a missing AMR, the patient must not have any claims for rescue or controller medications in the AMR calculation period. The higher proportion of patients with missing AMRs in the 3-month approach suggests that the 6-month AMR calculation approach might be superior to the 3-month approach. However, further investigation supports the 3-month approach. First, we demonstrated that children with missing AMRs are less likely to have an ED visit or hospitalization for asthma compared with children with both low-risk and high-risk AMRs. These children appear to have relatively inactive asthma, illustrated by their absence of asthma medication claims and low rate of emergent care visits. This raises the idea that some children do not necessarily remain persistent asthmatics for long periods, reflecting variation in their asthma control. They may require controller medications most of the time, but they may also have periods of disease inactivity where they do fine without controller medications. Recognizing these patients clinically will be a challenge, however. We suggest that children with missing AMRs can be treated like children with low-risk AMRs in future interventional studies. Second, our regression analysis illustrates that the 3-month AMR has a stronger relationship to the outcome of ED visit or hospitalization for asthma compared with the 6-month AMR. However, both are statistically significant, suggesting that a 6-month AMR approach would be valid and acceptable.

Differences in adherence patterns between patients with ICS versus those with ICS/LABA raise the question of whether or not these 2 populations can be handled the same way in an AMR-based intervention study. There are several potential reasons that the latter group has better controller medication adherence than the ICS group. The NHLBI Guidelines for the Diagnosis and Management of Asthma recommend subspecialist consultation for any child requiring “Step 4” therapy, which includes ICS/LABAs.

 Therefore, these children are likely to have more severe baseline disease and to have been seen by a subspecialist than are children on ICS therapy alone. The cumulative medication adherence messaging that ICS/LABA patients receive might be more extensive and more effective than the messaging that ICS patients receive.

Because of our method of index date assignment, our study highlights the issue of AMR inflation. In a rolling approach with shorter AMR calculation periods, an initial month with a controller claim would be dropped as soon as the next month’s claims data are available, allowing for a more accurate AMR-based risk assessment. There is less potential for inflation with shorter rolling calculation periods.

Articles

This longitudinal examination of the asthma medication ratio in a national sample of children has determined the predictive accuracy of a rolling 3-month ratio.

A Longitudinal Examination of the Asthma Medication Ratio in Children

To determine whether advanced electronic health record (EHR) use in hospitals is associated with lower cost of providing inpatient care.

National Inpatient Sample (NIS) and the Health Information Management Systems Society (HIMSS) Annual Survey are combined in the restrospective, cross-sectional analysis. We study patients who are 18 years or older and discharged from a general acute care hospital.

Using 2009 data and a cross-sectional design with a gamma distributed generalized linear model, a patient-level analysis is conducted with propensity scores to control for selection bias. Patient- and organizational-level variables are included as controls. The main outcome measure is total cost per patient admission and represents the amount that it costs the hospital to provide services based on the adjusted charges for an admission.

We include 5,047,089 individuals treated at 550 hospitals in the United States and represent a population-based sample. There are 104 (18.9%) hospitals included that use advanced EHRs. Patients treated in hospitals with advanced EHRs cost, on average, $731, or 9.66%, less than patients admitted to hospitals without advanced EHRs, after controlling for patient and hospital characteristics.

Hospitals that use advanced EHRs have lower cost per patient admission than comparable hospitals with similar case mix.

Advanced EHR use in hospitals has the potential to save money on patient care.

		This study supports the case for advanced EHR adoption and use, consistent with meaningful use.

		Advanced EHR use requires costly investment for hospitals that can be regained through savings in patient care.

		Patients treated in hospitals with advanced EHRs cost 9.66% less than those treated in hospitals without advanced EHRs.

Electronic health records (EHRs) have been suggested as a tool for improving the overall quality and cost of care in the United States.

Proponents and policy makers have created incentives through the Health Information Technology for Clinical and Economic Health (HITECH) Act to offset the cost of purchase in order to encourage the adoption and use of advanced EHR systems in a “meaningful” way.

Such criteria for use are based on previous studies that report improvements in quality.

 To qualify as a meaningful user and benefit from the related incentives, EHR systems must include electronic prescribing, health information exchange with other providers, automated reporting of quality data, electronic recording of patients’ history (demographics, vital signs, medication and diagnosis lists, and smoking status), created care summary documents, and at least 1 clinical decision support tool.

 Such meaningful use requirements are believed to improve the legibility of records, reduce prescription errors, improve adherence to best clinical practice guidelines, improve patient and clinician access to records, and allow exchange of health information.

 In addition to gains in quality, EHRs have been predicted to save $81 billion annually through safety improvement and increased efficiency of care,

 yet little is known about their impact on hospital cost, and no previous studies have examined the relationship between cost per admission and EHR use in a national sample of acute care hospitals for adults.

Cost savings associated with EHRs are expected to come through better coordination of care, reduction of medical errors and adverse drug events (ADEs), and increased efficiency and reduction of duplicate testing; previous studies have demonstrated the potential. Silow-Carroll and colleagues found that at 9 hospitals with comprehensive EHR use, “Faster, more accurate communication and streamlined processes have led to improved patient flow, fewer duplicative tests, faster responses to patient inquiries, redeployment of transcription and claims staff, more complete capture of charges, and federal incentive payments,” which lead to cost specific EHR components, including automated notes and records, order entry, and clinical decision support, are associated with fewer complications, lower mortality rates, and lower costs in Texas hospitals.

 In a single hospital, inpatient EHR with computerized provider order entry (CPOE) use was associated with a decrease in laboratory tests, radiology examinations, monthly transcription costs, medication errors, and paper costs.

 On the other hand, a study performed in California revealed that EHR use was associated with an increase in hospital costs, nursing staff levels, and complications, but a decrease in mortality for some conditions.

 This same study did not find any evidence that advanced EHR use reduced length of stay or demand for nurses. Similarly, another study examining EHR use in physician practices found that electronic access to patient lab and imaging results may actually increase the number of overall tests given to individual patients by 40% to 70%, thus increasing costs.

 In a study of Medicare patient—level billing data from 1998 to 2005, Agha reported that EHR adoption was associated with an initial 1.3% increase in billed charges, and saw no evidence of cost savings over the 5 years postadoption.

In a previous study of a single hospital, physician inpatient order writing on microcomputers was found to be associated with reduced resource utilization.

 These costs of providing care were approximately 12% lower in the area of overall charges, hospital charges, bed charges, diagnostic test charges, and drug charges, although the system required increased physician time. Similarly, the cost savings of advanced EHR use found in this study may be the result of several practices or benefits associated with advanced EHR use. First, the automated nature of advanced EHRs may reduce errors through overall coordination of care, less duplication of tests, and increased efficiency.

Dranove and colleagues examined the implementation of EHR as a business process innovation.

 They used data from Medicare cost reports combined with survey data on EHR adoption to assess the economic implications of EHR adoption on mean cost per admission. They reported that EHR adoption was associated with a rise in cost, but that cost reduction for hospitals in favorable locations—where there was high availability of expertise and complementary services&mdash;resulted in cost reductions after 3 years, while hospitals with unfavorable conditions had increased costs, even after 6 years.

The objective of this study was to determine if advanced EHR use is associated with lower cost of care in acute care general hospitals. In an era of increased pressure to adopt EHRs and other health information technology, it is important to understand the benefits and challenges of EHR use. Our study adds to the work by Agha and Dranove by using more recent individual-level patient data that includes all payers.

 Our cost measure reflects the variation in cost per admission observed at the level of the individual, as opposed to the mean for the hospital, which allowed us to control for variations in case mix. We limited our analysis to contrasting hospitals with advanced EHRs to all other levels of implementation based on the finding by Dranove and colleagues that cost savings due to EHR implementation are most likely observed for institutions with a critical mass of EHR expertise.

The data for this cross-sectional patient-level analysis were obtained from the National Inpatient Sample (NIS) 2009 and Health Information Management Systems Society (HIMSS) 2009. The NIS includes discharge data from more than 1000 hospitals in 45 states, which encompasses 96% of the United States population. The HIMSS 2009 data were used to measure hospital EHR use. The HIMSS data “represent a broad canvassing of acute care hospitals, chronic care facilities, ambulatory practices on their adoption and plans to adopt various HIT components” and have been widely used in previous research.

 Patients were included in the analysis if they were 18 years or older. In the NIS, some states do not release the American Hospital Association identifiers, and thus the individual patient cases cannot be included because EHR use cannot be determined. Costs were calculated for each admission using the total charges reported multiplied by the hospital-specific cost-to-charge ratio for 2009. Admissions with zero charges were excluded from the analysis.

The generalized linear model and propensity models controlled for patient age, gender, race, All Patient Refined Diagnosis Related Groups (APDRGs) mortality and severity, neonatal or maternal status, private insurance coverage, Medicare or Medicaid coverage, Diagnosis Related Group (DRG) case mix group, and whether the patient arrived as a transfer. Both models also controlled for the following hospital-level variables: teaching status, urban location, bed size, and geographical region. The main outcome measure was total cost of hospital admission per patient-billed hospitalization. Outcomes such as length of stay and the effects of nurse staffing variables on outcomes were explored using similar modeling approaches.

EHR use is measured using stages based on individual applications reported in the hospitals. These include stage 0 (no automation), stage 1 (automation of ancillary services including a clinical data repository, and pharmacy, laboratory, and radiology information systems), stage 2 (stage 1 + automation of nursing work flow with electronic nursing documentation, and medication administration records), and stage 3 (advanced EHR including: stages 1 and 2 + CPOE and clinical decision support). Since meaningful use criteria are consistent with stage 3 adoption of EHR, we chose to compare hospitals that have advanced EHRs with all others, and this staging system has been used in previous research.

Generalized linear modeling techniques were used to test the hypothesis that the total costs per admission were different between hospitals with and without advanced EHRs. To correct for the non-normal distribution of costs, gamma-distributed generalized linear models using a logarithmic transformation

 were analyzed using the PROC GENMOD module in the SAS statistical software (version 9.2; SAS Institute Inc, Cary, North Carolina). The use of a gamma-distributed generalized linear model with a log-transformed link function has been shown to be an accurate method to estimate healthcare cost distributions that are generally right-skewed, especially when the logtransformed dependent variables do not have heavy tails or excessive heteroscedasticity such as was found to be true in these data.

We controlled for potential selection bias of advanced EHR use in hospitals and potential differences in patient demographics, severity, and hospital case mix through the use of a propensity score stratification. To calculate the propensity score, a logistic regression analysis was performed to estimate the propensity of each patient to be seen in a hospital with advanced EHRs. Use of a propensity score approach can remove upward of 95% of bias from estimates.

 Generalized estimating equation (GEE) methods were used to confirm that results remained the same after accounting for correlated outcomes across patients treated within the same hospital. We examined the potential for selection bias due to case mix, patient characteristics, and hospital characteristics in a sensitivity analysis using a 5% random sample of the data. This sensitivity analysis using a portion of the data in a propensity score—matching methodology allowed us to estimate the potential selection bias of known confounders that might have remained after using propensity score stratification methods. The data were prohibitively large, thus limiting propensity score–matching sensitivity analyses to a 5% sample. An additional sensitivity analysis was performed to assess the potential hidden bias potentiated by unknown confounders.

 This allowed us to examine how much the inferential findings of cost differences could be altered by hidden biases of various magnitudes and how large these differences would have to be to alter the qualitative conclusions of our study.

The analysis included 5,047,089 individual patient cases. Of these, 1,509,610 (29.9%) patients were cared for in hospitals that use advanced EHRs. There were 550 hospitals included in the analysis, and of these, 104 (18.9%) use advanced EHRs. The mean overall total cost per admission was $10,790, with the mean cost for patient admissions at hospitals with advanced EHRs being $10,203 and the mean cost for patient admissions at hospitals without advanced EHRs being $11,010. Other descriptive information about the patients and hospitals are provided in 

When controlling for patient and hospital characteristics in the multivariable model, the mean cost per patient admission for hospitals without advanced EHRs was $7938. The mean cost per patient admission for hospitals with advanced EHRs was $7207. Based on the generalized linear regression model, presence of an advanced EHR was significantly associated with cost of admission, and the beta of —0.0966 indicates that patients treated in hospitals with advanced EHRs have costs that are on average 0.0966%, or $731, lower than those for patients treated in hospitals without an advanced EHR. Other significant predictors of cost include, as expected, patient age, race, gender, severity of illness, risk of death, transfer status, case mix, and insurance type, as well as hospital size, type of hospital, teaching status, rural or urban location, and region of the country. The estimates of the multivariable model are presented in 

. When the propensity score adjustment was included, the association between advanced EHR use and cost remained statistically significant, with cost estimates remaining the same, indicating that selection bias was likely minimal. The GEE model controlling for potential correlation between patients seen at the same hospital remained significant (

 of advanced EHR =.04) and garnered the same results as seen in the generalized linear model.

We developed a 5% sample of the data using propensity score matching based on the nearest neighbor-matching greedy algorithm approach. We limited the sample to 72,002 randomly selected observations from each group to keep computational time below 8 hours. Propensity scores in the original sample ranged from 0.012 to 0.921. The heterogeneity between the EHR and non-EHR admissions was well controlled by the matching, as illustrated by the decrease in the absolute standardized difference in the means for all the propensity score model variables before and after the match (

). The cost differences between the EHR and non-EHR admissions in the matched groups were slightly larger than the cost difference observed for the total data set (

). The use of the GEE approach on the main data analysis did not make a significant difference in the results. We explored whether the observed relationship between advanced EHR and cost of care was also present for other measures of hospital resource use, and found $3305 lower total charges per admission when mean charges were modeled. When we explored length of stay (LOS), adjusting for covariates similar to the cost model, we found a mean LOS of 4.1 days for both groups (

 =.84). The resource input data available in the data set were limited. However, hospitals with advanced EHRs appear to have had a slightly higher rate of nursing fulltime— equivalent personnel per 1000 adjusted admissions (RNFTEs). The mean number of RNFTEs was 4.27 for the advanced EHR admissions and 4.15 for the comparison group in the propensity score–matched sample, for a difference of 0.09 (

 <.0001) RNFTEs after controlling for covariates in the model.

When assessing hidden bias due to unmeasured variables, we used methods described by Rosenbaum to estimate the sensitivity parameter, Γ, the degree of departure from a study that is free of hidden bias.

In order to attribute the lower hospital cost in patients in an advanced EHR hospital to an unmeasured factor, a hidden bias or unobserved covariate would need to increase the odds of being in an advanced EHR hospital by more than a factor of gamma >100. The inferential finding of lower admission cost in patients seen in advanced EHR hospitals was, therefore, very insensitive to hidden bias.

Hospitals have struggled to adopt and use EHRs because they are costly to purchase, implement, and maintain, especially in the current environment of cost containment and increased efforts to be “lean.”

 Thus, the HITECH incentives were created to help offset the costs by offering hospitals hundreds of thousands of dollars in payments for purchase and “meaningful use” of EHRs. This analysis provides evidence that advanced EHR use may result in cost savings for hospitals in addition to meaningful use incentives, and supports the business case for the costly implementation and maintenance of an EHR system. Such implementation should include plans for advanced EHR use.

During the period of study, only a minority of hospitals were using advanced EHR systems.

 The additional value of this study compared specific EHR components that are associated with stage 3, or meaningful use, and can be related to cost savings, including CPOE, clinical decision support systems, automation of ancillary services including a clinical data repository, pharmacy, laboratory, and radiology information, and automation of nursing work flow with electronic nursing documentation and medication administration records. The staging model that was used demonstrates that cost savings may not be realized until multiple features are included and implemented. Since EHR systems are complex and costly to implement, it is often a multistage process to adopt and use EHRs.

Thus, hospitals must anticipate that the financial savings may not exist until advanced, “meaningful” use is attained. The majority of hospitals have yet to reach the stage of implementation where cost savings are possible, since they are not using advanced EHRs.

The cost differences observed in this study were measured based on costs estimated from charges using the hospital’s reported mean cost-to-charge ratio for the year of the data, and it is possible that this estimate of cost could bias cost estimates for an individual hospital. However, it is unlikely that systematic bias from this approach would have skewed our results. Given the findings of this study, patients and payer groups may consider selecting hospitals for care based on advanced EHR use, especially if they are paying for care based on a formula related to charges. This may be especially true if the efforts to promote consumer-driven healthcare are successful and if these hospitals demonstrate high levels of quality. However, given the current arrangement where hospital prices are not related to what patients have to pay, the cost savings of advanced EHR use may be most valuable to third-party payers. Our business case also does not include the costs of implementation or maintenance of an EHR system, which can be significant: they range from hundreds of thousands to millions of dollars.

This study had several limitations. First, EHR use is not randomly assigned, and thus there is the potential for selection bias. To overcome this, we included a propensity score adjustment. Second, the NIS data did not include a fully representative sample, and some of the states included in the sample did not provide hospital identifiers for patient-level data. Those cases were excluded from the study, as they could not be linked to EHR use data. Also, because the data are from 2009, we could not be certain that the systems classified as “advanced” in the adoption model absolutely meet the requirements for meaningful use according to the HITECH Act. However, given the definitions and components included in our definition, they are most consistent with meaningful use. There is a possibility that hospitals with EHRs are better able to capture charges. Better charge capture will result in higher estimated cost per admission in our data. Thus, if EHR use improves charge capture, costs (estimated as charges adjusted by the cost-to-charge ratio) will appear higher, resulting in a higher cost per admission in the advanced EHR group in the current study. Finally, some individual cases were excluded because the NIS data did not provide a hospital identifier to allow EHR use to be determined. While we did not find evidence that the excluded cases were systematically different than the included case, this does present a potential weakness.

Hospitals that use advanced EHRs report lower costs per patient admission than hospitals that do not use advanced EHRs. These cost savings will benefit many thirdparty payers, hospitals, and patients, and incentives such as those provided through the HITECH Act to promote EHR adoption and use will benefit hospitals. Since many previous studies have shown that EHRs can improve the safety and quality of care in hospitals, the projected cost savings in this study provides additional motivation and builds the business case for hospitals to make the large investment in adopting and maintaining an EHR system. This study is a very large multistate study that is likely representative of national trends and builds upon previous work of cost savings associated with EHR use. The results provide support for the continued adoption and use of EHRs to improve healthcare through cost savings and quality improvements.

 Department of Health Care Leadership and Management (ASK, ANS, KNS); Department of Pediatrics, Medical University of South Carolina, Charleston (RT).

The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

 Concept and design (ASK, KNS, RT); acquisition of data (ASK, KNS, RT); analysis and interpretation of data (ASK, ANS, KNS); drafting of the manuscript (ASK, ANS, KNS, RT); critical revision of the manuscript for important intellectual content (ASK, KNS, RT); statistical analysis (ANS, KNS); administrative, technical, or logistic support (ASK).

 Abby Swanson Kazley, PhD, Department of Health Care Leadership and Management, Medical University of South Carolina, 151 Rutledge Ave, Charleston, SC 29425. E-mail: swansoaj@ musc.edu.1. Silow-Carroll S, Edwards JN, Rodin D. Using electronic health records to improve quality and efficiency: the experiences of leading hospitals. Issue Brief (Commonwealth Fund). 2012;17:1-40.

2. Amarasingham R, Plantinga L, Diener-West M, Gaskin DJ, Powe NR. Clinical information technologies and inpatient outcomes: a multiple hospital study. Arch Intern Med. 2009;169(2):108-114.

3. Zlabek JA, Wickus JW, Mathiason MA. Early cost and safety benefits of an inpatient electronic health record. J Am Med Inform Assoc. 2011;18(2):169-172.

The authors examine the association between advanced electronic health record (EHR) use and cost in hospitals. Patients treated in hospitals with advanced EHRs cost 9.66% less.

Annie N. Simpson, PhD

Articles

A Longitudinal Examination of the Asthma Medication Ratio in Children

Association of Electronic Health Records With Cost Savings in a National Sample