Associations between Ambulatory Practice-Setting Characteristics and Demonstration of Stage 1 Meaningful Use Objectives for the EHR Incentive Program: A Descriptive Study at an Integrated Delivery System

Abstract

Background

The Medicare and Medicaid Meaningful Use (MU) incentive programs promote adoption and “meaningful use” of certified electronic health records among hospitals and eligible professionals (i.e., providers in outpatient settings), with a goal of improving quality of patient care. Despite the potential importance of MU for providers and patients, little is currently known about the characteristics of practice settings that facilitate providers’ demonstration of MU (i.e., meeting MU objectives). This study examined whether selected practice-setting characteristics were associated with providers’ meeting Stage 1 MU objectives at the end of year one in a single large North Carolina health center.

Methods

Our retrospective database analysis included all 702 eligible providers from 54 ambulatory care practices on the main campus of the University of North Carolina Health Care System. We assessed associations between providers’ ability to meet Stage 1 MU objectives as of December 2012 and the following characteristics of their practice settings: practice specialty, size, and mix of Medicare- and Medicaid-eligible providers.

Results

The following practice characteristics were associated with providers’ ability to meet MU objectives: primary care practices as compared to specialty practices (OR = 2.49, 95% CI [1.11, 5.62]), small practices as compared to medium-sized practices (OR = 0.29, [0.09, 0.89]), and the presence of only Medicare-eligible providers in the practice as compared to the presence of only Medicaid-eligible providers (OR = 6.48, [1.08, 38.97]).

Limitations

Because our sample was drawn from one integrated delivery system, results may not be generalizable to all ambulatory practice settings.

Conclusions

This study suggests that larger practices, primary care practices, and practices comprised of Medicare-eligible providers may be better able to meet MU objectives. Further research is needed to evaluate strategies that account for practice characteristics and other contextual factors in the MU implementation process.

INTRODUCTION

The Medicare and Medicaid electronic health record incentive programs are intended to promote adoption and “meaningful use” (MU) of certified electronic health records (EHR) among hospitals and eligible professionals (i.e., providers) in outpatient settings. Under the programs, hospitals and providers who are deemed eligible according to MU criteria receive incentive payments for satisfying specified EHR usage objectives, which are defined in three stages. In stage 1, eligible providers in outpatient settings must meet all of 14 required core objectives and five menu objectives selected from a set of 10 options [1]. Examples of these objectives include maintaining up-to-date medication lists for patients in the EHR (core) and using the EHR to provide patient-specific educational materials (menu) [2]. (See Table 1.) During the first year of participation, objectives must be met for a period of at least 90 days during the calendar year in order for a provider to “attest” to meaningful use demonstration (i.e., meeting MU objectives) and receive the incentive. MU also requires providers to report on clinical quality measures; however, there is no minimum threshold for performance for year one of stage 1 [3].

Table 1.

Eligible professional (EP) Meaningful Use stage 1 core and menu objectives

	Core Objectives	Menu Objectives
1	Computerized provider order entry (CPOE) for medication orders	Implement drug formulary checks
2	Drug-drug interaction and drug-allergy checks	Incorporate clinical lab test results into EHR as structured data
3	Maintain an up-to-date problem list of current and active diagnoses	Generate lists of patients by specific conditions to use for quality improvement, reduction of disparities, research, or outreach
4	Generate and transmit permissible prescriptions electronically (eRx)	Send reminders to patients per patient preference for preventive/follow-up care
5	Maintain active medication list	Provide patients with timely electronic access to their health information (including lab results, problem list, medication lists, and allergies) within 4 business days of the information being available to the EP
6	Maintain active medication allergy list	Use certified EHR technology to identify patient-specific education resources and provide those resources to the patient if appropriate
7	Record patient demographics: preferred language, gender, race, ethnicity, date of birth	The EP who receives a patient from another setting of care or provider of care or believes an encounter is relevant should perform medication reconciliation
8	Record and chart changes in vital signs: height, weight, blood pressure, calculate and display body mass index (BMI), plot and display growth charts for children 0–20 years including BMI	The EP who transitions their patient to another setting of care or provider of care or refers their patient to another provider of care should provide summary care record for each transition of care or referral
9	Record smoking status for patients 13 years old or older	Capability to submit electronic data to immunization registries or immunization information systems and actual submission according to applicable law and practice
10	Report ambulatory clinical quality measures to CMS	Capability to submit electronic syndromic surveillance data to public health agencies and actual submission according to applicable law and practice.
11	Implement one clinical decision support rule relevant to specialty or high clinical priority along with the ability to track compliance with that rule
12	Provide patients with an electronic copy of their health information (including diagnostic test results, problem list, medication lists, medication allergies) upon request
13	Provide clinical summaries for patients for each office visit
14	Protect electronic health information created or maintained by the certified EHR technology through the implementation of appropriate technical capabilities

Note: This table summarizes information provided by the Centers for Medicare & Medicaid Services. For a complete description of objectives and measures go to http://www.cms.gov/Regulations-and-Guidance/Legislation/EHRIncentivePrograms/Downloads/EP-MU-TOC.pdf

Although the underlying motivation behind MU is to improve the quality of patient care, the ability of providers to satisfy MU objectives will have substantial financial implications for providers and practices going forward. Current incentives are capped at a total of $44,000 over 5 years of participation for Medicare-eligible providers and $63,750 over 6 years of participation for Medicaid-eligible providers. Beginning in 2015, Medicare-eligible providers who do not demonstrate MU will receive cumulative penalties to physician fee schedule payments of 1% per year up to a maximum of 5% [4].

Despite the potential importance of meaningful use demonstration to patient care and population health [5] as well as the financial implications for providers and the practices in which they work, little is currently known about factors associated with providers’ ability to meet MU objectives. To best support providers and direct limited resources where they are needed most, policy makers, health systems, and practice leaders need information about characteristics of the practice setting that may be associated with greater or lesser success among providers in meeting MU objectives.

This study examines whether three practice-setting characteristics — primary care or specialty setting, size (i.e., small, medium, or large based on the number of providers), and the mix of Medicare- and Medicaid-eligible providers in the setting — are associated with providers’ demonstration of MU Stage 1 objectives during the first year of participating in the program. Regarding differences between primary care and specialty settings, evidence suggests that primary care physicians are more likely than specialists to believe that EHRs will improve the quality and efficiency of the health care system [6]. Furthermore, even though providers have some flexibility in terms of selecting menu objectives and quality measures for MU, there have been a substantial number of complaints about the appropriateness and/or difficulty of specific MU requirements depending on specialty. For example, some specialists have argued that maintaining the problem list falls in the domain of the primary care provider [7]; however, maintaining an updated problem list is a core stage 1 MU objective. Given that specialists may not value MU as much as primary care providers do, we hypothesize that providers in specialty practices will be less likely to meet MU objectives (i.e., demonstrate MU) than their colleagues in primary care practices, even within an integrated delivery system utilizing a system-wide MU implementation strategy.

Regarding practice size, evidence from previous EHR studies indicates that providers in smaller practice settings are concerned about having the capacity to effectively install EHRs, and these providers do not use MU-relevant functionalities available in their EHRs (e.g., e-prescribing) as regularly as do providers in larger practices [8]. Such findings suggest that larger practices may have more resources available (e.g., human and financial) to dedicate to the MU change effort as compared to smaller practices. Given the potential difference in resource availability, we hypothesize that providers in larger practices will be more likely to demonstrate MU than providers in smaller practices.

The relative mix of providers in a practice setting who are participating in the Medicare versus Medicaid incentive programs may be important because Medicare providers were required to demonstrate MU objectives for a 3-month period during the first year of Stage 1, whereas Medicaid providers were not required to demonstrate the objectives at all in the first year. Because individual providers who are eligible for both programs must choose which one to participate in, the implementation schedule for MU within a practice setting may depend on whether the setting has only providers pursuing the Medicare incentive program, only providers pursuing the Medicaid incentive program, or a combination of providers pursuing either program. Settings that have only Medicare providers likely have a more aggressive implementation schedule than settings with only Medicaid providers because Medicare providers must demonstrate MU in order to receive the incentive payments in the first year of Stage 1, whereas Medicaid providers do not need to demonstrate in the first year in order to receive the incentives. However, settings with a combination of both Medicare and Medicaid providers must decide whether to have one implementation schedule for both types or to allow providers to demonstrate MU on different schedules according to the requirements of their incentive program. Furthermore, in subsequent years, the MU expectations and schedule will be similar for Medicare and Medicaid providers. Therefore, the association between Medicare-Medicaid provider mix and provider demonstration of MU objectives will be an important aspect of assessing capacity for MU going forward and, to the extent that MU impacts care delivery, the quality of care for Medicare and Medicaid populations. We hypothesize that providers in Medicare-only settings would be more likely to demonstrate MU in the first year of Stage 1 than those in Medicaid-only and mixed Medicare-Medicaid settings, even within an integrated delivery system with a common MU implementation strategy. We also hypothesize that Medicaid-eligible providers will demonstrate MU sooner if they are in settings with Medicare-eligible providers than if they are in Medicaid-only settings due to spillover caused by the system-wide MU implementation strategy.

The overall goal of this study was to determine whether the specialty, size, and mix of Medicare and Medicaid eligible providers at the practice-setting level are associated with providers’ success in demonstrating MU Stage 1 objectives at the end of the first year of MU program participation. All of the practice settings in our study belonged to the same integrated delivery system, operated under the same system-level policies (including policies governing how MU incentives would be shared between individual providers and their practice settings), had access to the same MU implementation support, and used the same homegrown EHR system which had been certified for MU. Regarding the system-level policy for sharing of MU incentive payments between providers and their practice settings, the University of North Carolina Health Care System (UNC HCS) required both Medicare- and Medicaid-eligible providers to meet the same demonstration requirements (i.e., those required of Medicare-eligible providers as specified by the MU program) in order to receive a portion of the MU incentives. This policy is notable because MU program criteria require Medicaid providers only to adopt, implement, or upgrade MU-certified EHR technology in year one of Stage 1; they are not required to demonstrate MU objectives at all in the first year.[9]

METHODS

Sample and Data Source

Our sample included all eligible providers (n=702) from 54 academic ambulatory care clinics on the main campus of UNC HCS. We obtained our study data from a dashboard developed by UNC HCS to track MU demonstration status. This dashboard draws from the UNC HCS data warehouse and administrative systems. Using these data, we were able to identify how many providers were demonstrating MU (i.e., meeting or exceeding required measures for the core and menu objectives) at a particular point in time within a given practice setting. Our outcome of interest was demonstration of stage 1 MU as of December 2012.

It is important to note that demonstration at December 2012 is a different measure than attestation status. For the Medicare program, a provider could attest to meeting the MU objectives in year one of Stage 1 based on 90 days of performance during the calendar year. For the Medicaid program in year one of Stage 1, providers could attest by adopting, implementing, or upgrading to certified EHR technology, a lower threshold than meeting the stage 1 MU objectives. Therefore, for both programs, providers could have attested for MU but not have been demonstrating the stage 1 MU objectives at year-end. We selected December MU demonstration as the outcome of interest because we believed that Medicare providers demonstrating at this point in time would likely be better able to demonstrate MU for the following 12-month period, which is required for the second year of Stage 1. Furthermore, Medicaid providers demonstrating in December would be ahead of schedule with respect to their demonstration requirements and therefore better positioned to meet objectives for year two of Stage 1.

Data Analysis

We analyzed our data at the provider level using generalized estimating equation (GEE) methods with an independence working correlation matrix to account for clustering of providers within clinics. Because the downloaded data were initially aggregated at the clinic level, we were only able to include clinic-level predictors in the model. However, we chose to conduct our analysis at the provider level rather than an analysis aggregated at the clinic level because GEE appropriately weights the contribution from each clinic according to clinic size [10]. We first assessed associations between demonstration of meaningful use and the clinic characteristics by applying separate logistic regression models for each characteristic (analogous to conducting separate chi-square tests for each characteristic, albeit adjusted for clustering). We then fit a full model that simultaneously included all three clinic characteristics. We considered P<0.05 to indicate statistically significant associations, with no adjustments for multiple comparisons.

RESULTS

Summary statistics for the clinics and providers are presented in Table 2. Of all providers included in the study, 87.7% were demonstrating MU as of December 2012. Table 3 summarizes the logistic regression results. Providers in primary care practices were more likely (OR = 2.49, 95% CI [1.11, 5.62]) to be demonstrating MU than were providers in specialty practices. Providers in small clinics were significantly less likely than providers in medium-sized clinics (OR = 0.29, [0.09, 0.89]) to be demonstrating MU. Providers in small clinics also appeared to be less likely than providers in large clinics to be demonstrating MU, but the results were not statistically significant (OR = 0.41, [0.14, 1.20]). Providers in Medicare-only clinics were much more likely than providers in Medicaid-only clinics to be demonstrating MU (OR = 6.48, [1.08, 38.97]). Also, providers in combination Medicare-Medicaid clinics were more likely to be demonstrating MU than those in Medicaid-only clinics (OR = 2.89, [0.74, 11.21]), although the results were not statistically significant. Results were similar in the full model.

Table 2.

Characteristics of practice settings in the study sample

Clinic Characteristics	Number of Clinics	Number of Eligible Providers	% of Eligible Providers Demonstrating MU
Practice type
Primary care	5	148	93.9%
Specialty	49	554	86.1%
Practice size
<5 eligible providers (small)	14	34	73.5%
5–20 eligible providers (medium)	28	252	90.5%
>20 eligible providers (large)	12	416	87.3%
Incentive programs within clinic
Medicaid only	9	39	71.8%
Medicare only	15	70	94.3%
Both Medicaid and Medicare	30	593	88.0%

Note: “Medicaid only” refers to practices with 100% Medicaid-eligible providers. “Medicare only” refers to practices with 100% Medicare-eligible providers.

Table 3.

Estimated associations of structural characteristics with demonstrating meaningful use

	Unadjusted			Full Model*
Clinic Characteristics†	OR	95% CI	P-value	OR	95% CI	P-value
Practice type
Primary care vs. Specialty	2.49	(1.11, 5.62)	0.027	2.63	(1.13, 6.10)	0.025
Practice Size
Small vs. medium	0.29	(0.09, 0.89)	0.042	0.22	(0.06, 0.79)	0.020
Small vs. large	0.41	(0.14, 1.20)	0.102	0.37	(0.10, 1.39)	0.139
Medium vs. large	1.39	(0.60, 3.23)	0.448	1.65	(0.74, 3.71)	0.223
Incentive programs within clinic
Both vs. Medicare only	0.45	(0.12, 1.64)	0.224	0.29	(0.07, 1.15)	0.078
Both vs. Medicaid only	2.89	(0.74, 11.21)	0.125	2.58	(0.53, 12.44)	0.239
Medicare only vs. Medicaid only	6.48	(1.08, 38.97)	0.041	8.82	(1.58, 49.26)	0.013

^*Full model simultaneously includes all three characteristics.

^†For each characteristic, all pairwise comparisons were estimated from the same logistic regression model by specifying appropriate linear combinations of the model parameters. Thus, no single characteristic level need be specified as a “reference level”.

OR = odds ratio; CI = confidence interval.

IMPLICATIONS FOR PRACTICE

Given the emphasis at the national level on using EHRs as a tool for improving health care delivery and public health, it is important to assess whether characteristics of practice settings are associated with providers’ ability to demonstrate meaningful use of an EHR. Without such an assessment, it is difficult to tailor policies and allocate resources effectively to support providers in practice settings that are not as conducive to MU demonstration. The establishment of Regional Extension Centers, such as the one in the North Carolina Area Health Education Centers (NC AHEC), assists providers with MU primarily in small rural settings and are an example of such support [11]. However, our study shows that characteristics of practice settings even within the same integrated delivery system and located in the same geographic area are associated with variation in providers’ MU demonstration. Therefore, the small, rural settings may not be the only ones in need of additional, tailored support.

Primary Care vs. Specialty Settings

Consistent with evidence that primary care providers are more optimistic about the potential benefits of EHRs for improving care delivery [6] and have more widely adopted EHRs with MU capability than specialists have [12], we observed that providers in primary care settings were more likely to be demonstrating MU than providers in specialty settings. This finding is consistent with apparent variation in providers’ beliefs about the appropriateness and difficulty of some MU objectives depending on specialty [7]. An implication of this finding for policy makers is that incorporating input from various types of providers when developing MU objectives may facilitate faster demonstration of those objectives by all providers. For health system leadership, such findings suggest that specialty practices may need additional education about the importance of MU for the health system and perhaps additional support at the practice level to achieve MU demonstration.

Practice Size

Regarding practice size, in our sample, providers in small practices were less likely than their peers in medium and large practices to be demonstrating MU (although the latter comparison was non-significant). This finding may indicate that small practices have fewer resources and less slack in their operational systems to accommodate substantial process changes. These practices also may be less likely to have providers willing to champion the cause of MU because they have fewer providers. An implication of this finding for policy makers and system leaders is that small practices may continue to need external support to demonstrate MU, particularly as MU objectives require more sophisticated quality improvement activities and process changes.

Medicare-Eligible vs. Medicaid-Eligible

With respect to the mix of Medicare-eligible and Medicaid-eligible providers, providers in Medicare-only practices were much more likely than those in Medicaid-only practices to be demonstrating MU. There are a couple reasons that may explain these findings. First, the MU program does not require providers in the Medicaid program to demonstrate any Stage 1 objectives to attest for incentive payments in the first year of stage 1. However, UNC HCS had an internal policy that required Medicaid providers to demonstrate the same MU objectives in stage one as required for Medicare providers in order to receive their portion of the MU incentive payments. Despite policies to tie Medicaid incentives to meeting Medicare expectations, Medicaid providers are not at risk of penalties for their Medicaid reimbursement (although they are at risk of penalties for Medicare reimbursement if they see Medicare patients). Therefore, Medicare providers had more of a personal financial stake in demonstrating the objectives than did their Medicaid peers. It is notable, that the majority (71%) of eligible providers in Medicaid-only clinics did demonstrate MU. This finding suggests promise for an organized effort to support MU within an integrated delivery system, including institutional policies governing whether individual providers would receive a portion of MU incentives. In this case, these policies required Medicare- and Medicaid-eligible providers to meet the same demonstration requirements (i.e., those for the Medicare track) in order to receive a portion of the MU incentives.

Limitations

This study has some limitations. First, there may be practice-level contextual factors (e.g., leadership, culture) and provider-level characteristics (e.g., age) that influence providers’ MU demonstration but were not measured in this study. Although our database did not capture provider-level characteristics, our analysis accounted for practice-level factors by clustering providers within practices. Second, because our data come from an integrated delivery system with a system-wide approach to providing MU support to practices, our results may not be generalizable to practices that do not have similar policies or support. Furthermore, the efforts of one provider to demonstrate MU may have positive spillover effects for other providers who see the same patients; therefore, results may differ in a practice setting in which specialists and primary care providers do not share an EHR. Finally, because practices in our study were all using the same EHR, our results do not account for differences in design across EHR systems that may influence providers’ ability to successfully demonstrate MU.

CONCLUSIONS

Our study indicated that practice-setting characteristics were associated with providers’ success in demonstrating Stage 1 MU objectives at the end of year one, even within an integrated delivery system that has employed a system-wide MU implementation strategy. Policy makers and health system leaders should account for such structural differences when specifying MU objectives and allocating resources to support providers in their MU efforts. Future research could evaluate strategies that aim to account for these differences, providing a foundation for evidence-based MU implementation strategies tailored to practices with different structural characteristics.