Claude M. Setodji, PhD

 We examined differences in patient outcomes associated with 3 patient-centered medical home (PCMH) recognition programs—National Committee for Quality Assurance (NCQA) Level 3, The Joint Commission (TJC), and Accreditation Association for Ambulatory Health Care (AAAHC)&mdash;among Medicare beneficiaries receiving care at federally qualified health centers (FQHCs).

 We used data from CMS’ FQHC Advanced Primary Care Practice Demonstration, in which participating FQHCs received assistance to achieve NCQA Level 3 PCMH recognition. We assessed the impact of the 3 recognition programs on utilization, quality, and Medicare expenditures using a sample of 1108 demonstration and comparison FQHCs.

 Using propensity-weighted difference-in-differences analyses, we compared changes in outcomes over 3 years for beneficiaries attributed to FQHCs that achieved each type of recognition relative to beneficiaries attributed to FQHCs that did not achieve recognition.

 Recognized FQHCs, compared with nonrecognized FQHCs, were associated with significant 3-year changes in FQHC visits, non-FQHC primary care visits, specialty visits, emergency department (ED) visits, hospitalizations, a composite diabetes process measure, and Medicare expenditures. Changes varied in direction and strength by recognition type. In year 3, compared with nonrecognized sites, NCQA Level 3 sites were associated with greater increases in ambulatory visits and quality and greater reductions in hospitalizations and expenditures (

<.01), TJC sites were associated with significant reductions in ED visits and hospitalizations (

<.01), and AAAHC sites had changes in the opposite direction of what we anticipated.

 Heterogeneous changes in beneficiary utilization, quality, and expenditures by recognition type may be explained by differences in recognition criteria, evaluation processes, and documentation requirements.

This study contributes to the literature assessing the effectiveness of medical home (MH) recognition programs on patient outcomes:

This analysis is the first to compare beneficiary outcomes across multiple MH recognition programs.

We document heterogeneity in the association between achievement of MH recognition and beneficiary outcomes among 3 MH recognition programs.

These study findings support the need to better understand how different components of MH recognition programs contribute to site-level changes and patient outcomes.

Many medical practices are pursuing primary care transformation using a medical home (MH) model.

 These models incorporate the joint principles of the patient-centered medical home (PCMH), which describe features of a strong primary care delivery system, including enhanced access, coordinated and comprehensive care, and continuous quality improvement.

 With time, additional administrative and financial burdens associated with primary care transformation have prompted a need to distinguish practices that systematically adhered to MH principles from those that did not.

Over time, payers have encouraged practices to pursue MH recognition in order to codify practices’ use of PCMH principles.

 As of 2017, 3 organizations offer 3 common forms of MH recognition: National Committee for Quality Assurance (NCQA; 12,000 practice sites), the Accreditation Association for Ambulatory Health Care (AAAHC; 6000 practice sites), and The Joint Commission (TJC; 1400 practice sites).

Despite the rapid growth of these programs, limited evidence exists on the relative effectiveness of different recognition types on patient outcomes.

 Although each program’s recognition standards align with core elements of the MH model, required elements vary for each program and may contribute to differences in outcomes. For example, recognition programs have differing criteria related to use of health information technology (IT), care coordination, and medication management. Additionally, each program assigns a different weight to these areas when measuring the extent of implementation of each program’s requirements.

Although descriptive work has compared recognition programs’ application procedures and recognition standards,

no analyses have examined patient outcomes associated with these programs.

 Using data from a nationwide evaluation of an initiative to assist federally qualified health centers (FQHCs) in becoming PCMHs, we examine changes in utilization, quality, and Medicare expenditures associated with the terminal (ie, highest) MH recognition status of 3 recognition programs: NCQA 2011 Level 3 PCMH, TJC Primary Care MH, and AAAHC MH. As NCQA 2011 Level 1 and Level 2 PCMH are not terminal recognition levels, they are not included in this analysis. We hypothesized that changes in visits to FQHCs, primary care outside of FQHCs, specialists, and emergency departments (EDs); inpatient admissions; quality of care; and expenditures would vary by MH recognition type.

The 3 recognition programs examined differ in the content and specificity of their criteria, evaluation processes, and documentation requirements. We reviewed descriptive studies comparing these programs, as well as their online resources, to provide context for our analysis.

NCQA has a practice site—level recognition program, meaning that sites affiliated with the same medical practice must individually pursue recognition. Practices pursuing recognition may achieve Level 1, 2, or 3 recognition based on the extent to which they meet specific standards.

 NCQA emphasizes meaningful use requirements; almost half of its recognition score is derived from IT capabilities.

In contrast, TJC recognition is awarded at the organizational level, meaning that a single recognition award is given to the organization spanning all of its practice sites. Practices seeking MH recognition must also meet criteria for TJC’s ambulatory care accreditation, which requires sites to implement foundational components for MH content areas; these do not disproportionately emphasize health IT capabilities. Uniquely, TJC requires sites to collect data on patient experience, which is not required for NCQA recognition.

Similar to TJC, AAAHC operates an organizational-level program allowing practices to pursue MH recognition in addition to a foundational ambulatory care certification.

 Burton et al found that a large proportion of AAAHC recognition criteria require practices to develop policies related to patient care, staff development, and prevention practices. However, limited guidance is provided about the specification of these policies, which allows variance in site-level interpretation and implementation.

 Almost a quarter of the AAAHC recognition score derives from the development of these policies.

The site and beneficiary sample used in this analysis was drawn from our evaluation of CMS’ FQHC Advanced Primary Care Practice (APCP) Demonstration.

 Demonstration sites received financial support and technical assistance to achieve NCQA Level 3 recognition within a 3-year period. Independent of this demonstration, both demonstration and comparison sites utilized other technical and financial resources through the Health Resources & Services Administration and CMS during the same period.

 Details of this demonstration’s evaluation methodology have been published and are supplemented in the 

We used Medicare claims data from 2010 to 2014 to attribute beneficiaries to FQHCs based on the plurality of their primary care visits during the year before the demonstration. The attribution process was repeated in each subsequent year for beneficiaries who utilized primary care in the years following the start of the demonstration. This analysis used data on beneficiaries attributed to 1 of 1108 of the evaluation’s 1330 FQHCs. We included 393 FQHCs with NCQA Level 3 (n = 232,990 beneficiaries), 100 FQHCs with TJC (n = 55,471), and 14 FQHCs with AAAHC (n = 14,020) recognition, and a group of 601 FQHCs that did not achieve recognition (n = 302,480) by the end of 2014 (

). Each of the 4 groups included demonstration and comparison sites under the FQHC APCP Demonstration (

). This analysis focuses on outcomes of beneficiaries attributed to 3 terminal recognition programs as distinct from foundational programs (eg, NCQA PCMH Level 1 and 2, TJC accreditation, and AAAHC certification). The remaining 222 of 1330 sites were excluded because they received a foundational recognition (n = 147), achieved multiple recognition types (n = 59), or were recognized by a program other than the 3 considered in this analysis (n = 16).

Utilization measures included FQHC visits, non-FQHC primary care visits (ie, visits to rural health clinics or physician offices but not FQHCs), and specialist visits. In each category, the visit could be to a physician, nurse practitioner, or physician assistant. Also included were ED visits, inpatient admissions, admissions for ambulatory care—sensitive conditions (ACSCs), and 30-day unplanned readmissions. We used 1 composite and 4 diabetes process quality measures (glycated hemoglobin [A1C] testing, low-density lipoprotein cholesterol testing, retinal eye exams, and nephropathy testing within the past year) and a measure of annual lipid tests for patients with ischemic vascular disease. We assessed effects on Medicare expenditures in 3 categories: inpatient, Part B (physician and supplier), and total Medicare expenditures (consisting of inpatient, outpatient, skilled nursing facility, home health, hospice, durable medical equipment, and Part B expenditures).

We used difference-in-differences (DID) analyses to compare changes in outcomes from baseline to the end of each demonstration year for beneficiaries attributed to NCQA Level 3—, TJC-, or AAAHC-recognized FQHCs compared with beneficiaries attributed to nonrecognized sites. A positive DID estimate can result from either a greater increase or smaller decrease in outcome measures for recognized versus nonrecognized sites. We used 2-part models to estimate effects on utilization and Medicare expenditures, with the first part using logistic regression and the second part using either negative binomial (utilization) or linear models (expenditures).

Readmissions and process measures were modeled using logistic regression. We used multinomial logistic regression to create propensity score weights reflecting each beneficiary’s propensity to be attributed to a site that achieved each recognition based on beneficiary and site-level characteristics measured the year prior to each beneficiary’s attribution (eAppendix). To provide context for the DID estimates, we present information on trends occurring between the baseline year and year 3 for all 4 groups. Lastly, we analyzed trends among groups for 2 years prior to the study to confirm DID model assumptions (eAppendix).

Sites associated with the 4 recognition outcomes were similar with respect to the demographics of patients they serve (

). The majority of FQHC users were female; half were disabled. When using propensity score weights to balance baseline characteristics across the 4 groups, we found only a few remaining differences: Beneficiaries attributed to AAAHC-recognized sites were more likely to be white and less likely to be black than those attributed to nonrecognized sites; beneficiaries attributed to TJC-recognized sites were more likely to live in metropolitan areas and less likely to live in rural areas than those attributed to nonrecognized sites; beneficiaries attributed to AAAHC-recognized sites were less likely to live in metropolitan or rural areas, more likely to live in urban areas, and less likely to live in high-poverty areas relative to those attributed to nonrecognized sites; AAAHC sites were likely to have fewer specialists relative to sites without recognition; and AAAHC sites had approximately 3 times the average number of Medicare beneficiaries as nonrecognized sites.

Several patterns in outcome measures emerged from baseline to the end of each demonstration year for beneficiaries attributed to recognized sites relative to nonrecognized sites. The first 4 columns in 

 show the direction of 3-year trends for each outcome measure for beneficiaries associated with each group. The remainder of Table 3 shows DID estimates and 

 values for analyses comparing sites with each recognition type with nonrecognized sites.

For FQHC visits, the first 4 columns show a statistically significant decrease across all 4 groups. The remaining columns show a net increase of 88, 172, and 180 more visits per 1000 beneficiaries for years 1, 2, and 3, respectively (

<.001). These estimates reflect a smaller reduction in FQHC visits among NCQA Level 3 sites compared with nonrecognized sites. For TJC- and AAAHC-recognized sites, compared with nonrecognized sites, all significant estimates are negative. This reflects a larger reduction among TJC- and AAAHC-recognized sites relative to nonrecognized sites.

All 4 groups were associated with an increase in non-FQHC primary care visits, but a smaller rate of increase for NCQA Level 3 sites and a larger rate of increase for AAAHC sites led to a relative reduction of 90 visits and a relative increase of 166 visits per 1000 beneficiaries in year 3, respectively, compared with nonrecognized sites (

<.01). Specialist visits similarly increased over time for all 4 groups, but only NCQA Level 3 sites were associated with a relative decrease over time (45 fewer visits per 1000 beneficiaries in year 3) relative to nonrecognized sites (

For nonambulatory utilization measures, only TJC sites achieved greater reductions in ED visits than nonrecognized sites (39 fewer ED visits per 1000 beneficiaries; 

<.01). By contrast, AAAHC sites were associated with an increase of 78 visits per 1000 beneficiaries in year 3 relative to nonrecognized sites (

<.01). This was due to a smaller reduction in ED visits compared with nonrecognized sites, rather than an increase in ED utilization over time. Although inpatient admissions increased in all 4 groups, NCQA Level 3 and TJC sites had smaller increases (11 and 18 fewer admissions per 1000 beneficiaries in year 3, respectively; 

<.001). Recognized sites were not associated with reductions in admissions for ACSCs or readmissions.

Few clear trends emerged among process measures. Both NCQA Level 3 and AAAHC sites were associated with larger increases over time in the diabetes composite measure (1.6 and 2.7 percentage points, respectively, in year 3) relative to nonrecognized sites (

<.05). However, for the individual diabetes measures, only NCQA Level 3 sites were associated with relative increases in A1C tests (1.0 percentage points), eye exams (1.4 percentage points), and nephropathy tests (2.1 percentage points). These were largely driven by smaller decreases in rates over time relative to nonrecognized sites.

Medicare expenditures increased over time for all 4 groups in each of the 3 categories we examined. However, only NCQA Level 3 sites had a smaller rate of increase than nonrecognized sites, leading to a relative decrease in total Medicare ($300 per beneficiary), inpatient ($214 per beneficiary), and Part B ($72 per beneficiary) expenditures in year 3 (

We compared trends in utilization, quality, and expenditures over 3 years for Medicare beneficiaries attributed to sites that received 1 of 3 types of MH recognition relative to sites with no recognition. The former were more likely to demonstrate changes in outcomes consistent with better access and quality of care, lower inpatient utilization, and lower expenditures in year 3. However, the direction and strength of these associations varied by recognition type.

Each recognition program was associated with a distinct pattern of relative change in FQHC utilization rates: a net increase, net decrease, or no difference relative to trends among nonrecognized sites. Relative changes in non-FQHC primary care visit rates were in the opposite direction to those of FQHC visits, suggesting that beneficiaries are exhibiting greater loyalty to FQHCs. Only NCQA Level 3—recognized sites achieved significant relative reductions in specialty visits by year 3, which may reflect reduced need due to the consolidation of primary care at these recognized FQHCs or implementation of more efficient referral systems.

Among non—ambulatory care utilization measures, NCQA- and TJC-recognized sites were associated with reductions in all-cause inpatient admissions. These changes are unlikely to be driven by trends in hospitalizations for ACSCs, where we observed no significant differences between recognized and nonrecognized sites. Instead, these reductions in rates of all-cause admissions may be related to improved primary care utilization or enhanced coordination within FQHCs or between FQHCs and outside specialists, which may lead to better management of patients with complex conditions or lower rates of specialty referrals. These factors might also explain TJC-recognized sites’ larger reductions in ED utilization over time. Large relative increases in ED utilization among AAAHC-recognized sites, along with reductions in FQHC visits and increases in non-FQHC visits, raise some concern about unintended consequences of MH implementation in these sites. However, a small number of AAAHC sites were included in this study, and with a future larger study, current null or small differences might become statistically significant.

Across all recognition programs, quality of care was more likely to improve relative to trends among nonrecognized sites. In year 3, we observed many statistically significant improvements for NCQA Level 3 sites and in the composite diabetes measure for AAAHC sites. Small sample sizes for TJC- and AAAHC-recognized sites may limit the ability to detect statistically significant associations and may explain results for Medicare expenditures, where we observed statistically significant changes only for NCQA Level 3 sites.

Some of the observed heterogeneity in associations with patient outcomes may be due to the specific recognition program criteria, which vary with respect to their content and specificity, as well as the evaluation process and required documentation. One major difference related to the content and specificity of criteria is that NCQA places a heavier emphasis on the use of electronic health records (EHRs) that incorporate meaningful use criteria.

 Requirements that focus on the utilization of EHRs can enhance primary care delivery and strengthen population health management. In comparison, TJC scoring criteria are weighted fairly consistently across multiple domains, rather than heavily emphasizing any one content area.

 Sites pursuing AAAHC recognition have to implement internal policies, but they receive limited guidance on the specificity of these policies. This may lead sites to implement practice changes that are not associated with patient outcomes measured in this analysis. Nevertheless, there may be unmeasured beneficial or adverse effects of these changes at the clinic or patient level.

Additionally, differences in the evaluation process and the associated documentation requirements may contribute to heterogeneity in patient outcomes. For example, sites pursuing NCQA recognition are responsible for completing worksheets, responding to follow-up questions, and providing documentation, a process that takes approximately 40 to 80 hours to complete and upload online.

 In comparison, TJC and AAAHC use external surveyors to perform on-site evaluations without requiring additional time to upload documentation materials. Sites pursuing AAAHC recognition provide supplemental documentation to the external surveyors for recognition, but sites seeking TJC recognition are only required to provide documentation to receive ambulatory care accreditation, a prerequisite for MH recognition. These required tasks add an administrative burden to staff who also are involved in the implementation of PCMH strategies and daily operational activities.

This analysis had several limitations. First, the small number of sites achieving AAAHC recognition limited our ability to identify statistically significant associations with many outcome measures. Second, we were unable to randomize sites to recognition programs. Accordingly, we used DID analyses to control for patient- and site-level differences. Nevertheless, it is possible that unmeasured confounders could bias the effects of recognition types, including self-selection of sites to recognition programs. For example, heterogeneity in the criteria and evaluation process among recognition programs, and how those align with a site’s existing practice structure, may influence sites’ decisions to pursue one recognition type over another. Sites’ self-selection to recognition programs may explain differences by program if these factors are correlated with beneficiary outcomes.

Third, a large proportion of sites achieving NCQA Level 3 recognition participated in the CMS FQHC APCP Demonstration and received support to pursue NCQA recognition. Despite our DID design, including examination of parallel trends (eAppendix), this group may differ from sites receiving other types of recognition. This may explain the strong associations with many outcomes for NCQA-recognized sites. To estimate the effect of recognition on outcomes independent of the effect of the demonstration, we conducted a sensitivity analysis replicating these with 687 comparison sites. We found fewer statistically significant results, although they were consistent with outcomes reported in this analysis (eAppendix).

Fourth, we measured associations between MH recognition and beneficiary outcomes over a 3-year period, but many sites obtained recognition toward the end of the study period. A longer duration of follow-up after achieving recognition may be needed to observe effects on outcomes.

 Fifth, our analyses were limited to claims-based outcomes as our sample sizes for patient experience measures were inadequate to support comparisons across the 3 recognition groups. Sixth, we achieved high levels of balance by using propensity score weights, but it is possible that unobserved differences among groups could bias the results if these unobserved factors changed over time. Lastly, we examined the effects of these recognition programs among FQHCs but do not know the extent to which these findings may generalize to non—safety-net clinics. Nevertheless, our findings suggest value in additional examination of how specific components of recognition programs affect patient outcomes across settings.

Despite these limitations, this analysis is the first to compare beneficiary outcomes across recognition programs, document heterogeneity in the association among recognition programs with beneficiary outcomes, and provide insight into the reasons for these differences. We note that our analyses pertain to recognition programs at a particular point in time, and all 3 programs have evolved over time.

 Given the extensive resources required by organizations to support the development and maintenance of recognition programs, and by sites to use them, this empirical evidence may help future studies focus on how changes in recognition criteria, evaluation processes, or documentation requirements may additionally affect patient outcomes.

As practices continue to pursue MH recognition, it is important to understand how components of each recognition program influence the types of changes that sites pursue and the degree to which they translate to changes in patient outcomes. NCQA Level 3 sites were associated with improvements across a greater number of outcomes, and there were fewer improved outcomes among sites with TJC recognition over a 3-year period. AAAHC-recognized sites were associated with limited improvements, primarily due to a small sample size. Although this analysis of variation in outcomes contributes to the literature on the effectiveness of recognition programs, there is a need for additional comparative analyses of types and components of recognition programs to better understand the associations among specific program requirements and patient outcomes.

RAND Corporation (AM, JWT, RM, CMS, AK, LH, PM, KLK), RAND Corporation, Arlington, VA; David Geffen School of Medicine at UCLA (KLK), Los Angeles, CA.

 Funding was provided by CMS (contract: HHSM-500-2005-00028I, task #T0008.) The contents of this publication are solely the responsibility of the authors and do not necessarily represent the official views of HHS or any of its agencies.

 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 (JWT, CMS, AK, LH, PM, KLK); acquisition of data (AK, LH, KLK); analysis and interpretation of data (AM, JWT, RM, CMS, LH, PM, KLK); drafting of the manuscript (AM, JWT, KLK); critical revision of the manuscript for important intellectual content (AM, JWT, CMS, AK, PM, KLK); statistical analysis (JWT, RM, CMS, KLK); obtaining funding (KLK); administrative, technical, or logistic support (AM, RM, KLK); and supervision (JWT, KLK).

Articles

We examine utilization, quality, and expenditures among Medicare beneficiaries receiving care at federally qualified health centers and compare outcomes among those attributed to 1 of 3 recognition programs versus none.

Examining Differential Performance of 3 Medical Home Recognition Programs

 To compare home health utilization and clinical outcomes between Medicare beneficiaries in the fee-for-service (FFS) and Medicare Advantage (MA) programs, and to compare regional variation.

 We used the 2010 and 2011 Outcome and Assessment Information Set to identify all home health episodes begun in 2010 and to measure 7 clinical home health outcomes that are defined by CMS for public reporting.

 We modeled the probability of home health use, the duration of home health episodes, and each clinical outcome measure as a function of MA versus FFS enrollment and model-specific risk adjustors. Empirical Bayes predictions from generalized linear mixed models were aggregated by hospital referral region (HRR) to create standardized regional measures of home health utilization and mean episode duration

We identified 30,837,130 FFS and 10,594,658 MA beneficiaries (excluding those dually eligible for Medicaid). After adjusting for demographic and clinical patient characteristics, the odds of receiving home health among FFS enrollees were 1.83 times those of MA (95% CI, 1.82-1.84). Adjusted home health duration was 34% longer for FFS (95% CI, 32%-34%). Outcomes differences were small in magnitude and inconsistent across measures. Regional variations in use and duration were substantial for both FFS and MA enrollees. Within HRRs, correlations between FFS and MA utilization rates and between FFS and MA episode durations were 0.51 and 0.94, respectively.

 MA beneficiaries use less home health than their FFS counterparts, but regional factors affect utilization, independent of insurance status.

Home health utilization is a substantial source of Medicare costs; however, whether enrollment in a Medicare Advantage (MA) plan affects home heathcare use is unknown. We compared nationwide and intra-regional use of home health between beneficiaries enrolled in fee-for-service (FFS) Medicare and MA plans. Our findings were as follows:

		FFS beneficiaries used substantially more care, even after adjusting for patient and demographic characteristics.

		Functional outcomes were similar, although FFS episodes were 33% longer in duration.

		Adjusted utilization varied markedly across regions for both FFS and MA, suggesting that MA plans cannot entirely overcome non—health-related regional forces driving use.

Although more than 30% of all Medicare beneficiaries are enrolled in the Medicare Advantage (MA) program,

 studies focusing on the MA patient population are relatively rare, in large part because there are no claims to use as a data source. The capitated payment model introduces a financial incentive to limit wasteful care, whereas none exists in the fee-for-service (FFS) system. However, few published studies have directly measured and compared how much a particular service is used in the 2 systems.

One venue where care is documented for all Medicare beneficiaries is home health. All patients receiving care through CMS-certified home health agencies are required to have a comprehensive clinical assessment at the start and end of care, using the Outcome and Assessment Information Set (OASIS) instrument. For both FFS and MA beneficiaries, there is therefore a record not only that care occurred, but also of the clinical status at the start of care and of the functional outcome. Thus, home health provides a unique window through which the 2 programs can be compared.

The use of home health within Medicare is a topic ripe for review. Home health accounted for more than 5% of all FFS Medicare spending ($18.4 billion) in 2011,

 and an Institute of Medicine (IOM) report identified postacute care as the leading source of unexplained regional variation in healthcare spending via FFS.

 In a report to Congress, the Medicare Payment Advisory Commission (MedPAC) suggested that much of the variation in home health use reflected “fraud, waste, and abuse,” and noted that “the broad program standards and fragmented nature of the FFS program do not encourage effective targeting of the benefit.”

 Whether MA plans, with their incentives to avoid unnecessary care, are more effective at managing home health use is a natural next question. As a first step toward answering it, we compared home health utilization, regional variation, and clinical outcomes among beneficiaries enrolled in the MA and FFS programs.

We identified all home health episodes among Medicare beneficiaries that started in 2010 and ended by December 31, 2011. We first performed a beneficiary-level analysis, comparing the probability (after adjusting for patient characteristics) that an FFS versus an MA beneficiary would have at least 1 home health episode. For those who received home health, we also compared the total number of days enrolled and 7 clinical outcomes that are publicly reported by CMS on the “Home Health Compare” website: improvement in the 3 measures of activities of daily living; acute care hospitalizations; and improvements in pain, dyspnea, and management of oral medications.

 Each of these measures is risk-adjusted (per CMS specifications) using the results of the OASIS assessment at the start of care.

 With the exception of the hospital utilization measure, outcomes are defined by comparing end-of-care and start-of-care OASIS assessments, and are defined only for patients who remain living in the community at the end of the episode.

We first compared each study endpoint as adjusted averages across the FFS and MA populations. Then, to better understand patterns of utilization differences between FFS and MA, we compared FFS and MA beneficiaries within each of the 306 hospital referral regions (HRRs). HRRs were chosen as the geographic unit of interest because the system partitions the country into mutually exclusive regional healthcare markets, which were felt to be the best available approximation for regions serviced by home health agencies.

Using the 100% OASIS file, we grouped start-of-care and end-of-care assessments into episodes according to CMS specifications.

 OASIS episodes represent a complete cycle of care that ends only when a patient is hospitalized, dies, or is discharged to the community. An OASIS episode may encompass 1 or more 60-day billing episodes, as defined by the home health prospective payment system.

 To each OASIS episode (observation), we merged in data from the  100% denominator file and from a file containing Hierarchical Condition Category (HCC) scores (which represent the ratio of an individual’s expected cost to that of an average beneficiary).

The Medicare health insurance claim number or the combination of Social Security number and date of birth were used as unique identifiers.  We also added demographic data from the American Community Survey, using 5-year estimates for 2007 to 2011, by linking via the zip code of each beneficiary.

Because patients who were dually eligible for Medicare and Medicaid may have received home health benefits through Medicaid—which would not have been visible to us&mdash;we excluded dual-eligible beneficiaries from the analysis. All others were included.

We used logistic regression with robust standard errors to model the odds of receiving home health and the odds of each of the 7 clinical home health outcomes. We used a generalized linear model with a log link and Gaussian distribution to model the geometric mean number of days enrolled in home health (for those who received care). For models of utilization, risk adjustors included age, sex, HCC score, and 2 socioeconomic indicators from the 2007 to 2011 American Community Survey: deciles of median household income in the Zip Code Tabulation Area (ZCTA) and the proportion of households in the ZCTA with only 1 resident (among those with 1 or more residents aged ≥65), as surrogates for the beneficiary’s income and the probability that they live alone. For home health duration, we used 2 additional risk adjustors: whether the initial home health episode started after a hospitalization (postacute) and whether it followed a surgical procedure. For each clinical outcome, we used an extensive set of outcome-specific risk adjustors (derived from the baseline OASIS assessment) that is specified by CMS for the Home Health Compare program.

To generate region-specific standardized utilization rates, we used generalized linear mixed models (with logit link for the probability of receiving home health and log link for home health duration) to model home health use as described for the national analysis, but added a random intercept for HRR. We then aggregated recycled empirical Bayes predictions to create a standardized utilization rate (or mean duration) for MA and FFS, for each of the 306 HRRs.

 We presented our results graphically by plotting each region’s FFS utilization rate against the MA utilization rate, and reported the correlation coefficients for MA versus FFS across the 306 HRRs.

In keeping with recent reports by MedPAC and the IOM, we quantified regional variation by reporting the ratio of the 90th percentile versus the 10th percentile of regional utilization.

 We reported these ratios separately for the FFS and MA populations.

After excluding 8,254,838 individuals dually eligible for Medicare and Medicaid, we identified 30,837,130 FFS beneficiaries and 10,594,658 MA beneficiaries. Baseline characteristics and unadjusted home health utilization rates are summarized in 

. Among the general population, MA patients were slightly older, whereas in the population using home health, the opposite was true. HCC scores were higher for MA in both populations.

Of those enrolled in FFS, 6.3% had at least 1 home health episode; of those in MA, 3.9% did. Among FFS beneficiaries who used home health, the average number of enrolled days was 75.8; among MA beneficiaries, it was 56.3 days. Although MA patients using home health had higher HCC scores than their FFS counterparts (suggesting a higher disease burden), they had less baseline disability according to each of 6 different measures at the start-of-care OASIS assessment. Home health more commonly started after a hospitalization and/or a surgical procedure for MA patients.

FFS Versus MA: Adjusted National Estimates

. We estimated that among the general Medicare population, the odds of starting home health during 2010 were 1.82-fold higher for those enrolled in FFS versus MA, after adjusting for patient characteristics and socioeconomic indicators. Among those who did start home health in 2010, we estimated that the FFS patients were enrolled an average of 34% more days than their MA counterparts, after risk adjustment.

The relationship between FFS versus MA, in regard to clinical home health outcomes, was inconsistent, and observed differences were small in magnitude: FFS outcomes were better with regard to pain (odds ratio [OR], 1.08; 95% CI, 1.07-1.09), ambulation (OR, 1.01; 95% CI, 1.00-1.02; 

 = .03), and management of medications (OR, 1.06;  95% CI, 1.04-1.07); and MA outcomes were better with regard to bed transfer (OR, 0.98;  95% CI, 0.97-0.99), dyspnea (OR, 0.99;  95% CI, 0.98-1.00; 

 = .02), and hospital utilization (an adverse outcome) (OR, 1.16; 95% CI, 1.15-1.17). Improved bathing was not significantly related to FFS versus MA enrollment (OR, 1.00; 95% CI, 0.98-1.02). Hospital utilization, which measures whether home health ended in hospitalization rather than discharge to the community, was the outcome with the greatest difference (favoring MA). Of note, this outcome is not adjusted for duration of home health. Because FFS patients—who have longer home health episodes&mdash;spend more time at risk for hospitalization while enrolled, this relationship may be confounded.

Among the 10% of HRRs (ie, 31 of 306 HRRs) with the highest proportions of FFS beneficiaries with 1 or more home health episodes (90th-percentile FFS), we estimated that 9.4% of all FFS beneficiaries used home health, whereas among the 31 HRRs with the lowest proportion (10th-percentile FFS), only 4.2% did. The ratio of the 90th percentile/10th percentile for FFS was, therefore, 2.2. For MA patients, the 90th percentile for use of home health was 6.3%, the 10th percentile was 2.2%, and the ratio was 2.8. The higher 90th/10th percentile ratio for MA means that by this measure, regional variation in home health use was higher for MA patients than for FFS patients.

An analogous evaluation of the duration of home health episodes yielded the following results: in the 90th-percentile HRR (with regard to average episode duration), mean duration was estimated to be 92 days for FFS and 67 days for MA. The 10th percentile was 39 days for FFS and 31 days for MA. The ratios of 90th/10th percentiles were 2.4 for FFS and 2.2 for MA.

 shows the standardized proportions of FFS and MA beneficiaries receiving home health in each of the 306 HRRs. 

 shows the analogous mean (standardized) duration of home health episodes. These figures illustrate the correlation between FFS and MA rates of home health use (Figure 1), and average home health duration (Figure 2) across regions. Correlation coefficients for MA versus FFS across regions were 0.51 for the proportion using home health and 0.94 for mean duration, meaning regions with longer home health episodes among MA patients also tend to have longer episodes among FFS patients and that this relationship exists to a lesser extent with regard to the probability that an MA or FFS beneficiary in a given region will use home health.

We have shown that Medicare beneficiaries enrolled in FFS use substantially more home health than those in MA, both in terms of the proportion of beneficiaries who receive care and the duration of home health among those who do. This is true after adjusting for age, disease burden, and regional differences in household size and income. We found that despite the average home health episode duration being 34% longer for FFS, differences in clinical outcomes were small in magnitude and inconsistent. Of the 7 outcomes that we measured, FFS fared slightly better on 3, MA fared slightly better on 3 others, and there was no difference on 1.

Assertions of “fraud, waste, and abuse” in the home health industry are based on data from the FFS program and have been made in the context of a history of rapid fluctuations in home health use that have closely paralleled changes in home health agency reimbursement policy.

 In 1997, after Medicare spending for home health quadrupled over a 5-year period, cost-based reimbursement was abruptly replaced with a forerunner of today’s home health prospective payment system, whereupon spending and utilization rates were cut in half within a year. Subsequently, utilization and expenditures have again crept upward, and since 2002, total FFS home health expenditures have nearly doubled.

The MA versus FFS comparison of home health utilization across regions is of interest because MA plans, which are paid a capitated rate to cover all needed services, have a direct incentive to minimize financial costs, whereas there is no real equivalent incentive for FFS. It is unclear how financial incentives might affect an MA plan’s “propensity to provide home health.” On the one hand, MA plans might be expected to minimize the direct costs of paying home health agencies and restrict the use of home health. On the other, they might substitute home health for more costly inpatient services and be more inclined to promote its use. Whichever the predominant effect, whether MA enrollees actually use less home health than their FFS counterparts also depends on the proportion of FFS home health that is truly wasteful and the practical ability of MA plans to curb such waste. Small studies from the pre-prospective payment system era suggested that Medicare health maintenance organizations (the forerunner of today’s MA plans) used substantially less home health than their FFS counterparts at the time, and our study confirms that this is still true (albeit to a lesser degree).

Regional variation is of interest because it provides insight into the practical limitations to optimizing the use of a service: if MA plans were perfectly able to optimize health outcomes or their own financial interests, then unexplained regional variation (ie, beyond what can be explained by differences between patients) would not be expected. The strong correlation between the use of home health by MA and FFS patients within regions therefore implies that MA plans’ ability to manage utilization is limited; MA plans limit use, but cannot entirely overcome the same regional forces (eg, healthcare market, local practice styles) that drive variation across regions in the FFS population.

We found a greater correlation between the duration of home health episodes for MA versus FFS patients within a region compared with the regional correlation in the proportion of beneficiaries enrolled in MA versus FFS who receive any home health at all. One explanation might be that home health agencies have more leverage in determining duration of care than they do in determining whether beneficiaries are referred to them. If most home health agencies treat a mix of MA and FFS beneficiaries, and if they provide care in a “payer-blind” way, then this finding might be expected.

There has been little previous study of regional variation in the use of a service by MA beneficiaries. One study did compare the use of cardiovascular procedures across 32 of 306 HRRs and found regional FFS and MA rates to be correlated, albeit less so than we find here.

 An IOM report has also noted regional correlation (0.66) in total per-capita spending.

 Regional variation among the FFS population has been extensively studied, but we found more variation in FFS home health use than has been recently described for most other services or for postacute care in general.

 We noted that postacute care encompasses more than home health and that not all home health is post acute. However, a 1996 study of FFS home health use during an era prior to the prospective payment reform found a degree of regional variation that appears to be on par with what we find here (methodological differences preclude an exact comparison).

We cannot directly determine whether greater home health use by FFS beneficiaries means that they receive unnecessary care or whether MA beneficiaries are not receiving necessary care; it is likely that both are sometimes true (as would sometimes be the converse). Measurement limitations notwithstanding, our finding that MA and FFS patients have similar clinical outcomes suggests that the shorter MA episodes may not be detrimental to health. We do not address patient preferences, however; nor do we explore the clinical effect of differing rates of home health use (compared with differing episode durations).

Our study has several other limitations. First, because MA plans have a financial incentive to select for healthier patients in ways that are not captured by CMS payment models,

 it is possible that unobservable differences between the populations remain after risk adjustment. Although this could account for some of the difference that we see in a beneficiary’s probability of receiving home health, we would expect less influence on our estimates of home health duration or outcomes, which are conditional on receiving home health and which use a larger set of observables (from OASIS) for risk adjustment. Second, functional outcome measures rely on clinical assessments at the start and end of each home health episode. Because each agency’s performance on these measures is publicly reported, there may be an incentive for agencies to underestimate functional status at the start of care or overestimate it at the end of care, biasing the results toward better outcomes, which would have the effect of minimizing any true outcomes differences between the FFS and MA populations. Moreover, since the end-of-care assessment does not take place when patients die or are transferred, outcomes are only measured for patients who are discharged back to the community at the end of home health. Finally, we do not have a direct window onto the actual number of home health visits that a patient receives, and rely instead on the duration of a home health episode as a proxy. We cannot be certain that FFS and MA beneficiaries receive the same number of visits per unit of time enrolled in home health.

MedPAC has reported to Congress that the FFS system is not structured to limit wasteful use of the home health benefit.

 Our study uses MA as a counterexample of care under an alternative reimbursement structure. The Government Accountability Office cites the lack of standards for what constitutes appropriate and necessary care as one of the challenges to curbing excess. Although we do not offer such a standard, we suggest that the type of within-region, between-payer comparison performed here can help to better define a reasonable range of use and to identify high-use and low-use regional outliers (for both FFS and MA). This methodology may be more widely applicable using MA encounter data, which are still being collected.

The authors acknowledge Patty Smith for editorial assistance, Cheryl Damberg and Rodger Madison for assistance with data access, and Emmett Keeler and Jeffrey Wasserman for comments on a previous version of this paper. 

Department of Emergency Medicine (DAW) and David Geffen School of Medicine (NSW, DAG), University of California, Los Angeles, Los Angeles, CA; RAND Corporation (DAW, LM, CMS, MH, NSW, DAG), Santa Monica, CA; University of Michigan Medical School (LM), Ann Arbor, MI; VA Ann Arbor Healthcare System (LM), Ann Arbor, MI; VA Greater Los Angeles Healthcare System (DAG), Los Angeles, CA.

The analyses upon which this article is based were performed under contract HHSM-500-2005-00028I, task order number HHSM-500-T0004, entitled “Evaluation and Development of Outcome Measures for Quality Assessment in MAOs and SNPs,” funded by CMS and HHS. The content of this article does not necessarily reflect the views or policies of HHS, nor does the mention of trade names, commercial products, or organizations imply endorsement by the US government. The author assumes full responsibility for the accuracy and completeness of the ideas presented.

Dr Waxman was supported by the Department of Veterans Affairs (VA) Office of Academic Affiliations through the VA Health Services Research and Development Advanced Fellowship Program; he had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Dr Ganz was part of the project team on contracts awarded to RAND from CMS and the Health Services Advisory Group, and was part of the project team on a contract awarded to UCLA from SCAN Health Plan. The remaining 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 (DAG, MH, LM, DAW); acquisition of data (DAW); analysis and interpretation of data (DAG, MH, LM, CMS, DAW); drafting of the manuscript (LM, CMS, DAW); critical revision of the manuscript for important intellectual content (DAG, LM, DAW); statistical analysis (CMS, DAW); obtaining funding (DAG, DAW); administrative, technical, or logistic support (MH); and supervision (DAW). 

Daniel A. Waxman, MD, PhD, RAND Corporation, 1776 Main St, Santa Monica, CA 90407. E-mail: dwaxman@rand.org.

1. Medicare Advantage. Kaiser Family Foundation website. http://kff.org/medicare/fact-sheet/medicare-advantage/. Published May 11, 2016. Accessed February 13, 2016.

Medicare Advantage beneficiaries use less home healthcare than do their fee-for-service counterparts, but there is marked regional variation in use by both groups.

Does Medicare Advantage Enrollment Affect Home Healthcare Use?

To compare antibiotic prescribing among retail clinics, primary care practices, and emergency departments (EDs) for acute respiratory infections (ARIs): antibiotics-may-be-appropriate ARIs (eg, sinusitis) and antibiotics-never-appropriate ARIs (eg, acute bronchitis).

 We analyzed retail clinic data from the electronic health records of the 3 largest retail clinic chains in the United States, and data on visits to primary care practices and EDs from the nationally representative National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey.

 Using multivariate models, we estimated an adjusted antibiotic prescribing rate for each site of care, controlling for differences in patient characteristics and diagnosis.

From 2007 to 2009 in the United States, there were 3 million, 167 million, and 29 million ARI visits at retail clinics, primary care practices, and EDs, respectively. For all ARI visits, the adjusted antibiotic prescribing rate at retail clinics (58%) was similar to the rate at primary care practices (62%; 

= .48). For antibiotics-may-be-appropriate ARI visits, the adjusted antibiotic prescribing rate (95%) at retail clinics was higher than at primary care practices (85%; 

<.01). For antibiotics-never-appropriate ARI visits, the adjusted antibiotic prescribing rate (34%) at retail clinics was lower than at primary care practices (51%; 

Compared with primary care practices and EDs, there was no difference at retail clinics in overall ARI antibiotic prescribing. At retail clinics, antibiotic prescribing was more diagnosis-appropriate.

There is concern that retail clinics could increase inappropriate antibiotic prescribing because they have financial incentives to overprescribe medications and they employ nurse practitioners. In the first evaluation of this question using medical record data, we found that the overall antibiotic prescribing rate was similar among retail clinics, physician offices, and emergency departments; furthermore, care at retail clinics was more guideline-concordant. For example, in the case of conditions for which antibiotics are never indicated, retail clinics had a lower antibiotic prescribing rate.

		There are concerns that retail clinics provide an inferior quality of care, but our results do not support these concerns.

		No difference was found in quality of antibiotic prescribing among retail clinics, physician offices, and emergency departments.

Acute respiratory infections (ARIs) such as pharyngitis, rhinosinusitis, and bronchitis are the most common symptomatic reasons why patients seek care in the United States.

ARIs account for half of all adult antibiotic prescriptions and three-fourths of all pediatric antibiotic prescriptions.

Physicians overprescribe antibiotics for ARIs and unnecessarily use broad-spectrum antibiotics, both of which lead to increased antibiotic resistance, adverse drug reactions, and increased cost.

A growing fraction of ARI visits occur at retail clinics, which are located in drugstores and grocery stores and are typically staffed by nurse practitioners (NPs) as opposed to physicians. Retail clinics provide walk-in care for a limited set of conditions for which they have treatment protocols, including ARIs, other minor illnesses, immunizations, sports physicals, and other screening services—ARIs account for 60% of all retail clinic visits.

 Providers at retail clinics independently manage and prescribe medications. State regulations vary on whether NPs at retail clinics must maintain a collaborative or supervisory relationship with a physician.

Currently, there are almost 6 million yearly visits to the 1600 retail clinics in the United States,

and the number of retail clinics is expected to grow rapidly in the coming years.

The impact of this shift to retail clinics in the care of ARIs, and subsequent antibiotic prescribing, is unknown. It is possible this shift may worsen existing problems with ARI antibiotic prescribing patterns, as most retail clinics are owned by for-profit pharmacy chains that have a financial incentive to provide prescription medicines.

Alternatively, it is possible this shift will improve antibiotic prescribing, as NPs often provide care that is more consistent with guidelines

 and retail clinics strictly incorporate evidence-based guidelines into their electronic health records (EHRs).

In prior studies comparing rates of antibiotic prescribing, little difference was observed between retail clinics and outpatient physician practices

; however, these were small studies that used a limited set of diagnoses and health plan claims data. In this study, we compared antibiotic prescribing at retail clinics, primary care practices, and emergency departments (EDs) for all ARIs using national data collected directly from the EHR or the medical chart.

We obtained de-identified data from all visits from 2007 to 2009 (2007, 0.8 million visits; 2008, 2.4 million; 2009, 5.4 million) from 3 retail clinic operators: MinuteClinic, TakeCare, and LittleClinic. Together they operate 81% of all US retail clinics.

These data came directly from the operators’ EHRs.

We obtained 2007 to 2009 data (to match the years of available retail clinic data) on visits to ambulatory practices from the National Ambulatory Medical Care Survey (NAMCS) and outpatient hospital practices from the National Hospital Ambulatory Medical Care Survey (NHAMCS). NAMCS and NHAMCS employ a complex sampling methodology to obtain a nationally representative estimate of outpatient visits in the United States, and the 2 surveys have been used in many studies to characterize physician care patterns in the United States.

 However, neither survey includes retail clinics. For each visit, standard 1-page encounter forms are completed by the clinician, office staff, or an outside coder using the medical chart.

We defined primary care visits in NAMCS as those associated with the physician specialties of family/general practice, internal medicine, or pediatrics, and in NHAMCS as visits to general medicine and pediatric clinics. We excluded the small number of visits with a nurse practitioner because we did not want to confound training and practice setting; we included these visits in a sensitivity analysis, and it had no impact on our findings. To exclude follow-up visits for ARIs, we limited our sample to visits where the treating provider indicated that the major reason for the visit was a new problem. After these exclusions, we sampled 6289 ARI visits. We also conducted a sensitivity analysis in which we examined all ARI visits, including those where it was not a new problem, and we found no substantive differences.

We obtained data on 7629 sampled ED visits for ARIs from 2007 to 2009 from the ED component of the NHAMCS survey. Similar to the physician outpatient visits in NAMCS, a standard encounter form is completed for each visit using the medical chart.

Across the 3 data sources, we looked at patient demographics, reasons for visit, diagnoses, and medications prescribed. We did not include patient temperature because for almost half of ARI visits at primary care practices, the temperature field was blank. Patient drug allergies were not available.

International Classification of Diseases, Ninth Revision, Clinical Modification

we divided the 6 ARI diagnoses into “antibiotics-may-be-appropriate” diagnoses (streptococcal pharyngitis [034.x], otitis media [381.x, 382.x], sinusitis [461.x]) and “antibiotics-never-appropriate” diagnoses (upper respiratory infection [460.x, 465.x], nonstreptococcal pharyngitis and laryngitis [462.x, 464.x], bronchitis [466.x, 490.x, 491.21]). These definitions of antibiotic-appropriateness are forgiving to prescribing of antibiotics. For example, although we consider sinusitis an antibiotics-may-be-appropriate diagnosis, it is not clear that sinusitis ever requires antibiotic treatment.

Excluding Visits With Competing Diagnoses

Patients may have multiple problems that require antibiotics, and the fraction of ARI patients with such competing diagnoses might differ across the 3 care sites. To address this potential concern, we excluded visits with a diagnosis code for a competing diagnosis that might require antibiotics (list of 

). This excluded 3%, 6%, and 10% of the ARI visits at retail clinics, physician offices, and EDs, respectively. In a sensitivity analysis, including such visits had no substantive impact on the results.

Among all eligible ARI visits, we identified any oral antibiotic prescription associated with the visit. In the retail clinic, data prescriptions are identified using the National Drug Code system, while in the NAMCS and NHAMCS surveys, prescriptions are classified using the Multum Lexicon system. We defined broad-spectrum antibiotics as azithromycin and clarithromycin, quinolones, amoxicillin- clavulanate, ampicillin-sulbactam, and second- and third-generation cephalosporins.

We calculated the antibiotic prescribing rate for all ARI visits, for antibiotics-may-be-appropriate diagnoses, for antibiotics-never-appropriate diagnoses, and for each individual ARI diagnosis. We limited our analysis to patients 2 years and older because retail clinics rarely see children under that age.

We used multivariable models to assess the independent association between site of care and antibiotic prescribing, where the outcome variable is a binary (yes/no) indicator for whether an antibiotic was prescribed. Building on prior work,

we combined the data from the 3 clinical sites into a single data set. In the physician office and ED data, each visit is given a sampling weight so that one can calculate nationally representative estimates. Since retail clinics represent a full national sample of visits, in the analyses, these visits were given a sampling weight of 1. The NAMCS and NHAMCS samples include identifiers allowing for control for clustering within individual clinicians and practices and EDs. For the retail clinic data, we could not identify individual clinicians and practice sites associated with a visit; we therefore accounted for a higher level of clustering within each of 3 retail clinic chains. Because it assumes that all retail clinic visits within a given retail clinic chain are similar, the intercluster correlation was large (0.02) and therefore standard errors were conservatively large.

We used separate models for all ARI visits, antibiotics-may-be-appropriate ARI diagnosis visits, and antibiotics-never-appropriate ARI diagnosis visits. Among all ARI visits where an antibiotic was prescribed, we created an analogous logistic regression model where the dependent variable was broad-spectrum antibiotic prescription. In each of the multivariate models we adjusted for gender, age, site of care, and whether the patient had a chronic illness. In the model where we focused on all ARI visits, we also adjusted for diagnosis type (antibiotics-may-be-appropriate diagnoses vs antibiotics-never-appropriate). We conducted sensitivity analyses where we: 1) did not adjust for diagnosis; 2) adjusted for the 6 specific ARI diagnoses instead of diagnosis type to provide a higher level of specificity; 3) examined only children (aged ≤17 years) as this is an important subpopulation; 4) examined only adults; and 5) examined only patients without a reported chronic illness, as this is the population most likely to be seen at retail clinics. The odds ratio results from all multivariate models are reported in the eAppendix.

To facilitate interpretation of the regression results, we used the method of predictive margins (also called recycled predictions) to report the predicted antibiotic prescribing rate and the predicted broad-spectrum antibiotic prescribing rate for each of the 3 care sites while adjusting for other covariates.

 Specifically, we used the estimated coefficients from the multivariate regression models to predict antibiotic prescribing for each patient visit, alternatively assigning all visits to a given value of the variable (eg, retail clinic, primary care practice, ED), but leaving all other explanatory variables at their original values. We then averaged the predictions across all patient visits. This allowed us to estimate the mean antibiotic prescribing rate if all patient visits occurred at a given site, but otherwise retained the original values of all their other characteristics. 

values less than .05 were considered significant. All analyses were performed using SAS 9.2 (Cary, North Carolina).

Between 2007 and 2009, there were 3 million, 162 million, and 29 million ARI visits, respectively, at retail clinics, primary care practices, and EDs. There were several notable differences in patient characteristics across the 3 care sites (

). Children and adolescents (aged ≤17 years) made up a smaller fraction of ARI visits to retail clinics (27%) than to primary care practices (50%; 

<.01). A smaller fraction of patients reported a chronic illness at retail clinics (4%) compared with primary care practices (20%; 

The mix of ARI visits varied by site of care. At retail clinics, a higher proportion of ARI visits (53%) led to an antibiotics-may-be-appropriate diagnosis (sinusitis, otitis media, and streptococcal pharyngitis) than at primary care physician offices (32%; 

Unadjusted Differences in ARI Antibiotic Prescribing

Not adjusting for other factors, the fraction of all ARI visits at which an antibiotic was prescribed was similar at retail clinics (66%), primary care practices (62%; 

= .28 comparison with retail clinics), and EDs (60%; 

= .11) (Table 2). This pattern varied by antibiotic appropriateness. For antibiotics-may-be-appropriate diagnoses, retail clinic providers were more likely to prescribe antibiotics (95%) than primary care physicians (85%; 

<.01). In contrast, for antibiotics-never-appropriate diagnoses, retail clinics providers were less likely to prescribe antibiotics (33%) than primary care physicians (50%; 

<.01). One notable exception among antibiotics-never-appropriate diagnoses was for acute bronchitis/bronchiolitis where antibiotic prescribing at retail clinics, physicians, and EDs was 77%, 76%, and 70%, respectively. Among all ARI visits where an antibiotic was prescribed, retail clinic providers were less likely to prescribe a broad-spectrum antibiotic than primary care practices (40% vs 49%; 

<.01), but there was no difference between retail clinics and EDs (40% vs 44%; 

Adjusted Differences in ARI Antibiotic Prescribing

The adjusted antibiotic prescribing rate for all ARI visits at retail clinics (58%) was similar to the adjusted antibiotic prescribing rate at primary care practices (62%; 

). For antibiotics-may-be-appropriate ARI visits, the adjusted antibiotic prescribing rate for retail clinics (95%) was higher than at primary care practices (86%; 

<.01). For antibiotics-never-appropriate ARI visits, the adjusted antibiotic prescribing rate at retail clinics (34%) was lower than at primary care practices (50%; 

 <.01). For ARI visits at which an antibiotic was prescribed, the adjusted rate of broad-spectrum antibiotic prescribing at retail clinics was 42%, lower than the rate at primary care practices (49%; 

There are concerns that retail clinics provide inferior quality of care. The authors found no difference in quality among retail clinics, physician offices, and emergency departments.

Claude M. Setodji, PhD

Articles

Examining Differential Performance of 3 Medical Home Recognition Programs

Does Medicare Advantage Enrollment Affect Home Healthcare Use?

Antibiotic Prescribing for Respiratory Infections at Retail Clinics, Physician Practices, and Emergency Departments

Trends in Retail Clinic Use Among the Commercially Insured