Abstract
Context.
Published evaluations of medical home interventions have found limited effects on quality and utilization of care.
Objective.
To measure associations between participation in the Northeastern Pennsylvania Chronic Care Initiative and changes in quality and utilization of care.
Design, Setting, and Participants.
We analyzed medical claims for 17,363 patients attributed to 27 pilot and 29 comparison practices, using difference-in-difference methods to estimate changes in quality and utilization of care associated with pilot participation.
Exposure.
The intervention included learning collaboratives, disease registries, practice coaching, payments to support care manager salaries and practice transformation, and shared savings incentives (bonuses of up to 50% of any savings generated, contingent on meeting quality targets). As a condition of participation, pilot practices were required to attain recognition by the National Committee for Quality Assurance (NCQA) as medical homes.
Main Outcome Measures.
Performance on 6 quality measures for diabetes and preventive care; utilization of hospital, emergency department (ED), and ambulatory care.
Results.
All pilot practices received recognition as medical homes during the intervention. By intervention year 3, relative to comparison practices, pilot practices had statistically significantly better performance on 4 process measures of diabetes care and breast cancer screening; lower rates of all-cause hospitalization (8.5 vs. 10.2 per 1000 patients per month; difference −1.7; 95% CI −3.2 to −0.03), lower rates of all-cause ED visits (29.5 vs. 34.2; difference −4.7; −8.7 to −0.9), lower rates of ambulatory care-sensitive ED visits (16.2 vs. 19.4; difference −3.2; −5.7 to −0.9), lower rates of ambulatory visits to specialists (104.9 vs. 122.2; difference −17.3; −26.6 to −8.0); and higher rates of ambulatory primary care visits (349.0 vs. 271.5; difference 77.5; 37.3 to 120.5).
Conclusions.
Over a 3 year period, this medical home intervention, which included shared savings for participating practices, was associated with improvements in quality, increased primary care utilization, and lower use of ED, hospital, and specialty care. With further experimentation and evaluation, such interventions may continue to become more effective.
Background
The medical home concept, which encompasses a diverse set of primary care practice models intended to achieve high quality and efficiency of care, has gained wide support.1,2 Pilot interventions that encourage primary care practices to receive recognition as medical homes generally feature new resources (such as technical assistance and per-patient per-month fees to support practice transformation) and, more recently, new payment incentives such as shared savings.3,4
Systematic reviews and studies of medical home interventions have reported mixed effects on quality of care and little evidence of reductions in utilization or costs among community-based primary care practices.5–10 However, with few exceptions,11 these studies have evaluated interventions that lacked financial incentives for practices to control the utilization or costs of care. Moreover, conveners of early medical home pilots have modified their approaches over time, potentially enhancing the effectiveness of their interventions.
We evaluated the northeast region of the Pennsylvania Chronic Care Initiative (PACCI), a medical home intervention that was led by experienced conveners and that featured shared savings for participating practices, thereby creating direct financial incentives to reduce costs and utilization of care. We hypothesized that this intervention would be associated with improvements in quality and efficiency over a 3 year period.
Design of the Northeast PACCI
As detailed elsewhere,7,12,13 a broad coalition of payers, clinicians, delivery system, and government stakeholders formulated the PACCI as a series of regional medical home pilot interventions. The northeast region intervention began in October 2009, included 2 commercial health plans and 29 volunteering small primary care practice sites, and was designed to run for 36 months. The state selected these 29 practices from among all volunteering practices according to criteria available in Appendix A. As in other PACCI regions, the northeast intervention targeted diabetes care improvement among adult patients and included learning collaboratives, web-based registries to generate quality reports, and practice coaching to facilitate practice transformation.7 However, the intervention did not target diabetic patients exclusively; only 3 of the 14 quality benchmarks for participating practices [details in Appendix B] focused on diabetes.
Participating practices were required to obtain National Committee for Quality Assurance (NCQA) Physician-Practice Connections-Patient Centered Medical Home (PPC-PCMH) as medical homes, by intervention month 18, at “Level 1 Plus” or greater based on 2008 PPC-PCMH criteria [details in Appendix B]. Practices also received $1.50 per patient per month in “Care Management Payments,” which were earmarked for dedicated care manager salaries, and $1.50 per patient per month in “Practice Support Payments,” which could be used to support other costs of practice transformation.
Unlike most previous medical home interventions, practices participating in the northeast PACCI were eligible to receive shared savings bonuses, contingent on meeting quality benchmarks, if total spending on their patients was less than expected in a given year (i.e., if savings were observed). Each health plan could decide its own savings calculation method, and the bonus payments could range from 40 to 50 percent of calculated savings in each year [details in Appendix B]. Practices faced no financial penalties if observed spending was greater than expected. Participating health plans also provided each participating practice with semi-annual feedback on hospital and ED utilization.
On January 1, 2012 (27 months after the pilot began), the northeast PACCI joined the Medicare Advanced Primary Care Practice Demonstration, and Medicare became a participating payer. No other intervention components changed substantially during the first 3 years of this pilot.
The northeast PACCI intervention had multiple design features that were not present in the southeast PACCI intervention, which began 16 months earlier.7 Unlike the southeast PACCI, the northeast PACCI included shared savings incentives and provided utilization data to participating practices (including lists of each practice’s patients who had recent ED visits or hospitalizations). Also, the northeast PACCI placed less emphasis on early NCQA medical home recognition than the southeast PACCI, where monthly per-patient bonuses (which were larger for higher levels of NCQA recognition) began as soon as recognition was received.
Methods
We used a difference-in-differences design to compare changes over a 3 year period in the quality and utilization of care for patients attributed to practices that participated in the northeast PACCI and comparison practices that did not participate in this medical home intervention.
Comparison practices
Based on lists of practices provided by the two participating health plans, a state contractor selected 29 comparison practices in northeast Pennsylvania that had the same approximate composition as the pilot practices on practice size and specialty (family practice, internal medicine). The comparison practices were selected after the pilot practices were identified, but before the pilot intervention began. Data on quality and utilization of care were not used to select comparison practices.
Survey of practices
As detailed elsewhere, we developed a survey instrument to measure practices’ structural capabilities, including use of disease management, registries, and electronic health records (EHRs) [instrument available from authors upon request].7,14 We mailed this survey to one leader of each participating and comparison practice in September 2010, querying baseline capabilities that were present in September 2009, and we mailed a second survey in October 2012 to assess capabilities after the third year of the pilot.
Measures of quality and utilization
Both participating commercial plans supplied medical claims and enrollment data spanning October 1, 2007 to September 30, 2012 (2 years prior to and 3 years following the pilot inception date) for their members who, at any time during this 5-year period, had 1 or more medical claims for any service with a pilot or comparison practice.
To facilitate comparison of our evaluation with evaluations of other medical home interventions, we calculated claims-based performance measures following recommendations from Commonwealth Fund’s medical home evaluators’ collaborative.15,16 These included NCQA Healthcare Effectiveness and Data Information Set (HEDIS) process measures of quality, modified to account for duration of observation in some instances; rates of hospitalization (all-cause and ambulatory care-sensitive); ED visits (all-cause and ambulatory care-sensitive); and ambulatory visits. Appendix C presents measure specifications.
Patient attribution
Based on the qualifying services detailed in Appendix D, we attributed patients to the primary care clinicians who provided the plurality of qualifying services in each of 4 time periods (the pre-intervention period and intervention years 1, 2, and 3), with the most recent service breaking ties.7 In sensitivity analyses, we re-attributed patients based on the majority (>50%) of qualifying services.
Analyses
We compared the pre-intervention characteristics and patient populations of pilot and comparison practices using Wilcoxon-rank-sum and Fisher exact tests. We compared practices’ baseline and post-intervention possession of structural capabilities using Liddell exact tests.
It is possible that the intervention could affect patients’ likelihood of leaving or staying with their primary care practices. We tested for this possibility by fitting a linear probability model with patient retention (i.e., whether a given patient attributed to a practice in the pre-intervention period remained so attributed during each intervention year) as the dependent variable and pilot participation interacted with intervention year as the independent variables. We reasoned that if there was evidence of selection (manifesting as differential patient retention during the intervention), patients should be assigned to practices based on pre-intervention attribution only (i.e., an intent-to-treat approach). In the presence of selection, patient assignment using sequential cross-sectional attribution (i.e., re-attributing patients annually) could lead to biased estimates of intervention effects.
Among continuously enrolled patients attributed to a study practice at baseline, the observed rate of patient retention (i.e., same-practice attribution in pilot year 3) was 57.2% among pilot practices and 50.2% among control practices (difference 7.0%; 95% C.I. −1.2% to 15.2%; P for difference 0.094). Due to this difference in patient retention rates, we performed an intent-to-treat analysis using pre-intervention attribution of patients to practices.
We evaluated associations between practice exposure to the northeast PACCI pilot and changes in quality measure performance by fitting linear probability models with “average treatment effect on the treated” (ATT) propensity weights to balance pilot and comparison practices’ baseline shares of patients from each health plan and performance on each measure.17 For each quality measure, the dependent variable was receipt of the indicated service, and independent variables were indicators for time period (pre-intervention and each intervention year), interactions between time period and practice participation in the pilot, indicators for the health plan contributing each observation and patient enrollment in an HMO, and fixed effects (dummy variables) for each practice.
For measures of utilization, we fit 2-part logistic and negative binomial models, using propensity weights to balance practices’ shares of patients from each health plan and baseline utilization rates. The dependent variables were utilization counts in each time period, and independent variables were indicators for time period, interaction between time period and pilot/comparison status, indicators for the health plan contributing each observation and patient enrollment in an HMO, patient age, gender, and pre-intervention Charlson comorbidity score,18 and practice fixed effects.
In all models, we used generalized estimating equations with robust standard errors to account for practice-level clustering.19,20 Because empirical standard error estimates can be sensitive to missing data, we included continuously enrolled health plan members in the regression models. To display adjusted data from non-linear regressions on their original measurement scales, we generated recycled predictions and used practice-level bootstrapping (1000 resamples) to generate single confidence intervals from 2 part models.21,22 To generate single P values for display purposes, we fit one-part negative binomial models; P values from each part of the 2 part models are available in Appendix E.
In sensitivity analyses, we substituted logistic for linear probability models and included patients who lacked continuous health plan enrollment. We considered two-tailed P values <0.05 significant and conducted data management and analyses using SAS 9.2 (SAS Institute Inc., Cary, NC) and SQL Server 2008 (Microsoft, Redmond, WA).
This study was approved by the RAND Human Subjects Protection Committee.
Results
Of the 29 practices that volunteered to participate in the pilot, 1 withdrew before the intervention began, and 1 withdrew during the first intervention year. The remaining 27 practices completed the 3-year intervention as planned and are included for analysis. The pilot and comparison practices were similar in baseline size, specialty, and patient case-mix [Table 1].
Table 1.
Baseline characteristics of pilot and comparison practices.
| Pilot (27 practices; 17921 attributed patients) | Comparison (29 practices; 12894 attributed patients) | P value* | |
|---|---|---|---|
| Practice size | Median (IQR) | ||
| Number of PCPs | 3 (1, 7) | 3 (1, 4) | 0.40 |
| Main practice specialty | Number of practices | ||
| Family practice | 10 | 12 | 0.39 |
| Internal medicine | 8 | 12 | |
| Mixed specialty** | 9 | 5 | |
| Patient panel characteristics*** | Median (IQR) | ||
| Number of attributed patients per practice | 400 (213, 1002) | 290 (170, 681) | 0.29 |
| % female | 52 (49, 57) | 54 (50, 61) | 0.37 |
| Median age among adults | 46 (43, 50) | 45 (42, 50) | 0.98 |
| % with age<17 | 10 (3, 20) | 6 (2,21) | 0.74 |
| % diabetic | 9 (7, 11) | 8 (6, 9) | 0.080 |
| Mean Charlson score | 1.25 (0.99, 1.80) | 1.24 (1, 1.69) | 0.77 |
| % health plan A (commercial) | 22 (0, 96) | 33 (0, 42) | 0.78 |
| % health plan B (commercial) | 78 (4, 100) | 67 (57, 100) | 0.78 |
Abbreviation: IQR, interquartile range; PCP, primary care physician or clinician (including MDs, DOs, and NPs).
Wilcoxon rank-sum test for counts, percentages, and other continuous variables; Pearson chi-square and Fisher exact tests for categorical variables. All variables are calculated at the practice level.
Any combination of family practice, internal medicine, and pediatrics.
Calculated at the practice level among patients attributed to study practices in the pre-intervention period based on plurality of qualifying visits.
Each pilot practice received NCQA PPC-PCMH recognition during the intervention: 23 at level 3, 2 at level 2, and 2 at level 1. Twelve pilot practices were recognized under the 2008 criteria and 15 under the 2011 criteria. Two of the comparison practices received NCQA recognition during the pilot.
Twenty-three pilot practices (85%) responded to both the baseline and year 3 structural surveys, but only six (21%) comparison practices did so, precluding analysis of their responses. Pilot practices adopted capabilities in performance feedback, registry use, care management, patient outreach, and electronic test ordering [detailed results in Appendix F]. All responding pilot practices had EHRs at baseline.
Pilot participation was statistically significantly associated with higher performance on all 4 examined measures of diabetes care quality and breast cancer screening but not colorectal cancer screening [Table 2]. These associations emerged in intervention year 1 for each of these measures, except LDL-C testing in patients with diabetes, for which performance was statistically significantly greater in intervention year 3 only.
Table 2.
Propensity-weighted, adjusted quality of care differences between pilot and comparison practices among continuously enrolled patients.
| Pilot | Comparison | Difference (95% CI) | P value | |
|---|---|---|---|---|
| Diabetes: HbA1c testing* | %** | |||
| Pre-intervention | 93.5 | 93.5 | NA*** | NA |
| Intervention year 1 | 94.4 | 88.2 | 6.2 (1.9, 10.5) | 0.005 |
| Intervention year 2 | 95.0 | 90.7 | 4.2 (−0.1, 9.2) | 0.092 |
| Intervention year 3 | 92.1 | 83.9 | 8.3 (2.3, 14.2) | 0.007 |
| Diabetes: LDL-C testing | ||||
| Pre-intervention | 90.5 | 90.5 | NA | NA |
| Intervention year 1 | 90.8 | 86.5 | 4.3 (−2.1, 10.6) | 0.188 |
| Intervention year 2 | 91.8 | 86.3 | 5.5 (−0.9, 11.9) | 0.093 |
| Intervention year 3 | 88.1 | 79.6 | 8.5 (0.3, 16.7) | 0.043 |
| Diabetes: Nephropathy monitoring | ||||
| Pre-intervention | 78.0 | 78.0 | NA | NA |
| Intervention year 1 | 87.7 | 66.2 | 21.5 (14.4, 29.6) | <0.001 |
| Intervention year 2 | 87.2 | 71.4 | 15.8 (7.0, 24.6) | 0.001 |
| Intervention year 3 | 85.6 | 70.2 | 15.5 (5.4, 25.5) | 0.003 |
| Diabetes: Eye exams | ||||
| Pre-intervention | 55.2 | 55.2 | NA | NA |
| Intervention year 1 | 58.0 | 42.5 | 15.5 (7.4, 23.6) | <0.001 |
| Intervention year 2 | 54.5 | 44.9 | 9.7 (−0.5, 19.8) | 0.061 |
| Intervention year 3 | 51.2 | 39.2 | 12.0 (2.8, 21.1) | 0.011 |
| Breast cancer screening | ||||
| Pre-intervention | 81.4 | 81.4 | NA | NA |
| Intervention year 1 | 83.0 | 78.9 | 4.1 (2.0, 6.2) | <0.001 |
| Intervention year 2 | 82.7 | 75.9 | 6.8 (4.5, 9.0) | <0.001 |
| Intervention year 3 | 80.5 | 74.9 | 5.6 (2.9, 8.3) | <0.001 |
| Colorectal cancer screening | ||||
| Pre-intervention | 37.1 | 37.1 | NA | NA |
| Intervention year 1 | 43.4 | 42.9 | 0.5 (−1.3, 2.3) | 0.57 |
| Intervention year 2 | 47.4 | 47.3 | 0.1 (−2.3, 2.5) | 0.95 |
| Intervention year 3 | 50.0 | 51.1 | −1.1 (−4.0, 1.7) | 0.44 |
Abbreviations: NA, Not Applicable; CI, confidence interval.
Measure denominators are 674 pilot and 258 comparison patients for diabetes measures, 2316 pilot and 1714 comparison patients for breast cancer screening, and 4072 pilot and 2410 comparison patients for colorectal cancer screening.
Percentage point estimates are propensity-weighted recycled predictions from linear probability models adjusting for practice baseline score, health plan contributing each observation, and whether each patient was in an HMO product at the time of the observation. Differences, confidence intervals, and p-values correspond to marginal differences from the linear probability models.
Due to the inclusion of fixed effects for practices, regression models do not estimate pre-intervention differences between pilot and comparison.
By year 3, pilot participation was statistically significantly associated with lower rates of all-cause hospitalization per 1000 patients per month (−1.7; 95% CI −3.2 to −0.03), all-cause ED visits (−4.7; −8.7 to −0.9), ambulatory care-sensitive ED visits (−3.2; −5.7 to −0.9), and ambulatory visits to specialists (−17.3; −26.6 to −8.0) and higher rates of ambulatory primary care visits (77.5; 37.3 to 120.5) [Table 3]. For all-cause hospitalizations, statistically significant differences between pilot and comparison practices emerged in year 2. For all-cause and ambulatory care-sensitive ED visits, statistically significant differences between pilot and comparison practices were present in year 3 only. Rates of ambulatory care-sensitive hospitalization also were lower among pilot practices, but this difference was not statistically significant.
Table 3.
Propensity-weighted, adjusted differences in utilization of care between pilot and comparison practices among continuously enrolled patients (n=10548 pilot and n=6815 comparison patients).
| Pilot | Comparison | Difference (95% CI) | P value | |
|---|---|---|---|---|
| Hospitalizations, all-cause | Rate per 1000 patients per month (95% CI)* | |||
| Pre-intervention | 7.0 | 7.0 | NA** | NA |
| Intervention year 1 | 7.3 | 8.8 | −1.5 (−3.1, 0.2) | 0.069 |
| Intervention year 2 | 7.4 | 9.2 | −1.8 (−3.3, −0.2) | 0.001 |
| Intervention year 3 | 8.5 | 10.2 | −1.7 (−3.2, −0.03) | 0.006 |
| Hospitalizations, ambulatory care-sensitive | ||||
| Pre-intervention | 0.5 | 0.5 | NA | NA |
| Intervention year 1 | 0.5 | 0.7 | −0.2 (−0.6, 0.2) | 0.21 |
| Intervention year 2 | 0.8 | 0.9 | −0.1 (−0.5, 0.4) | 0.67 |
| Intervention year 3 | 0.8 | 1.0 | −0.2 (−0.6, 0.3) | 0.52 |
| ED visits, all-cause | ||||
| Pre-intervention | 23.9 | 23.9 | NA | NA |
| Intervention year 1 | 24.5 | 27.5 | −3.0 (−6.6, 0.5) | 0.090 |
| Intervention year 2 | 26.3 | 28.4 | −2.1 (−5.6, 1.1) | 0.29 |
| Intervention year 3 | 29.5 | 34.2 | −4.7 (−8.7, −0.9) | 0.001 |
| ED visits, ambulatory care-sensitive | ||||
| Pre-intervention | 13.5 | 13.5 | NA | NA |
| Intervention year 1 | 13.5 | 15.0 | −1.4 (−3.8, 1.1) | 0.11 |
| Intervention year 2 | 14.5 | 15.6 | −1.2 (−3.8, 1.3) | 0.16 |
| Intervention year 3 | 16.2 | 19.4 | −3.2 (−5.7, −0.9) | <0.001 |
| Pre-intervention | 379.6 | 379.6 | NA** | NA |
| Intervention year 1 | 357.0 | 304.2 | 52.8 (9.1, 99.4) | 0.024 |
| Intervention year 2 | 357.1 | 250.0 | 107 (51.1, 178.5) | 0.002 |
| Intervention year 3 | 349.0 | 271.5 | 77.5 (37.3, 120.5) | 0.001 |
| Ambulatory visits, specialist | ||||
| Pre-intervention | 106.2 | 106.2 | NA | NA |
| Intervention year 1 | 108.7 | 117.3 | −8.7 (−16.2, −1.2) | 0.01 |
| Intervention year 2 | 104.8 | 121.3 | −16.5 (−27.5, −5.9) | <0.001 |
| Intervention year 3 | 104.9 | 122.2 | −17.3 (−26.6, −8.0) | <0.001 |
Abbreviations: NA, Not Applicable; CI, confidence interval.
Point estimates for utilization and utilization differences are propensity-weighted recycled predictions from two-part logistic and negative binomial regression models adjusting for baseline utilization rates; patient gender, age, Charlson comorbidity score; health plan contributing each observation, and whether each patient was in an HMO product at the time of the observation. Confidence intervals are bootstrap estimates from these two-part models; p-values are from one-part negative binomial regression models.
Due to the inclusion of fixed effects for practices, regression models do not estimate pre-intervention differences between pilot and comparison.
For primary care visits, only one-part negative binomial models converged (because, due to attribution methods, no patients had zero primary care visits in the pre-intervention period).
Sensitivity analyses differed from the main results in one way only: in logistic models, there was no statistically significant association between pilot participation and rates LDL-C testing among diabetics.
Discussion
To our knowledge, the northeast region of the Pennsylvania Chronic Care Initiative (PACCI) is the first evaluated multipayer medical home intervention to feature shared savings in addition to the financial resources, technical assistance, and recognition requirements typical of previously evaluated medical home interventions. In contrast to other recent evaluations of medical home interventions among small primary care practice sites,7–10 participation in the northeast PACCI was associated with improvements in the majority of quality measures examined, more use of ambulatory primary care visits, and lower use of hospital, ED, and ambulatory specialist visits.
Why did the northeast PACCI pilot intervention produce more quality improvements and utilization changes than previous medical home pilots evaluated by our team and others?7–10 Our study was not designed to identify specific mechanisms of improvement, but intervention attributes suggest several possibilities. First, the inclusion of a substantial shared savings incentive, with shared savings bonus payments being contingent on meeting quality measure benchmarks, may have been a particularly strong motivator for practices to invest and engage more effectively in care management efforts. Better care management may have contributed to the higher rate of patient retention that we observed among the pilot practices relative to comparison practices. Second, pilot practices received regular feedback from participating health plans on hospital, ED, and other medical utilization for their patients. Timely feedback may have enabled practices to more quickly adjust their efforts to meet quality and utilization benchmarks.
Third, the northeast PACCI intervention did not include a financial incentive tied to early achievement of medical home recognition, potentially enhancing participating practices’ abilities to focus on learning collaborative activities and other process improvement efforts. Fourth, all of the pilot practices had EHRs at baseline. Adopting new EHRs can be stressful for primary care practices and distract from other efforts to improve patient care.23 Fifth, pilot practices received relatively high levels of NCQA medical home recognition, with some recognized under the newer 2011 PPC-PCMH criteria released during the intervention. Thus they may have been better positioned to implement case management and other advanced capabilities.
Our study was designed to evaluate a particular medical home intervention (the northeast PACCI) rather than changes associated with practice-level implementation of a particular medical home model. Despite important differences in study design and setting, we note that studies of medical home implementation within the Veterans Health Administration (VHA) also have found associations with quality and utilization of care.24,25 Like the practices participating in the northeast PACCI, VHA primary care practices had access to data on their patients’ utilization of hospital and ED services, potentially enhancing their abilities to function effectively as medical homes.
To our knowledge there are four prior evaluations of medical home interventions that have found statistically significant reductions in one or more measures of hospitalization or ED utilization.9–11,26 The relative rate reductions in these evaluations (ranging from 6% to 18% for all-cause hospitalizations;11,26 approximately 30% for all-cause ED visits;10,26 and 12% for ambulatory care-sensitive ED visits9), were comparable to or greater than those observed in the northeast PACCI, with mutually overlapping confidence intervals surrounding these point estimates. We note, however, that multiple other medical home intervention evaluations have not detected such effects.5–8 By examining design differences between medical home interventions, the reasons for these discrepant results may become clearer.
We saw declines in quality measure performance and primary care ambulatory visit rates among comparison practices. These decreases may have been due to prospective patient attribution, which required that patients visit their primary care practices in the pre-intervention period but not thereafter. Also, performance on most evaluated quality measures was high at baseline, limiting room for improvement in general. It is also possible that unobserved changes in northeast Pennsylvania during the study period (which coincided with an economic recession) could have reduced primary care visit rates and use of recommended treatment and preventive services.
Our study has limitations. First, unobserved differences between pilot and comparison practices could affect our results, despite application of propensity score weighting and statistical adjustment. Second, the findings we observed may not generalize to other settings, other types of primary care practices, and other medical home initiatives. Third, the range of quality measures for which sufficient sample size existed was limited, and we did not assess changes in patient or provider experience. Fourth, complete data on the costs of implementing the northeast PACCI intervention (e.g., the costs of practice coaching) and on shared savings payments made from health plans to participating practices were unavailable to us. Therefore financial effects of the pilot could not be estimated. Fifth, patients in the main evaluation models were enrolled continuously in commercial health plans, and no data on patients’ sociodemographic characteristics were available for analysis; these characteristics of the evaluation may limit the generalizability of findings to patients lacking continuous health insurance and to patient populations dissimilar to the one we studied. Sixth, we lacked data on any structural transformation that may have occurred among comparison practices.
Conclusion
Over 100 medical home interventions are under way in the United States.4 They vary considerably in the mix of new resources, technical assistance, contractual obligations, performance measures, and incentives available to primary care practices. We believe evaluation results from the first 3 years of the northeast Pennsylvania Chronic Care Initiative offer guidance for program designers and policy makers. Medical home interventions that incentivize activities in addition to structural transformation may produce larger improvements in patient care. In particular, providing shared savings incentives and timely availability of data on ED visits and hospitalizations may encourage and enable primary care practices to contain unnecessary or avoidable utilization in these settings. Additional studies will be needed to determine empirically whether these features or others are indeed the key “active ingredients” in medical home interventions. Continuing experimentation and careful evaluation of the features of medical home interventions can inform the design of future programs intended to strengthen primary care.
Acknowledgments
This study was sponsored by the Commonwealth Fund. No sponsor had a role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.
The authors gratefully acknowledge Aaron Kofner, MA, MS (RAND), Scott Ashwood, PhD (RAND), Scot Hickey, MA (RAND), and Samuel Hirshman, BA (RAND) for assistance with programming and data management; Claude Setodji, PhD (RAND) for statistical consultation; and Marcela Myers, MD (Commonwealth of Pennsylvania), and Michael Bailit, MM (Bailit Health Purchasing, LLC) for providing data on pilot intervention design and NCQA recognition levels and facilitating other data collection.
Mr. Kofner, Dr. Ashwood, Mr. Hickey, Mr. Hirshman, and Dr. Setodji received compensation for their roles in the study.
Dr. Friedberg had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Dr. Friedberg has received compensation from the United States Department of Veterans Affairs for consultation related to medical home implementation and research support from the Patient-Centered Outcomes Research Institute via subcontract to the National Committee for Quality Assurance. Dr. Volpp has received compensation as a consultant from CVS Caremark and VALHealth and has received research funding from CVS Caremark, Humana, Horizon Blue Cross Blue Shield, Weight Watchers, and Discovery (South Africa). No other author has a potential conflict of interest to disclose.
Appendix A. Criteria for Selecting Primary Care Practices for Participation in the Northeast Region of the Pennsylvania Chronic Care Initiative.
| 1. | Practitioners are varied among pediatrics, family practice and internal medicine. |
| 2. | Practice sites derive a significant portion of their revenue from the participating carriers. |
| 3. | Practice site locations are varied (e.g., urban, suburban and rural). |
| 4. | Practice ownership is varied (e.g., academic health system, independently owned private practices, community health centers). |
| 5. | Practice sites are generally smaller in order to accommodate the targeted average practice size of three lead clinician FTEs, although larger practice sites may be considered if they agree that a maximum of five clinician FTEs at the site will be eligible for supplemental payments. |
Appendix B. Shared Savings Design Used by the Northeast Region of the Pennsylvania Chronic Care Initiative.
In the northeast region of the Pennsylvania Chronic Care Initiative (PACCI), practice eligibility to receive shared savings bonus payments was determined by the number of performance criteria on which the practice met or exceeded, by pilot month 18, the performance thresholds shown in Appendix Table B1.
Appendix Table B1.
| Criterion | Threshold* |
|---|---|
| PPC-PCMH recognition by NCQA at no less than Level 1 Plus | Binary practice-level measure (yes or no)** |
| Percentage of diabetic patients with HbA1c below 9% | 2 percentage point improvement, relative to baseline (before pilot began)*** |
| Percentage of diabetic patients with LDL-C below 100 | 2 percentage point improvement, relative to baseline (before pilot began)*** |
| Percentage of diabetic patients with blood pressure < 130/80 | 2 percentage point improvement, relative to baseline (before pilot began)*** |
| Percentage of hypertensive patients with blood pressure <140/90 | 2 percentage point improvement, relative to baseline (before pilot began)*** |
| Percentage of coronary artery disease patients with LDL-C below 100 | 2 percentage point improvement, relative to baseline (before pilot began)*** |
| 30-day hospital readmission rate | 2 percentage point reduction, relative to baseline (before pilot began) |
| Ambulatory Care Sensitive Condition (ACSC) hospitalization rate | 2 percentage point reduction, relative to baseline (before pilot began) |
| Primary care practice visit rate | 2 percentage point increase, relative to baseline (before pilot began) |
| Emergency room visit rate | 5 percentage point reduction, relative to baseline (before pilot began) |
| Documented care plan | For all patients identified as high risk by each participating health plan |
| Documented self-management support goal setting | For 90% of patients identified as high risk by each participating health plan |
| Practice team clinical telephonic or face-to-face patient follow-up within 2 days after hospitalization discharge | For 75% of hospital discharges |
| Documentation that there is a care manager in place and that the care manager is operating consistently with the requirements set forth in the Participation Agreement | Binary practice-level measure (yes or no) |
Methods for determining practice performance on each criterion other than PPC-PCMH recognition (which was determined by the NCQA) were not specified in the Participation Agreement for the Northeast PACCI and are unknown to the authors.
Level 1 Plus defined as meeting the 2008 NCQA PPC-PCMH Level 1 standards, plus the following 2008 NCQA PPC-PCMH standards at the specified levels of performance: 3C (Care Management: Practice Organization) at 75%, 3D (Care Management for Important Conditions) at 100%, and 4B (Patient Self-Management Support) at 50%.
If a practice’s performance was equal to or exceeded the NCQA Mid-Atlantic region All Lines of Business 90th percentile rate for the measure for the most recently reported measurement year, the practice did not need to demonstrate a 2 percentage point improvement.
Using the thresholds in Appendix Table A1, shared savings bonuses for each practice were calculated as shown in Appendix Table A2.
Appendix Table B2.
| Number of performance thresholds met (of 14 possible thresholds) | Shared savings bonus payment |
|---|---|
| 14 | 50%* of observed savings** |
| 12–13 | 47%* of observed savings |
| 10–11 | 44%* of observed savings |
| 9 | 41%* of observed savings |
| 8 or fewer | Practice not eligible for shared savings |
Minus the combined value of the Care Management ($1.50 per patient per year) and Practice Support ($1.50 per patient per year) payments to the practice.
Observed savings were determined annually by each participating health plan. Each plan could develop its own savings calculation method within the following parameters specified in the Participation Agreement for the Northeast PACCI: “Savings [are] determined annually by comparing risk-adjusted actual to expected medical costs for the Practice’s patient population, with an adjustment for outliers, and subtracting from any resulting savings the value of Care Management and Practice Support Payments during the measurement year. Savings will be calculated based on the experience of the Practice’s patients who are enrolled with a Carrier. Carriers are solely responsible for determining the methodology for calculating actual and expected medical costs.”
Appendix C. Measure specifications.
| Quality measures | |
|---|---|
| Measure name | Definition |
| Breast Cancer Screening | Percentage of women aged 40–69 years who had at least one mammogram in the measurement year or year prior to the measurement year. |
| Colorectal Cancer Screening | Percentage of adults 50 to 80 years of age who had 1 or more of the following during the measurement year: fecal occult blood test, flexible sigmoidoscopy, double contrast barium enema or air contrast barium enema, or colonoscopy. |
| Comprehensive Diabetes Care: HbA1c Testing | Percentage of patients aged 18–75 years with diabetes (type 1 and type 2) who had a hemoglobin A1c test during the measurement year. |
| Comprehensive Diabetes Care: Eye Exams | Percentage of patients aged 18–75 years with diabetes (type 1 and type 2) who had a retinal or dilated eye exam by an eye care professional in the measurement year or a negative retinal exam (no evidence of retinopathy) by an eye care professional in the measurement year. |
| Comprehensive Diabetes Care: Cholesterol Screening | Percentage of patients aged 18–75 years with diabetes (type 1 and type 2) who had a low-density lipoprotein cholesterol test during the measurement year. |
| Comprehensive Diabetes Care: Monitoring Diabetic Nephropathy | Percentage of patients aged 18–75 years with diabetes (type 1 and type 2) who received nephropathy screening, had a nephrologist visit, or had evidence of nephropathy as documented through administrative data during the measurement year. |
Definitions taken from National Committee for Quality Assurance (NCQA). HEDIS 2009. Health plan employer data & information set. Vol. 2, Technical specifications. Washington (DC): National Committee for Quality Assurance (NCQA); 2009. Detailed lists of comorbid conditions, competing diagnoses, and other exclusion criteria are contained in the original measure documentation available from the NCQA.
| Utilization measures | |
|---|---|
| Measure name | Definition |
| Hospitalization rate, all-cause | Count of unique hospitalizations for any reason per month. |
| Hospitalization rate, ambulatory care-sensitive | Count of unique hospitalizations per month that meet one or more criteria for being ambulatory care-sensitive according to the Agency for Healthcare Research and Quality “Prevention Quality Indicators Technical Specifications,” Version 4.0. Specifications available from Agency for Healthcare Research and Quality, http://www.qualityindicators.ahrq.gov/Modules/PQI_TechSpec.aspx |
| Emergency department visit rate, all-cause | Count of unique emergency department visits for any reason per month. |
| Emergency department visit rate, ambulatory care-sensitive | Count of unique emergency department visits per month that have any evidence of being avoidable or primary care treatable according to the “NYU ED Algorithm,” specifications available from http://wagner.nyu.edu/faculty/billings/nyued-download. For each ED visit, the NYU algorithm assigns a probability that the visit is in one of 4 categories: 1- Non-Emergent; 2- Emergent, Primary Care Treatable; 3- Emergent, ED Care Needed, Preventable/Avoidable; 4- Emergent, ED Care Needed, Not Preventable/Avoidable. For this measure, we count an ED visit as “ambulatory care-sensitive” if it has a nonzero probability of belonging in any of the first 3 categories. |
| Ambulatory care visit rate | Count of unique ambulatory visits (excluding emergency department visits) for any reason per month. |
Appendix D. Qualifying services for patient attribution.
Services that qualified for patient attribution were those that were provided by primary care clinicians (specialty designations “family practice,” “general practice,” “internal medicine,” “pediatrics,” “adolescent medicine,” “geriatric medicine,” and “nurse practitioner”) and that had one of the following CPT codes: 9920×, 9921×, 9924×, 99381 – 99387, 99391 – 99397, 99401 – 99404, 99411 – 99412, 99420 – 99429, 99339 – 99340, 99341 – 99345, 99347 – 99350, G0402, G0438, G0439.
Appendix E. Propensity-weighted, adjusted differences in utilization of care between pilot and comparison practices among continuously enrolled patients (n=10548 pilot and n=6815 comparison patients) with P values from each part of the 2 part models.
| Pilot | Comparison | Difference (95% CI) | P value, logistic part | P value, negative binomial part | |
|---|---|---|---|---|---|
| Hospitalizations, all-cause | Rate per 1000 patients per month (95% CI)* | ||||
| Pre-intervention | 7.0 | 7.0 | NA** | NA | NA |
| Intervention year 1 | 7.3 | 8.8 | −1.5 (−3.1, 0.2) | 0.056 | 0.40 |
| Intervention year 2 | 7.4 | 9.2 | −1.8 (−3.3, −0.2) | <0.001 | 0.070 |
| Intervention year 3 | 8.5 | 10.2 | −1.7 (−3.2, −0.03) | 0.075 | 0.37 |
| Hospitalizations, ambulatory care-sensitive | |||||
| Pre-intervention | 0.5 | 0.5 | NA | NA | NA |
| Intervention year 1 | 0.5 | 0.7 | −0.2 (−0.6, 0.2) | 0.174 | 0.41 |
| Intervention year 2 | 0.8 | 0.9 | −0.1 (−0.5, 0.4) | 0.46 | 0.081 |
| Intervention year 3 | 0.8 | 1.0 | −0.2 (−0.6, 0.3) | 0.28 | 0.50 |
| ED visits, all-cause | |||||
| Pre-intervention | 23.9 | 23.9 | NA | NA | NA |
| Intervention year 1 | 24.5 | 27.5 | −3.0 (−6.6, 0.5) | 0.55 | 0.028 |
| Intervention year 2 | 26.3 | 28.4 | −2.1 (−5.6, 1.1) | 0.074 | 0.92 |
| Intervention year 3 | 29.5 | 34.2 | −4.7 (−8.7, −0.9) | 0.27 | 0.005 |
| ED visits, ambulatory care-sensitive | |||||
| Pre-intervention | 13.5 | 13.5 | NA | NA | NA |
| Intervention year 1 | 13.5 | 15.0 | −1.4 (−3.8, 1.1) | 0.60 | 0.31 |
| Intervention year 2 | 14.5 | 15.6 | −1.2 (−3.8, 1.3) | 0.146 | 0.70 |
| Intervention year 3 | 16.2 | 19.4 | −3.2 (−5.7, −0.9) | 0.092 | 0.111 |
| Ambulatory visits, primary care | Rate per 1000 patients per month (95% CI)* | ||||
| Pre-intervention | 379.6 | 379.6 | NA** | NA | NA |
| Intervention year 1 | 357.0 | 304.2 | 52.8 (9.1, 99.4) | NA*** | 0.024 |
| Intervention year 2 | 357.1 | 250.0 | 107 (51.1, 178.5) | NA | 0.002 |
| Intervention year 3 | 349.0 | 271.5 | 77.5 (37.3, 120.5) | NA | 0.001 |
| Ambulatory visits, specialist | |||||
| Pre-intervention | 106.2 | 106.2 | NA | NA | NA |
| Intervention year 1 | 108.7 | 117.3 | −8.7 (−16.2, −1.2) | 0.27 | 0.01 |
| Intervention year 2 | 104.8 | 121.3 | −16.5 (−27.5, −5.9) | 0.34 | <0.001 |
| Intervention year 3 | 104.9 | 122.2 | −17.3 (−26.6, −8.0) | 0.23 | <0.001 |
Abbreviations: NA, Not Applicable; CI, confidence interval.
Point estimates for utilization and utilization differences are propensity-weighted recycled predictions from two-part logistic and negative binomial regression models adjusting for baseline utilization rates; patient gender, age, Charlson comorbidity score; health plan contributing each observation, and whether each patient was in an HMO product at the time of the observation. Confidence intervals are bootstrap estimates; p-values are from the regression models.
Due to the inclusion of fixed effects for practices, regression models do not estimate pre-intervention differences between pilot and comparison.
For primary care visits, only one-part negative binomial models converged (because, due to attribution methods, no patients had zero primary care visits in the pre-intervention period).
Appendix F. Selected structural changes among pilot practices.
| Baseline | Pilot year 3 | P value* | |
|---|---|---|---|
| Performance feedback | Number of practices (%)** | ||
| Quality feedback to PCPs (N=21 practices) | 19 (90%) | 21 (100%) | 0.50 |
| Utilization or cost feedback to PCPs (N=23) | 19 (83%) | 18 (78%) | 0.99 |
| Monthly or more frequent meetings about quality (N=22) | 7 (32%) | 21 (95%) | <0.001 |
| Monthly or more frequent meetings about utilization (N=23 | 7 (30%) | 13 (56%) | 0.070 |
| Registry use | |||
| Registry of patients who are overdue for screening services (N=23) | 14 (61%) | 22 (96%) | 0.022 |
| Registry of patients who are overdue for chronic disease services (N=22) | 14 (63%) | 22 (100%) | 0.008 |
| Registry of patients who are out of target range for chronic disease laboratory values (N=23) | 15 (65%) | 23 (100%) | 0.008 |
| Registry of patients at high risk of disease complications or hospitalization (N=23 | 16 (70%) | 23(100%) | 0.016 |
| Care management | |||
| Care management for patients at high risk of disease complications or hospitalization (N=23) | 6 (26%) | 24 (100%) | <0.001 |
| Specially-trained non-physician staff who help patients better manage their diabetes (N=22) | 16 (73%) | 21 (95%) | 0.124 |
| Specially-trained non-physician staff who help patients better manage their asthma (N=22) | 9 (41%) | 20 (91%) | 0.007 |
| Routine assessment of self-management needs of chronically ill patients (N=23) | 5(22%) | 23 (100%) | <0.001 |
| Referral system for linking patients to community programs (N=22) | 6 (27%) | 11 (50%) | 0.267 |
| Breast cancer screening (N=23) | 9 (39%) | 22 (96%) | <0.001 |
| Cervical cancer screening (N=23) | 9 (39%) | 21 (91%) | <0.001 |
| Colorectal cancer screening (N=23) | 9 (39%) | 22 (96%) | <0.001 |
| Diabetes: hemoglobin A1c testing (N=23) | 16 (70%) | 22 (96%) | 0.031 |
| Diabetes: cholesterol testing (N=23) | 17 (74%) | 22 (96%) | 0.062 |
| Diabetes: eye examination (N=23) | 10 (43%) | 22 (96%) | <0.001 |
| Diabetes: nephropathy monitoring (N=23) | 9 (39%) | 21 (91%) | <0.001 |
| Other outreach systems | |||
| Outreach to patients after hospitalization (N=23) | 10 (43%) | 23 (100%) | <0.001 |
| Outreach to patients with no appointment in for an extended period (longer than clinically appropriate) (N=22) | 8 (36%) | 21 (95%) | <0.001 |
| Electronic health record capabilities | |||
| Patient medication lists (N=23) | 23 (100%) | 23 (100%) | 0.99 |
| Patient problem lists (N=23) | 23 (100%) | 23 (100%) | 0.99 |
| Consultation notes from specialists (N=23) | 21 (91%) | 20 (87%) | 0.99 |
| Hospital discharge summaries (N=23) | 21 (91%) | 22 (96%) | 0.99 |
| Electronic medication prescribing (N=23) | 22 (96%) | 23 (100%) | 0.99 |
| Electronic laboratory test ordering (N=23) | 11 (48%) | 20 (87%) | 0.004 |
| Electronic radiology test ordering (N=23) | 11 (48%) | 19 (82%) | 0.022 |
| Alerts if ordered tests are not performed (N=23) | 6 (26%) | 16 (70%) | 0.006 |
| Secure electronic messaging to and from patients (N=23) | 7 (30%) | 9 (39%) | 0.50 |
| Access | |||
| Weekend care offered regularly (N=22) | 6 (27%) | 8 (36%) | 0.50 |
| Evening care offered ≥2 nights per week (N=23) | 10 (43%) | 13(57%) | 0.45 |
| Appointments for new patients within 2 weeks (N=22) | 5(23%) | 4(18%) | 0.99 |
Abbreviations: NCQA, National Committee for Quality Assurance; NA, Not Applicable; PCP, primary care physician or clinician (including MDs, DOs, and NPs).
Liddell exact test.
Due to item nonresponse, denominators for percentages are not the same for all entries in the table.
References
- 1.American Academy of Family Physicians, American Academy of Pediatrics, American College of Physicians, American Osteopathic Association. Joint principles of the patient-centered medical home. 2007; http://www.acponline.org/running_practice/delivery_and_payment_models/pcmh/demonstrations/jointprinc_05_17.pdf. Accessed January 12, 2015.
- 2.Patient-Centered Primary Care Collaborative. Statements of Support. 2014; http://www.pcpcc.net/content/statements-support. Accessed January 12, 2015.
- 3.Friedberg MW, Lai DJ, Hussey PS, Schneider EC. A guide to the medical home as a practice-level intervention. Am J Manag Care. December 2009;15(10 Suppl):S291–299. [PubMed] [Google Scholar]
- 4.Edwards ST, Bitton A, Hong J, Landon BE. Patient-centered medical home initiatives expanded in 2009–13: providers, patients, and payment incentives increased. Health Aff (Millwood). October 1 2014;33(10):1823–1831. [DOI] [PubMed] [Google Scholar]
- 5.Peikes D, Zutshi A, Genevro JL, Parchman ML, Meyers DS. Early evaluations of the medical home: building on a promising start. Am J Manag Care. 2012;18(2):105–116. [PubMed] [Google Scholar]
- 6.Jackson GL, Powers BJ, Chatterjee R, et al. The patient-centered medical home: a systematic review. Annals of Internal Medicine. February 5 2013;158(3):169–178. [DOI] [PubMed] [Google Scholar]
- 7.Friedberg MW, Schneider EC, Rosenthal MB, Volpp KG, Werner RM. Association between participation in a multipayer medical home intervention and changes in quality, utilization, and costs of care. JAMA. February 26 2014;311(8):815–825. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Werner RM, Duggan M, Duey K, Zhu J, Stuart EA. The Patient-centered Medical Home: An Evaluation of a Single Private Payer Demonstration in New Jersey. Med Care. June 2013;51(6):487–493. [DOI] [PubMed] [Google Scholar]
- 9.Rosenthal MB, Friedberg MW, Singer SJ, Eastman D, Li Z, Schneider EC. Effect of a multipayer patient-centered medical home on health care utilization and quality: the Rhode Island chronic care sustainability initiative pilot program. JAMA Intern Med. November 11 2013;173(20):1907–1913. [DOI] [PubMed] [Google Scholar]
- 10.Fifield J, Forrest DD, Burleson JA, Martin-Peele M, Gillespie W. Quality and Efficiency in Small Practices Transitioning to Patient Centered Medical Homes: A Randomized Trial. J Gen Intern Med. March 2 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gilfillan RJ, Tomcavage J, Rosenthal MB, et al. Value and the medical home: effects of transformed primary care. Am J Manag Care. 2010;16(8):607–614. [PubMed] [Google Scholar]
- 12.Gabbay RA, Bailit MH, Mauger DT, Wagner EH, Siminerio L. Multipayer patient-centered medical home implementation guided by the chronic care model. Jt Comm J Qual Patient Saf. June 2011;37(6):265–273. [DOI] [PubMed] [Google Scholar]
- 13.Gabbay RA, Friedberg MW, Miller-Day M, Cronholm PF, Adelman A, Schneider EC. A positive deviance approach to understanding key features to improving diabetes care in the medical home. Ann Fam Med. May-Jun 2013;11 Suppl 1:S99–107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Friedberg MW, Safran DG, Coltin KL, Dresser M, Schneider EC. Readiness for the patient-centered medical home: structural capabilities of Massachusetts primary care practices. J Gen Intern Med. February 2009;24(2):162–169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Rosenthal MB, Abrams MK, Bitton A. Recommended core measures for evaluating the patient-centered medical home: cost, utilization, and clinical quality. New York, NY: Commonwealth Fund; 2012. Available from http://www.commonwealthfund.org/~/media/files/publications/data-brief/2012/1601_rosenthal_recommended_core_measures_pcmh_v2.pdf. Accessed January 12, 2015. [Google Scholar]
- 16.Rosenthal MB, Beckman HB, Forrest DD, Huang ES, Landon BE, Lewis S. Will the patient-centered medical home improve efficiency and reduce costs of care? A measurement and research agenda. Med Care Res Rev. August 2010;67(4):476–484. [DOI] [PubMed] [Google Scholar]
- 17.Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70(1):41–55. [Google Scholar]
- 18.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–383. [DOI] [PubMed] [Google Scholar]
- 19.Liang K, Zeger S. Longitudinal data analysis using generalized linear models. Biometrika. 1986;73:13–22. [Google Scholar]
- 20.Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics. March 1986;42(1):121–130. [PubMed] [Google Scholar]
- 21.Graubard BI, Korn EL. Predictive margins with survey data. Biometrics. June 1999;55(2):652–659. [DOI] [PubMed] [Google Scholar]
- 22.Setodji CM, Scheuner M, Pankow JS, Blumenthal RS, Chen H, Keeler E. A graphical method for assessing risk factor threshold values using the generalized additive model: the multi-ethnic study of atherosclerosis. Health Serv Outcomes Res Methodol. March 2012;12(1):62–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Nutting PA, Miller WL, Crabtree BF, Jaen CR, Stewart EE, Stange KC. Initial lessons from the first national demonstration project on practice transformation to a patient-centered medical home. Ann Fam Med. May-Jun 2009;7(3):254–260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Nelson KM, Helfrich C, Sun H, et al. Implementation of the patient-centered medical home in the Veterans Health Administration: associations with patient satisfaction, quality of care, staff burnout, and hospital and emergency department use. JAMA Intern Med. August 2014;174(8):1350–1358. [DOI] [PubMed] [Google Scholar]
- 25.Yoon J, Rose DE, Canelo I, et al. Medical Home Features of VHA Primary Care Clinics and Avoidable Hospitalizations. J Gen Intern Med. September 2013;28(9):1188–1194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Reid RJ, Coleman K, Johnson EA, et al. The group health medical home at year two: cost savings, higher patient satisfaction, and less burnout for providers. Health Aff. May 1, 2010. 2010;29(5):835–843. [DOI] [PubMed] [Google Scholar]