Research
Original Investigation
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 Karin M. Nelson, MD, MSHS; Christian Helfrich, MPH, PhD; Haili Sun, PhD; Paul L. Hebert, PhD; Chuan-Fen Liu, MPH, PhD; Emily Dolan, PhD; Leslie Taylor, PhD; Edwin Wong, PhD; Charles Maynard, PhD; Susan E. Hernandez, MPA; William Sanders, AA, AS; Ian Randall, MHSA; Idamay Curtis, BA; Gordon Schectman, MD; Richard Stark, MD; Stephan D. Fihn, MD, MPH
IMPORTANCE In 2010, the Veterans Health Administration (VHA) began implementing the
patient-centered medical home (PCMH) model. The Patient Aligned Care Team (PACT) initiative aims to improve health outcomes through team-based care, improved access, and care management. To track progress and evaluate outcomes at all VHA primary care clinics, we developed and validated a method to assess PCMH implementation.
Invited Commentary page 1358 Supplemental content at jamainternalmedicine.com
OBJECTIVES To create an index that measures the extent of PCMH implementation, describe variation in implementation, and examine the association between the implementation index and key outcomes. DESIGN, SETTING, AND PARTICIPANTS We conducted an observational study using data on more than 5.6 million veterans who received care at 913 VHA hospital-based and community-based primary care clinics and 5404 primary care staff from (1) VHA clinical and administrative databases, (2) a national patient survey administered to a weighted random sample of veterans who received outpatient care from June 1 to December 31, 2012, and (3) a survey of all VHA primary care staff in June 2012. Composite scores were constructed for 8 core domains of PACT: access, continuity, care coordination, comprehensiveness, self-management support, patient-centered care and communication, shared decision making, and team-based care. MAIN OUTCOMES AND MEASURES Patient satisfaction, rates of hospitalization and emergency department use, quality of care, and staff burnout. RESULTS Fifty-three items were included in the PACT Implementation Progress Index (Pi2).
Compared with the 87 clinics in the lowest decile of the Pi2, the 77 sites in the top decile exhibited significantly higher patient satisfaction (9.33 vs 7.53; P < .001), higher performance on 41 of 48 measures of clinical quality, lower staff burnout (Maslach Burnout Inventory emotional exhaustion subscale, 2.29 vs 2.80; P = .02), lower hospitalization rates for ambulatory care–sensitive conditions (4.42 vs 3.68 quarterly admissions for veterans 65 years or older per 1000 patients; P < .001), and lower emergency department use (188 vs 245 visits per 1000 patients; P < .001). CONCLUSIONS AND RELEVANCE The extent of PCMH implementation, as measured by the Pi2,
was highly associated with important outcomes for both patients and providers. This measure will be used to track the effectiveness of implementing PACT over time and to elucidate the correlates of desired health outcomes.
JAMA Intern Med. 2014;174(8):1350-1358. doi:10.1001/jamainternmed.2014.2488 Published online June 23, 2014. 1350
Author Affiliations: Author affiliations are listed at the end of this article. Corresponding Author: Karin M. Nelson, MD, MSHS, Seattle Center of Innovation for Veteran-Centered and Value-Driven Care, VA Puget Sound Health Care System, 1100 Olive Way, Ste 1400, Seattle, WA 98108 (karin [email protected]). jamainternalmedicine.com
Copyright 2014 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ by a Ndsu Library Periodicals User on 05/18/2015
Patient-Centered Medical Home Implementation
A
lthough the patient-centered medical home (PCMH) has been endorsed by most major primary care groups as a promising model to strengthen primary care, decrease costs, and improve quality,1 early assessment of PCMH impact have yielded mixed results.2-10 Since 2010, the Veterans Health Administration (VHA) has undertaken national adoption of a PCMH model, called PACT (Patient Aligned Care Team).11 The focus of PACT has been to restructure primary care to provide team-based care that is more comprehensive, coordinated, and patient centered.11 The PACT initiative is a multifaceted and complex intervention, creating challenges to measuring implementation across diverse clinic sites. One of the most widely recognized PCMH recognition tools is the National Committee for Quality Assurance (NCQA) certification process, which focuses on practice infrastructure and health information technology,12 an area in which the VHA has made considerable past investments.13,14 The VHA has a universally deployed electronic health record, electronic prescribing, patient registries, and a national quality improvement and performance measurement infrastructure for which all clinics in the VHA would receive “credit.” Many national programs for coordinating care, such as home-based primary care, integrated mental health services, and palliative care, were already widely available before PACT was initiated. The focus within the VHA has been on how effectively these extensive resources are being applied and coordinated to fulfill the goals of the PACT initiative.11 Our goal was to derive a comprehensive index from existing data and survey instruments that would have a low respondent burden and would reflect processes and attributes that are essential to effective primary care. Our approach differs from other PCMH measurement tools15 by incorporating multiple data sources, including a primary care personnel survey, patient surveys, and administrative data. We sought to develop a measure to represent areas of focus of the PACT initiative, including continuity through team-based care, patient access, care coordination, and patient-centered care.11 We desired an instrument that would facilitate comparisons across clinical sites within the VHA, assist in identifying sites that had most effectively implemented PACT, and determine the relationship between effective implementation and important outcomes, such as patient satisfaction, quality of care, provider experience, and use of health care services.
Methods
(0.61).16 To test convergent validity, we used information on patient satisfaction from another sample of veterans from the Survey of the Health Experiences of Patients, an ongoing national mailed US Department of Veterans Affairs (VA) survey that assesses the health care experiences of veterans who receive care at the VHA and uses a stratified random sampling method.17 The evaluation efforts are part of an going quality improvement effort at the VHA and are not considered research activity; they are thus not subject to institutional review board review or waiver. Primary Care Personnel Survey The PACT Primary Care Personnel Survey was an internally developed instrument designed to measure team functioning in PACT and has been described elsewhere.18 The target population of the survey was all VHA primary care personnel, including the 4 occupations included in PACT teams: primary care providers, nurse care managers, medical associates (eg, licensed practical nurses and medical technicians), and administrative clerks. Data were collected from May 21 through June 29, 2012. Team-based care was represented by items from the primary care personnel survey related to delegation, staffing, team functioning, and team assignment.18 Administrative Data Information about demographics, clinical characteristics, and use of health services was obtained from the VHA Corporate Data Warehouse for fiscal year 2012 (n = 5 653 616). Using data from the Primary Care Management Module contained within the Corporate Data Warehouse, we identified all patients who were enrolled in primary care and assigned to a primary care provider.19 We included administrative data for important PACT programmatic goals,11 including (1) access to care and use of non–face-to-face care, such as telephone clinics and secure messaging; (2) continuity of care; and (3) use of VHA programs to support care coordination (eg, home telemonitoring, 2-day posthospital follow-up). We used data collected by the VHA External Peer Review Program (EPRP) during fiscal year 2012 to assess quality of care. The EPRP is an audit program designed to assess clinical performance using standard performance criteria. National data are collected through manual abstraction of electronic health records by an independent external contractor.20 Previous studies have found high interrater reliability (κ = 0.9) within the EPRP program.14
Construction of the PACT Implementation Progress Index
Survey Instruments and Data Sources Patient Survey We used data from the previously validated Consumer Assessment of Health Plans–Patient Centered Medical Home (CAHPS PCMH) survey16 that was administered to a nationally weighted random sample of veterans who received outpatient care from June 1 to December 31, 2012. The CAHPS PCMH scales have acceptable internal consistency reliability estimates for access (Cronbach α, 0.74), comprehensiveness (0.68), self-management support (0.62), patient-centered care and communication (0.91), and shared decision making jamainternalmedicine.com
Original Investigation Research
The method for developing the PACT Implementation Progress Index (Pi2) and a full description of all items are provided in eTable 1 in the Supplement. Briefly, we mapped data items to PACT conceptual domains, calculated domain scores based on these items, and then generated site-level rankings for each domain. Table 1 outlines Pi2 domains and provides examples of representative variable items. A Pi2 score was assigned to each clinic based on the number of domains in the top and bottom quartiles for the domain scores, ranging from 8 (all domain scores in the top quartile) to –8 (all domain scores in the bottom quartile). Using these scores, we categorized sites in JAMA Internal Medicine August 2014 Volume 174, Number 8
Copyright 2014 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ by a Ndsu Library Periodicals User on 05/18/2015
1351
Research Original Investigation
Patient-Centered Medical Home Implementation
Table 1. PCMH Concepts and Pi2 Domains Items From Each Data Source, No. PACT Initiative Goals Accessible, continuous, and coordinated care
Patient-centered care
Team-based care
Total Items, No.
CAHPS PCMHa
CDWb
11
6
5
0
How often did you get an appointment as soon as you needed?a When you phoned this provider’s office, how often did you get an answer to your medical question that same day?a Same-day access to appointments (% of clinics)b Enhanced access: telephone clinics (%)b
Continuity of care
3
1
2
0
How long have you been going to this provider?a Proportion of visits to assigned primary care providerb
Care coordination
7
5
2
0
When this provider ordered a test, how often did someone from this provider’s office follow up to give you those results?a Did the provider seem informed and up to date about the care you got from specialists?a Percentage of patients contacted 2 d after hospital dischargeb
Comprehensiveness
3
3
0
Did you and anyone in this provider’s office talk about things in your life that worry you or cause you stress?a
Self-management support
2
2
0
Did anyone in this provider’s office talk with you about specific goals for your health?a
Patient-centered care and communication
6
6
0
How often did this provider explain things in a way that was easy to understand?a
Shared decision making
3
3
0
When you talked about starting a prescription medicine, did this provider ask you what you thought was best for you?a
Delegation, staffing, and team functioning
18
0
0
18
Primary care provider relies on registered nurse care manager for tasks including gathering patient preventive services, responding to prescription refillsc Percentage reporting recommended staffing ratioc Time spent in team huddles: percentage spending >30 min/dc
…
53
26
9
18
…
Pi2 Domains Access
Total
Abbreviations: CAHPS PCMH, Consumer Assessment of Health Plans–Patient Centered Medical Home; CDW, Corporate Data Warehouse; PACT, Patient Aligned Care Team; PCMH, patient-centered medical home; Pi2, PACT Implementation Progress Index. a
CAHPS PCMH patient survey.
the top decile of the Pi2 (score, 5 to 8) as having achieved effective implementation and those in the lowest decile of the Pi2 (score, –7 to –5) as having been less effective.
Patient- and Provider-Level Outcome Measures Patient satisfaction was assessed by using a single item from the CAHPS PCMH survey16 as follows: “Using any number from 0 to 10, where 0 is the worst provider possible and 10 is the best provider possible, what number would you use to rate this provider?” Staff burnout was assessed with both a single-item measure and the emotional exhaustion subscale of the Maslach Burnout Inventory, a widely used measure of burnout.21-23 The single item measure asks: “Overall, based on your definition of burnout, how would you rate your level of burnout,” with 5 ordinal response options.24 We defined burnout as a response of 3 or higher, where 3 corresponds to “I am definitely burning out and have one or more symptoms of burnout, such as physical and emotional exhaustion.”24 We also used a 3-item version of the Maslach Burnout Inventory subscale.25 Items reflecting burnout symptoms are scored using a Likert scale ranging from 0 (never) to 6 (every day) and summed to form a scale score. We defined burnout as a score of 10 or higher (range, 0-18). To assess quality of care, we examined outpatient measurements from the EPRP for chronic disease management, behavioral health screening, and prevention services. These in1352
Personnel Surveyc Example Itemd
b
CDW data from US Department of Veterans Affairs.
c
PACT Primary Care Personnel Survey.
d
Full description in eTable 1 in the Supplement. The questions have been shortened for the sake of brevity.
dicators include frequently used measures of the quality of prevention (eg, vaccinations, screening tests) and outpatient care of chronic diseases (eg, annual retinal examinations in patients with diabetes mellitus). The performance measurement for preventive services and chronic disease cohorts, sampling frame, and criterion for meeting the measurement are provided in eTable 2 in the Supplement. The EPRP selects a random sample of patient records from VHA facilities to monitor quality and appropriateness of medical care.26 The sample includes veterans who used VHA health care at least once in the 2 years before the assessment. Patients who were sampled had at least 1 primary care or specialty medical visit in the month being sampled. Among eligible patients, a random sample is drawn with oversampling of prevalent chronic conditions (eg, diabetes, heart failure).26 For patients at each primary care site, we determined the numbers of emergency department or urgent care visits, VA hospital admissions, and hospitalizations for ambulatory care– sensitive conditions (ACSCs), which are postulated to be most avoidable through provision of effective primary care.27 Hospitalizations for ACSCs were based on Agency for Healthcare Research and Quality Prevention Quality Indicators and were identified through standardized protocols using International Classification of Diseases, Ninth Revision, diagnoses and Current Procedural Terminology codes from inpatient VA records.27
JAMA Internal Medicine August 2014 Volume 174, Number 8
Copyright 2014 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ by a Ndsu Library Periodicals User on 05/18/2015
jamainternalmedicine.com
Patient-Centered Medical Home Implementation
Statistical Analysis PI2 Properties To test internal consistency reliability, we calculated the Cronbach α for all items in each domain and all 53 items that make up the total scale. Variation in PCMH Adoption We evaluated bivariate comparisons of facility characteristics and level of implementation by using χ2 tests for categorical variables and t tests for continuous variables. We compared sites assessed to have effectively implemented PACT with those assessed as less effective according to type of facility (hospital or community-based outpatient clinic), number of patients, demographic characteristics, and Elixhauser comorbidity score.28 Associations With Patient and Provider Outcomes We used a nonparametric test of trend for the ranks across ordered groups (an extension of the Wilcoxon rank sum test) to test for trends in patient satisfaction and staff burnout by Pi2 scores. We tested differences in the proportions of eligible patients at each VHA clinic fulfilling each of the 48 quality indicators according to the success of PACT implementation as measured by the Pi2. We calculated rates of services at the facility level by dividing the number of patients who satisfied the EPRP quality measure by the number who met inclusion criteria for each quality measure (eTable 3 in the Supplement). For each of the 48 facility-level quality indicators, we tested the trend in proportions of patients fulfilling the EPRP quality guideline by the level of PACT implementation. We used the nonparametric test for trend developed by Cuzick, which is an extension of the Wilcoxon test.29 We adjusted for multiple comparisons using a method described by Benjamini and Yekutieli.30 To determine whether more effective implementation (as measured by Pi2) corresponded to higher performance overall, we included all 48 outcome measures in a linear mixed-effects model that accounted for correlation among outcomes from the same facility and estimated an overall implementation effect. We adjusted for implementation in this model as a linear term ranging from 1 to 5, corresponding to the grouped Pi2 scores. This approach was possible because all 48 outcomes were measured on the same scale. We examined fiscal year 2012 emergency department and urgent care visits and total hospitalizations for sites with more effective vs less effective implementation, adjusting for patient age, community-based outreach clinics, and Elixhauser comorbidity scores.28 To account for temporal trends, we modeled facility-level trends for hospitalization from 2003 to 2012. The method for examining such trends has been described elsewhere.31 We estimated interrupted time-series models of ACSC and all-cause hospitalizations from October 1, 2003, through September 20, 2012, for each facility and assessed how the trends in hospitalizations changed after the start of the PACT initiative in April 2010. All regression models adjusted for facility-level patient characteristics, unemployment rate in the VA market area, quarterly dummy variables to capture seasonal variation, and a linear time trend. Patient risk was measured using mean facility-level Elixhauser comorbidity scores. jamainternalmedicine.com
Original Investigation Research
Changes in admissions for ACSC and all-cause hospitalizations after implementation of the PACT initiative were calculated as the difference between the observed rate of admissions and the predicted rate had the initiative not been implemented during the 2½-year period between April 1, 2010, and September 30, 2012. In this way, we estimated changes in admissions that might be attributed to the PACT initiative. Trend analyses for hospitalizations were stratified by age (≥65 and
Original Investigation
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 Karin M. Nelson, MD, MSHS; Christian Helfrich, MPH, PhD; Haili Sun, PhD; Paul L. Hebert, PhD; Chuan-Fen Liu, MPH, PhD; Emily Dolan, PhD; Leslie Taylor, PhD; Edwin Wong, PhD; Charles Maynard, PhD; Susan E. Hernandez, MPA; William Sanders, AA, AS; Ian Randall, MHSA; Idamay Curtis, BA; Gordon Schectman, MD; Richard Stark, MD; Stephan D. Fihn, MD, MPH
IMPORTANCE In 2010, the Veterans Health Administration (VHA) began implementing the
patient-centered medical home (PCMH) model. The Patient Aligned Care Team (PACT) initiative aims to improve health outcomes through team-based care, improved access, and care management. To track progress and evaluate outcomes at all VHA primary care clinics, we developed and validated a method to assess PCMH implementation.
Invited Commentary page 1358 Supplemental content at jamainternalmedicine.com
OBJECTIVES To create an index that measures the extent of PCMH implementation, describe variation in implementation, and examine the association between the implementation index and key outcomes. DESIGN, SETTING, AND PARTICIPANTS We conducted an observational study using data on more than 5.6 million veterans who received care at 913 VHA hospital-based and community-based primary care clinics and 5404 primary care staff from (1) VHA clinical and administrative databases, (2) a national patient survey administered to a weighted random sample of veterans who received outpatient care from June 1 to December 31, 2012, and (3) a survey of all VHA primary care staff in June 2012. Composite scores were constructed for 8 core domains of PACT: access, continuity, care coordination, comprehensiveness, self-management support, patient-centered care and communication, shared decision making, and team-based care. MAIN OUTCOMES AND MEASURES Patient satisfaction, rates of hospitalization and emergency department use, quality of care, and staff burnout. RESULTS Fifty-three items were included in the PACT Implementation Progress Index (Pi2).
Compared with the 87 clinics in the lowest decile of the Pi2, the 77 sites in the top decile exhibited significantly higher patient satisfaction (9.33 vs 7.53; P < .001), higher performance on 41 of 48 measures of clinical quality, lower staff burnout (Maslach Burnout Inventory emotional exhaustion subscale, 2.29 vs 2.80; P = .02), lower hospitalization rates for ambulatory care–sensitive conditions (4.42 vs 3.68 quarterly admissions for veterans 65 years or older per 1000 patients; P < .001), and lower emergency department use (188 vs 245 visits per 1000 patients; P < .001). CONCLUSIONS AND RELEVANCE The extent of PCMH implementation, as measured by the Pi2,
was highly associated with important outcomes for both patients and providers. This measure will be used to track the effectiveness of implementing PACT over time and to elucidate the correlates of desired health outcomes.
JAMA Intern Med. 2014;174(8):1350-1358. doi:10.1001/jamainternmed.2014.2488 Published online June 23, 2014. 1350
Author Affiliations: Author affiliations are listed at the end of this article. Corresponding Author: Karin M. Nelson, MD, MSHS, Seattle Center of Innovation for Veteran-Centered and Value-Driven Care, VA Puget Sound Health Care System, 1100 Olive Way, Ste 1400, Seattle, WA 98108 (karin [email protected]). jamainternalmedicine.com
Copyright 2014 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ by a Ndsu Library Periodicals User on 05/18/2015
Patient-Centered Medical Home Implementation
A
lthough the patient-centered medical home (PCMH) has been endorsed by most major primary care groups as a promising model to strengthen primary care, decrease costs, and improve quality,1 early assessment of PCMH impact have yielded mixed results.2-10 Since 2010, the Veterans Health Administration (VHA) has undertaken national adoption of a PCMH model, called PACT (Patient Aligned Care Team).11 The focus of PACT has been to restructure primary care to provide team-based care that is more comprehensive, coordinated, and patient centered.11 The PACT initiative is a multifaceted and complex intervention, creating challenges to measuring implementation across diverse clinic sites. One of the most widely recognized PCMH recognition tools is the National Committee for Quality Assurance (NCQA) certification process, which focuses on practice infrastructure and health information technology,12 an area in which the VHA has made considerable past investments.13,14 The VHA has a universally deployed electronic health record, electronic prescribing, patient registries, and a national quality improvement and performance measurement infrastructure for which all clinics in the VHA would receive “credit.” Many national programs for coordinating care, such as home-based primary care, integrated mental health services, and palliative care, were already widely available before PACT was initiated. The focus within the VHA has been on how effectively these extensive resources are being applied and coordinated to fulfill the goals of the PACT initiative.11 Our goal was to derive a comprehensive index from existing data and survey instruments that would have a low respondent burden and would reflect processes and attributes that are essential to effective primary care. Our approach differs from other PCMH measurement tools15 by incorporating multiple data sources, including a primary care personnel survey, patient surveys, and administrative data. We sought to develop a measure to represent areas of focus of the PACT initiative, including continuity through team-based care, patient access, care coordination, and patient-centered care.11 We desired an instrument that would facilitate comparisons across clinical sites within the VHA, assist in identifying sites that had most effectively implemented PACT, and determine the relationship between effective implementation and important outcomes, such as patient satisfaction, quality of care, provider experience, and use of health care services.
Methods
(0.61).16 To test convergent validity, we used information on patient satisfaction from another sample of veterans from the Survey of the Health Experiences of Patients, an ongoing national mailed US Department of Veterans Affairs (VA) survey that assesses the health care experiences of veterans who receive care at the VHA and uses a stratified random sampling method.17 The evaluation efforts are part of an going quality improvement effort at the VHA and are not considered research activity; they are thus not subject to institutional review board review or waiver. Primary Care Personnel Survey The PACT Primary Care Personnel Survey was an internally developed instrument designed to measure team functioning in PACT and has been described elsewhere.18 The target population of the survey was all VHA primary care personnel, including the 4 occupations included in PACT teams: primary care providers, nurse care managers, medical associates (eg, licensed practical nurses and medical technicians), and administrative clerks. Data were collected from May 21 through June 29, 2012. Team-based care was represented by items from the primary care personnel survey related to delegation, staffing, team functioning, and team assignment.18 Administrative Data Information about demographics, clinical characteristics, and use of health services was obtained from the VHA Corporate Data Warehouse for fiscal year 2012 (n = 5 653 616). Using data from the Primary Care Management Module contained within the Corporate Data Warehouse, we identified all patients who were enrolled in primary care and assigned to a primary care provider.19 We included administrative data for important PACT programmatic goals,11 including (1) access to care and use of non–face-to-face care, such as telephone clinics and secure messaging; (2) continuity of care; and (3) use of VHA programs to support care coordination (eg, home telemonitoring, 2-day posthospital follow-up). We used data collected by the VHA External Peer Review Program (EPRP) during fiscal year 2012 to assess quality of care. The EPRP is an audit program designed to assess clinical performance using standard performance criteria. National data are collected through manual abstraction of electronic health records by an independent external contractor.20 Previous studies have found high interrater reliability (κ = 0.9) within the EPRP program.14
Construction of the PACT Implementation Progress Index
Survey Instruments and Data Sources Patient Survey We used data from the previously validated Consumer Assessment of Health Plans–Patient Centered Medical Home (CAHPS PCMH) survey16 that was administered to a nationally weighted random sample of veterans who received outpatient care from June 1 to December 31, 2012. The CAHPS PCMH scales have acceptable internal consistency reliability estimates for access (Cronbach α, 0.74), comprehensiveness (0.68), self-management support (0.62), patient-centered care and communication (0.91), and shared decision making jamainternalmedicine.com
Original Investigation Research
The method for developing the PACT Implementation Progress Index (Pi2) and a full description of all items are provided in eTable 1 in the Supplement. Briefly, we mapped data items to PACT conceptual domains, calculated domain scores based on these items, and then generated site-level rankings for each domain. Table 1 outlines Pi2 domains and provides examples of representative variable items. A Pi2 score was assigned to each clinic based on the number of domains in the top and bottom quartiles for the domain scores, ranging from 8 (all domain scores in the top quartile) to –8 (all domain scores in the bottom quartile). Using these scores, we categorized sites in JAMA Internal Medicine August 2014 Volume 174, Number 8
Copyright 2014 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ by a Ndsu Library Periodicals User on 05/18/2015
1351
Research Original Investigation
Patient-Centered Medical Home Implementation
Table 1. PCMH Concepts and Pi2 Domains Items From Each Data Source, No. PACT Initiative Goals Accessible, continuous, and coordinated care
Patient-centered care
Team-based care
Total Items, No.
CAHPS PCMHa
CDWb
11
6
5
0
How often did you get an appointment as soon as you needed?a When you phoned this provider’s office, how often did you get an answer to your medical question that same day?a Same-day access to appointments (% of clinics)b Enhanced access: telephone clinics (%)b
Continuity of care
3
1
2
0
How long have you been going to this provider?a Proportion of visits to assigned primary care providerb
Care coordination
7
5
2
0
When this provider ordered a test, how often did someone from this provider’s office follow up to give you those results?a Did the provider seem informed and up to date about the care you got from specialists?a Percentage of patients contacted 2 d after hospital dischargeb
Comprehensiveness
3
3
0
Did you and anyone in this provider’s office talk about things in your life that worry you or cause you stress?a
Self-management support
2
2
0
Did anyone in this provider’s office talk with you about specific goals for your health?a
Patient-centered care and communication
6
6
0
How often did this provider explain things in a way that was easy to understand?a
Shared decision making
3
3
0
When you talked about starting a prescription medicine, did this provider ask you what you thought was best for you?a
Delegation, staffing, and team functioning
18
0
0
18
Primary care provider relies on registered nurse care manager for tasks including gathering patient preventive services, responding to prescription refillsc Percentage reporting recommended staffing ratioc Time spent in team huddles: percentage spending >30 min/dc
…
53
26
9
18
…
Pi2 Domains Access
Total
Abbreviations: CAHPS PCMH, Consumer Assessment of Health Plans–Patient Centered Medical Home; CDW, Corporate Data Warehouse; PACT, Patient Aligned Care Team; PCMH, patient-centered medical home; Pi2, PACT Implementation Progress Index. a
CAHPS PCMH patient survey.
the top decile of the Pi2 (score, 5 to 8) as having achieved effective implementation and those in the lowest decile of the Pi2 (score, –7 to –5) as having been less effective.
Patient- and Provider-Level Outcome Measures Patient satisfaction was assessed by using a single item from the CAHPS PCMH survey16 as follows: “Using any number from 0 to 10, where 0 is the worst provider possible and 10 is the best provider possible, what number would you use to rate this provider?” Staff burnout was assessed with both a single-item measure and the emotional exhaustion subscale of the Maslach Burnout Inventory, a widely used measure of burnout.21-23 The single item measure asks: “Overall, based on your definition of burnout, how would you rate your level of burnout,” with 5 ordinal response options.24 We defined burnout as a response of 3 or higher, where 3 corresponds to “I am definitely burning out and have one or more symptoms of burnout, such as physical and emotional exhaustion.”24 We also used a 3-item version of the Maslach Burnout Inventory subscale.25 Items reflecting burnout symptoms are scored using a Likert scale ranging from 0 (never) to 6 (every day) and summed to form a scale score. We defined burnout as a score of 10 or higher (range, 0-18). To assess quality of care, we examined outpatient measurements from the EPRP for chronic disease management, behavioral health screening, and prevention services. These in1352
Personnel Surveyc Example Itemd
b
CDW data from US Department of Veterans Affairs.
c
PACT Primary Care Personnel Survey.
d
Full description in eTable 1 in the Supplement. The questions have been shortened for the sake of brevity.
dicators include frequently used measures of the quality of prevention (eg, vaccinations, screening tests) and outpatient care of chronic diseases (eg, annual retinal examinations in patients with diabetes mellitus). The performance measurement for preventive services and chronic disease cohorts, sampling frame, and criterion for meeting the measurement are provided in eTable 2 in the Supplement. The EPRP selects a random sample of patient records from VHA facilities to monitor quality and appropriateness of medical care.26 The sample includes veterans who used VHA health care at least once in the 2 years before the assessment. Patients who were sampled had at least 1 primary care or specialty medical visit in the month being sampled. Among eligible patients, a random sample is drawn with oversampling of prevalent chronic conditions (eg, diabetes, heart failure).26 For patients at each primary care site, we determined the numbers of emergency department or urgent care visits, VA hospital admissions, and hospitalizations for ambulatory care– sensitive conditions (ACSCs), which are postulated to be most avoidable through provision of effective primary care.27 Hospitalizations for ACSCs were based on Agency for Healthcare Research and Quality Prevention Quality Indicators and were identified through standardized protocols using International Classification of Diseases, Ninth Revision, diagnoses and Current Procedural Terminology codes from inpatient VA records.27
JAMA Internal Medicine August 2014 Volume 174, Number 8
Copyright 2014 American Medical Association. All rights reserved.
Downloaded From: http://archinte.jamanetwork.com/ by a Ndsu Library Periodicals User on 05/18/2015
jamainternalmedicine.com
Patient-Centered Medical Home Implementation
Statistical Analysis PI2 Properties To test internal consistency reliability, we calculated the Cronbach α for all items in each domain and all 53 items that make up the total scale. Variation in PCMH Adoption We evaluated bivariate comparisons of facility characteristics and level of implementation by using χ2 tests for categorical variables and t tests for continuous variables. We compared sites assessed to have effectively implemented PACT with those assessed as less effective according to type of facility (hospital or community-based outpatient clinic), number of patients, demographic characteristics, and Elixhauser comorbidity score.28 Associations With Patient and Provider Outcomes We used a nonparametric test of trend for the ranks across ordered groups (an extension of the Wilcoxon rank sum test) to test for trends in patient satisfaction and staff burnout by Pi2 scores. We tested differences in the proportions of eligible patients at each VHA clinic fulfilling each of the 48 quality indicators according to the success of PACT implementation as measured by the Pi2. We calculated rates of services at the facility level by dividing the number of patients who satisfied the EPRP quality measure by the number who met inclusion criteria for each quality measure (eTable 3 in the Supplement). For each of the 48 facility-level quality indicators, we tested the trend in proportions of patients fulfilling the EPRP quality guideline by the level of PACT implementation. We used the nonparametric test for trend developed by Cuzick, which is an extension of the Wilcoxon test.29 We adjusted for multiple comparisons using a method described by Benjamini and Yekutieli.30 To determine whether more effective implementation (as measured by Pi2) corresponded to higher performance overall, we included all 48 outcome measures in a linear mixed-effects model that accounted for correlation among outcomes from the same facility and estimated an overall implementation effect. We adjusted for implementation in this model as a linear term ranging from 1 to 5, corresponding to the grouped Pi2 scores. This approach was possible because all 48 outcomes were measured on the same scale. We examined fiscal year 2012 emergency department and urgent care visits and total hospitalizations for sites with more effective vs less effective implementation, adjusting for patient age, community-based outreach clinics, and Elixhauser comorbidity scores.28 To account for temporal trends, we modeled facility-level trends for hospitalization from 2003 to 2012. The method for examining such trends has been described elsewhere.31 We estimated interrupted time-series models of ACSC and all-cause hospitalizations from October 1, 2003, through September 20, 2012, for each facility and assessed how the trends in hospitalizations changed after the start of the PACT initiative in April 2010. All regression models adjusted for facility-level patient characteristics, unemployment rate in the VA market area, quarterly dummy variables to capture seasonal variation, and a linear time trend. Patient risk was measured using mean facility-level Elixhauser comorbidity scores. jamainternalmedicine.com
Original Investigation Research
Changes in admissions for ACSC and all-cause hospitalizations after implementation of the PACT initiative were calculated as the difference between the observed rate of admissions and the predicted rate had the initiative not been implemented during the 2½-year period between April 1, 2010, and September 30, 2012. In this way, we estimated changes in admissions that might be attributed to the PACT initiative. Trend analyses for hospitalizations were stratified by age (≥65 and
