Abstract
Objectives
To examine how a targeted six-month interventions impacted Best Practice/Patient Outcomes for minority patients receiving primary care in physician practices participating in a pay-for-performance (P4P) program.
Methods
P4P Practices were invited to participate in a pilot intervention study designed to improve care for minority patients with hypertension, diabetes or pediatric asthma. Patient medical records were reviewed to assess how the interventions impacted (n=7 practices): Body mass index, diet and exercise, smoking, compliance with visits as recommended, blood pressure, sodium intake and weight management counseling, medication reconciliation, HbA1c testing, annual lipid profile, and anti-inflammatory medications.
Results
Significant improvements in various clinical quality measures were observed in all seven practices. Of the 19 specified interventions, 13 were statistically significant at α=0.05 level and 14 met the target proportion. This suggests that the best practice intervention had a significant impact on some of the health care processes in the physician practices.
Conclusions
The most impactful interventions were those related to face-to-face educational discussions, patient medical chart documentations rather than those pertaining to medication adherence. Improvements in measuring reporting and recording of data at post-intervention were also observed.
Keywords: Pay-for-Performance, Healthcare Health Disparities, BMI and Diabetes, Primary Care Physicians, Smoking
INTRODUCTION
The 2011 National Healthcare Disparities Report documents that racial and ethnic disparities remain pervasive [1]. A decade prior, in 2001, the Institute of Medicine released the Crossing the Quality Chasm report that recommends incentives to encourage quality health care and better health outcomes [2]. To address these concerns, numerous private-sector initiatives to link portion of payments made to physicians and hospitals dependent on improved quality are underway. These initiatives are commonly referred to as pay-for-performance programs (P4P) [2]. P4P is broadly defined to include any type of performance-based provider payment arrangement, including those that target performance on cost measures [2]. The primary goal of P4P is rewarding or improving quality of care; a secondary goal is controlling cost by reducing errors and inappropriate utilizations [2]. P4P is only one of many proposed strategies that may improve quality and thus positively impact health care disparities [3-6]. A proposed strategy for addressing health disparities is engagement [7] as defined by general awareness, reflection/empowerment, and cues to action within their own practice and their communities in general.
It is well-recognized [3-4,7-8,10-12] that there are both advantages (increased payment, improved efficiency improved care quality, especially among disparate populations) and disadvantages (cost of acquiring information technology, multiple programs and guidelines, data collection) to P4P programs currently designed and implemented as reported by the American Academy of Family Physicians (AAFP) [8]. One challenge associated with P4P is its potential to increase disparities if some groups (like medically underserved groups) are less able to respond to specific interventions when compared to others [9]. Another reported challenge is to implement P4P in the small practice setting [2,4,8], which experience substantial barriers like limited staff and fewer resources than larger group practices, thus increasing the difficulty completing quality improvement projects [2,4,8]. According to AAFP there are seven P4P guiding principles that programs should adhere to [8]: (1) focus on improved quality of care; (2) support the patient relationship; (3) utilize performance measures based on evidence-based clinical guidelines; (4) involve practicing physicians in program design; (5) use reliable, accurate, and scientifically valid data; (6) provide positive incentives; and (7) offer voluntary participation. Highmark's Quality Blue Physician P4P program incorporates all of these principles. Specifically, the program recognizes and rewards primary care physicians (PCP) for collaborating with Highmark to improve the quality and effectiveness of health care services provided to our members.
Highmark has been engaged in both focused disparity related activities as well as P4P for several years. These efforts have been published elsewhere [16-18]. Notably, Highmark's health disparities work has been recognized in the National Health Plan Collaborative (NHPC) Toolkit. NHPC was a novel public-private partnership between eight health plans covering approximately 95 million lives. Sponsors included the Agency for Health Care Research and Quality, and the Robert Wood Johnson Foundation.
The Quality Blue Physician P4P program is offered in 49 counties of western and central Pennsylvania with approximately 6,300 primary care physicians (PCP) in more than 1,500 practices, providing services to more than 1,731,000 unique Highmark members. The primary care physician network is composed of 2,448 practices; 1,604 or 66 percent of all primary care practices participate in the Quality Blue program.
The Best Practice component of Highmark's Quality Blue Physician P4P Program rewards practices for conducting a population based clinical initiative in their office setting or for completing a professional organization certification/recognition activity. The primary goal of the Best Practice component is to improve patient care offered by requiring practices to apply knowledge and experience gained via Deming's Plan-Do-Study-Act methodology. The practices explore opportunities for clinical-based activities, thereby empowering them to integrate improvement models into daily office activities and promote proactive rather than reactive care. The Best Practice studies carried out by practices have varied over the years and have included some of the following improvements:
Improving Diabetic LDL and HbA1c control;
Reviewing medications with patients to address medication adherence;
Conducting spirometry testing to assess for asthma and COPD;
Counseling and other interventions to address childhood obesity; and
Improving rates of colorectal cancer screening.
Although P4P strategies are increasingly common, little is known about the impact of applying this payment method to reducing healthcare disparities (HCD) in general or by geography (urban and rural areas), including Pennsylvania [13]. In this study, we hypothesize that existing P4P programs have the potential to decrease disparities in minority groups. This study examined the impact of various interventions on the Best Practice measures: body mass index (BMI), diet and exercise, smoking, compliance with visits as recommended, measuring blood pressure, sodium intake and weight management counseling, medication reconciliation, HbA1c, lipid profile, and anti-inflammatory medications among participating network practices utilizing a P4P approach in the Pennsylvania counties in Highmark's service region.
METHODS
Participating Practices
Although all PCP practices that participate in the Quality Blue Physician P4P program are encouraged to conduct a Best Practice study, practices identified as serving sizable minority populations (e.g. African American and Latino/Hispanic) were recruited to participate in the pilot study. Highmark's Clinical Quality Consultants (CQC) and Provider Relations staff met with each of the practices to inform them of the pilot and to gauge their interest in participating. Efforts were made to recruit practices where African American and Hispanic Highmark members with known cardiovascular and diabetes disparities received their primary care. Since previous data analysis had identified disparities in pediatric asthma clinical indicators, efforts were also made to recruit pediatric practices that served minority children.
All PCPs serving minority (African American and Latino/Hispanic) patients previously identified with disparities were identified by Highmark trained staff in April 2010. These PCPs were identified through a query report matching all minority patients with known health care disparities in diabetes and cardiovascular disease HEDIS indicators with their Primary Care Physician practices who were participating in the Quality Blue Physician P4P program From these PCPs, practices were identified that included a total of at least five eligible minority patients. The five- patient criterion was used to allow sufficient numbers of minority patients for the statistical evaluation of the pilot and to help identify practices that may be more willing to participate. Figure 1 illustrates how the practices were selected for the pilot study. Of the 515 practices identified from the eligible PCPs, only 29 met the minimum 5 patient criterion. In addition, 30 practices were identified by Highmark staff between April – August 2010 that were not included in the query report but were recommended as having a larger volume of minority patients with hypertension or pediatric asthma. Thus a total of 59 practices were available for the pilot. Of those, 50 practices met the eligibility criteria of having previously participated in the best practice option. Of the 50 practices, 25 chose to participate in the pilot and among those a total of seven completed the pilot study. The Allegheny Singer Research Institute's Institutional Review Board approved this study.
Figure 1.
Data Flow for Pay-for-Performance Best Practices/Patient Outcomes for Minority Patients
Study Design
Best practice interventions were decided within each of the seven participating practices. Therefore, each practice participated in different interventions. The clinical Best Practices were measured at the end of each month for six consecutive months. A random sample of medical charts of minority patients’ with the identifiable healthcare disparity was selected to audit for evidence of the intervention specific measurement. The baseline measurement was taken before the start of the intervention and was compared with results at month six to evaluate the effectiveness of the various interventions within each practice. Interventions are described in the left most column of Table 1.
Table 1.
Practice Demographics
| PRACTICE | TYPE OF PRACTICE |
NUMBER OF PRACTICE SITES |
REGION | LOCATION DESCRIPTION |
NUMBER OF PHYSICIANS |
PHYSICIAN SPECIALTY |
PATIENT AGE |
ELETRONIC MEDICAL RECORDS (EMR) &/or Electronic Prescribing (erx) |
EVENING & WEEKEND HOURS |
LANGUAGES SPOKEN BY PHYSICIANS |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Hospital owned Multi-specialty | 2 | Southwestern | Urban located inside two teaching hospitals | 6 Physicians | Emergency Medicine Internal Medicine Pediatrics |
6 – 100+ Years | eRx | No | English Spanish Polish Castilian |
| 2 | Independent Solo Practitioner | 2 | Southwestern | Suburban | 1 Physicians | Family Practice | 6 – 100+ Years | NNo | No (E) Yes (W) |
English Spanish Castilian |
| 3 | Hospital owned Multi-specialty | 1 | Northwestern | Urban | 5 Physicians 1 Nurse Practitioner |
Family Practice Orthopedic Surgery Neonatology Podiatry Pediatrics |
0 – 100+ Years | EEMR eRx |
Yes | English Hindi Telugu |
| 4 | Independent | 1 | Northwestern | Urban | 2 Physicians | Family Practice Ob-Gynecology |
0 – 100+ Years | EEMR eRx |
No | English |
| 5 | Independent | 1 | Northwestern | Rural | 2 Physicians 1 Physician Assistant |
Family Practice Pediatrics |
6 – 100+ Years | EEMR eRx |
No | English |
| 6 | Independent | 1 | Central | Rural | 6 Physicians 1 Physician Assistant |
Family Practice General Practice |
0 – 100+ Years | EEMR eRx |
No | English |
| 7 | Independent | 1 | Central | Urban | 3 Physicians 1 Physician Assistant |
Family Practice General Practice |
0 – 100+ Years | EEMR eRx |
No | English Philippine Spanish Castilian |
Statistical Analyses
A two-tailed, two-sample test of proportions was performed, which compares the baseline results (proportion of patients satisfying the BPPO measure) for each BPPO with the final results at month six. The baseline and final samples were treated as independent samples. A statistically significant result is indicated if the p-value is less than or equal to 0.05. Due to the exploratory nature of this pilot no adjustments were made to account for multiple comparisons. It is possible that some of the subjects were measured at both baseline and month six; in this scenario, the independent two-sample test of proportions will likely yield a conservative p-value. Additionally, most of the outcomes are measures of documentation rate of the practice which would be unaffected by the particular patient being documented. As such, even if patients did overlap in the samples the resulting pre and post documentation rates of the practice would be independent of each other. Practices were identified if it met or exceeded the target proportion. Targets were specified separately by each practice.
RESULTS
Practice Demographics
Of the 50 eligible intervention practices that met the requirements for this study, only 25 chose to participate with seven completing the P4P program. As shown in Table 1, two of the seven practices were hospital-owned, multi-specialty, and the remaining five were independently owned. Four practices were located in urban Pennsylvania while two were in rural areas; one was located in a suburban area. Four of the seven practices had physician(s) reporting that he/she spoke more than one language. Most practices offered only weekday hours. Only one practice reported not having an electronic medical record (EMRs) system.
Patient Demographics
Table 2 shows a description of the patient population of the participating practices. Three geographic regions were represented across practices: Southwestern, Northwestern, and Central Pennsylvania. In general, there were equal representation of both males (48%) and female (52%) patients across practices, but there was slightly more females. Patients were heterogeneous relative to race/ethnicity across all practices. Approximately one in five patients in Practices 1, 3 and 4 were black patients. For practices 6 and 7, one in three patients was Hispanic. The average income level was $33,000 (range: $28,197 to $39, 529).
Table 2.
Patient Demographics for Each of the Seven Practices
| PRACTICE | GEOGRAPHIC REGION | GENDER DISTRIBUTION | RACE | AGE DISTRIBUTION | MEDIAN HOUSEHOLD INCOME | EDUCATION LEVEL FOR ≥ 25 YEAR OLD |
|---|---|---|---|---|---|---|
| 1 | Southwestern Pennsylvania | 52% Female 48% Male |
66% White 26% Black 4% Asian 2% Hispanic 3% Other |
0 - 19 = 21% 20 - 64 = 65% 65 + = 14% |
$28,588 | 88% = high school graduates 34% = Bachelors degree |
| 2 | Southwestern Pennsylvania | 54% Female 46% Male |
91% White 5% Black 1% Asian 2% Hispanic 1% Other |
0 - 19 = 23% 20 - 64 = 59% 65 + = 18% |
$39,529 | 95% = high school graduates 33% = Bachelors degree |
| 3 | Northwestern Pennsylvania | 52% Female 48% Male |
72% White 17% Black 2% Asian 7% Hispanic 2% Other |
0 - 19 = 28% 20 - 64 = 60% 65 + = 12% |
$32,218 | 85% = high school graduates 20% = Bachelors degree |
| 4 | Northwestern Pennsylvania | 52% Female 48% Male |
72% White 17% Black 2% Asian 7% Hispanic 2% Other |
0 - 19 = 28% 20 - 64 = 60% 65 + = 12% |
$32,218 | 85% = high school graduates 20% = Bachelors degree |
| 5 | Northwestern Pennsylvania | 54% Female 46% Male |
93% White 3% Black 1% Asian 1% Hispanic 4% Other |
0- 19 = 24% 20 - 64 = 57% 65 + = 19% |
$34,732 | 91% = high school graduates 21% = Bachelors degree |
| 6 | Central Pennsylvania | 52% Female 48% Male |
62% White 4% Black 1% Asian 30% Hispanic 3% Other |
0 - 19 = 29% 20 - 64 = 57% 65 + = 14% |
$37,438 | 76% = high school graduates 9% = Bachelors degree |
| 7 | Central Pennsylvania | 52% Female 48% Male |
48% White 13% Black 1% Asian 37% Hispanic 1% Other |
0- 19 = 36% 20 - 64 = 54% 65 + = 10% |
$28,197 | 65% = high school graduates 10% = Bachelor's degree |
Best Practice/Patient Outcomes
As shown in Table 3, most of the differences in the proportion of patients satisfying the BPPO measure between the baseline, which was measured before the start of the intervention, and month six were statistically significant. Of the 19 specified interventions, 13 were statistically significant at α=0.05 level, and 14 met the target proportion. This suggests that the best practice intervention had a significant impact on some of the health care processes in the physician practices.
Table 3.
Final Reporting and Re-measurement of Minority Patients of a Seven Practice Pilot in a Pay For Performances ( P4P)-Based Program in Pennsylvania
| Best Practice Intervention | Variable Assessed from Patient Medical Chart | Total patients participated: n | Baseline measurement: x1, p1a | Month Six measurement (final): x2, p2a | p-vaue for Ha: p1≠p2 | Targatb: p | Did final measurement meet or exceed target? |
|---|---|---|---|---|---|---|---|
| Practice 1 | |||||||
| BMI Measured within parameters - follow up plan for healthy habits | Both Documented | 19 | 3, 15.8% | 19, 100% | <0.00001* | 85% | Yes |
| BMI Measured outside parameters- exercise & eating plan developed | Both Documented | 19 | 3, 15.8% | 19, 100% | <0.00001* | 85% | Yes |
| Patient adherence to diet and exercise documented | 19 | 3, 15.8% | 16, 84.2% | <0.00001* | 85% | No | |
| Screen for tobacco use | Assessed Documented | 19 | 14, 73.7% | 19, 100% | 0.017* | 95% | Yes |
| Counseling Documented | 19 | 14, 73.7% | 19, 100% | 0.017* | 95% | Yes | |
| Patients that quit Documented | 14 | 6, 42.9% | 10, 71.4% | 0.128 | 75% | No | |
| Compliance with at least 2 visits or as recommended | Following initiation of TX uncontrolled/ Seen every 3mo | 19 | 11, 57.9% | 16, 84.2% | 0.074 | 85% | No |
| Controlled BP – seen every 6 months | 4 | 4, 100% | 4, 100% | 1.00 | 100% | Yes | |
| BP Control < 140/90 | Documented Measurement | 19 | 4, 21.1% | 17, 89.5% | <0.00001* | 85% | Yes |
| Practice 2 | |||||||
| Hypertension Dietary/Weight Management | Sodium Restriction discussion documented | 20 | 0, 0% | 20, 100% | <0.00001* | 80% | Yes |
| Weight reduction/management discussions documented | 20 | 0, 0% | 20, 100% | <0.00001* | 80% | Yes | |
| BMI documented | 20 | 20, 100% | 20, 100% | 1.00 | 100% | yes | |
| Hypertension Medication Management | Medication reconciliation documented | 20 | 0, 0% | 20, 100% | <0.00001* | 80% | Yes |
| Practice 3 | |||||||
| Hypertension Medication Management | Medication reconciliation documented | 40 | 1, 2.5% | 35, 87.5% | <0.00001* | 85% | Yes |
| Practice 4 | |||||||
| Diabetes Management HbA1c Testing Rate | HbA1c Testing documented | 23 | 5, 21.7% | 7, 30.4% | 0.502 | 40% | No |
| Practice 5 | |||||||
| Hypertension Management Lipid Profile Testing | Annual Lipid Profile documented | 20 | 12, 60.0% | 18, 90.0% | 0.019* | 90% | Yes |
| Practice 6 | |||||||
| Persistent Pediatric Asthma treatment | Documentation of Anti-Inflammatory Meds prescribed | 30 | 27, 90.0% | 28, 93.3% | 0.644 | 95% | No |
| Documentation of Spacer device prescribed | 48 | 21, 43.8% | 35, 72.9% | <0.00001* | 50% | Yes | |
| Practice 7 | |||||||
| Management of Pediatric Asthma | Asthma Education and parent(s) verbalized understanding documented | 9 | 2, 22.2% | 9, 100% | 0.0007* | 95% | Yes |
p-value is significant at α=0.05 level
p1=x1/n and p2=x2/n
Target p was determined by each practice practitioner separately for each intervention.
For example, in Practice 1, only three of the 19 patients had their BMI measured and documented before the start of the intervention. At the end of the intervention, all 19 of the patients selected in month six had their BMI measured and documented. The most impactful interventions were those related to education, discussions, and patient medical documentations rather than those pertaining to medication adherence.
DISCUSSION
We examined the change in selected Best Practice measures among participating practices utilizing a P4P approach in the Pennsylvania counties in Highmark's service region. We found significant improvements in blood pressure measurement and documentation (from 15.8% to 100%); exercise & eating plan documentation (from 15.8% to 100%) and patient adherence to diet and exercise (from 15.8% to 84.2%); smoking assessment and counseling provided (from 73.7% to 100% for each parameter); sodium and weight discussion (from 0 to 100% for each parameter); medication reconciliation ( from 2.5% to 87.5%); annual lipid profile testing (from 60% to 90%); and asthma education (from 22.2% to 100%) (see Table 3). While there was no formal comparison group, the magnitude of these findings suggest that P4P had a significant impact on some of the physician practices. This study supports the findings of others reporting the use of P4P as a proposed strategy that may improve quality and thus positively impact health care disparities [1-6].
Similar to findings reported by Coleman14, improvements were observed in diabetes for our patients that tested their HbA1c (from 21.7% to 30.4%). However, these changes were not statistically significant. Coleman's study showed that the P4P program significantly increased the likelihood that patients received two HbA1c scores per year as recommended by the American Diabetes Association, but do not contribute to improved blood sugar control [14]. Though we only measured HbA1c testing rate and did not directly measure diabetes control parameters, we do agree with Coleman's reporting that it appears that P4P programs may help with improvement in compliance with HbA1c testing recommendations, however, a more comprehensive strategy [14] may be necessary to better understand the link between P4P and diabetes control. One proposed strategy presented by Coleman is to include a patient social support [14] component in the P4P approach. This may be a reasonable starting point for examining this link as minorities report lack of social support as a barrier to their health care behaviors (cues to action), including compliance [14].
Because in our study the most impactful interventions were those related face to face educational discussions and patient medical chart documentations rather than those pertaining to medications, we propose that strategies for examining this particular link also includes an education, discussion, and patient medical documentation component. Whether these components should be tested alone or in combination needs further research.
Strengths
There are five notable strengths to this study. First, this study included feedback from a disparate population that could benefit from improved health care quality. Second, this study builds upon Highmark's existing P4P Program capacity to focus on disparity related activities and provides a catalyst for extending its Best Practice component to address health care disparities among minority patients residing in urban and rural areas in Pennsylvania. Thirdly, this study provided an opportunity for one health insurer and several local practices serving minority patients to work together and identify potential strategies aimed at rewarding or improving quality of care. Another strength of the pilot is the absence of selection bias. Because the P4P incentive was based solely on completing the intervention and not on the results, there was no pre-selection of patients within each practice.
Lastly, all doctors and staff were required to complete a cultural competency course as a prerequisite for the participating in the pilot study. Completion of the course was optional but encouraged for nurses, physician assistants and other staff at the practices. The course, Quality Interactions®, is a proprietary e-learning program that provides case-based instruction on cross-cultural health care. This interactive program focuses on common clinical and/or cross-cultural scenarios that build a framework of knowledge and skills for delivering quality care to diverse patient populations. A paired t-test analysis was performed on the pre- and post- intervention test scores for the nurses, doctors and physicians’ assistants. The results were all statistically significant at α=0.001 level, suggesting that the course improved their cultural awareness within and around the practice.
We attribute the statistically significant improvements observed in the Best Practices measures following intervention to our P4P program's study design, which was informed by the literature. For example, our P4P program focused on improving quality of care as defined by the selected biological and behavioral endpoints at baseline and monthly assessments over a six-month period. Our program supported the physician/patient relationship by providing a direct and open-line of communications between our Highmark staff and each practice as needed. Our program utilized performance measures based on evidence-based clinical guidelines. It also included ongoing input during each phase of the study from Highmark's CQC and Provider Relations staff. One of the primary roles of Highmark's CQC staff is to act as practice coaches educating and guiding primary care physicians and their staff through quality improvement and practice transformation activities throughout the year. Staff from Highmark's Health Equity and Quality Services (HEQS) department collaborated on the pilot. The pilot physician and nurse participants were given access to a complimentary E-learning Quality Interactions Continuing Medical Education (CME) course and other cultural competency educational resources and the tools to implement the healthcare disparities activities. These resources remained available to network providers post-intervention.
Based on the opinion of and feedback from these intervention (pilot) practices, this P4P program can be easily adopted by other health plans provided they have an engagement strategy in place to support the spread of evidence based guidelines, process and quality improvement activities and a commitment to eliminating health care disparities. If true, then this Quality Blue P4P Best Practice program can potentially be used as a template for practices aiming to reduce racial and ethnic disparities in health care among minority patients with diabetes, pediatric asthma and/or hypertension.
Results of this study may also help to inform next steps for public health initiatives aiming to improve patient medical chart documentations. Reported strategies include [21]: (1) to focus on how proper documentation can improve quality, have impact is important. This may help to encourage and provide added reasons for physicians to improve their medical documentation; (2) to provide training opportunities for the practices. Medical documentation is mostly the responsibility of a physician and medical staff, and thus across-the-board trainings become increasingly important. This may help provide consistency in documentations across staff members; and (3) to improve existing coding structures used by a practice. Review of coding structures should be revisited on an ongoing basis. This may help improve accuracy or consistency of claims data that is needed when codes change and/or do not provide the necessary level of detail. It is well documented that targeted interventions yield better outcomes when compared to non-targeted approaches. If true, then future studies examining effective strategies for improving patient medical documentation by physician specialty, practice size and geography, patient's race/ethnicity, gender, and staff trainings options to name a few are warranted.
Limitations
Three limitations must be noted. First, this was a one-time assessment of seven of 50 potential practices in Highmark service areas. Thus the patients who participated in this study might not have been representative of all practices, including those serving outside the Highmark service area. However, many of these health professionals had experience working within and outside Highmark and stated that they had no reason to believe that the P4P approach used would not work in other practices serving minority patients. We did not assess whether their minority patients were any different compared to other minority patients. This was outside the scope of this study. Second, of 25 practices that committed to participation, only seven completed the pilot study. One explanation for the low completion rate by practices was due to internal changes/dynamics that were not foreseen. A major programming change in the Quality Blue P4P program requirements and scoring was implemented in 2011. The focus of the Quality Blue P4P component shifted to support Meaningful Use and Patient Centered Medical Home activities. We do not believe that this reason is associated with any of our measured outcomes. As such, the low completion rate does not pose a statistical concern of biased results.
A great deal of the CQC and Provider Relations staff focus was devoted to communicating the program changes to network PCP practices during the latter part of the year. This may have contributed to practices who had originally committed to the disparity pilot to withdraw; or, they may have selected a different best practice study activity instead of ours. Also, some practices that we recruited for the disparity pilots were not due to start a new best practice until 2012 and they may have decided not to complete a best practice. This may also be due to structural features at practices that are associated with participation and may provide leverage points for (or for not) fostering participation as reported by others [20]. Future studies assessing barriers to self-enrolled participation completion rates among practices participating in P4P programs are warranted.
Lastly, this was a onetime assessment feedback about how to reduce health disparities among minority populations using a P4P approach. Whether health disparities can be further improved with continuous or repeated cyclical feedback from practices is unknown, but is worthy of further study.
CONCLUSION
This was a pilot study of participating networks utilizing a P4P approach in the Pennsylvania counties in Highmark's service region. The results of this study showed that there were significant improvements in the selected Best Practice measures. Of the 19 specified interventions, 13 were statistically significant at α=0.05 level and 14 met the target proportion. The most impactful interventions were those related to education, discussions, and patient medical documentations rather than those pertaining to medication adherence. Improvement in measuring reporting and recording of data at post-intervention were also observed.
ACKNOWLEDGEMENT
The authors are grateful to our participating Pay-for-Performance (P4P) practices and former Highmark employee Christine Heasley that served our minority communities in Pennsylvania. This study was funded by Highmark, Inc. Authors Rhonda Johnson, MD, MPH, Oralia Garcia Dominic, PhD, MS, MA, and Twyla Johnson, MPH, MBA are employees of Health Equity and Quality Services at Highmark, Inc. Oralia Garcia Dominic is an National Institutes of Health (NIH) National Cancer Institute (NCI) K01 grantee. Authors Gary Marsh, PhD, FACE and Sarah D. Zimmerman, BS, MS are Highmark Consultants. The authors thank Judah Abberbock, MS for his technical support.
Footnotes
CONFLICT OF INTEREST
Authors Rhonda M. Johnson, MD, MPH, Twyla Johnson, MPH, MBA, and Sarah D.Zimmerman, BS, MS declares that they gave no conflict of interest.
Gary M. Marsh, PhD, FACE declares that he has received grants. He is a principal investigator of a sponsored research project at the University of Pittsburgh funded by Highmark, Inc.
Oralia Garcia Dominic, PhD, MS, MA declares that she has received grants from NIH, PA DOH, ACS. She is an NIH NCI K01 Grantee and principal investigator of this sponsored research project at the Allegheny Singer Research Institute (ASRI) funded by NIH NCI CRCHD. She is an employee and the primary lead for health disparities and clinical interventions for Health Equity and Quality Services at Highmark, Inc., Camp Hill, Pennsylvania.
INFORMED CONSENT
The Allegheny Singer Research Institute's Institutional Review Board approved this study. All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000. Informed consent was obtained from all participating practices/physicians participating in a pay-for-performance (P4P) program for being included in the study.
No identifying information about patients is included in the article.
Animal Studies
No animal or human studies were carried out by the authors for this article.
Contributor Information
Rhonda M. Johnson, Medical Director of Health Equity & Quality Services at Highmark, Pittsburgh, Pennsylvania, Rhonda.moore.johnson@highmark.com; 120 Fifth Avenue, Suite FAP 733, Pittsburgh, PA 15222 Phone: 412.544.1027; Fax: 412-544-6792.
Twyla Johnson, Manager, Provider Engagement, Performance & Partnerships at Highmark, Pittsburgh, Pennsylvania, twyla.johnson@highmark.com; 120 Fifth Avenue, Suite 893, Fifth Avenue Place Pittsburgh, PA 15222-3099; Phone: (412) 544- 5167; Fax: (412) 544 - 8255.
Sarah D. Zimmerman, Research Specialist V at the Center for Occupational Biostatistics and Epidemiology, Department of Biostatistics at University of Pittsburgh. scd27@pitt.edu; A410 Crabtree Hall A411 DeSoto Street, Pittsburgh, PA 15261 Phone: (412)624-9498.
Gary M. Marsh, Professor of Biostatistics, Epidemiology and Clinical & Translational Sciences, and Director of the Center for Occupational Biostatistics and Epidemiology at the Department of Biostatistics, Graduate School of Public Health, University of Pittsburgh, Director, gmarsh@pitt.edu A410 Crabtree Hall 130 DeSoto Street, Pittsburgh, PA 15261 Phone: 412-624-1294; Fax: 412-624-9969.
Oralia Garcia-Dominic, Highmark Health Equity Quality Consultant; NIH NCI CRCHD K01 grantee; and Adjunct Assistant Professor of Public Health Sciences, and Biobehavioral Health at The Pennsylvania State University, Milton S. Hershey Medical Center, College of Medicine, Hershey, Pennsylvania, 100 Senate Avenue, Suite SP6N Camp Hill, PA 17011.
REFERENCES
- 1.Reports: Agency for Healthcare Research and Quality . National healthcare disparities report. Rockville, MD: 2011. [April 10, 2014]. http://ahrq.gov/qual/nhdr11/key.htm. [Google Scholar]
- 2.Reports: Institute of Medicine . Crossing the quality chasm: a new health system for the 21st century. National Academy Press; Washington, DC: 2001. [PubMed] [Google Scholar]
- 3.Klein J, Schubiner L, Gadon M, Wynia M. Physicians’ experience and opinions regarding strategies to improve care for minority patients. JHDRP. 2008;2(2):75–90. [Google Scholar]
- 4.Reports: Institute of Medicine . Unequal Treatment: Confronting Racial & Ethnic Disparities in Health Care. National Academy of Sciences; [April 10, 2014]. http://www.iom.edu/~/media/Files/Report%20Files/2003/Unequal-Treatment-Confronting-Racial-and-Ethnic-Disparities-in-Health-Care/Disparitieshcproviders8pgFINAL.pdf. [Google Scholar]
- 5.Reports: National Health Plan Collaborative . Toolkit to reduce racial and ethnic disparities in health care 2008. AHRQ Rockville, MD: [April 10, 2014]. http://www.rwjf.org/files/research/nhpctoolkit.pdf. [Google Scholar]
- 6.Ho K, Moy E, Clancy C. Can incentives to improve quality reduce disparities? Health Serv Res. 2010;45(1):1–5. doi: 10.1111/j.1475-6773.2009.01076.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Alexander GC, Lin S, Sayla MA, Wynia MK. Development of measure of physician engagement in addressing racial and ethnic health care disparities. Health Serv Res. 2008;43(2):773–84. doi: 10.1111/j.1475-6773.2007.00780.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.American Academy of Family Physicians [April 10, 2014];Pay-for-performance. http://www.aafp.org/online/en/home/policy/policies/p/payforperformance.html. http://www.aafp.org/about/policies/all/pay-performance.html.
- 9.Casolino LP, Elster A. Will pay-for-performance and quality reporting affect health care disparities? Health Aff. 2007;26(3):w405–w414. doi: 10.1377/hlthaff.26.3.w405. [DOI] [PubMed] [Google Scholar]
- 10.Baker G, Carter B. Case studies in health plan pay for performance (introduction) Atlantic Information Services; Washington, DC: 2004. The evolution of pay for performance models for rewarding providers. [Google Scholar]
- 11.Epstein AM, Lee TH, Hamel MB. Paying physicians for high-quality care. N Engl J Med. 2004;350:406–410. doi: 10.1056/NEJMsb035374. [DOI] [PubMed] [Google Scholar]
- 12.Christianson JB, Knutson DJ, Mazze RS. Physician pay-for-performance: implementation and research issues. J Gen Intern Med. 2006;21(Suppl 2):S9–S13. doi: 10.1111/j.1525-1497.2006.00356.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Pennsylvania Department of Health [April 10, 2014];Minority Health Disparities. http://www.portal.state.pa.us/portal/server.pt?open=514&objID=596652&mode=2.
- 14.Coleman K, Reiter KL, Fulwiler D. The impact of pay-for-performance on diabetes care in a large network of community health centers. J Health Care Poor Underserved. 2007;18(4):966–983. doi: 10.1353/hpu.2007.0090. [DOI] [PubMed] [Google Scholar]
- 15.Rees CA, Karter AJ, Young BA. Race/ethnicity, social support, and association with diabetes self-care and clinical outcomes in NHANES. Diabetes Edu. 2010:435–445. doi: 10.1177/0145721710364419. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lurie N, Fremont A, Somers SA, Coltin K, Gelzer A, Johnson R, Rawlins W, Ting G, Wong W, Zimmerman D. The national health plan collaborative to reduce disparities and improve quality. The Joint Commission Journal on Quality and Patient Safety. 2008;34(5):256–265. doi: 10.1016/s1553-7250(08)34032-x. [DOI] [PubMed] [Google Scholar]
- 17.America's Health Insurance Plan (AHIP) Innovations in recognizing and rewarding quality. Recognizing Highmark Blue Cross Blue Shield – Quality BLUE Physician Pay for Performances Program. 2009 Mar;:30–31. [Google Scholar]
- 18.Highmark's Quality Blue 2011 Achievement Compendium . Quality Blue Hospital Payfor-Performance Program. Copyright Highmark Inc.; 2011. [April 10, 2014]. Available at: www.highmark.com. [Google Scholar]
- 19. [April 10, 2014];Your Health Care Partner: Press, Quality Blue. 2012 Available at: https://www.highmark.com/hmk2/pdf/presskits/qualityblue2012.pdf.
- 20.Herald LR, Alexander JA, Shi Y, Casalino LP. Pay-for-Performance and public reporting program participation and administrative challenges among small- and medium-sized physician practices. J Medical Care Research and Review. 2014;71:299–312. doi: 10.1177/1077558713509018. [DOI] [PubMed] [Google Scholar]
- 21.Leon-Chisen N. Coding and quality reporting: resolving the discrepancies, finding opportunities. Journal of AHIMA. 2007;78(7):26–30. [PubMed] [Google Scholar]

