William H. Shrank, MD, MSHS

 Physician satisfaction is associated with patient satisfaction, adherence to treatment recommendations, and quality. However, burnout is prevalent, and physician experience with health plans is likely a key contributor. We explored physician satisfaction with health plans and assessed physician and plan characteristics associated with greater satisfaction.

 Cross-sectional analysis of physician satisfaction surveys for US health plans in 2016.

 We assessed the association between health plan/physician characteristics and physician satisfaction domains using multivariable linear regression. The following satisfaction domains were outcomes of interest, measured by 5-point Likert scales: overall health plan rating, finance, utilization/quality management, network/care coordination, pharmacy, call center, provider relations, and recommendation of the plan to others’ practices.

 We analyzed surveys from 3158 physicians on 74 health plans, representing a 12.6% response rate. We observed highest satisfaction in overall plan rating, finance, and call center domains (adjusted means = 3.25) and lowest satisfaction in the pharmacy domain (adjusted mean = 3.02). The largest and smallest plans and vertically integrated plans had the highest satisfaction; 76% and 66% of physicians recommended vertically integrated plans and non—vertically integrated plans, respectively, to others (

 <.001). Solo practitioners rated overall plan rating, finance, utilization/quality management, and pharmacy domains more favorably than did physicians in larger practices, whereas primary care physicians rated overall plan rating, finance, and utilization/quality management more favorably than did specialists.

 Our findings demonstrate opportunity to improve physician satisfaction with health plans, specifically in pharmacy/formulary management. As provider satisfaction is increasingly recognized as a critical outcome, our findings highlight intervention targets.

Significant opportunity exists to improve physician satisfaction with health plans, specifically in pharmacy/formulary management.

Vertically integrated health plans and the largest and smallest plans had the highest physician satisfaction.

Solo practitioners rated overall health plan ratings, finance, utilization/quality management, and pharmacy domains more favorably than did physicians in larger practices, whereas primary care physicians rated overall health plan ratings, finance, and utilization/quality management domains more favorably than did specialists.

As provider satisfaction is becoming increasingly recognized as a critical outcome of its own, our findings highlight potential intervention targets.

Studies have demonstrated that physician satisfaction is associated with patient satisfaction, adherence to chronic disease management recommendations, and quality of care.

 However, physician burnout is a growing concern; the majority of physicians in the United States experience burnout symptoms, and between 2011 and 2014, the prevalence of burnout among physicians increased by 9% while remaining stable in other fields.

 Given the importance of physician satisfaction, some have suggested that the Triple Aim (representing patient experience, quality of care, and cost reduction) be expanded to the Quadruple Aim, which adds provider satisfaction as the fourth key tenet.

A key source of physicians’ frustration is related to their relationships with payers. Challenges include administrative burden, documentation pressures, payment and delivery changes, and publicly reported quality metrics,

 and they represent central causes of physician burnout. It is therefore important to better understand the associations between physician satisfaction and characteristics of the health plan, provider, and practice.

Our existing knowledge of physician satisfaction with health plans is limited, mixed, and outdated. Studies from the 1990s that focused on managed care found contradictory results regarding the association between managed care and physician satisfaction.

 More recent studies note that increased administrative burden and productivity requirements are associated with reduced satisfaction, but limited evidence is focused on physician satisfaction with health plans specifically.

 Moreover, little is known about what provider and plan characteristics are associated with greater provider satisfaction with health plans. Considering the rapid evolution of new payment models and the emergence of vertically integrated health plans (ie, plans in which healthcare providers and payers are integrated), we need to refresh our understanding of key provider and payer relationships and their association with physician satisfaction. Accordingly, we sought to explore physician satisfaction with health plans and assess the physician and plan characteristics that are associated with greater satisfaction.

In partnership with SPH Analytics, a provider of patient/member experience and population health solutions charged with measuring, analyzing, and interpreting patient and provider satisfaction for US health plans, we conducted a cross-sectional analysis of provider satisfaction surveys from 2016. As a requirement for National Committee for Quality Assurance (NCQA) accreditation, all health plans solicit feedback from their providers with the goal of improving care coordination and quality. Unlike patient satisfaction surveys, NCQA does not specify which questions are asked of providers about health plans. SPH Analytics therefore developed its Provider Satisfaction Benchmark Survey Tool to help health plans assess provider satisfaction. The tool is administered by SPH Analytics on behalf of health plans to providers, including physicians, nurses, office managers, and behavioral health clinicians. Our study used results from the 2016 SPH Analytics Provider Satisfaction Benchmark Survey Tool, which was completed by providers for 130 health plans, representing approximately 30% of health plans across the United States. Of the 130 health plans, 54 had customized surveys with customized questions/responses. Because responses to these questions could not be compared among health plans, we excluded plans with customized surveys from our analyses. We included only physician responses in this analysis (ie, we excluded office staff responses) to simplify result interpretation because our specific focus was the intersection between health plan practices and physician satisfaction, and as a group, physicians may have different responses compared with other provider types.

SPH Analytics developed its original Provider Satisfaction survey in 2001 based on the reporting, regulatory, and accreditation requirements of its health plan clients. In 2012-2013, SPH Analytics re-evaluated the utility of this existing survey, seeking to ensure that it covered meaningful topics and provided valuable results by which health plans could measure and compare their performance and identify areas of opportunity. The organization therefore conducted focus groups with 10 physicians and interviews with 12 office managers from diverse practice settings (representing a combination of primary care and specialist offices and large and small groups) to obtain feedback on the survey, key domains, administration methods and frequency, and result sharing. SPH Analytics then used focus group and interview feedback to revise its original tool. It specifically reordered the domains by most important to least important according to the focus groups and, to reduce respondent burden, removed follow-up questions asking providers to rate other plans they encounter for each domain.

The final survey included 7 demographic questions, 33 five-point Likert scale questions (with response options of well above average, somewhat above average, average, somewhat below average, and well below average), 1 yes/no question about whether the provider would recommend the sponsor plan to other physicians, and 1 free-text question where responders could leave comments. Key domains included (1) overall health plan rating, which reflected general satisfaction with the respective health plan compared with all other plans they work with; (2) financial issues, which included questions on provider reimbursement, fee consistency, and claims processing and resolution; (3) utilization and quality management, which included questions on access to knowledgeable staff, procedures and timeliness of preauthorization information, facilitation of appropriate clinical care, access to care managers, and preventive care and wellness coverage; (4) network/coordination of care, which included questions on number, quality, and timeliness of reports from specialists and behavioral health clinicians in the plan’s provider network; (5) pharmacy issues, which included questions on formulary consistency over time, reflection of current care standards, variety of branded drugs, ease of prescribing preferred medications, and availability of substitutions; (6) call center experiences, which included questions on ease of reaching the call center, helpfulness in referrals, and overall satisfaction with call center staff; (7) provider relations, which included questions on availability and capacity of provider relations representatives, quality of provider orientation, and value of written communications; and (8) whether providers would recommend the respective health plan to other practices.

Piloting and Validity/Reliability Testing

SPH Analytics then piloted the final survey with 8 health plans that included 1524 provider responders, representing physicians, nurses, office managers, behavioral health clinicians, and other staff. SPH Analytics conducted reliability testing via internal consistency analysis (Cronbach’s alpha) and validity testing via factor analysis. In internal consistency analyses, the Cronbach’s alpha values for the key domains (finance issues, utilization and quality management, network/coordination of care, pharmacy, call center services, and provider relations) ranged from 0.884 to 0.957, indicating that the domains represented reliable measures of provider satisfaction. Factor analysis suggested 4 underlying factors that matched with survey domains—provider relations, quality and accessibility, pharmacy, and finance issues.

Sampling Strategy and Survey Administration

SPH Analytics works with health plans to determine their provider sampling strategies. Strategies generally start with a sample of 1500 providers per health plan, with stratification efforts to include 60% primary care physicians, 30% specialists, and 10% behavioral health clinicians. Physicians with high patient volumes are targeted first because they likely have more interaction with health plans. Based on focus group feedback on desired administration methods, SPH Analytics administers the survey to providers using mail, email, and phone. Of 114,880 email and mail surveys from 2016, 10,240 providers responded (response rate, 8.9%); of 62,632 phone surveys, 12,178 providers responded (response rate, 18.8%). The total response rate was 12.6% for all providers.

SPH Analytics received survey results from providers and built a deidentified data set at the provider—plan level using their survey data and AIS health plan demographic data. Each plan was assigned a unique numerical identifier so responses could be compiled for each plan. SPH Analytics then shared the deidentified data with the University of Pittsburgh Medical Center researchers who conducted analyses. The data set included 22,418 provider surveys for 76 health plans, of which 3158 were completed by physicians.

Outcomes/Covariates of Interest and Statistical Analysis

For each physician—plan dyad, we included the following provider satisfaction domains described previously as outcomes of interest: (1) overall health plan rating, (2) financial issues, (3) utilization and quality management, (4) network/coordination of care, (5) pharmacy, (6) call center experiences, (7) provider relations, and (8) provider recommendation of the sponsor plan to other practices.

We assessed the association between the following characteristics and outcomes of interest using multivariable linear regression, weighted by the number of providers who completed surveys for each plan: vertical integration status, defined as the integration of provider and payer systems and categorized using publicly available lists from Robert Wood Johnson Foundation, McKinsey, and Avalere (

; health plan size (stratified by ≤100,000, 100,001-500,000, 500,001-1 million, 1,000,001-2 million, and >2 million enrollees); practice size (stratified by practices having 1 physician, 2-5 physicians, and >5 physicians); provider type (stratified by primary care, specialists, and behavioral health physicians); and years of practice (stratified by <5 years, 5-15 years, and >15 years). In addition to these variables, we also adjusted our models for HHS region, number of insurance companies accepted by a respective provider’s practice, and proportion of a practice’s managed care volume represented by a respective health plan. We clustered deidentified provider responses based on the sponsored health plans they were assessing and, because we did not have unique identifiers for provider responders, used robust standard error estimation to account for potential correlation in a scenario in which the same provider responded to surveys about multiple health plans. Missing data were deemed missing completely at random through a multistep statistical verification process that included visual inspection of missing data, tabulating missing data for each variable, and performing Little’s test to assess the assumption that missing data were missing completely at random. Analyses were performed using Stata 14 (StataCorp LP; College Station, Texas) and SAS 9.4 (SAS Institute Inc; Cary, North Carolina).

We received surveys from 3158 physicians for 74 health plans (

). Primary care physicians represented 62% of responders and specialists represented 38%. Fifty-one percent of responders were solo practitioners, and 55% were in practice for more than 15 years. Thirty-eight percent of physicians’ practices participated with more than 15 insurance companies. In terms of plan demographics, 11% of plans were vertically integrated and 89% of plans were not. Most plans had 500,000 or fewer enrollees and belonged to HHS regions 9 (San Francisco), 6 (Dallas), 5 (Chicago), and 4 (Atlanta).

Overall provider satisfaction varied little across domains, with highest satisfaction in the overall health plan rating (adjusted mean = 3.25), finance (adjusted mean = 3.25), and call center (adjusted mean = 3.25) domains and lowest satisfaction in the pharmacy domain (adjusted mean = 3.02) (eAppendix Table). Providers rated pharmacy significantly lower than all other domains (

 <.05). The call center domain received the most “somewhat above average” and “well above average” ratings, whereas the pharmacy domain received the fewest “somewhat above average” and “well above average” ratings and the most “average” ratings (

In multivariable models, a number of health plan, provider, and practice characteristics were associated with physician satisfaction. Vertically integrated plans had significantly higher adjusted means than non—vertically integrated plans for all satisfaction domains (

; eAppendix Table), and the largest difference between vertically integrated and non—vertically integrated plans was observed in the overall health plan rating (β = 0.89; 95% CI, 0.61-1.17). Of note, 76% of providers would recommend vertically integrated plans and 66% of providers would recommend non–vertically integrated plans to other practices (

 <.001). By plan size, physician satisfaction was highest for the largest (1,000,0001-2 million enrollees and >2 million enrollees) and smallest (≤100,000 enrollees) plans (Figure 2). Compared with practices with more than 1 physician, solo practitioners rated plans more favorably in overall health plan rating, finance, utilization/quality management, and pharmacy domains (

). Compared with primary care responders, specialists rated plans less favorably in overall health plan rating (β = —0.20; 95% CI, –0.31 to –0.11), finance (β = –0.21; 95% CI, –0.33 to –0.11), and utilization/quality management (β = –0.17; 95% CI, –0.27 to –0.06) (eAppendix Table).

Our findings demonstrate that with a mean overall plan rating of 3.25 of 5 points, significant opportunity exists to improve physician satisfaction with health plans, specifically in pharmacy/formulary management. Furthermore, we found that vertically integrated plans and the largest and smallest plans had the highest physician satisfaction overall. Solo practitioners rated overall health plan ratings, finance, utilization/quality management, and pharmacy domains more favorably than did physicians in larger practices, whereas primary care physicians rated overall health plan ratings, finance, and utilization/quality management domains more favorably than did specialists.

Vertical integration was a strong predictor of satisfaction; this could be due to perceived increased efficiency and streamlining of processes and initiatives, alignment of incentives, or engagement and participation in coverage and benefits policies. Other studies indicate that vertically integrated plans have better quality and patient satisfaction,

 although these studies did not include provider satisfaction as an outcome.

Plan size was an important predictor of physician satisfaction. Plans that serve more enrollees could have higher physician satisfaction because of increased organizational commitment and potentially more experience interacting with and supporting providers. These findings are consistent with those of studies on patient satisfaction that suggest that larger plan size is positively associated with higher satisfaction.

 Interestingly, we found that smaller plans also had high physician satisfaction, potentially because smaller plans tend to be community-based plans and thus may have more intimate community-based relationships and increased ability to customize care compared with larger plans.

Physician practice characteristics were also associated with satisfaction with health plans. Solo practitioners rated plans more favorably than did physicians in practices with more than 1 physician. This finding is consistent with those of studies demonstrating that practices with fewer physicians have higher physician satisfaction in general.

 Grembowski et al and Blechter et al explain this observation by the notion that as primary care physicians shift from solo to group practice, they become salaried employees who are more likely to have increased bureaucratic controls and reduced autonomy imposed on them.

 Physicians in larger practices may be less satisfied with health plans due to similar pressures.

 Our finding that primary care physicians rated finance and utilization/quality management domains higher than did specialists could be related to multiple factors. First, specialists may be more affected by utilization management than primary care physicians, specifically those in procedure-based specialties whose relative value units and incomes are based on procedures that require prior authorization from health plans. Second, new alternative payment models that prioritize value over volume may disproportionately benefit primary care physicians over specialists.

Additionally, because primary care is more heavily influenced by payment reform, health plans potentially have had to intensify their outreach efforts with primary care providers more so than with specialists.

Our study has limitations. First, we could not include plans that use customized provider satisfaction surveys because responses could not be fairly compared. Therefore, the potential for selection bias exists for plans that chose standardized surveys because they may not be representative of all plans. Second, the response rate for the survey was 12.6% among providers. It is unclear how our responders might differ from nonresponders, and our responders may not be representative of all physicians responding for all health plans: Although our physician years of experience and practice size seem to be reflective of the general physician population

 and health plan size distribution and vertical integration status seem to be reflective of the general health plan population,

 our representation from primary care physicians is higher than primary care representation in the United States.

 Importantly, we adjusted for whether physicians were primary care providers versus specialists in our multivariable model (

) and display adjusted means for primary care and specialist physicians separately (Figure 3). Similarly, selection bias may exist among physicians, in that physicians who may be more or less satisfied with respective plans might complete these optional surveys more frequently. Interestingly, although we anticipated that providers may be more likely to respond to surveys when a higher proportion of their patients were enrolled in sponsor health plans, this was not the case; most providers had less than 20% of their patients enrolled in respective plans, likely because the majority of providers’ patients were represented by more than 11 health plans (Table 1). Nevertheless, we adjusted for differences in the proportion of patients represented by sponsor health plans in our multivariable models. Despite potential selection bias, to our knowledge, no other study has assessed such a large volume of physician satisfaction surveys.

 Finally, although we adjusted for regional differences, certain regions had very little representation by plans. For example, 0 plans were associated with region 8 (Denver), whereas 21 plans were associated with region 9 (San Francisco).

To our knowledge, this is the first study to assess contextual factors associated with provider satisfaction with health plans. As part of the largest physician satisfaction study to date,

 our results contribute to a growing foundation of work related to contextual factors associated with provider satisfaction. Our findings have important implications for policy and practice. We demonstrate that provider ratings of health plans were, in general, low. As provider satisfaction is increasingly recognized as crucial for health systems, significant opportunity exists to improve the relationship between providers and health plans. In 2005, the Association for Community Affiliated Plans collaborated with The Commonwealth Fund to identify best practices for health plans to recruit and retain providers.

 These best practices include improving payment practices, incentives, and financial assistance for claims; strengthening referral and authorization practices through streamlining processes and incorporating technology; improving communication through in-person meetings and regular written communications; simplifying administrative burdens, such as credentialing, eligibility requirements, and data required for encounters; and enhancing support for high-risk patients, including assistance with patient transportation, child care, and interpreter services. Importantly, the report did not address health plans’ approach to pharmacy. Our findings build on these existing recommendations by suggesting that plans could additionally focus on their formulary consistency, availability, affordability, messaging, prior authorization procedures, and ease of navigation to improve provider satisfaction.

Taken together, rather than exacerbating physician burnout through administrative burden, documentation requirements, and cumbersome utilization management processes, our findings suggest that health plans have an opportunity to improve physician satisfaction through prioritizing provider relationships, reducing administrative burden, and strengthening resource support. As physicians are required to adapt to a rapidly transforming healthcare landscape, it will be imperative for health plans to prioritize physician satisfaction moving forward.

 University of Pittsburgh Medical Center (UPMC) Center for High-Value Health Care (NP, WHS) and Insurance Services Division (NP, AH, SM, WHS), Pittsburgh, PA; Division of General Internal Medicine, University of Pittsburgh (NP), Pittsburgh, PA; SPH Analytics (SS, PA), Alpharetta, GA; University of Pittsburgh Graduate School of Public Health (IDM), Pittsburgh, PA.

 Dr Parekh is employed by the UPMC Center for High-Value Health Care within the UPMC Insurance Services Division. Dr Helwig and Ms McAnallen are employed by the UPMC Insurance Services Division. Dr Shrank was employed by the UPMC Insurance Services Division during the development, analysis, drafting, and acceptance of the article. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

 Concept and design (NP, AH, WHS); acquisition of data (SS, PA, SM); analysis and interpretation of data (NP, SS, AH, PA, IDM, WHS); drafting of the manuscript (NP, AH, WHS); critical revision of the manuscript for important intellectual content (SS, AH, IDM, SM, WHS); statistical analysis (NP, PA, IDM); administrative, technical, or logistic support (SM); and supervision (SM).

 Natasha Parekh, MD, MS, UPMC Center for High-Value Health Care, 600 Grant St, 40th Floor, Pittsburgh, PA 15219. Email: natashaparekh1@gmail.com.

1. Grol R, Mokkink H, Smits A, et al. Work satisfaction of general practitioners and the quality of patient care. 

. 1985;2(3):128-135. doi: 10.1093/fampra/2.3.128.

2. DiMatteo MR, Sherbourne CD, Hays RD, et al. Physicians’ characteristics influence patients’ adherence to medical treatment: results from the Medical Outcomes Study. 

. 1993;12(2):93-102. doi: 10.1037/0278-6133.12.2.93.

3. Grumbach K, Osmond D, Vranizan K, Jaffe D, Bindman AB. Primary care physicians’ experience of financial incentives in managed-care systems. 

. 1998;339(21):1516-1521. doi: 10.1056/NEJM199811193392106.

4. Barr DA. The effects of organizational structure on primary care outcomes under managed care. 

. 1995;122(5):353-359. doi: 10.7326/0003-4819-122-5-199503010-00006.

. 1998;339(21):1543-1545. doi: 10.1056/NEJM199811193392109.

6. Dyrbye LN, Shanafelt TD, Sinsky CA, et al. Burnout among health care professionals: a call to explore and address this underrecognized threat to safe, high-quality care. National Academy of Medicine website. nam.edu/burnout-among-health-care-professionals-a-call-to-explore-and-address-this-underrecognized-threat-to-safe-high-quality-care. Published July 5, 2017. Accessed August 28, 2018.

7. Bodenheimer T, Sinsky C. From triple to quadruple aim: care of the patient requires care of the provider. 

. 2014;12(6):573-576. doi: 10.1370/afm.1713.

Articles

Several physician and payer characteristics are associated with physician satisfaction with health plans. There is opportunity to improve physician satisfaction with payers, specifically in pharmacy.

Physician Satisfaction With Health Plans: Results From a National Survey

 To evaluate how changes in generic drug prices and the incidence of abrupt price increases varied with the number of manufacturers supplying each drug.

 Analysis of 2005 to 2016 monthly wholesale acquisition costs (WACs) and University of Pittsburgh Medical Center Health Plan counts of pharmacy claims for National Drug Codes (NDCs) for generic drugs.

 Each year, NDCs were categorized according to the number of manufacturers offering each combination of active ingredient and dosage form: 1 to 3, 4 to 7, and more than 7. For every month from January 2006 to January 2017, we estimated the 12-month change in WAC (eg, 12-month change in January 2006 was calculated as the difference in WAC between January 2006 and January 2005, divided by the WAC in January 2005), before and after weighting each NDC by counts of pharmacy claims. We evaluated the proportion of NDCs that had large price increases, greater than 20%, 50%, 100%, and 500% within a year.

 Before 2010, price changes were higher for drugs supplied by a lower number of manufacturers; however, after 2010, prices increased sharply, and drugs supplied by 4 to 7 manufacturers showed increases similar to or higher than those supplied by 1 to 3. In 2013, prices increased by an average of 29% for drugs supplied by 1 to 3 and 4 to 7 manufacturers, and 10% for more than 7. Price changes increased after weighting by counts of pharmacy claims, demonstrating that price increases disproportionately affected widely used drugs. The proportion of NDCs from drugs supplied by 1 to 3 manufacturers that doubled in price within a year was 3.6 times higher in 2012 to 2015 than in 2005 to 2009 (4.6% vs 1.3%, respectively).

 Increases in generic drug prices are concerning because they affected widely used drugs and suggest that generic drug prices may be increasingly insensitive to competition.

Before 2010, changes in prices of generic drugs were higher for drugs supplied by a lower number of manufacturers. However, after 2010, price changes increased abruptly, and drugs supplied by 4 to 7 manufacturers showed increases similar to or even higher than those supplied by 1 to 3 manufacturers.

In 2013 and 2014, price changes averaged 29% and 24%, respectively, for drugs supplied by 1 to 3 manufacturers, 29% and 28% for 4 to 7 manufacturers, and 10% and 15% for more than 7 manufacturers.

Recent increases in generic drug prices are concerning because they affected multisource, widely used drugs. Rising generic drug prices may be due not only to insufficient market competition but also to the increasing insensitivity of prices to competition.

Several studies have quantified drug price inflation over the past decade, finding that although brand name and specialty drugs experienced the largest price increases, generic prices also increased over time.

 Rising generic drug prices have been attributed to the relatively low competition present in most generic markets—the average generic is only supplied by 2 manufacturers,

 and generic drug prices are not responsive to competition until at least 3 manufacturers have entered the market.

 However, prices of widely used generic drugs supplied by numerous manufacturers have recently experienced abrupt increases, in some cases triggering allegations of price fixing.

We quantified how 12-month changes in prices of all generic drugs available from 2005 to 2016 changed with the number of generic manufacturers supplying each drug, and we also evaluated the incidence of abrupt price increases across the study period. We hypothesized that price increases would be higher for generic drugs supplied by a lower number of manufacturers.

We obtained monthly wholesale acquisition costs (WACs) for all National Drug Codes (NDCs) available from 2005 to 2016 from AnalySource (reprinted with permission by First Databank, Inc) and annual counts of pharmacy claims from the University of Pittsburgh Medical Center (UPMC) Health Plan over the same time frame. WACs represent the list price from a manufacturer to a wholesaler or a direct purchaser and do not account for any type of manufacturer rebate or discount. An NDC is a unique and universal 10-digit number that identifies human drug products in the United States. UPMC Health Plan is a health insurer offering a variety of insurance products to more than 3.2 million members in Pennsylvania and parts of Ohio, West Virginia, and Maryland. Consequently, UPMC Health Plan offers a representative expression of medication use in an insured population. We standardized annual counts of pharmacy claims for each NDC, as if enrollment had remained constant since 2005.

Using data provided by First Databank, we identified all NDCs for generic drugs with WAC information available any time from 2005 to 2016. Generics entering the market within this time frame were included in the study sample starting the first year when their list price was available. The sample included 24,944 NDCs representing 1851 combinations of active ingredient and dosage form. Each year, NDCs were categorized according to the number of generic manufacturers—1 to 3, 4 to 7, and more than 7&mdash;offering each combination of active ingredient and dosage form in January. For example, flecainide tablets were supplied by 3 manufacturers in 2006 to 2008, 4 in 2009 to 2011, 6 in 2012 to 2014, 8 in 2015, and 9 in 2016. Consequently, NDCs for flecainide tablets were categorized under the 1 to 3 manufacturer group in 2006 to 2008, the 4 to 7 group in 2009 to 2014, and the more than 7 group in 2015 and 2016. We selected 3 as a cutoff because the FDA expedites review of generic drug applications in markets served by 3 or fewer manufacturers.

For every month from January 2006 to January 2017, we estimated the 12-month change in WAC for each NDC. For example, the 12-month change measured in January 2006 was calculated as the difference in WAC between January 2006 and January 2005, divided by the WAC in January 2005. This is the reason why, although we had 2005 to 2016 data, we report 12-month changes in WAC for 2006 to 2016. Because our results could be disproportionately affected by rarely used drugs with large price increases, we further calculated the average 12-month change after weighting each NDC by the count of pharmacy claims from UPMC Health Plan in the respective year. Finally, we calculated and reported the average of the 12-month change in WAC for every year of the study period.

 To assess the association between number of manufacturers and price changes while controlling for other factors that can affect price changes,

 we constructed a linear mixed model. The linear mixed model regressed 12-month change in WAC for each NDC against indicator variables for year, for groups defined by number of manufacturers, route of administration, therapeutic classes (list in the 

]), and interactions between the variables for number of manufacturers and for years. To account for the correlation in price changes between NDCs for a given product, we specified random intercepts for generic names. An autoregressive covariance structure was used to model the correlation between repeated observations for each NDC. We constructed a second version of this linear mixed model by weighting each NDC by the count of pharmacy claims from UPMC Health Plan in the respective year.

Finally, using the 12-month change in WAC measured in January of every year, we calculated the proportion of NDCs that had abrupt price increases, which we defined as increases greater than 20% within a year. We further quantified the proportion of NDCs that had even higher price increases (ie, greater than 50%, 100%, and 500%) within a year.

Before 2010, average 12-month price changes were consistently higher for drugs supplied by a lower number of manufacturers (

). Specifically, average 12-month price changes were at least 3 percentage points higher for drugs supplied by 1 to 3 manufacturers than those supplied by 4 to 7, and price changes were negative for those supplied by more than 7. For instance, in 2008, the average 12-month price change was 7% for drugs with 1 to 3 generic manufacturers, 4% for 4 to 7 manufacturers, and —8% for more than 7 manufacturers. However, after 2010, average 12-month price changes increased sharply for all groups, and drugs supplied by 4 to 7 manufacturers showed price increases similar to or even higher than those supplied by 1 to 3 (Figure 1). Drugs supplied by more than 7 manufacturers presented lower average price increases than those supplied by 7 or fewer, yet they exceeded 10% in 2013 and 2014. For instance, in 2013 and 2014, 12-month price changes averaged 29% and 24%, respectively, for drugs supplied by 1 to 3 manufacturers, 29% and 28% for those supplied by 4 to 7 manufacturers, and 10% and 15% for those supplied by more than 7. The magnitude of these estimates increased when we weighted NDCs by counts of pharmacy claims (Figure 1).

Results from mixed models were consistent with these observations (

): Average 12-month price changes in 2007 to 2010 did not significantly differ from those observed in 2006; however, price changes in 2011 to 2016 were significantly higher compared with 2006 (

 <.01). The effect of the number of manufacturers on price was different in 2012 and 2013 compared with 2006, but not in the remaining years (eAppendix Table). To facilitate interpretation of the results from these models, Figure 1 shows the price changes predicted by linear mixed models. In 2006 to 2011, predicted price changes were consistently higher for drugs supplied by a lower number of manufacturers; however, from 2012 to 2013, predicted price changes were similar for drugs supplied by 4 to 7 manufacturers and those supplied by 1 to 3 (Figure 1). The magnitude of the predicted estimates for price changes increased when each NDC was weighted by counts of pharmacy claims, and predicted price changes for drugs supplied by 4 to 7 manufacturers in 2012 to 2015 exceeded those for drugs supplied by 1 to 3 (Figure 1).

The proportion of NDCs experiencing abrupt price increases showed a similar trend (

). In 2005 to 2009 the proportion that experienced price increases of more than 20% within a year included around 4% of NDCs from drugs supplied by 1 to 3 manufacturers, 1.5% of those supplied by 4 to 7, and 0.4% of those supplied by more than 7. In this time period, an average of 1.3% of NDCs from drugs supplied by 1 to 3 manufacturers, 0.7% for 4 to 7, and 0.01% for more than 7 doubled in price within a year. After 2010, the proportion of NDCs experiencing abrupt price increases raised considerably: In 2012 to 2015, an average of 4.6% of NDCs from drugs supplied by 1 to 3 manufacturers, 3.4% for 4 to 7, and 0.15% for more than 7 doubled in price within a year. In other words, the proportion of NDCs that doubled in price within a year was 3.6 times higher in 2012 to 2015 than in 2005 to 2009 for drugs supplied by 1 to 3 manufacturers, and 5.1 times higher for those supplied by 4 to 7. NDCs from drugs supplied by 1 to 3 manufacturers experienced particularly abrupt increases in 2012: As many as 14.7% experienced a price increase of more than 20%, 7.4% doubled in price, and 2.0% quintupled. NDCs from drugs supplied by 4 to 7 manufacturers presented the strongest increases in 2014: 11.1% experienced a price increase of more than 20%, 5.0% doubled in price, and 2.2% quintupled. In 2016, the proportion of NDCs experiencing abrupt price increases decreased to levels similar to those observed in 2005 to 2009.

Before 2010, the increased competition associated with a greater number of manufacturers consistently resulted in lower price increases. However, in 2012 to 2015, price increases were similar or even higher for drugs supplied by 4 to 7 manufacturers compared with those supplied by 1 to 3, and were as high as 25%. In this time period, increases in prices of drugs supplied by more than 7 manufacturers averaged 10%. In addition, when NDCs were weighted by utilization, prices increased more dramatically, demonstrating that price increases disproportionately affected widely used drugs. The proportions of NDCs offered by 1 to 3 and by 4 to 7 manufacturers that doubled in price within a year were 3.7 and 5.1 times higher, respectively, in 2012 to 2015 than in 2005 to 2009.

Our findings add to the existing literature because they demonstrate that large increases in prices of off-patent drugs have not been limited to single-source drugs used in rare conditions,

 but have also affected commonly used multisource drugs. These price increases are particularly concerning because they can have a large impact on spending, as illustrated by our estimates that account for the relative utilization of drugs. With generic drugs supplied by 4 to 7 and more than 7 manufacturers experiencing average increases of 25% and 10%, respectively, in 2012 to 2015, our results suggest that rising prices of generic drugs may be due not only to limited competition in the generic market,

 but also to the increasing lack of sensitivity of generic drug prices to competition.

Although our study did not assess the reasons behind this increasing insensitivity of prices to competition, the high price increases observed in 2012 to 2015 could reflect the occurrence of several cases currently under investigation for alleged price collusion.

 Alternatively, they could be related to the so-called 2012 patent cliff, when several blockbuster brand name drugs—including atorvastatin, clopidogrel, olanzapine, quetiapine, escitalopram, montelukast, esomeprazole, and methylphenidate&mdash;lost patent protection.

 Specifically, manufacturers may have increased generic drug prices because they had fewer opportunities to generate revenue through 180-day exclusivity periods after the end of the patent cliff. Regardless of the underlying reasons, the sharp increase in the incidence of abrupt price increases indicates distorted pricing dynamics in the pharmaceutical market during this time period. However, we found that trends in price changes stabilized after 2015. The reason for price stabilization is also unclear, although it could be due to increased social and political pressure or the filing of several complaints alleging collusion in rising prices of generic drugs.

Future research should further evaluate whether the price trends observed in 2012 to 2015 for generic drugs are isolated and whether the price stabilization observed after 2015 is maintained over time. This will clarify whether competition in the generic market is enough to maintain low inflation rates or whether the implementation of policy solutions designed to cap inflation is needed.

First, in our analyses, we used WACs as pricing estimates, which do not account for manufacturer discounts or rebates. Nevertheless, list prices are often used to provide directional information about drug prices, and uninsured or underinsured patients are often exposed to these prices. Additionally, our focus on generic drugs also reduces the likelihood of the impact of rebates. Second, we used counts of pharmacy claims from UPMC Health Plan to weight each NDC by its relative utilization. The use of data from only 1 insurance carrier limits the generalizability of our results; however, the use of weights was necessary to confirm that the observed average increases in prices were not driven by disproportionate increases in prices of rarely used drugs.

We observed double-digit increases in prices of all generic drugs in 2012 to 2015, regardless of the number of manufacturers. Our results signal market failures in the generic drug market and warrant the consideration of policy strategies to increase the sensitivity of generic drug prices to competition.

 Department of Pharmacy and Therapeutics, School of Pharmacy, University of Pittsburgh (IH, MH), Pittsburgh, PA; Insurance Services Division, University of Pittsburgh Medical Center (CBG, NP, WHS), Pittsburgh, PA; Division of General Internal Medicine, School of Medicine, University of Pittsburgh (WFG), Pittsburgh, PA.

 The authors acknowledge funding from the Myers Family Foundation. Dr Hernandez is funded by the National Heart, Lung, and Blood Institute (grant number K01HL142847).

 Dr Parekh and Dr Good are employed by the University of Pittsburgh Medical Center (UPMC) Insurance Service Division, Center for Value-Based Pharmacy Initiatives. Dr Shrank is employed by UPMC Health Plan. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article. This study was presented at the International Society for Pharmacoeconomics and Outcomes Research Europe 2018 Annual Meeting in Barcelona, Spain, on November 13, 2018.

 Concept and design (IH, CBG, WFG, NP, MH, WHS); acquisition of data (IH, CBG, MH); analysis and interpretation of data (IH, CBG, WFG, NP, MH, WHS); drafting of the manuscript (IH, CBG); critical revision of the manuscript for important intellectual content (WFG, NP, MH, WHS); statistical analysis (IH); provision of patients or study materials (IH); obtaining funding (IH); and administrative, technical, or logistic support (IH).

 Inmaculada Hernandez, PharmD, PhD, Department of Pharmacy and Therapeutics, School of Pharmacy, University of Pittsburgh, 3501 Terrace St, 637 Salk Hall, Pittsburgh, PA 15261. Email: inh3@pitt.edu.

1. Berndt ER, Conti RM, Murphy SJ. Small revenue concentrated markets, expedited reviews, and incentives to ensure a balance between innovation and access. FDA website. www.fda.gov/media/106398/download. Published July 18, 2017. Accessed June 5, 2018.

2. Schondelmeyer SW, Purvis L. Trends in retail prices of generic prescription drugs widely used by older Americans, 2006 to 2015. AARP website. aarp.org/content/dam/aarp/ppi/2017/08/trends-in-retail-prices-of-generic-prescription-drugs-widely-used-by-older-americans.pdf. Published September 2017. Accessed February 5, 2018.

3. Dave CV, Hartzema A, Kesselheim AS. Prices of generic drugs associated with numbers of manufacturers. 

. 2017;377(26):2597-2598. doi: 10.1056/NEJMc1711899.

4. Dave CV, Kesselheim AS, Fox ER, Qiu P, Hartzema A. High generic drug prices and market competition: a retrospective cohort study. 

. 2017;167(3):145-151. doi: 10.7326/M16-1432.

5. Generic competition and drug prices. FDA website. fda.gov/aboutfda/centersoffices/officeofmedicalproductsandtobacco/cder/ucm129385.htm. Published November 20, 2017. Accessed June 6, 2018.

6. Generic drugs—expanded complaint. Maryland Attorney General website. marylandattorneygeneral.gov/news%20documents/GenericDrugsExpandedComplaint.pdf. Published October 31, 2017. Accessed June 4, 2018.

7. United States District Court for the Eastern District of Pennsylvania. Direct Purchaser Class Action Complaint: case 2:16-cv-06671-CMR. Citron Research website. citronresearch.com/wp-content/uploads/2017/01/Rochester-vs-Lannett-Complaint.pdf. Published December 28, 2016. Accessed June 4, 2018.

8. Policy and procedures: prioritization of the review of original ANDAs, amendments, and supplements. FDA website. www.fda.gov/media/89061/download. Published November 9, 2017. Accessed September 11, 2018.

9. Hernandez I, Sampathkumar S, Good CB, Kesselheim AS, Shrank WH. Changes in drug pricing after drug shortages in the United States. 

. 2019;170(1):74-76. doi: 10.7326/m18-1137.

10. Hernandez I, Good CB, Cutler DM, Gellad WF, Parekh N, Shrank WH. The contribution of new product entry versus existing product inflation in the rising costs of drugs. 

. 2019;38(1):76-83. doi: 10.1377/hlthaff.2018.05147.

11. Hernandez I, Bott SW, Patel AS, et al. Pricing of monoclonal antibody therapies in the US: higher if used for cancer? 

12. Collins SM, McCaskill C. Sudden price spikes in off-patent prescription drugs: the monopoly business model that harms patients, taxpayers, and the U.S. health care system. Susan Collins, United States Senator for Maine website. collins.senate.gov/sites/default/files/DP%20Report.pdf. Published December 2016. Accessed November 3, 2017.

13. Harrison C. The patent cliff steepens. 

. 2011;10(1):12-13. doi: 10.1038/nrd3356.

14. Sivashanker K, Fanikos J, Kachalia A. Addressing the lack of competition in generic drugs to improve healthcare quality and safety. 

. 2018;33(11):2005-2007. doi: 10.1007/s11606-018-4548-x.

Regardless of the number of manufacturers, generic drug prices presented double-digit average increases from 2012 to 2015.

Number of Manufacturers and Generic Drug Pricing From 2005 to 2017

 To compare healthcare quality, utilization, and patient satisfaction between provider-led health plans (PLHPs) and non-PLHPs.

 Observational study of 2016 Medicare Advantage (MA) plans.

 We included 3 quality outcomes (MA Star Rating System, Healthcare Effectiveness Data and Information Set [HEDIS] effectiveness aggregate score, and HEDIS access aggregate score), 4 utilization outcomes (HEDIS average procedure rates, discharge rates, inpatient days, and readmission probability), and 1 patient satisfaction outcome (National Committee for Quality Assurance consumer satisfaction rating). We performed regression analysis to compare the 8 selected outcomes between PLHPs and non-PLHPs, controlling for key covariates, including region, profit status, patient risk, and patient-related and provider-related demographics.

 Our sample included 64 contracts offered by 31 PLHPs (representing 3,197,284 enrollees) and 311 contracts offered by 55 non-PLHPs (representing 13,881,210 enrollees). Compared with non-PLHPs, in our primary multivariable model, PLHPs were associated with higher star ratings (β = 0.41; 95% CI, 0.15-0.67), effectiveness scores (β = 3.11; 95% CI, 1.43-4.80), and patient satisfaction (β = 0.57; 95% CI, 0.30-0.84), and lower procedure rates (β = —0.47; 95% CI, –0.79 to –0.16). There were no significant differences in access, discharges, inpatient days, and readmission probability. The association between PLHPs and outcomes differed by plan size, nonprofit status, and region.

 Receipt of care within a PLHP was associated with improved quality, effectiveness, and patient satisfaction, as well as lower procedure rates. As providers bear increasing financial risk under alternative payment models, there is momentum to integrate healthcare provision and payment through PLHPs. Our results demonstrate the potential of such organizations to deliver high-quality care, although opportunities remain to optimize utilization.

In this cross-sectional study of 2016 Medicare Advantage plans, provider-led health plans (PLHPs) were associated with higher quality, effectiveness, and patient satisfaction and decreased procedure rates compared with non-PLHPs.

The association between PLHPs and outcomes differed by plan size, nonprofit status, and region.

There were no significant differences between PLHPs and non-PLHPs in access, number of inpatient discharges, duration of stay, and readmission probability.

As alternative payment models grow in popularity and momentum builds for providers to start their own health plans, our results demonstrate the potential of PLHPs to deliver higher-quality care and patient satisfaction, although opportunities remain to optimize utilization.

As healthcare providers accept increasing financial risk in alternative payment models, more provider organizations are expected to operate their own health insurance plans, known as provider-led health plans (PLHPs). Over the past 2 decades, PLHPs have become increasingly popular in the United States, with more than 100 plans covering more than 15 million individuals.

 Often referred to as vertical integration, the integration of healthcare providers and payers offers potential advantages to the patient, provider, and system. By inherently aligning payer—provider incentives and managing healthcare across a continuum of services, PLHPs may be particularly advantageous in population health management and, therefore, may have superior outcomes with lower premiums compared with non-PLHPs.

Our knowledge of the impact of PLHPs on outcomes remains limited and inconsistent.

 For instance, critics of PLHPs argue that they are not consistently associated with higher-quality healthcare and can lead to increased costs due to greater market power and administrative costs.

 Furthermore, it remains unknown how PLHP characteristics, including size, region, and nonprofit status, may affect outcomes. For example, nonprofit plans may perform better than for-profit plans,

 and larger plans may perform better than smaller plans through increased experience.

The objectives of this study were therefore to (1) determine the association between PLHP status and healthcare quality, utilization, and patient satisfaction and (2) determine whether these associations differed by plan size, nonprofit status, and region.

We conducted an observational study of Medicare Advantage (MA) contracts using December 2016 MA enrollment data from CMS. We focused on MA due to its large population and available outcome data that allow for standardized comparisons.

 We identified all MA contracts offered in 2016 with more than 20,000 enrollees to increase generalizability. For each contract, we obtained information on 3 quality outcomes, 4 utilization outcomes, and 1 patient satisfaction outcome. The quality outcomes were the 2017 MA Star Rating System (5-star maximum); 2016 Healthcare Effectiveness Data and Information Set (HEDIS) effectiveness aggregate score, defined as the average of 55 HEDIS Effectiveness of Care measures (100% maximum); and 2016 HEDIS access aggregate score, defined as the average of 2 HEDIS Access of Care measures (100% maximum). The utilization outcomes were 2016 HEDIS measures and included procedure rates, defined as average procedure rates per 1000 members for 13 selected procedures; discharge rates, defined as risk-adjusted discharges per 1000 members; inpatient days, defined as inpatient days per 1000 member-months; and risk-adjusted readmission probability. The patient satisfaction outcome was the 2016-2017 National Committee for Quality Assurance consumer satisfaction ratings, which are based on 2016-2017 Consumer Assessment of Healthcare Providers and Systems (CAHPS) surveys (5-point maximum) (see 

We categorized each MA contract as belonging to a PLHP or non-PLHP based on a publicly available list from the Robert Wood Johnson Foundation,

 which we enhanced to include additional PLHPs based on lists from McKinsey and Avalere (

 When there was uncertainty about PLHP status, we conducted an internet search to verify. We obtained region and profit status from the December 2016 MA enrollment list and patient risk from 2015 CMS plan payment data.

To compare how outcomes differed between PLHP and non-PLHP contracts, we constructed multivariable linear regression models using generalized estimating equations with exchangeable correlation matrices to account for correlation between contracts within health plans. For example, Aetna’s health plan offered 25 MA contracts in our data set. The model controlled for accessible covariates identified as meaningful from existing literature,

 including MA region, contract profit status, average MA patient risk score, and the following covariates, all of which were derived from Area Health Resources Files

 and weighted for county contract enrollment: percent urban residence, percent black/African American, mean per capita income, college education among population 25 years or older, percent poverty among population 65 years or older, population 65 years or older per 1000 population, hospital beds per 1000 population, and active physicians per 1000 population. Each contract was analytically weighted by enrollee number.

We conducted subgroup analyses to evaluate how the association between PLHP contracts and outcomes differed by PLHP size, profit status (for-profit vs nonprofit), and MA region. To assess outcome differences by size, we compared outcomes of the 6 PLHPs with at least 100,000 enrollees (Kaiser Permanente, UPMC, Healthfirst, Spectrum, Innovacare, and Tufts) with those of the remaining PLHPs. To assess PLHP effects stratified by region, we mapped our model results for each MA region, differentiating areas where PLHPs performed significantly better than non-PLHPs, worse than non-PLHPs, or where there was no difference. Subgroup analyses were based on the multivariable model above, except for regional analyses. Regional analyses only adjusted for profit status and patient risk score because the inclusion of additional covariates prevented the model from producing estimates for many regions. Finally, to explore whether our findings were driven by Kaiser Permanente, a notably high-quality plan, we ran our base-case models after excluding Kaiser Permanente contracts.

Analyses were performed using Stata 14 (StataCorp LP; College Station, Texas) and SAS 9.4 (SAS Institute Inc; Cary, North Carolina). Further information on data sources, variable definitions, and missing data is available in eAppendix Table 1 and 

Our study population included 64 contracts offered by 31 PLHPs (representing 3,197,284 enrollees) and 311 contracts offered by 55 non-PLHPs (representing 13,881,210 enrollees) (

). Unadjusted mean star ratings, effectiveness, access, and patient satisfaction were higher among PLHPs compared with non-PLHPs, whereas procedure rates, inpatient discharges, and inpatient days were lower.

In adjusted models, PLHPs were associated with higher star ratings (β = 0.41; 95% CI, 0.15-0.67), effectiveness (β = 3.11; 95% CI, 1.43-4.80), and patient satisfaction (β = 0.57; 95% CI, 0.30-0.84) compared with non-PLHPs. Procedure rates were lower for PLHPs than non-PLHPs (β = —0.47; 95% CI, –0.79 to –0.16). There were no significant differences in access, inpatient discharges, inpatient days, and readmission probability (Table 1).

 illustrates the results from subgroup analyses, which demonstrated that larger PLHPs had significantly higher star ratings (β = 0.57; 95% CI, 0.28-0.86), effectiveness (β = 4.63; 95% CI, 2.45-6.80), and patient satisfaction (β = 0.78; 95% CI, 0.54-1.02) compared with smaller PLHPs. After excluding Kaiser Permanente, PLHPs still had significantly higher star ratings (β = 0.22; 95% CI, 0.01-0.43), effectiveness (β = 1.22; 95% CI, 0.47-1.97), and patient satisfaction (β = 0.43; 95% CI, 0.12-0.73). Compared with nonprofit PLHPs, for-profit PLHPs had significantly lower star ratings (β = —1.07; 95% CI, –1.62 to –0.52), effectiveness (β = –5.59; 95% CI, –8.22 to –2.96), access (β = –7.66; 95% CI, –12.27 to –3.06), and patient satisfaction (β = –1.38; 95% CI, –2.70 to –0.06).

reflects the significance and direction of differences between PLHPs and non-PLHPs for each outcome by MA region. PLHPs performed significantly better than non-PLHPs in the following number of regions and outcomes, respectively: 6 of 16 regions for effectiveness, 5 of 16 for procedure frequency, 4 of 16 for inpatient days, 3 of 16 for star ratings and inpatient discharges, 3 of 12 for patient satisfaction, and 2 of 16 for access and readmission probability. PLHPs performed significantly worse than non-PLHPs in the following number of regions and outcomes, respectively: 2 of 16 for access, inpatient days, and readmission probability; 1 of 16 for selected procedure frequency; 1 of 12 for patient satisfaction; and 0 of 16 for star ratings, effectiveness, and inpatient discharges. Although regions in which PLHPs performed better than non-PLHPs generally varied by outcome, PLHPs performed consistently better in most outcomes in regions that included Texas, Illinois, and Wisconsin (

The results of our analyses show that MA contracts offered by PLHPs are associated with greater quality and patient satisfaction and decreased procedures. We further found that the effects of PLHP vary by size, nonprofit status, and region, with larger and nonprofit PLHPs performing better than their smaller and for-profit counterparts, respectively.

Our results on healthcare quality and patient satisfaction are consistent with findings by Johnson et al and Lyon et al, who found that PLHP MA contracts were associated with superior performance in quality and satisfaction measures.

 Prior literature on the impact of vertical integration on utilization has shown mixed results, with some studies’ results suggesting decreased utilization

 and others’ suggesting increased utilization.

 Although few studies have investigated access, Lyon et al demonstrated decreased provider access among PLHPs.

 Comparison with prior studies is nevertheless challenging, as some failed to differentiate between vertical and other forms of integration (eg, horizontal integration through provider consolidation), did not distinguish between separate outcome domains, and lacked standardized, timely outcomes. Our results are an important and distinctive contribution to existing literature because we evaluated the association between PLHPs and separate but standardized outcomes that reflect quality, access, utilization, and patient satisfaction.

PLHPs’ higher quality and satisfaction performance could be due to multiple factors. PLHPs may leverage the strengths and resources of insurers and providers to achieve common goals of delivering high-quality patient care. Potential resources include enhanced coordination between insurers and providers, use of unified electronic health records, integration of initiatives focused on high-value care, and streamlined interactions between patients and providers. Importantly, the populations served by PLHPs were lower risk than non-PLHPs and had nonsignificant trends toward being wealthier, more educated, and with fewer minorities. Although we adjusted for these differences in multivariable models, it is possible that we observed superior quality outcomes among PLHPs due to demographic-based differences.

Nevertheless, we observed no differences in access, inpatient days, discharges, or readmissions, thus identifying a need for PLHPs to streamline and optimize utilization.

We uniquely assessed how the effect of PLHPs differs with plan characteristics, identifying important differences in outcomes based on size, nonprofit status, and region. This suggests that not all PLHPs are alike, with the

heterogeneity potentially being caused by multiple factors. First, plans could have differences in organizational commitment to their populations. For example, commitment could be stronger in regions that emphasize population health and among larger plans that bear increased risk. Second, the complex relationship between plan enrollment and quality is important to consider: On one hand, larger plans seem more established as PLHPs and, therefore, may have more experience with the model, resulting in higher quality and efficiency; on the other hand, large plan size could result from the plan itself being high quality, as higher-quality plans tend to have higher enrollment.

 Third, although we adjusted for several demographic-based covariates, it is possible that the effect of PLHPs varies by their populations’ needs. For example, populations in Texas, Wisconsin, and Illinois may benefit more from vertical integration due to varying clinical needs and demographics. Fourth, our findings that nonprofit PHLPs performed better than for-profit PLHPs are consistent with previous literature

 and suggest that population health approaches differ by profit status, potentially due to differences in underlying incentives. Fifth, there is heterogeneity in the extent of integration employed by PLHPs. For example, some PLHPs restrict providers and enrollees within their own PLHP systems, whereas in others, providers and enrollees are permitted to see non-PLHP patients and be seen by non-PLHP providers, respectively. As a result, PLHPs with less mutual exclusivity in respective provider and payer markets may have less alignment of payer—provider incentives compared with more restrictive plans. Finally, PLHPs may take advantage of varying aspects of integration in their initiatives.

 For instance, some PLHPs might have a more unified electronic health record than others, whereas other PLHPs might have more collaborative payer—provider care management programs.

Although prior studies identified characteristics associated with health plan and accountable care organization success,

 future research could investigate factors and approaches associated with successful PLHPs.

Given that this is a cross-sectional study, we can make no inferences regarding causality. It is possible that high-performing providers practice in PLHPs or that health-seeking patients enroll in such plans. Second, some plans did not have data available for certain outcomes. However, the likelihood of missing data did not significantly vary between PLHPs and non-PLHPs and is therefore unlikely to bias results (eAppendix Table 3). Third, although we adjusted for differences between PLHPs and non-PLHPs, our findings may still be subject to residual confounding due to unobserved effects. Fourth, we did not adjust for market competition, which can impact pricing, networks, and population health approaches. Finally, our outcomes are not independent because star ratings include components of HEDIS and CAHPS in their calculations. Nevertheless, assessing PLHP effects on these outcomes is valuable because star ratings represent important composite measures, whereas HEDIS and CAHPS scores inform which domains may drive star ratings differences.

As alternative payment models grow, momentum is building to integrate provision and payment of care through PLHPs. Our analysis of 2016 MA plans demonstrates the potential of such organizations to deliver high-quality care, although opportunities remain in optimizing utilization.

 Division of General Internal Medicine (NP, TRR) and School of Pharmacy (IH), University of Pittsburgh, Pittsburgh, PA; Center for High-Value Health Care, UPMC Health Plan (NP, WHS), Pittsburgh, PA.

 The authors’ time was supported by the Health Resources and Services Administration National Research Services Award T32 for Primary Medical Care Research (T32NP22240) and by the National Center for Advancing Translational Sciences of the National Institutes of Health (KL2TR001856).

 Drs Parekh and Shrank are employed by UPMC. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

 Concept and design (NP, IH, TRR, WHS); acquisition of data (NP); analysis and interpretation of data (NP, IH, WHS); drafting of the manuscript (NP, TRR); critical revision of the manuscript for important intellectual content (NP, IH, TRR, WHS); statistical analysis (NP, IH); and administrative, technical, or logistic support (WHS).

 Natasha Parekh, MD, MS, Division of General Internal Medicine, University of Pittsburgh, 200 Lothrop St, Pittsburgh, PA 15213. Email: nkp10@pitt.edu.

1. Johnson G, Lyon ZM, Frakt A. Provider-offered Medicare Advantage plans: recent growth and care quality. 

. 2017;36(3):539-547. doi: 10.1377/hlthaff.2016.0722.

2. Khanna G, Narula D, Rao N. The market evolution of provider-led health plans. McKinsey & Company website. healthcare.mckinsey.com/sites/default/files/The%20market%20evolution%20of%20provider-led%20health%20plans.pdf. Published April 2016. Accessed July 1, 2017.

3. Enthoven AC. Integrated delivery systems: the cure for fragmentation. 

4. Friedberg MW, Safran DG, Coltin KL, Dresser M, Schneider EC. Readiness for the patient-centered medical home: structural capabilities of Massachusetts primary care practices. 

. 2009;24(2):162-169. doi: 10.1007/s11606-008-0856-x.

The results of the study demonstrate the potential of provider-led health plans to deliver high-quality care and patient satisfaction. The relationships between these plans and outcomes differed by plan size, nonprofit status, and region.

William H. Shrank, MD, MSHS

Articles

Physician Satisfaction With Health Plans: Results From a National Survey

Number of Manufacturers and Generic Drug Pricing From 2005 to 2017

Relationships Between Provider-Led Health Plans and Quality, Utilization, and Satisfaction

Delivering on the Value Proposition of Precision Medicine: The View From Healthcare Payers