Quality of Medication Cost Conversations and Interest in Future Cost Conversations Among Older Adults.

BACKGROUND: Medication cost conversations occur less frequently than patients prefer, and it is unclear whether patients have positive experiences with them when they do occur. OBJECTIVE: To describe patients' experiences discussing their medication costs with their health care team. DESIGN: Cross-sectional survey. SETTING: Nationally representative survey fielded in the United States in 2022 (response rate = 48.5%). PATIENTS: 1020 adults over age 65. MEASUREMENTS: Primary measures were adapted from Clinician and Group Consumer Assessment of Healthcare Providers Survey visit survey v4.0 and captured patients' experiences of medication cost conversations. Additional measures captured patients' interest in future cost conversations, the type of clinicians with whom they would be comfortable discussing costs, and sociodemographic characteristics. RESULTS: Among 1020 respondents who discussed medication prices with their health care team, 39.3% were 75 or older and 78.6% were non-Hispanic White. Forty-three percent of respondents indicated that their prior medication cost conversation was not easy to understand; 3% indicated their health care team was not respectful and 26% indicated their health care team was somewhat respectful during their last conversation; 48% indicated that there was not enough time. Those reporting that their prior discussion was not easy to understand or that their clinician was not definitely respectful were less likely to be interested in future discussions. Only 6% and 10% of respondents indicated being comfortable discussing medication prices with financial counselors or social workers, respectively. Few differences in responses were observed by survey participant characteristics. LIMITATIONS: This cross-sectional survey of prior experiences may be subject to recall bias. CONCLUSION: Among older adults who engaged in prior medication cost conversations, many report that these conversations are not easy to understand and that almost one-third of clinicians were somewhat or not respectful. Efforts to increase the frequency of medication cost conversations should consider parallel interventions to ensure the discussions are effective at informing prescribing decisions and reducing cost-related medication nonadherence.

Physician and facility drivers of spending variation in locoregional prostate cancer.

BACKGROUND: Prostate cancer is the most common male cancer, with a wide range of treatment options. Payment reform to reduce unnecessary spending variation is an important strategy for reducing waste, but its magnitude and drivers within prostate cancer are unknown. METHODS: In total, 38,971 men aged ≥66 years with localized prostate cancer who were enrolled in Medicare fee-for-service and were included in the Surveillance, Epidemiology, and End Results-Medicare database from 2009 to 2014 were included. Multilevel linear regression with physician and facility random effects was used to examine the contributions of urologists, radiation oncologists, and their affiliated facilities to variation in total patient spending in the year after diagnosis within geographic region. The authors assessed whether spending variation was driven by patient characteristics, disease risk, or treatments. Physicians and facilities were sorted into quintiles of adjusted patient-level spending, and differences between those that were high-spending and low-spending were examined. RESULTS: Substantial variation in spending was driven by physician and facility factors. Differences in cancer treatment modalities drove more variation across physicians than differences in patient and disease characteristics (72% vs 2% for urologists, 20% vs 18% for radiation oncologists). The highest spending physicians spent 46% more than the lowest and had more imaging tests, inpatient care, and radiotherapy spending. There were no differences across spending quintiles in the use of robotic surgery by urologists or the use of brachytherapy by radiation oncologists. CONCLUSIONS: Significant differences were observed for patients with similar demographics and disease characteristics. This variation across both physicians and facilities suggests that efforts to reduce unnecessary spending must address decision making at both levels.

A Randomized Trial of Displaying Paid Price Information on Imaging Study and Procedure Ordering Rates.

BACKGROUND: Prior studies have demonstrated how price transparency lowers the test-ordering rates of trainees in hospitals, and physician-targeted price transparency efforts have been viewed as a promising cost-controlling strategy. OBJECTIVE: To examine the effect of displaying paid-price information on test-ordering rates for common imaging studies and procedures within an accountable care organization (ACO). DESIGN: Block randomized controlled trial for 1 year. SUBJECTS: A total of 1205 fully licensed clinicians (728 primary care, 477 specialists). INTERVENTION: Starting January 2014, clinicians in the Control arm received no price display; those in the intervention arms received Single or Paired Internal/External Median Prices in the test-ordering screen of their electronic health record. Internal prices were the amounts paid by insurers for the ACO's services; external paid prices were the amounts paid by insurers for the same services when delivered by unaffiliated providers. MAIN MEASURES: Ordering rates (orders per 100 face-to-face encounters with adult patients): overall, designated to be completed internally within the ACO, considered "inappropriate" (e.g., MRI for simple headache), and thought to be "appropriate" (e.g., screening colonoscopy). KEY RESULTS: We found no significant difference in overall ordering rates across the Control, Single Median Price, or Paired Internal/External Median Prices study arms. For every 100 encounters, clinicians in the Control arm ordered 15.0 (SD 31.1) tests, those in the Single Median Price arm ordered 15.0 (SD 16.2) tests, and those in the Paired Prices arms ordered 15.7 (SD 20.5) tests (one-way ANOVA p-value 0.88). There was no difference in ordering rates for tests designated to be completed internally or considered to be inappropriate or appropriate. CONCLUSIONS: Displaying paid-price information did not alter how frequently primary care and specialist clinicians ordered imaging studies and procedures within an ACO. Those with a particular interest in removing waste from the health care system may want to consider a variety of contextual factors that can affect physician-targeted price transparency.

Impact of the Rochester Medical Home Initiative on Primary Care Practices, Quality, Utilization, and Costs.

BACKGROUND: Patient-centered medical homes (PCMH) may improve the quality of primary care while reducing costs and utilization. Early evidence on the effectiveness of PCMH has been mixed. OBJECTIVES: We analyze the impact of a PCMH intervention in Rochester NY on costs, utilization, and quality of care. RESEARCH DESIGN: A propensity score-matched difference-in-differences analysis of the effect of the PCMH intervention relative to a comparison group of practices. Qualitative interviews with PCMH practice managers on their experiences and challenges with PCMH practice transformation. SUBJECTS: Seven pilot practices and 61 comparison practices (average of 36,531 and 30,192 attributed member months per practice, respectively). Interviews with practice leaders at all pilot sites. MEASURES: Individual HEDIS quality measures of preventive care, diabetes care, and care for coronary artery disease. Utilization measures of hospital use, office visits, imaging and laboratory tests, and prescription drug use. Cost measures are inpatient, prescription drug, and total spending. RESULTS: After 3 years, PCMH practices reported decreased ambulatory care sensitive emergency room visits and use of imaging tests, and increased primary care visits and laboratory tests. Utilization of prescription drugs increased but drug spending decreased. PCMH practices reported increased rates of breast cancer screening and low-density lipid screening for diabetes patients, and decreased rates of any prevention quality indicator. CONCLUSIONS: The PCMH model leads to significant changes in patient care, with reductions in some services and increases in others. This study joins a growing body of work that finds no effect of PCMH transformation on total health care spending.

Dominated choices and Medicare Advantage enrollment.

Research in behavioral economics suggests that certain circumstances, such as large numbers of complex options or revisiting prior choices, can lead to decision errors. This paper explores the enrollment decisions of Medicare beneficiaries in the Medicare Advantage (MA) program. During the time period we study (2007-2010), private fee-for-service (PFFS) plans offered enhanced benefits beyond those of traditional Medicare (TM) without any restrictions on physician networks, making TM a dominated choice relative to PFFS. Yet more than three quarters of Medicare beneficiaries remained in TM during our study period. We analyze the role of status quo bias in explaining this pattern of enrollment. Our results suggest that status quo bias plays an important role; the rate of MA enrollment was significantly higher among new Medicare beneficiaries than among incumbents. These results illustrate the importance of the choice environment that is in place when enrollees first enter the Medicare program.

Tiered network health plans and changes in physician practice intensity.

OBJECTIVE: To examine whether physicians in tiered physician networks where tier assignments are based on "intensity" of care, which is the quantity of resources used per-episode of care, change their intensity after learning detailed information about how their intensity compares to their peers. DATA SOURCES: Administrative data on intensity and quality at the physician-episode level for all physicians included in a tiered physician network offered through the Massachusetts Group Insurance Commission (GIC) in 2010-2015. Data on physicians' share of revenue from GIC patients from the 2012 Massachusetts All-Payer Claims Database. STUDY DESIGN: For 21,086 physicians in seven specialties, we estimate the impact of the dissemination of detailed intensity performance information in 2014 on physician intensity per episode of care overall and decomposed into physician services, facility, and pharmaceutical subcomponents. Intensity outcomes were measured using a standardized price schedule. Using a difference-in-differences regression, we compared physicians with high exposure to the tiered network via a large share of their revenue coming from GIC patients ("GIC share") to physicians who were less exposed. Measures of intensity of care and GIC share were log-transformed, and models controlled for physician-episode type fixed effects. DATA EXTRACTION METHODS: We linked GIC share to administrative data using National Provider Identifier. PRINCIPAL FINDINGS: There were no statistically significant differences in total intensity of care with the informational intervention for physicians in procedure-based specialties (-0.12 elasticity of intensity per episode with respect to GIC patient share, 95% CI -0.30 to 0.06) or in relationship-based specialties (0.09, 95% CI -0.15 to 0.33). There were also no differences in intensity of subcomponents of care following the intervention. CONCLUSIONS: Tiered network incentives had no detectable impact on intensity of care that physicians provided to patients.

Quality-of-Care Outcomes in Vertical Relationships Between Physicians and Health Systems.

Importance: Vertical relationships (ownership, affiliations, joint contracting) between physicians and health systems are increasing in the US. Many proponents of vertical relationships argue that increased spending associated with consolidation is accompanied by improvements in quality of care. Objective: To assess the association of vertical relationships between primary care physicians (PCPs) and large health systems and quality of care. Design, Setting, and Participants: This stacked difference-in-differences study compared outcomes for patients whose attributed PCP entered a vertical relationship with a large system in 2015 or 2017 to patients whose PCP was either never or always in a vertical relationship with a large system from 2013 to 2017. Models account for differences between PCPs, patient characteristics, market concentration, and secular trends. Data were derived from the 2013 to 2017 Massachusetts All-Payer Claims Database. The study population included commercially insured individuals attributed to a PCP in the Massachusetts Health Quality Partners' Massachusetts Provider Database in 2013, 2015, or 2017. Analyses were conducted between January 2021 and January 2024. Exposure: PCPs attributed to patients in the study entering a vertical relationship with a large health system in 2015 or 2017. Main Outcomes and Measures: Low-value care utilization, posthospitalization follow-up, utilization among patients with ambulatory care-sensitive conditions, practice site visit fragmentation, and timeliness of specialty care. Results: The study population included 4 603 172 patient-year observations from 2013 to 2017. Among all patients in the study, 53.5% were female, 35.3% had any chronic condition, and the mean (SD) age was 38.9 (20.3) years. There was no association between vertical relationships and low-value care or ambulatory care-sensitive conditions utilization. A patient's PCP entering a vertical relationship had no association with the probability of follow-up within 90 days of cancer diagnosis with any oncologist but was associated with a 7.34-percentage point (pp) (95% CI, 2.28-12.40; P = .01) increase in the probability of follow-up with an oncologist in the health system. Vertical relationships were associated with increased posthospitalization follow-up with a physician in the health system by 7.51 pp (95% CI, 2.96-12.06: P = .001) in the 2015 subgroup. PCP-health system vertical relationships were associated with a significant decrease in fragmentation of practice site visits of -1.05 pp (95% CI, -2.05 to 0.05; P = .04). Conclusions and Relevance: In this study, vertical relationships between PCPs and large health systems were associated with patient steering and changes in care delivery processes, but not necessarily improvements in patient outcomes.

Tiered Physician Network Plans and Patient Choices of Specialist Physicians.

Importance: Tiered physician network (TPN) health plans sort physicians into tiers based on their cost and quality, and patients pay lower copays for visits with physicians in the lower-cost and better-quality tiers. When the plans are first introduced, they lead patients to seek care from higher-value physicians. Objectives: To examine whether TPNs are associated with patient choice of physician when the plans have been in place for 8 to 12 years and whether there are inequities in patient out-of-pocket costs associated with inequities in access to physicians in lower-copay tiers. Design, Setting, and Participants: This cross-sectional study comprising 46 645 physicians and 585 399 patients in TPNs, including 54 683 patients who had a new patient visit with a physician in a TPN, used health insurance claims data from a large employer purchaser from July 1, 2014, to June 30, 2019. Statistical analysis was performed from November 2020 to August 2023. Exposure: Evaluation and management visit with a physician in a TPN. Main Outcomes and Measures: Main outcomes were new patient market share per physician-carrier-zip code-year, distance from centroid of patient zip code to centroid of zip code of nearest low- or medium-copay physician, and mean TPN physician office visit copay per patient. A regression discontinuity design was used to estimate the association of a physician's tier ranking, and a difference-in-differences analysis was used to estimate the association of copayment differences across tiers with market share among new patients. Equity in access was measured by comparing travel distance to the nearest physician in a low-copay or medium-copay tier and mean copayments across patient incomes. Results: The main analysis sample included 46 645 physician-carrier-zip code-year observations, 9506 (20.4%) of which were in the low-copay tier, 31 798 (68.2%) in the medium-copay tier, and 5341 (11.5%) in the high-copay tier. The 54 683 new patients in the sample had a mean (SD) age of 46.4 (16.7) years and included 33 542 women (61.3%). There was no association of having a worse tier ranking (0.045 percentage points [95% CI, -0.058 to 0.148 percentage points]) or of copayment differences between tiers (0.001 percentage points [95% CI, -0.002 to 0.004 percentage points]) with physician market share among new patients. The patients with the lowest income paid slightly lower mean (SD) copayments for office visits to a TPN physician than the patients with high income ($48.08 [$16.42] vs $51.59 [$16.79], a 6.8% difference). Conclusions and Relevance: In this cross-sectional study of TPN health plans, there was no association between physician tier ranking and physician market share among any group of patients. These findings suggest there are limitations in TPNs' steering of patients toward high-value physicians. These plans were not associated with exacerbated health inequity in this setting.

Variation in tiered network health plan penetration and local provider market characteristics.

OBJECTIVE: To understand variation in enrollment in tiered network health plans (TNPs) and the local provider market characteristics associated with TNP penetration. DATA SOURCES AND STUDY SETTING: We used 2013-2017 Massachusetts three-digit ZIP code level employer-sponsored health insurance enrollment data, data on physician horizontal and vertical affiliations from the Massachusetts Provider Database, state hospital reports in 2013, 2015, and 2017, and the 2013-2017 Massachusetts All-Payer Claims database. STUDY DESIGN: Linear regressions were used to estimate associations between TNP and local provider market characteristics. DATA EXTRACTION: We constructed measures of TNP penetration and local provider market characteristics and linked these data using three-digit ZIP code. PRINCIPAL FINDINGS: TNP penetration was at least 10% in all employer market sectors and highest among jumbo sized employers. All state employee health plan enrollees were in a tiered network health plan. Among enrollees not in the state employee health plan, TNP penetration varied from 6.0% to 19.6% across three-digit ZIP codes in Massachusetts. TNP penetration was higher in areas with less horizontal and vertical physician market concentration. CONCLUSIONS: Market competition, rather than the absolute quantity of physicians in an area, is associated with TNP penetration.

Utilization, Steering, and Spending in Vertical Relationships Between Physicians and Health Systems.

Importance: Vertical relationships (eg, ownership or affiliations, including joint contracting) between physicians and health systems are increasing in the US. Objective: To analyze how vertical relationships between primary care physicians (PCPs) and large health systems are associated with changes in ambulatory and acute care utilization, referral patterns, readmissions, and total medical spending for commercially insured individuals. Design, Setting, and Participants: This case-control study with a repeated cross-section, stacked event design analyzed outcomes of patients whose attributed PCP entered a vertical relationship with a large health care system in 2015 or 2017 compared with patients whose attributed PCP was either never or always in a vertical relationship with a large health system from 2013 to 2017 in the state of Massachusetts. The sample consisted of commercially insured patients who met enrollment criteria and who were attributed to PCPs who were included in the Massachusetts Provider Database in 2013, 2015, and 2017 and for whom vertical relationships were measured. Enrollee and claims data were obtained from the 2013 to 2017 Massachusetts All-Payer Claims Database. Statistical analyses were conducted between January 5, 2021, and June 5, 2023. Exposure: Evaluation-and-management visit with attributed PCP in 2015 to 2017. Main Outcomes and Measures: Outcomes (which were measured per patient-year [ie, per patient per year from January to December] in this sample) were utilization (count of specialist physician visits, emergency department [ED] visits, and hospitalizations overall and within attributed PCP's health system), spending (total medical expenditures and use of high-price hospitals), and readmissions (readmission rate and use of hospitals with a low readmission rate). Results: The sample of 4 030 224 observations included 2 147 303 females (53.3%) and 1 881 921 males (46.7%) with a mean (SD) age of 35.07 (19.95) years. Vertical relationships between PCPs and large health systems were associated with an increase of 0.69 (95% CI, 0.34-1.04; P < .001) in specialist visits per patient-year, a 22.64% increase vs the comparison group mean of 3.06 visits, and a $356.67 (95% CI, $77.16-$636.18; P = .01) increase in total medical expenditures per patient-year, a 6.26% increase vs the comparison group mean of $5700.07. Within the health care system of the attributed PCPs, the number of specialist visits changed by 0.80 (95% CI, 0.56-1.05) per patient year (P < .001), a 29.38% increase vs the comparison group mean of 2.73 specialist visits per patient-year. The number of ED visits changed by 0.02 (95% CI, 0.01-0.03) per patient year (P = .001), a 14.19% increase over the comparison group mean of 0.15 ED visits per patient-year. The number of hospitalizations changed by 0.01 (95% CI, 0.00-0.01) per patient-year (P < .001), a 22.36% increase over the comparison group mean of 0.03 hospitalizations per patient-year. There were no differences in readmission outcomes. Conclusions: Results of this case-control study suggest that vertical relationships between PCPs and large health systems were associated with steering of patients into health systems and increased spending on patient care, but no difference in readmissions was found.

Utilization and Spending With Preventive Drug Lists for Asthma Medications in High-Deductible Health Plans.

Importance: High-deductible health plans with health savings accounts (HDHP-HSAs) incentivize patients to use less health care, including necessary care. Preventive drug lists (PDLs) exempt high-value medications from the deductible, reducing out-of-pocket cost sharing; the associations of PDLs with health outcomes among patients with asthma is unknown. Objective: To evaluate the associations of a PDL for asthma medications on utilization, adverse outcomes, and patient spending for HDHP-HSA enrollees with asthma. Design, Setting, and Participants: This case-control study used matched groups of patients with asthma before and after an insurance design change using a national commercial health insurance claims data set from 2004-2017. Participants included patients aged 4 to 64 years enrolled for 1 year in an HDHP-HSA without a PDL in which asthma medications were subject to the deductible who then transitioned to an HDHP-HSA with a PDL that included asthma medications; these patients were compared with a matched weighted sample of patients with 2 years of continuous enrollment in an HDHP-HSA without a PDL. Models controlled for patient demographics and asthma severity and were stratified by neighborhood income. Analyses were conducted from October 2020 to June 2023. Exposures: Employer-mandated addition of a PDL that included asthma medications to an existing HDHP-HSA. Main Outcomes and Measures: Outcomes of interest were utilization of asthma medications on the PDL (controllers and albuterol), asthma exacerbations (oral steroid bursts and asthma-related emergency department use), and out-of-pocket spending (all and asthma-specific). Results: A total of 12 174 participants (mean [SD] age, 36.9 [16.9] years; 6848 [56.25%] female) were included in analyses. Compared with no PDL, PDLs were associated with increased rates of 30-day fills per enrollee for any controller medication (change, 0.10 [95% CI, 0.03 to 0.17] fills per enrollee; 12.9% increase) and for combination inhaled corticosteroid long-acting ß2-agonist (ICS-LABA) medications (change, 0.06 [95% CI, 0.01 to 0.10] fills per enrollee; 25.4% increase), and increased proportion of days covered with ICS-LABA (6.0% [0.7% to 11.3%] of days; 15.6% increase). Gaining a PDL was associated with decreased out-of-pocket spending on asthma care (change, -$34 [95% CI, -$47 to -$21] per enrollee; 28.4% difference), but there was no significant change in asthma exacerbations and no difference in results by income. Conclusions and Relevance: In this case-control study, reducing cost-sharing for asthma medications through a PDL was associated with increased adherence to controller medications, notably ICS-LABA medications used by patients with more severe asthma, but was not associated with improved clinical outcomes. These findings suggest that PDLs are a potential strategy to improve access and affordability of asthma care for patients in HDHP-HSAs.

The association between care integration and care quality.

OBJECTIVE: The study aims to analyze the relationship between care integration and care quality, and to examine if the relationship varies by patient risk. DATA SOURCES AND STUDY SETTING: The key independent variables used validated measures derived from a provider survey of functional (i.e., administrative and clinical systems) and social (i.e., patient integration, professional cooperation, professional coordination) integration. Survey responses represented data from a stratified sample of 59 practice sites from 17 health systems. Dependent variables included three quality measures constructed from patient-level Medicare data: colorectal cancer screening among patients at risk, patient-level 30-day readmission, and a practice-level Healthcare Effectiveness Data and Information Set (HEDIS) composite measure of publicly reported, individual measures of ambulatory clinical quality performance. DATA COLLECTION/EXTRACTION METHOD: We obtained quality- and beneficiary-level covariate data for the 41,966 Medicare beneficiaries served by the 59 practices in our survey sample. STUDY DESIGN: We estimated hierarchical linear models to examine the association between care integration and care quality and the moderating effect of patients' clinical risk score. We graphically visualized the moderating effects at ±1 standard deviation of our z-standardized independent and moderating variables and performed simple slope tests. PRINCIPAL FINDINGS: Our analyses uncovered a strong positive relationship between social integration, specifically patient integration, and the quality of care a patient receives (e.g., a 1-point increase in a practice's patient integration was associated with 0.31-point higher HEDIS composite score, p < 0.01). Further, we documented positive and significant associations between aspects of social and functional integration on quality of care based on patient risk. CONCLUSIONS: The findings suggest social integration matters for improving the quality of care and that the relationship of integration to quality is not uniform for all patients. Policymakers and practitioners considering structural integrations of health systems should direct attention beyond structure to consider the potential for social integration to impact outcomes and how that might be achieved.

Cost-Related Medication Nonadherence and Desire for Medication Cost Information Among Adults Aged 65 Years and Older in the US in 2022.

Importance: Rising prescription drug costs and increasing prices for consumer goods may increase cost-related medication nonadherence. Cost-conscious prescribing can be supported by real-time benefit tools, but patient views on real-time benefit tool use and their potential benefits and harms are largely unexplored. Objective: To assess older adults' cost-related medication nonadherence, cost-coping strategies, and views on the use of real-time benefit tools in clinical practice. Design, Setting, and Participants: A weighted, nationally representative survey of adults aged 65 years and older administered via the internet and telephone from June 2022 to September 2022. Main Outcomes and Measures: Cost-related medication nonadherence; cost coping strategies; desire for cost conversations; potential benefits and harms from real-time benefit tool use. Results: Among 2005 respondents, most were female (54.7%) and partnered (59.7%); 40.4% were 75 years or older. Cost-related medication nonadherence was reported by 20.2% of participants. Some respondents used extreme forms of cost-coping, including foregoing basic needs (8.5%) or going into debt (4.8%) to afford medications. Of respondents, 89.0% reported being comfortable or neutral about being screened before a physician's visit for wanting to have medication cost conversations and 89.5% indicated a desire for their physician to use a real-time benefit tool. Respondents expressed concern if prices were inaccurate, with 49.9% of those with cost-related nonadherence and 39.3% of those without reporting they would be extremely upset if their actual medication price was more than what their physician estimated with a real-time benefit tool. If the actual price was much more than the estimated real-time benefit tool price, nearly 80% of respondents with cost-related nonadherence reported that it would affect their decision to start or keep taking a medication. Furthermore, 54.2% of those with any cost-related nonadherence and 30% of those without reported they would be moderately or extremely upset if their physicians used a medication price tool but chose not to discuss prices with them. Conclusions and Relevance: In 2022, approximately 1 in 5 older adults reported cost-related nonadherence. Real-time benefit tools may support medication cost conversations and cost-conscious prescribing, and patients are enthusiastic about their use. However, if disclosed prices are inaccurate, there is potential for harm through loss of confidence in the physician and nonadherence to prescribed medications.

Price Effects Of Vertical Integration And Joint Contracting Between Physicians And Hospitals In Massachusetts.

Vertical integration in health care has recently garnered scrutiny by antitrust authorities and state regulators. We examined trends, geographic variation, and price effects of vertical integration and joint contracting between physicians and hospitals, using physician affiliations and all-payer claims data from Massachusetts from the period 2013-17. Vertical integration and joint contracting with small and medium health systems rose from 19.5 percent in 2013 to 32.8 percent in 2017 for primary care physicians and from 26.1 percent to 37.8 percent for specialists. Vertical integration and joint contracting with large health systems slightly declined, whereas geographic variation in these physician affiliations rose. We found that vertical integration and joint contracting led to price increases from 2013 to 2017, from 2.1 percent to 12.0 percent for primary care physicians and from 0.7 percent to 6.0 percent for specialists, with the greatest increases seen in large health systems. These findings can inform policy makers seeking to limit growth in health care prices.

Affordability and Access Challenges Among US Subscribers to Nongroup Insurance Plans.

This cohort study assesses cost-related experiences in non-group plans purchased on or off Marketplace and variation by Marketplace enrollment, decision support use, and other characteristics.

Physician Perspectives on Implementation of Real-Time Benefit Tools: A Qualitative Study.

OBJECTIVES: Congress and Medicare have required real-time benefit tools (RTBT) to provide patient-specific medication price information during prescribing to decrease the cost of medications. We sought physicians' perspectives on how these tools might most effectively improve their selection of low-cost medication. METHODS: We conducted 15 semi-structured interviews of physicians (6 oncologists, 1 endocrinologist, 4 rheumatologists, and 4 from internal medicine) and identified key themes across interviews during coding and analysis. RESULTS: Although physicians saw value in real-time medication price information, they were wary of the complexity of obtaining specific information and the potential for inaccuracies. Physicians described how medication price information would be used in various prescribing scenarios including from simple substitutions (different drug formulations) to more complex decisions (different drug classes). In more complex situations, physicians were concerned that price information might only be available after discussing options with the patient, which would be too late to inform decisions. Concern about adding more information to the electronic health record was common. CONCLUSION: While most physicians saw value in implementation of RTBTs, they also expressed concerns related to the accuracy of information, the availability of information at the right time in the clinical workflow, and the most effective format for information. Many concerns raised paralleled the "Five Rights of Clinical Decision Support" framework and indicate the need for additional design work to achieve benefit from RTBTs. Beyond the public policy that has supported the availability of RTBTs, substantial development will be required to ensure that information is used to improve prescribing decisions.

Social Features of Integration in Health Systems and Their Relationship to Provider Experience, Care Quality and Clinical Integration.

More is known about the structural features of health system integration than the social features-elements of normative integration (alignment of norms) and interpersonal integration (collaboration among professionals and with patients). We surveyed practice managers and 1,360 staff and physicians at 59 practice sites within 17 health systems (828 responses; 61%). Building on prior theory, we developed and established the psychometric properties of survey measures describing normative and interpersonal integration. Normative and interpersonal integration were both consistently related to better provider experience, perceived care quality, and clinical integration (e.g., a 1-point increase in a practice's normative integration was associated with 0.53-point higher job satisfaction and 0.77-point higher perceived care quality in the practice, measured on 1 to 5 scales, p < .01). Variation in social features of integration may help explain why some health systems better integrate care, pointing to normative and interpersonal integration as potential resources for improvement.