Long-Term Clinical Outcomes Following the WATCHMAN Device Use in Medicare Beneficiaries.

BACKGROUND: Long-term outcomes following left atrial appendage occlusion outside clinical trials and small registries are largely unknown. Collecting these data was a condition of US market authorization of the WATCHMAN device. The aim of this analysis was to evaluate the rates of stroke, bleeding, and death among Medicare beneficiaries following left atrial appendage occlusion implantation during initial commercial availability of the WATCHMAN left atrial appendage occlusion device overall and in important subgroups. METHODS: All Medicare fee-for-service beneficiaries ≥65 years of age who underwent left atrial appendage occlusion from April 1, 2016, to August 31, 2020, were included based on the International Classification of Diseases, Tenth Revision, and Current Procedural Terminology codes. Over a 5-year follow-up period, the cumulative incidence over time of mortality, ischemic stroke, and major bleeding were calculated using the International Classification of Diseases, Tenth Revision, diagnosis codes for the full study cohort and within important prespecified subgroups. RESULTS: WATCHMAN recipients (n=48 763) were a median of 77 (interquartile range, 72-82) years of age, 42% female, and mostly White (93%). The median CHA2DS2VASc score was 4 (interquartile range, 3-5) with prior major bleeding in 42% and prior stroke in 12%. At 5 years, death occurred in 44%, bleeding in 15% (with higher risk early following implantation), and ischemic stroke in 7%. Each of these end points was more common with greater baseline age. Male patients had greater 5-year mortality than female patients (46.9% versus 40.6%), but there was no difference between sexes in the rates of ischemic stroke (6.6% versus 7.5%) or major bleeding (14.9% for both). WATCHMAN recipients with prior ischemic stroke or a major bleeding event were older and frailer; these groups had higher rates of ischemic stroke, major bleeding, and death. CONCLUSIONS: Compared with patients enrolled in the pivotal clinical trials, Medicare beneficiaries undergoing WATCHMAN implantation were older, more female, and had more comorbid conditions. Substantial long-term mortality and major bleeding following WATCHMAN reflect the high-risk nature of the patient population, while the ischemic stroke rate was relatively low (<1.5% per year).

Equity considerations for the implementation of health insurance benefit package in Ethiopia: result of expert Delphi exercise.

BACKGROUND: Efficiency, equity and financial risk protection are key health systems objectives. Equitable distribution of health care is among the priority strategic initiative of the government of Ethiopia. However, data on the distribution of interventions benefits or on disease burden disaggregated by subpopulations to guide health care priority setting is not available in Ethiopia. METHODS: Aligned with policy documents, we identified the following groups to be the worse off in the Ethiopian context: under-five children, women of reproductive age, the poor, and rural residents. We used the Delphi technique by a panel of 28 experts to assign a score for 253 diseases/conditions over a period of two days, in phases. The expert panel represented different institutes and professional mix. Experts assigned a score 1 to 4; where 4 indicates disease/condition predominantly affecting the poor and rural residents and 1 indicates a condition more prevalent among the wealthy and urban residents. Subsequently, the average equity score was computed for each disease/condition. RESULTS: The average scores ranged from 1.11 (for vitiligo) to 3.79 (for obstetric fistula). We standardized the scores to be bounded between 1 and 2; 1 the lowest equity score and 2 the highest equity score. The scores for each disease/condition were then assigned to their corresponding interventions. We used these equity scores to adjust the CEA values for each of the interventions. To adjust the CEA values for equity, we multiplied the health benefits (the denominator of the cost-effectiveness value) of each intervention by the corresponding equity scores, resulting in equity adjusted CEA values. The equity adjusted CEA was then used to rank the interventions using a league table. CONCLUSIONS: The Delphi method can be useful in generating equity scores for prioritizing health interventions where disaggregated data on the distribution of diseases or access to interventions by subpopulation groups are not available.

Relationship between universal health insurance benefits and prostate cancer mortality in Colombia.

PURPOSE: Prostate cancer is the most common cause for cancer mortality among men in Colombia. Law 100, in 1993, created a contributory regime (private insurance) and subsidized regime (public insurance) in which the subsidized regime had fewer benefits. However, Ruling T760 in July 2012 mandated that both systems must offer equal quality and access to healthcare. This study examines the impact of this change on prostate cancer mortality rates before and after 2012. METHODOLOGY: Prostate cancer mortality records from 2006 to 2020 were collected from Colombia's National Administrative Department of Statistics (DANE). Crude mortality was calculated by health insurance for different geographic areas and analyzed for changes between 2006 and 2012 and 2013-2020. Join-Point regressions were used to analyze trends by health insurance. RESULTS: Crude mortality rates in the contributory regime had a non-statistically significant decrease from 2006 to 2012 (AAPC= -1.32%, P = 0.14, 95% CI= -3.12, 0.52). In contrast, between 2013 and 2020 there was a non-statistically significant increase in crude mortality (AAPC 1.10%, P = 0.07, 95% CI= -0.09, 2.31). Comparatively, crude mortality in the subsidized regime, from 2006 to 2012, increased with a statistically significant AAPC of 2.51% (P < 0.001, 95% CI = 1.21, 3.83). From 2013 to 2020, mortality continued to increase with statistically significant AAPC of 5.52% (P < 0.001, 95% CI = 4.77, 6.27). Compared to their crude mortality differences from 2006 to 2020, from 2013 to 2020, the departments of Atlántico, Córdoba, Sucre, Arauca, Cesar, and Cauca had the highest rates in prostate cancer mortality in the subsidized regime compared to the contributory regime. CONCLUSION: Ruling T760 did not positively impact prostate cancer mortality, particularly of men in the subsidized regime.

County-Level Enrollment in Medicare Advantage Plans Offering Expanded Supplemental Benefits.

Importance: Since 2019 and 2020, Medicare Advantage (MA) plans have been able to offer supplemental benefits that address long-term services and supports (LTSS) and social determinants of health (SDOH). Objective: To examine the temporal trends and geographic variation in enrollment in MA plans offering LTSS and SDOH benefits. Design, Setting, and Participants: This cross-sectional study used publicly available data to examine changes in beneficiary enrollment and plan offerings of LTSS and SDOH benefits from the benefits data from the second quarter of each year and other data from April of each year except 2024, for which the first quarter was the latest for benefits data and January the latest for other data at the time of analysis. Multivariable linear regression models for each type of benefit were used to investigate associations between county characteristics and enrollment in 2024. Analyses were stratified for (1) Dual Eligible Special Needs Plans (D-SNPs) that exclusively enroll dual-eligible beneficiaries and (2) non-D-SNPs. Main Outcomes and Measures: The percentage of MA enrollees in plans offering LTSS or SDOH benefits at the county level. Results: This study included 2 631 697 D-SNP and 20 114 506 non-D-SNP enrollees in 2020, which increased to 5 494 426 and 25 561 455, respectively, in 2024. From 2020 to 2024, the percentage of D-SNP enrollees in plans offering SDOH benefits increased from 9% to 46%, whereas the percentage fluctuated between 23% and 39% for LTSS benefits. There was an increase in non-D-SNP enrollees with LTSS (from 9% to 22%) and SDOH (from 4% to 20%) benefits from 2020 to 2023, which decreased in 2024. In 2024, the most offered LTSS benefit was in-home support services, and the most offered SDOH benefit was food and produce. The percentage of enrollees with these benefits varied across counties in 2024. In multivariable linear regression models, among D-SNPs, enrollment in plans offering any SDOH benefits was higher in counties with greater MA penetration (coefficient, 5.0 percentage points [pp] per 10-pp change; 95% CI, 2.1-7.9 pp), in urban counties (coefficient, 7.2 pp vs rural counties; 95% CI, 3.8-10.6 pp), in counties with greater enrollment in fully integrated D-SNPs (coefficient, 3.0 pp per 10-pp change; 95% CI, 2.2-3.9 pp), and in counties in states with approved Medicaid home- and community-based services waivers for individuals 65 years or older or those with disabilities (coefficient, 10.8 pp; 95% CI, 4.0-17.6 pp). Enrollment in D-SNPs offering LTSS benefits was also higher in counties with greater MA penetration (coefficient, 5.9 pp per 10-pp change; 95% CI, 2.4-9.5 pp), urban vs rural counties (coefficient, 4.6 pp; 95% CI, 1.1-8.1 pp), and counties with greater enrollment in fully integrated D-SNPs (coefficient, 3.0 pp per 10-pp change; 95% CI, 2.1-3.9 pp) in addition to counties with greater social vulnerability scores (coefficient, 1.4 pp per 10-pp change; 95% CI, 0.3-2.5 pp). Conclusions and Relevance: In this cross-sectional study of MA plans and enrollees, an increase in enrollment was most consistent in D-SNPs offering SDOH benefits compared with LTSS benefits and in D-SNPs compared with non-D-SNPs. Geographic variation in enrollment patterns highlights potential gaps in access to LTSS and SDOH benefits for rural MA beneficiaries and dual-eligible enrollees living in counties with lower enrollment in fully integrated D-SNPs and states with more limited Medicaid home- and community-based services coverage.

Prescribed Minimum Benefits complaints: a five-year retrospective review.

BACKGROUND: No matter which benefit option members have chosen, medical schemes are required by the Medical Schemes Act no. 131 of 1998 to pay costs associated with the diagnosis, treatment, or care of a specified set of benefits known as Prescribed Minimum Benefits (PMBs). Medical scheme beneficiaries have the right to lodge complaints with the Council for Medical Schemes (CMS) when their claims are denied. OBJECTIVE:  To determine and describe the pattern of PMBs complaints received by CMS from January 2014 to December 2018. METHODS:  This was a cross-sectional study that utilised the CMS' clinical complaints. Data for PMBs, complainants, medical scheme types, and reasons for payment denial were extracted. The CMS' lists of chronic conditions, PMBs, and registered schemes were used to confirm PMBs and to categorise schemes as either restricted (i.e., to only members of specific organisations) or open (i.e., to all South Africans). Extracted and coded data were analysed using SAS v.9.4 software. RESULTS:  A total of 2141 complaints were retrieved and 1124 PMBs complaints were included in the study. The median of PMBs complaints per year was 225. Most of the complaints (43.6%, n=490/1124) were lodged by members themselves. Non-Communicable Diseases (NCDs) constituted most of the PMBs conditions that members complained about. Medicine and surgery were the services that were mostly denied full payment by medical schemes. Open medical schemes accounted for more (73.8%, n=830/1124) of the complaints. CONCLUSION:  Chronic conditions are the main diseases that medical scheme members complained about. Member education and clear definition of PMBs should be prioritised by medical schemes and the Council for Medical Schemes.

The impacts of supplemental benefits on Medicare Advantage plan composition.

OBJECTIVES: In 2019 and 2020, Medicare Advantage (MA) plans received historic flexibility to begin to address members' nonmedical and social needs through a set of primarily health-related benefits (PHRBs) and Special Supplemental Benefits for the Chronically Ill (SSBCIs). We aimed to evaluate the impact of adoption on the number and composition of new MA plan enrollees. STUDY DESIGN: A difference-in-differences design of retrospective Medicare enrollment data linked to publicly available plan and county-level data. METHODS: We linked individual-level Medicare enrollment data to publicly available, plan-level MA benefit, crosswalk, and penetration files from 2016 to 2020. We compared the number of new enrollees and the proportion of new enrollees who were Black, Hispanic, younger than 65 years, partially and fully Medicare and Medicaid dual eligible, and disabled in plans that adopted a PHRB or SSBCI vs a set of matched control plans that did not. RESULTS: In fully adjusted models, PHRB adoption was associated with a 2.2% decrease in the proportion of fully dual-eligible new members (95% CI, -4.0% to -0.5%). SSBCI adoption was associated with a 2.3% decrease in the proportion of new members younger than 65 years (95% CI, -3.6% to -0.9%). After accounting for multiple comparisons, these results were no longer statistically significant. CONCLUSION: We determined that supplemental benefit adoption was not associated with demographic shifts in MA plan enrollment.

New Supplemental Benefits and Plan Ratings Among Medicare Advantage Enrollees.

Importance: In 2018, the US Congress gave Medicare Advantage (MA) historic flexibility to address members' social needs with a set of Special Supplemental Benefits for the Chronically Ill (SSBCIs). In response, the Centers for Medicare & Medicaid Services expanded the definition of primarily health-related benefits (PHRBs) to include nonmedical services in 2019. Uptake has been modest; MA plans cited a lack of evidence as a limiting factor. Objective: To evaluate the association between adopting the expanded supplemental benefits designed to address MA enrollees' nonmedical and social needs and enrollees' plan ratings. Design, Setting, and Participants: This cohort study compared the plan ratings of MA enrollees in plans that adopted an expanded PHRB, SSBCI, or both using difference-in-differences estimators with MA Consumer Assessment of Health Care Providers and Systems survey data from March to June 2017, 2018, 2019, and 2021 linked to Medicare administrative claims and publicly available benefits and enrollment data. Data analysis was performed between April 2023 and March 2024. Exposure: Enrollees in MA plans that adopted a PHRB and/or SSBCI in 2021. Main Outcomes and Measures: Enrollee plan rating on a 0- to 10-point scale, with 0 indicating the worst health plan possible and 10 indicating the best health plan possible. Results: The study sample included 388 356 responses representing 467 MA contracts and 2558 plans in 2021. Within the weighted population of responders, the mean (SD) age was 74.6 (8.7) years, 57.2% were female, 8.9% were fully Medicare-Medicaid dual eligible, 74.6% had at least 1 chronic medical condition, 13.7% had not graduated high school, 9.7% were helped by a proxy, 45.1% reported fair or poor physical health, and 15.6% were entitled to Medicare due to disability. Adopting both a new PHRB and SSBCI benefit in 2021 was associated with an increase of 0.22 out of 10 points (95% CI, 0.4-4.0 points) in mean enrollee plan ratings. There was no association between adoption of only a PHRB (adjusted difference, -0.12 points; 95% CI, -0.26 to 0.02 points) or SSBCI (adjusted difference, 0.09 points; 95% CI, -0.03 to 0.21 points) and plan rating. Conclusions and Relevance: Medicare Advantage plans that adopted both benefits saw modest increases in mean enrollee plan ratings. This evidence suggests that more investments in supplemental benefits were associated with improved plan experiences, which could contribute to improved plan quality ratings.

Medical insurance benefits and health inequality: evidence from rural China.

Alleviating health inequality among different income groups has become a significant policy goal in China to promote common prosperity. Based on the data from the China Health and Retirement Longitudinal Study (CHARLS) covering the period from 2013 to 2018, this study empirically examines the impact of Integrated Medical Insurance System (URRBMI) on the health and health inequality of older adult rural residents. The following conclusions are drawn: First, URRBMI have elevated the level of medical security, reduced the frailty index of rural residents, and improved the health status of rural residents. Second, China exhibits "pro-rich" health inequality, and URRBMI exacerbates health inequality among rural residents with different incomes. This result remains robust when replacing the frailty index with different health modules. Third, the analysis of influencing mechanisms indicates that the URRBMI exacerbate inequality in the utilization of medical services among rural residents, resulting in a phenomenon of "subsidizing the rich by the poor" and intensifying health inequality. Fourth, in terms of heterogeneity, URRBMI have significantly widened health inequality among the older adult and in regions with a higher proportion of multiple-tiered medical insurance schemes. Finally, it is suggested that China consider establishing a medical financing and benefit assurance system that is related to income and age and separately construct a unified public medical insurance system for the older adult population.

Home Health Care Use and Outcomes After Coronary Artery Bypass Grafting Among Medicare Beneficiaries.

BACKGROUND: Home health care (HHC) has been increasingly used to improve care transitions and avoid poor outcomes, but there is limited data on its use and efficacy following coronary artery bypass grafting. The purpose of this study was to describe HHC use and its association with outcomes among Medicare beneficiaries undergoing coronary artery bypass grafting. METHODS: Retrospective analysis of 100% of Medicare fee-for-service files identified 77 331 beneficiaries undergoing coronary artery bypass grafting and discharged to home between July 2016 and December 2018. The primary exposure of HHC use was defined as the presence of paid HHC claims within 30 days of discharge. Hierarchical logistic regression identified predictors of HHC use and the percentage of variation in HHC use attributed to the hospital. Propensity-matched logistic regression compared mortality, readmissions, emergency department visits, and cardiac rehabilitation enrollment at 30 and 90 days after discharge between HHC users and nonusers. RESULTS: A total of 26 751 (34.6%) of beneficiaries used HHC within 30 days of discharge, which was more common among beneficiaries who were older (72.9 versus 72.5 years), male (79.4% versus 77.4%), White (90.2% versus 89.2%), and not Medicare-Medicaid dual eligible (6.7% versus 8.8%). The median hospital-level rate of HHC use was 31.0% (interquartile range, 13.7%-54.5%) and ranged from 0% to 94.2%. Nearly 30% of the interhospital variation in HHC use was attributed to the discharging hospital (intraclass correlation coefficient, 0.296 [95% CI, 0.275-0.318]). Compared with non-HHC users, those using HHC were less likely to have a readmission or emergency department visit, were more likely to enroll in cardiac rehabilitation, and had modestly higher mortality within 30 or 90 days of discharge. CONCLUSIONS: A third of Medicare beneficiaries undergoing coronary artery bypass grafting used HHC within 30 days of discharge, with wide interhospital variation in use and mixed associations with clinical outcomes and health care utilization.

Does the increase in health insurance benefits have different effects on the health of middle-aged and older people individuals in rural areas? Analysis based on quantile difference-in-differences method.

In the context of healthy aging, enhancing health performance is an intrinsic requirement for the development and reform of the health insurance system. This paper mainly discusses the health effects of increasing medical insurance benefits on people with different levels of health. So this paper utilizes multiple rounds of data from the China Health and Retirement Longitudinal Study (CHARLS) and employs the quantile difference-in-differences method to systematically investigate the impact effects of the integration of urban and rural residents' health insurance on the frailty levels of rural middle-aged and older people individuals. The research findings are as follows: Firstly, the integration of urban and rural resident health insurance has mitigated the frailty level of rural older people individuals, with a more pronounced impact on those with poorer health statuses. Secondly, in terms of heterogeneity analysis, the health performance effects of the urban-rural health insurance integration policy are more significant among the older people population and in the western regions. Thirdly, the integration of urban and rural resident health insurance primarily improves health by reducing the burden of medical expenses, with a greater impact on the older people population with poorer health statuses. Based on the research findings, we recommend addressing the disparities in healthcare benefits across various insurance systems, alleviating the financial burden of healthcare for impoverished individuals, and consistently improving the coordination of healthcare insurance policies for both urban and rural residents.

Partner Plan Choices and Medicare Advantage Enrollment Decisions Among Older Adults.

This study examines the association of partner Medicare Advantage plan status over 1 year with beneficiary and plan characteristics.

The Rise and Risks of Medicare Advantage "Affinity Plans".

This Viewpoint discusses the proliferation of Medicare Advantage plans targeting specific groups of individuals and whether these plans will improve quality of care for beneficiaries.

Medicare Eligibility and Reported Support for Proposals to Expand Medicare.

This study estimates the association between Medicare eligibility and support for recent proposals to expand program participation and benefits.

Implementation of state health insurance benefit mandates for cancer-related fertility preservation: following policy through a complex system.

BACKGROUND: A myriad of federal, state, and organizational policies are designed to improve access to evidence-based healthcare, but the impact of these policies likely varies due to contextual determinants of, reinterpretations of, and poor compliance with policy requirements throughout implementation. Strategies enhancing implementation and compliance with policy intent can improve population health. Critically assessing the multi-level environments where health policies and their related health services are implemented is essential to designing effective policy-level implementation strategies. California passed a 2019 health insurance benefit mandate requiring coverage of fertility preservation services for individuals at risk of infertility due to medical treatments, in order to improve access to services that are otherwise cost prohibitive. Our objective was to document and understand the multi-level environment, relationships, and activities involved in using state benefit mandates to facilitate patient access to fertility preservation services. METHODS: We conducted a mixed-methods study and used the policy-optimized exploration, preparation, implementation, and sustainment (EPIS) framework to analyze the implementation of California's fertility preservation benefit mandate (SB 600) at and between the state insurance regulator, insurer, and clinic levels. RESULTS: Seventeen publicly available fertility preservation benefit mandate-relevant documents were reviewed. Interviews were conducted with four insurers; 25 financial, administrative, and provider participants from 16 oncology and fertility clinics; three fertility pharmaceutical representatives; and two patient advocates. The mandate and insurance regulator guidance represented two "Big P" (system level) policies that gave rise to a host of "little p" (organizational) policies by and between the regulator, insurers, clinics, and patients. Many little p policies were bridging factors to support implementation across levels and fertility preservation service access. Characterizing the mandate's functions (i.e., policy goals) and forms (i.e., ways that policies were enacted) led to identification of (1) intended and unintended implementation, service, and patient outcomes, (2) implementation processes by level and EPIS phase, (3) actor-delineated key processes and heterogeneity among them, and (4) inner and outer context determinants that drove adaptations. CONCLUSIONS: Following the midstream and downstream implementation of a state health insurance benefit mandate, data generated will enable development of policy-level implementation strategies, evaluation of determinants and important outcomes of effective implementation, and design of future mandates to improve fit and fidelity.

Physician responses to insurance benefit restrictions: The case of ophthalmology.

This study examines the impact of social insurance benefit restrictions on physician behaviour, using ophthalmologists as a case study. We examine whether ophthalmologists use their market power to alter their fees and rebates across services to compensate for potential policy-induced income losses. The results show that ophthalmologists substantially reduced their fees and rebates for services directly targeted by the benefit restriction compared to other medical specialists' fees and rebates. There is also some evidence that they increased their fees for services that were not targeted. High-fee charging ophthalmologists exhibited larger fee and rebate responses while the low-fee charging group raise their rebates to match the reference price provided by the policy environment.

Hip fractures after 60 years of age in France in 2005-2017: Nationwide sample of statutory-health-insurance beneficiaries.

BACKGROUND: Proximal femoral factures (PFFs) constitute a heavy medical, social, and economic burden. Overall, orthopaedic conditions vary widely in France regarding the patients involved and treatments applied. For PFFs specifically, data are limited. Moreover, the ongoing expansion of geriatric orthopaedics holds promise for improving overall postoperative survival. The objectives of this retrospective study of a nationwide French database were: 1) to describe the pathway of patients with PFFs regarding access to care, healthcare institutions involved, and times to management; 2) to look for associations linking these parameters to post-operative mortality. HYPOTHESIS: Across France, variations exist in healthcare service availability and time to management for patients with PFFs. MATERIAL AND METHODS: A retrospective analysis of data in a de-identified representative sample of statutory-health-insurance beneficiaries in France (Échantillon généraliste des bénéficiaires, EGB, containing data for 1/97 beneficiaries) was conducted. All patients older than 60 years of age who were managed for PFFs between 2005 and 2017 were included. The following data were collected for each patient: age, management method, Charlson's Comorbidity Index (CCI), home-to-hospital distance by road, and type of hospital (public, non-profit private, or for-profit private), and time to surgery were collected. The study outcomes were the incidence of PFF, mortality during the first postoperative year, changes in mortality between 2005 and 2017, and prognostic factors. RESULTS: In total 8026 fractures were included. The 7561 patients had a median age of 83.8 years and a mean CCI of 4.6; both parameters increased steadily over time, by 0.18 years and 0.06 points per year, respectively (p<10-4 for both comparisons). Management was by total hip replacement in 3299 cases and internal fixation in 4262 cases; this information was not available for 465 fractures. The overall incidence increased from 90/100,000 in 2008 to 116/100,000 in 2017 (p=0.03). Of the 8026 fractures, 5865 (73.1%) were managed in public hospitals (and this proportion increased significantly over time), 1629 (20.3%) in non-profit private hospitals (decrease over time), and 264 (3.3%) in for-profit private hospitals. The home-to-hospital distance ranged from 7.5 to 38.5km and increased over time by 0.26km/year (95% confidence interval [95%CI]: 0.15-0.38) (p<10-4). Median time to surgery was 1 day [1-3 days], with no significant difference across hospital types. Mortality rates at 90 days and 1 year were 10.5% (843/8026) and 20.8% (1673/8026), respectively. Two factors were significantly associated with day-90 mortality: the CCI (hazard ratio [HR], 1.087 [95%CI: 1.07-1.10] [p<10-4]) and time to surgery>1 day (HR 1.35 [95%CI: 1.15-1.50] [p<0.0001]). Day-90 mortality decreased significantly from 2005 to 2017 (HR 0.95 [95%CI: 0.92-0.97] [p<10-4]), with no centre effect. CONCLUSION: The management of PFF in patients older than 60 varied widely across France. Time to surgery longer than 1 day was a major adverse prognostic factor whose effects persisted throughout the first year. This factor was present in over half the patients. Day-90 mortality decreased significantly from 2005 to 2017 despite increases in age and comorbidities. LEVEL OF EVIDENCE: IV Retrospective cohort study.