Patient-level Associations Research Articles

The Reach of Calls and Text Messages for Mailed FIT Outreach in the PROMPT Stepped-Wedge Colorectal Cancer Screening Trial.

Mailed fecal immunochemical test (FIT) outreach can improve colorectal cancer screening participation. We assessed the reach and effectiveness of adding notifications to mailed FIT programs. We conducted secondary analyses of a stepped-wedge evaluation of an enhanced mailed FIT program (n = 15 clinics). Patients were stratified by prior FIT completion. Those with prior FIT were sent a text message (Group 1); those without were randomized 1:1 to receive a text message (Group 2) or live phone call (Group 3). All groups were sent automated phone call reminders. In stratified analysis, we measured reach and effectiveness (FIT completion within 6 months) and assessed patient-level associations using generalized estimating equations. Patients (n = 16,934; 83% Latino; 72% completed prior FIT) were reached most often by text messages (78%), followed by live phone calls (71%), then automated phone calls (56%). FIT completion was higher in patients with prior FIT completion versus without [44% (Group 1) vs. 19% (Group 2 + Group 3); P < 0.01]. For patients without prior FIT, effectiveness was higher in those allocated to a live phone call [20% (Group 3) vs. 18% (Group 2) for text message; P = 0.04] and in those who personally answered the live call (28% vs. 9% no call completed; P < 0.01). Text messages reached the most patients, yet effectiveness was highest in those who personally answered the live phone call. Despite the broad reach and low cost of text messages, personalized approaches may more successfully boost FIT completion.

Re-Evaluation of Pathologic Complete Response as a Surrogate for Event-Free and Overall Survival in Human Epidermal Growth Factor Receptor 2-Positive, Early Breast Cancer Treated With Neoadjuvant Therapy Including Anti-Human Epidermal Growth Factor Receptor 2 Therapy.

Pathologic complete response (pCR) has prognostic importance and is frequently used as a primary end point, but doubts remain about its validity as a surrogate for event-free survival (EFS) and overall survival (OS) in human epidermal growth factor receptor 2 (HER2)-positive, early breast cancer. We obtained individual-patient data from randomized trials of neoadjuvant anti-HER2 therapy that enrolled at least 100 patients, had data for pCR, EFS, and OS, and a median follow-up of at least 3 years. We quantified the patient-level association between pCR (defined as ypT0/Tis ypN0) and both EFS and OS using odds ratios (ORs, with ORs >1.00 indicating a benefit from achieving a pCR). We quantified the trial-level association between treatment effects on pCR and on EFS and OS using R2 (with values above 0.75 considered as indicating strong associations). Eleven of 15 eligible trials had data for analysis (3,980 patients, with a median follow-up of 62 months). Considering all trials, we found strong patient-level associations, with ORs of 2.64 (95% CI, 2.20 to 3.07) for EFS and 3.15 (95% CI, 2.38 to 3.91) for OS; however, trial-level associations were weak, with an unadjusted R2 of 0.23 (95% CI, 0 to 0.66) for EFS and 0.02 (95% CI, 0 to 0.17) for OS. We found qualitatively similar results when grouping trials according to different clinical questions, when analyzing only patients with hormone receptor-negative disease, and when using a more stringent definition of pCR (ypT0 ypN0). Although pCR may be useful for patient management, it cannot be considered as a surrogate for EFS or OS in neoadjuvant trials of HER2-positive, operable breast cancer.

Worldwide Prevalence of Antibiotic-Associated Stevens-Johnson Syndrome and Toxic Epidermal Necrolysis

Antibiotics are an important risk for Stevens-Johnson syndrome and toxic epidermal necrolysis (SJS/TEN), which are the most severe types of drug hypersensitivity reaction with a mortality rate up to 50%. To our knowledge, no global systematic review has described antibiotic-associated SJS/TEN. To evaluate the prevalence of antibiotics associated with SJS/TEN worldwide. The MEDLINE and Embase databases were searched for experimental and observational studies that described SJS/TEN risks since database inception to February 22, 2022. Included studies adequately described SJS/TEN origins and specified the antibiotics associated with SJS/TEN. Two reviewers (E.Y.L. and C.K.) independently selected the studies, extracted the data, and assessed the risk of bias. A meta-analysis using a random-effects model was performed in the studies that described patient-level associations. Subgroup analyses were performed to explore the heterogeneity. The risk of bias was assessed using the Joanna Briggs Institute checklist, and the certainty of evidence was rated using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach. Prevalence of antibiotic-associated SJS/TEN was presented as pooled proportions with 95% CIs. Among the 64 studies included in the systematic review, there were 38 studies that described patient-level associations; the meta-analysis included these 38 studies with 2917 patients to determine the prevalence of single antibiotics associated with SJS/TEN. The pooled proportion of antibiotics associated with SJS/TEN was 28% (95% CI, 24%-33%), with moderate certainty of evidence. Among antibiotic-associated SJS/TEN, the sulfonamide class was associated with 32% (95% CI, 22%-44%) of cases, followed by penicillins (22%; 95% CI, 17%-28%), cephalosporins (11%; 95% CI, 6%-17%), fluoroquinolones (4%; 95% CI, 1%-7%), and macrolides (2%; 95% CI, 1%-5%). There was a statistically significant heterogeneity in the meta-analysis, which could be partially explained in the subgroup analysis by continents. The overall risk of bias was low using the Joanna Briggs Institute checklist for case series. In this systematic review and meta-analysis of all case series, antibiotics were associated with more than one-quarter of SJS/TEN cases described worldwide, and sulfonamide antibiotics remained the most important association. These findings highlight the importance of antibiotic stewardship, clinician education and awareness, and weighing the risk-benefit assessment of antibiotic choice and duration.

Tooth Wear Epidemiology and Its Associated Periodontal Health and Sociodemographic Factors in a Cluster of Senior Citizens in Northern Greece.

Tooth wear (TW) is an irreversible and cumulative phenomenon causing aesthetic and functional compromise. Increasing wear has been associated with age, and various other factors have been reported to influence its type and/or severity both in individuals and groups. Increased TW may constitute a major future problem for the elderly. The present cross-sectional study aims at determining the prevalence of TW in senior citizens from Northern Greece and evaluating the patient-level associations between TW, periodontal condition, and sociodemographic factors. A sample of 363 dentate individuals, aged between 65 and 74 years, was considered according to the WHO guidelines for national pathfinder surveys and three different dentists examined the representative population groups from different urban and rural areas in Northern Greece. The examiners were calibrated prior to the survey, with an interexaminer agreement of over 85%. The simplified TWI, community periodontal index (CPI), attachment loss (AL), plaque index (DI), calculus index (CI), and sociodemographic factors were detected and measured. TW is very prevalent among senior citizens in Northern Greece, with males having been found to experience more wear. Age and oral health status, when measured by the periodontal indices AL, DI, and CI, are significantly combined with TW in both urban and rural areas.

Association of progression-free survival and overall response rate with overall survival in first-line randomized trials of immune checkpoint inhibitor–based regimens for metastatic non–small cell lung cancer (NSCLC): An FDA pooled analysis.

9029 Background: Overall Survival (OS) has represented the endpoint of choice to support approvals of Immune Checkpoint Inhibitors (ICIs) in first-line metastatic NSCLC. Despite continued interest in the use of earlier clinical endpoints as true surrogates for OS in this setting, the correlation of Progression-Free Survival (PFS) and Overall Response Rate (ORR) with OS remain an area of active investigation. We conducted a patient-level and trial-level pooled analysis of first-line randomized trials of ICI-based regimens to assess correlations of early clinical endpoints of PFS and ORR with OS in first-line metastatic NSCLC. Methods: The analysis included randomized trials comparing ICI-based regimens (anti-PD-(L)1 with or without anti-CTLA4 antibodies, with or without platinum-based chemotherapy) to platinum-based chemotherapy alone for the first-line treatment of patients with metastatic NSCLC that were submitted to the U.S. Food and Drug Administration between July 2016 to March 2021 to support a marketing application. Patient-level associations were estimated using Spearman (rs) correlation coefficients for PFS and OS, and Cox Proportional Hazards models in RECIST response-based subgroups for ORR and OS. At the trial level, associations were estimated using R2 coefficients from weighted linear regression models, using the log of hazard-ratio (HRs) for PFS and OS and log-odds ratio for ORR. Results: The pooled analysis included 13 trials enrolling 9,285 patients total. Seven trials compared ICIs combined with chemotherapy vs chemotherapy; 6 trials compared ICIs alone vs chemotherapy. Among all patients, the distribution of PD-L1 expression was 31%, 66%, and 32% for PD-L1 &lt;1%, ≥1%, and ≥50%, respectively. The table shows the correlation coefficients for PFS and OS, and ORR and OS at the patient and trial levels. At the patient level, the OS HR comparing ICI-based regimens to chemotherapy was 0.54 (95% CI: 0.48, 0.61) for RECIST responders and 0.96 (95% CI: 0.90, 1.02) for non-responders. Conclusions: This pooled analysis did not indicate a strong correlation between endpoints of PFS and ORR with OS at the patient and trial levels in first-line randomized trials of ICI-based regimens for metastatic NSCLC, potentially because of use of subsequent therapies, cross-over to ICIs, and continuation of ICIs beyond progression. Future research will explore the correlation of alternative endpoints with OS, such as time to treatment discontinuation. Our analysis supports the continued importance of OS as an endpoint for first-line NSCLC trials of ICI-based regimens.[Table: see text]

OA39 Online Patient Reported Outcome Measure (PROM) engagement is dependent on demographics and locality: findings from an observational cohort

Abstract Background/Aims Online patient-reported outcome measures (PROMs) enable remote collection of perceptions of health status, function and wellbeing. Single-centre studies report that PROM completion is dependent on patient demographics. Here we use a large national prospective dataset to explore patterns of PROM completion. Methods Patients with a new diagnosis of early inflammatory arthritis (EIA) were recruited to the National Early Inflammatory Arthritis Audit (NEIAA) between May 2018 and March 2020 and followed up for one year. PROMs capturing the impact of disease using the MSK-HQ, disability using HAQ, mental health using PHQ4ADS, and work using the WPAI were requested at baseline, 3 and 12 months. Multivariable logistic regression was constructed to examine patient level associations with PROM completion. Mean PROM completion by clinical commissioning groups (CCG) were mapped across England and spatial regression models used to identify regional patterns. Results A total of 34,856 patients with suspected EIA were recruited to NEIAA. Of these, 11,986 were diagnosed with EIA across England, recruited from 205 CCGs. The cohort was evenly distributed across the indices of multiple deprivation (IMD) deciles. Thirteen percent were Black, Asian, or minority ethnic (BAME). At least one PROM was completed by 5,331 (44.5%) patients. Of these, 3,846 (72.1%) had completed half of all the available PROMs whilst 412 (7.7%) had completed all available PROMs. BAME participants were less likely to return a PROM in adjusted regression modelling (adj OR 0.57, 95% CI 0.48-0.66). Deprivation quintile (adj OR 0.73, 95% CI 0.64-0.83), male gender (adj OR 0.86, 95% CI 0.78-0.94), comorbidity burden (adj OR 0.95, 95% CI 0.91-0.99) and current smoker status (adj OR 0.73, 95% CI 0.64-0.82) also reduced the odds of PROM completion. The proportion with PROM completion varied across CCGs. Spatial analysis identified two autocorrelated regions in England with a hotspot of high PROM completion in the North and a cold spot in the South East. Conclusion This study defines key participant characteristics that influence PROM engagement in a national clinical audit and identifies a novel association between locality and PROM completion. Integral to the utility of PROMs in NEIAA is that they are representative of the population with EIA. Our results suggest there are under-represented groups with the potential to bias in favour of those likely to experience better outcome. Disclosure K. Bechman: None. M. Yates: None. H. Wright: None. M. Russell: None. D. Nagra: None. B.D. Clarke: None. J. Ledingham: None. S. Norton: None. J. Galloway: Consultancies; Abbvie, Sanofi, Novartis, Pfizer, UCB, Janssen. Grants/research support; Eli Lilly.

A national evaluation of geographic accessibility and provider availability of obesity medicine diplomates in the United States between 2011 and 2019.

Background/Objectives:Obesity is a pressing health concern within the United States (US). Obesity medicine “diplomates” receive specialized training, yet it is unclear if their accessibility and availability adequately serves the need. The purpose of this research was to understand how accessibility has evolved over time and assess the practicality of serving an estimated patient population with the current distribution and quantity of diplomates.Methods:Population-weighted Census tracts in US counties were mapped to the nearest facility on a road network with at least one diplomate who specialized in adult (including geriatric) care between 2011 and 2019. The median travel time for all Census tracts within a county represented the primary geographic access measure. Availability was assessed by estimating the number of diplomates per 100 000 patients with obesity and the number of facilities able to serve assigned patients under three clinical guidelines.Results:Of the 3 371 diplomates certified since 2019, 3 036 were included. The median travel time (weighted for county population) fell from 28.5 minutes [IQR: 13.7, 68.1] in 2011 to 9.95 minutes [IQR: 7.49, 18.1] in 2019. There were distinct intra- and inter-year travel time variations by race, ethnicity, education, median household income, rurality, and Census region (all P < .001). The median number of diplomates per 100 000 with obesity grew from 1 [IQR: 0.39, 1.59] in 2011 to 5 [IQR: 2.74, 11.4] in 2019. In 2019, an estimated 1.7% of facilities could meet the recommended number of visits for all mapped patients with obesity, up from 0% in 2011.Conclusions:Diplomate geographic access and availability has improved over time, yet there is still not a high enough supply to serve the potential patient demand. Future studies should quantify patient-level associations between travel time and health outcomes, including whether the number of available diplomates impacts utilization.

Response Rate, Event-Free Survival, and Overall Survival in Newly Diagnosed Acute Myeloid Leukemia: US Food and Drug Administration Trial-Level and Patient-Level Analyses.

To explore trial-level and patient-level associations between response (complete remission [CR] and CR + CR with incomplete hematologic [CRi] or platelet [CRp] recovery), event-free survival (EFS), and overall survival (OS) in newly diagnosed acute myeloid leukemia (AML) trials of intensive chemotherapy. We identified data from eight randomized, active-controlled trials of intensive chemotherapy submitted to the US Food and Drug Administration for treatment of newly diagnosed AML (N = 4,482). Associations between trial-level odds ratios (ORs) for CR and CR + CRi or CRp, and hazard ratios (HRs) for EFS and OS were analyzed using weighted linear regression models. We performed patient-level responder analyses to compare OS by response using pooled data from all studies. In trial-level analyses, association between HR for OS and OR for CR was moderate (R2 = 0.49; 95% CI, 0.05 to 0.86), as was the association with OR for CR + CRi or CRp (R2 = 0.48; 95% CI, 0.05 to 0.99). For OS versus EFS, a strong association was observed (R2 = 0.87; 95% CI, 0.47 to 0.98) when EFS definitions were harmonized across trials using raw data. In the patient-level responder analyses, patients who achieved CR had better OS compared with CRi or CRp responders (0.73; 95% CI, 0.64 to 0.84) and nonresponders (HR, 0.33; 95% CI, 0.31 to 0.37). On a trial level, there is a moderate association between OS and CR rate. A strong association between EFS and OS was observed. However, CIs were wide, and results became moderate using alternative definitions for EFS. Patient-level analyses showed CR responders have better OS compared with CRi or CRp responders and nonresponders. A therapy in newly diagnosed AML with benefit in EFS or substantial benefit in CR rate would be likely to have an OS effect.

Abstract PS3-17: Patient and provider determinants of breast cancer screening among Ontario women age 40-49: A population based retrospective cohort study

Abstract Background: For women aged 40-49, Canadian guidelines recommend individualized-decision making (based on risk assessment, values, and preferences) rather than routine screening mammography (SM). In this age group, family physicians are the gatekeepers to access SM; however, studies indicate substantial variability in practice patterns. There are few population-based data regarding uptake and patient/provider determinants of SM in this age group. We describe the uptake and frequency of SM and identify patient and provider level associations with SM in Ontario women aged 40-49. We hypothesized that SM would vary by provider characteristics and women’s demographics, suggesting lack of guideline-concordant care. Methods: This population-based retrospective cohort study linked health administrative databases to form a cohort of all Ontario women aged 40-49 between April 1, 2009 to March 31, 2019. Mammograms were identified using Ontario Health Billing codes. In order to identify mammograms that were specifically for screening, women were excluded if they had any prior breast MRI, mammogram, cancer diagnosis, oncologist visit, or breast surgical procedure. Sub-cohorts were created to identify women who had (a) at least one SM (“screen cohort”) and (b) 3 or more SM (“routine screen cohort”). Following SM, women were censored from cohorts if they had any cancer diagnosis, breast surgical procedure, oncologist visit, or death; however, breast cancer related outcomes were tracked for 6 months following SM, regardless of whether a censoring event occurred. Patient and provider characteristics were extracted for women in each cohort. A multivariable regression model was used to identify predictors of routine SM. Results: Of 2 million eligible women, 743 274 (35.6%) received a mammogram, 532 782 (25.5%) received at least one SM, and 90 651 (4.3%) received routine SM (3 or more). Table 1 demonstrates cohort characteristics. There were 0.32 and 0.52 mammograms per woman per year in the screen and routine screen cohort respectively. Call-backs were similar for women after the first SM compared to the third SM (9.5% vs 9.3%); however, there were more biopsies (3.2% vs 1.8%) and breast cancers diagnosed (1.2% vs 0.45%) within six months of the first SM. Compared to the full cohort, women in the routine screen cohort were more likely to have a family physician at cohort entry, be in higher-income quintiles, receive annual health exams, and receive have pap smears (p&amp;lt;0.001). Women in the screen cohort were more likely to have female providers and providers that were primarily paid fee for service versus capitation (p&amp;lt;0.001). Multivariable analysis will be reported at the meeting. Conclusions: Less than 5% of Ontario women 40-49 undergo routine SM. SM is associated with patient demographics related to higher socioeconomic status which could be related to higher risk of breast cancer and/or increased access to care. SM is also associated with some provider demographics which could be independent of breast cancer risk, suggesting lack of individualized risk assessment. Qualitative work is ongoing to explore this hypothesis. This information can inform guideline implementation strategies. Table 1: Patient and Provider Characteristics of ScreeningCharacteristicpercent (%) unless otherwise notedAll Women 40-49Screen Cohort (1 or more SM)Routine Screen Cohort (3 or more SM)Patient CharacteristicsAge at Cohort entryMean (median)42.5 (40)42.8 (42)41.7 (41)Age at Cohort exitMean (median)47.3 (50)48.5 (50)49.1 (50)Family Physician at Cohort entry90.0 %92.3%94.8%Income – top quintile20.7%22.3%24%Rurality Category – most urban76.3%79.0%81.2%Birth Location – Canada74.4%72.2%71.6%Annual Health Exams: at least one51.1%72.1%75.3%Pap Smear: at least one68.4%86.6%92.9%Provider CharacteristicsFemale44.8%47.2%52.1%Canadian Graduates70.3%70.3%72.1%Fee for Service model48.0%55.0%58.8% Citation Format: Michelle B Nadler, Alex Marchand-Austin, Peter C. Austin, Alexandra Desnoyers, Brooke E. Wilson, Aisha Lofters, Noah Ivers, Eitan Amir. Patient and provider determinants of breast cancer screening among Ontario women age 40-49: A population based retrospective cohort study [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr PS3-17.

Development of the PriCARE classification for potentially preventable emergency department visits by ambulance: a RAND/UCLA modified Delphi study protocol

IntroductionOntario ambulances are restricted from patient transportation to sub-acute levels of care when these facilities may be more suitable than emergency departments for non-emergent conditions. There is no known patient...

FAMILY AFFLUENCE RELATIONSHIP TO SPORTS SPECIALIZATION IN YOUTH ATHLETES

Background:Specializing in a single sport while excluding others, also known as sports specialization, is increasing in prevalence in the United States. The effect of socioeconomic status (SES) on rates of sports specialization has been previously described in the specialty clinic setting using zip code as a surrogate for SES. Patient-reported family affluence, a validated surrogate for SES, may provide more specific information about how family SES is associated with sport specialization.Purpose/Hypothesis:Our purpose was to examine the association between patient-reported SES and sport specialization level. We hypothesized that athletes who reported higher levels of sports specialization would report greater family affluence.Methods:We conducted a cross-sectional study of youth athletes undergoing a pre-participation evaluation. During the assessment, participants specified their level of sport specialization on a 3-point scale described originally by Jayanthi and colleagues. Participants also completed the Family Affluence Scale, a 6-item scale which is a valid correlate of parental income. We compared family affluence among the sport specialization groups (low/medium/high) using ANOVA with Tukey post-hoc comparisons. We then constructed a multivariable linear regression model to assess the association between family affluence and sport specialization level while controlling for potential confounding variables.Results:A total of 195 youth athletes participated in the study. The majority (52%) reported low levels of sport specialization, 33% reported medium sport specialization, and 15% reported high sport specialization. The sport specialization groups were similar in age, proportion of females, and time spent training in their sport (Table 1). A significantly larger proportion of highly specialized athletes were varsity athletes compared to the low specialization group (Table 1). The high sport specialization group reported significantly greater family affluence than the low sport specialization group (Figure 1). After covariate adjustment, higher levels of sport specialization remained significantly associated with greater reported family affluence (Table 2).Conclusion:In our population of youth athletes, high sports specialization was associated with greater family affluence. Additionally, we found a significantly higher proportion of varsity level athletes in the high sports specialization groups. This study extends previous studies identifying greater sports specialization levels among those with higher SES. By using the Family Affluence Scale and sampling from a large and diverse population during a pre-participation evaluation, our findings reflect accurate patient-level associations. Examining SES and sports specialization affords the sports medicine community an opportunity to better identify and monitor populations who are likely to specialize in a single sport during high school.Table 1.Characteristics of the three sport specialization groups. Continuous variables are presented as means (95% confidence intervals), categorical variables are presented as n (%).VariableLow Sport Specialized(n=101)Medium Sport Specialized(n=64)High Sport Specialized(n=30)P valueAge (years)15.2 (14.8, 15.6)15.3 (15.0, 15.6)15.7 (15.3, 16.1)0.26Female sex49 (49%)29 (45%)17 (57%)0.60Time spent training in sport(hours per week)11.9 (10.8, 13.1)11.6 (10.3, 13.0)12.5 (10.8, 14.1)0.75Varsity athlete28 (28%)26 (41%)15 (50%)0.04** A significantly higher proportion of high sport specialized athletes reported competing at the varsity level compared to low specialized athletes.Figure 1.Family Affluence Ratings between the low, medium, and high sports specialization groups. Data are presented as means (95% confidence intervals). Higher values on the family affluence scale reflect higher socioeconomic status. ANOVA analysis indicated a significant effect between groups (F=5.73; p= 0.0038). Tukey post-hoc comparisons indicated the high sport specialization group had significantly greater family affluence compared to the low sport specialization group (contrast= 1.19; p=0.005; 95% confidence interval for difference= 0.30, 2.08).Table 2.Multivariable regression model results for the association between sport specialization level and family affluence, while controlling for potential confounding variables.Variableβ coefficientStandard error95% confidence intervalP valueSport Specialization Level0.480.190.11, 0.860.012*Varsity athlete0.340.32-0.29, 0.980.29Age-0.040.09-0.21, 0.140.69Sex0.160.29-0.40, 0.730.57Hours per week training in sport-0.020.03-0.08, 0.030.41*After covariate adjustment, higher levels of sport specialization were significantly associated with greater family affluence.

Disease-free survival as a surrogate for overall survival in patients with HER2-positive, early breast cancer in trials of adjuvant trastuzumab for up to 1 year: a systematic review and meta-analysis.

Although frequently used as a primary endpoint, disease-free survival has not been validated as a surrogate for overall survival in early breast cancer. We investigated this surrogacy in the adjuvant setting of treatment with anti-HER2 antibodies. In a systematic review and meta-analysis, we identified published and non-published randomised controlled trials with completed accrual and available disease-free survival and overall survival results for the intention-to-treat population as of September 2016. Bibliographic databases (MEDLINE, Embase, and Cochrane Central Register of Controlled Trials), clinical trial registries (Clinicaltrials.gov, EU Clinical Trials Register, WHO International Clinical Trials Registry Platform, and PharmNet.Bund), and trial registries from relevant pharmaceutical companies were searched. Eligibility for treatment of HER2-positive early breast cancer required at least one group to have an anti-HER antibody treatment (ie, trastuzumab, pertuzumab, or trastuzumab emtansine) planned for 12 months, and at least one control arm with chemotherapy without the antibody, a lower total dose or duration of the antibody, or observation alone. Units of analysis were contrasts: two-group trials gave rise to one contrast, whereas trials with more than two groups gave rise to more than one contrast. We excluded trials enrolling patients with recurrent, metastatic, or non-invasive disease, and those testing neoadjuvant therapy exclusively. Our primary objective was to estimate patient-level and trial-level correlations between disease-free survival and overall survival. We measured the association between disease-free survival and overall survival using Spearman's correlation coefficient (rs), and the association between hazard ratios (HRs) for disease-free survival and overall survival using R2. We computed the surrogate threshold effect, the maximum HR for disease-free survival that statistically predicts an HR for overall survival less than 1·00 in a future trial. Eight trials (n=21 480 patients) gave rise to a full set (12 contrasts). Patient-level associations between disease-free and overall survival were strong (rs=0·90 [95% CI 0·89-0·90]). Trial-level associations gave rise to values of R2 of 0·75 (95% CI 0·50-1·00) for the full set. Subgroups defined by nodal status and hormone receptor status yielded qualitatively similar results. Depending on the expected number of deaths in a future trial, the surrogate threshold effects ranged from 0·56 to 0·81, based on the full set. These findings suggest that it is appropriate to continue to use disease-free survival as a surrogate for overall survival in trials in HER-2-positive, early breast cancer. The key limitation of this study is the dependence of its results on the trials included and on the existence of an outlying trial. Roche Pharma AG.

Surrogate endpoints in early prostate cancer research.

Clinical research into clinically-localized prostate cancer (PC) is a highly challenging environment. The protracted durations and large numbers required to achieve survival endpoints have placed much pressure on validating early surrogate endpoints. Further confounding is the predominance of deaths from causes other than PC. The analysis of multiple randomized clinical trials in early PC has shown MFS to be a robust surrogate for OS, using a contemporary analytic framework that identify patient-level and trial-level associations. This could potentially save around one year of trial follow-up in some therapies. Identification of a similarly robust surrogate at a substantially earlier timepoint remains a major challenge. Multiple biochemical indices based on PSA have been proposed in the literature, but all remain to be validated at the trial-level. Operationally, many of these indices have inherent biases such as immortal-time bias (ITB) and interval censoring that potentially weakens associations and the individual- or trial-level. The complexity of a failure definition can also impact the reliability of the derived outcomes. Confounding issues such as the impact of comorbidities leading to non-cancer deaths have been largely dealt with by their exclusion using cancer-specific endpoints and advanced statistical methods, while issues such as PSA “bounce” and recovery from androgen deprivation therapy remain important to account for in cohorts treated with radiotherapy. Several potential surrogate endpoints based on serum prostate-specific antigen (PSA) levels show promising associations with PC-specific and overall survival (OS) in individual studies. Further large collaborative projects will continue to refine potential indices with these issues in mind, and explore the objective of an early surrogate of OS.

Features of Patient-Centered Primary Care and the Use of Ambulatory Care.

This study explores the association between patients' use of ambulatory care resources and features of patient-centered primary care (PCPC), specifically clinic-level National Committee for Quality Assurance (NCQA) recognition of PCPC, continuity of care, and care team communication. Data for this study were compiled from the electronic health records of a large multispecialty group practice in California, covering the period between 2009 and 2010 for 37,042 nonelderly patients under capitated managed care plans. Regression analysis of these data was performed using a generalized linear model, comparing measures of patient-level annual resource use (in total relative value units [tRVUs]) against measures of both clinic- and patient-level PCPC, and patient-level risk (eg, age, comorbidities). Patients linked to NCQA Recognition Level III versus Level II clinics used 4.8% (P < 0.001) fewer ambulatory care tRVUs. Patients with a 1 standard deviation increase in primary care continuity used 3.9% (P < 0.001) fewer ambulatory care tRVUs. Patients who switched primary care physicians used 17.4% (P < 0.001) more ambulatory care tRVUs. These results indicate that PCPC is associated with reductions in resource use related to both clinic- and patient-level features. The patient-level associations document within-clinic heterogeneity in PCPC. Based on these findings, practices can be encouraged to perform consistent with high NCQA recognition, promote primary care continuity, and be mindful of differences at an individual patient level.

Intraoperative Fluid Resuscitation Strategies in Pancreatectomy: Results from 38 Hospitals in Michigan.

Fluid administration practices may affect complication rates in some abdominal surgeries, but effects in patients undergoing pancreatectomy are not understood well. We sought to determine whether amount of intraoperative fluid administered to patients undergoing pancreatectomy is associated with postoperative complication rates and to determine whether hospitals vary in their fluid administration practices. Data for 504 patients undergoing pancreatectomy at 38 hospitals between 2012 and 2015 were evaluated. The main exposure was intraoperative fluid administration (≤10, 10-15, >15mL/kg/h). Mortality, complications, and length of stay were the main outcomes of interest. Patient-level associations between exposure and outcome were tested, with adjustment for potentially confounding patient and surgical factors, using random intercept, mixed-effects linear or logistic regression models. Hospitals were then categorized as having a restrictive, intermediate, or liberal resuscitation practice, and adjusted outcomes were compared. A total of 167 (33.1%), 185 (36.7%) and 152 (30.2%) patients received restrictive, intermediate, or liberal fluid administration, respectively. Hospitals with more restrictive practices had significantly lower adjusted 30-day mortality than those with more liberal practices (2.7 vs. 6.6%; P<0.001). Hospitals with more restrictive practices had the lowest rates of severe (Grade 2 and 3) complications (15.4% restrictive vs. 25.3% intermediate vs. 44.3% liberal; P<0.001). More restrictive hospitals had decreased adjusted mean length of stay (9.5days vs. 12.7days intermediate vs. 11.6days liberal; P<0.001). More restrictive intraoperative resuscitation practices in pancreatectomy are associated with decreased hospital-level mortality, severe complications, and length of stay.

Overall response rate, progression-free survival, and overall survival with targeted and standard therapies in advanced non-small-cell lung cancer: US Food and Drug Administration trial-level and patient-level analyses.

To conduct analyses exploring trial-level and patient-level associations between overall response rate (ORR), progression-free survival (PFS), and overall survival (OS) in advanced non-small-cell lung cancer (NSCLC) trials. We identified 14 trials (N = 12,567) submitted to US Food and Drug Administration since 2003 of treatments for advanced NSCLC. Only randomized, active-controlled trials with more than 150 patients were included. Associations between trial-level PFS hazard ratio (HR), OS HR, and ORR odds ratio were analyzed using a weighted linear regression model. Patient-level responder analyses comparing PFS and OS between patients with and without an objective response were performed using pooled data from all studies. In the trial-level analysis, the association between PFS and ORR was strong (R(2) = 0.89; 95% CI, 0.80 to 0.98). There was no association between OS and ORR (R(2) = 0.09; 95% CI, 0 to 0.33) and OS and PFS (R(2) = 0.08; 95% CI, 0 to 0.31). In the patient-level responder analyses, patients who achieved a response had better PFS and OS compared with nonresponders (PFS: HR, 0.40; 95% CI, 0.38 to 0.42; OS: HR, 0.40; 95% CI, 0.38 to 0.43). On a trial level, there is a strong association between ORR and PFS. An association between ORR and OS and between PFS and OS was not established, possibly because of cross-over and longer survival after progression in the targeted therapy and first-line trials. The patient-level analysis showed that responders have a better PFS and OS compared with nonresponders. A therapy in advanced NSCLC with a large magnitude of effect on ORR may have a large PFS effect.

Tumor response and progression‐free survival as potential surrogate endpoints for overall survival in extensive stage small‐cell lung cancer

AbstractBACKGROUND:The authors investigated the putative surrogate endpoints of best response, complete response (CR), confirmed response, and progression‐free survival (PFS) for associations with overall survival (OS), and as possible surrogate endpoints for OS.METHODS:Individual patient data from 870 untreated extensive stage small‐cell lung cancer patients participating in 6 single‐arm (274 patients) and 3 randomized trials (596 patients) were pooled. Patient‐level associations between putative surrogate endpoints and OS were assessed by Cox models using landmark analyses. Trial‐level surrogacy of putative surrogate endpoints were assessed by the association of treatment effects on OS and individual putative surrogate endpoints. Trial‐level surrogacy measures included: R2 from weighted least squares regression model, Spearman correlation coefficient, and R2 from bivariate survival model (Copula R2).RESULTS:Median OS and PFS were 9.6 (95% confidence interval [CI], 9.1‐10.0) and 5.5 (95% CI, 5.2‐5.9) months, respectively; best response, CR, and confirmed response rates were 44%, 22%, and 34%, respectively. Patient‐level associations showed that PFS status at 4 months was a strong predictor of subsequent survival (hazard ratio [HR], 0.42; 95% CI, 0.35‐0.51; concordance index 0.63; P &lt; .01), with 6‐month PFS being the strongest (HR, 0.41; 95% CI, 0.35‐0.49; concordance index, 0.66, P &lt; .01). At the trial level, PFS showed the highest level of surrogacy for OS (weighted least squares R2 = 0.79; Copula R2 = 0.80), explaining 79% of the variance in OS. Tumor response endpoints showed lower surrogacy levels (weighted least squares R2≤0.48).CONCLUSIONS:PFS was strongly associated with OS at both the patient and trial levels. PFS also shows promise as a potential surrogate for OS, but further validation is needed using data from a larger number of randomized phase 3 trials. Cancer 2011. © 2010 American Cancer Society.

Tumor response and progression-free survival (PFS) as potential surrogate endpoints for overall survival (OS) in extensive-stage small cell lung cancer (ES-SCLC): Findings based on North Central Cancer Treatment Group (NCCTG) trials.

7637 Background: We investigated the putative surrogate endpoints (PSEs) of best response (BR), confirmed response (CoR), and PFS for association with OS, and as possible surrogate endpoints for OS. Methods: Individual patient (pt) data for 870 untreated ES-SCLC pts from 6 single-arm (274 pts) and 3 randomized trials (596 pts from 32 participating centers) containing platinum or paclitaxel-based regimens were pooled. Patient-level associations between the PSEs and OS were assessed by Cox models using a landmark analyses at 2, 4, and 6 months post-study entry. Model discrimination was evaluated using the concordance (c) index. Trial-level surrogacy of PSEs was assessed by the association of the treatment effects on OS and individual PSEs. Treatment effects were estimated through log hazard ratios and log odds ratios. Trial-level surrogacy measures included: R2 from a weighted least squares regression model (R2 WLS), Spearman rank correlation coefficient (Spearman), and R2 from a bivariate survival model (R2 Copula). Results: The median OS and PFS were 9.6 and 5.5 months, respectively; the BR and CoR rates were 44% and 34%, respectively. Patient-level associations showed that PFS status at 4 months was a strong predictor of subsequent survival (HR=0.42 [95% CI: 0.35, 0.51]; c-index = 0.63; p < 0.01), with 6-month PFS being the strongest (HR=0.41 [95% CI: 0.35, 0.49]; c-index = 0.66; p < 0.01). At the trial level, PFS showed the highest level of surrogacy for OS (R2 WLS = 0.79; R2 Copula = 0.80), which indicates that PFS explained 79% of the variance in OS and is a good surrogacy measure (see Table). Conclusions: PFS was strongly associated with OS, where PFS status as early as 4 months was a strong predictor of subsequent survival. PFS also shows promise as a potential surrogate for OS, and needs further validation using data from a larger number of randomized phase III trials. This data, if substantiated, will allow faster yet valid evaluation of drugs for ES- SCLC. Trial-level surrogacy measures Endpoint R2 WLS R2 Copula Spearman PFS 0.79 0.80 (95% CI: 0.67, 0.93) 0.75 BR 0.21 NA 0.52 CoR 0.40 NA 0.60 No significant financial relationships to disclose.

Does Meeting the HEDIS Substance Abuse Treatment Engagement Criterion Predict Patient Outcomes?

This study examines the patient-level associations between the Health Plan Employer Data and Information Set (HEDIS) substance use disorder (SUD) treatment engagement quality indicator and improvements in clinical outcomes. Administrative and survey data from 2,789 US Department of Veterans Affairs SUD patients were used to estimate the effects of meeting the HEDIS engagement criterion on improvements in Addiction Severity Index Alcohol, Drug, and Legal composite scores. Patients meeting the engagement indicator improved significantly more in all domains than patients who did not engage, and the relationship was stronger for alcohol and legal outcomes for patients seen in outpatient settings. The benefit accrued by those who engaged was statistically significant but clinically modest. These results add to the literature documenting the clinical benefits of treatment entry and engagement. Although these findings only indirectly support the use of the HEDIS engagement measure for its intended purpose-discriminating quality at the facility or system level-they confirm that the processes of care captured by the measure are associated with important patient outcomes.