Publications

Estimating the magnitude of placebo responses across pharmacological and nonpharmacological trials is important for understanding their influence on trial outcomes. Yet, the extent to which more intense placebo interventions like sham acupuncture yield larger analgesic responses than placebo pills, and the factors predicting these responses, remain unclear. This meta-analysis investigated the magnitude and predictors of placebo analgesia responses in pharmacological vs acupuncture trials. Analyses included individual patient data from the placebo arm of 11 randomized controlled trials (RCTs): 9 pharmacological RCTs using placebo pills (N = 2021) and 2 acupuncture RCTs using sham acupuncture (N = 747). All trials were conducted in patients with chronic nociceptive pain (osteoarthritis, N = 2068; low back pain, N = 700). The placebo response was calculated as the change in pain intensity (0-100) between baseline and week 12. A random effects model demonstrated that placebo pills and patients with osteoarthritis exhibited smaller placebo responses than sham acupuncture and patients with low back pain (both P < 0.001, marginal effects). A mixed effects model showed that route of administration interacted significantly with baseline pain, premature termination, and the presence of adverse events. Together, predictors explained 20% to 25% of the individual variance in placebo responses, whereas 75% to 80% remained unaccounted for. In summary, sham acupuncture accounted for slightly larger placebo responses than placebo pills. Since basic trial and patient parameters explained only a small portion of this variability, we might need to start considering the patient's perception of the treatment-including cognition and emotions-to better predict placebo analgesia responses.

BACKGROUND: Age-related cognitive decline results in significant clinical and public health costs. Cognitive decline predisposes individuals to dementia and leads to difficulties with activities of daily living and increased hospitalizations. While a few dietary interventions have shown considerable promise at delaying cognitive decline, studies of the effects of long-term blueberry consumption on brain health among older adults have been limited and equivocal.

OBJECTIVE: To test the hypothesis that a 24-week intervention with blueberry powder versus placebo will improve (i) plasma biomarkers of brain health and (ii) cognitive performance in older adults.

METHODS: This randomized, double-blind, placebo-controlled trial will be conducted in older Americans aged 65-99 years old. Sixty-seven participants will be randomly assigned to receive either blueberry powder (20 g/d - equivalent of one (1) cup of fresh blueberries) or a placebo powder (20 g/d) for 24 weeks. Participants will undergo four clinic visits (baseline and 8 weeks, 16 weeks, and 24 weeks after randomization) for cognitive assessments, blood pressure measurements, and blood sample collections. Primary outcomes will be plasma levels of neurofilament light chain (NFL) and phosphorylated tau protein (pTau-181). Secondary outcomes will be plasma levels of glial fibrillary acidic protein (GFAP) and non-esterified fatty acids (NEFA); blood pressure during visits; and cognitive function.

CONCLUSION: Findings from this trial will answer the question whether a long-term intervention with blueberry powder can improve (i) plasma biomarkers of brain health and (ii) cognitive function in older adults.

INTRODUCTION: Triple-negative breast cancer (TNBC) often shows significant response to neoadjuvant chemotherapy (NACT), creating opportunities for axillary de-escalation. This study investigates axillary de-escalation and the factors influencing its use in TNBC.

METHODS: Patients with cT1-cT2, cN1 TNBC treated between 2012 and 2020 were identified from the National Cancer Database. Nodal response rates, including achievement of ypN0, were analyzed. Logistic regression identified clinical and sociodemographic factors associated with axillary management.

RESULTS: Among 12,742 patients undergoing axillary lymph node dissection (ALND), 31.7% achieved nodal pathologic complete response (pCR). These patients were more also likely to achieve breast tumor pCR. Treatment across multiple facilities was associated with a higher likelihood of attempted axillary de-escalation. ALND rates steadily declined between 2014 and 2020.

CONCLUSION: Despite achieving nodal pCR after NACT, many patients with TNBC underwent ALND. These findings underscore the need to refine criteria and increase adoption of axillary de-escalation strategies in TNBC.

BACKGROUND: Mechanisms of injury due to hypertension (HTN) in pregnancy remain poorly characterized. This study examined trends in markers of cardiac injury (high-sensitivity troponin I, hs-cTnI), strain (N-terminal pro-B-type natriuretic peptide, NT-proBNP), and inflammation (high-sensitivity C-reactive protein, hs-CRP) in pregnancies with and without HTN.

METHODS: This prospective, 1:1 case-control study enrolled pregnant women with and without HTN (24-32 weeks gestation) from 2019-2022. HTN was defined by a clinical diagnosis of HTN or baseline blood pressure (BP)≥140/90 mmHg. Serum was collected at baseline, pre-delivery, and postpartum day 1. Mixed effects tobit models compared log-transformed hs-cTnI, NT-proBNP, and hs-CRP across HTN groups and over time, adjusted for age and BMI.

RESULTS: Mean baseline BP was 130.5 (17.5)/88.2 (13.5) mmHg for the HTN group (n=38, 86.8% chronic, 13.2% gestational HTN), and 112.0 (9.8)/70.9 (8.2) mmHg for those without HTN (n=38). Over pregnancy, the HTN group had higher hs-cTnI than those without HTN (2.12 [0.43] vs. 1.07 [0.25], Δ1.05 [95%CI: 0.07-2.03] ng/L). Compared to baseline, hs-cTnI increased at pre-delivery and postpartum for both groups. Overall, the two groups had similar NT-proBNP (HTN: 39.0 [4.5] vs. no HTN: 35.6 [4.3] pg/mL) and hs-CRP (HTN: 12.0 [1.7] vs. no HTN: 9.9 [1.5] mg/L). For both groups, NT-proBNP and hs-CRP increased from baseline to postpartum (NT-proBNP, HTN: 127% [58-227%], no HTN: 120% [51-219%]; hs-CRP: HTN: 550% [343-853%], no HTN: 664% [415-1034%]).

CONCLUSION: HTN was associated with markers of cardiac injury during pregnancy, while delivery alone led to increases in markers of strain and inflammation. These biomarker changes associated with HTN in pregnancy may represent potential mechanisms to explain adverse cardiovascular events.

BackgroundAdvance care planning (ACP) helps older adults make end-of-life medical decisions. While ACP discussions are associated with improved patient outcomes, overall engagement remains low in the emergency department (ED).ObjectivesThis study assessed ACP engagement in older ED patients.MethodsWe conducted a questionnaire study among adults ≥65 in a Boston ED (July-Oct 2023). Our primary outcome was ACP Engagement as measured by a validated 9-item instrument with three ACP self- efficacy and six ACP readiness questions. Secondary outcomes included participants' preferences for learning about five ACP topics in the ED-medical decision makers, what matters most, leeway and flexibility for decision makers, sharing wishes, and asking questions)-as well as their favored learning formats (eg, pamphlets, videos, clinician conversations). Each was rated on a 5-point Likert scale. We examined the relationship between ACP engagement and existing electronic health record (EHR) documentation.ResultsNinety-nine older adults participated (mean age 75.5; 53.5% women). On the 9-item ACP Engagement Survey, participants reported high overall scores with a mean of 4.1 (95% CI: 4.0-4.2). Among ACP readiness topics, 80 (81.6%) named a decision-maker; 37 (40.2%) discussed end-of-life wishes with doctors. Participants preferred ED team conversations. Among the 51 participants who reported having signed paperwork regarding end-of-life wishes, only 7 (13.7%) had forms documented in the EHR.ConclusionsAmong older adults in a large Boston-based ED, ACP engagement was high, but few patients had documentation of end-of-life wishes available in the EHR. Findings highlight the need for better ACP documentation in EDs.

BACKGROUND: Many patient portals allow patients to authorize a care partner to use the portal on their behalf, with evidence suggesting a range of benefits to patients, care partners, and clinicians. Shared or proxy access aligns with patient- and family-centered care and supports care partners' legitimacy and identification by clinicians in patient portal interactions. As shared access uptake remains low, the Coalition for Care Partners ( https://coalitionforcarepartners.org ) and three healthcare delivery organizations co-designed an initiative promoting shared access to the patient portals of older adults.

OBJECTIVE: To evaluate an initiative's demonstration through users' perspectives.

DESIGN: The 12-month demonstration was launched at five clinics (geriatric oncology, primary care, and geriatric medicine) across the three organizations. Clinicians and staff were interviewed mid- and post-demonstration via focus groups and individually; clinic patients and care partners responded to an anonymous post-demonstration online portal survey.

PARTICIPANTS: Demonstration users included established patients from the five clinics and their care partners, as well as clinic physicians, nurses, social workers, care managers, patient-facing staff, administrators, and information technology specialists.

APPROACH: We followed the Consolidated Framework for Implementation Research to develop interview guides and inform our analysis of the survey's open-ended responses and interview transcripts. We analyzed 11 focus groups, 10 individual interviews, and 392 patients' and 79 care partners' survey responses employing rapid assessment procedures methodology.

KEY RESULTS: The demonstration was appropriate, useful, routinized in the clinics, and well received by patients and care partners. The demonstration was perceived as undemanding and low-cost, utilizing existing infrastructure and organizational processes. Facilitators included alignments of shared access with clinics' practice and philosophy, organizations' policies, and needs of patients and care partners. Identified barriers included clinicians' competing priorities and patients' and care partners' low awareness and motivation for using shared access.

CONCLUSIONS: The Coalition for Care Partners can spread this appropriate, useful, undemanding, and low-cost initiative. Further efforts might be supported by policies that ensure that shared access benefits are bolstered, potential harms of unidentifiable access are emphasized, and registration is conceptualized as an opt-out versus opt-in process.

BACKGROUND: Racial and ethnic disparities in hypertension care persist. Shared decision making (SDM) is promoted in hypertension guidelines. However, evidence is lacking on how race, ethnicity, and SDM relate, and the effect of SDM on hypertension control in diverse groups. We aimed to explore the relationships among SDM, blood pressure (BP), race and ethnicity, and other decision-making factors in patients with hypertension.

METHODS AND RESULTS: Longitudinal analysis of data from the RICH LIFE (Reducing Inequities in Care of Hypertension: Lifestyle Improvement for Everyone) project's participants (n=1426) with uncontrolled hypertension was performed using descriptive statistics, linear regression, and generalized estimating equations. Participants were middle-aged (mean age 60±11.6 years), predominantly women (59.4%, 847), non-Latino Black (59%, 844), and high school graduates or below (65%, 931). The mean SDM score was 7.2±2.6 out of 9, and the mean baseline systolic and diastolic BP were 152.2±12.0 and 85.3±12.1 mm Hg. Non-Latino Black people had 0.14 points higher mean SDM score (P<0.001) than non-Latino White people. Systolic BP reduction over 12 months was greater with a higher SDM mean score (β=-0.42, P=0.035). Baseline characteristics associated with SDM included more than high school education (β=-0.08, P=0.045), hypertension knowledge (β=-0.05, P=0.046), considering taking BP medication as very important (β=0.06, P=0.022), and patient activation (β=0.09, P=0.001).

CONCLUSIONS: There was greater BP reduction for patients with higher SDM score at follow-up, and associations between SDM and race and ethnicity, education, hypertension knowledge and attitude, and patient activation. Future research should further explore SDM differences among racial and ethnic groups to better align hypertension care with patients' needs.

BACKGROUND: Postoperative delirium is one of the most common complications in the older surgical population, but its pathogenesis and biomarkers are largely undetermined. Retinal layer thickness has been demonstrated to be associated with cognitive function in mild cognitive impairment and patients with Alzheimer's disease. However, relatively little is known about possible retinal layer thickness among patients with postoperative delirium.

AIMS: We aimed to investigate the relationship between retinal layer thickness and postoperative delirium in this cross-sectional study.

METHODS: The participants (≥65 years old) having elective surgery under general anaesthesia were screened via medical records from Shanghai 10th People's Hospital. Preoperative macular thickness and peripapillary retinal nerve fibre layer (RNFL) thickness were measured using optical coherence tomography (OCT). The Confusion Assessment Method (CAM) algorithm and CAM-Severity (CAM-S) were used to assess the incidence and severity of postoperative delirium on the first, second and third days after surgery.

RESULTS: Among 169 participants (mean (standard deviation (SD) 71.15 (4.36) years), 40 (24%) developed postoperative delirium. Notably, individuals who developed postoperative delirium exhibited thicker preoperative macular thickness in the right eye compared with those who did not (mean (SD) 283.35 (27.97) µm vs 273.84 (20.14) µm, p=0.013). Furthermore, the thicker preoperative macular thickness of the right eye was associated with a higher incidence of postoperative delirium (adjusted odds ratio 1.593, 95% confidence interval (CI) 1.093 to 2.322, p=0.015) and greater severity (adjusted mean difference (β)=0.256, 95% CI 0.037 to 0.476, p=0.022) after adjustment for age, sex and Mini-Mental State Examination (MMSE) scores. However, such a difference or association did not appear in the left macular or bilateral peripapillary RNFL thicknesses.

CONCLUSIONS: Current findings demonstrated that preoperative macular thickness might serve as a potential non-invasive marker for the vulnerability of developing postoperative delirium in older surgical patients. Further large-scale validation studies should be performed to confirm these results.

BACKGROUND: The Dietary Approaches to Stop Hypertension (DASH) diet lowers estimated 10-year ASCVD (atherosclerotic cardiovascular disease) risk. The effects of dietary sodium reduction on ASCVD risk are uncertain. This study aims to evaluate the impact of sodium reduction, alone and combined with the DASH diet, on 10-year ASCVD risk scores.

METHODS: The DASH-Sodium trial randomized adults with elevated blood pressure (average systolic blood pressure of 120 to 159 mm Hg and average diastolic blood pressure of 80 to 95 mm Hg) to the DASH diet or typical American diet. Within each arm, individuals consumed 3 different levels of sodium in random order: low, medium, and high. Each period lasted 30 days. Pooled cohort equation-estimated 10-year ASCVD risk scores were calculated at baseline and at the end of each feeding period. The primary outcomes of interest were the absolute and relative differences in 10-year ASCVD risk scores from baseline.

RESULTS: Among the 412 participants (mean age 48 ± 10 years; 57 % female, 57 % Black), sodium reduction decreased ASCVD risk scores in both dietary arms. Compared to high sodium intake, low sodium intake changed ASCVD risk by -9.4 % (95 % CI -11.7, -7.0). When compared to a typical American diet, the DASH diet changed 10-year ASCVD by -5.3 % (95 % CI -9.3, -1.2). Compared to a high sodium-control diet, the combination of both low sodium intake with DASH changed ASCVD risk by -14.1 % (95 % CI -18.6, -9.3).

CONCLUSIONS: Sodium reduction and the DASH diet both independently reduced 10-year ASCVD risk scores. Moreover, the combined impact was additive. These findings support dietary sodium reduction in addition to the DASH diet for ASCVD prevention.

STUDY OBJECTIVES: Time-restricted eating (TRE) is a novel dietary intervention targeting weight loss and cardiometabolic risk factors. The impact of TRE on sleep patterns remains under-explored.

METHODS: This was a post-hoc analysis of a parallel-arm, controlled feeding trial in 41 adults with obesity and prediabetes/diabetes, randomized to TRE (8am-6pm) or usual eating pattern (UEP; 8am-12am) for 12 weeks. We objectively determined sleep/wake patterns from 7-day wrist actigraphy data obtained at baseline and week-12. From this data, we derived total sleep time (TST) and sleep midpoint over a 24-hour period, sleep onset/offset, and sleep continuity measures. We used paired t-tests or Wilcoxon signed rank tests to compare data between baseline and week-12 within intervention arms and Mann-Whitney U tests or Wilcoxon signed rank tests to compare changes between intervention arms.

RESULTS: 38 participants (20 UEP; 18 TRE; 93% of those randomized in the parent trial) with adequate actigraphy data (mean age 59.6 ± 7.3 years, 92% female, 92% Black, mean BMI 36.3 ± 4.7 kg/m2) were analyzed. Compared to UEP, TRE increased TST by 55 minutes (p=0.03). TRE shifted sleep midpoint to 44 minutes earlier, from 3:24am to 2:40am (p=0.01), while UEP maintained the same sleep midpoint at 3:15am. Sleep onset shifted from a median of 12:22am to 11:52pm in TRE (p=0.03) while it remained stable in UEP (p=0.97). There were no differences in sleep offset and sleep continuity within and between intervention arms.

CONCLUSIONS: TRE increased sleep time and caused earlier sleep onset compared to UEP, revealing how timing of eating may affect sleep timing and duration.