Category: Statistics

JMIR Med Educ. 2026 Apr 2;12:e87524. doi: 10.2196/87524.

ABSTRACT

BACKGROUND: Women and visible minorities (VMs) continue to face discrimination while working in health care. These instances of discrimination can range from those perpetrated by individuals, such as misidentification, to prejudices held by institutions, such as a lack of commitment to hiring VMs. Furthermore, many residents face unique experiences of discrimination due to the intersection of gender and race. Although numerous studies have been published on the experiences of physicians with discrimination, there is a limited literature specifically on the experiences of residents.

OBJECTIVE: This study aimed to explore and analyze the impact of gender and racial discrimination on medical residents during their residency training.

METHODS: This scoping review was conducted in accordance with the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines. A health sciences librarian searched 5 databases: Ovid Medline(R) ALL, Embase (via Ovid), APA PsycInfo (via Ovid), CINAHL Plus with Full Text (via EBSCOhost), and Scopus. No study design, country, or date restrictions were applied. Studies that explored the impact of both gender and racial discrimination on residents during their residency training were included. Two reviewers conducted title and abstract screening, followed by full-text screening, and any discrepancies were resolved through group consensus. This study’s content was extracted using Microsoft Word to create tables for organizing and managing the information from the included studies.

RESULTS: After removing duplicates, the literature search revealed 2435 papers for title and abstract screening. A total of 340 papers were selected for full-text review, and ultimately, 26 papers met the inclusion criteria. Included papers were published between 1995 and 2024 in the United States (n=16), Canada (n=4), Australia and New Zealand (n=2), Saudi Arabia (n=1), Mexico (n=1), South Africa (n=1), and the United Kingdom (n=1). Our analysis revealed four themes: (1) forms of discrimination faced by VMs, (2) sources of discrimination, (3) ramifications of discrimination, and (4) ways for individuals and institutions to reduce gender and racial discrimination.

CONCLUSIONS: This scoping review identified that discrimination primarily manifests as microaggressions against residents and revealed the negative impact it continues to have on their careers. Raising awareness about these issues can help programs and institutions develop tailored solutions to tackle these problems and provide a safe, inclusive training environment for all residents.

PMID:41926156 | DOI:10.2196/87524

JMIR Form Res. 2026 Apr 2;10:e84947. doi: 10.2196/84947.

ABSTRACT

BACKGROUND: Nipah virus (NiV) infection is considered one of the deadliest infectious diseases, with a case fatality rate of approximately 71%. In Bangladesh, the primary risk factor for NiV infection is the consumption of raw date palm sap (DPS) contaminated with excreta from fruit bats (ie, members of the Pteropodidae family). Recently, the increasing use of social media among Bangladeshi youth has enabled business groups to widely advertise and sell raw DPS. This increased access, combined with young people perceiving consumption of raw DPS as an “adventurous event,” may facilitate an increase in incidences of NiV infection.

OBJECTIVE: We aimed to explore and document data from social media regarding raw DPS advertisements and consumption. Furthermore, we aimed to investigate the commercial distribution of raw DPS across the country.

METHODS: The data were accessed from common social media platforms used in Bangladesh, including Facebook and YouTube, between November 10, 2023, and January 31, 2024. We considered this period to capture public opinions, discussions, and reported incidents during the peak harvesting and consumption season of DPS. While DPS harvesting continues until March, early monitoring is essential for identifying the emerging concerns related to NiV transmission. Extracted variables were post dates and times, source locations, types of sources and posts, provider details (eg, sellers and gachis), post descriptions, user reactions, views, comments, and shares. Particular emphasis was placed on identifying the districts of both DPS suppliers and recipients. We used R version 4.3.2 and Stata version 15 for analyzing statistical data and QGIS for geographic data.

RESULTS: Of 556 social media posts, 361 (64.9%) were advertisements promoting raw DPS. Few DPS-related posts (n= 10, 1.8%) were associated with raising awareness about DPS consumption and NiV infection. In total, 473,724 people interacted with the social media posts. The identification of supplier and recipient districts revealed 14 source districts of raw DPS. The majority of raw DPS were distributed from Rajshahi, Naogaon, Rajbari, and Faridpur, which are among the most NiV-prone districts. During the data collection period, we observed an average of 996 (SD 377) liters of raw DPS sold per day from multiple gachhis (sap collectors) and sellers (vendors) in Rajshahi.

CONCLUSIONS: The marketing of raw DPS through digital media platforms has increased customer interest, as evidenced by the notably high level of public engagement observed on this topic within the Bangladeshi social media context. Distributing raw DPS using digital platforms is a marketing tool that significantly increases the availability of raw DPS to previously hard-to-reach markets and potentially increases individuals’ exposure to NiV infection. This study recommends a multidisciplinary approach, which incorporates context-specific public health investigations, policymaking, and digital media surveillance to address emerging public health concerns.

PMID:41926155 | DOI:10.2196/84947

JAMA Netw Open. 2026 Apr 1;9(4):e263758. doi: 10.1001/jamanetworkopen.2026.3758.

ABSTRACT

IMPORTANCE: Platform trials have emerged as a novel approach to clinical trials and are often described as a more resource- and cost-efficient design. Despite their growing relevance, the experiences and challenges of teams conducting platform trials remain underexplored.

OBJECTIVES: To examine the experiences and views of platform trial teams regarding the barriers and facilitators raised when conducting platform trials.

DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional survey study was carried out from June to August 2023 and was analyzed with a mixed-methods approach. The survey was sent to 127 principal investigators of platform trials and 11 other stakeholders who are experienced in the methodology and conduct of platform trials. The survey could be forwarded to other relevant team members of a platform trial.

MAIN OUTCOMES AND MEASURES: Survey data on barriers and facilitators to conducting platform trials were analyzed using descriptive statistics and standard qualitative content analysis.

RESULTS: A total of 40 completed surveys were returned. Respondents’ roles in platform trials included trial management (22 respondents [55%]), principal investigator (12 respondents [30%]), statistics and trial methodology (11 respondents [28%]), and data management (11 respondents [28%]). Respondents reported efficacy gains from using a shared coordinating infrastructure for multiple intervention arms as the key advantage of platform trials, although these require more planning time and greater team expertise than traditional randomized clinical trials. There was no consensus on whether the perceived costs and resources needed for adding a comparison within a platform trial were lower or higher compared with setting up a traditional randomized clinical trial. A repeatedly raised challenge was that current regulatory and organizational processes are set up for intervention-specific clinical trials and are incompatible with adaptive and open-ended platform trials. Targeted supports such as guidelines, templates, funding mechanisms, and infrastructure setup for platform trials were identified as facilitators for their conduct.

CONCLUSIONS AND RELEVANCE: In this survey study of platform trial experts, it was found that the clinical research landscape is currently not set up for adaptive platform trials, which leads to substantial challenges conducting such trials. These findings suggest that stakeholders need to take concerted action to enable the integration of platform trials.

PMID:41926125 | DOI:10.1001/jamanetworkopen.2026.3758

JAMA Netw Open. 2026 Apr 1;9(4):e264421. doi: 10.1001/jamanetworkopen.2026.4421.

ABSTRACT

IMPORTANCE: Low back pain (LBP) is a leading cause of disability, and there is limited evidence from clinical practice to support the effectiveness of alternative care models.

OBJECTIVE: To compare a sequenced care pathway (SCP) with a pain navigator pathway (PNP) for patients with LBP.

DESIGN, SETTING, AND PARTICIPANTS: In this embedded cluster randomized clinical trial, 19 primary care clinics in the Veterans Health Administration were randomized to deliver 1 of 2 multimodal guideline-supported care pathways, with primary outcomes assessed in their electronic health records (EHRs) at 3 months. Between February 8, 2021 (first enrolled), and January 31, 2024 (last enrolled), 1817 participants were referred by primary care clinicians and attended an initial AIM-Back trial visit. A subset of 799 participants consented to complete additional questionnaires for secondary analyses (March 8, 2021 [first survey collected], to January 10, 2025 [final secondary outcome collected by survey]).

INTERVENTIONS: The SCP included pain education and modulation, physical activity coaching, risk stratification, and psychologically informed physical therapy. The PNP included shared decision-making and facilitated referrals to nondrug treatments.

MAIN OUTCOMES AND MEASURES: Pain interference and physical function were coprimary outcomes, assessed with the Patient-Reported Outcomes Measurement Information Systems 4-item Short Forms (PROMIS-SF; potential score range for pain interference, 41.6-75.6, where lower scores indicated less interference with daily activities due to pain; and potential score range for physical function, 22.5-57.0, where higher scores indicated higher physical functioning during daily activities). Secondary EHR outcomes included sleep disturbance and National Institutes of Health pain intensity, and survey outcomes included the coprimary outcomes and additional measures of pain, function, and quality of life. Analysis was performed in the intent-to-treat population.

RESULTS: There were 1817 enrolled participants (SCP, 811; PNP, 1006; mean [SD] age, 53.0 [15.7] years; 1597 men [87.9%]). At 3 months, 461 of 811 patients (56.8%) in the SCP group and 537 of 1006 (53.4%) in the PNP group had analyzable primary outcomes. The estimated baseline mean PROMIS-SF score was 63.2 points (97.5% CI, 62.7-63.6 points) for pain interference and 37.1 points (97.5% CI, 36.7-37.4 points) for physical function. The 3-month mean PROMIS-SF score for pain interference was 60.5 points (97.5% CI, 59.7-61.3 points) in the SCP group and 61.1 points (97.5% CI, 60.4-61.8 points) in the PNP group. The 3-month mean PROMIS-SF score for physical function was 39.1 points (97.5% CI, 38.4-39.7 points) in the SCP group and 38.5 points (97.5% CI, 37.8-39.1 points) in the PNP group. There was no SCP superiority, with estimated 3-month differences of -0.6 points (97.5% CI, -1.6 to 0.4 points) for pain interference and 0.6 points (97.5% CI, -0.3 to 1.5 points) for physical function. There were no pathway differences in secondary outcomes.

CONCLUSION AND RELEVANCE: In this cluster randomized trial, the SCP was not superior for the primary outcomes of pain interference and physical function. Future research should consider designs that optimize pathway adherence, assess the effectiveness in other settings, and investigate patient-level factors indicative of a favorable response to the SCP or PNP.

TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT04411420.

PMID:41926124 | DOI:10.1001/jamanetworkopen.2026.4421

JAMA Netw Open. 2026 Apr 1;9(4):e264636. doi: 10.1001/jamanetworkopen.2026.4636.

ABSTRACT

IMPORTANCE: Direct care workers (DCWs) provide essential support for millions of older individuals in the US, but high workforce turnover-the rate at which workers leave the DCW workforce-undermines care access and quality. Structural factors associated with DCW working conditions, such as unionization and employer ownership status, may play an important role in DCW retention, but their association with DCW workforce turnover is not known.

OBJECTIVE: To assess the association of unionization and ownership with workforce turnover among DCWs.

DESIGN, SETTING, AND PARTICIPANTS: Data on DCW turnover for this cross-sectional study were obtained from the Outgoing Rotation Groups of the Current Population Survey, an annual survey of 60 000 US households, from January 1, 2009, to December 31, 2024. The study population included individuals aged 15 years or older who were employed as DCWs.

EXPOSURES: Unionization (whether the respondent was covered by a union through their DCW role) and employer ownership status (whether their employer was for profit, nonprofit, or publicly owned).

MAIN OUTCOMES AND MEASURES: The main outcome of interest was workforce turnover at 1 year, defined as reporting a non-DCW occupation or no occupation 1 year after the initial survey. Bivariate analyses and logistic regression were used to compare DCW turnover rates across union status and employer ownership type and test whether ownership type moderates the association between unionization and turnover.

RESULTS: The overall sample included 18 175 DCWs (mean [SD] age, 44.0 [14.7] years; 15 860 female DCWs [86.5%]). In the fully adjusted models, the estimated probability of turnover was significantly lower among unionized DCWs than nonunionized DCWs overall (37.4% vs 45.0%; odds ratio [OR], 0.72 [95% CI, 0.64-0.81]), at nonprofit organizations (33.6% vs 47.1%; OR, 0.56 [95% CI, 0.39-0.80]) and at for-profit organizations (35.2% vs 45.9%; OR, 0.63 [95% CI, 0.54-0.75]), but not at public employers (39.8% vs 41.0%; OR, 0.95 [95% CI, 0.78-1.16]). Public ownership was also directly associated with lower turnover (39.1% vs 41.8%; OR, 0.89 [95% CI, 0.80-1.00]) compared with for-profit ownership.

CONCLUSIONS AND RELEVANCE: In this cross-sectional study of DCWs, employer ownership status and unionization were independently and jointly associated with DCW workforce turnover rates, suggesting that these structural factors may play an important role in DCW retention. State and federal policies that facilitate DCW unionization or public employment of DCWs may significantly improve DCW retention.

PMID:41926123 | DOI:10.1001/jamanetworkopen.2026.4636

JAMA Netw Open. 2026 Apr 1;9(4):e264642. doi: 10.1001/jamanetworkopen.2026.4642.

ABSTRACT

IMPORTANCE: Thiazide diuretics are a cornerstone for the treatment of hypertension, but their use is associated with development of hyponatremia. Women and older adults are particularly vulnerable, but population-based estimates of absolute risk are largely absent. Such data are a prerequisite for a robust risk-benefit assessment in the clinical setting.

OBJECTIVE: To compare new use of thiazide diuretics with calcium channel blockers (CCBs) and subsequent risk of hyponatremia among different age groups and between the sexes.

DESIGN, SETTING, AND PARTICIPANTS: This propensity score-matched cohort study included 159 080 individuals 18 years or older in the Stockholm Sodium Cohort, a research database established to investigate the association between thiazide and hyponatremia among individuals in Stockholm, Sweden, between July 1, 2006, and December 31, 2018. Statistical analysis was performed between January 2025 and January 2026.

EXPOSURE: Newly initiated treatment with thiazide diuretics and CCBs.

MAIN OUTCOMES AND MEASURES: The primary outcome was profound hyponatremia (ie, a sodium concentration <125 mEq/L). Secondary outcomes were sodium concentrations less than 130 mEq/L and less than 135 mEq/L.

RESULTS: A total of 79 540 individuals (median age, 63 years [IQR, 54-72 years]; 41 275 women [51.9%]) initiating thiazide treatment were propensity score matched with 79 540 individuals (median age, 63 years [IQR, 54-72 years]; 41 168 women [51.8%]) receiving CCBs. The cumulative incidence of profound hyponatremia was 0.80% (95% CI, 0.74%-0.87%) for thiazide users and 0.46% (95% CI, 0.41%-0.51%) for CCB users during the first 2 years of treatment. The occurrence of profound hyponatremia with thiazide treatment was higher among women (cumulative incidence, 1.04% [95% CI, 0.94%-1.15%) and individuals 80 years or older (cumulative incidence, 2.40% [95% CI, 2.07%-2.73%]). Thus, among women 80 years or older, the number needed to harm (NNH) was 53 (95% CI, 41-73) for developing sodium concentrations less than 125 mEq/L, 28 (95% CI, 22-38) for concentrations less than 130 mEq/L, and 16 (95% CI, 13-20) for concentrations less than 135 mEq/L. This was in marked contrast with women younger than 65 years, for whom the corresponding NNH was 790 (95% CI, 408-11 966) for developing sodium concentrations less than 125 mEq/L, 818 (95% CI, 303-∞) for concentrations less than 130 mEq/L, and 120 (95% CI, 78-261) for concentrations less than 135 mEq/L.

CONCLUSIONS AND RELEVANCE: In this cohort study comprising 159 080 individuals, the association between newly initiated thiazide diuretics and hyponatremia was negligible among individuals younger than 65 years of age. In contrast, among older adults, especially among women, the association was substantial. The results may incentivize the prescriber toward choosing an alternative antihypertensive treatment. If thiazides are initiated, subsequent monitoring of serum sodium concentrations should be considered.

PMID:41926122 | DOI:10.1001/jamanetworkopen.2026.4642

JAMA Netw Open. 2026 Apr 1;9(4):e264780. doi: 10.1001/jamanetworkopen.2026.4780.

ABSTRACT

IMPORTANCE: Latino/a adults in the US are at higher risk for developing type 2 diabetes and are less likely to achieve goals in the National Diabetes Prevention Program (NDPP) than non-Latino/a White adults. Little is known about the barriers and facilitators to diabetes prevention behaviors among Latino/a adults.

OBJECTIVE: To describe the perspectives of Latino/a NDPP participants regarding barriers and facilitators to diabetes prevention behaviors.

DESIGN, SETTING, AND PARTICIPANTS: This qualitative study using semistructured phone interviews with Spanish-speaking Latino/a adults who participated in a community-based NDPP in an urban and suburban area of Colorado was conducted from June 2022 to August 2023.

MAIN OUTCOME AND MEASURES: The primary outcome was themes and subthemes reflecting individual and family support perspectives. Interviews were audio-recorded and transcribed. An inductive thematic analysis was conducted.

RESULTS: In total, 22 Spanish-speaking Latino/a participants, including 17 women (77%) and 5 men (22%) with a mean (SD) age of 44 (8) years, were interviewed. The interviews revealed 5 themes with respective subthemes. The first theme and subthemes described the perceived impact of culture and self-perception (frustration and presence of self-critical beliefs, perceived traditional gender roles, and internalized cultural biases of health behaviors) on diabetes prevention behaviors. The next 2 themes described family as reciprocal catalysts for diabetes prevention action (perceived family exercise engagement, dietary education influencing family eating habits, and collective emotional impact and accountability) and having a culturally and linguistically aligned program to support participants (perceived support and motivation from community health workers, culturally and linguistically concordant information as essential, and perceived peer support through shared experiences). The last 2 themes described reflection on personal motivation for behavior change (family diabetes lived experience awareness and motivation, desire to model lifestyle behaviors, and perceived mental health and spiritual appreciation) and desired program components for family inclusion (perceived role of knowledge for family participation, structured family-friendly wellness activities, and preventing diabetes among children).

CONCLUSIONS AND RELEVANCE: In this qualitative study of Spanish-speaking Latino/a adults, participating in a diabetes prevention program, gender roles and internalized cultural biases were perceived as barriers to engaging in diabetes prevention behaviors, while family support and culturally aligned programming served as key facilitators. These findings may be used to improve diabetes prevention programs in Latino communities.

PMID:41926121 | DOI:10.1001/jamanetworkopen.2026.4780

JAMA Netw Open. 2026 Apr 1;9(4):e264788. doi: 10.1001/jamanetworkopen.2026.4788.

ABSTRACT

IMPORTANCE: The Outcome and Assessment Information Set (OASIS) is essential for measuring home health quality and outcomes, and accurate linkage between OASIS, Medicare enrollment, and claims is crucial for monitoring utilization and informing payment policy. Preliminary documentation suggested a decrease in OASIS-beneficiary match rates beginning in 2019, but the extent and implications of this decrease have not been quantified.

OBJECTIVE: To quantify annual match rates between OASIS assessments and Medicare enrollment and fee-for-service (FFS) home health claims from 2017 through 2023.

DESIGN, SETTING, AND PARTICIPANTS: This cohort study used 100% OASIS assessments from 2017-2023 linked to the Master Beneficiary Summary File (MBSF) and FFS home health claims. OASIS assessments included patients receiving home health care in any payer category, while claims analyses were restricted to FFS beneficiaries. Data were analyzed from June 2025 to January 2026.

MAIN OUTCOMES AND MEASURES: The yearly number and proportion of OASIS assessments that can be matched to a person in the MBSF and the yearly number and proportion of FFS home health claims that can be matched to an OASIS assessment.

RESULTS: Across 2017-2023, approximately 18 million OASIS assessments and 6 million unique assessed beneficiaries were recorded annually. The proportion of assessments linked to a Medicare beneficiary decreased from 89.8% in 2017 to 76.4% in 2023. Among unmatched assessments, the share with Medicare as the recorded payer increased, while the share with Medicaid decreased. For FFS beneficiaries, the number of claims increased due to payment cycle changes, yet the number of unique FFS claimants decreased from 3 424 394 in 2017 to 2 636 931 in 2023. The match rate of FFS claims to an OASIS assessment decreased from 96.8% to 73.9%, with variation across states; by 2023, all state match rates were below 90%.

CONCLUSIONS AND RELEVANCE: The findings of this study suggest a reduced ability to accurately attribute OASIS assessments to individual Medicare beneficiaries or matching claims to corresponding OASIS assessments using Centers for Medicare & Medicaid Services (CMS) Research Identifiable Files. Until these issues are addressed by CMS, researchers should cautiously interpret findings utilizing individually linked Medicare claims and OASIS assessment data.

PMID:41926120 | DOI:10.1001/jamanetworkopen.2026.4788

JAMA Netw Open. 2026 Apr 1;9(4):e265039. doi: 10.1001/jamanetworkopen.2026.5039.

ABSTRACT

IMPORTANCE: Liver transplants are performed around the clock, often associated with substantial disutility for patients and clinicians. While short-duration dual hypothermic oxygenated machine perfusion (short-DHOPE) mitigates ischemia-reperfusion injury and related complications, prolonged DHOPE (DHOPE-PRO) may further extend preservation time and facilitate daytime liver transplant.

OBJECTIVE: To assess whether the use of DHOPE-PRO is associated with an increased proportion of daytime liver transplants without compromising graft or patient outcomes.

DESIGN, SETTING, AND PARTICIPANTS: This prospective cohort study conducted at a large academic liver transplant center in the Netherlands included adult and pediatric recipients of liver grafts received from donation after brain death (DBD), donation after circulatory death (DCD), or living donors. The study compared all liver transplants performed between January 1, 2023, and December 31, 2024, following routine DHOPE-PRO implementation, with a control cohort of all consecutive liver transplants performed between January 1, 2021, and December 31, 2022. Follow-up continued through December 31, 2025. Outcomes were stratified by graft type.

EXPOSURES: Grafts underwent DHOPE-PRO, with or without viability assessment depending on graft risk, or short-DHOPE (for DCD livers) or no perfusion (for DBD livers).

MAIN OUTCOMES AND MEASURES: The primary outcome was the difference in the percentage of daytime liver transplants, defined as surgery starting at or after 8 am and either (1) reperfusion occurring before 8 pm or (2) completion by midnight, between the 2021-2022 and 2023-2024 cohorts. Secondary outcomes included 1-year patient and graft survival, intraoperative parameters, and postoperative complications, stratified by graft type.

RESULTS: A total of 330 liver transplants (median [IQR] age, 45 [13-62] years; 186 [56.4%] male) were included in the analyses: 155 transplants in the 2021-2022 cohort and 175 transplants in the 2023-2024 cohort. Following DHOPE-PRO implementation, the percentage of daytime transplants increased from 48.4% (75 of 155) to 84.6% (148 of 175) for reperfusion (P < .001) and from 53.5% (83 of 155) to 89.1% (156 of 175) for completion (P < .001). Median (IQR) DHOPE duration increased from 2.1 (1.6-4.1) to 10.2 (5.1-13.1) hours (P < .001), with total preservation times up to 31.4 hours. Use of DHOPE-PRO showed no association with postoperative complications (eg, new-onset acute kidney injury; odds ratio, 0.64 [95% CI, 0.37-1.07]; P = .09]). One-year patient survival exceeded 90% in all subgroups. Use of DHOPE-PRO was not associated with either graft (hazard ratio, 1.28 [95% CI, 0.59-2.74]; P = .53) or patient (hazard ratio, 2.05 [95% CI, 0.75-5.59]; P = .16) survival.

CONCLUSIONS AND RELEVANCE: In this prospective cohort study, routine implementation of DHOPE-PRO was associated with increased proportions of daytime liver transplants and with improved surgical logistics. It was also associated with outcomes similar to those after short-duration DHOPE, supporting its broader application in clinical practice.

PMID:41926119 | DOI:10.1001/jamanetworkopen.2026.5039

JAMA Otolaryngol Head Neck Surg. 2026 Apr 2. doi: 10.1001/jamaoto.2026.0166. Online ahead of print.

ABSTRACT

IMPORTANCE: Malnutrition, common in patients with head and neck squamous cell carcinoma (HNSCC), may impair the effectiveness of immunotherapy. Understanding whether nutritional status affects outcomes can emphasize the importance of pretreatment nutritional optimization.

OBJECTIVE: To evaluate the associations of body mass index (BMI), pretreatment BMI change, and prognostic nutritional index (PNI) with progression-free survival (PFS) in patients with advanced HNSCC treated with immunotherapy.

DESIGN, SETTING, AND PARTICIPANTS: Cohort study using deidentified electronic health records from the Flatiron Health database between January 2014 to January 2024, with follow-up of 3000 days in multiple cancer institutions across the US. Patients were from community and academic practices in the US, diagnosed with all types of advanced head and neck cancer who received immunotherapy (nivolumab, pembrolizumab, cemiplimab, durvalumab, atezolizumab, avelumab, or ipilimumab) between January 2014 and January 2024. Exclusion criteria included age younger than 18 years, patients without a diagnosis of squamous cell carcinoma, missing stage information, missing treatment information (eg, a gap in documented care of ≥90 days or more), and incomplete data needed to calculate BMI or PNI prior to initiation of therapy.

EXPOSURES: Baseline BMI, pretreatment BMI change (≥2% decrease vs stable), and PNI (low <45 vs normal ≥45).

MAIN OUTCOMES AND MEASURES: PFS following treatment with immunotherapy.

RESULTS: Among 1108 patients (mean [SD] age, 66.2 [10.2] years; 236 [21.3%] female; 872 [78.7%] male), 214 (79%) experienced 2% or more pretreatment BMI loss. BMI loss was associated with worse PFS (hazard ratio, 1.17; 95% CI, 1.02-1.35); median PFS was 271 (IQR, 121-603) days with loss vs 415 days with stable BMI. In the 471 patients with laboratory data, 320 (67.9%) had a low PNI. Low PNI was associated with shorter PFS (adjusted hazard ratio, 1.58; 95% CI, 1.39-1.79); median PFS was 213 (IQR, 98-445) days for low vs 566 (IQR, 307-1094) days for patients with a normal PNI. Baseline BMI category was not independently associated with PFS.

CONCLUSIONS AND RELEVANCE: In this large, nationally representative cohort of patients with advanced head and neck squamous cell carcinoma treated with immunotherapy, pretreatment BMI loss and low PNI were independently associated with PFS, while baseline BMI was not. These findings suggest that dynamic measures of nutrition and immune status provide more meaningful prognostic information than static measures. Nutritional optimization may represent a modifiable factor to improve outcomes in patients receiving immunotherapy for advanced head and neck squamous cell carcinoma.

PMID:41926098 | DOI:10.1001/jamaoto.2026.0166