OBJECTIVE: Public support for nutrition policies depends on beliefs about responsibility for dietary behaviour. We explored how perceptions of responsibility for healthy and sustainable eating relate to policy acceptabil...OBJECTIVE: Public support for nutrition policies depends on beliefs about responsibility for dietary behaviour. We explored how perceptions of responsibility for healthy and sustainable eating relate to policy acceptability in a representative German sample. METHODS: Cross-sectional survey data were collected online (December 2023-January 2024). Participants rated the responsibility of the state, individuals, cafeterias, food industry, and parents for healthy and sustainable eating among adults and children, and the acceptability of nutrition policies. K-means cluster analyses identified patterns of responsibility attribution across four domains: adult health, adult sustainability, child health, and child sustainability. Differences in demographics and acceptability were assessed using χ tests, t-tests, and linear regression. RESULTS: Among 2001 participants, individuals were perceived as most and the state as least responsible for dietary behaviour. Two clusters emerged: a "Shared Responsibility" cluster assigning responsibility to all actors and a "Directed Responsibility" cluster assigning responsibility to industry (adult sustainability) or individuals/parents (other domains). Compared with the Directed Responsibility cluster, participants in the Shared Responsibility cluster reported greater policy acceptability (b = 0.45-0.56, all p < .001). Directed Responsibility participants were older, reported poorer health, and differed politically. CONCLUSION: Emphasising shared responsibility for dietary behaviour may increase policy acceptability.
This analytic essay reviews the transition from unidimensional relative reinforcing efficacy to multidimensional behavioral economic demand analysis in drug reinforcement, focusing on tobacco science. It synthesizes oper...This analytic essay reviews the transition from unidimensional relative reinforcing efficacy to multidimensional behavioral economic demand analysis in drug reinforcement, focusing on tobacco science. It synthesizes operant laboratory studies in human smokers that use operant demand curve analysis. Reviewed studies used single- and concurrent-commodity sessions in which participants performed operant responses to earn the self-administration of controlled cigarettes puffs, and other commodities included money, and self-administration of denicotinized cigarette puffs, reduced-nicotine cigarette puffs, and nicotine gum. Unlike traditional measures purporting to assess relative reinforcing efficacy, demand curves showed multidimensional reinforcement. Operant demand studies resulted in findings relevant to tobacco science, therapeutics, and regulation. Despite greater demand for money at lower prices, cigarette puffs showed greater resistance to increasing prices. Denicotinized cigarette puffs were more effective substitutes (than nicotine gum) for nicotine-containing cigarettes. Reduced-nicotine and full-nicotine experimental cigarette exposure yielded modest, non-dose-dependent demand reductions. Greenwashing tactics increased demand intensity at low prices, enhancing unconstrained preference. Behavioral economic demand analysis offers a superior framework for understanding and assessing reinforcement over the concept of relative reinforcing efficacy. These findings underscore demand analysis as a versatile tool for investigating abuse liability, therapeutics, and tobacco regulatory science.
OBJECTIVE: Adverse childhood experiences (ACEs) are important determinants of long-term health and behavior, but limited research has examined their association with household firearm ownership. METHODS: We analyzed data...OBJECTIVE: Adverse childhood experiences (ACEs) are important determinants of long-term health and behavior, but limited research has examined their association with household firearm ownership. METHODS: We analyzed data from the 2023 Behavioral Risk Factor Surveillance System in Nevada, New Jersey, Oregon, and Virginia. ACEs and household firearm ownership were self-reported. Weighted multivariable logistic regression was performed to examine the associations. RESULTS: This study included 15,771 participants, of whom 4739 (30.1%, unweighted) self-reported household firearm ownership. Compared with individuals reporting no ACEs, those with 1-2 types of ACEs (adjusted odds ratio [aOR]: 1.21, 95% CI: 1.06, 1.39), 3-5 types of ACEs (aOR: 1.24, 95% CI: 1.06, 1.46), and 6 or more types of ACEs (aOR: 1.30, 95% CI: 1.02, 1.65) had higher odds of household firearm ownership. Among specific types of ACEs, household illegal drug use (aOR: 1.37, 95% CI: 1.13, 1.66), household incarceration (aOR: 1.28, 95% CI: 1.02, 1.59), and verbal abuse (aOR: 1.20, 95% CI: 1.06, 1.36) were each associated with increased likelihood of household firearm ownership. CONCLUSIONS: In these four states, ACEs were positively associated with household firearm ownership, suggesting early-life adversity may shape firearm-related behaviors and highlight its relevance for public health interventions.
OBJECTIVE: Diabetic ketoacidosis (DKA) is a life-threatening complication of type 1 diabetes (T1D) and a marker of delayed diagnosis, associated with significant morbidity, mortality, and healthcare costs. Global estimat...OBJECTIVE: Diabetic ketoacidosis (DKA) is a life-threatening complication of type 1 diabetes (T1D) and a marker of delayed diagnosis, associated with significant morbidity, mortality, and healthcare costs. Global estimates of DKA at T1D onset remain inconsistent. METHODS: We conducted a systematic review and meta-analysis, searching PubMed (1996), Web of Science (1966), Embase (1947), and Scopus (1980), each from its inception to October 2025, without restrictions. Studies reporting DKA prevalence at diagnosis in individuals <19 years with newly diagnosed T1D were included. Pooled prevalence was calculated using random-effects meta-analysis. Subgroup analyses explored variations by country, region, World Bank income level, and pre- versus post-COVID-19 period; meta-regression examined continuous moderators. RESULTS: A total of 233 studies from 58 countries (380,191 participants) were included. The pooled global prevalence was 41.9% (CI = 39.7, 44.0; I = 99.3%; prediction interval: 12.9, 74.3), ranging from 15.6% (Sweden) to 78.5% (Thailand). Rates were lower in high-income (38.6%) countries (p < 0.001). Younger age and higher female proportion were independently associated with higher prevalence. CONCLUSIONS: Approximately two in five children with new-onset T1D present in DKA. The rising trend is indicative of ongoing challenges in achieving timely diagnosis, demanding urgent public health strategies encompassing clinical vigilance and innovative early detection approaches.
OBJECTIVE: This study aimed to examine whether employment transitions are associated with within-person variation in binge drinking and whether unemployment insurance may buffer this risk. METHODS: Using 19 waves of data...OBJECTIVE: This study aimed to examine whether employment transitions are associated with within-person variation in binge drinking and whether unemployment insurance may buffer this risk. METHODS: Using 19 waves of data from the Understanding America Study (May 2020-March 2021; N = 1710) in the United States, mixed-effects models examined within-person associations between employment transitions and binge drinking days. Postestimation contrasts evaluated restorative changes associated with upward employment transitions. RESULTS: Shifts from secure employment to underemployment (β = 0.033, p < .10) or unemployment without benefits (β = 0.055, p < .10) were associated with more binge drinking. Loss of benefits while unemployed (β = 0.130, p < .05) and remaining unemployed without benefits (β = 0.083, p < .01) were also associated with more binge drinking. Patterns consistent with restorative changes were observed only for shifts to secure employment and not for those involving underemployment. Gaining unemployment benefits while unemployed was also associated with less binge drinking, compared to remaining unemployed without benefits (β = -0.116, p < .05). CONCLUSIONS: Employment transitions are differentially associated with binge drinking. Underemployment may represent a distinct state for alcohol-related risk, and unemployment insurance may mitigate binge drinking during employment insecurity.
E-cigarettes are an effective way to stop smoking but there is disagreement regarding their use as treatment. One concern is that recommending e-cigarettes for adults who smoke (AWS) could result in sustained dual use of...E-cigarettes are an effective way to stop smoking but there is disagreement regarding their use as treatment. One concern is that recommending e-cigarettes for adults who smoke (AWS) could result in sustained dual use of both products among those who fail to quit smoking. Given most quit attempts end in failure, it is important to understand the likelihood of developing dual use and its consequences when considering e-cigarettes for AWS. We describe variability in dual use within a behavioral economics framework, including literature which addresses the likelihood that trying to quit smoking with e-cigarettes will lead to established dual use as well as the impact that dual use has on smoking heaviness and smoking cessation. Limited evidence suggests that a meaningful minority of AWS who initiate e-cigarettes with support to quit smoking establish dual use. Compared to exclusive smoking, dual use appears to be associated with an overall reduction in smoking and an increase in the likelihood of cessation. However, outcomes appear to vary based on substitutive versus complementary use patterns, with the potential for dual use to lead to sustained or increased smoking for a small proportion of AWS who initiate e-cigarettes. Compared to exclusive smoking, developing dual use is likely associated with reduced smoking for the majority of AWS, providing further support for including e-cigarettes as a smoking cessation treatment option. Future prospective and controlled research is needed to explore predictors of developing substitutive versus complementary patterns of dual use and subsequent smoking cessation.
OBJECTIVE: To estimate the association between maternal exposure to Head Start and offspring mental health. METHODS: In this retrospective cohort study using a difference-in-differences approach, we leveraged the quasi-r...OBJECTIVE: To estimate the association between maternal exposure to Head Start and offspring mental health. METHODS: In this retrospective cohort study using a difference-in-differences approach, we leveraged the quasi-random rollout of Head Start in the United States in the 1960's and the timing of birth cohorts in the National Longitudinal Survey of Youth 1979 to estimate associations between maternal Head Start exposure and offspring depression (among those aged 15+ in years 1994-2020) and suicidal ideation (among those aged 15+ in years 2012-2020). We examined heterogeneity by offspring race, ethnicity, and sex. RESULTS: Maternal Head Start exposure was associated with 0.79 (95% CI = 0.62, 0.99) times the risk of depression among female Hispanic/Latina offspring and 0.66 (95% CI = 0.50, 0.87) times the risk of depression among female White offspring. We found null associations for other races, ethnicities, and males, and for the outcome of suicidal ideation. CONCLUSIONS: Maternal Head Start exposure may meaningfully reduce risk of depression among Hispanic/Latina and White female offspring-groups who tend to report higher risk of depression. POLICY IMPLICATIONS: Findings suggest social policies that invest in the social, economic, and human capital of vulnerable children can improve mental health outcomes across generations.
OBJECTIVE: To examine whether objective audiometric hearing loss and subjective hearing difficulty show differential associations with depressive symptoms in a nationally representative Korean population. METHODS: Data w...OBJECTIVE: To examine whether objective audiometric hearing loss and subjective hearing difficulty show differential associations with depressive symptoms in a nationally representative Korean population. METHODS: Data were obtained from 6589 participants aged ≥40 years in the Korea National Health and Nutrition Examination Survey (2020 and 2022). Hearing status was classified into four groups based on pure-tone audiometry (≥26 dB HL at 0.5-4 kHz in the better ear) and self-reported hearing difficulty. Depressive symptoms were defined as a Patient Health Questionnaire-9 (PHQ-9) score ≥ 10, with symptom-level analyses conducted for each PHQ-9 item. Survey-weighted logistic regression estimated odds ratios (ORs) and 95% confidence intervals (CIs). RESULTS: Overall, 4.3% of participants had depressive symptoms. Compared with normal hearing, participants with subjective hearing difficulty but no audiometric hearing loss had higher odds of depressive symptoms (OR = 1.74, 95% CI = 1.01, 3.00). Symptom-level analyses showed significant associations between subjective hearing difficulty without audiometric loss and psychomotor changes, suicidal ideation, concentration difficulties, low self-esteem, and appetite changes. CONCLUSIONS: Subjective hearing difficulty, even without audiometric loss, was associated with depressive symptoms. As an easily ascertainable measure, it may represent a marker of vulnerability to depressive symptoms and may warrant consideration in mental health screening.
OBJECTIVES: This umbrella review tested the hypothesis that increased sedentary time and behaviours (e.g., television viewing) would be associated with increased risk in all-cause mortality, even when accounting for phys...OBJECTIVES: This umbrella review tested the hypothesis that increased sedentary time and behaviours (e.g., television viewing) would be associated with increased risk in all-cause mortality, even when accounting for physical activity and measured using device-based and self-report measures. METHODS: Reviews were included if they conducted a systematic review and/or meta-analysis on the impact of sedentary time/behaviours on all-cause mortality in adults. Sources were searched from inception-March 2025 and included Scopus, EMBASE, MEDLINE, CINAHL, and Academic Search Premier with no date restrictions applied. The Joanna Briggs Institute and Risk of Bias in Systematic Reviews tools assessed study quality and PRISMA reporting were followed. RESULTS: Thirty reviews (n = 83 unique studies, n = 2,778,595 unique participants) were included and investigated total sedentary time (n = 28, meta-analyses n = 17) and television viewing time (n = 7, meta-analyses n = 7). Average study quality was 9.6 ± 1.6/11.0. n = 29/30 observed a positive relationship between sedentary time or television viewing time with all-cause mortality. Results remained significant but were attenuated after accounting for physical activity. Results were consistent across studies using self-report and device-based sedentary measures. CONCLUSION: This umbrella review of high-quality reviews provides evidence that increased sedentary time and television viewing represent a unique risk for all-cause mortality, even when accounting for physical activity.
OBJECTIVE: Human papillomavirus (HPV) causes oropharyngeal, anal, and penile cancers in men. In 2019, HPV vaccination for boys was included in Denmark's free-of-charge childhood vaccination program. This study examines s...OBJECTIVE: Human papillomavirus (HPV) causes oropharyngeal, anal, and penile cancers in men. In 2019, HPV vaccination for boys was included in Denmark's free-of-charge childhood vaccination program. This study examines sociodemographic and parental health factors associated with HPV vaccine initiation among Danish boys. METHODS: This nationwide, register-based cohort study included all boys living in Denmark and born July 2007-December 2010 (n = 113,519), representing the first eligible cohorts. Boys were followed up until December 2024, and vaccine initiation within 12 months after their 12th birthday was identified through national registers, which also provided sociodemographic factors and parental medical conditions. Logistic regression estimated odds ratios (ORs) and 95% confidence intervals (CIs) for associations with vaccine initiation. RESULTS: Prevalence of HPV vaccine initiation was 79.5%. Uptake was lower among boys of immigrant origin (OR = 0.85; 95%CI: 0.79, 0.90) and those not living with both legal parents. Uptake was higher among boys from families with higher education (OR = 1.29; 95%CI: 1.23, 1.35), higher income (OR = 1.95; 95%CI: 1.85, 2.05), or a parental history of cancer or HPV-associated precancer (OR = 1.11; 95%CI: 1.06, 1.17). CONCLUSIONS: Most eligible Danish boys initiate HPV vaccination. Social disparities persist despite universal free access, and key predictors of uptake were ethnicity, family structure, and socioeconomic status.
OBJECTIVE: This study aimed to assess vaccination coverage among individuals receiving HIV pre-exposure prophylaxis (PrEP). METHODS: A systematic review and meta-analysis were conducted following PRISMA guidelines. PubMe...OBJECTIVE: This study aimed to assess vaccination coverage among individuals receiving HIV pre-exposure prophylaxis (PrEP). METHODS: A systematic review and meta-analysis were conducted following PRISMA guidelines. PubMed, Scopus, and Web of Science were searched from January 2011 to March 2026. Studies assessing vaccination coverage among individuals receiving PrEP within routine care were included. Random-effects meta-analyses were performed for each vaccine, and subgroup analyses were conducted when adequate data were available. RESULTS: Twenty-four studies including 13,087 individuals were analysed. Vaccination coverage was highest for HBV (73.00%, 95% CI [59.00, 88.00]), followed by Monkeypox (61.00%, 95% CI [40.00, 83.00]) and HAV (59.00%, 95% CI [48.00, 71.00]), and lowest for HPV (43.00%, 95% CI [34.00, 52.00]). Evidence on other vaccines was limited. High heterogeneity was observed across studies, and vaccination status was frequently self-reported (63.00%). No consistent geographical pattern emerged. CONCLUSIONS: Vaccination coverage among individuals receiving PrEP remains relatively low across most key vaccines, highlighting missed opportunities for prevention. Strengthening the integration of vaccination within PrEP care pathways and enhancing proactive provider engagement are key priorities. Future efforts should focus on high-quality primary studies, standardized methodologies, evaluation of targeted interventions, and economic evidence to better inform policy decisions.
OBJECTIVES: Pregestational diabetes (PDM) and gestational diabetes mellitus (GDM) have increased in the U.S., with persistent racial/ethnic disparities. We examined PDM and GDM prevalence and associated social determinan...OBJECTIVES: Pregestational diabetes (PDM) and gestational diabetes mellitus (GDM) have increased in the U.S., with persistent racial/ethnic disparities. We examined PDM and GDM prevalence and associated social determinants of health (SDOH) across disaggregated racial/ethnic subgroups in California. METHODS: We included 1,627,679 pregnant women with singleton births (2013-2022) from the California birth certificate database. We estimated age-standardized prevalence, annual percent change (APC), and prevalence ratios (PRs) for SDOH associated with PDM and GDM. RESULTS: From 2013 to 2022, PDM prevalence increased 1.8-fold (0.4% to 0.7%; APC = 7.7%) and GDM increased 1.5-fold (4.8% to 7.2%; APC = 5.1%). Asian and Hawaiian/Pacific Islander experienced the largest APC for both PDM (10.6%) and GDM (6.4%). Hawaiian/Pacific Islander (2.1%) and Hmong (1.3%) women had the highest PDM prevalence, while Hmong (12.1%) and Asian Indian (10.5%) women had the highest GDM. Lower education level was associated with higher PDM (PR = 1.76 [1.67, 1.85]) and GDM (1.15 [1.13, 1.17]). Foreign-born was associated with lower PDM (0.91 [0.87, 0.96]), but higher GDM (1.24 [1.22, 1.25]), with large variations across subgroups. Prenatal Medicaid vs. private insurance was associated with higher GDM, excepting Korean and Vietnamese. CONCLUSIONS: PDM and GDM prevalence increased across all racial/ethnic groups in California, with pronounced heterogeneous SDOH associations across disaggregated subgroups.
OBJECTIVE: Despite racial/ethnic disparities in prostate cancer (PCa) outcomes, evidence on screening-specific outcomes across diverse populations remains limited. This study examined racial/ethnic differences in PCa scr...OBJECTIVE: Despite racial/ethnic disparities in prostate cancer (PCa) outcomes, evidence on screening-specific outcomes across diverse populations remains limited. This study examined racial/ethnic differences in PCa screening, tumor characteristics, and mortality, among White, Black, Hispanic, and Asian men Methods: Secondary analysis of the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (1993-2022) in the United States (N = 73,158). Chi-square tests compared screening outcomes, tumor characteristics, and mortality across racial/ethnic groups. Adjusted rate ratios (aRRs) were estimated for incidence through 2017, and mortality through 2022, and survival was assessed using Kaplan-Meier and Cox models. RESULTS: Screening outcomes differed by race and ethnicity. Black men had elevated false-positive prostate-specific antigen rates, biopsy rates, and prostate cancer-specific mortality (aRR = 1.97, 95% CI = 1.53, 2.53 vs. White). Asian men had the highest rate of high-grade tumors (p < 0.01) despite the lowest mortality. However, these mortality differences for Black and Asian men relative to White men were not observed in 2013-2022. Hispanic men had false-positive patterns similar to White men but lower biopsy conversion. CONCLUSIONS: PCa screening disparities reflected distinct risk profiles and tumor biology across groups. While mortality gaps attenuated over time, persistent differences underscore the need for population-specific screening strategies and research.
OBJECTIVE: Evidence regarding whether first responders have elevated suicide rates is mixed. We investigated if career firefighters and emergency medical service providers (EMS) had increased suicide mortality compared w...OBJECTIVE: Evidence regarding whether first responders have elevated suicide rates is mixed. We investigated if career firefighters and emergency medical service providers (EMS) had increased suicide mortality compared with general populations. METHODS: We included two occupational cohorts of responders employed by United States urban fire departments: 32,925 FDNY firefighters and EMS employed between 9/11/2001-12/31/2023, and 11,917 non-FDNY responders employed between 9/11/2001-12/31/2009. Follow-up started at the later of hire date or 9/11/2001 and ended at earliest of death date or 12/31/2023. We obtained death data from the National Death Index. Demographic-specific US and NYC suicide rates were used to calculate standardized mortality ratios (SMRs) and 95% CIs. Multivariable extended Cox models with time-dependent variables assessed associations between select characteristics and suicide in FDNY responders. RESULTS: FDNY firefighters had lower suicide rates than US (SMR = 0.38, 95% CI = 0.27, 0.51) and NYC adults (SMR = 0.57, 95% CI = 0.41, 0.77). FDNY EMS suicide rates and non-FDNY responders' rates were not significantly different from general population rates. Suicide risk factors included posttraumatic stress disorder (hazard ratio [HR] = 5.44, 95% CI = 2.60, 11.40) and depression symptoms (HR = 5.41, 95% CI = 2.72, 10.77). CONCLUSIONS: Among nearly 45,000 career first responders, suicide rates were not elevated compared with general populations.
Canada has been experiencing a public health crisis of high levels of overdose deaths mainly from toxic, synthetic opioids (e.g., fentanyl) drugs for more than a decade. While a wide range of prevention and treatment int...Canada has been experiencing a public health crisis of high levels of overdose deaths mainly from toxic, synthetic opioids (e.g., fentanyl) drugs for more than a decade. While a wide range of prevention and treatment interventions to reduce overdose risk have been scaled up, drug overdose death levels have overall increased - reflecting similar US-based trends - before abruptly decreasing since 2023. These developments raise the crucial question: What factors have been driving the recent decreases in drug overdose deaths? We considered North America-based studies/data that examined potential supply-related factors. Emerging data suggest that while drug user-oriented prevention and treatment interventions have averted substantial numbers of overdose deaths, recent decreases in deaths appear to be associated mainly with changes in the global supply dynamics of toxic drugs, possibly related to intensified source control efforts (e.g., on fentanyl precursors) in China. Available data plausibly suggest that these changes may have affected both the composition and/or potency of illicit synthetic opioids available for consumption, and thereby reduced the population-based risks for and burden of overdose fatalities from exposure. While requiring further examination, these developments raise important questions about effective health-oriented overdose prevention measures and policy that are independent of unpredictable global drug supply dynamics.
OBJECTIVE: To generate precise estimates of excess lifetime medical expenditures attributable to type 2 diabetes mellitus (T2DM) by leveraging statewide administrative, health, and insurance records. METHODS: Data were d...OBJECTIVE: To generate precise estimates of excess lifetime medical expenditures attributable to type 2 diabetes mellitus (T2DM) by leveraging statewide administrative, health, and insurance records. METHODS: Data were drawn from the Utah Population Database and included administrative and health records prior to 2015 and insurance records in 2014. Study population consisted of individuals born in 1970-1990 and their parents, who resided in four urban counties in Utah, United States, and had health insurance in 2014. Logistic and generalized linear regression models were used to estimate annual T2DM-attributable medical expenditures. Lifetime T2DM-attributable medical expenditures overall and by sex were calculated using population-specific age- and sex-adjusted survival probabilities. RESULTS: Our study cohort comprised 430,171 individuals, of whom 11% had a T2DM diagnosis. Expenditures declined with later age at T2DM diagnosis, with especially sharp reductions during early adulthood. Women incurred higher healthcare costs when diagnosed prior to age 40 compared with men, but lower costs when diagnosed between ages 40-59. CONCLUSION: Lifetime T2DM-attributable medical expenditures declined rapidly with later diagnosis age, with patterns varying by sex. Our study demonstrates the value of using administrative and insurance claims data to improve precision in assessing the economic burden associated with T2DM across the life course.
OBJECTIVE: The selection of accelerometer processing methods may influence the shape of the dose-response association between wearable-measured physical activity and health outcomes. We aimed to compare the association o...OBJECTIVE: The selection of accelerometer processing methods may influence the shape of the dose-response association between wearable-measured physical activity and health outcomes. We aimed to compare the association of stroke and myocardial infarction with Moderate-Vigorous Physical Activity (MVPA) assessed by three accelerometer-generated metrics: Low-pass Filtered Euclidean Norm Minus One (LFENMO), machine-learning, and activity counts. METHODS: We computed MVPA durations in the UK Biobank accelerometer sub-cohort recruited between 2013 and 2015 in the UK. The outcomes, incident stroke and myocardial infarction, were followed up until December 2022. We used Cox regression and a restricted cubic spline to estimate the dose-response association for each of the three MVPA metrics. RESULTS: There were 90,237 cardiovascular disease-free participants at baseline. We observed 1298 incident strokes and 2031 myocardial infarctions. For stroke, a linear decrease in hazard ratio was observed with machine-learning, but not with LFENMO and activity counts. For myocardial infarction, machine-learning and LFENMO showed a curvilinear decrease in hazard ratios, whereas activity counts showed a linear decrease. CONCLUSIONS: The dose-response associations between MVPA and cardiovascular disease varied markedly across the three accelerometer-derived MVPA metrics. Research using a single accelerometer metric may caution about the interpretation of the association.