Screen time of toddlers in Paris suburbs: Quantitative and qualitative analysis.

BACKGROUND: Our aim was to describe the exposure of French children aged 12-36 months to screens (time, content, age of first exposure) and to analyze different moderating factors: sociodemographic data, parents' screen time, and other factors (childcare arrangements, language spoken at home, book reading). POPULATION AND METHODS: We conducted an observational, cross-sectional, descriptive and analytical study based on 171 questionnaires from parents of children aged 12-36 months who consulted different hospitals in the Paris region during the summer of 2020. RESULTS: The median screen time was 1 h per day and was essentially television time. The median age of first exposure for children was 12 months. Among the most-watched sites, YouTube was in first place. One third of the children chose the content they watched alone, and the majority did so without any parental guidance (66%). Children watched a screen during mealtime every day in 25% of cases, before bedtime in 12.3% of cases, and 8.8% had a screen in their bedroom. More than one third of families left the television on in the background most of the time. In the multivariate analysis, a high level of screen time was notably linked to the child's age, the parents' screen time, and background television. However, the parental reason for exposure "to calm the child" was the most strongly correlated factor with significant child screen time. Reading books appeared to be a determining factor for less screen exposure. CONCLUSION: These results emphasize the importance of raising parents' awareness about the potential negative effects of screen exposure (particularly on children's cognitive and emotional development) as early as possible during the maternity period. Implementing this prevention in the maternity wards could be an effective way of informing and educating parents about the potential negative effects of screen time on their child's development.

Prolonged response after TPO-RA discontinuation in primary ITP: results of a prospective multicenter study.

Sustained response off treatment (SROT) after thrombopoietin receptor agonist (TPO-RA) discontinuation has been reported in immune thrombocytopenia (ITP). This prospective multicenter interventional study enrolled adults with persistent or chronic primary ITP and complete response (CR) on TPO-RAs. The primary end point was the proportion of patients achieving SROT (platelet count >30 × 109/L and no bleeding) at week 24 (W24) with no other ITP-specific medications. Secondary end points included the proportion of sustained CR off-treatment (SCROT, platelet count >100 × 109/L and no bleeding) and SROT at W52, bleeding events, and pattern of response to a new course of TPO-RAs. We included 48 patients with a median age of 58.5 years; 30 of 48 had chronic ITP at TPO-RA initiation. In the intention-to-treat analysis, 27 of 48 achieved SROT, 15 of 48 achieved SCROT at W24; 25 of 48 achieved SROT, and 14 of 48 achieved SCROT at W52. No severe bleeding episode occurred in patients who relapsed. Among patients rechallenged with TPO-RA, 11 of 12 achieved CR. We found no significant clinical predictors of SROT at W24. Single-cell RNA sequencing revealed enrichment of a tumor necrosis factor α signaling via NF-κB signature in CD8+ T cells of patients with no sustained response after TPO-RA discontinuation, which was further confirmed by a significant overexpression of CD69 on CD8+ T cells at baseline in these patients as compared with those achieving SCROT/SROT. Our results strongly support a strategy based on progressive tapering and discontinuation of TPO-RAs for patients with chronic ITP who achieved a stable CR on treatment. This trial was registered at www.clinicaltrials.gov as #NCT03119974.

The Relative Contributions of Occupational and Community Risk Factors for COVID-19 among Hospital Workers: The HOP-COVID Cohort Study.

The relative contributions of occupational and community sources of COVID-19 among health-care workers (HCWs) are still subject to debate. In a cohort study at a 2814-bed tertiary medical center (five hospitals) in the Paris area of France, we assessed the proportion of hospital-acquired cases among staff and identified risk factors. Between May 2020 and June 2021, HCWs were invited to complete a questionnaire on their COVID-19 risk factors. RT-PCR and serology test results were retrieved from the virology department. Mixed-effects logistic regression was used to account for clustering by hospital. The prevalence of COVID-19 was 15.6% (n = 213/1369 respondents) overall, 29.7% in the geriatric hospitals, and 56.8% of the infections were hospital-acquired. On multivariable analyses adjusted for COVID-19 incidence and contact in the community, a significantly higher risk was identified for staff providing patient care (especially nursing assistants), staff from radiology/functional assessment units and stretcher services, and staff working on wards with COVID-19 clusters among patients or HCWs. The likelihood of infection was greater in geriatric wards than in intensive care units. The presence of significant occupational risk factors after adjustment for community exposure is suggestive of a high in-hospital risk and emphasizes the need for stronger preventive measures-especially in geriatric settings. Clinicaltrials.gov NCT04386759.

Respiratory recovery trajectories after severe-to-critical COVID-19: a 1-year prospective multicentre study.

BACKGROUND: Survivors of severe-to-critical coronavirus disease 2019 (COVID-19) may have functional impairment, radiological sequelae and persistent symptoms requiring prolonged follow-up. This pragmatic study aimed to describe their clinical follow-up and determine their respiratory recovery trajectories, and the factors that could influence them and their health-related quality of life. METHODS: Adults hospitalised for severe-to-critical COVID-19 were evaluated at 3 months and up to 12 months post-hospital discharge in this prospective, multicentre, cohort study. RESULTS: Among 485 enrolled participants, 293 (60%) were reassessed at 6 months and 163 (35%) at 12 months; 89 (51%) and 47 (27%) of the 173 participants initially managed with standard oxygen were reassessed at 6 and 12 months, respectively. At 3 months, 34%, 70% and 56% of the participants had a restrictive lung defect, impaired diffusing capacity of the lung for carbon monoxide (D LCO) and significant radiological sequelae, respectively. During extended follow-up, both D LCO and forced vital capacity percentage predicted increased by means of +4 points at 6 months and +6 points at 12 months. Sex, body mass index, chronic respiratory disease, immunosuppression, pneumonia extent or corticosteroid use during acute COVID-19 and prolonged invasive mechanical ventilation (IMV) were associated with D LCO at 3 months, but not its trajectory thereafter. Among 475 (98%) patients with at least one chest computed tomography scan during follow-up, 196 (41%) had significant sequelae on their last images. CONCLUSIONS: Although pulmonary function and radiological abnormalities improved up to 1 year post-acute COVID-19, high percentages of severe-to-critical disease survivors, including a notable proportion of those managed with standard oxygen, had significant lung sequelae and residual symptoms justifying prolonged follow-up.

Immune thrombocytopenia and pregnancy: an exposed/nonexposed cohort study.

The risk of immune thrombocytopenia (ITP) worsening during pregnancy and neonatal ITP (NITP) have never been prospectively studied. We included 180 pregnant and 168 nonpregnant women with ITP in a prospective, multicenter, observational cohort study. A total of 131 pregnant women with ITP were matched to 131 nonpregnant women with ITP by history of splenectomy, ITP status (no response, response, complete response), and duration. Groups were followed for 15 months. The primary outcome was the first occurrence of ITP worsening defined by a composite end point including bleeding events and/or severe thrombocytopenia (<30 × 109/L) and/or ITP treatment modification. We also studied the recurrence of ITP worsening and the incidence of NITP and risk factors. The first occurrence of ITP worsening did not differ between pregnant and nonpregnant women with ITP (53.4 per 100 person-years [95% confidence interval {CI}, 40.8-69.9] vs 37.1 [95% CI, 27.5-50.0]; hazard ratio {HR}, 1.35 [95% CI, 0.89-2.03], P = .16). Pregnant women with ITP were more likely to have recurrence of severe thrombocytopenia and treatment modification (HR, 2.71 [95% CI, 1.41-5.23], P = .003; HR, 2.01 [95% CI, 1.14-3.57], P = .017, respectively). However, recurrence of severe bleeding events was not different between groups (P = .4). Nineteen (14%) neonates showed NITP <50 × 109/L. By multivariable analysis, NITP was associated with a previous offspring with NITP and maternal platelet count <50 × 109/L within 3 months before delivery (adjusted odds ratio, 5.55 [95% CI, 1.72-17.89], P = .004 and 4.07 [95% CI, 1.41-11.73], P = .009). To conclude, women with ITP do not increase their risk of severe bleeding during pregnancy. NITP is associated with NITP history and the severity of maternal ITP during pregnancy. These results will be useful for counseling women with ITP.

Prescription of benzodiazepines and Z-drugs among older patients in primary care: a French, national, cohort study.

BACKGROUND: In France, general practitioners (GPs) prescribe benzodiazepines and Z-drugs (BZD/ZDs) widely, and especially to older adults. Several characteristics of patients and/or GPs linked to BZD/ZD overprescription have been described in the general population but not among older patients in primary care. OBJECTIVES: To estimate the proportion of GP consultations by patients aged 65 and over that resulted in a BZD/ZD prescription, and determine whether any GP-related factors predicted BZD/ZD overprescription in this setting. METHODS: We analyzed sociodemographic and practice-related GP characteristics, and aggregated data on consultations recorded prospectively by 117 GPs in a database between 2000 and 2010. Next, we used logistic regression models to look for factors potentially associated with BZD/ZD overprescription (defined as an above-median prescription rate). RESULTS: The GPs' mean age at inclusion was 47.4 (7.1), and 87.9% were male. During the study period, the median (95% confidence interval) proportion of consultations with patients aged 65 and over resulting in a BZD/ZD prescription was 21.8% (18.1-26.1) (range per GP: 5-34.1%). In a multivariable analysis, a greater number of chronic disease (OR [95% CI] = 2.10 [1.22-3.64]), a greater number of drugs prescribed per consultation (5.29 [2.72-10.28]), and shorter study participation were independently associated with BZD/ZD overprescription. CONCLUSIONS: BZD/ZD overprescription was associated with a greater chronic disease burden and the number of drugs prescribed per consultation but not with any sociodemographic or practice-related GP characteristics. Targeted actions are needed to help GPs limit their prescription of BZD/ZDs to older patients with multiple comorbidities and polypharmacy.

In France, general practitioners (GPs) prescribe benzodiazepines and Z-drugs (BZD/ZDs) widely, and especially to older adults. Even though BZD/ZDs may not have a favorable risk­benefit ratio in older patients, we lack data on GP-related factors that might influence BZD/ZD overprescription in our population. The objectives of the present study were to (i) estimate the proportion of GP consultations by patients aged 65 and over that resulted in a BZD/ZD prescription and (ii) identify GP-related factors that were predictive of overprescription. To achieve this goal, we analyzed consultation notes registered by 117 GPs in a database curated by the French Society of General Practice between 2000 and 2010. About 22% of consultations by patients aged 65 and over resulted in a BZD/ZD prescription. With regard to the GPs, we did not find any sociodemographic or practice-related characteristics associated with overprescription. A greater chronic disease burden and the number of drug prescriptions (other than BZD/ZDs) per consultation was independently associated with overprescription. Targeted actions are therefore needed to help GPs limit their prescription of BZD/ZDs in older patients with multimorbidity and polypharmacy.

Menstrual Toxic Shock Syndrome: A French Nationwide Multicenter Retrospective Study.

BACKGROUND: Studies describing the clinical features and short-term prognosis of patients admitted to the intensive care unit (ICU) for menstrual toxic shock syndrome (m-TSS) are lacking. METHODS: This was a multicenter retrospective cohort study of patients with a clinical diagnosis of m-TSS admitted between 1 January 2005 and 31 December 2020 in 43 French pediatric (n = 7) or adult (n = 36) ICUs. The aim of the study was to describe the clinical features and short-term prognosis, as well as to assess the 2011 Centers for Disease and Control (CDC) diagnostic criteria, in critically ill patients with m-TSS. RESULTS: In total, 102 patients with m-TSS (median age, 18 years; interquartile range, 16-24 years) were admitted to 1 of the participating ICUs. All blood cultures (n = 102) were sterile. Methicillin-sensitive Staphylococcus aureus grew from 92 of 96 vaginal samples. Screening for superantigenic toxin gene sequences was performed for 76 of the 92 vaginal samples positive for S. aureus (83%), and toxic shock syndrome toxin 1 was isolated from 66 strains (87%). At ICU admission, no patient met the 2011 CDC criteria for confirmed m-TSS, and only 53 (52%) fulfilled the criteria for probable m-TSS. Eighty-one patients (79%) were treated with antitoxin antibiotic therapy, and 8 (8%) received intravenous immunoglobulins. Eighty-six (84%) patients required vasopressors, and 21 (21%) tracheal intubation. No patient required limb amputation or died in the ICU. CONCLUSIONS: In this large multicenter series of patients included in ICUs for m-TSS, none died or required limb amputation. The CDC criteria should not be used for the clinical diagnosis of m-TSS at ICU admission.

Impact of pre-eclampsia on renal outcome in sickle cell disease patients.

The long-term consequences of pre-eclampsia (PrE) for renal function have never been determined in patients with sickle cell disease (SCD). Between 2008 and 2015, we screened 306 pregnancies in women with SCD and identified 40 with PrE (13%). The control group consisted of 65 pregnant SCD patients without PrE. In multivariable analysis, PrE events were associated with an increase of 1 log of lactate dehydrogenase level (adjusted odds ratio, aOR = 3·83, P = 0·05), a decrease of 10 g/l of haemoglobin levels (aOR = 2·48, P = 0·006) and one or more vaso-occlusive crisis during pregnancy (aOR = 16·68, P = 0·002). Estimated glomerular filtration rate (eGFR) was similar in the two groups at steady state but was significantly lower in the PrE group after one year of follow-up and at last follow-up (130 vs 148 ml/min/1·73 m2 , P < 0·001 and 120 vs 130 ml/min/1·73 m2 , P < 0·001, respectively). In multivariable analysis, eGFR had returned to steady-state levels one year after pregnancy in patients without PrE but continued to decrease in patients with PrE (ß = -18·15 ml/min/1·73 m2 , P < 0·001). This decline was more marked at the end of follow-up (ß = -31·15 ml/min, P < 0·001). In conclusion, PrE episodes are associated with a significant risk of subsequent renal function decline in SCD patients.

Inappropriate direct oral anticoagulant prescriptions in patients with non-valvular atrial fibrillation: cross-sectional analysis of the French CACAO cohort study in primary care.

BACKGROUND: Direct oral anticoagulants (DOACs) account for an increasing proportion of prescriptions in patients with non-valvular atrial fibrillation (NVAF) in primary care. Inappropriate dosing of DOACs is a common problem, with under-dosing being a particular issue. However, conflicting results have been reported about the factors independently associated with inappropriate dosing. AIM: To describe inappropriate prescriptions of DOACs among patients in the CACAO French nationwide general practice cohort, and to identify the factors independently associated with inappropriate DOAC doses. DESIGN AND SETTING: Cross-sectional baseline analysis of the CACAO French national multicentre prospective cohort of adult patients in primary care receiving an oral anticoagulant who were recruited between April and October 2014. METHOD: A total of 1111 patients from the CACAO cohort who received a DOAC for NVAF were included in this study. Inappropriate prescriptions of DOACs were described (inappropriate dosage, contraindications, non-indications, interactions, and non-compliance with the precautions for use). Multivariate logistic models were used to investigate factors associated with inappropriate DOAC dosing (under-dosing and over-dosing). RESULTS: Overall, 438 patients (39.4%) received at least one inappropriate DOAC prescription. The most common inappropriate prescription was inappropriate dosage (n = 374, 33.7%), particularly under-dosing (n = 348, 31.3%). Multivariate analysis revealed that factors independently associated with under-dosing were older age, prescription of apixaban or dabigatran, and a CHA2DS2-VASc score ≥2 vs. a score = 1. Factors with over-dosing were kidney failure, a HAS-BLED score ≥3, and older age. CONCLUSION: The appropriateness of DOAC prescribing for NVAF can be improved, especially in older patients, and in patients with kidney failure, a higher risk of ischaemic stroke, and/or a higher risk of bleeding. GPs have a key role in increasing the proportion of appropriate DOAC prescriptions via informational, educational, and/or management strategies.

Use of Evidence-Based Practice Among Healthcare Professionals After the Implementation of a New Competency-Based Curriculum.

BACKGROUND: Although evidence-based practice (EBP) has been spreading since the 1990s, it has not yet been sufficiently implemented. AIM: Following the reform of initial training for healthcare professions in France 2012, we sought to determine whether the new curriculum was associated with more frequent use of EBP. METHODS: We performed an online, cross-sectional survey of nurses, occupational therapists, and podiatrists (divided into pre- and post-reform groups) in June 2018. The questionnaire covered demographic data, use of EBP, and the perception of EBP. As holding a master's degree may enhance knowledge and use of EBP, we adjusted for this variable. Categories to analyze qualitative data were created regarding the five steps in EBP and its definition. RESULTS: The total sample was N = 595 (pre-reform group n = 301; post-reform group n = 294). The proportion of respondents who frequently read the professional literature was lower in the post-reform group than in the pre-reform group (33% vs. 54%, respectively; OR [95% CI] = .52 [.37-.73]; p < .001). The main stated reasons for reading the professional literature were "keeping up to date with practice" and "making clinical decisions." Respondents in both groups mentioned a lack of time as the most frequent barrier to reading the literature (82%), a lack of access to bibliographical resources, and that EBP was not encouraged. Most professionals limited their definition of EBP to reading the literature and implementing research results. LINKING EVIDENCE TO ACTION: There is a need to teach the five steps of EBP more explicitly and to embed its position into daily practice, for example, through reflective analysis practice. Professional trainings about EBP should be offered on a regular basis. Guidance coming from the healthcare directorate should include expected daily practice time for reading and journal club and giving more access to international healthcare literature.

Evaluation of Outcomes and Quality of Care in Children with Sickle Cell Disease Diagnosed by Newborn Screening: A Real-World Nation-Wide Study in France.

This study's objective was to assess, on a national scale, residual risks of death, major disease-related events, and quality of care during the first five years in children diagnosed at birth with sickle cell disease (SCD). Data were retrospectively collected from medical files of all children with SCD born between 2006-2010 in France. Out of 1792 eligible subjects, 1620 patients (71.8% SS or S/beta°-thalassemia -SB°-) had available follow-up data, across 69 centers. Overall probability of survival by five years was 98.9%, with 12/18 deaths related to SCD. Probability of overt stroke by five years in SS/SB° patients was 1.1%, while transcranial Doppler (TCD) was performed in 81% before three years of age. A total of 26 patients had meningitis/septicemia (pneumococcal in eight cases). Prophylactic penicillin was started at a median age of 2.2 months and 87% of children had received appropriate conjugate pneumococcal vaccination at one year. By five years, the probability of survival without SCD-related events was 10.7% for SS/SB° patients. In contrast, hydroxyurea was prescribed in 13.7% and bone marrow transplant performed in nine patients only. In this study, residual risks of severe complications were low, probably resulting from a good national TCD, vaccination, and healthcare system coverage. Nonetheless, burden of disease remained high, stressing the need for disease-modifying or curative therapy.

Metabolic Positron Emission Tomography Response to Gamma Knife of the Ventral Intermediate Nucleus in Essential Tremor.

BACKGROUND: To understand the substrates of response and nonresponse and to identify potential biomarkers for the selection and follow-up of patients with essential tremor (ET) treated with Gamma Knife (Elekta AB, Stockholm, Sweden) of the ventral intermediate nucleus (GKVIM). OBJECTIVE: To characterize positron emission tomography (PET) changes in the metabolism of glucose and metabolic connectivity in patients with ET treated by GKVIM through observational study. METHODS: Forty-two patients with right ET were referred to 18F-fluorodesoxyglucose positron emission tomography (18F-FDG PET) imaging before and after left GKVIM. Statistical Parametric Mapping T-score map comparisons were performed between pre- and post-GKVIM groups and between clinical responders and nonresponders. Metabolic connectivity was evaluated by the interregional correlation analysis method. RESULTS: After GKVIM, patients with ET exhibited decreased left thalamic metabolism, which was associated with remote metabolic decreases in the right cerebellum, left temporal gyri, and bilateral frontal gyri (P < .05, family-wise error-corrected). Additionally, nonresponders (n = 7) showed metabolic decreases in the right temporo-occipital area (P < .005 corrected for cluster volume) after GKVIM. The metabolism in this area was already reduced in nonresponders before treatment in comparison to that in responders and was predictive of future response (sensitivity: 89%; specificity: 71%). In nonresponder patients, strong connectivity between the left thalamus and right temporo-occipital area was found before GKVIM and was lost after treatment, whereas this connectivity remained weak and stable in responders. CONCLUSION: These findings could lead to better knowledge of the variability in the metabolic PET profiles among patients with ET, particularly the integration of 18F-FDG PET imaging in the pretherapeutic evaluation of patients with refractory ET candidates for GKVIM.

Belatacept in renal transplant recipient with mild immunologic risk factor: A pilot prospective study (BELACOR).

The benefit of belatacept on antibody-mediated rejection (ABMR) incidence after kidney transplant with preformed donor-specific antibodies (DSAs) has never been assessed. Between 2014 and 2016, we conducted a multicenter prospective clinical trial with 49 patients to determine kidney allograft outcome in recipients with preformed DSAs (maximal mean fluorescence intensity 500 to 3000) treated with belatacept (BELACOR trial). Immunosuppressive strategy included antithymocyte globulin, belatacept, mycophenolate mofetil, and steroids. An ancillary control group was designed retrospectively, including patients fulfilling the same inclusion criteria treated with calcineurin inhibitors. In BELACOR group, no patient exhibited acute ABMR, patient and allograft survival at 1 year was 100% and 95.4%, respectively, and the estimated glomerular filtration rate was 53.2 mL/min/1.73 m2 . However, the 12-month incidence of acute T cell-mediated rejection was 25.4% (14.5% to 42.4%). Comparison with the control group showed significantly higher T cell-mediated rejection incidence only in the BELACOR group (P = .003). Considering the DSAs, the outcome was similar in the 2 groups except a significantly higher number of patients displayed a complete disappearance of class II DSAs in the BELACOR group (P = .001). Belatacept was not associated with an acute ABMR increased risk and may be considered as immunosuppressive strategy in transplant recipients with preformed DSAs (maximal mean fluorescence intensity 500 to 3000). Prospective randomized trials are needed to confirm these results.

Safety of renal transplantation in patients with bipolar or psychotic disorders: a retrospective study.

Solid organ transplantation societies recommend a relative contraindication of transplantation for people with bipolar or psychotic disorders. Very few data are available on the outcome of kidney transplantation and the increased risk of kidney disease in those patients. We conducted a retrospective multicenter cohort study (1979-2014) including kidney allograft recipients with either bipolar (BD) or psychotic disorders prior to transplant. Objectives were kidney allograft and patient outcomes compared to a matched control group without psychiatric disorders and the evolution of psychiatric disorder at 60 months after transplantation. Forty-seven patients including 25 women were identified, 34 with BD and 13 with psychotic disorder. Patients' overall cumulative death rates at 60 months were not significantly different in both groups [12.2%; 95% confidence interval: (4.5-24.1) in the group with psychiatric disorder versus 5.2%; (1.7-11.7) in control group P = 0.11] as for cumulative allograft loss rates [11.7% (3.5-25.2) vs. 9.4% (4.4-16.8) in control group (P = 0.91)]. Twenty-three patients (16 with BD and seven with psychotic disorder) experienced at least one psychiatric relapse [incidence rate: 1.8/100 persons- months; 95% CI; (1.2-2.7)] totaling 13 hospitalizations within 60 months of follow-up. Four patients stopped immunosuppressive therapy leading to allograft loss in three. Our study suggests that patients with BD or psychotic disorders have to be considered for renal transplantation with close psychiatric follow-up after transplant.

Combined Plasma Elevation of CRP, Intestinal-Type Fatty Acid-Binding Protein (I-FABP), and sCD14 Identify Older Patients at High Risk for Health Care-Associated Infections.

Background: We hypothesized that low-grade inflammation was driven by microbial translocation and associated with an increased risk of health care-associated infections (HAIs). Methods: We included 121 patients aged 75 years or over in this prospective cohort study. High-sensitivity C-reactive protein (hs-CRP), I-FABP, and sCD14-as markers for low-grade inflammation, intestinal epithelial barrier integrity, and monocyte activation, respectively-were measured at admission. Results: HAIs occurred during hospitalization in 62 (51%) patients. Elevated hs-CRP (≥6.02 mg/L, ie, the median) was associated with a significantly higher HAI risk when I-FABP was in the highest quartile (odds ratio [OR], 4; 95% confidence interval [95% CI], 1.39-11.49; p = .010). In patients with hs-CRP elevation and highest-quartile I-FABP, sCD14 elevation (≥0.65 µg/mL, ie, the median) was associated with an 11-fold higher HAI risk (OR, 10.8; 95% CI, 2.28-51.1; p = .003). Multivariate analyses adjusted for invasive procedures and comorbidities did not change the associations linking the three markers to the HAI risk. Conclusion: Increased levels of hs-CRP, I-FABP, and sCD14 may reflect loss of intestinal epithelial barrier integrity with microbial translocation leading to monocyte activation and low-grade inflammation. In our cohort, these markers identified patients at high risk for HAIs.

Interrelations between body mass index, frailty, and clinical adverse events in older community-dwelling women: The EPIDOS cohort study.

BACKGROUND: The hypothesis of reverse epidemiology holds that, obesity may reduce the risk of clinical adverse events in older subjects. However, this association is controversial and rarely explored according to the underlying health status. We tested this phenomenon by assessing the association between body mass index (BMI) and clinical adverse events in community dwelling older women according to their frailty status. METHODS: EPIDOS is a multicenter prospective cohort of community-dwelling women aged 75 and older recruited between 1992 and 1994. At baseline, we collected demographics, BMI (<21 kg/m2: underweight; 21-24.9: normal weight; 25-29.9: overweight and ≥30: obesity), frailty through Fried model, and clinical characteristics. All-cause mortality, falls, hip fractures, and hospital admission were collected within 5 years of follow-up and were analyzed using univariate and multivariate survival analysis by using Kaplan-Meier methods and Cox Hazard Proportional models. RESULTS: Of 6662 women (mean age, 80.4 years), 11.6%; 95% Confidence Interval (95% CI) CI [10.8%-12.3%] were frail. By multivariate analysis, the risk of death in frail women (compared to not-frail normal weight women) decreases with increase of BMI: adjusted Hazard Ratio (aHR)frail-underweight = 2.04 [1.23-3.39]; aHRfrail-normal weight = 3.07 [2.21-4.26]; aHRfrail-overweight = 1.83 [1.31-2.56]; aHRfrail-obese = 1.76 [1.15-2.70]; p < 0.001. Frail overweight and obese women had a significant lower risk of death than frail normal-weight women (p = 0.004). Similar features were found for fall risk and hip fracture and for not-frail women. The relative risks of hospital admission for normal weight, overweight and obese frail women were similar (aHRfrail-normal weight = 1.50 [1.22-1.84], aHR frail-overweight =1.48 [1.26-1.74] and aHR frail-obese =1.53 [1.24-1.89], respectively). CONCLUSION: Our results suggest that overweight and obesity reduce the risks of clinical adverse events in frail community-dwelling older women and that frailty definition through Fried model had to be re-calibrated for overweight and obese individuals.