Multilevel Models II: Testing the Difference in Group Means in Two-Level Multisite Randomized Trials

This article examines how teachers unions affect teachers’ well-being under various legal institutions. Using a district–teacher matched data set, this study identifies the union effects by three approaches. First, I contrast teacher outcomes across different state laws toward unions. Second, I compare the union–nonunion differentials within the same legal environment, using multilevel models and propensity score matching. Finally, unexpected legal changes restricting the collective bargaining of teachers in four states form a natural experiment, allowing me to use the difference-in-difference estimation to identify the causal effect of weakening unionism on teacher outcomes. I find that (a) many teachers join unions even when bargaining is rarely or never available, and meet-and-confer or union membership rate affects teachers’ lives in the absence of a bargaining contract; (b) how unions influence teacher outcomes vary greatly by different legal environment; and (c) the changes in public policy limiting teachers’ bargaining rights significantly decrease teacher compensation.

Download Full-text

Estimation of the Overall Treatment Effect in the Presence of Interference in Cluster-Randomized Trials of Infectious Disease Prevention

Epidemiologic Methods ◽

10.1515/em-2015-0016 ◽

2016 ◽

Vol 5 (1) ◽

Cited By ~ 4

Author(s):

Nicole Bohme Carnegie ◽

Rui Wang ◽

Victor De Gruttola

Keyword(s):

Infectious Disease ◽

Disease Prevention ◽

Social Interactions ◽

Treatment Effect ◽

Randomized Trials ◽

Effect Estimate ◽

Treatment Effect Estimate ◽

Partial Interference ◽

The Difference ◽

Cluster Randomized

AbstractAn issue that remains challenging in the field of causal inference is how to relax the assumption of no interference between units. Interference occurs when the treatment of one unit can affect the outcome of another, a situation which is likely to arise with outcomes that may depend on social interactions, such as occurrence of infectious disease. Existing methods to accommodate interference largely depend upon an assumption of “partial interference” – interference only within identifiable groups but not among them. There remains a considerable need for development of methods that allow further relaxation of the no-interference assumption. This paper focuses on an estimand that is the difference in the outcome that one would observe if the treatment were provided to all clusters compared to that outcome if treatment were provided to none – referred as the overall treatment effect. In trials of infectious disease prevention, the randomized treatment effect estimate will be attenuated relative to this overall treatment effect if a fraction of the exposures in the treatment clusters come from individuals who are outside these clusters. This source of interference – contacts sufficient for transmission that are with treated clusters – is potentially measurable. In this manuscript, we leverage epidemic models to infer the way in which a given level of interference affects the incidence of infection in clusters. This leads naturally to an estimator of the overall treatment effect that is easily implemented using existing software.

Download Full-text

Perceived sleep need, cumulative sleep deficit, and associations with daily affect in adolescents: A 28-day study over school and vacation periods

10.31234/osf.io/pagzc ◽

2021 ◽

Author(s):

Lin Shen ◽

Joshua F. Wiley ◽

Bei Bei

Keyword(s):

Sleep Duration ◽

Multilevel Models ◽

Weighted Average ◽

Sleep Restriction ◽

Positive And Negative Affect ◽

The Past ◽

Daily Affect ◽

Day Of The Week ◽

The Difference ◽

Sleep Deficit

Study Objectives: To describe trajectories of daily perceived sleep need (PSNeed) and sleep deficit across 28 consecutive days, and examine if cumulative sleep deficit predicts next-day affect.Methods: Daily sleep and affect were measured over 2 weeks of school and 2 weeks of vacation in 205 adolescents (54.1% females, Mage = 16.9 years). Each day, participants wore actigraphs and self-reported the amount of sleep needed to function well the next day (i.e., perceived sleep need), sleep duration, and high- and low-arousal positive and negative affect (PA, NA). Cumulative actigraphy and diary sleep deficit were calculated as the weighted average of the difference between PSNeed and sleep duration over the past 3 days. Cross-lagged, multilevel models were used to test cumulative sleep deficit as a predictor of next-day affect. Lagged affect, day of the week, study day, and sociodemographics were controlled.Results: PSNeed was lower early in the school week, before increasing in the second half of the week. Adolescents accumulated sleep deficit across school days and reduced it during weekends. During weekends and vacations, adolescents’ self-reported, but not actigraphy sleep duration, met PSNeed. Higher cumulative actigraphy sleep deficit predicted higher next-day high arousal NA; higher cumulative diary sleep deficit predicted higher NA (regardless of arousal), and lower low arousal PA the following day.Conclusions: Adolescents experienced sustained cumulative sleep deficit across school days. Non-school nights appeared to be opportunities for reducing sleep deficit. Trajectories of sleep deficit during vacation suggested recovery from school-related sleep restriction.

Download Full-text

Daily Versus Weekly Levothyroxine Tablet Replacement in Adults With Hypothyroidism: A Meta-Analysis

Journal of the Endocrine Society ◽

10.1210/jendso/bvab048.1686 ◽

2021 ◽

Vol 5 (Supplement_1) ◽

pp. A827-A827

Author(s):

Harold Henrison Chang Chiu ◽

Ramon Bagaporo Larrazabal ◽

Angelique Bea Cariaga Uy ◽

Cecilia A Jimeno

Keyword(s):

Adverse Events ◽

Randomized Trials ◽

Meta Analysis ◽

Hormone Replacement ◽

Hormone Deficiency ◽

Significant Heterogeneity ◽

Suitable Alternative ◽

Significant Difference ◽

The Difference ◽

Tsh Levels

Abstract Background and Objectives: Hypothyroidism is a common hormone deficiency with a prevalence ranging from 4-5% worldwide. It is a very treatable condition with treatment in the form of thyroid hormone replacement, with an overall excellent prognosis if patients are compliant to regular treatment. Daily levothyroxine (LT4) is the treatment of choice and standard of care, sufficient to restore the thyroid stimulating hormone (TSH) to the normal range. For many patients, daily and lifelong therapy is required, and compliance/adherence then becomes a major issue. In such cases, weekly replacement may be a suitable alternative in terms of improving patient compliance. In this study, we aimed to determine the efficacy and safety of weekly versus daily levothyroxine replacement in patients with hypothyroidism. Methods: Electronic databases were searched, supplemented with manual searches. Two reviewers independently screened the abstracts, reviewed full-text papers, independently critically appraised the quality of included studies and abstracted the data. A meta-analysis was performed using the random-effects model on randomized controlled trials (RCTs) that reported standard doses of daily versus weekly levothyroxine administration in the treatment of hypothyroidism. The primary outcome is the difference in serum TSH levels between daily versus weekly levothyroxine administration, while secondary outcomes included clinical symptoms and adverse events using the hypothyroidism symptom scale. Results: The study included two randomized trials (N = 109) in the primary analysis. The difference in TSH levels was 1.78 mIU/mL higher (95% CI: 1.28, 2.28; P < 0.00001) at 6 weeks and 1.22 mIU/mL higher (95% CI: 0.76,1.67; P < 0.00001) at 12 weeks for the weekly replacement regimen, respectively. There was no significant heterogeneity noted between the two groups. There was no significant difference in terms hyperthyroid symptoms and adverse events measured by the hypothyroid symptom scales and echocardiographic parameters, respectively, before and after LT4 within each group. Conclusions: Our results showed that weekly LT4 administration has less suppression of TSH levels, while still remaining within the reference range of normal. It may be an alternative for patients especially in setting of noncompliance. However, more randomized trials with larger sample sizes and a longer duration of follow-up are needed to firmly establish the definite role of weekly LT4.

Download Full-text

1518Preterm birth and trajectories of cardiometabolic health measures from birth to adulthood

International Journal of Epidemiology ◽

10.1093/ije/dyab168.137 ◽

2021 ◽

Vol 50 (Supplement_1) ◽

Author(s):

Gemma Clayton ◽

Laura Howe ◽

Linda O’Keeffe ◽

Adam Lewandowski ◽

Deborah Lawlor ◽

...

Keyword(s):

Fat Mass ◽

Lean Mass ◽

Multilevel Models ◽

Disease Risk ◽

Cardiovascular Disease Risk ◽

Birth Cohort Study ◽

Cardiometabolic Health ◽

Health Measures ◽

The Difference

Abstract Background Preterm birth is associated with increased cardiovascular disease risk, but it is unclear when this risk emerges. We compared life course trajectories from birth to adulthood of multiple cardiometabolic health measures in adults born preterm compared with term. Methods Multilevel models were used to compare trajectories from birth to age 25 years of BMI, fat mass, lean mass, systolic and diastolic blood pressure (BP), lipids, glucose and insulin, between participants born preterm (N = 311-560) and term (N = 5365-9262) in a contemporary UK birth cohort study. We also investigated gestational age as a continuum. Results Systolic and diastolic BP were higher at age 7 (mean predicted differences 0.8mmHg; 95%CI -0.1, 1.7 and 0.7mmHg; 95%CI 0.03, 1.3, respectively). The difference in SBP persisted and in DBP disappeared by 25 years. Preterm participants had lower BMI between 7 and 18 years, but by age 25, there was no difference. Lean mass and fat mass (measured from age 9 only) trajectories were consistent with BMI. HDL-c was higher, and triglycerides lower at birth in those born preterm, but this difference disappeared by 25 years. No other differences were found. Conclusions There were few, modest differences in cardiometabolic health measures during early life in those born preterm versus term. All disappeared by age 25, except SBP. Longer follow-up is needed to establish if and when trajectories diverge. Key messages By age 25, there were no differences in measures of cardiometabolic health between people born preterm and term except for a higher SBP in those born preterm.

Download Full-text

Bevacizumab: Analysis of clinical benefit in females across trials in colorectal cancer and non-small cell lung cancer

Journal of Clinical Oncology ◽

10.1200/jco.2007.25.18_suppl.7634 ◽

2007 ◽

Vol 25 (18_suppl) ◽

pp. 7634-7634 ◽

Cited By ~ 2

Author(s):

D. A. Ramies ◽

A. Sandler ◽

R. Gray ◽

B. Giantonio ◽

J. Brahmer ◽

...

Keyword(s):

Clinical Benefit ◽

Randomized Trials ◽

Metastatic Breast ◽

Progression Free Survival ◽

Small Cell ◽

Small Cell Lung ◽

Open Label ◽

Free Survival ◽

Analysis Of Results ◽

The Difference

7634 Background: Bevacizumab (Bv) has demonstrated significant improvement in overall survival (OS) in randomized trials in first- and second-line (FL and SL) metastatic CRC (mCRC), FL NSCLC, and progression-free survival (PFS) in metastatic breast cancer (mBC). However, in FL metastatic non-squamous NSCLC, OS benefit was less robust in females treated with Bv + chemotherapy (CTx). In order to examine whether there is a gender effect upon efficacy in other trials, an analysis of results from randomized ph II and open-label ph III trials with Bv in mCRC (and a randomized ph II trial in SL NSCLC) was conducted for females. Methods: RR, PFS and OS are summarized for females in the following trials (primary endpoint): 1] ph II FL mCRC (PFS); 2] ph III: FL mCRC (OS); 3] ph III SL mCRC (OS); 4] ph II SL NSCLC (PFS); 5] ph III FL NSCLC (OS). Results: Table 1 presents results of Studies 1 - 5. RR is presented as the difference in RR between Bv and respective control arms. Conclusions: Clinical benefit with Bv + CTx compared to CTx-alone was observed in OS for females in studies which included Bv in treatment of mCRC. OS observed in females in Study 5 (FL NSCLC) is inconsistent with findings in mCRC, mBC and SL NSCLC, and the improvement for PFS and RR for females in Study 5. As such, there is no compelling evidence to suggest lack of benefit with Bv in females. [Table: see text] [Table: see text]

Download Full-text

Are weak or negative clinical recommendations associated with higher geographical variation in utilisation than strong or positive recommendations? Cross-sectional study of 24 healthcare services

BMJ Open ◽

10.1136/bmjopen-2020-044090 ◽

2021 ◽

Vol 11 (5) ◽

pp. e044090

Author(s):

Agne Ulyte ◽

Wenjia Wei ◽

Oliver Gruebner ◽

Caroline Bähler ◽

Beat Brüngger ◽

...

Keyword(s):

Geographical Variation ◽

Multilevel Models ◽

Healthcare Services ◽

Cross Sectional Study ◽

Future Research ◽

Sectional Study ◽

Cross Sectional ◽

Mandatory Health Insurance ◽

The Difference ◽

Unwarranted Variation

ObjectivesWhen research evidence is lacking, patient and provider preferences, expected to vary geographically, might have a stronger role in clinical decisions. We investigated whether the strength or the direction of recommendation is associated with the degree of geographic variation in utilisation.DesignIn this cross-sectional study, we selected 24 services following a comprehensive approach. The strength and direction of recommendations were assessed in duplicate. Multilevel models were used to adjust for demographic and clinical characteristics and estimate unwarranted variation.SettingObservational study of claims to mandatory health insurance in Switzerland in 2014.ParticipantsEnrolees eligible for the 24 healthcare services.Primary outcome measuresThe variances of regional random effects, also expressed as median odds ratios (MOR). Services grouped by strength and direction of recommendations were compared with Welch’s t-test.ResultsThe sizes of the eligible populations ranged from 1992 to 409 960 patients. MOR ranged between 1.13 for aspirin in secondary prevention of myocardial infarction to 1.68 for minor surgical procedures performed in inpatient instead of outpatient settings. Services with weak recommendations had a negligibly higher variance and MOR (difference in means (95% CI) 0.03 (−0.06 to 0.11) and 0.05 (−0.11 to 0.21), respectively) compared with strong recommendations. Services with negative recommendations had a slightly higher variance and MOR (difference in means (95% CI) 0.07 (−0.03 to 0.18) and 0.14 (−0.06 to 0.34), respectively) compared with positive recommendations.ConclusionsIn this exploratory study, the geographical variation in the utilisation of services associated with strong vs weak and negative vs positive recommendations was not substantially different, although the difference was somewhat larger for negative vs positive recommendations. The relationships between the strength or direction of recommendations and the variation may be indirect or modified by other characteristics of services. As initiatives discouraging low-value care are gaining attention worldwide, these findings may inform future research in this area.

Download Full-text