Disparities in Physical Therapy Outcomes Based on Race and Ethnicity: A Scoping Review.

OBJECTIVE: This scoping review synthesizes and summarizes the evidence on racial and ethnic disparities in outcomes after physical therapist treatment. METHODS: Four databases from 2001 through 2021 were searched for articles reporting physical therapy outcomes across racial and ethnic groups. The Arksey and O'Malley's methodological framework was adapted for this scoping review. Two reviewers screened the abstracts and 5 reviewers screened full texts for inclusion. Five reviewers extracted information including study design, diagnoses, setting, outcomes reported, the domains the outcomes measured, and racial and ethnic groups included. To identify disparities, summarized differences in outcomes (better, worse, no difference) for each racial and ethnic group compared to White patients were calculated. RESULTS: Of 1511 abstracts screened, 65 met inclusion criteria, 57 of which were observational designs. All 65 articles included non-Hispanic White patients as the reference group. A majority of the physical therapy outcomes reported by race were for Black patients and/or Hispanic or Latino patients, whereas outcomes for Asian, American Indian, Alaskan Native, and/or Native Hawaiians or Pacific Islander patients were reported infrequently. Most articles reported disparities in health outcomes for patients in the inpatient rehabilitation setting (n = 48) and for adults (n = 59) with neurologic diagnoses (n = 36). Compared to White patients, worse outcomes were reported more frequently for all marginalized racial and ethnic groups after physical therapy, with the exception of marginalized groups having the same or better outcomes for successful post-rehabilitation community discharge. CONCLUSION: Gaps remain in understanding outcome disparities beyond older adult and neurologic populations as well as for musculoskeletal diagnoses frequently treated by physical therapists. IMPACT: The presence of racial and ethnic disparities in physical therapy outcomes should motivate physical therapists to understand the mechanisms underlying disparities and focus on social and structural drivers of health inequity in their clinical decision-making.

Diagnostic rate estimation from Medicare records: Dependence on claim numbers and latent clinical features.

OBJECTIVE: International Classification of Disorders version 10 (ICD-10) codes contribute heavily to healthcare data. Medicare claims and other data-sources are used to constitute study populations and appraise healthcare processes. How variability in claims-per-beneficiary impacts diagnostic determinations is inadequately understood. The objective of this study is so assess distributional properties of Medicare claims, and examine claim rates impact on code utilization and rate determinations. METHODS: The study population was Medicare beneficiaries aged 75-79.99 with claim(s) in the 5% standard analytical Carrier and Outpatient files, alive and participating in Medicare part B for all 12 months of 2017. Medicare beneficiary files were processed to create records containing all ICD-10 codes specified, key demographics, Part B and vital status, and the total claims for each 2017 beneficiary. Claim number cohorts were characterized. RESULTS: Beneficiaries meeting inclusion criteria totaled 221,625, these having 7,617,503 claims; 96.4% had between 1 and 120 claims. Median claims were 24 for males (females 25); modal claims were 11 (13). Average distinct codes per beneficiary increased with claims number. The assignment of ICD-10 codes, i.e., 'diagnostic rate estimates' (DRE), increased as claim numbers increased for most codes among those most commonly utilized. For some conditions, mostly benign and age-related, DREs plateaued as claim numbers increased. For other conditions, typically associated with clinical acuity, e.g., chest pain, DREs increased steeply with claims. CONCLUSIONS: Older adult Medicare beneficiaries aged 75-80 exhibited varying claims activity over the course of a year. Although DRE dependence on claim numbers varies across ICD-10 codes, rate estimates are higher for beneficiaries with claim numbers above the median.

Issues in Implementing Regression Calibration Analyses.

Regression calibration is a popular approach for correcting biases in estimated regression parameters when exposure variables are measured with error. This approach involves building a calibration equation to estimate the value of the unknown true exposure given the error-prone measurement and other covariates. The estimated, or calibrated, exposure is then substituted for the unknown true exposure in the health outcome regression model. When used properly, regression calibration can greatly reduce the bias induced by exposure measurement error. Here, we first provide an overview of the statistical framework for regression calibration, specifically discussing how a special type of error, called Berkson error, arises in the estimated exposure. We then present practical issues to consider when applying regression calibration, including: 1) how to develop the calibration equation and which covariates to include; 2) valid ways to calculate standard errors of estimated regression coefficients; and 3) problems arising if one of the covariates in the calibration model is a mediator of the relationship between the exposure and outcome. Throughout, we provide illustrative examples using data from the Hispanic Community Health Study/Study of Latinos (United States, 2008-2011) and simulations. We conclude with recommendations for how to perform regression calibration.

Synoptic review on existing and potential sources for bias in dental research methodology with methods on their prevention and remedies.

The results of years of dental study serve as the foundation for the practise of medicine and, for that matter, dentistry. Doctors may have their own preferences for techniques and materials, but whether directly or indirectly, their decisions are influenced by systematic reviews and meta-analyses. However, due to poorly conducted or presented research, this very basic foundation may not be reliable. Bias in research is one of several factors that might make study results or research itself unreliable. Bias can be introduced into research at many stages, deliberately or unknowingly. Bias can appear at any point during the research process, even before the study itself begins. There are many biases in research, but some of them are more relevant to dentistry research than others. Because it is said that "eyes see what the mind knows", it is essential to have a complete understanding of the different types of bias, how and when they get entrenched, and what steps may be taken to prevent or lessen them if they do occur. This comprehensive summary of bias in dentistry research is provided by this synoptic review. The goal is to identify gaps and measures that have been taken-or that should have been taken-by providing both descriptive and evaluative summaries, as well as examples from the literature, when needed.

Metformin use is not associated with colorectal cancer incidence in type-2 diabetes patients: evidence from methods that avoid immortal time bias.

PURPOSE: Immortal time bias (ITB) continues to distort many observational studies on metformin use and cancer risk. Our objective was to employ three statistical methods proven to avoid ITB and compare their results to that of a naïve time-fixed analysis in order to provide further evidence of metformin's association, or none thereof, with colorectal cancer (CRC) incidence. METHODS: A total of 41,533 Korean subjects with newly diagnosed type-2 diabetes in 2005-2015 were selected from a prospectively maintained cohort (median follow-up of 6.3 years). Time-to-CRC incidence was regressed upon metformin use (yes/no, average prescription days/year) using time-dependent Cox, landmark, nested case-control, and time-fixed Cox analyses. Other CRC risk factors were included to adjust for possible confounding. RESULTS: Neither metformin ever-use nor average metformin prescription days/year was associated with incident CRC hazard in time-dependent Cox, landmark, and nested case-control analyses with HR (95% CI) of 0.88 (0.68-1.13), 0.86 (0.65-1.12), and 1.10 (0.86-1.40) for metformin ever-use, and 0.97 (0.90-1.04), 0.95 (0.88-1.04), and 1.02 (0.95-1.10) for average metformin prescription days/year, respectively. In contrast, time-fixed Cox regression showed a falsely exaggerated protective effect of metformin on CRC incidence. CONCLUSION: The association between metformin use and subsequent CRC incidence was statistically nonsignificant after accounting for time-related biases such as ITB. Previous studies that avoided these biases and meta-analyses of RCTs on metformin and cancer incidence were in agreement with our results. A definitive, large-scale RCT is needed to clarify this topic, and future observational studies should be explicit in avoiding ITB and other time-related biases.

Misconceptions About the Direction of Bias From Nondifferential Misclassification.

Measurement error is pervasive in epidemiologic research. Epidemiologists often assume that mismeasurement of study variables is nondifferential with respect to other analytical variables and then rely on the heuristic that "nondifferential misclassification will bias estimates towards the null." However, there are many exceptions to the heuristic for which bias towards the null cannot be assumed. In this paper, we compile and characterize 7 exceptions to this rule and encourage analysts to take a more critical and nuanced approach to evaluating and discussing bias from nondifferential mismeasurement.

Immortal time bias for life-long conditions in retrospective observational studies using electronic health records.

BACKGROUND: Immortal time bias is common in observational studies but is typically described for pharmacoepidemiology studies where there is a delay between cohort entry and treatment initiation. METHODS: This study used the Clinical Practice Research Datalink (CPRD) and linked national mortality data in England from 2000 to 2019 to investigate immortal time bias for a specific life-long condition, intellectual disability. Life expectancy (Chiang's abridged life table approach) was compared for 33,867 exposed and 980,586 unexposed individuals aged 10+ years using five methods: (1) treating immortal time as observation time; (2) excluding time before date of first exposure diagnosis; (3) matching cohort entry to first exposure diagnosis; (4) excluding time before proxy date of inputting first exposure diagnosis (by the physician); and (5) treating exposure as a time-dependent measure. RESULTS: When not considered in the design or analysis (Method 1), immortal time bias led to disproportionately high life expectancy for the exposed population during the first calendar period (additional years expected to live: 2000-2004: 65.6 [95% CI: 63.6,67.6]) compared to the later calendar periods (2005-2009: 59.9 [58.8,60.9]; 2010-2014: 58.0 [57.1,58.9]; 2015-2019: 58.2 [56.8,59.7]). Date of entry of diagnosis (Method 4) was unreliable in this CPRD cohort. The final methods (Method 2, 3 and 5) appeared to solve the main theoretical problem but residual bias may have remained. CONCLUSIONS: We conclude that immortal time bias is a significant issue for studies of life-long conditions that use electronic health record data and requires careful consideration of how clinical diagnoses are entered onto electronic health record systems.

Influence of Sponsorship Bias on Treatment Effect Size Estimates in Randomized Trials of Oral Health Interventions: A Meta-epidemiological Study.

BACKGROUND: In this meta-epidemiological study, we aimed to examine associations between treatment effect size estimates and sponsorship bias in oral health randomized clinical trials. METHODS: We selected oral health related meta-analyses that included a minimum of five randomized controlled trials. We extracted data, in duplicate, related to influence of sponsorship bias. We quantified the extent of bias associated with influence of sponsorship on the magnitude of effect size estimates of continuous variables using a two-level meta-meta-analytic approach with random-effects models to allow for intra- and inter-meta-analysis heterogeneity. RESULTS: We initially identified 540 randomized trials included in 64 meta-analyses. Risk of sponsorship bias was judged as being "unclear" in 72.8% (n = 393) of the trials, while it was assessed as "low" in 16.7% (n = 90) and as "high" in 10.6% (n = 57) of the trials. Using a meta-epidemiological analysis (37 meta-analyses, including 328 trials that analyzed 85,934 patients), we identified statistically significant larger treatment effect size estimates in trials that had "high or unclear" risk of sponsorship bias (difference in treatment effect size estimates=0.10; 95% confidence intervals: 0.02 to 0.19) than in trials that had "low" risk of sponsorship bias. CONCLUSIONS: We identified significant differences in treatment effect size estimates between dental trials based on sponsorship bias. Treatment effect size estimates were 0.10 larger in trials with "high or unclear" risk of sponsorship bias. PRACTICAL IMPLICATIONS: Clinicians should have an adequate knowledge of sponsorship bias in a clinical trial and be able to estimate the degree to which the conclusions of a systematic review are synthesized and interpreted, based on trials with low risk of sponsorship bias.

The Validity of Danish Prescription Data to Measure Use of Aspirin and Other Non-Steroidal Anti-Inflammatory Drugs and Quantification of Bias Due to Non-Prescription Drug Use.

PURPOSE: To evaluate the potential of Danish prescription registries to capture aspirin and non-aspirin non-steroidal anti-inflammatory drug (NSAID) use and to quantitatively evaluate the magnitude of bias from misclassification of true NSAID and aspirin use as apparent non-use in drug outcome studies. PATIENTS AND METHODS: In a population-based cohort study, we retrieved sales statistics for NSAIDs and aspirins based on nationwide data from the Danish Health Data Authority and the Danish National Prescription Registry. We estimated prevalence of recorded and non-recorded NSAID use in the prescription registry and resulting proportions of true NSAID and aspirin use misclassified as apparent non-use from 1999 to 2019 at population and patient levels. RESULTS: The prevalence of true use misclassified as non-use (mainly due to over-the-counter use) peaked at 4.7% in 2012 for NSAIDs overall, 5.5% in 2012 for ibuprofen, and at 5.9% in 2002 for high-dose aspirin. Misclassification of other individual NSAIDs was near null. Misclassification of true low-dose aspirin use as non-use declined during the study period but remained around 1% since 2005. In subgroups of cardiac patients, the highest prevalence of true NSAID use misclassified as non-use was 5.0% in 2002 and 4.3% in 2017. Quantitative bias analyses showed how such misclassification of true NSAID and aspirin use as non-use remained minimal both at population and patient levels. In hypothetical examples simulating real study populations with differing exposure prevalence and prevalence of true NSAID and aspirin use misclassified as apparent non-use, the approximate percentage change due to misclassification of use as non-use did not exceed 5% and in most scenarios stayed around 1%. CONCLUSION: The Danish prescription registries are valid data sources for assessing the effects of aspirin and NSAID use. The influence of non-recorded NSAID and aspirin use on estimates of association is virtually negligible.

Propensity Score Weighting and Trimming Strategies for Reducing Variance and Bias of Treatment Effect Estimates: A Simulation Study.

To extend previous simulations on the performance of propensity score (PS) weighting and trimming methods to settings without and with unmeasured confounding, Poisson outcomes, and various strengths of treatment prediction (PS c statistic), we simulated studies with a binary intended treatment T as a function of 4 measured covariates. We mimicked treatment withheld and last-resort treatment by adding 2 "unmeasured" dichotomous factors that directed treatment to change for some patients in both tails of the PS distribution. The number of outcomes Y was simulated as a Poisson function of T and confounders. We estimated the PS as a function of measured covariates and trimmed the tails of the PS distribution using 3 strategies ("Crump," "Stürmer," and "Walker"). After trimming and reestimation, we used alternative PS weights to estimate the treatment effect (rate ratio): inverse probability of treatment weighting, standardized mortality ratio (SMR)-treated, SMR-untreated, the average treatment effect in the overlap population (ATO), matching, and entropy. With no unmeasured confounding, the ATO (123%) and "Crump" trimming (112%) improved relative efficiency compared with untrimmed inverse probability of treatment weighting. With unmeasured confounding, untrimmed estimates were biased irrespective of weighting method, and only Stürmer and Walker trimming consistently reduced bias. In settings where unmeasured confounding (e.g., frailty) may lead physicians to withhold treatment, Stürmer and Walker trimming should be considered before primary analysis.

Preanalytical Sample Handling Conditions and Their Effects on the Human Serum Metabolome in Epidemiologic Studies.

Many epidemiologic studies use metabolomics for discovery-based research. The degree to which sample handling may influence findings, however, is poorly understood. In 2016, serum samples from 13 volunteers from the US Department of Agriculture's Beltsville Human Nutrition Research Center were subjected to different clotting (30 minutes/120 minutes) and refrigeration (0 minutes/24 hours) conditions, as well as different numbers (0/1/4) and temperatures (ice/refrigerator/room temperature) of thaws. The median absolute percent difference (APD) between metabolite levels and correlations between levels across conditions were estimated for 628 metabolites. The potential for handling artifacts to induce false-positive associations was estimated using variable hypothetical scenarios in which 1%-100% of case samples had different handling than control samples. All handling conditions influenced metabolite levels. Across metabolites, the median APD when extending clotting time was 9.08%. When increasing the number of thaws from 0 to 4, the median APD was 10.05% for ice and 5.54% for room temperature. Metabolite levels were correlated highly across conditions (all r's ≥ 0.84), indicating that relative ranks were preserved. However, if handling varied even modestly by case status, our hypotheticals showed that results can be biased and can result in false-positive findings. Sample handling affects levels of metabolites, and special care should be taken to minimize effects. Shorter room-temperature thaws should be preferred over longer ice thaws, and handling should be meticulously matched by case status.

Potential Biases Arising From Epidemic Dynamics in Observational Seroprotection Studies.

The extent and duration of immunity following infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are critical outstanding questions about the epidemiology of this novel virus, and studies are needed to evaluate the effects of serostatus on reinfection. Understanding the potential sources of bias and methods for alleviating biases in these studies is important for informing their design and analysis. Confounding by individual-level risk factors in observational studies like these is relatively well appreciated. Here, we show how geographic structure and the underlying, natural dynamics of epidemics can also induce noncausal associations. We take the approach of simulating serological studies in the context of an uncontrolled or controlled epidemic, under different assumptions about whether prior infection does or does not protect an individual against subsequent infection, and using various designs and analytical approaches to analyze the simulated data. We find that in studies assessing whether seropositivity confers protection against future infection, comparing seropositive persons with seronegative persons with similar time-dependent patterns of exposure to infection by stratifying or matching on geographic location and time of enrollment is essential in order to prevent bias.

Methodological Considerations for Preterm Birth Research.

Complications from preterm birth are a leading cause of infant mortality, with long-term implications for morbidity and quality of life of preterm infants. There are many important risk factors for preterm births however in this article, we focus on the maternal infection etiological pathway, given its significance in low-to-middle income countries. In high preterm birth settings such as sub-Saharan Africa, maternal HIV infection and antiretroviral therapy (ART) use have been associated with an increased risk of preterm births. Consequently, we highlight methodological considerations related to selection and measurement bias in preterm birth research. We further illustrate the potential impact of these biases in studies investigating the relationship between HIV/ART and preterm births. We also briefly discuss issues related to population-level estimations based on routinely collected clinical or civil registration data. We conclude by emphasizing the importance of strengthening of antenatal care services to improve quality of population data as well as optimizing current and future study designs, by taking into account the important methodological considerations described in this article.

[Dentification and correction of the lead-time bias and length bias in cancer screening studies].

Lead-time bias and length bias were common systematic errors in observational screening studies, which might be a common cause of overstating or distorting the true screening effects. One of key concerns in observational screening studies was how to estimate the screening effects based on the consideration of these two biases. This paper illustrated how to identify and correct the lead-time bias using the tumor volume doubling time and the non-homogeneous Poisson process, and how to correct the length bias using a weighted method. The application conditions of each method were also discussed to present several useful toolboxes to correct the lead-time bias and length bias appropriately and evaluate the effectiveness of the cancer screening program accurately.

History of Benzodiazepine Prescriptions and Risk of Dementia: Possible Bias Due to Prevalent Users and Covariate Measurement Timing in a Nested Case-Control Study.

Previous estimates of whether long-term exposure to benzodiazepines increases dementia risk are conflicting and are compromised by the difficulty of controlling for confounders and by reverse causation. We investigated how estimates for the association between benzodiazepine use and later dementia incidence varied based on study design choices, using a case-control study nested within the United Kingdom's Clinical Practice Research Datalink. A total of 40,770 dementia cases diagnosed between April 2006 and July 2015 were matched on age, sex, available data history, and deprivation to 283,933 control subjects. Benzodiazepines and Z-drug prescriptions were ascertained in a drug-exposure period 4-20 years before dementia diagnosis. Estimates varied with the inclusion of new or prevalent users, with the timing of covariate ascertainment, and with varying time between exposure and outcome. There was no association between any new prescription of benzodiazepines and dementia (adjusted odds ratio (OR) = 1.03, 95% confidence interval (CI): 1.00, 1.07), whereas an inverse association was observed among prevalent users (adjusted OR = 0.91, 95% CI: 0.87, 0.95), although this was likely induced by unintentional adjustment for colliders. By considering the choice of confounders and timing of exposure and covariate measurement, our findings overall are consistent with no causal effect of benzodiazepines or Z-drugs on dementia incidence.

Prevalence of periodontitis: misclassification, under-recognition or over-diagnosis using partial and full-mouth periodontal examination protocols.

OBJECTIVE: The aim of this cross-sectional study was to assess the bias in estimating the prevalence of periodontitis due to partial-mouth periodontal examination protocols (PMPE) and to relate the severity and extent of periodontal damage to periodontitis misclassification when applying case definitions by Centres of Disease Control and Prevention and American Academy of Periodontology (CDC/AAP). MATERIALS AND METHODS: A full-mouth periodontal examination (FMPE) was performed in 721 adults living in North Italy to identify moderate and severe periodontitis. These results were compared with those obtained with two PMPE protocols analyzing two interproximal sites on all teeth (fMB-DL) or four interproximal sites in two random diagonal quadrants (pMDB-MDL). RESULTS: Both PMPE systems estimated the prevalence of moderate periodontitis with limited bias (-2.79% for pMDB-MDL and -3.49% for fMB-DL), whereas produced larger relative biases for severe periodontitis (-28.74% versus - 14.55%). The percentage of under-recognition of existing periodontal disease was 8.9% under fMB-DL and 15.5% under pMDB-MDL. The diagnosis of moderate and severe periodontal disease was correctly assigned to individuals with on average 8% and 30% of pathological sites, respectively. CONCLUSION: These findings suggest that PMPE systems provide high level of bias when using CDC/AAP case definitions.

Use of self-reported vaccination status can bias vaccine effectiveness estimates from test-negative studies.

INTRODUCTION: A number of national/multi-national networks provide annual estimates of influenza vaccine effectiveness (VE) based on the test-negative design. Most of these networks use subject self-reports to define influenza vaccination history. In this study, we used simulations to estimate the degree to which self-reported vaccination status can bias test-negative VE estimates. METHODS: We simulated a population whose members are at risk for acute respiratory illness (ARI) due to influenza and for ARI due to other respiratory pathogens. Vaccination was assumed to reduce the risk of influenza but not of non-influenza ARI. We simulated a range of possible values for VE and for vaccine coverage. Across simulations, we varied the sensitivity and specificity of self-reported vaccination status relative to true vaccination. We estimated bias as the percent difference in VE in the presence of misclassification relative to true simulated VE. RESULTS: Assuming self-report has sensitivity of 95% and specificity of 90%, estimated VE underestimated true VE by 16% (95% confidence interval, 4-30%). Decreasing specificity of self-reports resulted in greater bias than decreasing sensitivity of self-reports. Bias also increased as vaccine coverage decreased. CONCLUSIONS: The use of self-reported influenza vaccination history can meaningfully bias influenza VE in test-negative studies. Researchers using test-negative designs should attempt to supplement or validate self-reported vaccination history using additional data sources.

Gestational Age at Arrest of Development: An Alternative Approach for Assigning Time at Risk in Studies of Time-Varying Exposures and Miscarriage.

The time between arrest of pregnancy development and miscarriage represents a window in which the pregnancy is nonviable and not developing. In effect, the pregnancy loss has already occurred, and additional exposure cannot influence its outcome. However, epidemiologic studies of miscarriage traditionally use gestational age at miscarriage (GAM) to assign time in survival analyses, which overestimates duration of exposure and time at risk. In Right From the Start, a pregnancy cohort study (2000-2012), we characterized the gap between estimated gestational age at arrest of development (GAAD) and miscarriage using transvaginal ultrasound in 500 women recruited from 3 states (North Carolina, Tennessee, and Texas). We compared effect estimates from models using GAAD with GAM to assign time at risk through a simulation study of several exposure patterns with varying effect sizes. The median gap between GAAD and miscarriage was 23 days (interquartile range, 15-32). Use of GAAD decreased the bias and variance of the estimated association for time-varying exposures, whereas half the time using GAM led to estimates that differed from the true effect by more than 20%. Using GAAD to assign time at risk should result in more accurate and consistent characterization of miscarriage risk associated with time-varying exposures.

Structure and Control of Healthy Worker Effects in Studies of Pregnancy Outcomes.

Much of the literature on the healthy worker effect focuses on studies of chronic disease and mortality; however, when studying pregnancy outcomes, these effects might differ because of the short, defined risk periods of most pregnancy outcomes. Three pregnancy-specific healthy worker effects have also been described, but the structure of these effects has not yet been investigated when occupational exposure, and not employment status, is the exposure of interest. We used directed acyclic graphs to examine healthy worker effects in studies of occupational exposures and pregnancy outcomes: the healthy hire effect, the healthy worker survivor effect, the desperation/privilege effect (differential workforce reentry after pregnancy), the reproductively unhealthy worker effect (women with live births leave the workforce, while women with nonlive births do not), and the insecure pregnancy effect (women with adverse pregnancy outcomes reduce their exposures in subsequent pregnancies). Given our assumptions, we conclude that the healthy hire effect, the desperation/privilege effect, the reproductively unhealthy worker effect, and the insecure pregnancy effect result from confounding that can be addressed if data on measured confounders, such as employment status, are available. The presence of the healthy worker survivor effect, however, varies by study design. Different types of healthy worker effects can be present in studies of occupational exposure and pregnancy outcomes, and many of them are easily addressed analytically.