- Browse by Subject
Browsing by Subject "Latent class analysis"
Now showing 1 - 10 of 10
Results Per Page
Sort Options
Item A simple two-step procedure using the Fellegi-Sunter model for frequency-based record linkage(Taylor & Francis, 2021-05-04) Xu, Huiping; Li, Xiaochun; Grannis, Shaun; Biostatistics, School of Public HealthThe widely used Fellegi-Sunter model for probabilistic record linkage does not leverage information contained in field values and consequently leads to identical classification of match status regardless of whether records agree on rare or common values. Since agreement on rare values is less likely to occur by chance than agreement on common values, records agreeing on rare values are more likely to be matches. Existing frequency-based methods typically rely on knowledge of error probabilities associated with field values and frequencies of agreed field values among matches, often derived using prior studies or training data. When such information is unavailable, applications of these methods are challenging. In this paper, we propose a simple two-step procedure for frequency-based matching using the Fellegi-Sunter framework to overcome these challenges. Matching weights are adjusted based on frequency distributions of the agreed field values among matches and non-matches, estimated by the Fellegi-Sunter model without relying on prior studies or training data. Through a real-world application and simulation, our method is found to produce comparable or better performance than the unadjusted method. Furthermore, frequency-based matching provides greater improvement in matching accuracy when using poorly discriminating fields with diminished benefit as the discriminating power of matching fields increases.Item Chromosome 17q12-21 Variants Are Associated with Multiple Wheezing Phenotypes in Childhood(American Thoracic Society, 2021) Hallmark, Brian; Wegienka, Ganesa; Havstad, Suzanne; Billheimer, Dean; Ownby, Dennis; Mendonca, Eneida A.; Gress, Lisa; Stern, Debra A.; Biagini Myers, Jocelyn; Khurana Hershey, Gurjit K.; Hoepner, Lori; Miller, Rachel L.; Lemanske, Robert F.; Jackson, Daniel J.; Gold, Diane R.; O’Connor, George T.; Nicolae, Dan L.; Gern, James E.; Ober, Carole; Wright, Anne L.; Martinez, Fernando D.; ECHO-CREW; Pediatrics, School of MedicineRationale: Birth cohort studies have identified several temporal patterns of wheezing, only some of which are associated with asthma. Whether 17q12-21 genetic variants, which are closely associated with asthma, are also associated with childhood wheezing phenotypes remains poorly explored. Objectives: To determine whether wheezing phenotypes, defined by latent class analysis (LCA), are associated with nine 17q12-21 SNPs and if so, whether these relationships differ by race/ancestry. Methods: Data from seven U.S. birth cohorts (n = 3,786) from the CREW (Children’s Respiratory Research and Environment Workgroup) were harmonized to represent whether subjects wheezed in each year of life from birth until age 11 years. LCA was then performed to identify wheeze phenotypes. Genetic associations between SNPs and wheeze phenotypes were assessed separately in European American (EA) (n = 1,308) and, for the first time, in African American (AA) (n = 620) children. Measurements and Main Results: The LCA best supported four latent classes of wheeze: infrequent, transient, late-onset, and persistent. Odds of belonging to any of the three wheezing classes (vs. infrequent) increased with the risk alleles for multiple SNPs in EA children. Only one SNP, rs2305480, showed increased odds of belonging to any wheezing class in both AA and EA children. Conclusions: These results indicate that 17q12-21 is a “wheezing locus,” and this association may reflect an early life susceptibility to respiratory viruses common to all wheezing children. Which children will have their symptoms remit or reoccur during childhood may be independent of the influence of rs2305480.Item Differential typologies of current substance use among Black and White high-school adolescents: A latent class analysis(Elsevier, 2020-07) Banks, Devin E.; Bello, Mariel S.; Crichlow, Queenisha; Leventhal, Adam M.; Barnes-Najor, Jessica; Zapolski, Tamika C.B.; Psychology, School of ScienceBlack and White adolescents demonstrate different prototypical profiles (i.e., typologies) of substance use, with Blacks demonstrating lower risk for concurrent use of two or more substances. Despite knowledge of these differences, typologies of adolescent substance use identified by person-centered methods, such as latent class analysis, have not characterized profiles by racial group. The current study examined typologies of substance use among Black and White youth separately using person-centered methods to identify common patterns of substance use among subjects. Data were drawn from a 5-year parent study examining adolescent health outcomes. The current study examined high-school aged White (n = 7271, 45.4% male) and Black youth (n = 1301, 40.1% male) who reported past-30-day frequency of cigarette, alcohol, marijuana, inhalant, and other drug use. Latent class analysis was used to examine substance use typologies among each group adjusting for grade and sex. Black and White youth demonstrated different typologies such that four typologies emerged among Blacks: Non-Use (87.8%), Alcohol and Marijuana Use (6.3%), Alcohol, Marijuana, and Cigarette Use (3.8%), and Frequent Polysubstance Use (2.0%). Conversely, five typologies emerged among Whites: Non-Use (73.4%), Predominant Alcohol Use (13.9%), Alcohol, Marijuana, and Cigarette Use (9.4%), Moderate Polysubstance Use (1.6%), and Frequent Polysubstance Use (1.7%). Findings suggest that Black and White youth engage in similar rates of concurrent substance use. Given that Black youth face greater risk for adverse consequences from substance use, prevention efforts are needed to prevent related health disparities related to concurrent substance use.Item Examining Training Motivations Among Public Health Workers(Wolters Kluwer, 2019-03) Apathy, Nate C.; Yeager, Valerie A.; Health Policy and Management, School of Public HealthCONTEXT: As public health needs and priorities evolve, maintaining a trained public health workforce is critical to the success of public health efforts. Researchers have examined training needs in various contexts and subpopulations, but a nationally representative study of what motivates public health workers to seek out training has yet to be conducted. By understanding these motivations, public health agencies and policy makers can appeal to worker motivations in both training programs and organizational incentives. OBJECTIVE: The purpose of this article was to describe overall training motivations and identify patterns of training motivations among public health workers. This study also explored whether or not training needs differ across prevalent motivational patterns. DESIGN AND PARTICIPANTS: Using data from the 2017 Public Health Workforce Interests and Needs Survey (PH WINS), the study used latent class analysis (LCA) to identify motivational patterns and logistic regression to analyze associations with training needs. RESULTS: The most prominent motivation to seek training was personal growth (82.7% of respondents). LCA identified 4 motivational classes of public health workers: those motivated by organizational pressure and requirements (31.8%), those motivated indiscriminately by all factors (28.4%), those motivated primarily by personal growth (21.7%), and those motivated by organizational accommodations and supports (18.2%). Motivational class was not associated with indicating training needs in any of 8 training domains, nor was it associated with indicating any training need in any domain. CONCLUSIONS: Public health agencies should consider the different motivational classes present in the public health workforce. In particular, motivational classes that represent organizational choices suggest that public health agencies should both motivate workers with organizational requirements and pressure from managers and offer institutional support via paid travel and covered time for training.Item Exploring Parental Influence on the Progression of Alcohol Use in Mexican-Heritage Youth: a Latent Transition Analysis(Springer, 2016-02) Shin, YoungJu; Lee, Jeong-Kyu; Lu, Yu; Hecht, Michael; Department of Communication Studies, School of Liberal ArtsMexican-heritage youth are members of the fastest growing minority group and are at particular risk for substance use including alcohol consumption. Youth face numerous risk factors including positive descriptions of substance use on media and peer offers that are potentially ameliorated by parental anti-substance use socialization efforts. Guided by primary socialization theory and the theory of planned behavior, the present study posited eight research questions to identify discrete subgroups/patterns of Mexican-heritage youth alcohol use behavior and parental influence on youth outcomes. Longitudinal survey data (n = 1147) from youth in 29 public schools located in Phoenix, Arizona, were collected over 3 years. Latent class and transition analyses identified four discrete subgroups characterized by response patterns of alcohol use behaviors and perceptions in Mexican-heritage youth: (1) non-drinker, (2) potential drinker, (3) experimenter, and (4) regular drinker. Targeted parent-child communication about alcohol and parental monitoring were found to be significant predictors for youth alcohol use. Research implications and future directions are suggested.Item Latent Class Symptom Profiles of Colorectal Cancer Survivors with Cancer-Related Cognitive Impairment(Research Square, 2023-04-12) Han, Claire; Saligan, Leorey; Crouch, Adele; Kalady, Matthew; Noonan, Anne; Lee, Lena; Von Ah, Diane; School of NursingPurpose: Colorectal cancer (CRC) survivors experience cancer-related cognitive impairment and cooccurring symptoms after cancer treatments. There has been little data to inform the risk factors of complex symptom phenotypes in CRC survivors. Objectives: To determine if subgroups of CRC survivors after cancer treatments could be identified based on the cognitive impairment and common co-occurring symptoms (depression, anxiety, sleep disturbance, fatigue, and pain); and to explore risk factors (sociodemographic and clinical characteristics, perceived stress, and social support) of these subgroups. Methods: Latent class profile analysis (LCPA) was used to identify subgroups based on self-reported symptoms in 64 CRC survivors. Cognitive impairment was measured by assessing subjective cognitive function using the Patient-Reported Outcome Measurement Information System (PROMIS) measure. The Kruskal-Wallis test and regression analyses were performed. Results: Three distinct latent classes were identified (Class 1: All Low ‘28.1%’; Class 2: High Psychological Symptoms (depression/anxiety) ‘25%’; Class 3: High Somatic Symptoms (fatigue, sleep disturbance, and pain) with High Cognitive Impairment’46.9%’). Pain was the most distinguishable symptom across the latent classes. The high symptom burden group was associated with less time since cancer diagnosis, higher perceived stress levels, and poor emotional social support. Conclusion: Our study adds to the information on interindividual variability in symptom experience of CRC survivors with cognitive impairment. Findings suggest a need for increased attention to screening for cooccurring symptoms (e.g., high pain) and future interventions focused on stress management and social supports.Item Patterns of opioid use behaviors among patients seen in the emergency department: Latent class analysis of baseline data from the POINT pragmatic trial(Elsevier, 2023) Bray, Bethany C.; Watson, Dennis P.; Salisbury-Afshar, Elizabeth; Taylor, Lisa; McGuire, Alan; Social and Behavioral Sciences, School of Public HealthIntroduction: The nation's overdose epidemic has been characterized by increasingly potent opioids resulting in more emergency department (ED) encounters over time. ED-based opioid use interventions are growing in popularity; however, they tend to treat people who use opioids as a homogenous population. The current study sought to understand heterogeneity among people who use opioids who encounter the ED by identifying qualitatively different subgroups among participants in an opioid use intervention clinical trial at baseline and examining associations between subgroup membership and multiple correlates. Methods: Participants were from a larger pragmatic clinical trial of the Planned Outreach, Intervention, Naloxone, and Treatment (POINT) intervention (n = 212; 59.2 % male, 85.3 % Non-Hispanic White, mean age = 36.6 years). The study employed latent class analysis (LCA) using five indicators of opioid use behavior: preference for opioids, preference for stimulants, usually use drugs alone, injection drug use, and opioid-related problem at ED encounter. Correlates of interest included participants' demographics, prescription histories, health care contact histories, and recovery capital (e.g., social support, naloxone knowledge). Results: The study identified three classes: (1) noninjecting opioid preferers, (2) injecting opioid and stimulant preferers, and (3) social nonopioid preferers. We identified limited significant differences in correlates across the classes: differences existed for select demographics, prescription histories, and recovery capital but not for health care contact histories. For example, members of Class 1 were the most likely to be a race/ethnicity other than non-Hispanic White, oldest on average, and most likely to have received a benzodiazepine prescription, whereas members of Class 2 had the highest average barriers to treatment and members of Class 3 were the least likely to have been diagnosed with a major mental health illness and had the lowest average barriers to treatment. Conclusions: LCA identified distinct subgroups among POINT trial participants. Knowledge of such subgroups assists with the development of better-targeted interventions and can help staff to identify the most appropriate treatment and recovery pathways for patients.Item Symptom experiences in post-treatment cancer survivors: associations with acceptance and commitment therapy constructs(Springer, 2021) Lewson, Ashley B.; Johns, Shelley A.; Krueger, Ellen; Chinh, Kelly; Kidwell, Kelley M.; Mosher, Catherine E.; Psychology, School of SciencePurpose: Acceptance and Commitment Therapy (ACT) has improved symptom and quality-of-life outcomes in pilot research with post-treatment cancer survivors. To further test the ACT model, the present study examined relationships between ACT constructs and subgroups of post-treatment survivors based on the severity of common symptoms. Methods: Survivors who had completed primary treatment for stage I or II cancer (N = 203) participated in this one-time survey. Latent class analysis (LCA) was used to identify subgroups of survivors based on the severity of fatigue, sleep disturbance, pain, anxiety, and depressive symptoms. Multinomial logistic regressions employing Vermunt's 3-step approach were used to examine ACT constructs (e.g., mindfulness, acceptance, values progress) as correlates of survivor subgroups based on symptoms. Results: The LCA showed three survivor classes: (1) mild-to-moderate levels of all symptoms except for normal pain intensity; (2) mild anxiety, moderate fatigue, and normal levels of all other symptoms; and (3) normal levels of all symptoms. Lower mindfulness, acceptance, and values progress and higher cognitive fusion, psychological inflexibility, and values obstruction were associated with a greater likelihood of being in class 1 or 2 than in class 3. Conclusion: Findings are consistent with the ACT model. Survivors with greater symptom burden reported greater withdrawal from personally meaningful activities and less acceptance of their cancer diagnosis and internal experiences (e.g., thoughts, feelings, symptoms). Findings provide strong justification for further testing of ACT to reduce symptom-related suffering in cancer survivors.Item Symptom experiences in post-treatment cancer survivors: Associations with acceptance and commitment therapy constructs(SpringerLink, 2020-11-02) Lewson, Ashley B.; Johns, Shelley A.; Krueger, Ellen; Chinh, Kelly; Kidwell, Kelley M.; Mosher, Catherine E.; Psychology, School of SciencePurpose: Acceptance and commitment therapy (ACT) has improved symptom and quality-of-life outcomes in pilot research with post-treatment cancer survivors. To further test the ACT model, the present study examined relationships between ACT constructs and subgroups of post-treatment survivors based on the severity of common symptoms. Methods: Survivors who had completed primary treatment for stage I or II cancer (N=203) participated in this one-time survey. Latent class analysis (LCA) was used to identify subgroups of survivors based on the severity of fatigue, sleep disturbance, pain, anxiety, and depressive symptoms. Multinomial logistic regressions employing Vermunt’s 3-step approach were used to examine ACT constructs (e.g., mindfulness, acceptance, values progress) as correlates of survivor subgroups based on symptoms. Results: LCA showed three survivor classes: (1) mild to moderate levels of all symptoms except for normal pain intensity, (2) mild anxiety, moderate fatigue, and normal levels of all other symptoms, and (3) normal levels of all symptoms. Lower mindfulness, acceptance, and values progress and higher cognitive fusion, psychological inflexibility, and values obstruction were associated with a greater likelihood of being in class 1 or 2 than class 3. Conclusion: Findings are consistent with the ACT model. Survivors with greater symptom burden reported greater withdrawal from personally meaningful activities and less acceptance of their cancer diagnosis and internal experiences (e.g., thoughts, feelings, symptoms). Findings provide strong justification for further testing of ACT to reduce symptom-related suffering in cancer survivors.Item Understanding the Better Than Average Effect on Altruism(Frontiers Media, 2021-01-07) Xiao, Yunyu; Wong, Kelly; Cheng, Qijin; Yip, Paul S. F.; School of Social WorkPrior research suggests that most people perceive themselves to be more altruistic than the average population, an observation known as the better-than-average (BTA) effect. Understanding the BTA effect carries significant public health implications, as self-perceived altruism is closely related to altruistic behaviors, which plays a significant role in individual and societal well-being. However, little is known about whether subpopulations with specific sociodemographic profiles are more likely to hold BTA altruistic self-perceptions, making it difficult to design targeted programs based on multiple sociodemographic characteristics to promote altruistic behaviors. This study addresses this gap by identifying the sociodemographic profiles of populations who are more likely to exhibit BTA effects on trait altruism. Data were derived from a representative sample of Hong Kong citizens (n = 1,185) in the 2017 Hong Kong Altruism Survey. A latent class analysis was performed using four domains of sociodemographic characteristics: sex, age, religion, and socioeconomic status. Multivariate multinomial logistic regressions were conducted to examine associations between class membership, BTA effect, and altruistic behaviors. The results yielded four classes of sociodemographic profiles. Middle-aged, Christian/Catholic, highly educated, and high-income individuals (Class 4, 17.8%) were most likely to exhibit BTA effects and behave altruistically; Class 3 (14.0%) were older, male, no/other religious belief, low education, and least likely to exhibit BTA effects and behave altruistically. Findings improve the understanding of the sociodemographic profiles of people showing BTA effects and facilitate targeted policy development to effectively promote altruism.