Browsing by Author "Li, Zhuokai"
Now showing 1 - 10 of 12
Results Per Page
Sort Options
Item Choice of Dose Level for a Randomized Clinical Trial of Low-Dose Bevacizumab vs Laser for Type 1 Retinopathy of Prematurity(American Medical Association, 2021) Kraker, Raymond T.; Wallace, David K.; Beck, Roy W.; Saunders, Christina T.; Lorenzi, Elizabeth; Melia, B. Michele; Li, Zhuokai; Ophthalmology, School of MedicineThis study determines which of 8 doses of bevacizumab are effective in treating severe retinopathy of prematurity to carry forward to a large-scale randomized clinical trial.Item Depressive symptoms are associated with fasting insulin resistance in obese youth(Wiley Online Library, 2014-10) Hannon, Tamara S.; Li, Zhuokai; Tu, Wanzhu; Humber, Jordan N.; Carroll, Aaron E.; Lagges, Ann M.; Gupta, Sandeep; Department of Pediatrics, School of MedicineBACKGROUND: In adults, depressive symptoms are positively associated with insulin resistance. OBJECTIVE: To determine whether an association exists between depressive symptoms and markers of insulin resistance in youth. METHODS: This study used a retrospective review of data from an obesity clinic. We evaluated the association between depressive symptoms (Children's Depression Inventory, CDI) and fasting insulin and homeostatic model assessment-insulin resistance (HOMA-IR) in obese youth (n = 207, age 10-18 years). Individuals with lower vs. higher CDI T-scores (<65 vs. ≥65) were compared; this cut-point is accepted as indicating the possibility of clinical depression. Multiple linear regression was used to evaluate relationships between CDI T-scores and insulin resistance. RESULTS: Fasting insulin and HOMA-IR values were 40% higher in patients with higher CDI T-scores (P = 0.04). After accounting for gender, race, age and body mass index, CDI T-score remained associated with HOMA-IR, although the strength of the association was small (b = 0.007, P = 0.049). CONCLUSIONS: Relationships between depressive symptoms and insulin resistance should be considered when evaluating obese youth.Item The Effect of Body Mass Index on Blood Pressure Varies by Race among Children(Office of the Vice Chancellor for Research, 2013-04-05) Li, Zhuokai; Eckert, George; Tu, Wanzhu; Gupta, Sandeep; Carroll, Aaron; Pratt, J. Howard; Hannon, Tamara S.The effect of adiposity on blood pressure (BP) intensifies as children become obese, and black children tend to have greater body mass index (BMI) and higher BP than age-matched white children. But few studies have compared the magnitude of the effect of BMI on BP in obese black and white children. We used a novel analytic technique to examine the influence of age and BMI on BP in children seen at a hospital-based obesity clinic. The study sample included 821 overweight and obese children (age and sex adjusted BMI% ranged from 87% to 100%; 306 males, 515 females, 362 blacks, and 459 whites). The mean age of the study subjects was 11.72 ± 3.48 years, the mean BMI was 36.22 ± 8.51 kg/m2, and the mean systolic and diastolic BP were 109.36 ± 16.10 and 69.99 ± 10.48 mmHg, respectively. In comparison, blacks and whites were similar in age (11.89 vs 11.58; p=0.197); while black patients had higher BMI (37.32 vs 35.34 kg/m2; p=0.0010), and higher systolic BP% than whites (58.71 vs 50.72 mmHg; p=0.00062). Semiparametric regression models showed that while age and BMI were significantly associated with systolic BP% in both race groups, black children had significantly higher BP% values as compared with white children of the same age and BMI (Fig 1 (a) and (b)). Although BP% values have taken into account the effect of age, there continued to be a significant effect of age on BP% in black children. In conclusion, among children referred for treatment of obesity, black children are at a significantly greater risk for having elevated BP as compared with their white peers of similar age and severity of obesity. Further research is needed to better understand this population-specific intensification of the adiposity effect on BP in obese black children.Item The effect of body mass index on blood pressure varies by race among obese children(De Gruyter, 2015-05) Hannon, Tamara S.; Gupta, Sandeep; Li, Zhuokai; Eckert, George; Carroll, Aaron E.; Pratt, J. Howard; Tu, Wanzhu; Department of Pediatrics, Indiana University School of MedicineObjective: Previous studies have shown that the effect of adiposity on blood pressure (BP) intensifies as children become increasingly obese. Black children tend to have greater body mass index (BMI) and higher BP than age-matched white children. It is unclear whether the BP effects of BMI are race-specific among black and white children, and data on obese Hispanic children are sparse. We compared the BP effect of BMI in obese white, black, and Hispanic children. Methods: We examined the medical records of children enrolled in a pediatric obesity clinic. Height, weight, BP, and fasting insulin were assessed as part of routine clinical care. The concurrent effects of age and BMI on BP percentile values were examined using semiparametric regression, which allows the accommodation of nonlinear effects. Results: The study included 873 children (338 male; 354 black, 447 white, 72 Hispanic; 11.7±3.5 years, BMI 36.2±8.5 kg/m2). While BMI Z-scores were similar among the groups, systolic BP (SBP) was higher in black children and Hispanic children (white: 107 mm Hg; black: 112 mm Hg; Hispanic: 112 mm Hg; p=0.0001). Age, sex, and height-adjusted SBP percentiles were significantly different among the three groups (white: 50; black: 59; Hispanic: 59; p=0.0006). In children of the same age, BP was higher at any given BMI in black children and Hispanic children. Conclusions: Among children referred for treatment of obesity, black children and Hispanic children are at a greater risk for having elevated BP when compared to white children of similar age and BMI.Item A generalized semiparametric mixed model for analysis of multivariate health care utilization data(Sage, 2018-12) Li, Zhuokai; Liu, Hai; Tu, Wanzhu; Biostatistics, School of Public HealthHealth care utilization is an outcome of interest in health services research. Two frequently studied forms of utilization are counts of emergency department (ED) visits and hospital admissions. These counts collectively convey a sense of disease exacerbation and cost escalation. Different types of event counts from the same patient form a vector of correlated outcomes. Traditional analysis typically model such outcomes one at a time, ignoring the natural correlations between different events, and thus failing to provide a full picture of patient care utilization. In this research, we propose a multivariate semiparametric modeling framework for the analysis of multiple health care events following the exponential family of distributions in a longitudinal setting. Bivariate nonparametric functions are incorporated to assess the concurrent nonlinear influences of independent variables as well as their interaction effects on the outcomes. The smooth functions are estimated using the thin plate regression splines. A maximum penalized likelihood method is used for parameter estimation. The performance of the proposed method was evaluated through simulation studies. To illustrate the method, we analyzed data from a clinical trial in which ED visits and hospital admissions were considered as bivariate outcomes.Item Low- and Very Low-Dose Bevacizumab for Retinopathy of Prematurity: Reactivations, Additional Treatments, and 12-Month Outcomes(Elsevier, 2022) Freedman, Sharon F.; Hercinovic, Amra; Wallace, David K.; Kraker, Raymond T.; Li, Zhuokai; Bhatt, Amit R.; Boente, Charline S.; Crouch, Eric R.; Hubbard, G. Baker; Rogers, David L.; VanderVeen, Deborah; Yang, Michael B.; Cheung, Nathan L.; Cotter, Susan A.; Holmes, Jonathan M.; Pediatric Eye Disease Investigator Group; Ophthalmology, School of MedicinePurpose: Low-dose and very low-dose intravitreal bevacizumab (IVB) have been reported to be successful in short-term treatment of type 1 retinopathy of prematurity (ROP), down to an initial dose of 0.004 mg. We now report 12-month outcomes for these infants. Design: Masked, multicenter, dose de-escalation study. Participants: One hundred twenty prematurely born infants with type 1 ROP. Methods: A cohort of 120 infants with type 1 ROP in at least 1 eye from 2 sequential dose de-escalation studies of low-dose IVB (0.25 mg, 0.125 mg, 0.063 mg, and 0.031 mg) or very low-dose IVB (0.016 mg, 0.008 mg, 0.004 mg, and 0.002 mg) to the study eye; the fellow eye (if also type 1) received 1 dose level higher of IVB. After primary success or failure at 4 weeks, clinical management was at investigator discretion, including all additional treatment. Main outcome measures: Reactivation of severe ROP by 6 months corrected age, additional treatments, retinal and other ocular structural outcomes, and refractive error at 12 months corrected age. Results: Sixty-two of 113 study eyes (55%) and 55 of 98 fellow eyes (56%) received additional treatment. Of the study eyes, 31 (27%) received additional ROP treatment, and 31 (27%) received prophylactic laser therapy for persistent avascular retina. No trend toward a higher risk of additional ROP treatment related to initial IVB doses was found. However, time to reactivation among study eyes was shorter in eyes that received very low-dose IVB (mean, 76.4 days) than in those that received low-dose IVB (mean, 85.7 days). At 12 months, poor retinal outcomes and anterior segment abnormalities both were uncommon (3% and 5%, respectively), optic atrophy was noted in 10%, median refraction was mildly myopic (-0.31 diopter), and strabismus was present in 29% of infants. Conclusions: Retinal structural outcomes were very good after low- and very low-dose IVB as initial treatment for type 1 ROP, although many eyes received additional treatment. The rate of reactivation of severe ROP was not associated with dose; however, a post hoc data-driven analysis suggested that reactivation was sooner with very low doses.Item Multivariate semiparametric regression models for longitudinal data(2014) Li, Zhuokai; Tu, Wanzhu; Liu, Hai; Katz, Barry P.; Fortenberry, J. DennisMultiple-outcome longitudinal data are abundant in clinical investigations. For example, infections with different pathogenic organisms are often tested concurrently, and assessments are usually taken repeatedly over time. It is therefore natural to consider a multivariate modeling approach to accommodate the underlying interrelationship among the multiple longitudinally measured outcomes. This dissertation proposes a multivariate semiparametric modeling framework for such data. Relevant estimation and inference procedures as well as model selection tools are discussed within this modeling framework. The first part of this research focuses on the analytical issues concerning binary data. The second part extends the binary model to a more general situation for data from the exponential family of distributions. The proposed model accounts for the correlations across the outcomes as well as the temporal dependency among the repeated measures of each outcome within an individual. An important feature of the proposed model is the addition of a bivariate smooth function for the depiction of concurrent nonlinear and possibly interacting influences of two independent variables on each outcome. For model implementation, a general approach for parameter estimation is developed by using the maximum penalized likelihood method. For statistical inference, a likelihood-based resampling procedure is proposed to compare the bivariate nonlinear effect surfaces across the outcomes. The final part of the dissertation presents a variable selection tool to facilitate model development in practical data analysis. Using the adaptive least absolute shrinkage and selection operator (LASSO) penalty, the variable selection tool simultaneously identifies important fixed effects and random effects, determines the correlation structure of the outcomes, and selects the interaction effects in the bivariate smooth functions. Model selection and estimation are performed through a two-stage procedure based on an expectation-maximization (EM) algorithm. Simulation studies are conducted to evaluate the performance of the proposed methods. The utility of the methods is demonstrated through several clinical applications.Item Plasma Levels of Bevacizumab and Vascular Endothelial Growth Factor After Low-Dose Bevacizumab Treatment for Retinopathy of Prematurity in Infants(American Medical Association, 2022) Hartnett, M. Elizabeth; Wallace, David K.; Dean, Trevano W.; Li, Zhuokai; Boente, Charline S.; Dosunmu, Eniolami O.; Freedman, Sharon F.; Golden, Richard P.; Kong, Lingkun; Prakalapakorn, S. Grace; Repka, Michael X.; Smith, Lois E.; Wang, Haibo; Kraker, Raymond T.; Cotter, Susan A.; Holmes, Jonathan M.; Ophthalmology, School of MedicineImportance: Intravitreal bevacizumab effectively treats severe retinopathy of prematurity (ROP), but it enters the bloodstream and may reduce serum vascular endothelial growth factor (VEGF), potentially causing detrimental effects on developing organs in the premature infant. Objective: To evaluate the association of intravitreal bevacizumab with plasma bevacizumab and VEGF concentrations at 2 and 4 weeks after predefined, de-escalating doses of intravitreal bevacizumab were administered to infants with severe ROP. Design, setting, and participants: This phase 1 dose de-escalation case series study was conducted at 10 US hospitals of ophthalmology institutions from May 21, 2015, to May 7, 2019. Blood samples were collected 2 and 4 weeks after intravitreal bevacizumab injection. Participants included 83 premature infants with type 1 ROP in 1 or both eyes and no previous ROP treatment. Data were analyzed from April 2017 to August 2021. Interventions: Study eyes received a single bevacizumab injection of 0.250 mg, 0.125 mg, 0.063 mg, 0.031 mg, 0.016 mg, 0.008 mg, 0.004 mg, or 0.002 mg. When the fellow eye required treatment, one dose higher was administered. Total dose administered at baseline was defined as the sum of doses given to each eye within 3 days of initial study-eye injection. Main outcomes and measures: Plasma bevacizumab concentration at 2 and 4 weeks after injection and the percentage change in plasma VEGF concentrations from pretreatment levels. Results: A total of 83 infants (mean [SD] age, 25 [2] weeks; 48 boys [58%]) were included in this study. Higher doses of bevacizumab administered at baseline were associated with higher plasma bevacizumab concentrations at 2 weeks (ρ, 0.53; 95% CI, 0.31-0.70) and 4 weeks (ρ, 0.44; 95% CI, 0.18-0.64). Plasma VEGF concentrations decreased by 50% or more from pretreatment levels in 40 of 66 infants (61%) at 2 weeks and 31 of 61 infants (51%) at 4 weeks, but no association was observed between the total dose of bevacizumab administered at baseline and percentage change in plasma VEGF concentrations 2 weeks (ρ, -0.04; 95% CI, -0.28 to 0.20) or 4 weeks (ρ, -0.17; 95% CI, -0.41 to 0.08) after injection. Conclusions and relevance: Results of this phase 1 dose de-escalation case series study revealed that bevacizumab doses as low as 0.002 mg were associated with reduced plasma VEGF levels for most infants at 2 and 4 weeks after intravitreal administration; however, no association was observed between total bevacizumab dose administered and reductions in plasma VEGF levels from preinjection to 2 weeks or 4 weeks. Additional studies are needed to evaluate the long-term effects of low-dose bevacizumab on neurodevelopment and retinal structure.Item Proportion of Osteoporotic Women Remaining at Risk for Fracture Despite Adherence to Oral Bisphosphonates(Elsevier, 2016-02) Imel, Erik A.; Eckert, George; Modi, Ankita; Li, Zhuokai; Martin, Joel; de Papp, Anne; Allen, Katie; Johnston, C. Conrad; Hui, Siu L.; Liu, Ziyue; Medicine, School of MedicineBackground Adherence to oral bisphosphonates is often low, but even adherent patients may remain at elevated fracture risk. The goal of this study was to estimate the proportion of bisphosphonate-adherent women remaining at high risk of fracture. Methods A retrospective cohort of women aged 50 years and older, adherent to oral bisphosphonates for at least two years was identified, and data were extracted from a multi-system health information exchange. Adherence was defined as having a dispensed medication possession ratio ≥ 0.8. The primary outcome was clinical occurrence of: low trauma fracture (months 7–36), persistent T-score ≤ − 2.5 (months 13–36), decrease in bone mineral density (BMD) at any skeletal site ≥ 5%, or the composite of any one of these outcomes. Results Of 7435 adherent women, 3110 had either pre- or post-adherent DXA data. In the full cohort, 7% had an incident osteoporotic fracture. In 601 women having both pre- and post-adherent DXA to evaluate BMD change, 6% had fractures, 22% had a post-treatment T-score ≤ − 2.5, and 16% had BMD decrease by ≥ 5%. The composite outcomes occurred in 35%. Incident fracture was predicted by age, previous fracture, and a variety of co-morbidities, but not by race, glucocorticoid treatment or type of bisphosphonate. Conclusion Despite bisphosphonate adherence, 7% had incident osteoporotic fractures and 35% had either fracture, decreases in BMD, or persistent osteoporotic BMD, representing a substantial proportion of treated patients in clinical practices remaining at risk for future fractures. Further studies are required to determine the best achievable goals for osteoporosis therapy, and which patients would benefit from alternate therapies.Item A sexually transmitted infection screening algorithm based on semiparametric regression models(Wiley, 2015-09-10) Li, Zhuokai; Liu, Hai; Tu, Wanzhu; Department of Biostatistics, Richard M. Fairbanks School of Public HealthSexually transmitted infections (STIs) with Chlamydia trachomatis, Neisseria gonorrhoeae, and Trichomonas vaginalis are among the most common infectious diseases in the United States, disproportionately affecting young women. Because a significant portion of the infections present no symptoms, infection control relies primarily on disease screening. However, universal STI screening in a large population can be expensive. In this paper, we propose a semiparametric model-based screening algorithm. The model quantifies organism-specific infection risks in individual subjects and accounts for the within-subject interdependence of the infection outcomes of different organisms and the serial correlations among the repeated assessments of the same organism. Bivariate thin-plate regression spline surfaces are incorporated to depict the concurrent influences of age and sexual partners on infection acquisition. Model parameters are estimated by using a penalized likelihood method. For inference, we develop a likelihood-based resampling procedure to compare the bivariate effect surfaces across outcomes. Simulation studies are conducted to evaluate the model fitting performance. A screening algorithm is developed using data collected from an epidemiological study of young women at increased risk of STIs. We present evidence that the three organisms have distinct age and partner effect patterns; for C. trachomatis, the partner effect is more pronounced in younger adolescents. Predictive performance of the proposed screening algorithm is assessed through a receiver operating characteristic analysis. We show that the model-based screening algorithm has excellent accuracy in identifying individuals at increased risk, and thus can be used to assist STI screening in clinical practice.