Walter W. Hauck

Johns Hopkins University, Baltimore, Maryland, United States

Are you Walter W. Hauck?

Claim your profile

Publications (255)1122.11 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose of the study: To facilitate replication, we examined delivery characteristics, acceptability, and depression outcomes of a home-based intervention, Get Busy Get Better, Helping Older Adults Beat the Blues (GBGB). GBGB, previously tested in a randomized trial, reduced depressive symptoms and enhanced quality of life in African Americans. Design and methods: A total of 208 African Americans aged above 55 years with Patient Health Questionnaire (PHQ-9) scores ≥5 on two subsequent screenings were randomized to receive GBGB immediately or 4 months later. GBGB involves up to 10 home sessions consisting of care management, referral/linkage, depression education/symptom recognition, stress reduction, and behavioral activation. Interventionists recorded delivery characteristics (dose, intensity) and perceived acceptability of sessions. Baseline and post-tests were used to characterize participants and examine associations between dose/intensity and depression scores. Participant satisfaction and perceived benefits were examined at 8 months. Results: Of 208 participants, 181 (87%, mean age = 69.6) had treatment data. Of these, 165 (91.2%) had ≥3 treatment sessions (minimal dose). Participants had on average 8.1 sessions (SD = 2.6) for an average of 65.4min (SD = 18.3) each. Behavioral activation and care management were provided the most (average of six sessions for average duration = 17.9 and 22.2min per session respectively), although all participants received each treatment component. GBGB was perceived as highly acceptable and beneficial by interventionists and participants. More sessions and time in program were associated with greater symptom reduction. Implications: GBGB treatment components were highly acceptable to participants. Future implementation and sustainability challenges include staffing, training requirements, reimbursement limitations, competing agency programmatic priorities, and generalizability to other groups.
    No preview · Article · Nov 2015 · The Gerontologist
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In prior works, this group demonstrated the feasibility of valid adaptive sequential designs for crossover bioequivalence studies. In this paper, we extend the prior work to optimize adaptive sequential designs over a range of geometric mean test/reference ratios (GMRs) of 70-143% within each of two ranges of intra-subject coefficient of variation (10-30% and 30-55%). These designs also introduce a futility decision for stopping the study after the first stage if there is sufficiently low likelihood of meeting bioequivalence criteria if the second stage were completed, as well as an upper limit on total study size. The optimized designs exhibited substantially improved performance characteristics over our previous adaptive sequential designs. Even though the optimized designs avoided undue inflation of type I error and maintained power at ≥ 80%, their average sample sizes were similar to or less than those of conventional single stage designs. Copyright © 2015 John Wiley & Sons, Ltd.
    Full-text · Article · Nov 2015 · Pharmaceutical Statistics
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this report, we emphasize the importance of public monographs with reference materials, coupled with careful process and change control and attention to GMPs, as a means of advancing access to good quality, safe, and effective medicines, with emphasis on available and incoming biologic medicines. With adequate control of articles covered by a monograph, these public standards can form the basis for a global public quality platform that covers reference products, non-interchangeable reference products, biosimilars, and interchangeable biosimilars. Working collaboratively with all stakeholders, new approaches allow these public standards to emerge nationally and globally in a timely way. Yet, there are increasing limitations in the availability of public standards for biologic medicines, which may reverse many decades of progress. Solutions are considered in this report.
    No preview · Article · Apr 2014 · The AAPS Journal
  • Source
    Nancy Hodgson · Laura N Gitlin · Laraine Winter · Walter W Hauck
    [Show abstract] [Hide abstract]
    ABSTRACT: Pain is under recognized and under managed in older adults with dementia. Because dementia patients have a diminished capacity to communicate discomfort, untreated pain may be expressed in the form of behavioral and psychiatric symptoms. The goal of the present study was to examine the relationship between pain and behavioral and psychiatric symptoms of dementia in community residing older adults from the perspective of the family caregiver. Dyads composed of 272 dementia patients and their family caregivers were assessed to determine dementia patient's mental status, and family caregiver's assessment of care recipient's pain, functional dependence and number of behavioral symptoms, analgesic use and demographic information. Hierarchical multiple regression analysis controlling for age, marital status, race, functional disability, and analgesic use showed that pain explained a small but significant percent of variance in the number of behavioral symptoms (3%, P<0.001). Pain had a stronger influence on the number of BPDS among those with severe cognitive impairment (F [1, 69]=11.75, P<0.001) compared to those with low to moderate cognitive impairment (F [1,199]=4.543, P=0.034.). The findings indicate that pain is a risk factor for behavioral symptoms in individuals with dementia and suggest that pain is a more significant predictor of behavior for individuals with severe dementia, compared to those with mild/moderate stage dementia. These results reinforce the importance of proper pain assessment and its management as part of dementia care planning.
    Full-text · Article · Nov 2013 · The Clinical journal of pain
  • Robert C Van Winkle · Walter W Hauck · Stephen J McGeady
    [Show abstract] [Hide abstract]
    ABSTRACT: Infants with recurrent infection may be found to have hypogammaglobulinemia without impaired specific antibody responses. Many will be diagnosed with transient hypogammaglobulinemia of infancy. This study used a parametric survival analysis of 100 infants with hypogammaglobulinemia to predict time to normalization. Aggregate initial immunoglobulins (IgG + IgA + IgM), as a percentage of age-adjusted normal, predicted time to resolution: median time to resolution for the infants in the lowest quartile of aggregate levels (≤81 % of age-adjusted lower limits) was greater than 5 years, with 34 % resolving in 3 years. For infants in the highest quartile (≥130 % of age-adjusted lower limits), the median was 9.9 months, with 77 % resolving in 3 years (P = 0.008). Initial IgG level, as a percentage of age-adjusted normal, also predicted resolution: the median time in the lowest quartile (≤78 % of age-adjusted lower limits) was greater than 5 years, with 36 % resolving in 3 years. In the highest quartile (≥128 %), the median time was 14.5 months, with 70 % resolving in 3 years (P = 0.010). Male sex was associated with more rapid resolution. The median time in males was 13 months, with 73 % resolution in 3 years. The median time in females was greater than 5 years, with 32 % resolution in 3 years. These results suggest that if a term infant presents with hypogammaglobulinemia, protective specific antibody titers, and an absence of other known immune deficiency, initial immunoglobulin levels and sex may predict time to normalization.
    No preview · Article · Sep 2013 · Journal of Clinical Immunology
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Chinese translation Effective care models for treating older African Americans with depressive symptoms are needed. To determine whether a home-based intervention alleviates depressive symptoms and improves quality of life in older African Americans. Parallel, randomized trial stratified by recruitment site. Interviewers assessing outcomes were blinded to treatment assignment. (ClinicalTrials.gov: NCT00511680) SETTING: A senior center and participants' homes from 2008 to 2010. African Americans aged 55 years or older with depressive symptoms. A multicomponent, home-based intervention delivered by social workers or a wait-list control group that received the intervention at 4 months. Self-reported depression severity at 4 months (primary outcome) and depression knowledge, quality of life, behavioral activation, anxiety, function, and remission at 4 and 8 months. Of 208 participants (106 and 102 in the intervention and wait-list groups, respectively), 182 (89 and 93, respectively) completed 4 months and 160 (79 and 81, respectively) completed 8 months. At 4 months, participants in the intervention group showed reduced depression severity (difference in mean change in Patient Health Questionnaire-9 score from baseline, -2.9 [95% CI, -4.6 to -1.2]; difference in mean change in Center for Epidemiologic Studies Depression Scale score from baseline, -3.7 [CI, -5.4 to -2.1]); improved depression knowledge, quality of life, behavioral activation, and anxiety (P < 0.001); and improved function (P = 0.014) compared with wait-list participants. More intervention than wait-list participants entered remission at 4 months (43.8% vs. 26.9%). After treatment, control participants showed benefits similar in magnitude to those of participants in the initial intervention group. Those in the initial intervention group maintained benefits at 8 months. The study had a small sample, short duration, and differential withdrawal rate. A home-based intervention delivered by social workers could reduce depressive symptoms and enhance quality of life in most older African Americans. National Institute of Mental Health.
    Full-text · Article · Aug 2013 · Annals of internal medicine
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To examine prevalence of modifiable risk factors and their contribution to patient quality of life (QoL) as rated by dementia patients and family caregivers. Cross-sectional. Home environment. 88 patients and their caregivers. Modifiable characteristics of home environments, patients, and caregivers were observed or obtained through interview. Demographics and ratings of patients' QoL were obtained from patients and caregivers. Patients had mean Mini-mental Status Examination (MMSE) score = 17.7 ± 4.6, (range: 10-28) on an average 7.7 ± 2.4 neuropsychiatric behaviors, 6.0 ± 3.1 health conditions and moderate functional challenges; 70.7% (N = 58) had fall risk; 60.5% (N = 52) had sleep problems at least once weekly; and 42.5% (N = 37) had pain. An average of 8.1 ± 5.2 home hazards and 5.4 ± 4.1 adaptations were observed; 51.7% had unmet device/navigation needs. Patients' and caregivers' QoL ratings were unrelated to MMSE; and patients' self-rated QoL was higher than rated by caregivers. Number of health conditions and unmet device/navigation needs were inversely associated with patient self-rated QoL, and number of health conditions, frequency of behaviors, and level of negative communications were inversely associated with caregiver's assessment of patient QoL. Positive endorsement of caregiving was positively associated with caregiver's appraisal of patient QoL. Other factors were unrelated. Most patients lived at home with high fall risk, unmanaged behavioral symptoms, pain, sleep disturbances, environmental challenges, and multiple hazards. Except for health, factors associated with lower QoL differed for patients and caregivers. Results suggest need to improve QoL by addressing modifiable risk factors and tailoring interventions to patient and caregiver perspectives.
    Full-text · Article · Jul 2013 · The American journal of geriatric psychiatry: official journal of the American Association for Geriatric Psychiatry
  • [Show abstract] [Hide abstract]
    ABSTRACT: The United States Pharmacopeial Convention has released Prednisone Tablets Reference Standard lot Q0H398 (Q0) to replace lot P1I300 (P1) for performance verification testing of dissolution Apparatus 1 and 2 as mandated by United States Pharmacopeia General Chapter Dissolution <711>. This report presents the collaborative study data, development of the acceptance limits, and results from supporting work for this lot. The collaborative study involved 21 collaborators who provided data for both Apparatus 1 and Apparatus 2. The study also evaluated the stability of Lot Q0 based on the dissolution results using Apparatus 1 and 2. The authors determined the geometric mean and percent coefficient of variation acceptance limits for 6-, 7-, 8-, 12-, and 14-position dissolution testers. The results of the collaborative study were used to establish the acceptance limits for the PVT for Apparatus 1 and 2 using lot Q0. Analysts found similarity in the data as well as in the acceptance limits for lots Q0 and P1, and dissolution results for Lot Q0 with Apparatus 1 appear to be stable. As with Lot P1, the dissolution results for Lot Q0 with Apparatus 2 show some decrease over time, and the geometric mean acceptance limits were adjusted accordingly. Results yielded acceptance limits for Apparatus 1 and Apparatus 2 that are different from those for Lot P1-but not markedly so-and are suitable for continued use in the PVT.
    No preview · Article · Feb 2013 · Dissolution Technologies
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The purpose of this study was to identify operational parameters of the vertical diffusion cell (VDC) apparatus that have an influence on results of drug release testing from semisolid dosage forms, which are one of three types of topically applied products (see USP General Chapter Topical and Transdermal Drug Products—Product Quality Tests <3>). The VDC apparatus operates using a static diffusion cell, a synthetic membrane, and an appropriate receptor medium. A cream formulation con-taining 1% hydrocortisone was used for the current study. The operational parameters investigated were stirring rate, mixing helix, stirring while sampling, medium degassing, membrane wetting with Ethomeen (a surfactant), and membrane wetting time. Stat-Ease design of experiment software was used to create partial factorial experimental designs to evaluate these parameters. The effects of the operational parameters were evaluated using mean drug release rate (slope, µg/cm 2 /min ½) and the standard deviation (SD) of six individual release rates for each experimental setup. Results of the study indicate that one parameter, the presence of Ethomeen for wetting the membrane, had a large and significant effect on both drug release rates and SD. Two parameters, stirring while sampling and mixing helix, had a signifi-cant impact on the drug release rate when Ethomeen was not used. Two parameters, medium degassing and stirring while sampling, had significant effects on the variability of the results (SD). Additionally, instrument-specific parameters (e.g., mixing helix) also contributed significantly to the variability of drug release rates.
    Full-text · Article · Aug 2012 · Dissolution Technologies
  • Source

    Full-text · Article · Aug 2012 · Dissolution Technologies
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: : To examine older African American's recognition of and beliefs about depressive symptoms, preferred symptom management strategies, and factors associated with willingness to use mental health treatments. Differences between the depressed and nondepressed and men and women were examined. : Cross-sectional survey. : Home, senior center. : A total of 153 senior center members (56 male, 97 female) 55 years and older. : Using a depression vignette, participants indicated if the person was depressed and their endorsement of items reflecting beliefs, stigma, symptom management, and willingness to use treatments (yes/no). A 9-item Patient Health Questionnaire assessed current symptomatology. : Overall, 24.2% reported depressive symptoms (≥5); 88.2% correctly identified the person in the vignette as depressed. Most (≥75%) endorsed active symptom management strategies, preference for treatment in physician and therapist offices, and willingness to take medications, seek therapy, see doctor, and attend support groups; less than 33% viewed depression as stigmatizing, whereas 48% viewed depression as normal aging. Logistic regressions revealed lower education, higher physical function, and feeling okay if community knew of depression diagnosis were associated with willingness to see physician if feeling depressed; being married and believing antidepressant medications are beneficial were related to willingness to use medications. Different associations emerged for depressed/nondepressed and men and women. : Overall, this older African American sample had positive attitudes and beliefs and endorsed traditional treatment modalities suggesting that beliefs alone are unlikely barriers to underutilization of mental health services. Because different factors were associated with willingness to seek physician help and use medications and factors differed for depressed/nondepressed and by sex, interventions should be tailored.
    Full-text · Article · May 2012 · The American journal of geriatric psychiatry: official journal of the American Association for Geriatric Psychiatry
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This study compared in vitro dissolution characteristics and other quality measures of different amoxicillin, metronidazole, and zidovudine products purchased in the Americas to a comparator pharmaceutical product (CPP). These three drugs are classified as Biopharmaceutics Classification System Class I drugs with the possibility that dissolution findings might be used to document bioequivalence. All investigated zidovudine products were found to be in vitro equivalent to the CPP. Only 3 of 12 tested amoxicillin products were found to be in vitro equivalent to the CPP. None of the tested metronidazole products were in vitro equivalent to the CPP. These findings suggest but do not confirm bioinequivalence where in vitro comparisons failed, given that an in vivo blood level study might have confirmed bioequivalence. At times, identifying a CPP in one of the selected markets proved difficult. The study demonstrates that products sold across national markets may not be bioequivalent. When coupled with the challenge of identifying a CPP in different countries, the results of this study suggest the value of an international CPP as well as increased use of BCS approaches as means of either documenting bioequivalence or signaling the need for further in vivo studies. Because of increased movement of medicines across national borders, practitioners and patients would benefit from these approaches.
    Full-text · Article · Apr 2012 · The AAPS Journal
  • Source
    Walter W Hauck
    [Show abstract] [Hide abstract]
    ABSTRACT: At present a complex global patchwork of private and public monographs and reference materials is variously available to help ensure the quality of medicines and foods. The relationship of these monographs and reference materials, one to another, frequently is inconsistently understood and documented.This article considers the complexity of monographs and reference materials with a focus on qualifying one reference material relative to another.
    Preview · Article · Feb 2012 · Pharmaceutical Research
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this work, we develop modeling and estimation approach for the analysis of cross-sectional clustered data with multimodal conditional distributions where the main interest is in analysis of subpopulations. It is proposed to model such data in a hierarchical model with conditional distributions viewed as finite mixtures of normal components. With a large number of observations in the lowest level clusters, a two-stage estimation approach is used. In the first stage, the normal mixture parameters in each lowest level cluster are estimated using robust methods. Robust alternatives to the maximum likelihood estimation are used to provide stable results even for data with conditional distributions such that their components may not quite meet normality assumptions. Then the lowest level cluster-specific means and standard deviations are modeled in a mixed effects model in the second stage. A small simulation study was conducted to compare performance of finite normal mixture population parameter estimates based on robust and maximum likelihood estimation in stage 1. The proposed modeling approach is illustrated through the analysis of mice tendon fibril diameters data. Analyses results address genotype differences between corresponding components in the mixtures and demonstrate advantages of robust estimation in stage 1.
    Full-text · Article · Feb 2012 · Journal of Applied Statistics
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In 2008, this group published a paper on approaches for two-stage crossover bioequivalence (BE) studies that allowed for the reestimation of the second-stage sample size based on the variance estimated from the first-stage results. The sequential methods considered used an assumed GMR of 0.95 as part of the method for determining power and sample size. This note adds results for an assumed GMR = 0.90. Two of the methods recommended for GMR = 0.95 in the earlier paper have some unacceptable increases in Type I error rate when the GMR is changed to 0.90. If a sponsor wants to assume 0.90 for the GMR, Method D is recommended. Copyright © 2011 John Wiley & Sons, Ltd.
    Full-text · Article · Jan 2012 · Pharmaceutical Statistics
  • W.W. Hauck · A.J. DeStefano · P. Tyle · R.L. Williams
    [Show abstract] [Hide abstract]
    ABSTRACT: During the past two years compendial scientists have devoted increasing attention to how uniformity-and particularly content uniformity-is assessed in chapters of the US Pharmacopeia (USP). This Stimuli article reports USP proposals about consistent approaches for all types of uniformity tests and for all dosage forms that have monographs in USP. The article solicits public comments about how USP should proceed with modifying general chapters that include assessment of uniformity.
    No preview · Article · Jan 2012
  • Joseph W Eaton · Daren Tran · Walter W Hauck · Erika S Stippler
    [Show abstract] [Hide abstract]
    ABSTRACT: To evaluate salicylic acid tablets as a candidate reference material in a Performance Verification Test (PVT) when a USP performance test for dissolution (General Chapter <711>) relies on USP Apparatus 4 (flow-through cell). We developed a dissolution procedure relying on Apparatus 4 and salicylic acid tablets. Thereafter, a designed experiment was conducted to identify operational variables that significantly affect salicylic acid dissolution in this apparatus. Four variables (size of glass beads, cell temperature, flow rate, level of deaeration) and one combination effect (deaeration/bead size) were significant for mean percent dissolved. Two variables (tablet orientation, level of deaeration) were significant for standard deviation results, but these effects were less pronounced than those for mean percent dissolved results. Three variables (analyst, tester manufacturer, amount of glass beads) had no statistically significant effects on either the mean or standard deviation of the responses. The proposed PVT is capable of probing effects of changes in several critical operational parameters of Apparatus 4. Salicylic acid tablets were shown to be a suitable reference material for the PVT. The PVT using salicylic acid tablets satisfies important aspects of a PVT.
    No preview · Article · Aug 2011 · Pharmaceutical Research
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Purpose On 1 March 2010, the US Pharmacopeial Convention released into commerce Lot P1I300 of its Prednisone Tablets Reference Standard for use in periodic performance verification testing (PVT) of dissolution Apparatus 1 and 2. This report presents the collaborative study data, development of the acceptance limits, and results from supporting work for this Lot. Methods The collaborative study involved 25 collaborators who provided data for Apparatus 1 and 31 who provided data for Apparatus 2. These limits are for the geometric mean and percent coefficient of variation (%CV) instead of per-individual results as for prior lots. Stability of results and sensitivity to test performance parameters were also studied. Results To determine new PVT acceptance limits, the authors calculated geometric mean and variance components as percent coefficient of variation. The move to the geometric mean and %CV criteria brings the acceptance criteria in line with current accepted statistics and provides a more realistic assessment of the system’s performance. Results for Apparatus 1 are stable over time, but for Apparatus 2, the mean decreases over time. Acceptance criteria are adjusted for this trend. Lot P1 demonstrates sensitivity to test performance parameters (vessels and degassing). Conclusions Apparatus 1 results are stable over time. Those in Apparatus 2 show a decrease over time in the geometric mean but show no trend in variability. The current tablets are shown to remain sensitive to two operational parameters, degassing and vessel dimensions, not covered by mechanical calibration. The new acceptance limits for Lot P1 are based on geometric mean and %CV for Prednisone Tablets Reference Standard Lot P1I300. The limits better control variability than the prior per-individual-result limits.
    Full-text · Article · Feb 2011 · Pharmaceutical Research
  • Laura N. Gitlin · Walter W. Hauck · Laraine Winter

    No preview · Article · Dec 2010 · JAMA The Journal of the American Medical Association
  • [Show abstract] [Hide abstract]
    ABSTRACT: On 1 March 2010, the US Pharmacopeial Convention released into commerce Lot P1I300 of its Prednisone Tablets Reference Standard for use in periodic performance verification testing (PVT) of dissolution Apparatus 1 and 2. This report presents the collaborative study data, development of the acceptance limits, and results from supporting work for this Lot. The collaborative study involved 25 collaborators who provided data for Apparatus 1 and 31 who provided data for Apparatus 2. These limits are for the geometric mean and percent coefficient of variation (%CV) instead of per-individual results as for prior lots. Stability of results and sensitivity to test performance parameters were also studied. To determine new PVT acceptance limits, the authors calculated geometric mean and variance components as percent coefficient of variation. The move to the geometric mean and %CV criteria brings the acceptance criteria in line with current accepted statistics and provides a more realistic assessment of the system's performance. Results for Apparatus 1 are stable over time, but for Apparatus 2, the mean decreases over time. Acceptance criteria are adjusted for this trend. Lot P1 demonstrates sensitivity to test performance parameters (vessels and degassing). Apparatus 1 results are stable over time. Those in Apparatus 2 show a decrease over time in the geometric mean but show no trend in variability. The current tablets are shown to remain sensitive to two operational parameters, degassing and vessel dimensions, not covered by mechanical calibration. The new acceptance limits for Lot P1 are based on geometric mean and %CV for Prednisone Tablets Reference Standard Lot P1I300. The limits better control variability than the prior per-individual-result limits.
    No preview · Article · Oct 2010 · Pharmaceutical Research

Publication Stats

9k Citations
1,122.11 Total Impact Points

Institutions

  • 2012
    • Johns Hopkins University
      Baltimore, Maryland, United States
  • 1995-2008
    • Thomas Jefferson University
      • • Division of Clinical Pharmacology
      • • Department of Pharmacology & Experimental Therapeutics
      • • Division of Biostatics
      • • Department of Radiation Oncology
      • • Department of Medicine
      Philadelphia, Pennsylvania, United States
  • 1991-2008
    • University of Waterloo
      • Department of Statistics and Actuarial Science
      Waterloo, Ontario, Canada
  • 2006-2007
    • University of Pittsburgh
      Pittsburgh, Pennsylvania, United States
    • George Washington University
      • School of Medicine and Health Sciences
      Washington, Washington, D.C., United States
  • 1996-2007
    • Jefferson College
      Хиллсборо, Missouri, United States
  • 2002
    • Allegheny General Hospital
      Pittsburgh, Pennsylvania, United States
  • 2001-2002
    • University of Pennsylvania
      • Division of General Internal Medicine
      Filadelfia, Pennsylvania, United States
    • National Cancer Institute (USA)
      • Laboratory of Experimental Carcinogenesis
      Maryland, United States
  • 1996-2000
    • Thomas Jefferson University Hospitals
      • • Division of Clinical Pharmacology
      • • Department of Radiation Oncology
      Philadelphia, Pennsylvania, United States
  • 1986-1997
    • University of California, San Francisco
      • • School of Pharmacy
      • • Department of Epidemiology and Biostatistics
      • • Department of Family Health Care Nursing
      • • Division of Hospital Medicine
      • • Department of Medicine
      San Francisco, California, United States
  • 1992
    • University of Chicago
      • Department of Obstetrics & Gynecology
      Chicago, Illinois, United States
  • 1989
    • The University of Western Ontario
      • Department of Epidemiology and Biostatistics
      London, Ontario, Canada
  • 1983
    • Northwestern University
      Evanston, Illinois, United States