Mr. Jesse A. Canchola, BA, MS, PStat - Roche Molecular Systems, Inc. - Manager, Biostatistics

Mr. Jesse A. Canchola

BA, MS, PStat

Roche Molecular Systems, Inc.

Manager, Biostatistics

Pleasanton, CA | United States

Main Specialties: Biotechnology, Epidemiology, Other, Public Health, Statistics

Additional Specialties: linear models, SAS, R

Mr. Jesse A. Canchola, BA, MS, PStat - Roche Molecular Systems, Inc. - Manager, Biostatistics

Mr. Jesse A. Canchola

BA, MS, PStat

Introduction

Jesse Canchola is a native Californian, his parents having emigrated from Mexico in the 1950's and 1960's and has been a statistician for over 25 years. Jesse is a statistician by training from UC Berkeley, UC Davis and UCLA.

He started his career as a bio/statistician at Alpha Therapeutics Corporation, located in Los Angeles, before moving to the Bay Area in 1992.

He worked as a statistician for 14 years in the Dept. of Medicine at UCSF where he was a consulting statistician to DOM departments and ran the intern program at the Center for AIDS Prevention Studies (CAPS), mentoring and supervising up to 10 interns, before moving to the molecular diagnostics industry at Siemens Healthcare Diagnostics in Berkeley for about 6 years where he worked on the VERSANT HCV and HIV-1 quantitative assays.

He then moved on to Tethys Bioscience, a small startup in Emeryville, California for two years working on the PreDx test - a preventive test that used 7 assay results in one algorithm for assessing your five year risk for conversion to Type II Diabetes mellitus.

After the company went under, Jesse landed at Roche Molecular Systems in Pleasanton, California.

Jesse was also adjunct faculty in the Department of Statistics at the California State University, East Bay in Hayward, California for 12 years from 1997 to 2008, teaching both undergraduate and graduate courses in statistics as well as being the SAS instructor for the Department.

Over the years, he has presented, and continues to present, at statistics and SAS conferences and written substantially in peer reviewed journals and in proceedings of statistics (including The Joint Statistical Meetings) and SAS conferences (including The SAS Global Forum/SUGI, PharmaSUG and WUSS; over 40 publications with over 2300 citations).

In his spare time you will find Jesse volunteering for the Alameda County Community Food Bank, as an unpaid Board member (and past President and Vice-President) for the Latin America Community Assistance Foundation (MyLACA.Org) which serves and empowers the rural poor of Latin America, and the All Saints Parish of Hayward in various past roles as School Technology Committee President and School and Church websites webmaster, videographer and photographer. Jesse also donates his statistical expertise to non-profits that cannot afford a statistician.

Primary Affiliation: Roche Molecular Systems, Inc. - Pleasanton, CA , United States

Specialties:

Additional Specialties:

Research Interests:


View Mr. Jesse A. Canchola’s Resume / CV

Publications

33Publications

76Reads

1236Profile Views

29PubMed Central Citations

PrecMod: An Automated Precision SAS® Macro for Random Effects Models

PharmaSUG 2016 May 8

Proceedings of The PharmaSUG Conference 2016

Typical random effects linear model estimation involves fitting a model with main effects that may or may not include nesting with other study factors. Part of the challenge comes when having to calculate the confidence intervals for the variance components using the correct effective degrees of freedom (Satterthwaite, 1946) and then iterating the macro over different grouping levels (if they exist). The current precision macro, PrecMod, surmounts these challenges and provides a clear and concise path towards efficient and timely calculations ready for reporting.

View Article
May 2016
6 Reads

Limit of Detection (LoD) Estimation Using Parametric Curve Fitting to PCR (Hit) Rate Data: The LoD_Est SAS Macro

SAS Global Forum 2016 Apr 18

Proceedings of The SAS Global Forum Conference 2016

The Limit of Detection (LoD) is defined as the lowest concentration or amount of material, target or analyte that is consistently detectable (for PCR quantitative studies, in at least 95% of the samples tested)1. In practice, the estimation of the LoD uses a parametric curve fit to a set of panel member (PM1, PM2, PM3, etc.) data where the responses are binary. Typically, the parametric curve fit to the percent detection levels takes on the form of a probit or logistic distribution. For this, the SAS PROBIT procedure can be used to fit a variety of distributions, including both the probit and logistic. We introduce the LOD_EST SAS macro that takes advantage of the SAS PROBIT procedure’s strengths and returns an information-rich graphic as well as a percent detection table with associated 95% exact (Clopper-Pearson) confidence intervals for the hit rates at each level.

View Article
April 2016
2 Reads

Using SAS for Error Grid Analysis (EGA) of Glycated Hemoglobin A1c (HbA1c)

pp78, WUSS 2013.

Proceedings of the Twenty First Western Users of SAS Software Conference

Error Grid Analysis (EGA) uses graphical error grids for performance evaluations of medical devices* or for comparing candidate and comparative measurement methodologies**. Though not as widely used as we would like, EGA is an instructive and visually appealing performance evaluation new or improved test on the y-axis with a paired measurement of the reference test on the colorful grid of clinical regions (as established by experienced clinicians glucose (mg/dL and mmol/l), hematocrit (%) and hemoglobin (g/dL) have been proposed, there is no known EGA for glycated hemoglobin A1c (HbA1c or, simply, A1c glycated hemoglobin A1c (%) using The SAS® System v9.3, including SAS/GRAPH® and the SAS/GRAPH® Annotate facility.

View Article
October 2013
15 Reads

Mediators of childhood sexual abuse and high-risk sex among men-who-have-sex-with-men.

Child Abuse Negl 2008 Oct 7;32(10):925-40. Epub 2008 Nov 7.

Oregon State University, College of Health Sciences, Department of Public Health, Waldo Hall, Corvallis, OR, USA.

View Article

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.chiabu.2007.12.010DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2701627PMC
October 2008
8 Reads
20 Citations

Reply to Handsfield and Marrazzo

J Inf Dis 2006;194:307–315

The Journal of Infectious Diseases

To the Editor—We wish to thank Drs. Handsfield and Marrazzo for their very thoughtful comments and questions [1] regarding our recently published article [2] assessing sociodemographic markers and behavioral correlates of sexually transmitted infections (STIs) in young adult women. Although the original goal of our research was to examine correlates of 3 prevalent STIs in young women in a single logistic‐regression model, Drs. Handsfield and Marrazzo make a valid point about the potential variation in correlates by each STI assessed. In response to their queries, we performed 3 separate logistic‐regression models to examine multivariate correlates of Chlamydia trachomatis, Neisseria gonorrhoeae, and Trichomonas vaginalis. All data analyses were executed using the same procedures described in our original article.

View Article
June 2007
7 Reads

Microalbuminuria in HIV infection.

AIDS 2007 May;21(8):1003-9

Duke University Medical Center, Department of Medicine, Division of Nephrology, Durham, North Carolina, USA.

View Article

Download full-text PDF

Source
http://dx.doi.org/10.1097/QAD.0b013e3280d3587fDOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3189480PMC
May 2007
4 Reads
5.554 Impact Factor

Understanding the demographic characteristics of urban men who have sex with men.

J Homosex 2006 ;51(3):33-51

Health Survey Research Unit, Center for AIDS Prevention Studies, Department of Medicine, University of California-San Francisco, CA 94105, USA.

View Article

Download full-text PDF

Source
http://dx.doi.org/10.1300/J082v51n03_03DOI Listing
January 2007
4 Reads

Improving rescreening in community clinics: does a system approach work?

J Community Health 2006 Dec;31(6):497-519

Institute for Health and Aging, University of California, San Francisco, Laurel Heights Campus, San Francisco, CA 94143-0646, USA.

View Article

Download full-text PDF

Source
December 2006
1 Read

Analysis of Group-Randomized Designs: A clinic-based breast cancer rescreening example.

WUSS 2005

Proceedings of the Thirteenth Annual Western Users of SAS Software Conference

Assigning multiple treatments/interventions to clinic s or communities over time (rather than to individual subjects over time) involves contamination issues and several key reasons for considering the group-randomized designs (GRD) (MRC, 2002; Donner & Klar, 2000; and Murray, 1998): 1. The treatment/intervention to be studied is itself delivered to and affects groups of people rather than individuals; 2. The treatment/intervention is targeted at health professionals with the aim of studying its impact on patient outcomes; 3. The treatment/intervention is given to individuals but might affect others within that cluster (i.e., so-called contamination); 4. The target population is highly mobile so that individual follow-up would be prohibitively expensive. We utilize the SAS MIXED and GLIMMIX procedures to analyze group randomized designs. We use several illustrations, including a clinic -based breast cancer rescreening example. Rationale for the strategies and procedures used is given and limitations are discussed.

View Article
October 2006
12 Reads

Mixed Model Selection via the Information-Theoretic Approach using SAS

WUSS 14

Proceedings of the Fourth Annual Western Users of SAS Software

Model-building via the hypothesis-testing method (i.e., variable selection) has been the de-facto standard in the statistical analyst’s toolbox for some time. Other methods that include Bayesian, cross-validation and the bootstrap are excellent competing methods but can be complex and resource intensive. Still, superior predictive algorithms such as neural networks, random forests or support vector machines can be used but suffer from interpretability limitations (Breiman, 2001). The information-theoretic approach to model selection, as codified by Burnham and Anderson (2002), involves averaging over a class of eligible models to provide robust inference. The IT approach side-steps the (sometimes controversial) issue of judging whether individual variables are important or not based on the traditional hypothesis testing dichotomy. We utilize the SAS COMPMIX macro to carry out the IT paradigm. We illustrate with an example from AIDS research. A rationale for the strategies is given and we discuss the limitations.

View Article
September 2006
3 Reads

Sociodemographic markers and behavioral correlates of sexually transmitted infections in a nonclinical sample of adolescent and young adult women.

J Infect Dis 2006 Aug 23;194(3):307-15. Epub 2006 Jun 23.

Department of Pediatrics, Division of Adolescent Medicine, University of California, San Francisco, San Francisco, CA 94143-0503, USA.

View Article

Download full-text PDF

Source
http://dx.doi.org/10.1086/506328DOI Listing
August 2006
6 Reads
6.000 Impact Factor

Commentary on Schroder et al. (2003a, 2003b).

Ann Behav Med 2005 Apr;29(2):86-95; discussion 96-9

AIDS Research Institute and Center for AIDS Prevention Studies, Department of Medicine, University of California, San Francisco 94105, USA.

View Article

Download full-text PDF

Source
http://dx.doi.org/10.1207/s15324796abm2902_2DOI Listing
April 2005
11 Reads

Use of videos by directors of medical ethics courses.

J Clin Ethics 2004 ;15(2):201-8

Department of General Internal Medicine, University of California, San Francisco, USA.

View Article

Download full-text PDF

Source
January 2005
1 Read

Mammography rescreening among women of diverse ethnicities: patient, provider, and health care system factors.

J Health Care Poor Underserved 2004 Aug;15(3):390-412

University of California, San Francisco, School of Nursing, Institute for Health and Aging, USA.

View Article

Download full-text PDF

Source
August 2004
4 Reads

Distress and depression in men who have sex with men: the Urban Men's Health Study.

Am J Psychiatry 2004 Feb;161(2):278-85

Department of Psychiatry, University of California at San Francisco, USA.

View Article

Download full-text PDF

Source
http://dx.doi.org/10.1176/appi.ajp.161.2.278DOI Listing
February 2004
4 Reads
12.295 Impact Factor

Recalling sexual behavior: a methodological analysis of memory recall bias via interview using the diary as the gold standard.

J Sex Res 2003 Nov;40(4):325-32

The Kinsey Institute for Research in Sex, Gender, and Reproduction, Morrison Hall 313, Indiana University, Bloomington, IN 47405, USA.

View Article

Download full-text PDF

Source
http://dx.doi.org/10.1080/00224490209552198DOI Listing
November 2003
7 Reads

Battering victimization among a probability-based sample of men who have sex with men.

Am J Public Health 2002 Dec;92(12):1964-9

Center for AIDS Prevention Studies, University of California, San Francisco, USA.

View Article

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1447360PMC
December 2002
13 Reads
4.552 Impact Factor

Imputation Strategies for Sexual Orientation using SAS Proc MI

WUSS 2002

Proceedings of the Tenth Annual Western Users of SAS Software Conference

Studies of sexual behavior have demonstrated behaviorally defined sexual orientation is a correlate of HIV and STD/STI risk behavior. Not all survey respondents report their sexual orientations, however. These incomplete cases result in loss of statistical power, untrustworthy population inferences, and stymie prevention efforts that target specific subpopulations defined by sexual orientation. To mitigate the effects of non-response to sexual orientation survey items, it can be of benefit to impute sexual orientation status from responses to sexual behavior variables. Traditional imputation methods use a single imputation whereas newer imputation methodology as implemented in the SAS MI procedure performs multiple imputation. The present study compares single and multiple imputation approaches for addressing sexual orientation item non-response. We discuss the limitations and benefits of each approach.

View Article
October 2002
6 Reads

A response to "developing standards in behavioral surveillance for HIV/STD prevention".

AIDS Educ Prev 2002 Aug;14(4):343-7

Center for Aids Prevention Studies, Universtiy of California-San Francisco 94105, USA.

View Article

Download full-text PDF

Source
August 2002
7 Reads

A response to "developing standards in behavioral surveillance for HIV/STD prevention".

AIDS Educ Prev 2002 Aug;14(4):343-7

Center for Aids Prevention Studies, Universtiy of California-San Francisco 94105, USA.

View Article

Download full-text PDF

Source
August 2002
7 Reads

Paul et al. Respond

Am J Public Health 2002;92: 1338-1345

American Journal of Public Health

We are gratified to note that Mathy does not have any substantive criticisms of our research. However, we would like to address 2 of Mathy’s points: first, that other population-based studies on this topic have been conducted, and second, that our analyses lacked cultural sensitivity...

View Article
June 2002
6 Reads

Computation of expected risk ratios in a longitudinal study of sexually transmitted diseases on per-contact risk of HIV using SAS/IML

WUSS 9

Proceedings of The Ninth Annual (WUSS) Western Users of SAS Software Conference

Using the method of Hayes et al. (1995), we provide a SAS/IML program to calculate the expected risk ratio associating an STD as a cofactor effect of the per-contact risk of acquiring HIV. Inputs are length of follow-up, duration of STD, basic mv transmission probability, total number of exposures during follow-up to mv positive (InV+) partners, and cofactor effect values.

View Article
October 2001
7 Reads

Health-related characteristics of men who have sex with men: a comparison of those living in "gay ghettos" with those who live elsewhere

Am J Pub Hlth 91:13-17

American Journal of Public Health

OBJECTIVES: This study investigated the limitations of probability samples of men who have sex with men (MSM), limited to single cities and to the areas of highest concentrations of MSM ("gay ghettos"). METHODS: A probability sample of 2881 MSM in 4 American cities completed interviews by telephone. RESULTS: MSM who resided in ghettos differed from other MSM, although in different ways in each city. Non-ghetto-dwelling MSM were less involved in the gay and lesbian community. They were also less likely to have only male sexual partners, to identify as gay, and to have been tested for HIV. CONCLUSIONS: These differences between MSM who live in gay ghettos and those who live elsewhere have clear implications for HIV prevention efforts and health care planning.

View Article
June 2001
13 Reads

The continuing HIV epidemic among men who have sex with men

Am J Pub Hlth 91:907-914

American Journal of Public Health

OBJECTIVE This study characterized the AIDS epidemic among urban men who have sex with men (MSM). METHODS: A probability sample of MSM was obtained in 1997 (n = 2881; 18 years and older) from New York, Los Angeles, Chicago, and San Francisco, and HIV status was determined through self-report and biological measures. RESULTS: HIV prevalence was 17% (95% confidence interval = 15%, 19%) overall, with extremely high levels in African Americans (29%), MSM who used injection drugs (40%), "ultraheavy" noninjection drug users (32%), and less educated men (< high school, 37%). City-level HIV differences were non-significant once these other factors were controlled for. In comparing the present findings with historical data based on public records and modeling, HIV prevalence appears to have declined as a result of high mortality (69%) and stable, but high, incidence rates (1%-2%). CONCLUSIONS: Although the findings suggest that HIV prevalence has declined significantly from the mid-1980s, current levels among urban MSM in the United States approximate those of sub-Saharan countries (e.g., 14%-25%) and are extremely high in many population subsegments. Despite years of progress, the AIDS epidemic continues unabated among subsegments of the MSM community.

View Article
June 2001
5 Reads

Priorities in HIV prevention. Response to Behrman and Mokotoff

Science, 291:45-46

Science

One of the scientific anomalies of the AIDS epidemic is the large difference in infection rates across populations. Given limited resources and segregated epidemics, prevention funding should be directed to population segments with high HIV prevalence and incidence. However, recent surveys of U.S. populations indicate that the allocation of prevention dollars is not consistent with the distribution of HIV in the population

View Article
June 2001
6 Reads

Obtaining HIV test results with a home collection test kit in a community telephone sample

J Acq Imm Def Syn Hum Ret, 24:363-368

Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology

OBJECTIVES: To test the feasibility of obtaining HIV test results by home collection kit from a probability telephone sample of men who have sex with men (MSM). METHODS: A quota sample of 615 MSM previously interviewed by the Urban Men's Health Study phone survey in Chicago, Los Angeles, New York City, and San Francisco were re-contacted and offered an HIV test using an oral specimen (Orasure) home collection kit. RESULTS: Eighty percent consented to be mailed a kit, and 84% returned a specimen, for a 67% participation rate. All self-reported HIV-positive persons tested positive (77 of 77); 4 of 266 (1.5%) with a prior negative test and 2 of 69 (2.9%) with no prior positive HIV test result. Participation was associated with self-reported prior HIV test status-HIV-positive (83%), HIV-negative (68%), or no prior HIV test result (54%)-and marginally associated with New York City residence after adjustment for HIV status (odds ratio = 0.7; 95% confidence interval, 0.4-1.1; p =.08). CONCLUSIONS: These results suggest that urban MSM identified and interviewed by telephone will participate in home collection HIV testing. This methodology could be used to produce population-based estimates of HIV seroprevalence and seroincidence in MSM and could probably be extended to other populations and other viral infections.

View Article
June 2000
3 Reads

U.S. priorities-HIV prevention

Science, 290:717

Science

One of the scientific anomalies of the AIDS epidemic is the large difference in infection rates across populations. Given limited resources and segregated epidemics, prevention funding should be directed to population segments with high HIV prevalence and incidence. However, recent surveys of U.S. populations indicate that the allocation of prevention dollars is not consistent with the distribution of HIV in the population.

View Article
June 2000
6 Reads

Random selection in a national telephone survey: a comparison of the Kish, Next-Birthday, and Last-Birthday Methods

J Off Stat, 16:53-59

Journal of Official Statistics

The process of randomly selecting a respondent in household surveys is one juncture in the interview session that often results in a large proportion of refusals. Do the standard methods (Kish, "last-birthday," and "next-birthday") differentially contribute to the dropout rate? At what stage in the screening process are informants/respondents more likely to drop out? The literature that addresses these questions is mixed. Utilizing a national Random Digit Dialing (RDD) sample design, we compared the differential dropout rate in the standard respondent selection methods; in particular, the magnitude of the dropout rate in each phase of the screening process. Contrary to what would be expected, we found a significant difference between the three methods in the dropout rate during the initial stages of the screening process. The highest proportion of dropout rates occurred in the screening interview before the informant was asked questions unique to one of the three selection methods. The higher rates were in the Kish and ``last-birthday' conditions, with the highest number in the Kish condition. We suggest that interviewers rather than respondents are a primary source of the higher rate of refusals when using the Kish method.

View Article
June 2000
5 Reads

Logistic Regression Modeling in Complex Surveys: Examples Using the LOGITSE SAS Macro

SUGI 1997, 1282-1287

Proceedings of the Fifth Annual Western Users of SAS Software Conference

Traditional formulae for standard errors and subsequent statistical significance tests implemented in various popular statistical packages are based on the premise that the data are a simple random sample (SRS) of observations from a superpopulation. Equivalently, the observations are assumed to be independent and identically distributed (IID). For complex analytic surveys, these assumptions are almost always invalid leading to potentially incorrect inferences due to the failure to adjust relevant standard errors of the parameters. Here, we concentrate on binary responses where a logistic regression analysis would be meaningful and introduce a SAS ® macro, LOGITSE, that takes the cluster-correlated nature of the complex survey design into account, thus providing for correct inference.

View Article
March 1997
4 Reads

Using the GEE v2.02 SAS Macro to Adjust for Clustering of Respondents within Interviewer in a Methodological National Probability Survey/Study

Authors:
JA Canchola

WUSS 4, 343-344

Proceedings of the Fourth Annual Western Users of SAS Software

Hierarchical data structures are readily found in survey research. For example, households are found within blocks, cities within area codes and respondents are nested within interviewers. Although, here, we are interested in the latter, one can generalize the use of the currently proposed method to similar forms of clustering found in research. Various authors have criticized existing studies on their statistical model failures to account (statistically) for interviewer effects when making inferences based on respondent characteristics (Hagenaars and Heinen, 1982; Dijkstra, 1983; Groves and Fultz, 1985). We show how the SAS® macro by Karim and Zeger, updated by Groemping, is utilized in a national behavioral methodological survey/study to control for interviewer effects (clustering of respondents within interviewer) when making inferences based on items influenced by interviewer manipulated study variables. We briefly describe the limitations and give conclusions.

View Article
October 1996
2 Reads

Effects of Interviewer Gender, Interviewer Choice, and Item Context on Responses to Questions Concerning Sexual Behavior

Pub Op Quart, 60:345-375

Public Opinion Quarterly

We examined factors influencing responses to questions on sexual behavior among adult respondents 18-49 years old (unweighted N=2,030) obtained through a random-digt dialing survey. Based on self-disclosure and perceived control theory, we hypothesized that giving people a choice in selecting the gender of their interviewer rather than being assigned an interviewer, and using questions that are "supportive" of what may be perceived of as nonnormative behavior (enhanced items), would increase data quality relative to, respectively, matchedor opposite-gender interviewer conditions and standard worded items. The enhanced items facilitated responding to a number of sensitive topics. However, the effects of item wording on item response are often mediated by interviewer conditions. The "choice" results suggest that giving respondents greater control decreases question threat. However, the overall findings argue for matching respondents and interviewers on gender over opposite-gender interviewers or allowing respondents to select their interviewer's gender. Wording and interviewer manipulations reduced the discrepancies between men's and women's self-reports of sexual behavior, but they did not eliminate them, and in some cases they had no effect. The present findings suggest that males tend to be influenced by variations in item wording, interviewer gender, and respondent control across a somewhat wider range of sexual topics. In general, the findings recommend matching respondents and interviewers on gender and the use of more supportive wording in sexual behavior question. However, for assessment of some topics (e.g., sexual violence) in particular segments of the population (e.g., men), other procedures, such as increasing respondent control, may be a better choice. Overall, the data support the view that in terms of preferred procedures, not all sexual topics are created equal.

View Article
June 1996
7 Reads

Risk Factors for HIV and Other Sexually Transmitted Diseases and Prevention Practices Among US Heterosexual Adults: Changes from 1990 to 1992

Am J Pub Hlth, 85:1492-1499

American Journal of Public Health

OBJECTIVES. The National AIDS Behavioral Survey (1990-1992) of heterosexual adults (18-49 years) measured human immunodeficiency virus (HIV) risk factors, condom use, and HIV antibody testing, with a focus on major "high-risk" cities. METHODS. A longitudinal survey was conducted. RESULTS. There was little reduction in the overall prevalence of HIV risk factors in the national or high-risk cities cohorts over time. Despite this picture of stability, approximately 39% of the population at risk for HIV because of multiple sexual partners turns over annually. There was little change in HIV test-seeking or in consistent condom use with primary sexual partners. Although the majority of at-risk respondents used condoms sporadically or not at all (65%), a significant increase in condom use was found among those reporting multiple sexual partners in both waves, particularly among Black heterosexuals. Data from other surveys and condom sales nationally support the findings. CONCLUSIONS. There is a need for a series of surveys in this area to assess the reliability of the present findings and to monitor the general US population's response to prevention programs.

View Article
June 1995
6 Reads

Followers