Kenneth Portier

Emory University, Atlanta, Georgia, United States

Are you Kenneth Portier?

Claim your profile

Publications (21)34.99 Total impact

  • Liora Sahar, Kenneth M. Portier, Elizabeth Ward, Marcia Watts
    [Show abstract] [Hide abstract]
    ABSTRACT: Background: American Cancer Society (ACS) staff engages in multiple activities to support cancer patients, survivors and caregivers and influence cancer policies in the US. GIS is a key tool in Public Health program planning, delivery and evaluation. Non-profit programs, such as those in ACS, have not typically used geospatial analysis in planning resource allocation and implementing services, advocacy activities and research funding. While leadership identifies objectives, program managers determine local population targets and activities. Many programs utilize volunteers and staff, and hence program success requires effective communication methodologies. Objective/Purpose: Report on geospatial analyses and visualizations needs within ACS and describe how reports containing geospatial products for pilot programs are being distributed across the organization to inform decision making at all levels of the organization. Methods: A needs-assessment to evaluate geospatial data and analysis requirements was performed as part of a larger organizational transformation in ACS. Program directors were surveyed about data and analysis required to support effective and quality program delivery. Internal and external sources of data and the appropriate tools to satisfy these needs were identified and acquired. Results: Advanced spatial analysis and modeling are used to help target program delivery to focus populations. A transportation shortage index that identifies geographic areas for expansion of patient transportation services (ACS Road To Recovery program) is presented. A geospatial analyses that supports volunteer and patient recruitment efforts for a key quality of life program (the Look Good Feel Better personal appearance and self-image improvement sessions) is also presented. Discussion/Conclusions: Establishing robust geospatial analysis and visualization within health non-profits such as ACS enables informed decision making and better planning and resource allocation. Continued development of programs using geospatial analysis and visualizations are needed to improve program processes and outcomes, and to train staff and volunteers to better target populations in need.
    142nd APHA Annual Meeting and Exposition 2014; 11/2014
  • [Show abstract] [Hide abstract]
    ABSTRACT: Physician characteristics and changes in drug reimbursement rates have been shown to influence practice patterns regardless of clinical guidelines, patient, clinical, or sociodemographic factors. We concurrently examined the association between urologists׳ characteristics and non-evidence-based use of primary medical androgen deprivation therapy (ADT) for clinically localized patients with prostate cancer, before and after the 2003 Medicare Modernization Act׳s reductions in ADT reimbursement rates.
    Urologic oncology. 05/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: Objectives Physician characteristics and changes in drug reimbursement rates have been shown to influence practice patterns regardless of clinical guidelines, patient, clinical, or sociodemographic factors. We concurrently examined the association between urologists׳ characteristics and non–evidence-based use of primary medical androgen deprivation therapy (ADT) for clinically localized patients with prostate cancer, before and after the 2003 Medicare Modernization Act׳s reductions in ADT reimbursement rates. Methods and materials The Surveillance, Epidemiology, and End Results-Medicare–linked database and the American Medical Association Physician Masterfile are used in a retrospective analysis of 12,255 patients diagnosed between 2001 and 2007 with clinical stage T1-T2, low- to intermediate-grade prostate cancer, and the 1,863 urologists who treated them. Logistic multilevel regression analyses are used to evaluate the association of urologists׳ characteristics on ADT use among patients within 6 months of diagnosis. Results Overall, 3,866 (32%) patients received non–evidence-based ADT. After adjusting for patient and urologist characteristics, patients treated by urologists with no medical school affiliations, compared with those treated by urologists with major medical school affiliations, are significantly more likely to receive non–evidence-based medical ADT (odds ratio = 2.35; 95% CI: 1.71–3.23; P<0.0001). Non–US-trained urologists are also more likely to prescribe non–evidence-based medical ADT (odds ratio = 1.64; 95% CI: 1.33–2.04; P<0.0001). Conclusions Patients treated by non–medical school–affiliated or non–US-trained urologists or both are significantly more likely to receive non–evidence-based ADT before and after the passage of the Medicare Modernization Act. Better strategies to encourage evidence-based ADT use on clinically localized patients with prostate cancer may be of benefit especially among non–medical school–affiliated or non–US-trained urologists or both.
    Urologic Oncology: Seminars and Original Investigations. 01/2014;
  • [Show abstract] [Hide abstract]
    ABSTRACT: Research Objective: Prostate cancer treatment patterns have been shown to vary by physician and patient characteristics. For patients with low-risk localized prostate cancer, we examined the association between their region of residence and their radiation oncologists' practice affiliations with medical schools on the likelihood they would receive both external beam radiation therapy (EBRT) and brachytherapy (BT) a treatment regimen that is at variance with clinical guidelines and has not been shown to improve survival or other patient centered outcomes. Study Design: Multilevel regression analyses were used to evaluate the influence of patients' region of residence and radiation oncologists' practice affiliations with medical schools on the combined use of EBRT and BT on patients within 6 months of diagnosis. Population Studied: Using the Surveillance, Epidemiology and End Results Medicare linked database and the American Medical Association Physician Masterfile, we conducted a retrospective cohort study of 4,479 patients aged 66 years or older who were diagnosed between 2004 and 2007 with low-risk localized prostate cancer, and the 401 radiation oncologists who saw them. Principal Findings: Overall, 231 (5.2%) patients received combined EBRT and BT. After adjusting for patient, tumor and radiation oncologist characteristics, patients who saw radiation oncologists with no practice affiliation with medical schools were significantly more likely to receive combined EBRT and BT (odds ratio [OR], 3.14; 95% confidence interval [95% CI], 1.50-6.59, p=0.003). In addition, regional variations were observed; the odds of receiving combined therapy for patients residing in California (OR, 0.1; 95% CI, 0.03-0.33, p<0.0001) were significantly less than those residing in Georgia (OR, 1.0; referent). Conclusions: Low-risk localized prostate cancer patients residing in Georgia were significantly more likely to receive combined EBRT and BT when compared to patients in other SEER Regions. Radiation oncologists without practice affiliations with medical schools were significantly more likely to treat patients with combined EBRT and BT; such treatment patterns are not consistent with patient-centered clinical guidelines and unlikely to have significant survival benefit. Implications for Policy and Practice: In addition to increased health care spending, patients who receive combined radiation therapy for localized prostate cancer have been previously shown to suffer from a worse overall quality-of-life compared to those not receiving this combined treatment. The significant associations found in this study provide additional evidence for clinicians and policy makers regarding areas to target to reduce the overtreatment of low-risk localized prostate cancer patients and increase adherence to evidence-based guidelines.
    141st APHA Annual Meeting and Exposition 2013; 11/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: Research Objective: Physician academic affiliations and changes in drug reimbursement rates have been shown to influence physician practice patterns regardless of clinical guidelines, patient clinical or sociodemographic factors. We examined the association between urologists' practice affiliations with medical schools and the utilization of medical androgen deprivation therapy (ADT) before and after reductions in medical ADT reimbursement rates resulting from the 2003 Medicare Modernization Act (MMA). Study Design: Multilevel regression analyses were used to evaluate the influence of urologists' practice affiliations with medical schools on primary medical ADT use on patients within 6 months of diagnosis a treatment regimen that is at variance with clinical guidelines and has not been shown to improve survival or other patient-centered outcomes. Population Studied: Using the Surveillance, Epidemiology and End Results Medicare linked database and the American Medical Association Physician Masterfile, we conducted a retrospective cohort study of 10,301 patients aged 66 years or older who were diagnosed between 2003 and 2005 with localized, low-to-intermediate grade prostate cancer, and the 1,577 urologists who saw them. Principal Findings: Overall, 3,763 (37%) patients received medical ADT. After adjusting for patient, tumor and urologist characteristics, patients who saw urologists with no practice affiliation with medical schools were significantly more likely to receive medical ADT (odds ratio [OR], 2.03; 95% confidence interval [95% CI], 1.57-2.63). Compared to 2003, when the MMA went into effect, the odds of receiving medical ADT were significantly lower in 2004 (OR, 0.76; 95% CI, 0.68-0.85) and 2005 (OR, 0.51; 95% CI, 0.45-0.57). Conclusions: Even though the overall odds of patients receiving unnecessary medical ADT decreased after the MMA reimbursement reduction, urologists without practice affiliations with medical schools were still significantly more likely to prescribe medical ADT; such treatment patterns are not consistent with patient-centered clinical guidelines and unlikely to have significant survival benefit. Implications for Policy, Delivery and Practice: The significant associations found in this study between urologists' practice affiliations with medical schools and the utilization of medical ADT provides further insights into what efforts may be successful in reducing overtreatment of localized prostate cancer patients with primary medical ADT following Medicare reimbursement reductions. This study also provides additional evidence for clinicians and policy makers regarding factors including, physician reimbursement, that may influence adherence to evidence-based guidelines.
    141st APHA Annual Meeting and Exposition 2013; 11/2013
  • Ruben G. W. Quek, Kenneth M. Portier
    [Show abstract] [Hide abstract]
    ABSTRACT: Description: Across the US and Puerto Rico, the American Cancer Society (ACS) provides free accommodation, a full spectrum of supportive services and activities at the 31 Hope Lodges (HL) for cancer patients and caregivers who seek treatment at medical facilities forty or more miles away from their homes. Issues: The financial commitment borne by ACS to operate the HL program is substantial. A return on investment (ROI) evaluation was conducted to identify all the components associated with the returns and investments associated with the HL programs in order to inform future ACS HL program expansion and in defining partnership strategies with external providers. The HL programs in Atlanta, Boston and New York City were chosen as part of this ROI evaluation. Results: The ROI evaluation demonstrated that all three HL programs yielded positive returns for 2011. For every 1 USD invested in the HL programs, 1.21 to 1.97 USD was returned in accommodations, donations, volunteer services, and activities. Recommendations: The methodology used in this evaluation can be extended for all existing HL programs to provide a more complete understanding on the return ACS receives for the HL investment. Understanding how the variability of ROIs across other HL is associated with differences in geographic locality and HL characteristics will assist ACS in assessing future investments and HL locations. In addition, the ROI evaluation can be extended to other non-tangible returns like patients' and caregivers' psychosocial benefits and ACS brand value achieved through the HL program.
    141st APHA Annual Meeting and Exposition 2013; 11/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: We examine online peer-to-peer cancer community discussion boards to learn about issues of importance to people with cancer and cancer caregivers. The ACS Cancer Survivors Network(SM)(reference CSN), launched in 2000, is the oldest and largest online peer support community for cancer survivors and caregivers with over 160,000 registered members and 85,063 discussion board posts between 2008 and 2012. Text from forum posts are processed to support topic model analysis based on the assumption that each post is associated with one or more underlying latent topics. A Bayesian estimation algorithm is used to discover these latent topics and assign to each post posterior probabilities of it being related to each topic. Practical issues concerning the use and calibration of topic models are discussed as well as insight gained about the optimal number of topic classes. Topic models are applied to initiating posts from the CSN breast cancer and colorectal cancer discussion forums. The two most frequent topics initiated in the breast cancer forum are decisions after treatment (7.7%) and surgery/mastectomy/reconstruction decisions (6.4%). The most frequent topics initiated in the colorectal cancer forum were drugs used in colon cancer treatment (6.3%) and lung scan results (6.4%). Changes in topics over time and the entropy of topic distributions are also discussed.
    141st APHA Annual Meeting and Exposition 2013; 11/2013
  • International journal of radiation oncology, biology, physics 10/2013; 87(2):S485-S486. · 4.59 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Sentiment analysis has been widely researched in the domain of online review sites with the aim of getting summarized opinions of product users about different aspects of the products. However, there has been little work focusing on identifying the polarity of sentiments expressed by users in online health communities such as cancer support forums, etc. Online health communities act as a medium through which people share their health concerns with fellow members of the community and get social support. Identifying sentiments expressed by members in a health community can be helpful in understanding dynamics of the community such as dominant health issues, emotional impacts of interactions on members, etc. In this work, we perform sentiment classification of user posts in an online cancer support community (Cancer Survivors Network). We use Domain-dependent and Domain-independent sentiment features as the two complementary views of a post and use them for post classification in a semi-supervised setting using the co-training algorithm. Experimental results demonstrate effectiveness of our methods.
    08/2013;
  • [Show abstract] [Hide abstract]
    ABSTRACT: Prostate cancer treatment choices have been shown to vary by physician and patient characteristics. For patients with low-risk, clinically localized prostate cancer, the authors examined the impact of their clinical, sociodemographic, and radiation oncologists' (RO) characteristics on the likelihood that the patients would receive combined external beam radiotherapy and brachytherapy, a treatment regimen that is at variance with clinical guidelines. The Surveillance, Epidemiology and End Results (SEER)-Medicare linked database and the American Medical Association Physician Masterfile were used in a retrospective analysis of 5531 patients with low-risk, clinically localized prostate cancer who were diagnosed between 2004 and 2007, and the 708 ROs who treated them. Hierarchical logistic regression analyses were used to evaluate the relationship between patient and RO characteristics and the use of combined therapy within 6 months of diagnosis. Overall, 356 patients (6.4%) received combined therapy. Nonclinical factors were found to be associated with combined therapy. After adjusting for patient and RO characteristics, the odds of receiving combined therapy for patients residing in Georgia were found to be significantly greater than for all other SEER regions. Black patients were significantly less likely to receive combined therapy (odds ratio, 0.62; 95% confidence interval, 0.40-0.96 [P = .03]) compared with white patients. In addition, ROs accounted for 36.6% of the variation in patients receiving combined therapy. Geographic and sociodemographic factors were found to be significantly associated with guideline-discordant combined therapy for patients diagnosed with low-risk, clinically localized prostate cancer. Which RO a patient consults is important in determining whether they receive combined therapy. Cancer 2013. © 2013 American Cancer Society.
    Cancer 07/2013; · 5.20 Impact Factor
  • Value in Health 05/2013; 16(3):A154. · 2.19 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Online Health Communities is a major source for patients and their family members in the process of gathering information and seeking social support. The American Cancer Society Cancer Survivors Network has many users and presents a large number of users' interactions with regards to coping with cancer. Sentiment analysis is an important step in understanding participants' needs and concerns and the impact of users' responses on other members. We present an automated approach for sentiment analysis in an online cancer survivor community and compare it with a previous sentiment analysis approach. Both approaches are machine learning based and are tested on the same dataset. However, this work uses features derived from a dynamic sentiment lexicon, whereas the previous work uses a general sentiment lexicon to extract features. Tested on several classifiers, with only six features (versus thirteen), our results show 2.3% improvement on average, in terms of accuracy, and greater improvement in F-measure and AUC. An additional experiment was conducted that showed a positive impact of dimensionality reduction by extracting abstract features, instead of using term frequency (TF) vector space as attribute values.
    Social Intelligence and Technology (SOCIETY), 2013 International Conference on; 01/2013
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Online cancer communities help members support one another, provide new perspectives about living with cancer, normalize experiences, and reduce isolation. The American Cancer Society's 166000-member Cancer Survivors Network (CSN) is the largest online peer support community for cancer patients, survivors, and caregivers. Sentiment analysis and topic modeling were applied to CSN breast and colorectal cancer discussion posts from 2005 to 2010 to examine how sentiment change of thread initiators, a measure of social support, varies by discussion topic. The support provided in CSN is highest for medical, lifestyle, and treatment issues. Threads related to 1) treatments and side effects, surgery, mastectomy and reconstruction, and decision making for breast cancer, 2) lung scans, and 3) treatment drugs in colon cancer initiate with high negative sentiment and produce high average sentiment change. Using text mining tools to assess sentiment, sentiment change, and thread topics provides new insights that community managers can use to facilitate member interactions and enhance support outcomes.
    JNCI Monographs 01/2013; 2013(47):195-8.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: What characterizes influential users in online health communities (OHCs)? We hypothesize that (1) the emotional support received by OHC members can be assessed from their sentiment ex-pressed in online interactions, and (2) such assessments can help to identify influential OHC members. Through text mining and sentiment analysis of users' online interactions, we propose a novel metric that directly measures a user's ability to affect the sentiment of others. Using dataset from an OHC, we demonstrate that this metric is highly effective in identifying influential users. In addition, combining the metric with other traditional measures further improves the identification of influential users. This study can facilitate online community management and advance our understanding of social influence in OHCs.
    Journal of the American Medical Informatics Association 11/2012; · 3.57 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This study examined racial/ethnic differences in spiritual well-being (SWB) among survivors of cancer. We hypothesized higher levels of Peace and Faith, but not Meaning, among Black and Hispanic survivors compared to White survivors, differences that would be reduced but remain significant after controlling for sociodemographic and medical factors. Hypotheses were tested with data from the American Cancer Society's Study of Cancer Survivors-II. The FACIT-Sp subscale scores, Meaning, Peace, and Faith assessed SWB, and the SF-36 Physical Component Summary measured functional status. In general, bivariate models supported our initial hypotheses. After adjustment for sociodemographic and medical factors, however, Blacks had higher scores on both Meaning and Peace compared to Hispanics and Whites, and Hispanics' scores on Peace were higher than Whites' scores. In contrast, sociodemographic and medical factors had weak associations with Faith scores. The pattern with Faith in bivariate models persisted in the fully adjusted models. Racial/ethnic differences in Meaning and in Peace, important dimensions of SWB, were even stronger after controlling for sociodemographic and medical factors. However, racial/ethnic differences in Faith appeared to remain stable. Further research is needed to determine if racial/ethnic differences in SWB are related to variations in quality of life in survivors of cancer.
    Journal of Behavioral Medicine 07/2012; · 3.10 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The current study was undertaken to evaluate the spatiotemporal projection models applied by the American Cancer Society to predict the number of new cancer cases. Adaptations of a model that has been used since 2007 were evaluated. Modeling is conducted in 3 steps. In step I, ecologic predictors of spatiotemporal variation are used to estimate age-specific incidence counts for every county in the country, providing an estimate even in those areas that are missing data for specific years. Step II adjusts the step I estimates for reporting delays. In step III, the delay-adjusted predictions are projected 4 years ahead to the current calendar year. Adaptations of the original model include updating covariates and evaluating alternative projection methods. Residual analysis and evaluation of 5 temporal projection methods were conducted. The differences between the spatiotemporal model-estimated case counts and the observed case counts for 2007 were < 1%. After delays in reporting of cases were considered, the difference was 2.5% for women and 3.3% for men. Residual analysis indicated no significant pattern that suggested the need for additional covariates. The vector autoregressive model was identified as the best temporal projection method. The current spatiotemporal prediction model is adequate to provide reasonable estimates of case counts. To project the estimated case counts ahead 4 years, the vector autoregressive model is recommended to be the best temporal projection method for producing estimates closest to the observed case counts.
    Cancer 02/2012; 118(4):1100-9. · 5.20 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: A study was undertaken to evaluate the temporal projection methods that are applied by the American Cancer Society to predict 4-year-ahead projections. Cancer mortality data recorded in each year from 1969 through 2007 for the United States overall and for each state from the National Center for Health Statistics was obtained. Based on the mortality data through 2000, 2001, 2002, and 2003, Projections were made 4 years ahead to estimate the expected number of cancer deaths in 2004, 2005, 2006, 2007, respectively, in the United States and in each state, using 5 projection methods. These predictive estimates were compared to the observed number of deaths that occurred for all cancers combined and 47 cancer sites at the national level, and 21 cancer sites at the state level. Among the models that were compared, the joinpoint regression model with modified Bayesian information criterion selection produced estimates that are closest to the actual number of deaths. Overall, results show the 4-year-ahead projection has larger error than 3-year-ahead projection of death counts when the same method is used. However, 4-year-ahead projection from the new method performed better than the 3-year-ahead projection from the current state-space method. The Joinpoint method with modified Bayesian information criterion model has the smallest error of all the models considered for 4-year-ahead projection of cancer deaths to the current year for the United States overall and for each state. This method will be used by the American Cancer Society to project the number of cancer deaths starting in 2012.
    Cancer 02/2012; 118(4):1091-9. · 5.20 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Many users join online health communities (OHC) to obtain information and seek social support. Understanding the emotional impacts of participation on patients and their informal caregivers is important for OHC managers. Ethnographical observations, interviews, and questionnaires have reported benefits from online health communities, but these approaches are too costly to adopt for large-scale analyses of emotional impacts. A computational approach using machine learning and text mining techniques is demonstrated using data from the American Cancer Society Cancer Survivors Network (CSN), an online forum of nearly a half million posts. This approach automatically estimates the sentiment of forum posts, discovers sentiment change patterns in CSN members, and allows investigation of factors that affect the sentiment change. This first study of sentiment benefits and dynamics in a large-scale health-related electronic community finds that an estimated 75\%--85\% of CSN forum participants change their sentiment in a positive direction through online interactions with other community members. Two new features, \textit{Name} and \textit{Slang}, not previously used in sentiment analysis, facilitate identifying positive sentiment in posts. This work establishes foundational concepts for further studies of sentiment impact of OHC participation and provides insight useful for the design of new OHC's or enhancement of existing OHCs in providing better emotional support to their members.
    PASSAT/SocialCom 2011, Privacy, Security, Risk and Trust (PASSAT), 2011 IEEE Third International Conference on and 2011 IEEE Third International Confernece on Social Computing (SocialCom), Boston, MA, USA, 9-11 Oct., 2011; 01/2011
  • 21st Annual Workshop on Information Technologies and Systems (WITS'11); 01/2011
  • [Show abstract] [Hide abstract]
    ABSTRACT: Recent confirmatory factor analysis (CFA) of the Functional Assessment of Chronic Illness Therapy-Spiritual Well-Being (FACIT-Sp) Scale in a sample of predominantly white women demonstrated that three factors, Meaning, Peace, and Faith, represented a psychometric improvement over the original 2-factor model. The present study tested these findings in a more diverse sample, assessed the stability of the model across racial/ethnic groups, and tested the contribution of a new item. In a study by the American Cancer Society, 8805 cancer survivors provided responses on the FACIT-Sp, which we tested using CFA. A 3-factor model provided a better fit to the data than the 2-factor model in the sample as a whole and in the racial/ethnic subgroups (Deltachi(2), p<0.001, for all comparisons), but was not invariant across the groups. The model with equal parameters for racial/ethnic groups was a poorer fit to the data than a model that allowed these parameters to vary (Deltachi(2)(81)=2440.54, p<0.001), suggesting that items and their associated constructs might be understood differently across racial/ethnic groups. The new item improved the model fit and loaded on the Faith factor. The 3-factor model is likely to provide more specific information for studies in the field. In the construction of scales for use with diverse samples, researchers need to pay greater attention to racial/ethnic differences in interpretation of items.
    Psycho-Oncology 04/2009; 19(3):264-72. · 3.51 Impact Factor