Timothy L Lash

McGill University, Montréal, Quebec, Canada

Are you Timothy L Lash?

Claim your profile

Publications (223)1109.12 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: The association between stress and cancer incidence has been studied for more than seven decades. Despite plausible biological mechanisms and evidence from laboratory studies, findings from clinical research are conflicting. The objective of this study was to examine the association between PTSD and various cancer outcomes. This nation-wide cohort study included all Danish-born residents of Denmark from 1995 to 2011. The exposure was PTSD diagnoses (n = 4131). The main outcomes were cancer diagnoses including: (1) all malignant neoplasms; (2) hematologic malignancies; (3) immune-related cancers; (4) smoking- and alcohol-related cancers; (5) cancers at all other sites. Standardized incidence ratios (SIR) were calculated. Null associations were found between PTSD and nearly all cancer diagnoses examined, both overall [SIR for all cancers = 1.0, 95 % confidence interval (CI) = 0.88, 1.2] and in analyses stratified by gender, age, substance abuse history and time since PTSD diagnosis. This study is the most comprehensive examination to date of PTSD as a predictor of many cancer types. Our data show no evidence of an association between PTSD and cancer in this nationwide cohort.
    European Journal of Epidemiology 05/2015; DOI:10.1007/s10654-015-0032-7 · 5.15 Impact Factor
  • Timothy L Lash, Jay S Kaufman
    Epidemiology (Cambridge, Mass.) 05/2015; 26(4). DOI:10.1097/EDE.0000000000000318 · 6.18 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: It remains unknown whether incident chronic diseases are more often fatal among breast cancer survivors than among women free of breast cancer. We conducted a nationwide matched cohort study of all Danish breast cancer patients diagnosed between 1994 and 2007, who survived for five years. We compared their long-term mortality with five times as many women from the general population without breast cancer, matched on age. We used time-varying methods to compute mortality rate ratios (MRRs) for incident diseases included in the Charlson Comorbidity Index (CCI). One third of five-year breast cancer survivors developed incident diseases during 14years of follow-up, with about the same incidence as women without breast cancer. Mortality associated with any incident disease was similar among breast cancer survivors (MRR=7.1, 95% confidence interval (CI): 6.7, 7.4) and comparison women (MRR=7.5, 95% CI: 7.3, 7.7). Among breast cancer patients, relative mortality associated with incident diseases was higher among patients treated with chemotherapy (MRR=10, 95% CI: 8.7, 12) and radiotherapy (MRR=9.8, 95% CI: 8.8, 11) than among patients who received surgery (MRR=7.0, 95% CI: 6.7, 7.4) or hormonal therapy (MRR=6.3, 95% CI: 5.8, 6.9). There were no marked differences in mortality of diseases among breast cancer survivors and women from the general population. Among breast cancer patients, new diseases were more often fatal in patients treated with chemotherapy and radiotherapy. Five-year breast cancer survivors have similar risk of dying from new chronic medical conditions as women from the general population without breast cancer. Copyright © 2015 Elsevier Ltd. All rights reserved.
    European journal of cancer (Oxford, England: 1990) 03/2015; 47(6). DOI:10.1016/j.ejca.2015.02.001 · 4.82 Impact Factor
  • Timothy L Lash
    Epidemiology (Cambridge, Mass.) 03/2015; 26(2):141-2. DOI:10.1097/EDE.0000000000000250 · 6.18 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Autoimmune diseases (ADs) comprise a large group of heterogeneous diseases in which the immune system attacks healthy organs. Both intrinsic changes in the body and AD treatment can compromise immune function. Impaired immune function could increase the risk of recurrent cancer. We aimed to investigate this hypothesis in a population-based epidemiological study. We examined the risk of breast cancer (BC) recurrence associated with an AD diagnosis among patients with incident stages I–III BC diagnosed during 1980–2007. Data were obtained from Danish population-based medical registries. ADs were categorized dichotomously and according to organ system of origin. Follow-up was up to 10 years or until 31 December 2009. Multivariate Cox proportional hazard regression was used to compute hazard ratios (HRs) and associated 95 % confidence intervals (95 % CIs) to evaluate the association between AD diagnosis and BC recurrence. 78,095 women with stages I–III BC were identified. Median age-at-diagnosis was 61 years (19–102 years), median follow-up was 5.7 years, and 13,545 women had a recurrence during follow-up. 6,716 women had at least one AD. In adjusted models, the association between ADs and BC recurrence was near null: HRadjusted 0.96 (95 % CI 0.89, 1.04). These results held in all AD subcategories, except for central nervous/neuromuscular system ADs, with HRadjusted 0.56 (95 % CI 0.40, 0.78). Among women with BC, a history of at least one AD diagnosis was not associated with BC recurrence, with the possible exception of ADs of the central nervous/neuromuscular system.
    Breast Cancer Research and Treatment 01/2015; 149(2). DOI:10.1007/s10549-014-3258-2 · 4.20 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To assess the validity of reaction to severe stress and adjustment disorder diagnoses registered in the Danish Psychiatric Central Research Register (DPCRR), to examine the documentation of stressful and traumatic events in the medical records, and to investigate the occurrence of stress diagnoses among persons not registered in the DPCRR. Among 101,633 patients diagnosed with International Classification of Diseases, 10th Edition (ICD-10) F43 diagnoses between 1995 and 2011, we selected 50 patients from two hospitals (100 total), comprising one above and one below median age for each diagnosis for five time periods, and reviewed their medical records. We calculated the positive predictive value, comparing registration in the DPCRR with the original medical records, and captured data on stressful life events. Two general practitioners were queried about 50 patients without a stress diagnosis in the DPCRR, regarding whether they had ever received a stress diagnosis. The positive predictive value was 58% for acute stress reaction, 83% for posttraumatic stress disorder, 94% for adjustment disorder, 71% for other reactions to severe stress, and 68% for reaction to severe stress, unspecified. In 80% of the records, a stressful or traumatic event was noted. Of 100 patients without an F43 diagnosis in the DPCRR, seven had a stress diagnosis. The DPCRR represents a valid and comprehensive resource for research on reaction to severe stress and adjustment disorders, particularly for posttraumatic stress disorder and adjustment disorder.
    Clinical Epidemiology 01/2015; 7:235. DOI:10.2147/CLEP.S80514
  • [Show abstract] [Hide abstract]
    ABSTRACT: Birth certificates are an important source of pre-pregnancy body mass index (BMI) and gestational weight gain (GWG) data for surveillance and aetiologic studies, but little is known about their validity in twin pregnancies. Twins experience high rates of adverse perinatal outcomes that have been associated with BMI and GWG in singletons. Our objective was to evaluate the accuracy of birth certificate-derived pre-pregnancy BMI and GWG compared with medical record-derived data in a sample of 186 twin pregnancies at a teaching hospital in Pennsylvania (2003–2010). Twelve strata were created by simultaneous stratification on pre-pregnancy BMI (underweight, normal weight/overweight, obese class 1, obese classes 2 and 3) and GWG (<20th, 20–80th, >80th percentile). The agreement of birth certificate-derived pre-pregnancy BMI category with medical record BMI category was lowest among underweight mothers [75% (95% confidence interval 51–91%) ] and highest among normal/overweight [97% (90–99%) ] and obese classes 2 and 3 mothers [97% (85–99%) ]. Agreement for GWG category from the birth certificate varied from 57% (41–70%) for GWG >80th percentile to 80% (65–91%) and 82% (72–89%) for GWG <20th and 20th–80th percentiles, respectively. The misclassification of BMI and GWG was primarily due to error in pre-pregnancy weight rather than weight at delivery or height. Agreement proportions for twins were not meaningfully different from the proportions in a comparable sample of singleton pregnancies. These data suggest that birth certificate-based BMI and GWG data are prone to error in twin pregnancies. Those who use these data should conduct internal validation studies and adjust their results using bias analyses.
    Maternal and Child Nutrition 01/2015; DOI:10.1111/mcn.12160 · 2.97 Impact Factor
  • Timothy L. Lash
    Paediatric and Perinatal Epidemiology 01/2015; 29(1). DOI:10.1111/ppe.12167 · 2.81 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Several studies have shown that uncertainty about disease and fear of disease progression affects psychosocial adjustment and quality of life. The purpose of this study was to validate a Norwegian short version of the "The Mishel Uncertainty in Illness Scale" (SF-MUIS) and to examine the impact of uncertainty in illness in breast cancer patients. 209 patients in breast cancer treatment completed questionnaires for SF-MUIS, Hospital Anxiety and Depression Scale (HADS), the Functional Assessment of Cancer Therapy-Breast (FACT-ES), and eight questions concerning quality of the patient information provided (IQP). Relationship between scores on uncertainty in illness and anxiety, depression, social support, emotional well-being, the quality of patient information provided, and age were studied by multiple regression analyses. Ordinal coefficient alpha for the Norwegian version of SF-MUIS was 0.70. Scores on SF-MUIS correlated significantly with scores on HADS (P = 0.001), FACT-ES (P = 0.001), and IQP (P = 0.001) indicating good convergent validity. The patients reported a moderate degree of uncertainty in illness. However, those who had been diagnosed with breast cancer for a year, reported higher scores than those newly diagnosed (P=<0.0001). Information provided was the sole significant predictor of illness uncertainty (P=<0.0001). The results of the present study confirm that the Norwegian version of the SF-MUIS is a suitable tool for assessment of uncertainty in breast cancer patients, who reported a moderate degree of uncertainty in illness. Copyright © 2014 Elsevier Ltd. All rights reserved.
    European journal of oncology nursing: the official journal of European Oncology Nursing Society 12/2014; 19(2). DOI:10.1016/j.ejon.2014.10.009 · 1.79 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To examine the association between exacerbation frequency and mortality following an acute exacerbation of chronic obstructive pulmonary disease (AECOPD). Cohort study using medical databases. Northern Denmark. On 1 January 2005, we identified all patients with prevalent hospital-diagnosed chronic obstructive pulmonary disease (COPD) who had at least one AECOPD during 1 January 2005 to 31 December 2009. We followed patients from the first AECOPD during this period until death, emigration or 31 December 2009, whichever came first. We flagged all AECOPD events during follow-up and characterised each by the exacerbation frequency (0, 1, 2 or 3+) in the prior 12-month period. Using Cox regression, we computed 0-30-day and 31-365-day age-adjusted, sex-adjusted, and comorbidity-adjusted mortality rate ratios (MRRs) with 95% CIs entering exacerbation frequency as a time-varying exposure. We identified 16 647 eligible patients with prevalent COPD, of whom 6664 (40%) developed an AECOPD and were thus included in the study cohort. The 0-30-day MRRs were 0.97 (95% CI 0.80 to 1.18), 0.90 (95% CI 0.70 to 1.15) and 1.03 (95% CI 0.81 to 1.32) among patients with AECOPD with 1, 2 and 3+ AECOPDs versus no AECOPD within the past 12 months, respectively. The corresponding MRRs were 1.47 (95% CI 1.30 to 1.66), 1.89 (95% CI 1.59 to 2.25) and 1.59 (95% CI 1.23 to 2.05) for days 31-365. Among patients with AECOPD, one or more exacerbations in the previous year were not associated with 30-day mortality but were associated with an increased 31-365-day mortality. Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://group.bmj.com/group/rights-licensing/permissions.
    BMJ Open 12/2014; 4(12):e006720. DOI:10.1136/bmjopen-2014-006720 · 2.06 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Recent studies show an association between statin therapy and a reduced risk of heart failure among breast cancer survivors. Our goal was to evaluate whether statin therapy for prevention of cardiovascular disease (CVD) would ameliorate declines in left ventricular ejection fraction (LVEF) often observed during anthracycline-based chemotherapy (Anth-bC).Methods In 51 participants (33 women and 18 men; aged 48±2 years), we performed CV magnetic resonance (CMR) measurements of LVEF before and 6 months after initiation of Anth-bC for patients with breast cancer, leukemia, or lymphoma. Fourteen individuals received statin therapy, and 37 received no statin. MR image analysts were blinded to participant identifiers.ResultsThose receiving statins were older and often had diabetes (DM), hypertension (HTN), and hyperlipidemia (HLD). For those receiving statins, LVEF was 56.6±1.4% at baseline and 54.1±1.3% 6 months after initiating anthracycline (p=0.15). For those not receiving a statin, LVEF was 57.5±1.4% at baseline and decreased to 52.4±1.2% over a similar 6 month interval (p=0.0003). In a multivariable model accounting for age, sex, DM, HTN, HLD, and cumulative amount of anthracycline received, LVEF remained unchanged in participants receiving a statin (+ 1.1±2.6%) versus a -6.5±1.5% decline among those not receiving a statin (p=0.03).Conclusion In conclusion, these data highlight that individuals receiving statin therapy for prevention of CVD may experience less deterioration in LVEF upon early receipt of Anth-bC than individuals not receiving a statin. Further studies with large numbers of participants are warranted to determine if statins protect against LVEF decline in patients receiving Anth-bC.
    The Canadian journal of cardiology 11/2014; 31(3). DOI:10.1016/j.cjca.2014.11.020 · 3.94 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: -In a murine anthracycline-related cardiotoxicity model, increases in cardiovascular magnetic resonance (CMR) myocardial contrast-enhanced T1-weighted signal intensity are associated with myocellular injury and decreases in left ventricular ejection fraction (LVEF). We sought to determine if T1- and T2-weighted measures of signal intensity associate with decreases in LVEF in human subjects receiving potentially cardiotoxic chemotherapy.
    Circulation Cardiovascular Imaging 10/2014; DOI:10.1161/CIRCIMAGING.114.002217 · 6.75 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Colorectal cancer recurrences are difficult to ascertain accurately and efficiently. We developed and validated an algorithm to identify recurrences that uses Danish medical registries. The algorithm uses metastasis and chemotherapy codes in the Danish National Patient Registry and codes indicating cancer recurrence in the Danish Pathology Registry. We applied the algorithm to a cohort (n=21,246) of colorectal cancer patients diagnosed 2001–2011 and followed through 2012. In a cohort (n=355) of two groups of actively followed patients, we compared the imputed recurrence data with recurrences diagnosed by regular follow-up. We compared cumulative incidence curves of imputed recurrence in local and regional stage patients from the large cohort, and of imputed and diagnosed recurrences in the actively followed cohort. In the 355 members of the actively followed cohort, our algorithm correctly identified 60 of 63 recurrences (sensitivity = 95%; 95% CI 87%, 99%) and misclassified only 10 of 292 without recurrence (specificity = 97%; 95% CI 94%, 98%). Cumulative incidence curves showed that members of the large cohort with regional disease had much higher incidence of imputed recurrence than those with local disease. In the actively followed cohort, the cumulative incidence of recurrence overlapped substantially when recurrence was imputed by our algorithm or using the follow-up data. Despite some limitations regarding ambiguous pathology codes, our algorithm showed excellent performance against actively followed recurrence data, and the expected relation between recurrence risk and cancer stage. It can be used in the Danish registries and adapted to similar registries elsewhere. © 2014 Wiley Periodicals, Inc.
    International Journal of Cancer 10/2014; 136(9). DOI:10.1002/ijc.29267 · 5.01 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: System delay (delay from emergency medical service call to reperfusion with primary percutaneous coronary intervention [PPCI]) is acknowledged as a performance measure in ST-elevation myocardial infarction (STEMI), as shorter system delay is associated with lower mortality. It is unknown whether system delay also impacts ability to stay in the labor market. Therefore, the aim of the study was to evaluate whether system delay is associated with duration of absence from work or time to retirement from work among patients with STEMI treated with PPCI. We conducted a population-based cohort study including patients ≤67 years of age who were admitted with STEMI from January 1, 1999, to December 1, 2011 and treated with PPCI. Data were derived from Danish population-based registries. Only patients who were full- or part-time employed before their STEMI admission were included. Association between system delay and time to return to the labor market was analyzed using a competing-risk regression analysis. Association between system delay and time to retirement from work was analyzed using a Cox regression model. A total of 4,061 patients were included. Ninety-three percent returned to the labor market during 4 years of follow-up, and 41% retired during 8 years of follow-up. After adjustment, system delay >120 minutes was associated with reduced resumption of work (subhazard ratio 0.86, 95% confidence interval 0.81 to 0.92) and earlier retirement from work (hazard ratio 1.21, 95% confidence interval 1.08 to 1.36). In conclusion, system delay was associated with reduced work resumption and earlier retirement. This highlights the value of system delay as a performance measure in treating patients with STEMI. Copyright © 2014 Elsevier Inc. All rights reserved.
    The American Journal of Cardiology 09/2014; 114(12):1810-1816. DOI:10.1016/j.amjcard.2014.09.018 · 3.43 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Treatment with synthetic glucocorticoids (GCs) depresses the immune response and may therefore modify cancer outcomes. We investigated the association between GC use and breast cancer recurrence.
    Annals of Oncology 09/2014; DOI:10.1093/annonc/mdu453 · 6.58 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Much preclinical and epidemiological evidence supports the anticancer effects of statins. Epidemiological evidence does not suggest an association between statin use and reduced incidence of breast cancer, but does support a protective effect of statins-especially simvastatin-on breast cancer recurrence. Here, we argue that the existing evidence base is sufficient to justify a clinical trial of breast cancer adjuvant therapy with statins and we advocate for such a trial to be initiated without delay. If a protective effect of statins on breast cancer recurrence is supported by trial evidence, then the indications for a safe, well tolerated, and inexpensive treatment can be expanded to improve outcomes for breast cancer survivors. We discuss several trial design opportunities-including candidate predictive biomarkers of statin safety and efficacy-and offer solutions to the key challenges involved in the enrolment, follow-up, and analysis of such a trial.
    The Lancet Oncology 09/2014; 15(10):e461-8. DOI:10.1016/S1470-2045(14)70119-6 · 24.73 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Quantitative bias analysis serves several objectives in epidemiological research. First, it provides a quantitative estimate of the direction, magnitude and uncertainty arising from systematic errors. Second, the acts of identifying sources of systematic error, writing down models to quantify them, assigning values to the bias parameters and interpreting the results combat the human tendency towards overconfidence in research results, syntheses and critiques and the inferences that rest upon them. Finally, by suggesting aspects that dominate uncertainty in a particular research result or topic area, bias analysis can guide efficient allocation of sparse research resources. The fundamental methods of bias analyses have been known for decades, and there have been calls for more widespread use for nearly as long. There was a time when some believed that bias analyses were rarely undertaken because the methods were not widely known and because automated computing tools were not readily available to implement the methods. These shortcomings have been largely resolved. We must, therefore, contemplate other barriers to implementation. One possibility is that practitioners avoid the analyses because they lack confidence in the practice of bias analysis. The purpose of this paper is therefore to describe what we view as good practices for applying quantitative bias analysis to epidemiological data, directed towards those familiar with the methods. We focus on answering questions often posed to those of us who advocate incorporation of bias analysis methods into teaching and research. These include the following. When is bias analysis practical and productive? How does one select the biases that ought to be addressed? How does one select a method to model biases? How does one assign values to the parameters of a bias model? How does one present and interpret a bias analysis?. We hope that our guide to good practices for conducting and presenting bias analyses will encourage more widespread use of bias analysis to estimate the potential magnitude and direction of biases, as well as the uncertainty in estimates potentially influenced by the biases.
    International Journal of Epidemiology 07/2014; 43(6). DOI:10.1093/ije/dyu149 · 9.20 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Five-year breast cancer survivors, diagnosed after 65 years of age, may develop more incident comorbidities than similar populations free of cancer. We investigated whether older breast cancer survivors have a similar comorbidity burden 6-15 years after cancer diagnosis to matched women free of breast cancer at start of follow-up and whether incident comorbidities are associated with all-cause mortality. In this prospective cohort study, 1,361 older 5-year early-stage breast cancer survivors diagnosed between 1990 and 1994 and 1,361 age- and health system-matched women were followed for 10 years. Adjudicated medical record review captured prevalent and incident comorbidities during follow-up or until death as collected from the National Death Index. Older 5-year breast cancer survivors did not acquire incident comorbidities more often than matched women free of breast cancer in the subsequent 10 years [hazard ratio (HR) 1.0, 95 % confidence interval (95 % CI) 0.93, 1.1]. Adjusted for cohort membership, women with incident comorbidities had a higher mortality rate than those without incident comorbidities (HR 4.8, 95 % CI 4.1, 5.6). A breast cancer history continued to be a hazard for mortality 6-15 years after diagnosis (HR 1.3, 95 % CI 1.1, 1.4). We found that older breast cancer survivors who developed comorbidities had an increased all-cause mortality rate even after adjusting for age and prevalent comorbidity burden. Additionally, survivors acquire comorbidities at a rate similar to older women free of breast cancer. These results highlight the association between comorbidity burden and long-term mortality risk among older breast cancer survivors and their need for appropriate oncology and primary care follow-up.
    Breast Cancer Research and Treatment 06/2014; 146(2). DOI:10.1007/s10549-014-3021-8 · 4.20 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Objectives To assess the interaction between comorbidity and breast cancer (BC) on the rate of venous thromboembolism (VTE) beyond what can be explained by the independent effects of BC and comorbidity. Design Population-based matched cohort study. Setting Denmark. Participants Danish patients with BC (n=62 376) diagnosed in 1995–2010 and a comparison cohort of women without BC (n=304 803) from the general population were matched to the patients with BC on year of birth in 5-year intervals and on the specific diseases included in the Charlson Comorbidity Index (CCI) and atrial fibrillation and obesity. Measures The rate ratios of VTE per 1000 person-years (PY) were computed by comorbidity levels using the CCI, and interaction contrasts (IC) were calculated as a measure of the excess or deficit VTE rate not explained by the independent effects of BC and comorbidity. Results Among patients with BC with a CCI score of 1, the 0–1 year VTE rate was 12/1000 PY, and interaction accounted for 10% of the rate (IC=3.2, 95% CI 0.5 to 5.9). Among patients with BC with CCI ≥4, the VTE rate was 17, and interaction accounted for 8% of the rate (IC=1.2, 95% CI −1.8 to 4.2). There was no interaction during 2–5 years of follow-up. Conclusions There was only little interaction between BC and the CCI score on the rate of VTE.
    BMJ Open 06/2014; 4(6):e005082. DOI:10.1136/bmjopen-2014-005082 · 2.06 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We created a registry of Danish-born citizens of Denmark with incident International Classification of Diseases (10th ed.; ICD-10) severe stress and adjustment disorder diagnoses between 1995 and 2011. A unique personal identifier was used to retrieve and merge data on demographic characteristics and diagnoses (ICD-10 codes F43.x). Here we report on the incidence of these disorders and the demographic characteristics of the subset of the Danish population who have received 1 of these diagnoses: 111,844 adults and children received a first diagnosis between 1995 and 2011. More women than men (60.1% vs. 39.9%) received a diagnosis. Diagnoses increased during the late teens through early 30s. Adjustment disorder was the most common diagnosis (65.7% of adults and 64% of children). Reaction to severe stress unspecified was the second most common (19.8% of adults and 23.8% of children), and there was a large increase in both, as well as acute stress reaction diagnoses, in 2007 (3,717-5,141, 1,248-2,520, and 348-1,024 in 2006 to 2007, respectively). Findings regarding gender and age of onset are similar to other westernized countries. This registry can be used for future research programs, contributing to the study of stress and trauma.
    Journal of Traumatic Stress 06/2014; 27(3):370-4. DOI:10.1002/jts.21926 · 2.72 Impact Factor

Publication Stats

4k Citations
1,109.12 Total Impact Points

Institutions

  • 2015
    • McGill University
      Montréal, Quebec, Canada
  • 2013–2015
    • Emory University
      • Department of Epidemiology
      Atlanta, Georgia, United States
  • 2014
    • Wake Forest School of Medicine
      • Division of Public Health Sciences
      Winston-Salem, North Carolina, United States
    • Georgia Department of Public Health
      Marietta, Georgia, United States
  • 2005–2014
    • Aarhus University Hospital
      • Department of Clinical Epidemiology
      Aarhus, Central Jutland, Denmark
  • 2012–2013
    • Aarhus University
      • Department of Clinical Epidemiology
      Aarhus, Central Jutland, Denmark
  • 1970–2012
    • Boston University
      • • Department of Medicine
      • • Department of Epidemiology
      • • School of Public Health
      Boston, Massachusetts, United States
  • 1999–2009
    • Boston Medical Center
      Boston, Massachusetts, United States
  • 2008
    • University of California, Los Angeles
      Los Ángeles, California, United States
  • 2007
    • Beverly Hospital, Boston MA
      Beverly, Massachusetts, United States
  • 1999–2007
    • University of Massachusetts Boston
      • Clinical Epidemiology Research and Training Unit
      Boston, Massachusetts, United States