Paula F Rosenbaum, PhD
- Associate Professor of Public Health and Preventive Medicine
Research Programs and Affiliations
- Cancer Research Program
- Center for Research and Evaluation (CRE)
- Public Health and Preventive Medicine
Education & Fellowships
- PhD: SUNY Buffalo, 1998, Epidemiology & Community Health
- MS: SUNY Buffalo, 1988, Epidemiology
- MS: Syracuse University, 1978, Library Science
Pediatric and adult cancers, woman's health, and the role of environmental exposures in reproductive health, endocrine diseases and immunologic disorders, including allergies and asthma.
Link to PubMed (Opens new window. Close the PubMed window to return to this page.)
Indoor mold and the risk of wheeze in the first year of life for infants at risk for asthma.*PF Rosenbaum, TM Hargrave, JL Abraham, JA Crawford, A Hunt, C Liu, G Hall, SE Anagnost, CM Catranis, AA Fernando, SR Morey, S Zhou, CJK Wang (SUNY Upstate Medical University Syracuse, NY).
The role of bioaerosols in the development of medically diagnosed wheeze was assessed in 103 inner-city infants as part of the Assessment of Urban Dwellings for Indoor Toxics Study and the Accurate Exposure Assessment of Bioaerosols in Indoor Environment Program. Infants born to mothers (6/01-9/02) with diagnosed asthma were followed quarterly for the first year of life.Environmental sampling included measurement of total viable fungi and bacteria. Colony forming units/m3 (CFU) were calculated from plate counts, with identification on microscopic examination. Unconditional logistic regression was used to model the associations of total CFUs, and the genera Aspergillus (ASP) or Penicillium (PEN) on ever wheeze with control for birth season (winter, spring, summer, fall), gender, insurance (private/Medicaid) and mother’s age at birth (years). ASP and PEN CFUs were categorized as non-detectable (reference), low-medium (> detection limit and < 75th percentile) and high (=>75th percentile). Adjusted odds ratios (AOR) and 95% confidence intervals (CI) were calculated.
The infants were 45% male, 50% nonwhite, with 82% on Medicaid. Mean maternal age (standard deviation) was 25.4 (5.8) years; 71% were unmarried. Total CFUs were not associated with wheeze. While ASP was significantly associated with wheeze in bivariate models, the AOR (95% CI) was 1.48 (0.49-4.45) for low-medium and 1.88 (0.56-6.31) for high exposure when modeled with demographics and birth season. High PEN remained significantly associated with wheeze in all models; the AOR (95% CI) for low-medium exposure was 1.72 (0.50-5.88) while for high, it was 5.99 (1.46-24.66). The findings demonstrate the importance of modeling each genus when assessing the association of airborne mold with wheeze. Funded by EPA R 82860501-0
Improvement in air quality parameters in intervention study homes of
asthmatic children. PF Rosenbaum, JL Abraham, G Siwinski, A Fernandez.
Minimization of exposure to household asthma triggers may improve the health of children with diagnosed asthma. We randomized 93 children to a low (LIH) or high (HIH) intensity 6-month intervention program. Both groups received asthma education while HIH participants also were given a home cleaning, a heat recovery ventilator (HRV) window unit, a room air cleaner (RAC) and vacuum cleaner. Home inspection at study start and end gathered data on property characteristics, appliances, ventilation and occupant smoking. HRVs and RACs were installed at baseline; verification of on/off status of devices was made at 6 months (FU). 24-hour air sampling utilized a multisensor AirAdvice™ (AA) monitor deployed in all homes for at least 7 days at baseline and FU. The monitor collected data on particulates (PM), total volatile organic compounds (TVOC), carbon dioxide (CO2), carbon monoxide (CO), temperature and relative humidity (RH).Overall means were calculated for each parameter at each monitoring period. Differences between HIH and LIH groups for home characteristics and air quality measures were ascertained at baseline and FU using frequencies, X2, t-tests or ANOVA. As not all HRVs or RACs were “on” at FU, analyses of AA parameters also were done grouping HIH OFF with LIH. Results indicated that baseline values for all AA parameters were similar in LIH and HIH. At FU, HIH had significantly (p<0.05) lower overall PM means than LIH (15.9 vs 28.1mg/m3). Within the HIH, significant declines between baseline and FU means were noted for PM, CO2 and CO; similar changes were not noted within LIH. Use of an HRV (ON vs OFF or LIH) at FU resulted in lower mean TVOC, PM, CO2 and RH (p<0.05); in addition, RAC use was associated with significantly lower mean PM levels. In conclusion, a battery of mitigation techniques resulted in lower mean levels of multiple air quality parameters in the homes of children with email@example.comFunding-HUD, NYSERDA, NYSTAR.
Prenatal and early childhood factors and the risk of childhood acute lymphoblastic leukemia . With GM Buck & ML Brecher
40 years, few definitive risk factors have been identified. It has been hypothesized that children with an abnormal pattern of infectious disease acquisition, particularly delays in the timing of infections, are at an increased risk of ALL (Greaves \& Chan, Brit J Haematol 1986;64:1-13). Other factors potentially associated with ALL under this hypothesis include breast feeding and exposure potential to infectious agents while in day care. Additionally, some studies have suggested that a history of allergy protects against cancer although limited data exist concerning allergy and childhood ALL. This population based case-control study evaluated the role of several prenatal exposures, maternal reproductive factors and early childhood events and illnesses in the development of childhood ALL. Cases (n=255) were \<15 years of age when diagnosed with ALL and resident in one of 31 counties in western and central New York. Controls (n=760) were a random sample of births from the same 31 county area and were frequency matched to cases by sex, race and birth year. Mailed questionnaires were completed by case and control parents and adjusted odds ratios (OR) and 95% confidence intervals (CI) were calculated using unconditional logistic regression.
Cases and controls were predominantly male (56%) and white (95%); 50% were breast-fed and 29% of mothers smoked during the index pregnancy. Illnesses reported before two years of age in cases and controls included: colds (85%), otitis media (57%), diarrhea (43%) and vomiting (29%); croup, influenza, bronchiolitis and pneumonia were reported in \<15% of children. Thirty-four percent of cases and 43% of controls reported allergies, defined as the presence of asthma, eczema, pollen-dust, dander or food-drug-bee sensitivities. Allergy data were divided into two periods, onset before or after the age at diagnostic bone marrow among cases and an equivalent date among controls. Logistic modeling indicated that diarrhea was associated with a reduced risk of ALL, with an adjusted OR (CI) of 0.75 (0.55-1.02). Early episodes of otitis media were suggestive of an increased risk; that OR (CI) was 1.25 (0.93-1.69). The other respiratory illnesses and vomiting were not associated with ALL. An adjusted OR (CI) of 0.58 (0.39-0.89) was observed for a positive history of any allergy prior to diagnosis of ALL. Prior pollen-dust or dander allergy and food-drug-bee sensitivities also were associated with a reduced risk of ALL. These data support a protective effect of prior allergy on the development of ALL while demonstrating little support for the hypothesis that delays in infection acquisition are associated with ALL.