With data from the Swedish Women's Lifestyle and Health (WLH) and Norwegian Women and Cancer (NOWAC) cohorts, we were the first to prospectively study pigmentation factors and UV exposure in relation to risk of cutaneous melanoma (Veierød et al, 2003; Veierød et al., 2010). Our research in NOWAC, include studies of of sunhabits (Ghiasvand et al., 2015); solarium use and risk of melanoma (Ghiasvand et al., 2017) and squamous cell carcinoma (Veierød et al., 2014; Lergenmuller et al., 2019); sunscreen use and risk of melanoma  (Ghiasvand et al., 2016; Rueegg et al., 2018) and squamous cell carcinoma (Lergenmuller et al., 2022); pigmentation factors, UV exposure and melanoma risk on different body sites (Ghiasvand et al., 2019) and pigmentation factors and melanoma tumor thickness (Ghiasvand et al., 2021); UV exposure and DNA methylation in blood leukocytes (Page et al., 2020); reproductive factors, menopausal hormone therapy and melanoma risk (Støer et al., 2019, Støer et al., 2021); and physical activity and melanoma risk (Perrier et al., 2021).

In collaboration with the Cancer Registry of Norway, we have studied skin cancer incidence and mortality (Robsahm et al. 2015), prognostic factors after melanoma diagnosis (Helsing et al., 2016; Robsahm et al., 2018), prevalence of indoor tanning in teenagers (Robsahm et el., 2020), and clinical suspicion sensitivity of nodular and superficial spreading melanoma (Robsahm et al., 2021).

Our collaboration with the Cancer Registry of Norway, also includes research on exposure-related risks of cancer in petroleum workers and firefighters. After publishing the cohort profile of the Norwegian Offshore Petroleum Workers (NOPW) cohort explaining how exposures are recorded for cohort and case-cohort analyses (Stenehjem et al., 2021), we have investigated UV exposure and skin cancer risk (Liu et al., 2021) and night shift work and male (Liu et al., 2021) and female (Liu et al., 2022) breast cancer risk. As regards firefighters, we have documented work conditions and practices in Norwegian fire departments from 1950 until today (Jakobsen et al., 2020), investigated cancer incidence in the newly established Norwegian Fire Departments Cohort with 58 years of follow-up (Marjerrison et al., 2022), compared standardized incidence and mortality ratios (Marjerrison et al., 2022), and investigated age at diagnosis, prognostic factors and survival of prostate cancer in Norwegian firefighters and three other occupations 1960-2017 (Jakobsen et al., 2022).

In a collaboration with the Cancer Registry of Norway, a protocol was established for prospective studies of 25-hydroxyvitamin D, leptin and body mass index in relation to cutaneous melanoma incidence and survival (Stenehjem et al., 2017). Since then, we have published the results for anthropometric factors and melanoma risk (Stenehjem et al., 2018) and Breslow thickness (Stenehjem et al., 2018), vitamin D and melanoma risk (Stenehjem et al., 2020) and death (Stenehjem et al., 2022). In the same collaboration, a protocol for a population-based case-control study, merging nation-wide registries to study selected drugs and melanoma risk was established (Berge et al., 2019), and so far antidepressants (Berge et al., 2020) and immunomodulating drugs have been investigated (Berge et al., 2020). 

Measurement errors, misclassification and missing data are important sources of bias in observational studies. Thus, we studied measurement errors in self-reported melanoma risk factors (Veierød et al., 2008; Parr et al., 2009), compared methods for handling missing data in food frequency questionnaires (Parr et al., 2008), and also errors in important prognostic factors such as Breslow thickness of melanoma tumours (Veierød et al., 2018) and errors in breast cancer diameter (Tsuruda et al., 2020). Our interest in this area has also included methodological research on bias in Poisson regression coefficients (Veierød et al., 2001), clarification of under what conditions a naive analysis using a misclassified predictor will induce bias for the regression coefficients of other perfectly measured predictors in the model (Buonaccorsi et al., 2005) and power of the Cochrane-Armitage test for trend in the presence of misclassification (Buonaccorsi et al., 2014). 

In NOWAC, we have studied studied meat intake and cooking methods in relation to colorectal cancer risk (Parr et al. 2013). UV exposure, via its effects on vitamin D synthesis, might have beneficial effects on several conditions including cancer, and we studied potential positive aspects of UV exposure (Veierød et al., 2010; Edvardsen et al., 2011; Yang et al., 2011). We also studied the use of skin care products among Norwegian women (Aniasson et al., 2016) and the use of skin care products and risk of cancer of the breast and endometrium (Rylander et al., 2019).

In collaboration with the Norwegian Radiation and Nuclear Safety Authority, we studied  spectral UVB and UVA solarium irradiances and documented that these are 0.5-3.7 and 3-26 times, respectively, higher than from Oslo summer sun (Nilsen et al., 2008; Nilsen et al., 2011; Nilsen et al., 2012; Nilsen et al., 2016).

The properties of the gestational diet have become increasingly associated with consequences for the future health of the offspring. In the MoBa study, we have assessed dietary intake, as well as demographic profile and lifestyle factors, in relation to nausea and vomiting in pregnancy (Chortatos et al., 2013; Chortatos et al., 2015; Chortatos et al., 2018).

In collaboration with the Norwegian Institute of Public Health, we have studied 30-day mortality as a quality indicator for Norwegian hospitals (Kristoffersen et al, 2012), and statistical methods for identifying hospitals with low or high mortality (Kristoffersen et al., 2018).

The association between BMI and colon cancer is established. In the Norwegian County Study we have studied differences between genders and colon subsites, and also whether weight change influence colon cancer risk (Laake et al., 2010) and prognosis (Laake et al., 2016). In this cohort, we also studied trans fatty acid intake and both cancer risk and CVD mortality (Laake et al., 2012; Laake et al., 2013).

Oral glucose tolerance tests (OGTT) are important in medical care/research, and often studied by simple summary measures. We extracted information inherent in the shape of OGTT curves, compared it with the information from simple summary measures, and explored the clinical usefulness of such information (Frøslie et al., 2013; Frøslie et al., 2014). In this project, we found an unexpectedly increase in glucose values during a 7-year period, likely due to bias during long-term use of a glucose measurement system, and concluded that biologically implausible trends in data can be corrected by using independent control values (Frøslie et al., 2011). We also investigated the potential mediating roles of Interleukin 6 and Interleukin 1 Respetor anatagonist in the association between maternal body mass index and birth weight  (Frøslie et al., 2010), and revisited the general statistical advice to keep continuous exposure variables as continuous in the analyses (Frøslie et al., 2010).

In collaboration with National Institute of Public Health, we applied the Systematic Coronary Risk Evaluation (SCORE) risk model to Norwegian population surveys (Lindman et al., 2006; Lindman et al., 2007). Proper adjustments were required before implementation in clinical practice, thus we developed a cardiovascular disease (CVD) risk model, NORISK (Selmer et al., 2008), that was implemented in the Norwegian guidelines for CVD prevention (Norwegian Directorate of Health, 2009 (in Norwegian)) and available as a risk calculator (in Norwegian).

Published Feb. 24, 2011 8:39 PM - Last modified May 22, 2022 6:48 PM