QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Tjønneland, A. Articles by Olsen, J. H. Search for Related Content PubMed PubMed Citation Articles by Tjønneland, A. Articles by Olsen, J. H.

---

Nutrition and Cancer

Lifetime Alcohol Consumption and Postmenopausal Breast Cancer Rate in Denmark: a Prospective Cohort Study¹

Institute of Cancer Epidemiology, Danish Cancer Society, Copenhagen, Denmark and ^* Department of Clinical Epidemiology, Aalborg Hospital and Department of Epidemiology and Social Medicine, Aarhus University Hospital, Aarhus, Denmark

² To whom correspondence should be addressed. E-mail: annet{at}cancer.dk.

	ABSTRACT

TOP ABSTRACT SUBJECTS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
Alcohol intake may be one of the few modifiable risk factors for breast cancer. In a prospective cohort of 29,875 women with 423 cases of breast cancer during 1993–2000, we examined the relationship between postmenopausal breast cancer incidence rate and alcohol consumption in different life periods. When alcohol intake during four age ranges, twenties, thirties, forties and fifties was evaluated, only the intake in the fifties increased the risk of breast cancer [rate ratio (RR) = 1.12 (95% CI: 1.05–1.19)] per 10 g/d increase in alcohol intake. After adjustment for intake at study entry, this association was no longer present [RR = 1.01 (95% CI: 0.91–1.13)]. The cumulative lifetime alcohol intake, adjusted for recent intake, showed no association with postmenopausal breast cancer risk. Recent alcohol intake, adjusted for the alcohol intake in the other life time periods, showed a significant association of RR = 1.09 (95% CI: 1.00–1.18) per 10 g/d. There was no indication of a higher risk among women with early drinking start, nor did women who started to drink before their first birth have a higher risk than women who started to drink later in life. Our results suggest that baseline intake of alcohol is a more important determinant of postmenopausal breast cancer risk than earlier lifetime exposure.

KEY WORDS: • alcohol intake • lifetime • breast cancer

Findings reported during the last two decades provide consistent evidence that breast cancer risk shows a significant dose-response relationship with recent alcohol intake (1–3). Data on the influence of consumption earlier in life, age at drinking initiation and cumulative lifetime consumption have been more controversial. In a 1987 case-control study of both pre- and postmenopausal women, Harvey (4) postulated that breast cancer was associated with alcohol consumption of >13 g/d only for those who drank at younger ages (<30 y) and regardless of recent consumption, whereas no appreciable increase in risk was related to recent consumption of an equivalent amount among women with low-to-moderate intake of alcohol before age 30 y. In a nested, case-control study by Hiatt (5) among mainly postmenopausal women with a recent intake of 3 drinks/d, association with breast cancer appeared stronger among women who began drinking "the most" before age 30 y. In contrast, a newly published Canadian case-control study among postmenopausal women showed no discernible pattern of odds ratios in relation to the women’s age at commencement of alcohol consumption or intake in earlier life periods, but did find a suggested increase in association with breast cancer by weekly consumption of alcohol for >40 y (6). Relatively few data, mainly from case-control studies on age at drinking initiation, consumption in different lifetime periods and cumulative alcohol intake among postmenopausal women have been published (4–13). These studies have hypothesized that alcohol could influence risk through age-susceptible mechanisms. It has been argued by Colditz (14) that a reduction in breast cancer risk could be obtained if onset of drinking is delayed (15), but the overall evidence linking consumption of alcohol during specific periods of life to breast cancer is still inconsistent (16). Because most studies were case-control studies, information and recall-bias could not be excluded. In only two studies (5,12) was the information on drinking habits based on data collected before the onset of breast cancer. Only one of those studies collected information on alcohol drinking in the different age groups before age 30 y (12).

If recent alcohol intake increases breast cancer incidence, early exposure may lead to effect-modification such that the dose-response relationship between recent alcohol intake and risk of breast cancer may depend on the timing of initiation of alcohol consumption.

In the present prospective study, we investigated the influence of the timing of alcohol consumption. We related the postmenopausal breast cancer incidence rate to alcohol consumption in different life periods, and examined the timing of alcohol consumption initiation as a possible effect modifier for the dose-response relationship between recent alcohol consumption and the breast cancer rate previously reported for this prospective cohort of Danish women (17). The timing of the initiation of alcohol consumption was considered both in terms of chronological age and relative to first birth.

	SUBJECTS AND METHODS

TOP ABSTRACT SUBJECTS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
Between December 1993 and May 1997, all 79,729 women aged 50–64 y living in specific municipality-defined areas were invited to participate in the prospective study "Diet, Cancer and Health." To be eligible for invitation, women had to be born in Denmark and live in the greater Copenhagen or Aarhus areas; they could not be registered as having cancer in the Danish Cancer Registry. Subjects were identified by the unique, 10-digit identification number composed of date of birth and a four-digit running number, which is allocated to every Danish citizen by the Central Population Registry. A total of 29,875 women were enrolled in the study, corresponding to 37% of the women invited.

Informed consent was obtained from all participants, and "Diet, Cancer and Health" and the present substudy were approved by the regional Ethical Committees on Human Studies in Copenhagen and Aarhus, and by The Danish Data Protection Agency.

Participants completed a detailed, 192-item, FFQ, which they received by mail before a visit to one of the two study clinics. A description of the development and validation of the questionnaire was published previously (18,19). During the visit to the study clinic, participants also completed a lifestyle questionnaire, which included questions about reproductive factors, health status, social factors and lifestyle habits. The two self-administered questionnaires were processed by optical scanning and checked for missing information during the visit at the clinic, so that unclear information in the questionnaires could be clarified with the participant, preferably before she left the study clinic. A few missing data were allowed in the background questionnaire but not in the dietary questionnaire.

In the FFQ, recent intake of six types of alcoholic beverages over the preceding year was recorded as the mean frequency of intake of three types of beer in bottles (330 mL); wine in glasses (125 mL); fortified wine in drinks (60 mL); and spirits in drinks (30 mL). There were 12 categories of predefined responses, ranging from "never" to "8 or more drinks per day." Alcohol content was calculated as follows: one bottle of light beer, 8.9 g ethanol; one bottle of regular beer, 12.2 g ethanol; one bottle of strong beer, 17.5 g ethanol; one glass of wine, 12.2 g ethanol; one drink of fortified wine, 9.3 g ethanol; and one drink of spirits, 9.9 g ethanol. We did not differentiate between red and white wine.

In the lifestyle questionnaire, participants were asked about their mean alcohol consumption during four different periods of life, i.e., their twenties, thirties, forties and from age 50 until 1 y before study entry. Alcohol intake was reported as the number of drinks/wk of wine, fortified wine, beer and spirits. Participants were asked to indicate periods in life when they abstained from drinking alcohol. On the basis of this information, a total lifetime consumption of alcohol was calculated.

The age at which the participant first started to drink at least 1 drink/mo for 6 mo was indicated within the following categories: before the age of 14 y; 14–16 y; 17–18 y; 19–20 y; 21–25 y; and >25 y of age. To evaluate drinking start in relation to first birth, three groups were formed: one with drinking start before first birth; one with drinking start after first birth; and one with drinking start approximately at the time of first birth.

From the lifestyle questionnaire, we obtained information about years of school education (short: 7 y, medium: 8–10 y, or long: 11 y); parity; age at first birth; history of benign breast tumor surgery (yes/no); use of hormone replacement therapy (HRT) at study entry (never, past or current); and duration of HRT in years. Furthermore, anthropometric measurements including height and weight were obtained by professional staff members at the study clinics. BMI was calculated as weight (kg) per height squared (m²).

A total of 326 women (1.1%), who later were reported in the Cancer Registry with a cancer before entry into the study, were excluded. Additionally, eight women were excluded from the main dataset because they did not want to complete major parts of the lifestyle questionnaire. Because analyses were restricted to postmenopausal women, a total of 4798 premenopausal women who had reported at least one natural menstruation no >12 mo before entry, and no use of HRT, were excluded. Nine women who reported a lifetime history of no menstruation, and 177 women (0.7%) with missing information about present or previous alcohol consumption (47 and 130 women, respectively) were also excluded. Thirty-seven women (0.16%) did not answer the questions about current use of HRT. From the resulting dataset of 24,734 assumed postmenopausal women, we also excluded 837 individuals (3.4%) who gave no information on reproductive events or length of schooling, leaving 23,683 postmenopausal women (95.8%) for study.

By use of their personal identification number, cohort members were linked to other registries. Information on vital status and migration was obtained from the Central Population Register. Information on cancer occurrence among cohort members was obtained through record-linkage to the Danish Cancer Registry, which collects information on all inhabitants with cancer in Denmark. Each cohort member was followed up for breast cancer occurrence from the date of entry (i.e., date of visit to the study center). Follow-up continued until diagnosis of any cancer (except for nonmelanoma skin cancer), date of death, date of emigration or 31 December 2000, whichever came first.

    Statistical analyses. Analyses of the relation between postmenopausal breast cancer incidence rate and exposure variables were based on the Cox Proportional Hazard model (including time-dependent variables), with age as the time axis to ensure that the estimated procedure was based on comparisons of individuals at the same age. Time-under-study was included as the time-dependent variable and was modeled by a linear spline (20) with a boundary at 1 y after entry into the cohort study. All models were adjusted for baseline values of known risk factors for breast cancer, such as parity (entered as two variables: the categorical variable parous/nulliparous, and the linear variable number of births), age at first birth (linear), history of benign breast tumor surgery (yes/no), length of school education (low, medium, high), use of HRT (never, past, current), duration of HRT in years (linear) and BMI (linear).

Each exposure period (twenties, the thirties, the forties, and from 50 until 1 y before study entry) and recent intake were considered separately in the initial analyses. Recent intake was defined as the average reported alcohol intake during the last year before enrollment. The pattern of the relationship between breast cancer rates and alcohol intake in the different periods was examined using linear splines in the Cox regression model. We used three boundaries placed at the quartiles among the cases (21). We found no significant, systematic departures from linearity such as inflections or U-forms; thus, the relationships between intake in the different exposure periods and breast cancer rate were adequately described by log-linear dose-response relations. Therefore, to optimize the power of the analyses and to facilitate relevant comparisons, results are reported as the estimated rate ratio (RR) corresponding to comparison of any two individuals/subjects with the same values for all adjusted variables, but with one unit difference in the exposure variable in question per unit. The unit for average intake was 10 g alcohol/d, and the unit for cumulated lifetime intake was drink-years, defined as the amount of alcohol corresponding to 1 drink/d for 1 y (3650 g).

The possible modifying effect of timing of initiation of alcohol consumption on the dose-response relation between recent alcohol intake and breast cancer rate was investigated by estimating different slopes corresponding to recent alcohol intake for different timings of the initiation. The statistical significance of indications of effect-modification was assessed by testing whether the different slopes could be considered equal.

All tests were based on the likelihood ratio test statistic. Two-sided 95% CI for the RR were calculated based on Wald’s test of the Cox regression parameter, that is, on the log RR scale. The SAS procedure PHREG was used for statistical analyses (release 6.12 for Unix; SAS Institute, Cary, NC).

	RESULTS

TOP ABSTRACT SUBJECTS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 
A total of 423 cases of breast cancer were identified among the 23,683 postmenopausal women during a median of 4.7 y of follow-up. The median age at entry was 57 y (range 50–65 y). The distribution of the known risk factors for breast cancer for the entire cohort and for those with breast cancer during follow-up differed in the expected directions (Table 1). More cases were nulliparous, had higher age at first birth, had previously had a benign breast tumor surgery, were more highly educated, had a higher prevalence of current use of HRT at baseline and had longer durations of HRT use.

We found no association between the breast cancer rate and the age decade below 50 y with the highest alcohol intake (Table 4). We investigated the risk of breast cancer according to age at drinking start (Table 5). The analysis showed an increasing risk with increasing age at drinking start, except for those women with drinking start after age 30 y. There was no indication of a higher risk among women with early drinking start. An insignificantly higher RR was shown for women who started to drink after the first birth [RR = 1.11 (95% CI: 0.84–1.46)] or within the same year as the birth of the first child [RR = 1.10 (95% CI: 0.81–1.49)] compared with women who started to drink before the first birth (Table 6). We also investigated whether the timing of drinking initiation in relation to first birth modified the dose-response relation between recent alcohol intake and breast cancer risk, but no effect modification could be shown (P for interaction = 0.24, results not shown).

	DISCUSSION

TOP ABSTRACT SUBJECTS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have no reason to believe that previous alcohol intake or start of alcohol drinking was subject to differential recall bias; because the data were collected in a prospective design, and because the reported level of consumption of wine, fortified wine, beer and spirits was almost identical for the cases and the total cohort during their twenties, thirties and forties, there may have been a certain amount of inaccuracy in re-porting intake in the distant past such that relatively small differences in intake could not be detected. Therefore, we could not exclude the possibility of nondifferential misclassification of alcohol intake at younger ages. If this was the case, it could have obscured a true association between breast cancer and alcohol intake at younger ages.

In the analyses, we controlled for a number of known risk factors for breast cancer, which did not change the overall results of our study. It is therefore not likely that our results could be explained by residual confounding by any of these variables. However, caution should be taken when generalizing the findings, because only 37% of the women invited participated in this study.

Some studies have indicated that women with an early maximum intake of alcohol may show a higher risk of later breast cancer (4,5). In a study by Hiatt et al. (5), alcohol data collected on 68,674 women during routine health care were used to study 303 women who subsequently developed breast cancer. The women were asked to estimate the time in their life when they drank "the most." For current drinkers of 3 drinks/d, women who reported drinking the most before age 30 y had a greater risk (RR = 2.6; 95% CI: 1.2–6.2) than women who reported drinking more in later periods in their lives (RR = 1.4; 95% CI: 0.7–2.8), both compared with those who were never drinkers. Among past drinkers, however, women who began drinking the most before age 30 y appeared to have no higher risk than those who began drinking after age 30 y. A more pronounced effect of early drinking than recent drinking was also found in other studies (7–9). When the women in our study were categorized according to the age period before age 50 y with the highest intake of alcohol, no increased risk for a maximum intake at early lifetime was shown. The inconsistent results could, in most of these studies, be explained by a rather low intake of alcohol in early lifetime periods compared with recent intake. Among women in our cohort, median reported intake of alcohol in the twenties was 3.5 g/d.

Our results on the importance of recent alcohol intake were supported by another prospective study by Holmberg (12) in which drinking later in life appeared to have a greater effect than did drinking earlier in life. Some retrospective studies also found that drinking later in life has a larger effect on breast cancer risk than drinking earlier in life. In a large case-control study among premenopausal women by Swanson et al. (22) a nonsignificant, increased risk of breast cancer among women, who drank more than 7 drinks/wk in their teens (OR 1.3; 95% CI: 0.7–2.6) could in large part be explained by recent intake.

One study found a higher risk with later start of drinking (13). When we related the age at start of regular alcohol consumption to the risk of breast cancer, a nonmonotonic, nonsignificant relationship to age at start was seen. The findings that early-age intake was related to later risk of breast cancer (4–9) could also be due to a cumulative effect of alcohol rather than an increased sensitivity for early age exposure. Lifetime average alcohol intake was associated with an increased risk in two case-control studies by Longnecker (10,11). In our study, a slightly increased risk was shown with cumulative intake; however, adjusting for recent intake reduced this association. None of the other prospectively designed studies (5,12) measured cumulative intake.

In a review by Colditz and Frazier (14), it was postulated that the time period before the first birth could be defined as the most crucial period in establishing future risk of breast cancer. A plausible explanation for a possible effect-modification by first full-time pregnancy was that genetic insults before first pregnancy are propagated by the breast tissue proliferation of pregnancy, whereas the differentiation of breast duct cells during pregnancy, as well as the altered rates of cell turnover after first pregnancy, may be protective against additional genetic insult. A recent study on smoking showed an increased risk of breast cancer for women who smoke before their first birth, compared with later (23). We were not able to show any increased risk of breast cancer for women who started drinking before first birth. We also investigated whether dose-response association between baseline alcohol consumption and breast cancer depended on timing of start of drinking relative to first birth. We found that women who started drinking before their first birth had a lower increase in risk with increasing recent alcohol intake, than women who started drinking after their first birth. We are not aware of any other study that has related the dose-response association between alcohol to the intake of alcohol in the crucial period before the first birth.

Our finding that early drinking start does not increase the later risk of postmenopausal breast cancer is in agreement with findings in a recent case-control study (6) in which age when initiating weekly drinking showed no clear pattern in the association with postmenopausal breast cancer risk, with ages 31–40 y having the highest risk estimate (RR = 1.6; 95% CI: 0.8–3.0) compared with those who were never drinkers. This finding was supported by Holmberg (12).

In conclusion, this study showed that recent intake of alcohol was the important predictor of risk of breast cancer associated with alcohol consumption, and the risk of breast cancer was related to reported drinking habits in early periods of life. These data, therefore, suggest that decreasing recent alcohol consumption, independent of early lifetime exposure to alcohol, may reduce the risk of breast cancer in postmenopausal women.

	FOOTNOTES

 
¹ Supported by grants from the Danish Cancer Society and "Europe against cancer": European Prospective Investigation into Cancer and Nutrition (EPIC).

Manuscript received 26 August 2003. Initial review completed 22 September 2003. Revision accepted 14 October 2003.

	LITERATURE CITED

TOP ABSTRACT SUBJECTS AND METHODS RESULTS DISCUSSION LITERATURE CITED

 

1. Longnecker, M. P. (1994) Alcoholic beverage consumption in relation to risk of breast cancer: meta-analysis and review. Cancer Causes Control 5:73-82.[Medline]

2. Smith-Warner, S. A., Spiegelman, D., Yaun, S. S., van den Brandt, P. A., Folsom, A. R., Goldbohm, R. A., Graham, S., Holmberg, L., Howe, G. R., Marshall, J. R., Miller, A. B., Potter, J. D., Speizer, F. E., Willett, W. C., Wolk, A. & Hunter, D. J. (1998) Alcohol and breast cancer in women: a pooled analysis of cohort studies. J. Am. Med. Assoc. 279:535-540.[Abstract/Free Full Text]

3. Ellison, R. C., Zhang, Y., McLennan, C. E. & Rothman, K. J. (2001) Exploring the relation of alcohol consumption to risk of breast cancer. Am. J. Epidemiol. 154:740-747.[Abstract/Free Full Text]

4. Harvey, E. B., Schairer, C., Brinton, L. A., Hoover, R. N. & Fraumeni, J. F., Jr (1987) Alcohol consumption and breast cancer. J. Natl. Cancer Inst. 78:657-661.

5. Hiatt, R. A., Klatsky, A. L. & Armstrong, M. A. (1988) Alcohol consumption and the risk of breast cancer in a prepaid health plan. Cancer Res. 48:2284-2287.[Abstract/Free Full Text]

6. Lenz, S. K., Goldberg, M. S., Labreche, F., Parent, M. E. & Valois, M. F. (2002) Association between alcohol consumption and postmenopausal breast cancer: results of a case-control study in Montreal, Quebec, Canada. Cancer Causes Control 13:701-710.[Medline]

7. Freudenheim, J. L., Marshall, J. R., Graham, S., Laughlin, R., Vena, J. E., Swanson, M., Ambrosone, C. & Nemoto, T. (1995) Lifetime alcohol consumption and risk of breast cancer. Nutr. Cancer 23:1-11.[Medline]

8. van’t Veer, P., Kok, F. J., Hermus, R. J. & Sturmans, F. (1989) Alcohol dose, frequency and age at first exposure in relation to the risk of breast cancer. Int. J. Epidemiol. 18:511-517.[Abstract/Free Full Text]

9. Young, T. B. (1989) A case-control study of breast cancer and alcohol consumption habits. Cancer 64:552-558.[Medline]

10. Longnecker, M. P., Newcomb, P. A., Mittendorf, R., Greenberg, E. R., Clapp, R. W., Bogdan, G. F., Baron, J., MacMahon, B. & Willett, W. C. (1995) Risk of breast cancer in relation to lifetime alcohol consumption. J. Natl. Cancer Inst. 87:923-929.[Abstract/Free Full Text]

11. Longnecker, M. P., Paganini-Hill, A. & Ross, R. K. (1995) Lifetime alcohol consumption and breast cancer risk among postmenopausal women in Los Angeles. Cancer Epidemiol. Biomark. Prev. 4:721-725.[Abstract]

12. Holmberg, L., Baron, J. A., Byers, T., Wolk, A., Ohlander, E. M., Zack, M. & Adami, H. O. (1995) Alcohol intake and breast cancer risk: effect of exposure from 15 years of age. Cancer Epidemiol. Biomark. Prev. 4:843-847.[Abstract]

13. Nasca, P. C., Baptiste, M. S., Field, N. A., Metzger, B. B., Black, M., Kwon, C. S. & Jacobson, H. (1990) An epidemiological case-control study of breast cancer and alcohol consumption. Int. J. Epidemiol. 19:532-538.[Abstract/Free Full Text]

14. Colditz, G. A. & Frazier, A. L. (1995) Models of breast cancer show that risk is set by events of early life: prevention efforts must shift focus. Cancer Epidemiol. Biomark. Prev. 4:567-571.[Abstract]

15. Stevens, R. G. & Hilakivi-Clarke, L. (2001) Alcohol exposure in utero and breast cancer risk later in life. Alcohol Alcohol. 36:276-277.[Free Full Text]

16. Singletary, K. W. & Gapstur, S. M. (2001) Alcohol and breast cancer: review of epidemiologic and experimental evidence and potential mechanisms. J. Am. Med. Assoc. 286:2143-2151.[Abstract/Free Full Text]

17. Tjønneland, A., Thomsen, BL., Stripp, C., Christensen, J., Overvad, K., Mellemkjaer, L., Gronbaek, M. & Olsen, J. H. (2003) Alcohol intake, drinking patterns and risk of postmenopausal breast cancer in Denmark: a prospective cohort study. Cancer Causes Control 14:277-284.[Medline]

18. Overvad, K., Tjønneland, A., Haraldsdottir, J., Ewertz, M. & Jensen, O. M. (1991) Development of a semiquantitative food frequency questionnaire to assess food, energy and nutrient intake in Denmark. Int. J. Epidemiol. 20:900-905.[Abstract/Free Full Text]

19. Tjønneland, A., Overvad, K., Haraldsdottir, J., Bang, S., Ewertz, M. & Jensen, O. M. (1991) Validation of a semiquantitative food frequency questionnaire developed in Denmark. Int. J. Epidemiol. 20:906-912.[Abstract/Free Full Text]

20. Greenland, S. (1995) Dose-response and trend analysis in epidemiology: alternatives to categorical analysis. Epidemiology 6:356-365.[Medline]

21. Greenland, S. (1995) Avoiding power loss associated with categorization and ordinal scores in dose-response and trend analysis. Epidemiology 6:450-454.[Medline]

22. Swanson, C. A., Coates, R. J., Malone, K. E., Gammon, M. D., Schoenberg, J. B., Brogan, D. J., McAdams, M., Potischman, N., Hoover, R. N. & Brinton, L. A. (1997) Alcohol consumption and breast cancer risk among women under age 45 years. Epidemiology 8:231-237.[Medline]

23. Band, P. R., Le, N. D., Fang, R. & Deschamps, M. (2002) Carcinogenic and endocrine disrupting effects of cigarette smoke and risk of breast cancer. Lancet 360:1044-1049.[Medline]

This article has been cited by other articles:

				U. Vogel, J. Christensen, B. A Nexo, H. Wallin, S. Friis, and A. Tjonneland Peroxisome profilerator-activated receptor{gamma}2 Pro12Ala, interaction with alcohol intake and NSAID use, in relation to risk of breast cancer in a prospective study of Danes Carcinogenesis, February 1, 2007; 28(2): 427 - 434. [Abstract] [Full Text] [PDF]

				G. Ravn-Haren, A. Olsen, A. Tjonneland, L. O. Dragsted, B. A. Nexo, H. Wallin, K. Overvad, O. Raaschou-Nielsen, and U. Vogel Associations between GPX1 Pro198Leu polymorphism, erythrocyte GPX activity, alcohol consumption and breast cancer risk in a prospective cohort study Carcinogenesis, April 1, 2006; 27(4): 820 - 825. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Tjønneland, A. Articles by Olsen, J. H. Search for Related Content PubMed PubMed Citation Articles by Tjønneland, A. Articles by Olsen, J. H.