Board logo

標題: [學科討論] biostatistics 求救 [打印本頁]

作者: goodkwong    時間: 2015-3-14 11:43 AM     標題: biostatistics 求救

Question 1Researchers compared the nutritional habits of athletes with stress fractures to athleteswithout stress fractures. They considered 25 nutritional factors measured with a foodfrequency questionnaire and 25 nutritional factors measured using 24-hour dietaryrecall. They reported that vitamin K (as measured on the food frequency questionnaireonly) was significantly lower (p=.02) in fractured athletes; and dietary fat (asmeasured by dietary recall only) was significantly higher (p=.03) in fractured athletesthan non-fractured athletes. What conclusion should we draw?
A. Low vitamin K intake and high dietary fat intake may be causally related tofracture in athletes.
B. Low vitamin K and high dietary fat intake are related to fractures in athletes,but the association may not be causal.
C. The significant associations observed between vitamin K and high dietary fatand fractures are most likely chance findings.
D. We should immediately begin supplementing athletes with vitamin K andrecommending that they consume lower fat diets based on the results of thisstudy.



Question 2
Researchers in Britain attempted to determine whether babies who are exposed to Xrayswhile in the womb have an increased risk of certain childhood cancers. The oddsratio for myeloid leukemia, comparing babies exposed to in utero X-rays versusbabies not exposed was: 2.44 (0.95 to 6.33), p>.05. What conclusions should wedraw?

A. In utero X-rays do not increase a baby’s risk of myeloid leukemia, since thisresult is not statistically significant.
B. Babies exposed to in utero X-rays may have as much as a 6-fold increased riskof myeloid leukemia.
C. In utero X-rays are safe.
D. In utero X-rays are harmful.

Question 3
Background: Recent research demonstrated that vitamin D, apart from calciumrelatedactions, has antiproliferative, prodifferentiative and immunomodulatoryactivities.

Objective: To determine whether actinic keratosis may benefit from theantiproliferative and prodifferentiative effects of tropical vitamin D.

Materials and Methods: The study was an investigator-blinded, half-sidecomparison trail. Patients applied calcipotriol cream to one side and Ultrabase creamas placebo to the other side of the scalp and/or face for 12 weeks. The total number ofactinic keratosis (AKs), diameters and total scores of the target lesions weredetermined at each visit.

Results: Nineteen patients were included, eighteen of whom completed the treatment.The mean total score of the target lesions reduced significantly at week 12 oncalcipotriol side; however, no significant reduction was found on placebo side(p=0.017 vs p=0.056). Although side effects were more common on calcipotriol side,the difference was not statistically significant.

Conclusion: Topical calcipotriol may show promise in the treatment of actinickeratosis. More studies are needed to confirm its efficacy.

What conclusion(s) might you infer from the results of this study?

A. Topical vitamin D (calcipotriol) was significantly better than the placebocream at reducing actinic keratosis.
B. Topical vitamin D (calcipotriol) was effective at reducing actinic keratosis,whereas the placebo cream was totally ineffective.
C. You should not infer any conclusions from a study with so few patients.
D. The difference between the creams may not be statistically significant.

Question 4
The following tables come from a 2006 study on breast cancer and coffee thatappeared in the Journal of Nutrition, entitled: “Consumption of Coffee, but Not BlackTea, is Associated with Decreased Risk of Premenopausal Breast Cancer”. J Nutr136: 166-171.Table 2. Adjusted Odds Ratios for premenopausal women (n=878) for the associationbetween regular coffee, decaffeinated coffee, or black tea consumption and risk ofbreast cancer.

Beverage consumption Adjusted OR (95% CI)
Regular coffee

None 1.0<1 cup/d 1.23 (0.73-2.07)
1 cup/d 0.95 (0.52-1.71)
2-3 cups/d 0.94 (0.65-1.39)
4+ cups/d 0.62 (0.39-0.98)
P for trend = 0.03

Decaffeinated coffee
None 1.0<1 cup/d 0.80 (0.54-1.19)
1 cup/d 1.05 (0.67-1.63)
2-3 cups/d 1.12 (0.071-1.76)
4+ cups/d 0.88 (0.51-1.52)
P for trend = 0.92
Black tea
None 1.0<1 cup/d 0.95 (0.68-1.35)
1 cup/d 0.94 (0.61-1.46)
2-3 cups/d 1.08 (0.73-1.65)
4+ cups/d 1.02 (0.56-1.86)
P for trend = 0.72

Table 3. Adjusted Odds Ratios for postmenopausal women (n=2949) for theassociation between regular coffee, decaffeinated coffee, or black tea consumptionand risk of breast cancer.Beverage consumption Adjusted OR (95% CI)

Regular coffee
None 1.0<1 cup/d 0.89 (0.69-1.15)
1 cup/d 0.93 (0.73-1.19)
2-3 cups/d 1.11 (0.92-1.34)
4+ cups/d 0.99 (0.79-1.23)
P for trend = 0.57

Decaffeinated coffee
None 1.0<1 cup/d 0.95 (0.76-1.18)
1 cup/d 0.92 (0.73-1.17)
2-3 cups/d 0.83 (0.67-1.01)
4+ cups/d 0.88 (0.65-1.20)
P for trend = 0.12

Black tea
None 1.0<1 cup/d 1.20 (0.97-1.48)
1 cup/d 1.03 (0.80-1.31)
2-3 cups/d 1.17 (0.92-1.49)
4+ cups/d 0.93 (0.67-1.29)
P for trend = 0.78

The authors also examined associations between coffee, black tea, and decaffeinatedcoffee within ductal and lobular histologic subtypes of breast cancer. They write inthe text of the paper that: “premenopausal women who consumed ≤1 cup of coffee/dshowed a significant increase in risk of lobular breast cancer (adjusted OR = 3.54,95% CI 1.17–10.7), although similar elevations were not noted for women whoconsumed ≥2 cups/d (OR = 1.84, 95% CI 0.65–5.19). In addition, when consumers ofany amount of black tea were compared with non-consumers of black tea, theobserved reduction in risk among premenopausal lobular types was also significant(OR = 0.52, 95% CI 0.28–0.97). In contrast, for postmenopausal women, there weregenerally no associations between any of the beverages and either ductal (1008 cases)or lobular carcinoma (181 cases; data not shown).”

The authors conclude that coffee consumption is associated with a decreased risk ofpremenopausal breast cancer. Do you think that the evidence supports theirconclusion? Why or why not?

A. Yes these data provide strong support that coffee consumption andpremenopausal breast cancer are linked.
B. Yes, these data provide modest evidence of a link between coffeeconsumption and premenopausal breast cancer.
C. No, the significant p-values likely represent chance findings.
D. No, there are no significant p-values so there is no evidence of an effect.

Quesiont 5:
As the situation in question 4,The observations in this study are correlated, Is it Ture or False?




歡迎光臨 小卒資訊論壇 (http://lsforum.net/board/) Powered by Discuz! 6.0.0