Critical appraisal:Tang SW, Liu J, Juay L, Czene K, Miao H, Salim A, et al 2016

From Cancer Guidelines Wiki

Critical Appraisal

Article being appraised

Tang SW, Liu J, Juay L, Czene K, Miao H, Salim A, et al. Birth rates among male cancer survivors and mortality rates among their offspring: a population-based study from Sweden. BMC Cancer 2016 Mar 8;16:196 Available from:

Applicable clinical question

Key Facts

Study Design

cohort study

Study aims:

evaluates trends in birth rates among male cancer survivors and mortality rates of their offspring.

Number of Patients:


-84,752 men ≤70 years with a history of cancer and 126,696 offspring of men who had cancer from Multi-Generation Register(MGR), the Swedish Cancer Register, the Cause of Death Register, and the Migration Register (database contains more than 11 million individuals, included more than a million individuals with cancer diagnosed between 1958 and 2001).

-included men 16 to 70 years, born after 1931, who were alive between 1961 and 2002
-partners or spouse of these men were identified and tracked for up to 10 months after the men have passed away/
emigrated to ensure all information of their offspring were included
-end of follow up (31st December 2002)

Controls: 4,032,096 cancer-free men in the background population were matched to the male cancer survivors according to attained age and year of birth, and the number of live child births of the spouse of these men - 883,548 were excluded because outside of the observed age range of 16 to 70 years or had emigrated
Reported outcome(s):

number and dates of live child births (multiple pregnancies were counted as one event/birth) - included all births >1 year after diagnosis
relative birth rates as compared to the background population(Standardized Birth Ratios, SBRs)
compared risks of death to the background population(Standardized Mortality Ratio, SMRs)

Results of outcome(s):

-in survivors, 4973 (5.9 %) men had a total of 7910 births 1 year after Dx. Higher proportion who were Dx in childhood (19.5 %) and adolescence (22.8 %) fathered a child after the diagnosis, compared with men diagnosed in adulthood (5.2 %).
-Men with a history of cancer were 23% less likely to father a child compared to the background population(SBR 0.77, 95%CI 0.75–0.79) - remained fairly consistent over calendar time.
-Men Dx in childhood had a significantly lower birth rate (SBR 0.62, 95% CI 0.57–0.67) than men diagnosed in adulthood (SBR 0.80, 95 % CI 0.78–0.82) when compared to the background population.
-skin, thoracic cancer and head and neck had a birth rate similar to the background population (SBR 0.98 [0.92–1.04],0.92 [0.74–1.11] and 0.90 [95 % CI 0.80–1.01] respectively);
-prostate, brain and eye, and hematopoietic cancers had a significant decrease in their birth rates (SBR 0.24 [95 %
CI 0.06–0.54], 0.65 [0.62–0.69] and 0.66 [0.62–0.70] respectively) compared to the background population.
-Nulliparous men were significantly more likely to father a child after diagnosis (SBR 0.81, 95 % CI 0.79–0.83) compared to parous men (SBR 0.68, 95 % CI 0.66–0.74):
~pronounced in parous men Dx with prostate, thoracic and haematopoietic cancers (SBR 0.22 [0.04–0.53], 0.48 [0.40– 0.88] and 0.53 [0.46–0.61] respectively).
~men who were nulliparous at Dx with thoracic, skin, head and neck and digestive cancers had birth rates which were similar to controls (SBR 1.20 [0.94–1.49], 1.10 [1.01–1.18], 1.06 [0.92– 1.22], 0.97 [0.87–1.08] respectively).
~nulliparous men Dx with brain and eye, haematopoietic and reproductive organs cancer still had significantly
lower birth rates compared to controls (SBR 0.65 [0.60–0.69], 0.70 [0.65–0.75], 0.75 [0.68–0.77] respectively).
-Multivariate Poisson modelling: Being parous at diagnosis, a diagnosis of prostate cancer, older attained age (41–70 years), diagnosis in childhood and adolescence (0–12 and 13–18 years) and a recent diagnosis (1–5 and 6–10 years) were significant and independent predictors of a reduced probability of giving birth after diagnosis (adjusting for calendar period, socioeconomic status, and year of diagnosis showed that parity status at diagnosis, cancer site, age at diagnosis, time since diagnosis and attained age)
-Men diagnosed in adulthood, increased time since diagnosis, cancer survivors aged 26–35 and survivors of thoracic and head and neck cancers were most likely to father an offspring

Comments on results:

-Most of the men (95.7 %) were diagnosed with cancer in adulthood (19–70 years), with the commonest site being digestive tract (n = 14,817) and prostate (n = 11,281)
-Between 1961 and 2003, 126,696 children of the total 3,746,823 children were born to men who have had a
diagnosis of cancer at any point in their lives Of the 126,696 children born to men who have had a diagnosis of cancer, 2604(2.06 %) died during follow up. -> where 115,938 children were born pre-cancer Dx, 2707 born around the time of Dx and a further 8051 children born >1 year after Dx
-The overall mortality rate was similar to the background population(SMR of 1.00, 95 %CI 0.96–1.04) and was not
affected by the timing of their birth in relation to father’s cancer diagnosis.

Evidence ratings

Level of evidence


Risk of bias
High risk of bias Comments: Registry linkage data
Result of appraisal

Jutta's tick icon.png Included

Completed by

Dr Jessica Harris

Jutta's tick icon.png This appraisal has been completed.

Tang SW, Liu J, Juay L, Czene K, Miao H, Salim A, et al. Birth rates among male cancer survivors and mortality rates among their offspring: a population-based study from Sweden. BMC Cancer 2016 Mar 8;16:196 Available from:
Assigned to
Topic area
Guidelines:COSA:Cancer fertility preservation guidelines/Impact of cancer on fertility
Clinical question
Form:Critical appraisal

Section below only relevant for Cancer Council Project Officer

Edit appraisal assignment