Lecture 19 : Heart Disease 2 Perinatal And Early Life Hypotheses Overview

1
Lecture 19 Heart Disease (2) Perinatal And
Early Life HypothesesOverview

• FORSDAHLS HYPOTHESIS
• OTHER ECOLOGICAL STUDIES
• INDIVIDUAL LEVEL STUDIES
• DISCUSSION

2
Forsdahls Hypothesis

• Forsdahl found major variations in heart disease
  rates between Norwegian counties, but they did
  not relate to contemporary living conditions.
• They were, however, rated to patterns of infant
  mortality half a century previously.
• Infant mortality rates were now similar in most
  areas.
• As infant mortality tends to be closely related
  to living conditions, he concluded that heart
  disease rates were highest in areas which had
  fastest rates of improvements in living
  conditions.
• He hypothesised that people brought up in poverty
  had a reduced tolerance to certain types of fat.

3
Criticisms Of Forsdahl

• Elford et al. (1992) criticise Forsdahl
• Forsdahl provides no direct evidence that
  prosperity following poverty is a factor the
  initial poverty may be sufficient in itself.
• Forsdahl did not consider factors in adult life
  (e.g. smoking). These may have the same geography
  as past infant mortality.

4
Three Possibilities

• Taking these criticisms inot account, we are left
  with three possibilities
• Adult heart disease is a function of initial
  poverty followed by later prosperity (as
  hypothesised by Forsdahl)
• Adult heart disease is a function of childhood
  poverty, irrespective of whether it is followed
  by prosperity or not and
• Adult heart disease is a function of risk factors
  (such as smoking) in adult life - the orthodox
  view.

5
Other Ecological Studies (1)

• Buck and Simpson (1982) found infant mortality in
  US in 1917 was correlated with adult mortality in
  1961 and 1971, but that infant mortality in 1927
  was not. This provides partial support for
  hypotheses A and B.
• Williams et al. (1979) found heart disease in
  England and wales was correlated with past infant
  mortality, but that the pattern of infant
  mortlaity had not changed. This provides support
  for B, but not for A.
• Barker and Osmond (1986) found high correlations
  between past infant mortality and a variety of
  adult causes of death at present. This suggests
  that mortality may be related to deprivation and
  is consistent with hypotheses B and C.

6
Other Ecological Studies (2)

• Barker and Osmond (1987) found that adult
  mortality in three similar and neighbouring towns
  (Burnley, Colne and Nelson) varied and was
  similar to infant mortality patterns in the early
  1900s. This in turn was associated with living
  conditions and life histories of the mothers at
  the time. This lends support to hypothesis B.
• Osmond et al. (1987) examined 2 million death
  certificates. Although about half of the sample
  had moved since birth, they found that their
  place of birth was a better predictor of heart
  disease than current location. This lends more
  support to hypothesis B than C.

7
Other Ecological Studies (3)

• Barker and Osmond (1986) found adult Bronchitis
  mortality was strongly correlated with previous
  postneonatal mortality, stroke with neonatal
  mortality and heart disease with both. Neonatal
  mortality tends to be associated with foetal
  development / maternal health, whereas
  postneonatal mortality is associated with living
  conditions after birth (e.g. overcrowding). This
  suggests conditions around time of birth have a
  long-lasting impact (i.e. hypothesis B).

8
Individual Level Studies (1)

• Barker et al. (1989) tracked down 6,500 men on
  whom there were existing records taken by health
  visitors when they were infants. There was a
  strong relationship between weight at 12 months
  and heart disease mortality (but not other
  causes). There was also an association with other
  risk factors (e.g. systolic blood pressure). This
  suggests that factors which promote prenatal and
  postnatal growth may reduce deaths from heart
  disease.

9
(No Transcript)
10
Individual Level Studies (2)

• In a second study of 449 men and women in
  Preston, Barker at al. (1990) found that present
  day blood pressure was correlated with placental
  weight. A large placenta often indicates poor
  foetal nutrition. This suggests that the risks of
  adult heart disease may be programmed before
  birth.

11
Discussion

• Forsdahl suggested that childhood factors may be
  a more important determinant of adult heart
  disease rates than present day living conditions
  / lifestyles.
• Barker and his associates suggested the critical
  time was immediately after birth, but later work
  has tended to suggest conditions in the womb may
  be more critical.
• Dutch studies of survivors of the Dutch famine
  winter at the end of World War II suggests the
  critical period may be within the first few weeks
  after conception.