Chapter 9linked with increased life- satisfaction over the eight- year
period. When loneliness is removed from the model, the as-
sociation between changes from being married to divorced/
widowed and life- satisfaction becomes negative (although
remaining nonsignificant).
In the health domain, we find that people who developed
coronary heart disease or a new long- standing illness over
the eight- year period experienced a fall in life- satisfaction.
As in the levels analysis (Figure 9.2), individual chronic dis-
eases are not related to life- satisfaction. Functional changes
were also important, since people whose eyesight improved
show a smaller drop in life- satisfaction. Finally, in regard to
mental health, a decrease in depressive symptoms and an im-
provement in the sense of control between 2004 and 2010
predict an improvement in life- satisfaction by 2012, while
changes in cognitive function have no effect.
Overall, these dynamic processes in people’s lives account
for 12% of the variation in changes in life- satisfaction. As
in the cross- sectional analysis, we wanted to explore any im-
pact of age or gender on the observed associations. These are
described in online Tables A9.2 and A9.3 and reveal little
difference by age and gender. The main factors related to
change in life- satisfaction are changes in loneliness and pos-
itive support, improvement in eyesight, and less depressive
symptoms. Limiting long- standing illness was associated
with life- satisfaction only in men and in people under 65.
Conclusions
In this chapter we explored which factors influence well-
being at older ages. In part, our results confirm what is