Page 24 - IJPS-4-2

Trends in healthy life expectancy among older adults in South Africa

Japan also observed the effect of contextual factors, i.e., the “Golden plan” during the period of 1986–2004 (Yong and
Saito, 2009). During this period, several policy reforms were implemented which improved the health and well-being of
older people in Japan. As a result, a positive trend in HLE was observed (Yong and Saito, 2009). Another possibility for
improvement in HLE may be related to decrements in retirement age for men from age 65 down to age 60 in 2008–2010
as the receiving pension increases well-being (Schatz, Gomez-Olive, Ralston et al., 2012).
The finding that a greater improvement in HLE observed in women is accordance with a study from Austria
(Doblhammer and Kytir, 2001). It is possibly due to better access to health and social services over the study period by
women. The General Household Survey (Lehohla, 2013) showed that a higher proportion of women had better health-
seeking behaviors compared to men, which included consulting a health worker when they were ill or injured. Another
possible reason could be resilience in older women (Kinsel, 2005). There is evidence showing that older women are
more resilient than their male counterparts in dealing with daily difficulties because women tend to have a better social
supporting network than men (Depp and Jeste, 2006). Furthermore, as we reviewed above, self-rated health involves
a process of self-perceptions and subjective judgments about various health domains (Feng, Zhu, Zheng et al., 2016;
Jylhä, 2009); it is possible, during the study period, some health outcomes of women in South Africa improved greater
than those of men. Unfortunately, this study did not analyze and isolate the independent effects of possible predictors
of poor health. More research is clearly warranted to shed light on the root causes on the gender differential in the
improvement of HLE.
One interesting finding of the present study is a decline in life expectancy for both women and men and a decline in
HLE for men from 2005 to 2008, although these declines were not statistically significant. The declines in life expectancy
and HLE in the period of 2005–2008 were probably linked to the financial crisis in South Africa. As the 2008 survey
was conducted from May 2008 to March 2009 (Shisana, Rehle, Simbayi et al., 2009), it is thus possible that a sizeable
portion of the sample provided the responses during the financial crisis in South Africa and across the global. There is
evidence that the global financial crisis had a severe impact on South Africa, and consequently, the country’s economy
went into recession between 2008 and 2009 for the 1 time in 19 years (Rena and Msoni, 2014). During this period, the
st
unemployment rate and poverty levels increased dramatically (Padayachee, 2010). We speculate that this could have a
negative impact on HLE in South Africa, especially for men. Such a global and local economic crisis should also have
reshaped people’s daily life and behaviors and produced a negative impact on health-care access and health outcomes
in South Africa, and such an impact might be more fatal to the older adults than the younger adults as the former needs
more medical treatment and is more vulnerable. Fortunately, South African Government implemented a policy related to
retirement age decrement from age 65 to age 60 for men in 2008–2010, which may partially offset the negative impact of
financial crisis, especially for those aged 65 or younger.
Notwithstanding, although there were improvements over time in women, our results show that the proportion of
women reporting “fair’/’poor” health [Table 1] was higher than men, and the proportion of life spent in poor health
[Table 3] was higher for women than men in the study period. Studies are needed to investigate contextual determinants
to which the health gaps between older men and women can be attributed in the South African setting, including gender
roles. Health interventions need to be gender sensitive to address the health differences between the sexes.
One strength of this study is to use a nationally representative sample to investigate the trend of HLE among South
Africans aged 50 or older for the 1 time. The consistencies in survey design, data collection methods, and same
st
wording of questions across three waves in 2005, 2008, and 2012 make it possible to evaluate trends in the health of
older people over the period. The study period is unique, not only covering the global financial crisis but, also covering
the major events such as nation’s pension reform and hosting the 2010 FIFA World Cup. All these events could largely
influence people’s assessment of their emotion, spirit, psychological well-being, and eventually self-rated health. We
recommend studies that can disentangle this complexity tempo change of self-rated health in referring to contextual
environments.
Several caveats are worthy of mentioning when interpreting the results. First, it is possible to have concluded otherwise
had we used other measures, for example, disability. Such a measure is, however, not included in the SABSSM surveys’
questionnaires used in this analysis. A study on HLE in Thailand found stagnation when using self-rated health and
improvements in health status when using disability measures based on activities of daily living (ADLs) limitations
(Karcharnubarn and Rees, 2009). It has been acknowledged that self-rated health and disability can actually follow
different trends (Crimmins, 1996; Robine and Michel, 2004; Gu, Dupre, Warner et al., 2009; Spiers, Jagger, and Clarke,
1996). Furthermore, the cross-national and cross-cultural comparison of healthy expectancy based on self-rated health
should be caution due to its subjectivity of self-rated health. Nevertheless, our results are in line with findings from the
Rapid mortality surveillance (Dorrington, Bradshaw and Laubscher, 2014) which, based on mortality indicators over the

18 International Journal of Population Studies | 2018, Volume 4, Issue 2