Highly engaged Vitality members experience lower death rates than unengaged Vitality members (13-year comparative review)

Authors

Discovery Health's Lizelle Steenkamp (Senior Healthcare Statistician), Matt Zylstra (Senior Actuary) and Shirley Collie (Chief Research Actuary)

What we investigated

Vitality is a behaviour change programme that rewards its members for making healthier choices. As members engage with the programme, they go from unengaged to highly engaged.

In March 2023, we carried out our initial analysis on death rate trends over time for members of the Discovery Health Medical Scheme (DHMS).

In this new analysis we assess the pattern of death rates in members by Vitality engagement level, over a 13-year period. And, we look at the impact of COVID-19 on Vitality members again by engagement level.

Whose data is included in the analysis?

The analysis considers data from Discovery Health Medical Scheme members who are Vitality members. We focused on unengaged and highly engaged Vitality members.

Members of the Discovery Vitality programme earn points during the year through engaging in healthy behaviours, such as doing exercise and screening checks. These points are banked for the following year, so our analysis is done on a calendar year basis - January to December.

Each Vitality member has worked to get to a particular Vitality engagement level in a calendar year and their outcomes are assessed in the following calendar year.

How many Deaths were included in this analysis?

The total number of deaths recorded over the 13-year period looked at (2010-2022) was 26,369 for unengaged and 841 for highly engaged Vitality members. These deaths represent 15.4% of total deaths recorded across all DHMS members over the period.

Figure 1: Number of deaths included per Vitality engagement group over time

How do we calculate death rates?

Death rates are strongly correlated with an individual's age, sex and chronic disease profile. Shifts in these demographic values over time can result in shifts in the rates of death. We therefore adjust for changes in these demographic shifts such as changes in the age distribution of the population over time.

Let's consider the impact of 'age' as one example (Figure 2).

Vitality engagement groups feature people across a range of age bands. Death rates vary by age, rather than to the effect of Vitality membership or engagement level.

• Unengaged Vitality members increased in average age from 32.0 years in 2010 to 36.3 years in 2022.
• Highly engaged Vitality members increased in average age from 32.7 years in 2010 to 35.1 years in 2022
• Highly engaged Vitality members were, on average, 1.8 years younger than unengaged Vitality members over our 13-year analysis

Overall, to account for changing demographics and varying profiles across and within Vitality engagement groups, we calculated an age- and sex-standardised rate of death using DHMS member overall 2019 age- and sex-population weightings.

Figure 2: Average age of unengaged versus highly engaged Vitality members over time

Death trends over our 13-year study period for unengaged and highly engaged Vitality members

Between 2010 and 2019, the age- and sex-standardised rate of death for highly engaged Vitality members increased by 19% (from 98 deaths per 100,000 life years in 2010 to 116 deaths per 100,000 life years in 2019). For unengaged Vitality members this rate fell by 6% (from 320 deaths per 100,000 life years in 2010 to 300 deaths per 100,000 life years in 2019).

Standardised rates of death for unengaged Vitality members, however, reached the highest levels during the second year of the COVID-19 pandemic in South Africa (2021) at 699 deaths per 100,000 life years compared to 300 deaths per 100,000 life years for the year 2019 (pre-pandemic) - an increase of 133%.

Standardised rates of death for highly engaged Vitality members increased by 15% from 116 deaths per 100,000 life years in 2019 to 134 deaths per 100,000 life years in 2021.

In addition to Vitality engagement, age, and sex, there are other factors that affect this difference such as underlying disease burden, socio-economic status as well as structural changes made to Vitality point awards.

When we removed from the analysis, deaths where healthcare workers determined that the underlying cause was related to COVID-19, we still noted an elevated death risk for unengaged Vitality members - at 370 deaths per 100,000 life years in 2021 (down from 699 deaths per 100,000 life years including COVID-19 deaths) relative to 300 deaths per 100,000 life years for 2019. This elevated 2021 experience might be because deaths that were, in fact, related to COVID-19 were not classified as such.

The death rate for highly engaged Vitality members in 2021 - 94 deaths per 100,000 life years - was in fact lower than death rates in this group were in pre-pandemic years at 116 deaths per 100,000 life years for 2019.

For both unengaged and highly engaged Vitality members, rates of death returned to lower than pre-pandemic levels in 2022.

Over the 10-years before COVID-19 (2009-2019) unengaged Vitality members had around a three-fold higher rate of death compared to highly engaged Vitality members.

Figure 3: All cause age- and sex-standardised death rates for DHMS including and excluding COVID-19 related deaths, split by Vitality engagement group from 2010 to 2022

What were the leading causes of death over the last 10 years for unengaged and highly engaged vitality members?

In 2021:

• For unengaged Vitality members, the leading cause of death was COVID-19 (48% of all deaths), followed closely by non-communicable diseases (43%)
• For highly engaged Vitality members non-communicable diseases were the leading cause of death (56% of deaths) followed by COVID-19 (33% of deaths)

Interestingly, looking at our "injuries" category, from 2018 to 2022 more highly engaged Vitality members died due to injuries than did unengaged Vitality members.

Table 1: Proportion of deaths for unengaged Vitality members by underlying cause of death from 2012 to 2022 (underlying cause of death data not available for 2016 and 2017)

Table 2: Proportion of deaths for highly engaged Vitality members by underlying cause of death from 2012 to 2022 (underlying cause of death data not available for 2016 and 2017).

From 2012 to 2022, the leading non-communicable disease cause of death was cancer for both unengaged and highly engaged Vitality members.

In 2022 cancer made up 61% of non-communicable disease deaths in highly engaged Vitality members and 37% in unengaged Vitality members. Cardiovascular disease caused more deaths in unengaged Vitality members (23%) than in to highly engaged Vitality members (18%). Diabetes was the third highest driver of death rates for unengaged Vitality members and eighth for highly engaged Vitality members (Tables 3 & 4).

Table 3: Proportion of top 5 non-communicable disease deaths for unengaged Vitality members from 2012 to 2022 (underlying cause of death data not available for 2016 and 2017)

Table 4: Proportion of top 5 non-communicable disease deaths for highly engaged Vitality members from 2012 to 2022 (underlying cause of death data not available for 2016 and 2017)

Conclusion

Unengaged Vitality members experienced higher rates of death than highly engaged Vitality members (after allowing for differences in ages and sex) over the period 2010-2022.

Many factors could contribute to these differences, including chronic disease status, COVID-19 vaccination status, socio-economic differences, or structural changes in Vitality points allocation differences over the years.

Comparing death rates pre-pandemic (2010-2019), to the pandemic period (2020 and 2021) we found:

1. Unengaged Vitality members experienced large increases in rates of death during the pandemic. The majority of this increase was explained by COVID-19 deaths. After removing COVID-related deaths we still found an elevated 2021 death rate experience for unengaged Vitality members, possibly due to COVID-deaths that were misclassified.
2. Highly engaged Vitality members remained at the same risk of death as pre-pandemic.

Higher levels of engagement in the Discovery Vitality programme mean more healthy behaviour (spanning physical activity to healthy eating, preventive health checks and more) and this way of life may lower the risk of experiencing serious COVID-19 disease and dying as a result. Findings from other research we have carried out on the way in which exercise protects against serious COVID-19 outcomes, support this notion:

• COVID-19 vaccine effectiveness significantly higher in people who exercise regularly
• Exercise protects against serious COVID-19 outcomes

Interested in knowing more or reporting on these findings?

Please contact us on MEDIA_RELATIONS_TEAM@discovery.co.za to request any updated data available since publication and to obtain any further context required.

Did you find this post interesting?

Please visit our Discovery Health Insights Hub for a range of analyses and insights shared by our Discovery Health Intelligence Team and spanning a variety of health-related themes.

Disclaimer