Although there are some locations the place the wording is cautious, there’s a robust need on this journal article, and the related information articles, to make the interpretation that having a greater angle towards ageing will end in higher goal ageing outcomes. That may be true, and appears more likely to no less than be true at with a low impact measurement. However the “affiliation” or correlation between ageing satisfaction (the measuring of angle towards ageing) and goal ageing outcomes might simply be an occasion of individuals having a way of their very own well being standing that goes past simply tallied goal outcomes and having a worse angle towards ageing as a result of they already know their well being is poor.
I don’t see how the impact of angle towards ageing on goal ageing outcomes could be recognized and not using a randomized managed trial of an intervention attempting to enhance attitudes towards ageing. Word that getting a placebo here’s a little tough. It requires one thing like an intervention to alter an angle regarded as irrelevant to ageing outcomes.
If one considers underlying well being standing to be distinct from simply tallied well being outcomes, then one might see this as “Cousin Causality” from underlying [health status] affecting each [aging satisfaction] and [later easily tallied bad health outcomes]. A advantage of seeing this for instance of cousin causality (generally known as “third-factor causality”) versus reverse causality is that it makes clearer that the ageing satisfaction being measured earlier in time than the simply tallied well being outcomes doesn’t imply the ageing satisfaction precipitated the well being outcomes. The ancestor issue affecting each can simply be earlier in time than each. However then on one department we’re saying that poor underlying well being begets dangerous simply tallied well being outcomes, which looks like a truism.
Within the conclusion, the authors Julia S. Nakamura, Joanna H. Hong, and Jacqui Smith write:
… there’s potential for confounding by third variables. Nonetheless, we addressed this concern by implementing a longitudinal research design, sturdy covariate adjustment, and E-value analyses.
“Longitudinal research design” might imply two issues: one, having ageing satisfaction measured earlier than simply tallied well being outcomes. That doesn’t clear up an issue of “confounding by third variables,” which is one other approach of describing cousin causality, for the rationale I mentioned above: the third ancestor variable can simply be earlier than each ageing satisfaction and simply tallied well being outcomes. The opposite factor “longitudinal research design alludes to is a give attention to adjustments in ageing satisfaction. However adjustments in ageing satisfaction can simply be as a result of adjustments in underlying well being standing, and every thing I say above goes by means of. “Strong covariate adjustment” doesn’t clear up the issue of confounding as a result of subjective well being measures comprise data that goes past the entire different well being measures within the Well being and Retirement Examine (HRS); analysis reveals that subjective well being measures add predictive energy for later simply tallied well being outcomes past different covariates within the HRS. I’m suggesting that ageing satisfaction must be thought of as being analogous to the subjective well being measures within the HRS. Word that doesn’t imply controlling for subjective well being would clear up the issue both, as a result of subjective well being is definitely measured with error. (See “Including a Variable Measured with Error to a Regression Solely Partially Controls for that Variable.” Having ageing satisfaction as a second measure of subjective well being ought to extend predictive energy over just one measure of subjective well being. Lastly, the “E-value analyses” merely say that the story for why there may be confounding must be one which makes a whole lot of confounding believable. I believe that’s happy by the story I give above.
