Adult-onset testosterone deficiency (TD) in men is diagnosed by the finding of low serum testosterone levels and recognised, associated symptoms. The condition has high prevalence in men over 50 years of age, particularly those with type 2 diabetes (T2DM). Accumulating data show adult-onset TD is associated with increased mortality risk. We review the literature and consider the evidence suggesting testosterone therapy (TTh) reduces mortality, especially in men with T2DM. We previously reported that in the Burntwood Lichfield Atherstone Sutton Coldfield Tamworth (BLAST) study screened cohort of men with adult-onset TD and T2DM adult-onset TD was associated with increased mortality with TTh decreasing this higher mortality. The data hinted that the effect was greater in older men. We confirmed this observation with statistical analyses to study the effect of age on the association between adult-onset TD and mortality; Cox regression analysis demonstrated that the reduced risk (hazard ratio: 0.61, 95% CI: 0.38–0.96) following TTh was restricted to men above the median age of 65.89 years. Finally, we speculate on putative mechanisms that may mediate these associations. Heterogeneity in men with adult-onset TD is expected in view of its definition of low testosterone levels together with associated clinical phenotypes that are not always directly related. Many of these classifying phenotypes are associated with increased mortality. Thus, it is perhaps possible that mechanism(s) of all-cause mortality reduction following TTh is via the impact on these associated phenotypes such as the metabolic syndrome (MetS), hyperglycaemia, hypertension, dyslipidaemia, low haematocrit, sex hormone binding levels, erectile dysfunction, etc. We propose that further research studying the effect of TTh takes heterogeneity into account.