Not finished the whole article yet, but at least the first paper that's cited has obviously not been read beyond the first few lines.
But individuals are not averages: Studies have found that anywhere from one-third to three-quarters of people classified as obese are metabolically healthy
The author cites this meta-analysis for this: The prevalence of metabolically healthy obesity: a systematic review and critical evaluation of the definitions used (can't post link in comments)
To quote the conclusion of the full paper:
The prevalence of the MHO phenotype has been widely
debated. This systematic review showed that the overall
prevalence of MHO varied from 6% to 75%. The preva-
lence seems to be higher in women, young people and
Asians. However, when only studies with at least a 70%
response rate were considered, the overall MHO preva-
lence ranged from 10% to 51%. Considering the marked
heterogeneity of MHO definitions described in the litera-
ture, it is clear that the establishment of a common MHO
definition is urgently needed, although this may not be
an easy task because we still do not know the precise
mechanisms that are involved with this phenotype and its
clinical implications in the long term.
A) there's no widely accepted definition for being metabolically healthy in obese people, this also ignores all non-metabolic issues, i.e. having a good HbA1c but arthrosis due to your joints being overwhelmed.
B) with rising trial/study quality, the rate of MHO obese people dropped
C) young people are more likely to be MHO, meaning that they simply haven't developed damage yet. Asian people are more likely to be MHO, because some of the trials in asian populations "define obesity was a lower
BMI cut-off point ≥25 kg/m² (22% of the studies)", whereas in trials with caucasian/western populations it was a BMI >30kg/m².
Another poignant quote from the original meta-analysis:
On one hand, high prevalence estimates (arbitrarily set in ≥ 33%) were obtained when the definition was based on less strict criteria, for example, that proposed by Meigs et al . (37) using only HOMA-IR. On the other hand, low prevalence estimates were found when a more stringent definition of MHO was used, for example, that proposed by Karelis et al . (36), which is based on five cardiometabolic factors: blood pressure, HDL-c, low- density lipoprotein cholesterol, total cholesterol and HOMA-IR.
Again, HMO prevalence is tied to low trial quality.
Unlike the HuffPos authors conclusion, mine is that MHO is by and large a myth and for most people, for most individuals, obesity equals being unhealthy.