Categorisation is inevitable and essential for routine medical practice.
I absolutely agree that apparently 'mild' symptoms can have a devastating effect on quality of life. Acknowledging and dealing with this fact is critical in any doctor-patient relationship.
Equally, I am uncomfortable with the notion that "classifying... severity of a disease" should not be done and that treatments should be discussed without reference to disease "severity scale".
Legally, not discussing the severity of disease would severely compromise "informed consent".
Medically, daily clinical practice of all specialities involves objective categorisation of diseases and pre-diseases into various categories such as mild/moderate/severe or low/intermediate/high risk. For example, discussion of anti-coagulation for atrial fibrillation, statins for primary prevention or adjuvant chemotherapy for breast cancer needs risk categorisation.  
Comparison of a patient with general population is not only inevitable but also essential as it can have a positive effect too. Prostate cancer patients with low-risk, grade 1 cancer feel immensely reassured by the fact that their life expectancy is likely to match that of their peers, thus avoiding unnecessary over-treatment of their cancer.
1 McHugh B. When you live with a degenerative illness, no symptoms are mild. BMJ 2018;361:k1670. doi:10.1136/bmj.k1670
2 Staerk L, Wang B, Preis SR, et al. Lifetime risk of atrial fibrillation according to optimal, borderline, or elevated levels of risk factors: cohort study based on longitudinal data from the Framingham Heart Study. BMJ 2018;361:k1453. doi:10.1136/bmj.k1453
3 Ryan A, Heath S, Cook P. Managing dyslipidaemia for the primary prevention of cardiovascular disease. BMJ 2018;360:k946. doi:10.1136/bmj.k946
4 Mayor S. Some breast cancer patients with low genetic risk may not need chemotherapy, trial shows. BMJ 2016;354:i4635. doi:10.1136/bmj.i4635
5 Seibert TM, Fan CC, Wang Y, et al. Polygenic hazard score to guide screening for aggressive prostate cancer: development and validation in large scale cohorts. BMJ 2018;360:j5757. doi:10.1136/bmj.j5757
Competing interests: No competing interests