Skip to main content

Table 2 Clinical subtypes identified by combination of multiple component and clustering analyses, and their relations with treatment subgroups

From: Multidimensional analyses to assess the relations between treatment choices by physicians and patients’ characteristics: the example of COPD

Clinical subtypes

Description, frequency and associated treatment type

1 : Exposed but not severely impaired patients

Exposed to tobacco smoke or occupational smokes, toxic gaz or dust. No severe airflow obstruction (VEMS > 60%). MRC grade 0 (maximum MRC grade = 2).

9.62%* of patients met the first three conditions.

Patients of this clinical subtype are underrepresented in all treatment subgroups and particularly respiratory support (treatment group 4: OR = 0.198). All treatment types are less prescribed to this clinical subtypes and particularly respiratory support (treatment subgroup 4: OR = 0.255).

2 : Overweight smokers with high blood pressure and other comorbidities

Sleep apnea. Men. Robust stature (weight >80 kg). Current smokers. Underwent polysomnography. High blood pressure. Large variety of other comorbidities. Rather mild or moderate MRC grade.

6.50%* of patients met the first three conditions.

Patients of this clinical subtype are overrepresented among prescriptions of respiratory support (treatment subgroup 4: OR = 3.625) and underrepresented among prescriptions of vaccines (treatment subgroup 6: OR = 0.581).

3 : Severe airflow obstruction

Severe dyspnea (MRC grade 4 or 5). Low FEV1 (< 50%) and FVC. Chronic right ventricular failure. Emphysema. Many investigations performed or prescribed including lung CT-scan, bronchoscopy, DLCO, 6-min walking test, sleep oxymetry, arterial blood gases, exercise testing, echocardiography, EKG, lung scintigraphy.

16.06%* of patients met the first two conditions.

Patients of this clinical subtype are overrepresented in all treatment subgroups and particularly respiratory support (treatment subgroup 4: OR = 6.214). All treatment subgroups are more prescribed to this clinical subtype, and particularly respiratory support (treatment subgroup 4: OR = 7.381).

4 : Women

Women. Small stature (< 66 kg, < 165 cm). Small respiratory capacity (FEV1 < 1 l, FVC < 2 l). 60 years old or more. MRC grade can be severe. Often living alone. Possibly depressive. Living in town. Associated asthma is possible. Were prescribed reversibility testing.

12.70% *of patients met the first three conditions.

Patients of this clinical subtype are not significantly overrepresented or underrepresented in any treatment subgroup and no treatment subgroup is significantly more (or less) prescribed to this clinical subtype (no special treatment for women).

5 : Patients with symptoms of chronic bronchitis

Chronic cough and sputum production, chronic bronchitis. Fibrous or cavity sequelae and bronchiectasis in some patients. Not strongly related to MRC grade but rather moderate.

67.07%* of patients met the first three conditions.

Patients of this clinical subtype are not significantly overrepresented or underrepresented in any treatment subgroup except for vaccines (type 6: OR = 2.210) and no treatment subgroup except for type 6 is significantly more (or less) prescribed to this clinical subtype..

6 : Elderly patients with cardiovascular comorbidity

Age > 75 years. Heart diseases: heart failure, ischemic heart disease or other cardiovascular diseases.Not strongly related to MRC grade but rather moderate. High blood pressure. Possible peripheral artery disease and cancer.

11.06%* of patients met the first and at least one of the 3 other conditions.

Patients of this clinical subtype are significantly underrepresented in nebulised treatment (treatment subgroup 1: OR = 0.686) and overrepresented in fixed combinations (treatment subgroup 2: OR = 1.335) and vaccines (treatment subgroup 6: OR = 1.335). Nebulised treatments are significantly less prescribed to this clinical subtype. Fixed combinations and vaccines are significantly more prescribed to these patients.

  1. Associated treatment subgroups (see Table 3) are those obtained through logistic regressions (see Table 4).
  2. * The sum of percentages is larger than 100% because clinical subtypes are not exclusive. For example the same patient may belong to subtype 1 and 6.