Table 1

Relative risk (RR) in patients with versus without Cushing's disease (CD) of dyads of conditions selected by clinical content expert opinion

OrderCondition 1Condition 2Development data setValidation data set
RRCDNon-CDRR
Frequency (%)Frequency (%)
1HypertensionHirsutism70.049 (2.613)0 (0.000)
2Serious infectionsAdrenal mass42 (2.240)0 (0.000)
3Type 2 diabetesHirsutism62.030 (1.600)0 (0.000)
4Uncontrolled type 2 diabetesPremature menopause44.015 (0.800)0 (0.000)
5Weakness/fatigueHirsutism35.064 (3.413)1 (0.027)128.0
6HyperlipidaemiaAdrenal mass56 (2.987)1 (0027)112.0
7Type 2 diabetesAdrenal mass52 (2.773)1 (0.027)104.0
8Psychiatric disordersHirsutism43.049 (2.613)1 (0.027)98.0
9Serious infectionsHirsutism44 (2.347)1 (0.027)88.0
10Sleep disordersAdrenal mass35 (1.867)1 (0.027)70.0
11Uncontrolled type 2 diabetesHypokalaemia15.032 (1.707)1 (0.027)64.0
12Psychiatric disordersAdrenal mass63 (3.360)2 (0.053)63.0
13Weakness/fatigueAdrenal mass56 (2.987)2 (0.053)56.0
14ObesityOsteoporosis10.026 (1.387)1 (0.027)52.0
15Metabolic syndrome/impaired glucose tolerance /pre-diabetesVertebral, long bone, rib, pelvic and foot fracture38.018 (0.960)1 (0.027)36.0
16Weakness/fatigueFemale balding15.330 (1.600)3 (0.080)20.0
17Type 2 diabetesWeakness/fatigue8.9184 (9.813)22 (0.587)16.7
18Type 2 diabetesHypokalaemia19.356 (2.987)7 (0.187)16.0
19ObesityWeakness/fatigue11.7150 (8.000)22 (0.587)13.6
20HypertensionOsteoporosis15.695 (5.067)19 (0.507)10.0
21Type 2 diabetesPremature menopause31.335 (1.867)7 (0187)10.0
22OsteoporosisSerious infections15.166 (3.520)16 (0.427)8.3
23HypertensionSerious infections4.6269 (14.347)106 (2.827)5.1
24Psychiatric disordersSerious infections4.5245 (13.067)119 (3.173)4.1
  • The 24 pairs of conditions in this table where chosen as follows: authors initially selected for analysis 47 conditions defined by literature search and experts as being associated with CD in order to focus on conditions common enough in the CD population that their identification could be an aid to early diagnosis. After generating all possible two-way combinations of these conditions, authors chose for retention those that were either present at ≥5% frequency among patients with CD or present at ≥1% frequency if the RR of the condition was ≥15. Using this frequency and RR data along with expert consultation, authors narrowed the list down to 24 pairs of conditions. RR was calculated for these pairs and confirmed in the validation data set. These results indicate that patients with CD have a fourfold or higher risk of having any one of these 24 pairs of conditions compared with non-CD patients.

  • Table reprinted from Broder et al,11 with permission from the American Association of Clinical Endocrinologists.