Although most patients are improved after shoulder arthroplasty, the degree of improvement is variable. The factors contributing to this variability are not well understood. In particular, little information is available regarding the preoperative characteristics of the patient that may influence the quality of the result. This study correlated patient demographics, preoperative health status, and preoperative shoulder function with 3 outcome metrics: comfort, physical role function, and shoulder-specific function. One hundred thirty-four shoulders having total shoulder arthroplasty for degenerative glenohumeral joint disease had an average follow-up of 3.4 +/- 1.8 years. The SF-36 Comfort score improved from 39 to 61 (P < .0001). The SF-36 Physical Role Function score improved from 30 to 52 (P < .0001). The average number of Simple Shoulder Test functions performable (out of 12) improved from 4 to 9 (P < .0001). The strongest correlates with postoperative comfort included preoperative physical function (P < .0001), general health (P < .0001), and social function (P < .001). The strongest correlates with postoperative physical role function included preoperative physical function (P < .0001) and general health (P < .001). The strongest correlates with postoperative shoulder function included male gender (P < .0001), and preoperative physical function (P < .0001), social function (P < .0001), mental health (P < .0001) and shoulder function (P < .0001). These data indicate that the overall well-being of the patient before surgery is strongly correlated with the quality of the outcome from total shoulder arthroplasty for degenerative glenohumeral joint disease.