Table 4 Qualitative themes and example quotes from participants on the mental health screening tools

(1) Amendments to tool structure and language

[3]—Ask if they are interested in information about MH referrals. Ask about drug and alcohol issues which are also very common

[57]—I think that the last part of question 11 could be laid out better

[27]—I would reorder the options for question 11 and start with anxiety or depression first rather than Schizophrenia. Q.13—Have 'other' option

[94]—Also interested about non-inclusion of carer question – e.g. do you have or would you like a support person or carer?

(2) Scale response options (3 vs. 5)

[49]—I think the 5 point scale is more likely to have more sensitivity in picking up problems. The second scale (SPHERE-12) is also liable to be misinterpreted more easily by a patient with a serious mental illness, is less clear and direct and appears more casual

[57]—I found the wording and the categories in questions 1–12 somewhat difficult in SPHERE e.g. choosing between answers a b and c present problems in understanding. I think that never, sometimes or always would be easier choices

[68]—Potential for patients to get confused / overwhelmed with 5 options, especially as filling out many forms in the preadmission clinic and requires increased mental load / capacity

[74]—I feel like the 3-point scale will cause people to err on the low side. Maybe that it ok, but perhaps a 4 point scale may be more accurate

(3) Difficulty with inclusion of somatic symptoms in SPHERE-12

[82]—Depending on surgical procedure, the SPHERE-12 is likely to increase scores and reduce specificity particularly with the somatic questions

[85]—Somatic symptom questions not specific to mental health in a pre-surgical setting is a problem

(4) Greater practicality and familiarity with using the K10 tool

[6]—In broad terms the SPHERE while locally developed is relatively infrequently used

[48]—As K10 is standard form for MH and has eMR version, more likely than SPHERE-12 to have comparative measures

(5) Challenges for specific patient cohorts to complete a tool

[31]—May be difficult to use for thought disordered patients and/or paranoid patients

[45]—There will be the obvious difficulty with NESB patients

(6) Preoperative timing consideration for surgical patients to complete the tool

[14]—I think surgical patients may be too geared up by an operation to answer the survey adequately

[80]—Patients would in general be willing to fill in survey but not in first week