LEGISLATIVE COUNCIL

Question on Notice 5895: (1) and (2)

Both questions 1 and 2 are answered in the table below:

Hospital 2009 2010 2011

Q1 Q2 Q1 Q2 Q1 Q2

(a) Royal Hospital 59 5.20% 106 7.80% 127 8.20%

(b) Sir Charles Gairdner 230 12.20% 233 11.80% 210 9.80% Hospital

(c) Armada le Health Service 44 6.00% 56 7.10% 43 5.30%

(d) Swan Health Service 42 6.60% 53 7.10% 52 6.80%

(e) Health Campus 141 13.60% 168 13.70% 184 13.00%

(I) Hospital 122 8.60% 177 10.80% 190 11.20%

(g) Kalgoorlie Regional 41 14.00% 38 12.30% 55 13.40% Hospital

(h) Geraldton Regional 13 7.30% 6 3.50% 15 6.80% Hospital

(i) Derby Regional Hospital < 5 4.00% 8 12.90% < 5 3.90%

(j) Broome Regional Hospital 11 6.70% 10 7.10% 9 6.10%

(k) Bunbury Regional 62 10.10% 64 10.30% 65 9.30% Hospital

(1) Albany Regional Hospital 101 18.10% 84 17.70% 83 15.80%

(m) Port Hedland Regional 9 9.30% 5 5.00% 6 5.00% Hospital

(n) Northam Regional < 5 2.80% 11 11.60% < 5 4.80% Hospital

(o) Graylands 118 8.20% 120 9.20% 119 9.00% Notes:

Data is provided by calendar year up to December 2011 due to incomplete clinical coding of 2011/12 financial year data. Includes both planned and unplanned returns to the same hospital within 28 days. Includes separations from specialised mental health wards and general wards where there is a primary mental health diagnosis or external cause codes of intentional self-harm (either prior to or during admission) or psychiatric care days. The inclusion of intentional self-harm is a new addition to the methodology from previous PQs and will result in more cases than was reported previously. This is a more robust method than used previously. Excludes "out of scope" separations as per national mental health KPI reporting standards (i.e. day cases, same-day readmissions, inter hospital transfers, and statistical discharges) Source: Mental Health Data Collection.

Health information is considered identifiable when it either directly identifies an individual or has the potential to indirectly identify an individual based on a combination of identifiers or based on small cell sizes. A cell size is considered small if the result in a statistical table shows a value that is less than 5.

For data to be considered non-identifiable there is no reasonable basis to believe that the information can be used to re-identify an individual patient when used in combination with other details held by the requestor or those who will view the information.

The Custodian of the Mental Health Data Collection upholds that patient confidentiality cannot be guaranteed with the release of inpatient data containing small cells sizes.