 |
| Depression often missed |
28-Jul-2009 |
|
|
|
By Jenny Pogson
<[hnj ADW Standard]><[stk -3]>LANCET Depressed patients have only a 50/50 chance of their condition being detected after a single GP consult, a meta-analysis suggests.<[etk]><[qj]><[blg on]>
<[stk -3]>However, GPs have a much better strike rate for ruling out the condition in non-depressed patients, the analysis of 41 studies found. <[etk]>
<[stk -1]>Where the prevalence of depression was 20%, a GP would pick up about 10 out of 20 cases of depression. They correctly reassured 65 out of 80 non-depressed patients, thereby falsely diagnosing 15 people as depressed, the study authors said.<[etk]>
<[stk -3]>The actual prevalence of depression -- determined mainly by standardised diagnostic interviews -- was 10% in Australia; the lowest among the countries included in the meta-analysis.<[etk]>
<[stk -3]>Reassessing those patients who had an unclear depression classification would increase the accuracy of the diagnosis, the authors said.<[etk]>
"Because one-off brief assessments only facilitate identification of about half of those with depression, we suggest that additional consultation time should be available for those likely to have depression," they said.
<[stk -3]>The results should not be interpreted as a criticism of GPs, the authors said, but a call to better understand the problems they face managing depression.<[etk]>
<[stk -3]>Professor Gordon Parker, executive director of the Black Dog Institute in Sydney, said the definition of depression was dimensional so there would always be over- and under-classification of the condition.<[etk]>
<[stk -3]>"After establishing someone is provisionally depressed, you need to move quickly into [asking whether it] is clinically significant … and how it is best accounted for," he said.<[etk]>
<[stk -3]>"Those are the more important issues, rather than simply saying: 'Is this person a depressed case or not?'" Professor Parker said.<[etk]>
<[stk -4]>Ways of identifying depression did not need to be complicated, he said, as just asking a patient: "Do you think you are depressed?" had an overall accurate classification rate of 80%.<[etk]><[stk -3]><[etk]>
<[stk -3]>Lancet 2009; online.<[etk]><[ql]>
|
Submit your feedback here: