September 02, 2016
1 min read
Save

Alternative statistical method more accurate for early detection of untreated psychosis

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Quantile regression analyses may more accurately measure the impact of early detection efforts on duration of untreated psychosis, according to recent findings.

“Prolonged duration of untreated psychosis is associated with poorer outcomes across multiple health care systems. Early detection efforts can reduce these delays but face a challenge: typically skewed [duration of untreated psychosis] distributions are poorly managed by commonly used statistical methods. This limits the strength and scope of inferences about [early detection]’s effectiveness,” Sinan Guloksuz, MD, PhD, of Maastricht University Medical Centre, the Netherlands, and colleagues wrote.

Researchers assessed efficacy of early detection for untreated psychosis using quantile regression as an alternative statistical method.

“Unlike ordinary least-squares regression, [quantile regression] can estimate the heterogeneous effects of predictors (eg, age, sex, education, or [early detection]) across different quantiles of outcome (duration of untreated psychosis), rather than presuming a uniform mean effect,” the researchers wrote. “Furthermore, unlike linear regression that relies on a normality assumption, [quantile regression] can provide more accurate estimates in samples with extreme outliers, as is common in [duration of untreated psychosis] distributions.”

Analysis was stratified by early (eg, 2006 to 2009) and late (eg, 2010 to 2012) epochs.

Ordinary lest-squares regression indicated a significant coefficient of –3 (P = .02) on predicted mean duration of untreated psychosis during the early period vs. late period (coefficient = 0.1; P = .9).

However, fit diagnostics indicated severe violation of the normality assumption and invalidated the analysis, according to researchers.

Quantile regression indicated a significant differential effect of education by duration of untreated psychosis quantile. More education was associated with lower levels of extreme duration of untreated psychosis during the early period but not the late period (P = .01).

No effect was found for other demographic variables.

“Quantile regression analyses can inform messaging and outreach efforts for first-episode services that are contemplating [early detection] efforts in their communities. A similar analysis, using [early detection] as an independent variable, awaits results from an ongoing study and will allow prospective assessment of the differential effect of this [early detection] campaign across different subpopulations and across the full distribution of [duration of untreated psychosis],” the researchers concluded. – by Amanda Oldt

Disclosure: The researchers report no relevant financial disclosures.