# Predicting avoidant coping in individuals recently diagnosed with serious illness: a cross-sectional study

## TL;DR

This study found that avoidant coping behaviors in people recently diagnosed with serious illness cannot be reliably predicted using health and socioeconomic factors.

## Contribution

The novel contribution is demonstrating that predictive models for avoidant coping have poor accuracy in recently diagnosed serious illness patients.

## Key findings

- 13.4% of recently diagnosed individuals showed high avoidant coping.
- High avoidance was linked to female sex, lower education, shorter life-expectancy, and poor social support.
- Predictive models had poor discriminative capacity (AUC 0.62).

## Abstract

Avoidant coping has been linked to poorer health outcomes. This cross-sectional study aimed to examine whether high avoidant coping can be reliably predicted from health parameters and socioeconomics among adults recently diagnosed with a serious illness.

A nationwide survey linked to national registers. Inclusion criteria were: (i) age ≥50 years, and (ii) diagnosed with cancer, neurological disease, and/or heart disease within the year preceding the survey. Coping was assessed using the Brief Approach/Avoidance Coping Questionnaire with high avoidance defined as > mean avoidance score +1 SD. Predictive models were developed using data from all survey respondents aged ≥50 years and subsequently tested in the recently diagnosed subsample. Area Under the Curve (AUC) and 95% Confidence Intervals (CI) were reported.

The recently diagnosed sample comprised 746 individuals, of whom 13.4% exhibited high avoidant coping. High avoidance was associated with female sex, lower educational level, shorter self-reported life-expectancy, and poorer perceived social support. However, the predictive models demonstrated poor discriminative capacity (AUC 0.62; CI: 0.57–0.68) for the recently diagnosed sample. Among adults aged 50 + years recently diagnosed with serious illness, high avoidance could not be reliably predicted from health parameters and socioeconomics.