Scholarlink

Background <p>Junior nurses (clinical experience ≤ 5 years) are prone to increased second victim distress after adverse events, yet the heterogeneity within this population remains poorly understood. This study used a novel combination of latent profile analysis (LPA) and association rule mining (ARM) to identify distinct second victim profiles, investigate factors associated with profile membership, and design corresponding care-bundle prototypes.</p> Methods <p>A cross-sectional, STROBE-compliant survey of 387 junior nurses with ≥ 1 self-reported adverse event in the previous 12 months was conducted in three Jilin Province hospitals (provincial, municipal, county) between March and September 2025. Validated scales assessed second victim experiences, resilience, social support, coping, and personality, alongside items on adverse-event characteristics, understanding from patients’ families, and institutional response. LPA identified latent subgroups; multinomial logistic regression identified factors independently associated with profile membership; ARM (Apriori algorithm) extracted multi-factor co-occurrence rules and characteristic networks.</p> Results <p>Three profiles emerged: Low Support–Low Distress (42.12%, <i>n</i> = 163), Moderate Support–Moderate Distress (39.28%, <i>n</i> = 152), Low Support–High Distress (18.60%, <i>n</i> = 72). Moderate Support–Moderate Distress membership was most strongly associated with monthly income &lt; 4,500 CNY (OR = 13.254), reflecting demand–reward mismatch. Low Support–High Distress membership was independently associated with low social support (OR = 7.707), poor resilience (OR = 6.777), and lack of understanding from patients’ families (OR = 5.080). For the group with the highest observed distress, ARM isolated a three-element combination, {≥10 night shifts/month + lack of understanding from patients’ families + high negative coping} (confidence 0.920, lift 4.945), which may represent a candidate signal requiring validation.</p> Conclusions <p>Junior nurses’ second victim experiences form three heterogeneous profiles. The LPA + ARM combination identified profile-specific multi-factor co-occurrence patterns and supports person-centred early identification. Three care-bundle prototypes (high-distress stabilisation, demand–reward rebalancing, peer-mentor capacity building) provide a preliminary, data-informed framework for tiered support, pending independent validation.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Latent profiles and characteristic networks of second victim experiences among junior nurses: a latent profile analysis and association rule mining study

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月28日

查看全文
  • Yaokang Xue,
  • Xiaoshuang Zhao,
  • Hairui Sun,
  • Baorong Zhang,
  • Shanshan Yu,
  • Hua Yuan,
  • Yuxuan Yang,
  • Ruixin Guan

摘要

Background

Junior nurses (clinical experience ≤ 5 years) are prone to increased second victim distress after adverse events, yet the heterogeneity within this population remains poorly understood. This study used a novel combination of latent profile analysis (LPA) and association rule mining (ARM) to identify distinct second victim profiles, investigate factors associated with profile membership, and design corresponding care-bundle prototypes.

Methods

A cross-sectional, STROBE-compliant survey of 387 junior nurses with ≥ 1 self-reported adverse event in the previous 12 months was conducted in three Jilin Province hospitals (provincial, municipal, county) between March and September 2025. Validated scales assessed second victim experiences, resilience, social support, coping, and personality, alongside items on adverse-event characteristics, understanding from patients’ families, and institutional response. LPA identified latent subgroups; multinomial logistic regression identified factors independently associated with profile membership; ARM (Apriori algorithm) extracted multi-factor co-occurrence rules and characteristic networks.

Results

Three profiles emerged: Low Support–Low Distress (42.12%, n = 163), Moderate Support–Moderate Distress (39.28%, n = 152), Low Support–High Distress (18.60%, n = 72). Moderate Support–Moderate Distress membership was most strongly associated with monthly income < 4,500 CNY (OR = 13.254), reflecting demand–reward mismatch. Low Support–High Distress membership was independently associated with low social support (OR = 7.707), poor resilience (OR = 6.777), and lack of understanding from patients’ families (OR = 5.080). For the group with the highest observed distress, ARM isolated a three-element combination, {≥10 night shifts/month + lack of understanding from patients’ families + high negative coping} (confidence 0.920, lift 4.945), which may represent a candidate signal requiring validation.

Conclusions

Junior nurses’ second victim experiences form three heterogeneous profiles. The LPA + ARM combination identified profile-specific multi-factor co-occurrence patterns and supports person-centred early identification. Three care-bundle prototypes (high-distress stabilisation, demand–reward rebalancing, peer-mentor capacity building) provide a preliminary, data-informed framework for tiered support, pending independent validation.