Investigation of time-dependent risk of mental disorders after infertility diagnosis, through survival analysis and data mining: a nationwide cohort study

Abstract

Objectives: Infertile patients are vulnerable to mental disorders. However, a time-dependent model predicting the onset of mental disorders specific to infertile patients is lacking. This study examined the risk factors for the development of mental disorders in infertile patients and measured the duration until the occurrence of mental disorders after a diagnosis of infertility.

Methods: A total of 13,317 infertile patients in the 2002–2013 Taiwan National Health Insurance Research Database were observed. The 11 independent variables included in the hypothesised model, together with the dates of infertility and mental disorder diagnoses, were analysed using Cox proportional hazards. Data-mining methods using C5.0 and Apriori supplemented the statistical analyses.

Results: The total prevalence rate of mental disorders among infertile patients in Taiwan was 12.41%, including anxiety (4.66%), depression (1.81%) and other mental disorders (5.94%). The average time interval for onset of mental illness identified using survival analysis was 1.67 years. Income, occupation, treatment method, co-morbidity, region and hospital level and ownership were significant predictors of development of mental illness (all p < .05).

Conclusions: The four categories of factors associated with time-dependent onset were demographics, health, health care provider and geographical characteristics. Certain patient characteristics may predict a higher likelihood of onset of a specific mental disorder. Clinical practitioners may use the findings to identify high-risk patients and make timely health interventions.

摘要

目的：不孕症患者易发生精神疾病。然而, 缺乏一种预测不孕症患者特有的精神疾病发病的时间依赖模型。本研究调查了不孕症患者发生精神疾病的危险因素, 并测量了诊断为不孕症后直到发生精神疾病的持续时间。

方法：对2002-2013年台湾全民健康保险研究资料库中的13317例不孕症患者进行调查。假设模型中包含11个自变量及不孕症和精神疾病的诊断时间, 使用Cox比例风险模型分析。数据挖掘方法使用C5.0和Apriori补充统计分析。

结果：台湾不孕症患者精神障碍总患病率为12.41%, 其中焦虑（4.66%）、抑郁（1.81%）及其他精神障碍（5.94%）。采用生存分析确定的精神疾病发病的平均时间间隔为1.67年。收入、职业、治疗方法、共病率、地区和医院水平及所有权是所有精神疾病发展的重要预测因子（均p < 0.05）。

结论：与发病时间依赖相关的四类因素是人口统计学、健康、卫生保健提供者和地理特征。某些患者的特征可能预示着某种特定精神疾病高发的可能性。临床医生可以利用这些发现来识别高危患者并及时进行健康干预。

Keywords:

• Data mining
• infertility
• mental disorder
• reproductive health
• risk factor
• survival analysis

Acknowledgements

The authors thank the National Health Research Institute, Taiwan, for permitting use of the NHIRD.

Disclosure statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this paper.

Additional information

Funding

This research was supported by the Ministry of Science and Technology of Taiwan (grant numbers NSC 100-2410-H-039-003-SS2, MOST 104-2410-H-039-006 and MOST 105-2410-H-039-008-MY2).