Using Negative Lived Experience Scale for the Early Detection of Suicidal Ideation

Results: Using linear structural relations (LISREL) Confirmatory Factor Analysis (CFA), a three-factor model was determined to have a good fit. The Cronbach’s α coefficient of the GNLES with a 3-factor structure was 0.83. The GNLES has acceptable criterion-related validity. The suicidal ideation-predictive rate of the GNLES was greater than 70% for older adults. The performance of the model was acceptable, with an area under the curve (AUC) of 0.704.


Introduction
Geriatric suicide is a serious social and public health problem.
In most countries, the suicide rate among individuals aged 65 or older is higher than that of any other age group. Suicides among older adults receive less attention in terms of social resources, research and media than do suicides among young people [1,2]. The suicide rate among older adults in Taiwan is much higher than that in Germany, the United Kingdom, Italy and Australia. Considering arrangement, economic status, social support, health conditions, physical disabilities, perceived and depressive symptoms [7,8].
Most people do not take the initiative to inform others of their suicidal intention. It is difficult for clinicians to fully assess suicidal ideation. In fact, accurately confirming a suicide case is a difficult and complex task, particularly when the victim is an older adult [9]. When a person wants to commit suicide, he or she will first carefully consider suicide-related behaviors and the possibility of death; thus, suicidal ideation is a crucially important key point in suicide [2]. In general, elderly people rarely seek mental health care, and they do not often take the initiative in expressing their depressive state. Their primary medical complaint may imply suicidal ideation. For example, they may mention depression, nervousness, restlessness, feeling easily angered, feelings of guilt, impulsiveness, eccentric behavior, and/or changes in sleep and appetite that are associated with anxiety. However, these symptoms are often easily attributed to other physiological diseases or related to the aging process Petkus et al. [2]. Thus, the early detection of suicidal ideation is essential.

Lived experience is a complex phenomenon. An individual
does not only exist in the present situation as they experience it; they also carry past experiences. Thus, lived experience is an ongoing and dynamic process. Each lived experience is unique and has its own characteristics and attributes. Lived experience is a subjective awareness of life that exists as a sense of the individual, without external influence [10]. If an individual lived in a difficult environment and could not positively re-construct or re-evaluate his or her situation during that time, he or she might not be able to alleviate stress-related symptoms, and negative emotions and thoughts may result [11]. People who experience suicidal ideation or commit suicide have often experienced many stressful life events.
These life events result in defeatism and a perceived inability to escape. These outcomes, combined with the lack of rescue factors, may increase the risk of suicidal behavior. People who struggle with suicidal ideation or who have attempted suicide typically interpret stressful life events as negative lived experiences [12]. Therefore, it is important to explore the relationship between negative lived experience and suicidal ideation.
The types of lived experiences that an elderly person has had over his or her lifetime can lead to the development of suicidal ideation, suicide attempts and suicidal behavior; thus, an in-depth exploration of lived experiences is warranted. Studies on the lived experience of the suicidal elderly population are scarce. Based on these studies findings, the common lived experience of elderly people who attempted or committed suicide involved negative feelings and experiences; the negative feelings and experiences included the feeling that no one cares for them, life as a psychache and burden, loss of love, power and oneself, and death is better than being alive [13][14][15][16]. The community-dwelling older adults who had experienced suicidal ideation primarily expressed negative feelings, emotions and thoughts when describing their life experiences, such as feeling useless, waiting to die, and feeling that life is hopeless [17]. How the negative lived experiences of older adults trigger suicidal ideation or suicidal behavior and whether negative lived experiences can predict suicidal ideation or behavior remain unclear. Currently, there is no proper instrument for measuring negative lived experiences. Therefore, the development of an instrument that can measure older adults' negative lived experiences is necessary and important.

Theoretical Framework
The framework of the lived experiences of older adults experiencing suicidal ideation was proposed by the first author [17]. In their study, grounded theory was used to explore the lived experiences of elderly individuals experiencing suicidal ideation.
Four categories of lived experiences were identified: loss, negative emotions without outlet, lingering or persistent suicidal ideation, and meaningless existence. Based on this framework, we developed a quantitative measure of negative lived experiences for use with elderly individuals (Figure 1).

Purpose
The purpose of the study was to develop and test the psychometric properties of the Geriatric Negative Lived Experience Scale (GNLES).

Methods
The GNLES was developed in three stages. First, items were generated based on the theoretical framework outlined by Wu et al. [17]. Second, the items were tested for content validity. Last, the psychometric properties of the instrument were tested as part of a larger study.

Stage 1: Item Generation
The following items were created by the first author to address   The original pool consisted of 32 items distributed across the 4 categories outlined above. All these items were reviewed by the authors for redundancy and relevance. A sorting process was applied to eliminate repetitive items and to combine items to condense the scale, resulting in a total of 18 items (4 pertaining to loss, 7 pertaining to negative emotions without outlet, 3 pertaining to lingering or persistent suicidal ideation, and 4 pertaining to meaningless existence). The questions were answered using a Likert-type scale ranging from 1 (strongly agree) to 5 (strongly disagree).

Stage 2: Analysis of Content Validity
A total of 6 professor-level experts in different areas, including psychiatry, psychiatric nursing, elderly care and suicide prevention, were invited to rate the questionnaire in terms of the appropriateness and the rhetorical and semantic expression of the text using a four-point rating scale ranging from 1 (unrelated) to 4 (indicating a strong association) [18]. Based on the opinions of various experts, the wording of some questions was modified.
Three items were omitted because they were only loosely related to negative lived experience; these items addressed the idea of ending one's own life, the event that intensified the suicidal thought, and experiencing suicidal ideation without action. The total number of items was reduced from 18 to 15, and the content validity index The scoring system ranges from 0 (none) to 5 (very severe).
The BSRS-5 is a self-report and the included items have a 7-day recall period. (Lung & Lee, 2008

Construct Validity
Construct validity was assessed using confirmatory factor analysis (CFA). CFA was used to test the degree of compliance of the factors and the items on the scale to obtain the most effective factor structure for the theoretical construct [21]. Using linear structural relations (LISREL) software, the overall fit of the three-factor model was evaluated using the following indicators: relative χ2 ( χ2/df), root mean square error of approximation (RMSEA), goodness of fit  [21]. Therefore, the model was determined to have an acceptable fit. The factor loadings ranged from 0.05 to 0.40 for loss, 0.29 to 0.55 for negative emotions, and 0.22 to 0.55 for meaningless existence, as shown in Figure 2.

Concurrent validity
Concurrent validity refers to a scenario in which two measurement tools are administered to the same participant and both tools measure the same or similar phenomena or concepts: the higher the correlation between the results of these two measurement instruments, the better the concurrent validity [18]. In the present study, the concurrent validity of the GNLES was tested using the GDS. The correlation coefficients of the three subscales and the total scale of the GNLES and the GDS were 0.416, 0.598, 0.609, and 0.642, respectively; the scales were significantly correlated (p < 0.01), as shown in Table 2. Cohen's effect size criteria, with Pearson's r of .10 considered small, .30 medium, and .50 large [22]. Thus, correlations with the GDS were strong for negative emotions (0.598), meaningless existence (0.609) and total GNLES (0.642) and moderate for loss (0.416).

Predictive Validity
Predictive validity refers to the correlation between the scores obtained at the current time and those obtained at a future time point for the same concept or behavior [18]. In the present study, scores of 0 and 1 or above from the BSRS-5 were subjected to logistic regression. The results indicated that the predictive rates of the total scale and its three subscales of loss, negative emotions and meaningless existence were 75.5%, 73.4%, 75.1%, and 76%, as shown in Table 3 shows. Therefore, the predictive rates of the GNLES were greater than 70%, and this scale can be applied as an early screening tool for suicidal ideation among communitydwelling older adults.

Sensitivity and Specificity
To test how accurately the instrument was able to predict risk or no risk of suicidal ideation, we used the sensitivity and specificity of the GNLES. Receiver operating characteristic (ROC) curve were used to determine the range of trade-offs. [23] Based on the total score the GNLES, which ranges from 15 to 75. The performance of the model was good, with an area under the curve (AUC) of 0.704. The estimated probability at the sensitivity and specificity maximum sum was a cut-off probability of 0.330. The sensitivity of the model was 77%, the specificity was 56%, and the optimal cut-off value was 41. In other words, a GNLES score greater than 41 points indicated the possibility of suicidal ideation.

Reliability
In the present study, the Cronbach's α coefficient was used to test the internal consistency of the GNLES subscales and total scale. Many scholars apply the following principles to make rough estimates: an α coefficient greater than 0.9 is considered excellent, approximately 0.8 is considered very good, 0.7 is considered moderate, 0.5 or higher is considered acceptable, and less than 0.5 is considered inadequate [18].

Relationship between Cronbach's α Coefficient and Instrument Homogeneity
The Cronbach's α coefficient for the total scale of the GNLES was 0.835, indicating the internal consistency was very good. The The reliability obtained using CFA can bypass the issues outlined above, indicating that the CFA result is superior to Cronbach's α coefficient [21]. In the present study, there were lower Cronbach reliability values for the loss subscales of the GNLES. In the future, actions aimed at improving the internal consistency of the GNLES should be investigated.

Avoid Being Labeled: Geriatric Suicidal Ideation Measurement Terminology
Currently, the items on the instrument measuring geriatric suicide use language pertaining to suicidal ideation, plans to commit suicide and suicidal behavior or ending one's own life [4,24,25]. Relevant studies investigating suicidal ideation among older adults [26,27]

Prediction of Suicidal Ideation Allows Early Intervention and the Prevention of Geriatric Suicide
Suicidal ideation can be used to evaluate existing and potential suicide attempts, thus predicting the future risk of suicide [4].
Some scholars believe that the predictors of geriatric suicide include a history of suicide attempts, severe depression, perceived health, perceived burden, hopelessness and impulsivity [28,29].
The suicidal ideation predictive rate among older adults using the GNLES was greater than 70%; therefore, negative lived experience is also a predictor of suicide in the elderly population. The GNLES is suitable for use as an early screening tool for suicidal ideation among community-dwelling older adults. If suicidal ideation in older adults can be detected early and interventions provided, suicidal behavior may be reduced or prevented.

Limitations
Because of limitations of time, manpower and material resources, only elderly individuals over the age of 65 within a specific administrative district of Taipei served as participants in the present study. Purposive sampling did not take into consideration differences between urban and rural areas or somatization and other presentations of depression and suicide risk among older adults. Because the sample was not representative of all older adults, the findings of the present study cannot be generalized to older adults in other areas of the country. Furthermore, because our study sample was Taiwanese and was not heterogeneous, we could not further investigate cross-cultural issues or applications.
At present, the GNLES is only at the reliability and validity testing stage. Because the present study was a pilot, it could not provide more comprehensive evidence of the GNLES construct validity or test-retest information to indicate the measure's stability.
A large-scale formal test of older adults within communities is recommended to re-test the reliability and validity of the scale. The content of the scale may be revised accordingly to ensure that it is more fully developed and complete.

Conclusion
In many countries, the suicide rate among older adults is the highest among all age groups. Elderly suicide is an important

Implications for Practice
The GNLES was developed based on the lived experiences of elderly individuals and it is suitable for use as a screening instrument for early suicidal ideation in the elderly population. The GNLES should be used as the basis for the future development of various care interventions and for the early prevention of suicidal ideation or behavior in the elderly population.