Kleinstäuber Maria, Frank Ina, Weise Cornelia
Philipps-University Marburg, Department of Clinical Psychology and Psychotherapy, Germany.
Johannes Gutenberg-University Mainz, Department of Clinical Psychology and Psychotherapy, Germany.
J Psychosom Res. 2015 Mar;78(3):277-84. doi: 10.1016/j.jpsychores.2014.12.001. Epub 2014 Dec 10.
Because the postulated three-factor structure of the internationally widely used Tinnitus Handicap Inventory (THI) has not been confirmed yet by a confirmatory factor analytic approach this was the central aim of the current study.
From a clinical setting, N=373 patients with chronic tinnitus completed the THI and further questionnaires assessing tinnitus-related and psychological variables. In order to analyze the psychometric properties of the THI, confirmatory factor analysis (CFA) and correlational analyses were conducted.
CFA provided a statistically significant support for a better fit of the data to the hypothesized three-factor structure (RMSEA=.049, WRMR=1.062, CFI=.965, TLI=.961) than to a general factor model (RMSEA=.062, WRMR=1.258, CFI=.942, TLI=.937). The calculation of Cronbach's alpha as indicator of internal consistency revealed satisfactory values (.80-.91) with the exception of the catastrophic subscale (.65). High positive correlations of the THI and its subscales with other measures of tinnitus distress, anxiety, and depression, high negative correlations with tinnitus acceptance, moderate positive correlations with anxiety sensitivity, sleeping difficulties, tinnitus loudness, and small correlations with the Big Five personality dimensions confirmed construct validity.
Results show that the THI is a highly reliable and valid measure of tinnitus-related handicap. In contrast to results of previous exploratory analyses the current findings speak for a three-factor in contrast to a unifactorial structure. Future research is needed to replicate this result in different tinnitus populations.
国际上广泛使用的耳鸣障碍量表（THI）假定的三因素结构尚未通过验证性因素分析方法得到证实，这是本研究的核心目标。
在临床环境中，373例慢性耳鸣患者完成了THI及其他评估耳鸣相关和心理变量的问卷。为了分析THI的心理测量特性，进行了验证性因素分析（CFA）和相关性分析。
与一般因素模型（RMSEA = 0.062，WRMR = 1.258，CFI = 0.942，TLI = 0.937）相比，CFA为数据与假设的三因素结构（RMSEA = 0.049，WRMR = 1.062，CFI = 0.965，TLI = 0.961）的更好拟合提供了统计学上的显著支持。以Cronbach's alpha作为内部一致性指标进行计算，除灾难性分量表（0.65）外，其余分量表的值均令人满意（0.80 - 0.91）。THI及其分量表与耳鸣困扰、焦虑和抑郁的其他测量指标高度正相关，与耳鸣接纳度高度负相关，与焦虑敏感性、睡眠困难、耳鸣响度中度正相关，与大五人格维度相关性较小，证实了结构效度。
结果表明，THI是一种高度可靠且有效的耳鸣相关障碍测量工具。与先前探索性分析的结果不同，当前研究结果支持三因素结构而非单因素结构。未来需要在不同的耳鸣人群中重复这一结果。
