Riemsma R P, Forbes C A, Glanville J M, Eastwood A J, Kleijnen J
NHS Centre for Reviews and Dissemination, University of York, York, UK.
Health Technol Assess. 2001;5(6):1-100. doi: 10.3310/hta5060.
Currently there is a wide range of health status measures that aim to assess general health status in people with cognitive impairment. However, the validity and/or applicability to this patient group are largely unknown. This has implications for the assessment of treatment outcomes and rehabilitation, for prognostic purposes, for planning services, and for determining the benefits and adverse effects of health technologies targeted at these patient groups.
(1) To identify the general health status measures that have been validated in patients with cognitive impairment. (2) To assess the extent to which these measures have been validated. (3) To draw out the implications of the findings for the use of existing measures and for future primary research in this area. METHODS.
Studies that assessed general health status in people with cognitive impairment due to acquired brain injury (traumatic brain injury, cerebro-vascular accident or multiple sclerosis (MS)) or learning disability (LD) were included in the review. Studies that used general health status instruments measuring only one general health dimension, and studies that only featured participants with cognitive impairment due to dementia were excluded. METHODS.
A wide range of relevant databases were searched for studies on cognitive impairment, general health status measures, and validation of health status measures. A handsearch of general health status bibliographies was also conducted. Data were collected on the general health status measure used, the population characteristics, aims of the study, validity details, and conclusions.
The review includes data from 71 studies, reported in 83 separate publications. In total 34 different general health status measures were described in the 83 publications, with the Sickness Impact Profile (SIP) and the Short Form-36 (SF-36) the most frequently used measures (20 and 19 studies, respectively). These studies included a total of 98 instrument validations, 52 of which definitely or probably included people with cognitive impairment. Six measures were extensively validated (quality scores ranged from 0.25 to 0.5, on a scale from 0 to 1) in studies in which more than 50% of the respondents were people with cognitive impairment. A further three measures were also validated in studies in which more than 50% of the respondents were people with cognitive impairment, but their level of validation was more limited (quality scores ranged from 0.1 to 0.2). Five measures were validated in studies in which 20-50% of the respondents were cognitively impaired, which may limit their relevance to participants with cognitive impairment (quality scores ranged from 0.1 to 0.6). The SF-36 was also validated in two studies in which 20-50% of the respondents were cognitively impaired and the quality score was 0.3. Finally, nine of the measures were only validated in studies in which less than 20% of the respondents were cognitively impaired. For these measures it was unclear whether the findings applied to people with cognitive impairment.
Very few measures have been validated specifically for cognitively impaired respondents. Studies where at least 50% of the respondents were cognitively impaired generally showed poorer validity results compared with studies with fewer cognitively impaired persons, indicating that general health status measures designed for the general population are not automatically suitable for people with cognitive impairment. The few measures that were specifically developed for people with cognitive impairment also reported poor validity results. Therefore, there are no validated instruments available for use in cognitively impaired respondents; existing measures, specifically designed for use in these populations, should be used with caution. The most promising measure is the MS-Quality of Life Interview (MS-QLI) for MS patients. The MS-QLI was thoroughly validated in 300 MS patients and the results were good, except for the 'social function' subscale. However, only 20-50% of the respondents in this study had cognitive impairment. Most information on the validity of general health status measures was found in studies among people with LD. For these patients, six measures were found that have been validated in a populations where more than 50% of the respondents were cognitively impaired LD patients.
(1) Existing general health status measures should be used with caution in individuals with cognitive impairments. (2) There is no evidence to indicate the most suitable general health status measure for use in economic evaluations of cognitive impairment. (3) There is little evidence to support the validity of proxy assessments in cognitively impaired populations. (ABSTRACT TRUNCATED)
目前有多种健康状况测量方法旨在评估认知障碍患者的总体健康状况。然而，这些方法对该患者群体的有效性和/或适用性在很大程度上尚不清楚。这对治疗效果和康复评估、预后判断、服务规划以及确定针对这些患者群体的健康技术的益处和不良反应均有影响。
（1）识别已在认知障碍患者中得到验证的总体健康状况测量方法。（2）评估这些测量方法的验证程度。（3）阐述研究结果对现有测量方法的使用以及该领域未来基础研究的启示。方法。
纳入评估因后天性脑损伤（创伤性脑损伤、脑血管意外或多发性硬化症（MS））或学习障碍（LD）导致认知障碍患者总体健康状况的研究。排除仅使用测量单一总体健康维度的总体健康状况工具的研究，以及仅纳入因痴呆导致认知障碍参与者的研究。方法。
检索了一系列相关数据库，以查找关于认知障碍、总体健康状况测量方法以及健康状况测量方法验证的研究。还对手动检索的总体健康状况文献目录进行了检索。收集了关于所使用的总体健康状况测量方法、人群特征、研究目的、有效性细节及结论的数据。
该综述纳入了71项研究的数据，这些数据发表在83篇独立的出版物中。83篇出版物共描述了34种不同的总体健康状况测量方法，其中疾病影响量表（SIP）和简短健康调查问卷（SF - 36）是最常用的测量方法（分别有20项和19项研究使用）。这些研究总共进行了98次工具验证，其中52次肯定或可能纳入了认知障碍患者。在超过50%的受访者为认知障碍患者的研究中，有6种测量方法得到了广泛验证（质量得分在0至1的量表上，范围为0.25至0.5）。另外3种测量方法也在超过50%的受访者为认知障碍患者的研究中得到了验证，但其验证程度较为有限（质量得分范围为0.1至0.2）。5种测量方法在20% - 50%的受访者为认知障碍患者的研究中得到了验证，这可能限制了它们与认知障碍参与者的相关性（质量得分范围为0.1至0.6）。SF - 36在两项20% - 50%的受访者为认知障碍患者的研究中也得到了验证，质量得分为0.3。最后，9种测量方法仅在不到20%的受访者为认知障碍患者的研究中得到了验证。对于这些测量方法而言，其研究结果是否适用于认知障碍患者尚不清楚。
专门针对认知障碍受访者进行验证的测量方法非常少。与认知障碍患者较少的研究相比，至少50%的受访者为认知障碍患者的研究通常显示出较差的有效性结果，这表明为一般人群设计的总体健康状况测量方法并非自动适用于认知障碍患者。专门为认知障碍患者开发的少数测量方法也报告了较差的有效性结果。因此，目前没有可用于认知障碍受访者的经过验证的工具；专门为这些人群设计的现有测量方法应谨慎使用。最有前景的测量方法是针对MS患者的MS生活质量访谈（MS - QLI）。MS - QLI在300名MS患者中得到了充分验证，结果良好，但“社会功能”子量表除外。然而，该研究中只有20% - 50%的受访者有认知障碍。关于总体健康状况测量方法有效性的大多数信息来自LD患者的研究。对于这些患者，发现有6种测量方法在超过50%的受访者为认知障碍LD患者的人群中得到了验证。
（1）对于认知障碍个体，应谨慎使用现有的总体健康状况测量方法。（2）没有证据表明哪种总体健康状况测量方法最适合用于认知障碍的经济评估。（3）几乎没有证据支持在认知障碍人群中代理评估的有效性。（摘要截断）
