Passi Reetu, Kaur Manmeet, Lakshmi P V M, Cheng Christina, Hawkins Melanie, Osborne Richard H
Department of Community Medicine and School of Public Health, Post-Graduate Institute of Medical Education and Research, Chandigarh, India.
Centre for Global Health and Equity, School of Health Sciences, Swinburne University of Technology, Melbourne, Australia.
PLOS Glob Public Health. 2023 Feb 17;3(2):e0001595. doi: 10.1371/journal.pgph.0001595. eCollection 2023.
Cluster analysis can complement and extend the information learned through epidemiological analysis. The aim of this study was to determine the relative merits of these two data analysis methods for describing the multidimensional health literacy strengths and challenges in a resource poor rural community in northern India. A cross-sectional survey (N = 510) using the Health Literacy Questionnaire (HLQ) was undertaken. Descriptive epidemiology included mean scores and effect sizes among sociodemographic characteristics. Cluster analysis was based on the nine HLQ scales to determine different health literacy profiles within the population. Participants reported highest mean scores for Scale 4. Social support for health (2.88) and Scale 6. Ability to actively engage with healthcare professionals (3.66). Lower scores were reported for Scale 3. Actively managing my health (1.81) and Scale 8. Ability to find good health information (2.65). Younger people (<35 years) had much higher scores than older people (ES >1.0) for social support. Eight clusters were identified. In Cluster A, educated younger men (mean age 27 years) reported higher scores on all scales except one (Scale 1. Feeling understood and supported by a healthcare professional) and were the cluster with the highest number (43%) of new hypertension diagnoses. In contrast, Cluster H also had young participants (mean age 30 years) but with low education (72% illiterate) who scored lowest across all nine scales. While epidemiological analysis provided overall health literacy scores and associations between health literacy and other characteristics, cluster analysis provided nuanced health literacy profiles with the potential to inform development of solutions tailored to the needs of specific population subgroups.
聚类分析可以补充和扩展通过流行病学分析所获得的信息。本研究的目的是确定这两种数据分析方法在描述印度北部资源匮乏的农村社区多维健康素养优势和挑战方面的相对优点。采用健康素养问卷(HLQ)进行了一项横断面调查(N = 510)。描述性流行病学包括社会人口学特征中的平均得分和效应量。聚类分析基于HLQ的九个量表来确定人群中的不同健康素养概况。参与者在量表4“对健康的社会支持”(2.88)和量表6“与医疗保健专业人员积极互动的能力”(3.66)上报告的平均得分最高。在量表3“积极管理我的健康”(1.81)和量表8“寻找良好健康信息的能力”(2.65)上报告的得分较低。年轻人(<35岁)在社会支持方面的得分比老年人高得多(效应量>1.0)。识别出了八个聚类。在聚类A中,受过教育的年轻男性(平均年龄27岁)在除一个量表(量表1“感到被医疗保健专业人员理解和支持”)之外的所有量表上报告的得分都较高,并且是新高血压诊断数量最多(43%)的聚类。相比之下,聚类H也有年轻参与者(平均年龄30岁),但教育程度较低(72%为文盲),在所有九个量表上得分最低。虽然流行病学分析提供了总体健康素养得分以及健康素养与其他特征之间的关联,但聚类分析提供了细致入微的健康素养概况,有可能为针对特定人群亚组需求的解决方案的制定提供信息。