Psychiatric Epidemiology and Social Issues Unit, Centre for Research on Ageing, Health and Wellbeing, The Australian National University, Building 63 Eggleston Road, Acton, ACT, Australia.
BMC Health Serv Res. 2012 Feb 14;12:34. doi: 10.1186/1472-6963-12-34.
National screening programs have reduced cervical cancer mortality; however participation in these programs varies according to women's personal and social characteristics. Research into these inequalities has been limited by reliance on self-reported service use data that is potentially biased, or administrative data that lacks personal detail. We address these limitations and extend existing research by examining rates and correlates of cervical screening in a large epidemiological survey with linked administrative data.
The cross-sectional sample included 1685 women aged 44-48 and 64-68 years from the Australian Capital Territory and Queanbeyan, Australia. Relative risk was assessed by logistic regression models and summary Population Attributable Risk (PAR) was used to quantify the effect of inequalities on rates of cervical cancer screening.
Overall, 60.5% of women participated in screening over the two-year period recommended by Australian guidelines. Screening participation was associated with having children, moderate or high use of health services, employment, reported lifetime history of drug use, and better physical functioning. Conversely, rates of cervical screening were lower amongst women who were older, reliant on welfare, obese, current smokers, reported childhood sexual abuse, and those with anxiety symptoms. A summary PAR showed that effective targeting of women with readily observable risk-factors (no children, no partner, receiving income support payments, not working, obese, current smoker, anxiety, poor physical health, and low overall health service use) could potentially reduce overall non-participation in screening by 74%.
This study illustrates a valuable method for investigating the personal determinants of health service use by combining representative survey data with linked administrative records. Reliable knowledge about the characteristics that predict uptake of cervical cancer screening services will inform targeted health promotion efforts.
国家筛查计划降低了宫颈癌死亡率;然而,这些计划的参与率因妇女的个人和社会特征而异。对这些不平等现象的研究受到依赖潜在有偏差的自我报告服务使用数据或缺乏个人详细信息的行政数据的限制。我们通过检查在具有关联行政数据的大型流行病学调查中进行宫颈癌筛查的比率和相关性,解决了这些限制并扩展了现有研究。
横断面样本包括来自澳大利亚首都领地和澳大利亚昆比恩的 1685 名 44-48 岁和 64-68 岁的女性。通过逻辑回归模型评估相对风险,使用总体归因风险(PAR)来量化不平等对宫颈癌筛查率的影响。
总体而言,60.5%的女性在澳大利亚指南建议的两年内参加了筛查。筛查参与与有孩子、中高健康服务使用、就业、报告终生药物使用史以及更好的身体功能有关。相反,在年龄较大、依赖福利、肥胖、当前吸烟者、报告童年性虐待以及有焦虑症状的女性中,宫颈癌筛查的比率较低。总体归因风险的汇总显示,有针对性地针对具有明显风险因素的妇女(没有孩子、没有伴侣、领取收入支持金、不工作、肥胖、当前吸烟者、焦虑、身体不健康和整体健康服务使用率低)可能会降低整体筛查参与率 74%。
本研究通过将代表性调查数据与关联的行政记录相结合,展示了一种调查健康服务使用个人决定因素的有价值方法。可靠地了解预测宫颈癌筛查服务利用率的特征将为有针对性的健康促进工作提供信息。