Philip R. Lee Institute for Health Policy Studies, University of California, San Francisco, 3333 California St., Ste. 265, Box 0936, San Francisco, CA, 94118, USA.
Division of Tuberculosis Elimination, Centers for Disease Control and Prevention, Mailstop E-10, 1600 Clifton Road, Atlanta, GA, 30333, USA.
BMC Public Health. 2018 Apr 12;18(1):485. doi: 10.1186/s12889-018-5393-x.
Homelessness increases the risk of tuberculosis (TB) disease and latent TB infection (LTBI), but persons experiencing homelessness often lack access to testing and treatment. We assessed the yield of TB testing and linkage to care for programs targeting homeless populations in the United States.
We conducted a comprehensive search of peer-reviewed and grey literature, adapting Cochrane systematic review methods. Two reviewers independently assessed study eligibility and abstracted key data on the testing to care cascade: number of persons reached, recruited for testing, tested for LTBI, with valid test results, referred to follow-up care, and initiating care. We used random effects to calculate pooled proportions and 95% confidence intervals (CI) of persons retained in each step via inverse-variance weighted meta-analysis, and cumulative proportions as products of adjacent step proportions.
We identified 23 studies published between 1986 and 2014, conducted in 12 states and 15 cities. Among studies using tuberculin skin tests (TST) we found that 93.7% (CI 72.4-100%) of persons reached were recruited, 97.9% (89.3-100%) of those recruited had tests placed, 85.5% (78.6-91.3%) of those with tests placed returned for reading, 99.9% (99.6-100%) of those with tests read had valid results, and 24.7% (21.0-28.5%) with valid results tested positive. All persons testing positive were referred to follow-up care, and 99.8% attended at least one session of follow-up care. Heterogeneity was high for most pooled proportions. For a hypothetical cohort of 1000 persons experiencing homelessness reached by a targeted testing program using TST, an estimated 917 were tested, 194 were positive, and all of these initiated follow-up care.
Targeted TB testing of persons experiencing homelessness appears effective in detecting LTBI and connecting persons to care and potential treatment. Future evaluations should assess diagnostic use of interferon gamma release assays and completion of treatment, and costs of testing and treatment.
无家可归会增加患结核病（TB）和潜伏性结核感染（LTBI）的风险，但无家可归者往往无法接受检测和治疗。我们评估了针对美国无家可归人群的项目中结核病检测和与治疗关联的效果。
我们对同行评审和灰色文献进行了全面搜索，采用 Cochrane 系统评价方法。两名审查员独立评估了研究的合格性，并提取了检测到护理级联的关键数据：接触到的人数、接受检测的人数、接受 LTBI 检测的人数、具有有效检测结果的人数、被转介到后续护理的人数以及开始护理的人数。我们使用随机效应计算了通过逆方差加权荟萃分析保留在每个步骤中的个体的汇总比例和 95%置信区间（CI），并通过相邻步骤比例的乘积计算累积比例。
我们确定了 1986 年至 2014 年期间发表的 23 项研究，这些研究在 12 个州和 15 个城市进行。在使用结核菌素皮肤试验（TST）的研究中，我们发现 93.7%（72.4-100%）接触到的人被招募，97.9%（89.3-100%）招募的人接受了检测，85.5%（78.6-91.3%）接受了检测的人返回进行检测，99.9%（99.6-100%）进行了检测的人有有效的检测结果，24.7%（21.0-28.5%）有有效检测结果的人检测呈阳性。所有检测呈阳性的人都被转介到后续护理，99.8%的人至少参加了一次后续护理。大多数汇总比例的异质性很高。对于一个通过 TST 进行目标性检测的 1000 名无家可归者的假设队列，估计有 917 人接受了检测，194 人呈阳性，所有这些人都开始接受后续护理。
针对无家可归者的目标性结核病检测似乎可以有效检测 LTBI，并将个体与护理和潜在治疗联系起来。未来的评估应该评估干扰素γ释放试验的诊断用途和治疗完成情况，以及检测和治疗的成本。
