MacAllister Sean, Miller Matthew, Swanson Sonja
Department of Epidemiology, University of Pittsburgh School of Public Health, 130 De Soto St, Pittsburgh, PA, 15261, USA.
Department of Public Health and Health Sciences, Northeastern University Bouvé College of Health Sciences, 360 Huntington Avenue, Boston, MA, 02115, USA.
Inj Epidemiol. 2025 Apr 23;12(1):21. doi: 10.1186/s40621-025-00577-x.
Recent publications on Child Access Prevention (CAP) laws suggest substantial protective effects on adolescent firearm suicide. However, these studies have also found comparable protective effect estimates on adolescent non-firearm suicide and adult firearm suicide, which may indicate residual confounding. Here we apply bias analysis techniques to assess the effects of CAP laws while accounting for potential unmeasured sources of bias using a negative control approach.
Using established bias formulas, we bias-adjust previously published point estimates and their 95% confidence intervals (CI) assuming that an arbitrary confounder biases all suicide-related effect estimates and that adolescent non-firearm suicide and adult firearm suicide are negative controls. Negative controls are outcomes or populations that prior subject matter suggests should not be meaningfully affected by the exposure and can be used to better understand and sometimes account for bias in the primary exposure-outcome relationship.
After bias adjustments, effect estimates were attenuated, with many of the confidence intervals including the null. Assuming that adolescent non-firearm suicide is a negative control outcome and taking a published point estimate as the bias parameter, the bias-adjusted effect estimate for adolescent firearm suicide decreased from an incidence rate ratio of 0.87 (95% CI: 0.78, 0.97) to 0.95 (95% CI: 0.85, 1.07). When adult firearm suicide was used as the negative control, the bias-adjusted estimate was 0.92 (95% CI: 0.82, 1.03).
Our findings suggest that CAP laws may have had a smaller public health impact on adolescent suicide than previously estimated. Given the strong evidence that reducing access to firearms can prevent suicide deaths, and that secure storage helps reduce access for many children, our findings underscore the need to continue to identify and promote effective ways to motivate adults to make household firearms inaccessible to children.
近期关于儿童接触预防(CAP)法律的出版物表明,这些法律对青少年枪支自杀具有显著的保护作用。然而,这些研究还发现,对青少年非枪支自杀和成人枪支自杀也有类似的保护作用估计,这可能表明存在残余混杂因素。在此,我们应用偏倚分析技术来评估CAP法律的效果,同时使用阴性对照方法考虑潜在的未测量偏倚来源。
使用既定的偏倚公式,我们对先前发表的点估计值及其95%置信区间(CI)进行偏倚调整,假设一个任意的混杂因素使所有与自杀相关的效应估计值产生偏倚,并且青少年非枪支自杀和成人枪支自杀为阴性对照。阴性对照是指先前的主题研究表明不应受到暴露有意义影响的结局或人群,可用于更好地理解并有时解释主要暴露-结局关系中的偏倚。
经过偏倚调整后,效应估计值减弱,许多置信区间包含无效值。假设青少年非枪支自杀是阴性对照结局,并将已发表的点估计值作为偏倚参数,青少年枪支自杀的偏倚调整效应估计值从发病率比0.87(95%CI:0.78,0.97)降至0.95(95%CI:0.85,1.07)。当将成人枪支自杀用作阴性对照时,偏倚调整估计值为0.92(95%CI:0.82,1.03)。
我们的研究结果表明,CAP法律对青少年自杀的公共卫生影响可能比先前估计的要小。鉴于有强有力的证据表明减少枪支获取可预防自杀死亡,且安全储存有助于减少许多儿童获取枪支的机会,我们的研究结果强调需要继续识别并推广有效的方法,以促使成年人使儿童无法接触到家庭枪支。