Effects of non-health-targeted policies on migrant health: a systematic review and meta-analysis

Abstract

Background: Government policies can strongly influence migrants' health. Using a Health in All Policies approach, we systematically reviewed evidence on the impact of public policies outside of the health-care system on migrant health.

Methods: We searched the PubMed, Embase, and Web of Science databases from Jan 1, 2000, to Sept 1, 2017, for quantitative studies comparing the health effects of non-health-targeted public policies on migrants with those on a relevant comparison population. We searched for articles written in English, Swedish, Danish, Norwegian, Finnish, French, Spanish, or Portuguese. Qualitative studies and grey literature were excluded. We evaluated policy effects by migration stage (entry, integration, and exit) and by health outcome using narrative synthesis (all included studies) and random-effects meta-analysis (all studies whose results were amenable to statistical pooling). We summarised meta-analysis outcomes as standardised mean difference (SMD, 95% CI) or odds ratio (OR, 95% CI). To assess certainty, we created tables containing a summary of the findings according to the Grading of Recommendations Assessment, Development, and Evaluation. Our study was registered with PROSPERO, number CRD42017076104.

Findings: We identified 43 243 potentially eligible records. 46 articles were narratively synthesised and 19 contributed to the meta-analysis. All studies were published in high-income countries and examined policies of entry (nine articles) and integration (37 articles). Restrictive entry policies (eg, temporary visa status, detention) were associated with poor mental health (SMD 0·44, 95% CI 0·13-0·75; I²=92·1%). In the integration phase, restrictive policies in general, and specifically regarding welfare eligibility and documentation requirements, were found to increase odds of poor self-rated health (OR 1·67, 95% CI 1·35-1·98; I²=82·0%) and mortality (1·38, 1·10-1·65; I²=98·9%). Restricted eligibility for welfare support decreased the odds of general health-care service use (0·92, 0·85-0·98; I²=0·0%), but did not reduce public health insurance coverage (0·89, 0·71-1·07; I²=99·4%), nor markedly affect proportions of people without health insurance (1·06, 0·90-1·21; I²=54·9%).

Interpretation: Restrictive entry and integration policies are linked to poor migrant health outcomes in high-income countries. Efforts to improve the health of migrants would benefit from adopting a Health in All Policies perspective.

Funding: Swedish Council for Health, Working Life, and Social Research; UK Medical Research Council; Scottish Government Chief Scientist Office.

Figure 1

Article selection process

Figure 2

Random-effects meta-analysis of the effects of entry policies on mental health among migrants Figure includes information on policy and comparison; study reference; migrant population and host country; and specific health outcome and measurement instrument. Fully-adjusted estimates were included from each study, with adjustment variables varying by study (data not shown). CES-D=Center for Epidemiological Studies-Depression Scale. HSCL-25=Hopkins Symptom Checklist (25 item). IRCA=Immigration Reform and Control Act. PTSD=post-traumatic stress disorder. RATS=Reactions of Adolescents to Traumatic Stress Inventory. SF-12=Short Form Health Survey (12 item). SMD=standardised mean difference.

Figure 3

Random-effects meta-analysis of the effects of general-integration and documentation policies on self-rated health (A) and all-cause mortality (B) among migrants Figure includes information on policy and comparison; study reference; migrant population, host country, and population counterfactual (if applicable). Fully-adjusted estimates were included from each study, with adjustment variables varying by study (data not shown). MIPEX=Migrant Integration Policy Index. OR=odds ratio. SB 1070=Arizona's Support Our Law Enforcement and Safe Neighborhoods Act.

Figure 4

Random-effects meta-analysis of the effects of US welfare restrictions on health-insurance (Medicaid) enrolment (A), the odds of being uninsured (B), health-care service use (C), and prenatal care use (D) among migrants Figure includes information on policy and comparison; study reference; migrant population, US state context, and population counterfactual; and specific health outcome (if applicable). Fully-adjusted estimates were included from each study, with adjustment variables varying by study (data not shown). DRA=Deficit Reduction Act. IIRAIRA=Illegal Immigration Reform and Immigrant Responsibility Act. OR=odds ratio. PRWORA=Personal Responsibility and Work Opportunity Reconciliation Act. SB 1070=Arizona's Support Our Law Enforcement and Safe Neighborhoods Act.

Figure 4

Random-effects meta-analysis of the effects of US welfare restrictions on health-insurance (Medicaid) enrolment (A), the odds of being uninsured (B), health-care service use (C), and prenatal care use (D) among migrants Figure includes information on policy and comparison; study reference; migrant population, US state context, and population counterfactual; and specific health outcome (if applicable). Fully-adjusted estimates were included from each study, with adjustment variables varying by study (data not shown). DRA=Deficit Reduction Act. IIRAIRA=Illegal Immigration Reform and Immigrant Responsibility Act. OR=odds ratio. PRWORA=Personal Responsibility and Work Opportunity Reconciliation Act. SB 1070=Arizona's Support Our Law Enforcement and Safe Neighborhoods Act.