The immigrant paradox: foreign-born Latinos have lower rates of psychiatric disorders

Margarita Alegría, Glorisa Canino, Patrick Shrout, Meghan Woo, Naihua Duan, Doryliz Vila, Maria Torres, Chih-nan Chen, and Xiao-Li Meng, write:

Although widely reported among Latino populations, contradictory evidence exists regarding the generalizability of the immigrant paradox, i.e., that foreign nativity protects against psychiatric disorders. The authors examined whether this paradox applies to all Latino groups by comparing estimates of lifetime psychiatric disorders among immigrant Latino subjects, U.S-born Latino subjects, and non-Latino white subjects.

The authors combined and examined data from the National Latino and Asian American Study and the National Comorbidity Survey Replication, two of the largest nationally representative samples of psychiatric information.

In the aggregate, risk of most psychiatric disorders was lower for Latino subjects than for non-Latino white subjects. Consistent with the immigrant paradox, U.S.-born Latino subjects reported higher rates for most psychiatric disorders than Latino immigrants. However, rates varied when data were stratified by nativity and disorder and adjusted for demographic and socioeconomic differences across groups. The immigrant paradox consistently held for Mexican subjects across mood, anxiety, and substance disorders, while it was only evident among Cuban and other Latino subjects for substance disorders. No differences were found in lifetime prevalence rates between migrant and U.S.-born Puerto Rican subjects.


They conclude:

Caution should be exercised in generalizing the immigrant paradox to all Latino groups and for all psychiatric disorders. Aggregating Latino subjects into a single group masks significant variability in lifetime risk of psychiatric disorders, with some subgroups, such as Puerto Rican subjects, suffering from psychiatric disorders at rates comparable to non-Latino white subjects. Our findings thus suggest that immigrants benefit from a protective context in their country of origin, possibly inoculating them against risk for substance disorders, particularly if they emigrated to the United States as adults.

This is really important stuff, and it’s a big contribution to put all these data in the same place. That said, I have a couple of problems with the article. I actually know several of the authors, so maybe they can help me out with this. Anyway, here are my questions:

1. Why is this called a “paradox”? For something to be a paradox, don’t you need some sort of apparent contradiction? X, but also Y? Here, X is the lower rate of psychiatric disorders among foreign-born Latinos, but what’s Y? I’d have a better appreciation for the paradox if I understood why it was called a paradox. As they write, “These higher rates are not surprising, given that psychiatric disorders are more prevalent in the United States than in many other parts of the world.”

2. The difference between “significant” and “not significant” is not itself statistically significant. With that in mind, I’m skeptical of claims made in the article such as:

Overall, the immigrant paradox was only reliably observed for Mexican subjects, and only evident for depressive and anxiety disorders. However, the paradox was consistently observed among Mexican, Cuban, and other Latino subjects for substance disorders. No evidence for the immigrant paradox was found among Puerto Rican subjects.

The graph above (as well as the other graph shown in the paper) shows differences in a consistent direction for all groups. The differences are of different sizes, and maybe some are statistically significant and some are not, but, remember, the difference between . . .

3. Given the sample size, could they also disaggregate non-Latino whites? If not by ethnicity (maybe such information is unavailable from the survey), at least by state or region of the country, or urban/rural? Also blacks, no?

4. They used the National Latino and Asian American Study, so could they display results for Asians as well?

5. I expect that researchers in mental health would have other breadowns they’re interested in. I understand that it is not the responsibility of Alegria et al. to show all possible subsets of the population–their article is focused on the differences between U.S.-born and other Latinos, which is an important issue in its own right–but showing a few other groups might help put this into perspective, especially if it doesn’t really take much work. Maybe this can be done in the next article.

6. I wish they’d talked to me before making the final version of their graph above. The right way to do this is to label the lines directly rather than using a color code. I’d also zero the graph, but that’s a judgment call. Otherwise, the graph is pretty good.

Finally, let me emphasize that I’m not criticizing Alegria et al. for only presenting data without explanation. As a data-presenter myself, I have a lot of respect for work where researchers clearly describe what’s happening, providing a solid foundation so that they and others can dig deeper.

7 thoughts on “The immigrant paradox: foreign-born Latinos have lower rates of psychiatric disorders

  1. Regarding the graph, the lines are intended to communicate differences between two exclusive (and finite) states – native born or immigrant. By using lines, though, the graph implies some continuity between these states, in which the slope of the lines communicates the severity of differences between them. But graphically the physical distance between "US Born" and "Immigrant" is arbitrary. The slopes of the lines could be minimized or maximized whimsically (acknowledging their relative slopes would hold). Seems like points or columns would be more faithful to the data.

  2. I haven't read the article, but it seems to me that this may just be a selection artifact. That is, the US will (or now, won't) take "your tired, your poor," but won't take your psychiatrically disordered. Moreover, those with such disabilities may find it difficult to make the transition to a new country, and therefore are less likely to emigrate.

  3. Dear Andy,

    I am really glad that you pick up on this. I hope this reading can help you to better understand what I meant in the "desired and feared" article.

    Otherwise silently yours,


  4. I wonder to what extent the mediating factor is family structure & support. It might also be interesting to compare the size of the gaps between english-speaking immigrants and their native-born peers to non-english speaking immigrants and their peers (again on the social support assumption).

    I also wondered where the paradox was. I thought that on quite a few health-related outcomes (e.g., obesity, cancer rates), immigrants do better than their native-born counterparts, so if anything the paradox would be if immigrants had higher rates of psychiatric health probs.

  5. I've seen this effect in New Zealand data for an immigrant group I looked at.

    The recently immigrated were healthier compared to those who immigrated earlier or were born here (adjusted for age).

    I like the line plot even though histograms, for this kind of data, are politically correct. With the line plot you can overlay the data and do two comparisons at once (compare ethnicities or compare time). With the histogram you have to line up the bars side by side either 1) all the ethnic groups togerther for each time point or 2) the two time points side by side for each ethnic group.

    Isn't the paradox that people are going to USA to be better off but it turns out that, at least health wise, that for those born in America are worse off (for mental health).

    However, I suspect that some of the mental health conditions aren't recognised in the countries of origin – especially around alcohol and drug addiction.

  6. What about selection effects?

    If we model the places of origin as poor places that many people struggle to get out of, and into the (mainland) USA, and hypothesize that not all those who struggle succeed, then it follows that the winners in this struggle will be those with the best reserves of mental and physical health to call on. People in poorer health fall back and remain at home. So we would expect to find that the immigrant population consists of people in finer health than a random population chosen by accident of birth.

    Also, I bet US Immigration is detectably less kind to people who look like they have problems, and people in poor health are noticeably more likely to misplace vital paperwork.

    Finally, the new immigrants, once in the USA, may be less visible to the health monitoring system. If they're not yet diagnosed with a psychiatric disorder, it doesn't necessarily follow that they're not suffering from a psychiatric disorder.

Comments are closed.