Frequently Asked Questions to Accompany the Estimates of at Least the Top 15 Languages Spoken by Individuals with Limited English Proficiency under Section 1557 of the Affordable Care Act (ACA)

1. Regulations implementing Section 1557 of the ACA require each covered entity to “post taglines in at least the top 15 languages spoken by individuals with limited English proficiency of the relevant State or States.” 45 C.F.R. § 92.8(d)(1). What are these languages in each State?

As a resource for covered entities, OCR has made available a table displaying OCR’s list of the top 15 languages spoken by individuals with limited English proficiency (LEP) in each State, the District of Columbia, Puerto Rico and each U.S. Territory. OCR created this list for covered entities’ use in identifying languages in which to provide translated taglines in implementing §92.8(d)(1)-(2). OCR’s list is organized first by State with the U.S. Territories at the end of the table and then by language. Languages estimated to be spoken by the most individuals with LEP are ranked as number 1.

Covered entities may use the information in this list to implement §92.8(d)(1)-(2), although nothing in the Section 1557 regulation requires covered entities to use OCR’s particular resource in doing so. In implementing §92.8(d)(1)-(2), covered entities may refer to sources other than OCR’s list if covered entities have a reasonable basis for relying on such sources when considering characteristics such as the currency, reliability, and stability of the data. Covered entities may use such sources even if the list of languages produced from those sources is different from OCR’s list or has variations in the relative rank of the languages.

As a reminder, nothing in the Section 1557 regulation bars a covered entity from including taglines in languages beyond those triggered by §92.8(d)(1)-(2). For instance, a covered entity may choose to include taglines in additional languages to maximize the breadth of national origin populations informed about the availability of language assistance services.

Please be aware that providing taglines as §92.8(d)(1)-(2) requires does not fulfill the covered entity’s obligation for complying with the prohibition of national origin discrimination under Section 1557 and the rule. Under §92.201(a), covered entities must also take reasonable steps to provide meaningful access to each individual with LEP eligible to be served or likely to be encountered in the entity’s health programs or activities. Independent of the tagline requirement at §92.8(d)(1)-(2), covered entities are responsible for providing timely and accurate language assistance services, such as oral interpretation and written translation, in non-English languages, even if the language is not shown on OCR’s list, when doing so is a reasonable step to provide meaningful access to an individual with LEP.

View the table created for OCR.

2. What data did OCR use to compile its list of the top 15 languages spoken by individuals with LEP in each of the 50 States, the District of Columbia, and Puerto Rico?

OCR used the U.S. Census Bureau’s (Bureau) American Community Survey’s (ACS) data set entitled “Language Spoken at Home by Ability to Speak English for the Population 5 Years and Older.” The tables for this data set display, among other variables and geographic units, the estimated number of individuals, who speak non-English languages in each State,who speak English less than “very well,” which OCR used as a proxy for the population with LEP

OCR used the most recent State-based data set available to OCR during the development of the Section 1557 rule – the 2014 ACS 5-year estimates, available as one of the Bureau’s American FactFinder tables.[i] The American FactFinder tables include estimates on a total of 39 individual languages and language groups. The language groups on the American FactFinder tables “bundle” more than one language in the groups’ estimates. For instance, the American FactFinder tables estimate the prevalence of individuals who speak English less than “very well” who speak “Other Indic languages,” “Other Pacific Island Languages,” “African Languages,” and “Other and unspecified languages,” among other groups.

For these groups where more detail was needed to identify the languages represented, OCR used the detailed State-based data set from the 2013 ACS 5-year estimates – which were the most recent estimates available during the development of the Section 1557 rule.[ii] The detailed tables display approximately 380 languages and language groups.

On OCR’s list, an asterisk after the language denotes that the estimate came from the Bureau’s detailed 2013 ACS 5-year estimates rather than from the 2014 ACS 5-year estimates.

3. Why did OCR rely on the ACS 5-year estimates rather than the 1-year estimates?

OCR used the 5-year estimates (which average five years of data) rather than the 1-year estimates (which represent one year of data). This choice best balanced the currency of the estimates with their reliability and stability. In analyzing small populations, such as some LEP populations, a data source that averages five years of data is more reliable and stable than a source that includes only one year of data.[iii]

4. What data did OCR use to compile its estimates of the top languages spoken by individuals with LEP for each of the U.S. Territories, other than Puerto Rico?

For the U.S. Territories other than Puerto Rico, OCR used the data available from the U.S. Census Bureau, 2010 Census of Population and Housing.[iv] OCR used this data source instead of the ACS because the ACS does not include data on the U.S. Territories, other than Puerto Rico.

The language data available from the 2010 Census of Population and Housing for the U.S. Territories do not allow for a one-to-one comparison to the data available from the Bureau’s ACS for the 50 States, the District of Columbia, and Puerto Rico.  In contrast to the Bureau’s ACS data estimating the number of individuals who speak non-English languages who speak English less than “very well,” the data available for Guam, the Commonwealth of Northern Mariana Islands, and American Samoa show the number of individuals who speak a non-English language.  These estimates do not indicate how many individuals who speak a non-English language also speak English and whether that English-speaking ability is limited.  The language estimates available for the U.S. Virgin Islands show individuals who speak a non-English language who speak English “not well.”  These estimates do not indicate how many individuals speak a particular non-English language who also speak English less than “very well.”

5. Did OCR make any technical adjustments to the data? For example, how did OCR address the fact that even the detailed 2013 ACS 5-year estimates by State contain information on language groups, as opposed to individual languages?

In compiling its list for covered entities’ use in implementing § 92.8(d)(1)-(2), OCR made two main types of technical adjustments to the data OCR used. First, OCR omitted languages from the detailed language tables that still represented a language group rather than a single language or list of single languages for which covered entities could identify a specific language in which to provide a translated tagline in implementing §92.8(d)(1)-(2). This technical adjustment resulted in the omission of language groups, such as “African” and “Mayan language,” from OCR’s list. Second, OCR omitted any spoken languages that do not have a written equivalent in which a translated tagline could be provided.  This technical adjustment resulted in the omission of specific languages, such as Crow, Dakota, and Inupik, from OCR’s list.

These technical adjustments made by OCR are limited to the provision of this particular resource. Consequently, the omission from this specific resource of language groups such as “African” and “Mayan language” does not relieve a covered entity from the separate obligation to take reasonable steps to provide meaningful access to an individual with LEP whose primary language for communication is a specific African or Mayan language. Similarly, the omission from this specific resource of spoken languages that do not have a written equivalent does not relieve a covered entity from the separate obligation to take reasonable steps to provide meaningful access to an individual with LEP whose primary language for communication is a spoken language that does not have a written equivalent.

6. Why do some of the States on OCR’s list have more than 15 languages listed?

Four States (Colorado, Maryland, Rhode Island, and Virginia) and the District of Columbia have 17 languages listed because the detailed 2013 ACS language tables identify an estimate of language speakers for a grouping of three languages – Kru, Ibo, and Yoruba. The detailed ACS language estimates do not further disaggregate the number of individuals speaking English less than “very well” who speak Kru, Ibo, and Yoruba, respectively.

Although all three languages are grouped as one entry on the detailed 2013 ACS 5-year language tables with one combined estimate of language speakers, OCR counted each of these three languages separately because they are distinct languages. As a result, these four States and the District of Columbia have 15 estimates displayed but more than 15 languages listed. OCR recommends that covered entities that rely on OCR’s list in implementing § 92.8(d)(1)-(2) and that serve individuals in these States or the District of Columbia post taglines in 17 languages. This recommendation is informed by the purpose underlying the ACA to expand access to health care, reduce barriers, and address health disparities.

7. In OCR’s list, why are there significantly fewer than 15 languages listed for each of the U.S. Territories other than Puerto Rico? For instance, only one language is listed for Guam.

OCR has listed fewer than 15 languages for the U.S. Territories other than Puerto Rico because the data OCR used to estimate the languages spoken by individuals with LEP in the U.S. Territories other than Puerto Rico identify a mere handful of languages or language groups.

Using Guam as an example, the table of estimates for “Language Spoken at Home, Frequency of English Usage, and Sex by Ethnic Origin or Race” from the 2010 Census of Population and Housing provides data on the number of individuals in Guam who speak one language (Chamorro) and the four language groups: “Philippine languages,” “Other Pacific Island languages,” “Asian languages,” and “Other languages.”[v] Because the language groups listed are not further disaggregated to identify the specific languages into which a tagline or other materials could be translated, OCR made technical adjustments to omit the aggregated language groups. Consequently, the only individual language estimated to be spoken by individuals with LEP in Guam that OCR could identify for inclusion in this specific resource is Chamorro. Covered entities operating health programs or activities that serve individuals in the U.S. Territories may be in a better position to identify specific languages spoken by individuals with LEP in which to provide translated taglines to meet their obligations under §92.8(d)(1)-(2).

The technical adjustments made by OCR are limited to the provision of this particular resource. Consequently, the omission from this specific resource of language groups, such as “Philippine languages,” “Other Pacific Islands languages,” and “Asian languages” for Guam, does not relieve a covered entity that serves individuals in Guam through a health program or activity from the separate obligation under §92.201(a) to take reasonable steps to provide meaningful access to an individual with LEP whose primary language for communication is not shown on OCR’s list.

8. Some languages listed in the table that OCR has made available have more than one spoken or written dialect, which means that the grammar, vocabulary, pronunciation, or usage may differ for the same language. For these languages, into which dialects did OCR translate its sample taglines and other materials?

For the following languages with more than one dialect, OCR translated its sample taglines and other materials into the dialects as follows:

• For Spanish, materials are translated into neutral Spanish for the United States.

• For Chinese, materials are translated into traditional Chinese.

• For French Creole, materials are translated into Haitian Creole.

• For French, materials are translated into the European dialect.

• For Portuguese, materials are translated into the European dialect.

• For Persian, materials are translated into Farsi.

• For Cushite, materials are translated into Oromo.

• For Serbo-Croatian, materials are translated into Serbian.

• For Syriac, materials are translated into Assyria.

• For Kru, materials are translated into Bassa.

• For Micronesian, materials are translated into Pohnpeian.

• For Bantu, materials are translated into Kirundi.

• For Sudanic, materials are translated into Fulfulde.

• For Nilotic, materials are translated into Dinka.

9. The population with LEP may change in future years, affecting the languages spoken by individuals with LEP that are ranked within the top 15. Will OCR update its list of the top 15 languages in the future based on newer data? If so, will OCR release the sample tagline, sample notice of nondiscrimination, and sample nondiscrimination statement in languages not captured within the 64 languages in which OCR has already made such translated materials available?

The Section 1557 rule does not specify when to revisit the languages triggered by the standard in §92.8(d)(1)-(2). As newer data from the Bureau's American Community Survey becomes available for the dataset entitled “Language Spoken at Home by Ability to Speak English for the Population 5 Years and Older,” OCR will determine if and when the standard in §92.8(d)(1)-(2) triggers languages in addition to the 64 already triggered by this standard. When additional languages are warranted, OCR will make available the sample tagline, nondiscrimination notice, and nondiscrimination statement in the additional non-English languages.

Endnotes