We observed that patients requiring IS had significantly higher patterns of inpatient
utilization (ED visits and hospitalizations) compared with patients who did not require
IS. Our findings are consistent with those of previous studies that documented higher
inpatient utilization among patients with LEP for psychiatric disorders 22], coronary artery disease, some surgical syndromes 21], and pediatric ED visits 31].
Persons with LEP are heterogeneous with regards to culture, ethnicity, race, and sociodemographic
factors 2]. Therefore, the reasons underlying increased inpatient utilization are likely multifaceted.
Furthermore, this study cannot fully assess whether the excess utilization is “too
much†care or the correct amount of care. Nevertheless, this study provides important
objective findings of utilization in the context of existing literature around determinants
of inpatient and emergency room utilization among patients with LEP.
Because most patients in our study had health insurance, were empanelled to a primary
care practice, and were regular utilizers of the outpatient practice, some of the
traditional barriers to health care access and availability cannot explain our findings.
Organizational solutions to promote more efficient health care utilization must consider
patient factors that frequently coexist with LEP, including low socioeconomic position,
preexisting health care norms, and low health literacy 32]–34].
Healthcare-seeking behaviors among patients with LEP may be influenced by the norms
of their countries of origin. For example, the notion of chronic disease management
and preventive care may be unfamiliar to some patients coming from a region where
healthcare is defined as an acute care model 35], 36]. This may be associated with delays in seeking care, and as noted in one pediatric
ED, LEP patients were more likely to be triaged to higher acuity, which led to hospitalization
37]. Our study lacks data on immigration status, which has been postulated to impact
utilization of healthcare services, leading to delay in seeking care and recourse
to ED visits 38]. However, other work suggests that undocumented immigrants in the US have similar
levels of ED use to other immigrant and non-immigrant groups 39].
Finally, LEP and low health literacy are interrelated and often occur together 40]. LEP is associated with lower health literacy across different diseases, ethnicities,
and ages 41]. Health literacy, in turn, has been independently linked to health care utilization
42], 43], and lower health literacy is associated with an inefficient mix of services, leading
to higher health care costs 44]. Nevertheless, one study of 48,000 patients showed LEP to be an even more important
risk factor for poor health than low health literacy 40]. Thus, in addition to language considerations, interventions aimed at improving the
efficiency of health care utilization among patients with LEP should also incorporate
principles of messaging and communication for patients with low health literacy 37], 45], 46].
Communication barriers and unmet health care needs may help explain the increased
ED visits among patients with LEP 7]. ED visits among patients empaneled to a primary care practice are frequently preceded
by communication between the patient and the outpatient health care team. Patients
with LEP may be less likely to initiate this communication (typically a telephone
call) if IS are less consistent. Indeed, language barriers during emergency telephone
communications can negatively affect communication and care outcomes 47]. Likewise, provision of language-concordant outpatient providers for patients with
diabetes mellitus in one study resulted in reduced ED visits and hospitalizations
48].
ED utilization may represent unmet health care needs among patients withLEP. Primary
care practices are increasingly developing medical home initiatives for care coordination,
integrated behavioral health, and care management for patients with complex medical
problems 39], 49]. However, if special efforts are not taken, these programs may inadvertently exclude
patients with LEP, who have difficulty navigating these complex systems-within-systems
50]. Therefore, ED visits may be a mechanism by which patients with LEP disproportionately
address these unmet health care needs 51], 52]. This conclusion is further supported by our finding of higher ED utilization among
IS patients for dental, eye, skin, and ENT concerns, systems that are typically addressed
in the outpatient setting. Primary care practices should aim to systematically identify
and manage patients with LEP who frequently utilize the ED and hospital, while improving
communication to patients at these critical transitions that are linguistically, culturally,
and health literacy–level appropriate.
Our finding of increased hospitalizations among IS patients is compelling in that
the decision to be hospitalized is influenced largely by diagnostic circumstance and
the decision making of the admitting physician, rather than the decision making of
the patient. Communication between patients and providers is a key factor in the evaluation
of patients at the point of care. Compared with English-proficient patients, more
tests are ordered for patients with LEP who present to the ED with abdominal pain
53] and acute respiratory illnesses, and patients with LEP are more likely to receive
antibiotics than non-LEP patients 54]. The decision to order extra tests or more aggressive therapy by ED providers may
be influenced by a need to compensate for communication barriers, and this approach
may then extend to the decision about whether a patient should be hospitalized. One
study of pediatric patients showed higher admission rates among patients with LEP
compared with English-proficient patients, even where acuity was similar at presentation
55].
Our study has several limitations. First, it was retrospective and relied on medical
records. However, we had minimal missing data, and charts with any ED visit and hospitalization
were reviewed to confirm the event. It is conceivable that ED visits and hospitalizations
outside the 3 main local hospitals would be missed; however, we suspect such events
would be minimal among these community-dwelling primary care patients. Limitations
of our administrative dataset precluded the assessment of potentially important confounding
variables such as socioeconomic position and health literacy. The use of IS need as
a proxy for LEP is incomplete and represents only a subset of patients who truly have
LEP 49]. Furthermore, the fact that IS status was assessed by self-report may have led to
misclassification of patients. In addition, we are not able to verify the percentage
of eligible patients who received IS services during health care events, though institutional
policy dictates that professional interpreters participate in every clinical encounter.
Also, insurance status, language and race/ethnicity are highly correlated and LEP
is a marker for these characteristics as well; therefore, it is not possible to separate
out the individual effects of these factors in our study. While we did calculate the
Charlton Comorbidity Index for each patient and incorporated it into our analyses,
we do not have access to data regarding the acuity of conditions that prompted each
utilization event. Finally, this study was conducted among patients and institutions
in a single geographic region within a medical home, with implications for generalizability
to other primary care practices. Likewise, these results may not be applicable to
practices with much higher percentages of IS patients that may have systems in place
that specifically target inpatient utilization among patients with LEP.