Patients rated visit satisfaction and resident Spanish ability higher for curriculum-trained residents

Description #

Among the same 55 Spanish-speaking LEP ED patients, patient satisfaction with the ED visit was rated "very good" in 40% of encounters with South Campus (SC) residents (curriculum-trained) versus 25% of encounters with University Campus (UC) residents (overall 35%). Patients also rated the residents' Spanish-speaking ability as "very good" more often for SC residents — 17% versus 5% for UC (overall 13%). The direction favors the curriculum on both subjective measures, though no statistical testing was performed.

"Patients rated resident physician Spanish as very good in 13% of all encounters, with 17% for SC versus 5% for UC. Patient satisfaction with their ED visit was rated as very good in 40% of encounters with SC and 25% of encounters with UC resident physicians." (Stoneking, 2016)

Methods Context #

What? #ⓘ

The observables: (1) patient-rated satisfaction with the ED visit and (2) patient-rated resident Spanish-speaking ability, each dichotomized at the "very good" rating. Satisfaction was collected on a follow-up phone call ~1 week later; perceived Spanish ability was collected on the real-time visit survey.

"Follow-up phone calls were performed to assess patient satisfaction and adherence to discharge instructions ~1 week later." (Stoneking, 2016)

"Our survey also collected data on patient perception of resident physician Spanish-speaking ability." (Stoneking, 2016)

How? #ⓘ

Cross-sectional convenience sample contrasting the curriculum-exposed SC program against the non-exposed UC program via day-of-week scheduling. Real-time Spanish surveys administered after discharge/transfer, plus a follow-up phone call by a Spanish- and English-proficient resident or medical student. Results reported as percentages with 95% CIs (Jeffreys method); no formal statistical testing.

"For these two after-visit questions, a resident physician or medical student, all proficient in Spanish and English, called the patient at the listed home phone number." (Stoneking, 2016)

Who? #ⓘ

55 patients who self-identified as primarily Spanish-speaking at registration in an urban academic ED (~45,000 visits/year) in Tucson, Arizona, treated by emergency medicine residents from the SC (n=35) and UC (n=20) programs.

"patients who self-identified as primarily Spanish-speaking in our electronic health record during registration were asked to complete a real-time survey regarding their visit" (Stoneking, 2016)

Other Notes #

The authors had anticipated a larger satisfaction difference given the added cultural competency training, but did not separately measure cultural awareness/competency.

Caveats #

• Single-center underpowered pilot with no statistical testing limits the medical Spanish curriculum findings This was a single-center quality-improvement pilot at one urban academic ED, limiting generalizability. An EHR default that auto-set primary language to English under-captured enrollment, so the sample (55 patients) was not large enough to detect a significant difference, and the adherence comparison rested on only 13 patients (10 SC vs 3 UC) — too few to draw statistically significant meaning. No formal statistical testing or power analysis was performed, and program exposure could be contaminated (some SC/UC mixing on afternoons after didactics; possible use of family members or uncertified residents attempting Spanish), so the curriculum-vs-no-curriculum contrast is confounded and statistically fragile.