NEPL retained increased odds of emergency surgery (OR 1.59) among commercially insured diverticulitis patients

Description #

In a sensitivity analysis restricted to commercially insured patients only, NEPL retained a significantly higher odds of emergency versus elective surgery for diverticulitis compared to EPL (OR 1.59; 95% CI 1.23–2.04; P < 0.001). Within this insurance-homogeneous subset, non-Spanish NEPL patients again had significantly increased odds (OR 1.84; 95% CI 1.33–2.55; P < 0.001), while Spanish NEPL patients showed no significant difference (OR 1.27; 95% CI 0.86–1.88). The authors read this as evidence that insurance status is not the primary driver of the language association.

"On multivariable analysis among patients with commercial insurance only, patients with NEPL had a significantly higher odds of emergency surgery versus patients with EPL (OR 1.59; 95% CI 1.23-2.04; P <0.001). Non-Spanish NEPL patients also continued to have a significantly increased odds of emergency surgery (OR 1.84; 95% CI 1.33-2.55; P <0.001), with no significant difference for Spanish NEPL patients (OR 1.27; 95% CI 0.86-1.88)." (Maurer, 2021, p. 645)

Methods Context #

What? #ⓘ

The observable: emergency (urgent/emergent) versus elective surgical admission type, estimated within the commercially insured stratum only.

"A sensitivity analysis was also performed only among patients with commercial insurance to determine the odds of emergency surgery in NEPL versus EPL patients, in addition to Spanish NEPL and non-Spanish NEPL versus EPL patients." (Maurer, 2021, p. 645)

How? #ⓘ

Multivariable logistic regression identical in structure to the primary model but restricted to patients with commercial insurance, isolating the language effect from insurance-type confounding.

"A sensitivity analysis was also performed only among patients with commercial insurance to determine the odds of emergency surgery in NEPL versus EPL patients." (Maurer, 2021, p. 645)

Who? #ⓘ

The commercially insured subset of the 9,453-patient New Jersey diverticulitis surgery cohort; overall 5,507 (58.3%) of patients had commercial insurance, of whom 265 were NEPL.

"Patients with NEPL were less likely to have commercial insurance (44.8% NEPL versus 59.2% EPL; P <0.001)" (Maurer, 2021, p. 645)

Other Notes #

Robustness/mechanism finding: because commercial insurance is held constant, the persisting NEPL association argues against insurance access as the sole explanation and supports language itself as a contributor. The Spanish NEPL point estimate rises to 1.27 here (from 0.80 overall) but remains non-significant.

Caveats #

• Retrospective single-database diverticulitis study precludes causal inference and leaves residual confounding The study is a retrospective analysis of a single administrative claims database, so it can establish association but not causation between non-English primary language and emergency surgery. Residual confounding is likely on two additional counts the authors flag: income was available only as an ecological, neighborhood-level (FIPS) measure rather than individual income, and the State Inpatient Database offers no longitudinal patient linkage, so prior diverticulitis admissions and the true denominator of overall disease burden per patient are unknown. Together these limit adjustment for socioeconomic status and disease chronicity, so the adjusted odds ratios should be read as associational estimates that may retain uncontrolled confounding.
• NEPL is a proxy for primary language, not measured English proficiency, biasing the diverticulitis association toward the null The exposure is recorded primary language, not measured English proficiency: the database indicates only that a patient's primary language is not English, not their actual fluency. Some NEPL patients may in fact be English-proficient, producing nondifferential exposure misclassification. The authors note this would bias the estimated association toward the null, so the true effect of genuine limited English proficiency may be larger than the reported odds ratios rather than smaller.