H Pylori QI Project
2026-5-19
Table 1
| Characteristic | HP+ on initial pathology N = 501 |
HP+ on immunohistochemistry addendum N = 501 |
p-value2 |
|---|---|---|---|
| age | 56 (47, 63) | 56 (38, 68) | 0.9 |
| sex | >0.9 | ||
| Female | 30 (60%) | 29 (58%) | |
| Male | 20 (40%) | 21 (42%) | |
| race | 0.5 | ||
| Asian | 1 (2.0%) | 0 (0%) | |
| Black | 10 (20%) | 5 (10%) | |
| Declined | 5 (10%) | 5 (10%) | |
| Other | 25 (50%) | 28 (56%) | |
| White | 9 (18%) | 12 (24%) | |
| ethnicity | 0.4 | ||
| Non-Hispanic | 16 (32%) | 13 (27%) | |
| Hispanic | 29 (58%) | 34 (69%) | |
| Declined | 5 (10%) | 2 (4.1%) | |
| Unknown | 0 | 1 | |
| insurance_status | >0.9 | ||
| Medicaid | 20 (40%) | 21 (42%) | |
| Other | 30 (60%) | 29 (58%) | |
| site_of_evaluation | <0.001 | ||
| CUIMC | 43 (86%) | 26 (52%) | |
| Allen | 2 (4.0%) | 19 (38%) | |
| Westchester | 5 (10%) | 5 (10%) | |
| egd_setting | <0.001 | ||
| Inpatient/OR | 7 (14%) | 27 (54%) | |
| Outpatient | 43 (86%) | 23 (46%) | |
| pcp_nyp | 42 (84%) | 40 (80%) | 0.8 |
| first_hp_diagnosis | 41 (82%) | 40 (80%) | >0.9 |
| provider_type | 0.8 | ||
| Fellow | 13 (26%) | 11 (22%) | |
| Faculty | 37 (74%) | 39 (78%) | |
| followup_established | 31 (62%) | 38 (76%) | 0.2 |
| 1 Median (Q1, Q3); n (%) | |||
| 2 Wilcoxon rank sum test; Fisher’s exact test | |||
Outcomes Table
| Outcome | HP+ on initial pathology N = 501 |
HP+ on immunohistochemistry addendum N = 501 |
p-value2 |
|---|---|---|---|
| quad_therapy_30d | 32 (65%) | 17 (34%) | 0.003 |
| eradication_testing_12m | 25 (50%) | 12 (24%) | 0.012 |
| 1 n (%) | |||
| 2 Fisher’s exact test | |||
Multivariable logistic regression: Quad Therapy
| Variable | OR | 95% CI | p-value |
|---|---|---|---|
| H. pylori Detection Method (IHC vs H&E) | |||
| HP+ on initial pathology | — | — | |
| HP+ on immunohistochemistry addendum | 3.84 | 1.57, 9.93 | 0.004 |
| EGD Setting (Inpatient/OR vs Outpatient) | |||
| Outpatient | — | — | |
| Inpatient/OR | 0.95 | 0.35, 2.48 | >0.9 |
| No. Obs. | 100 | ||
| AIC | 134 | ||
| Abbreviations: CI = Confidence Interval, OR = Odds Ratio | |||
Multivariable logistic regression: Eradication Testing
| Variable | OR | 95% CI | p-value |
|---|---|---|---|
| H. pylori Detection Method (IHC vs H&E) | |||
| HP+ on initial pathology | — | — | |
| HP+ on immunohistochemistry addendum | 2.03 | 0.56, 7.96 | 0.3 |
| EGD Setting (Inpatient/OR vs Outpatient) | |||
| Outpatient | — | — | |
| Inpatient/OR | 1.93 | 0.49, 8.49 | 0.4 |
| No. Obs. | 49 | ||
| AIC | 70.1 | ||
| Abbreviations: CI = Confidence Interval, OR = Odds Ratio | |||
Histograms
timing_summary <- df_time %>%
group_by(hp_detection_group) %>%
summarise(
n = n(),
median_days = median(days_to_result, na.rm = TRUE),
q1 = quantile(days_to_result, 0.25, na.rm = TRUE),
q3 = quantile(days_to_result, 0.75, na.rm = TRUE),
mean_days = mean(days_to_result, na.rm = TRUE),
sd_days = sd(days_to_result, na.rm = TRUE),
min_days = min(days_to_result, na.rm = TRUE),
max_days = max(days_to_result, na.rm = TRUE)
)
timing_summary## # A tibble: 2 × 9
## hp_detection_group n median_days q1 q3 mean_days sd_days min_days
## <fct> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 H&E 50 1 1 3 2.04 1.50 0
## 2 IHC 50 6 4.25 8 6.72 3.88 2
## # ℹ 1 more variable: max_days <dbl>wilcox_result <- wilcox.test(
days_to_result ~ hp_detection_group,
data = df_time,
exact = FALSE
)
wilcox_result##
## Wilcoxon rank sum test with continuity correction
##
## data: days_to_result by hp_detection_group
## W = 211, p-value = 4.8e-13
## alternative hypothesis: true location shift is not equal to 0