ABSTRACT
Objective
Identify predictors of patient no‐show at an urban safety‐net otolaryngology outpatient clinic.
Methods
Retrospective cohort study including all scheduled patients and appointments in 2023. Predictor variables included sociodemographic factors, primary diagnosis, insurance, and the neighborhood deprivation index (NDI) based on census tract information. The outcome was analyzed as a binary variable using univariate and multivariate mixed‐effects logistic regression models.
Results
Among 2339 patients and 4641 scheduled appointments, 1639 patients completed all scheduled visits and 700 (29.9%) missed at least 1 visit. Among all appointments, 18.4% were missed. The prior no‐show rate was 9% (IQR 4%–18%), and days from scheduling to appointment was 42 days (IQR 19–75). Univariate analysis demonstrated significant sociodemographic factors associated with higher odds of missing an appointment, including NDI (OR 1.03, p = 0.001), male gender (OR 1.35, p = 0.004), Black/African American race (OR 1.49, p = 0.022), unemployment and disability status (OR 1.45, p = 0.007 and OR 2.12, p < 0.001 respectively), unstable/unknown housing (OR 3.66, p < 0.001), and sexual orientation as lesbian or gay (OR 1.93, p = 0.003). NDI remained a significant factor in multivariate analysis (OR 1.03, p = 0.001). Patient portal inactivation and lead time were significant intervenable factors in multivariate analysis (OR 1.23, p = 0.049 and OR 1.26, p < 0.001, respectively).
Conclusion
NDI, patient portal activation, and time from appointment scheduling to visit are significant predictors of patient no‐show. This study offers insights into potential interventions addressing specific barriers to improving patient no‐show rates for an urban, safety‐net outpatient population.
Level of Evidence
3 (retrospective cohort study).
Keywords: no show, otolaryngology, quality improvement, safety‐net, sociodemographic factors
This is a retrospective cohort study evaluating predictors of patient no‐show in a safety‐net otolaryngology clinic using a repeated measures approach. Sociodemographic factors, the neighborhood deprivation index, patient portal activation, and lead time were identified as significant factors. These findings offer insights into potential targeted interventions to improve no‐show rates and access to care.
1. Introduction
Missed appointments reduce efficiency in outpatient clinics, leading to delayed care and unnecessary costs. Specifically, missed appointments have a large impact in settings that serve marginalized populations with structural barriers to accessing health care, potentially exacerbating health disparities. Studies have shown that patients more likely to no‐show are younger, belong to racial or ethnic minorities, and are more likely to have Medicaid or no insurance, highlighting the impact on these communities [1, 2]. Although many interventions relating to no‐show rates in outpatient clinics have been studied, there are currently few published reports assessing risk factors within an outpatient otolaryngology‐head and neck surgery clinical setting [3, 4].
As a safety net hospital, Zuckerberg San Francisco General (ZSFG) Hospital serves a diverse patient population of roughly 100,000 patients per year; the patient population is noted for over 30% primarily non‐English speaking, and 80%–90% of patients are covered by Medicare and Medicaid [5]. ZSFG provides comprehensive OHNS services through its outpatient clinic. A prior short‐term quality improvement project at the ZSFG OHNS outpatient clinic in 2023 revealed that 138 patients missed their appointments over a 2‐week period, accounting for 19% of scheduled patients. Notably, 38% of Black/African American patients missed their appointments, highlighting the potential impact of social and structural determinants on healthcare access [6]. Though this short‐term study hinted at possible socioeconomic and structural factors that influence missed appointment rates, a more robust study was needed to identify individual and systemic factors contributing to barriers to subspecialty OHNS care for this and similar urban, safety‐net populations.
This study aims to add to the limited literature identifying factors contributing to missed outpatient otolaryngology visits, in this case at an urban, safety‐net hospital setting. Specifically, the study sought to assess factors that could inform targeted interventions to improve access to outpatient OHNS care at ZSFG and similar clinical settings.
2. Materials and Methods
2.1. Study Design
This retrospective cohort analysis included all scheduled visits and patients at the ZSFG otolaryngology outpatient clinic from January 2023 to December 2023. The hospital is a public, safety‐net hospital servicing the population of San Francisco. The study included all appointments marked as either completed or missed (no‐show). Canceled appointments were excluded from the study. This study was determined to be a quality improvement study and was exempted by the Institutional Review Board at the University of San Francisco, California (UCSF).
2.2. Data Collection
Sociodemographic information, including age, gender identity, sexual orientation, race, ethnicity, primary language, zip code, employment status, and marital status, was extracted from the electronic medical record through standardized reports. These variables, informed by previously published studies, were hypothesized to potentially influence patients' risk of missing appointments. In addition, the number of appointments per patient, patient portal enrollment status, insurance, and primary diagnoses were also collected. The patient portal system is associated with the electronic health record that sends automated appointment reminders. Lead time is defined as the number of days from scheduling an appointment to the appointment date. Zip code median household income was extracted from the U.S. Census Bureau 2022 American Community Survey, and NDI was calculated as described in Andrews et al. [7]. All data was stored on a secure cloud database.
2.3. Data Analyses
For the logistic regression, categorical variables were consolidated as follows: Gender identity was grouped into male, female, and other, which included individuals who chose not to disclose or identified as non‐binary/genderqueer, transgender female, transgender male, or none of the above. Racial groups were consolidated into Asian, Black or African American, White, and multiracial. Primary language was consolidated into the most common languages spoken (English, Cantonese, and Spanish) and all other languages. Sexual orientation was grouped into bisexual, lesbian or gay, straight, and other, which included the options chosen not to disclose, don't know, and something else. Insurance categories included commercial insurance, Medi‐Cal (California's Medicaid program for low‐income individuals which includes managed care plans), Healthy San Francisco (a health access program for uninsured, low‐income residents of San Francisco), San Francisco Health Plan—Healthy Worker (insurance for employees in the In‐Home Supportive Services program in San Francisco County), San Francisco Health Plan (a local Medi‐Cal managed care plan), Medicare, and “other,” which includes unknown, unlisted, county indigent, jail, and none of the above. Primary diagnoses were grouped into benign and malignant cancers, ear and hearing disorders, rhinology and diseases of the upper airway, facial trauma, and other. Housing status was consolidated into stable, temporary, and unstable/unknown.
Descriptive statistics were performed. The outcome was analyzed as a binary variable using univariate and multivariate mixed‐effects logistic regression models clustered by patient.
Continuous variables including age and lead time were standardized to the median. All significant and clinically valuable variables were included in the multivariate model, which was optimized using the Akaike information criterion.
The geographic distribution of no‐show rates was visually analyzed through a map of San Francisco to account for both the proportion of patients who showed and the number of patients from each zip code. No‐show rates were separated into quartiles and differentiated by color. Zip codes outside of San Francisco were excluded from the map.
3. Results
3.1. Patient Demographics and Visit Descriptions
A total of 4641 visits and 2339 patients were analyzed from January to December of 2023. 1639 patients completed all scheduled visits and 700 patients missed at least one scheduled appointment (Table 1). Overall, 18.4% of appointments were missed by 29.9% of patients. The median age was 58 years (interquartile range [IQR] 43–67 years), with a prior missed appointment rate of 9% (IQR 4%–18%). The median number of days from scheduling to appointment was 45 days (IQR 19–75).
TABLE 1.
Demographics of patients scheduled at an outpatient otolaryngology clinic at an urban safety‐net clinic.
| Completed all visits | No showed ≥ 1 | Total | |
|---|---|---|---|
| (n = 1639) | (n = 700) | (n = 2339) | |
| Median age (IQR) | 60 (45 to 68) | 53 (38 to 64) | 58 (43 to 67) |
| Median prior no show rate | 7 (3 to 13)% | 18 (10 to 29)% | 9 (4 to 18)% |
| Median lead days | 45 (18 to 81) | 46 (21 to 70) | 45 (19 to 75) |
| Neighborhood deprivation index | −0.119 (−1.749 to 2.110) | 0.393 (−1.597 to 3.463) | −0.015 (−1.715 to 2.349) |
| Gender identity | |||
| Female | 816 (49.8%) | 292 (41.7%) | 1108 (47.4%) |
| Male | 778 (47.5%) | 383 (54.7%) | 1161 (49.6%) |
| Other | 45 (2.8%) | 25 (3.6%) | 70 (3.0%) |
| Sexual orientation | |||
| Straight | 1335 (81.5%) | 531 (75.9%) | 1866 (79.8%) |
| Bisexual | 37 (2.3%) | 21 (3.0%) | 58 (2.5%) |
| Lesbian or gay | 66 (4.0%) | 47 (6.7%) | 113 (4.8%) |
| Other | 201 (12.3%) | 101 (14.4%) | 302 (12.9%) |
| Race | |||
| White | 338 (20.6%) | 176 (25.1%) | 514 (22.0%) |
| Asian | 496 (30.3%) | 122 (17.4%) | 618 (26.4%) |
| Black or African American | 147 (9.0%) | 107 (15.3%) | 254 (10.9%) |
| Multiracial | 54 (3.3%) | 37 (5.3%) | 91 (3.9%) |
| Other | 604 (36.9%) | 258 (36.9%) | 862 (36.9%) |
| Ethnicity | |||
| Decline to answer | 14 (0.9%) | 2 (0.3%) | 16 (0.7%) |
| Not Hispanic, Latino/a, or Spanish origin | 1029 (62.8%) | 446 (63.7%) | 1475 (63.1%) |
| Hispanic, Latino/a, or Spanish origin | 596 (36.4%) | 252 (36.0%) | 848 (36.3%) |
| Primary language | |||
| English | 720 (43.9%) | 428 (61.1%) | 1148 (49.1%) |
| Spanish | 465 (28.4%) | 182 (26.0%) | 647 (27.7%) |
| Cantonese | 234 (14.3%) | 35 (5.0%) | 269 (11.5%) |
| Other | 220 (13.4%) | 55 (7.9%) | 275 (11.8%) |
| Employment status | |||
| Employed | 487 (29.7%) | 185 (26.4%) | 672 (28.7%) |
| Unemployed | 422 (25.8%) | 223 (31.9%) | 645 (27.6%) |
| Retired | 378 (23.1%) | 106 (15.1%) | 484 (20.7%) |
| Disabled | 131 (8.0%) | 89 (12.7%) | 220 (9.4%) |
| Student | 44 (2.7%) | 19 (2.7%) | 63 (2.7%) |
| Other | 177 (10.8%) | 78 (11.1%) | 255 (10.9%) |
| Insurance | |||
| Medi‐Cal | 776 (47.4%) | 407 (58.1%) | 1183 (50.6%) |
| Medicare | 470 (28.7%) | 166 (23.7%) | 636 (27.2%) |
| Healthy San Francisco | 131 (8.0%) | 58 (8.3%) | 189 (8.1%) |
| Healthy workers | 246 (15.0%) | 56 (8.0%) | 302 (12.9%) |
| HMO/PPO | 16 (1.0%) | 13 (1.9%) | 29 (1.2%) |
| Housing status | |||
| Stable | 1574 (96.0%) | 643 (91.9%) | 2217 (94.8%) |
| Temporary | 46 (2.8%) | 35 (5.0%) | 81 (3.5%) |
| Unstable/Unknown | 19 (1.2%) | 22 (3.1%) | 41 (1.8%) |
| Median household income in SF based on zip code | |||
| $47,386–$104,247 (1st quartile) | 500 (30.5%) | 236 (33.7%) | 736 (31.5%) |
| $104,537–$145,931 (2nd quartile) | 604 (36.9%) | 236 (33.7%) | 840 (35.9%) |
| $149,927–$184,671 (3rd quartile) | 383 (23.4%) | 162 (23.1%) | 545 (23.3%) |
| $190,444–$250,000+ (4th quartile) | 85 (5.2%) | 31 (4.4%) | 116 (5.0%) |
| Marital status | |||
| Single | 758 (46.3%) | 430 (61.4%) | 1188 (50.8%) |
| Married or significant other | 589 (35.9%) | 155 (22.1%) | 744 (31.8%) |
| Divorced | 172 (10.5%) | 76 (10.9%) | 248 (10.6%) |
| Widowed | 100 (6.1%) | 36 (5.1%) | 136 (5.8%) |
| Other | 20 (1.2%) | 3 (0.4%) | 23 (1.0%) |
| Primary diagnosis | |||
| Ear and hearing disorders | 737 (45.0%) | 238 (34.0%) | 975 (41.7%) |
| Rhinology and diseases of the upper airway | 147 (9.0%) | 87 (12.4%) | 234 (10.0%) |
| Benign and malignant cancers | 125 (7.6%) | 56 (8.0%) | 181 (7.7%) |
| Facial trauma | 26 (1.6%) | 17 (2.4%) | 43 (1.8%) |
| Other | 604 (36.9%) | 302 (43.1%) | 906 (38.7%) |
| Electronic medical record patient portal registration | |||
| Activated | 978 (59.7%) | 388 (55.4%) | 1366 (58.4%) |
| Inactivated | 661 (40.3%) | 312 (44.6%) | 973 (41.6%) |
3.2. Socioeconomic Risk Factors
Using a univariate mixed‐effects logistic regression analysis clustered by patient, several characteristics emerged as significant predictors of no‐show including a higher prior no‐show rate (OR 3.16, p < 0.001), male gender (OR 1.35, p = 0.004), Black/African American race (OR 1.49, p = 0.022), and sexual orientation as lesbian or gay (OR 1.93, p = 0.003) (Table 2).
TABLE 2.
Univariate analyses assessing predictors with risk of missing outpatient otolaryngology appointments using mixed‐effects model clustered by patient.
| Odds ratio | 95% Lower CI | 95% Upper CI | p | |
|---|---|---|---|---|
| Age | 0.69 | 0.61 | 0.79 | < 0.001* |
| Prior missed appointment rate | 3.16 | 2.76 | 3.62 | < 0.001* |
| Lead days | 1.16 | 1.04 | 1.30 | < 0.001* |
| Neighborhood deprivation index | 1.03 | 1.01 | 1.04 | 0.001* |
| Gender identity | ||||
| Female | Ref | Ref | Ref | Ref |
| Male | 1.35 | 1.10 | 1.67 | 0.004* |
| Other | 1.56 | 0.87 | 2.79 | 0.138 |
| Race | ||||
| White | Ref | Ref | Ref | Ref |
| Asian | 0.39 | 0.29 | 0.53 | < 0.001* |
| Black or African American | 1.49 | 1.06 | 2.09 | 0.022* |
| Multiracial | 1.13 | 0.69 | 1.84 | 0.638 |
| Other | 0.74 | 0.57 | 0.97 | 0.026* |
| Primary language | ||||
| English | Ref | Ref | Ref | Ref |
| Spanish | 0.59 | 0.47 | 0.74 | < 0.001* |
| Cantonese | 0.20 | 0.13 | 0.31 | < 0.001* |
| Other | 0.40 | 0.28 | 0.57 | < 0.001* |
| Employment status | ||||
| Employed | Ref | Ref | Ref | Ref |
| Unemployed | 1.45 | 1.11 | 1.90 | 0.007* |
| Retired | 0.85 | 0.62 | 1.16 | 0.302 |
| Disabled | 2.12 | 1.47 | 3.07 | < 0.001* |
| Student | 1.45 | 0.76 | 2.75 | 0.259 |
| Other | 1.17 | 0.81 | 1.68 | 0.404 |
| Insurance | ||||
| Medi‐Cal | Ref | Ref | Ref | Ref |
| Medicare | 0.77 | 0.60 | 0.98 | 0.033* |
| Healthy San Francisco | 0.77 | 0.53 | 1.11 | 0.159 |
| Healthy Workers | 0.40 | 0.27 | 0.58 | < 0.001* |
| HMO/PPO | 1.58 | 0.63 | 4.00 | 0.330 |
| Housing status | ||||
| Stable | Ref | Ref | Ref | Ref |
| Temporary | 3.10 | 1.87 | 5.13 | < 0.001* |
| Unstable/Unknown | 3.66 | 1.77 | 7.54 | < 0.001* |
| Median household income based on zip code | ||||
| $47,386–$104,247 (1st quartile) | Ref | Ref | Ref | Ref |
| $104,537–$145,931 (2nd quartile) | 0.84 | 0.66 | 1.07 | 0.168 |
| $149,927–$184,671 (3rd quartile) | 0.89 | 0.68 | 1.16 | 0.375 |
| $190,444–$250,000+ (4th quartile) | 0.82 | 0.49 | 1.37 | 0.450 |
| Marital status | ||||
| Single | Ref | Ref | Ref | Ref |
| Divorced | 0.74 | 0.53 | 1.03 | 0.076 |
| Married or significant other | 0.42 | 0.33 | 0.54 | < 0.001* |
| Widowed | 0.55 | 0.35 | 0.86 | 0.009* |
| Other | 0.24 | 0.06 | 0.92 | 0.037* |
| Electronic medical record patient portal registration | ||||
| Activated | Ref | Ref | Ref | Ref |
| Inactivated | 1.26 | 1.03 | 1.55 | 0.028* |
| Ethnicity | ||||
| Decline to answer | 0.38 | 0.07 | 2.00 | 0.254 |
| Not Hispanic, Latino/a, or Spanish origin | Ref | Ref | Ref | Ref |
| Hispanic, Latino/a, or Spanish origin | 0.91 | 0.73 | 1.12 | 0.368 |
| Primary diagnosis | ||||
| Ear and hearing disorders | 0.96 | 0.65 | 1.41 | 0.841 |
| Rhinology and diseases of the upper airway | 1.36 | 0.88 | 2.10 | 0.166 |
| Benign and malignant cancers | Ref | Ref | Ref | Ref |
| Facial trauma | 2.06 | 0.94 | 4.51 | 0.071 |
| Other | 1.31 | 0.90 | 1.90 | 0.160 |
| Sexual orientation | ||||
| Bisexual | 1.28 | 0.69 | 2.39 | 0.437 |
| Lesbian or gay | 1.93 | 1.24 | 2.98 | 0.003* |
| Straight | Ref | Ref | Ref | Ref |
| Other | 1.22 | 0.89 | 1.66 | 0.217 |
p < 0.05.
NDI was also significantly associated with higher odds of no‐show (OR 1.03, p = 0.001) on univariate analysis. This correlated with other sociodemographic factors including employment status as disabled (OR 2.12, p < 0.001) or unemployed (OR 1.45, p = 0.007) and housing status as temporary (OR 3.10, p < 0.001) or unstable (OR 3.66, p < 0.001).
Interestingly, there were a few factors that were significantly associated with lower odds of no‐show on univariate analysis, including older age (OR 0.69, p < 0.001), Asian race (OR 0.39, p < 0.001), primary languages other than English, and marital status as married (OR 0.42, p < 0.001) and widowed (OR 0.55, p = 0.009). Patients who were insured through Medicare (OR 0.40, p < 0.001) or Healthy Workers (OR 0.77, p 0.033) were also associated with lower odds of no‐show.
At the neighborhood level, a higher NDI demonstrated a positive correlation with no show (Figure 1), although this was not significant (R 2 < 0.01). A mixed‐effects logistic regression multivariate analysis (Table 3) revealed that NDI remained a significant predictor (OR = 1.03, p = 0.001) when accounting for age, gender, and race at an individual level.
FIGURE 1.
The relationship between neighborhood deprivation index and proportion of patients who missed an outpatient otolaryngology visit. Each circle represents a census tract, which is the unit of measure for the neighborhood deprivation index, an aggregate measure of socioeconomic and structural barriers. The size and color of the circle represent the number of patients from that census tract. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
TABLE 3.
Multivariate mixed‐effects model assessing the association between neighborhood deprivation index and risk of missed outpatient otolaryngology appointments.
| Univariate analysis | Multivariate analysis | |||||||
|---|---|---|---|---|---|---|---|---|
| Odds ratio | 95% Lower CI | 95% Upper CI | p | Odds ratio | 95% Lower CI | 95% Upper CI | p | |
| NDI | 1.03 | 1.01 | 1.04 | 0.001* | 1.03 | 1.01 | 1.04 | 0.001* |
| Age | 0.69 | 0.61 | 0.79 | < 0.001* | 0.72 | 0.62 | 0.83 | < 0.001* |
| Gender identity | ||||||||
| Female | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Male | 1.35 | 1.10 | 1.67 | 0.004* | 1.21 | 0.99 | 1.49 | 0.068 |
| Other | 1.56 | 0.87 | 2.79 | 0.138 | 0.93 | 0.52 | 1.67 | 0.817 |
| Race | ||||||||
| White | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Asian | 0.39 | 0.29 | 0.53 | < 0.001* | 0.42 | 0.31 | 0.58 | < 0.001 |
| Black or African American | 1.49 | 1.06 | 2.09 | 0.022* | 1.40 | 0.99 | 1.96 | 0.055 |
| Multiracial | 1.13 | 0.69 | 1.84 | 0.638 | 0.94 | 0.57 | 1.53 | 0.792 |
| Other | 0.74 | 0.57 | 0.97 | 0.026* | 0.65 | 0.50 | 0.85 | 0.002 |
p < 0.05.
3.3. Geographic Distribution of Patients Who Missed Appointments
The median no‐show rate across all zip codes was 30.0%. The geographic distribution of patients across San Francisco zip codes revealed that the highest proportion of patients resided in zip codes 94110 (13.3%) and 94112 (11.8%), which corresponds to the Mission District and Excelsior neighborhoods, respectively (Figure 2); these neighborhoods are known for diverse working‐class populations and immigrant residents. The no‐show rates for these areas were 28.1% for 94110 and 26.9% for 94112, both corresponding to the second quartile. Excluding those with < 100 patients, the neighborhoods with the highest proportion of missed appointments were the Tenderloin (38.4%) and South of Market (34.7%) both characterized by socioeconomic disparities and a high proportion of individuals experiencing housing insecurity.
FIGURE 2.
Proportion and volume of patients who no show from neighborhoods in San Francisco. Each zip code is represented by a circle with size corresponding to the number of patients who were scheduled for an outpatient appointment and color corresponding to the no‐show rate quartile. [Color figure can be viewed in the online issue, which is available at www.laryngoscope.com.]
3.4. Clinic‐Level Predictors
In the context of clinic intervenable factors, an inactivated patient portal (OR 1.26, p = 0.028) and longer lead time (OR = 1.16, p < 0.001) were both associated with a higher likelihood of no‐show on univariate analyses.
A multivariate model accounting for age, NDI, gender, race, insurance, housing status, and marital status revealed patient portal inactivation (OR 1.23, p = 0.049) as a significant predictor of patient no‐show (Table 4). A similar multivariate model that also accounted for primary diagnosis demonstrated lead time (OR 1.16, p < 0.001) remained a significant predictor of patient no‐show (Table 5).
TABLE 4.
Multivariate mixed‐effects model assessing the association between mychart status and risk of missed outpatient otolaryngology appointments.
| Univariate analyses | Multivariate analyses | |||||||
|---|---|---|---|---|---|---|---|---|
| Odds ratio | 95% Lower CI | 95% Upper CI | p | Odds ratio | 95% Lower CI | 95% Upper CI | p | |
| Patient portal registration | ||||||||
| Activate | Ref | Ref | Ref | Ref | ||||
| Inactivated | 1.26 | 1.03 | 1.55 | 0.028* | 1.23 | 1.00 | 1.52 | 0.049* |
| Age | 0.69 | 0.61 | 0.79 | < 0.001* | 0.71 | 0.59 | 0.84 | < 0.001* |
| Neighborhood deprivation index | 1.03 | 1.01 | 1.04 | 0.001* | 1.02 | 1.00 | 1.04 | 0.012* |
| Gender identity | ||||||||
| Female | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Male | 1.35 | 1.10 | 1.67 | 0.004* | 1.10 | 0.90 | 1.36 | 0.356 |
| Other | 1.56 | 0.87 | 2.79 | 0.138 | 0.83 | 0.46 | 1.48 | 0.524 |
| Race | ||||||||
| White | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Asian | 0.39 | 0.29 | 0.53 | < 0.001* | 0.50 | 0.36 | 0.69 | < 0.001* |
| Black or African American | 1.49 | 1.06 | 2.09 | 0.022* | 1.22 | 0.86 | 1.71 | 0.263 |
| Multiracial | 1.13 | 0.69 | 1.84 | 0.638 | 0.86 | 0.53 | 1.41 | 0.556 |
| Other | 0.74 | 0.57 | 0.97 | 0.026* | 0.66 | 0.50 | 0.87 | 0.003* |
| Insurance | ||||||||
| Medi‐Cal | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Medicare | 0.77 | 0.60 | 0.98 | 0.033* | 1.14 | 0.86 | 1.53 | 0.358 |
| Healthy San Francisco | 0.77 | 0.53 | 1.11 | 0.159 | 0.96 | 0.67 | 1.38 | 0.828 |
| Healthy workers | 0.40 | 0.27 | 0.58 | < 0.001* | 0.71 | 0.48 | 1.06 | 0.091 |
| HMO/PPO | 1.58 | 0.63 | 4.00 | 0.330 | 1.47 | 0.60 | 3.63 | 0.400 |
| Housing status | ||||||||
| Stable | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Temporary | 3.10 | 1.87 | 5.13 | < 0.001* | 2.23 | 1.36 | 3.67 | 0.002* |
| Unstable/Unknown | 3.66 | 1.77 | 7.54 | < 0.001* | 2.29 | 1.12 | 4.68 | 0.023* |
| Marital status | ||||||||
| Single | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Divorced | 0.74 | 0.53 | 1.03 | 0.076 | 1.05 | 0.75 | 1.48 | 0.777 |
| Married or significant other | 0.42 | 0.33 | 0.54 | < 0.001* | 0.68 | 0.52 | 0.89 | 0.004* |
| Widowed | 0.55 | 0.35 | 0.86 | 0.009* | 0.91 | 0.57 | 1.46 | 0.692 |
| Other | 0.24 | 0.06 | 0.92 | 0.037* | 0.30 | 0.08 | 1.15 | 0.079 |
p < 0.05.
TABLE 5.
Multivariate mixed‐effects model assessing the association between lead days and the risk of missed outpatient otolaryngology appointments.
| Univariate analyses | Multivariate analyses | |||||||
|---|---|---|---|---|---|---|---|---|
| Odds ratio | 95% Lower CI | 95% Upper CI | p | Odds ratio | 95% Lower CI | 95% Upper CI | p | |
| Lead days | 1.16 | 1.04 | 1.30 | < 0.001* | 1.26 | 1.13 | 1.41 | < 0.001* |
| Age | 0.69 | 0.61 | 0.79 | < 0.001* | 0.71 | 0.60 | 0.85 | < 0.001* |
| Neighborhood deprivation index | 1.03 | 1.01 | 1.04 | 0.001* | 1.02 | 1.01 | 1.04 | 0.005* |
| Gender identity | ||||||||
| Female | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Male | 1.35 | 1.10 | 1.67 | 0.004* | 1.14 | 0.92 | 1.41 | 0.228 |
| Other | 1.56 | 0.87 | 2.79 | 0.138 | 0.82 | 0.45 | 1.48 | 0.509 |
| Race | ||||||||
| White | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Asian | 0.39 | 0.29 | 0.53 | < 0.001* | 0.51 | 0.37 | 0.70 | < 0.001* |
| Black or African American | 1.49 | 1.06 | 2.09 | 0.022* | 1.23 | 0.87 | 1.74 | 0.239 |
| Multiracial | 1.13 | 0.69 | 1.84 | 0.638 | 0.88 | 0.53 | 1.44 | 0.610 |
| Other | 0.74 | 0.57 | 0.97 | 0.026* | 0.70 | 0.53 | 0.92 | 0.011* |
| Insurance | ||||||||
| Medi‐Cal | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Medicare | 0.77 | 0.60 | 0.98 | 0.033* | 1.15 | 0.86 | 1.54 | 0.348 |
| Healthy San Francisco | 0.77 | 0.53 | 1.11 | 0.159 | 0.93 | 0.64 | 1.34 | 0.686 |
| Healthy workers | 0.40 | 0.27 | 0.58 | < 0.001* | 0.70 | 0.47 | 1.04 | 0.075 |
| HMO/PPO | 1.58 | 0.63 | 4.00 | 0.330 | 1.50 | 0.60 | 3.74 | 0.386 |
| Housing status | ||||||||
| Stable | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Temporary | 3.10 | 1.87 | 5.13 | < 0.001* | 2.35 | 1.42 | 3.89 | 0.001* |
| Unstable/Unknown | 3.66 | 1.77 | 7.54 | < 0.001* | 2.68 | 1.30 | 5.52 | 0.008* |
| Marital status | ||||||||
| Single | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Divorced | 0.74 | 0.53 | 1.03 | 0.076 | 1.04 | 0.74 | 1.46 | 0.836 |
| Married or significant other | 0.42 | 0.33 | 0.54 | < 0.001* | 0.66 | 0.51 | 0.86 | 0.002* |
| Widowed | 0.55 | 0.35 | 0.86 | 0.009* | 0.92 | 0.57 | 1.48 | 0.724 |
| Other | 0.24 | 0.06 | 0.92 | 0.037* | 0.29 | 0.08 | 1.12 | 0.072 |
| Primary diagnosis | ||||||||
| Ear and hearing disorders | 0.96 | 0.65 | 1.41 | 0.841 | 1.05 | 0.72 | 1.55 | 0.788 |
| Rhinology/upper airway diseases | 1.36 | 0.88 | 2.10 | 0.166 | 1.31 | 0.85 | 2.01 | 0.222 |
| Benign and malignant cancers | Ref | Ref | Ref | Ref | Ref | Ref | Ref | Ref |
| Facial trauma | 2.06 | 0.94 | 4.51 | 0.071 | 1.49 | 0.69 | 3.22 | 0.313 |
| Other | 1.31 | 0.90 | 1.90 | 0.160 | 1.31 | 0.90 | 1.91 | 0.153 |
p < 0.05.
4. Discussion
This is among the first studies assessing predictors of patient no‐shows in an outpatient otolaryngology clinic serving a safety‐net population. This study highlights both clinic‐level and systemic sociodemographic factors that contribute to missed appointments, providing a multifaceted, data‐driven approach to future interventions. Significant modifiable factors found included an inactive patient portal and increased lead time to the scheduled appointment. Furthermore, sociodemographic factors and structural barriers known to persistently affect access to health care, such as housing and represented by NDI, are important to address when designing strategies to reduce no‐show rates in collaboration with policymakers and public health initiatives.
Among the few studies exploring missed appointments within otolaryngology, our univariate analysis supports findings from a similar study on initial visits, indicating that disability status and Black/African American race are associated with higher odds of missed appointments [8]. In contrast, our study demonstrates several other significant sociodemographic predictors of no‐show on univariate analysis, including male gender, unemployment status, sexual orientation as lesbian or gay, temporary or unstable housing, and a higher NDI. Similarly, another study investigating the impact of COVID‐19 on missed otolaryngology appointments identified male gender, Black/African American race, and Hispanic/Latino ethnicity as significant predictors using multivariate analysis [9]. While these studies accounted for some social determinants of health, our study utilized the NDI—a robust measure of socioeconomic disadvantage integrating several measures of education, housing, employment, wealth, and income. Using metrics such as NDI to capture the multiple facets of barriers to accessing health care may provide a more comprehensive perspective when integrating social determinants of health into future studies.
In addition to sociodemographic factors, our study also investigated modifiable clinic factors such as patient portal activation and lead time to inform targeted interventions. Prior literature outside of otolaryngology has also noted that increased time from when the appointment is scheduled to the appointment date is associated with an increased risk of missed appointments [10, 11]. One survey study in the United Kingdom found that patients missed their outpatient otolaryngology appointments due to transportation challenges, professional engagement, family issues, and financial constraints [12]. While these barriers are influenced by social determinants of health, they can also be addressed by the health system through telehealth options, appointments during non‐business hours, and flexible appointment rescheduling. Overall, these findings highlight the importance of addressing both individual and clinic‐level barriers to improve appointment attendance.
Findings from this study have implications for interventions and broader approaches to addressing access to outpatient subspecialty care. Based on these findings, interventions to address access to outpatient specialty care in this urban safety‐net population include improving patient communication such as through electronic patient portals and decreasing lead time to appointments [3, 13, 14]. Given the link between neighborhood deprivation and transportation disadvantage, targeted transportation support could be another strategy to improve access to specialty care [15, 16]. The majority of patients in this study live within 15 miles of the hospital; transportation access rather than overall distance may be a more significant contributor to missed appointments in this population. Additional studied interventions include standardized 24‐ and 48‐h pre‐visit reminder calls, which demonstrate a 12.2% reduction in no‐show rate [17]. Another multi‐practice quality improvement study investigating strategies to reduce missed appointment rates, including behavioral and structural interventions, identified reducing forward booking time as the most effective approach [18]. Taken together, interventions both addressing barriers to accessing care and clinic structural interventions successfully decrease missed appointments. Moving forward, mixed‐methods studies, which include interviews or focus groups, would help identify additional barriers from the patient perspective and could be used to co‐develop and adapt interventions to diverse health settings.
In addition to informing targeted interventions to improve access to care, addressing missed appointments has broader implications for ensuring the quality of care provided at the health systems level. Missed appointments are significantly associated with increased all‐cause mortality and poorer health outcomes such as poorly controlled hypertension and diabetes [19, 20, 21]. Though not directly studied in otolaryngology care, missed appointments conceivably could contribute to delayed diagnoses and management of diseases such as head and neck cancer and ear infections. In addition to the impact on the quality of care, there are significant financial implications for missed appointments. An estimated 3%–14% of total clinic income and $191 K–$384 K in revenue per year are estimated to be lost due to missed appointments [22, 23]. Therefore, maximizing appointment attendance helps to optimize the health system overall.
This study had several limitations. While this study represents a diverse, urban, safety‐net otolaryngology practice, this single center may not be fully representative of other populations, given differences in patient demographics and health infrastructure. California, for example, has a lower proportion of uninsured patients at 6.4% compared to the United States at 7.9%. Furthermore, this retrospective study is limited by medical record documentation. For example, while some social determinants were captured, not all causes of missed appointments nor the causal relationships of variables to missed appointments were captured. Direct patient interviews and surveys are needed to identify the exact reasons for patients missing their appointments, information valuable to developing future interventions.
This study highlights how both the health systems and patient‐level factors influence the accessibility of outpatient subspecialty care at an urban safety‐net hospital setting. In addition to clinic‐level modifiable factors, including automating appointment reminders and minimizing appointment lead time, addressing structural barriers to care, such as assistance with transportation, is also critical.
5. Conclusion
Clinic intervenable factors associated with the increased likelihood of no‐show include patient portal inactivation and increased lead time. Several sociodemographic factors are also predictors of patient no‐shows, with NDI remaining significant when accounting for age, gender, and race. This study informs future interventions in addressing specific barriers to improve patient no‐show rates in safety‐net and otolaryngology outpatient clinics.
Conflicts of Interest
J.C. is a consultant for Inspire Medical Systems and Nyxoah. J.C. is supported by the Veterans Affairs Medical Center, San Francisco, CA. The contents do not represent the views of the U.S. Department of Veterans Affairs or those of the U.S. government.
Acknowledgments
The authors would like to thank Dr. Ann A. Lazar and Dr. Yea‐Hung Chen for their guidance with statistical analysis. The authors would also like to thank Angela Casarez for her valuable feedback during the study design phase.
Husman T., Miraftab‐Salo O., Ilan A., et al., “Predictors of No‐Show in a Safety‐Net Otolaryngology Clinic: A Repeated Measures Approach,” The Laryngoscope 135, no. 9 (2025): 3123–3133, 10.1002/lary.32150.
Funding: The authors received no specific funding for this work.
This article was presented at the Triological Society 2025 Combined Sections Meeting, Orlando, Florida, USA January 23–25, 2025.
References
- 1. Kaplan‐Lewis E. and Percac‐Lima S., “No‐Show to Primary Care Appointments: Why Patients Do Not Come,” Journal of Primary Care and Community Health 4, no. 4 (2013): 251–255, 10.1177/2150131913498513. [DOI] [PubMed] [Google Scholar]
- 2. Greig E. C., Gonzalez‐Colaso R., and Nwanyanwu K., “Racial, Ethnic, and Socioeconomic Disparities Drive Appointment No‐Show in Patients With Chronic Eye Disease,” Journal of Racial and Ethnic Health Disparities 10, no. 4 (2023): 1790–1797, 10.1007/s40615-022-01363-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Ansell D., Crispo J. A. G., Simard B., and Bjerre L. M., “Interventions to Reduce Wait Times for Primary Care Appointments: A Systematic Review,” BMC Health Services Research 17, no. 1 (2017): 295, 10.1186/s12913-017-2219-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Parikh A., Gupta K., Wilson A. C., Fields K., Cosgrove N. M., and Kostis J. B., “The Effectiveness of Outpatient Appointment Reminder Systems in Reducing No‐Show Rates,” American Journal of Medicine 123, no. 6 (2010): 542–548, 10.1016/j.amjmed.2009.11.022. [DOI] [PubMed] [Google Scholar]
- 5. Zuckerberg San Francisco General Hospital and Trauma Center , “Annual Reports,” accessed September 11, 2024, https://www.zuckerbergsanfranciscogeneral.org/service‐area/annual‐reports/.
- 6. Barazza E. L., “Decreasing No‐Show Rates in an Outpatient Specialty Clinic [Master's Project],” accessed September 11, 2024, https://repository.usfca.edu/capstone/1585.
- 7. Andrews M. R., Tamura K., Claudel S. E., et al., “Geospatial Analysis of Neighborhood Deprivation Index (NDI) for the United States by County,” Journal of Maps 16, no. 1 (2020): 101–112, 10.1080/17445647.2020.1750066. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Wilson C. A., Jamil T. L., Velu P. S., and Levi J. R., “Patient Factors Associated With Missed Otolaryngology Appointments at an Urban Safety‐Net Hospital,” Laryngoscope 134, no. 9 (2024): 4003–4010, 10.1002/lary.31401. [DOI] [PubMed] [Google Scholar]
- 9. P E., S L., G N., et al., “The Impact of COVID‐19 and Socioeconomic Determinants on Appointment Non‐Attendance in an Urban Otolaryngology Clinic: A Retrospective Analysis From a Safety Net Hospital,” Annals of Otology, Rhinology and Laryngology 134, no. 2 (2025): 117–124, 10.1177/00034894241295475. [DOI] [PubMed] [Google Scholar]
- 10. Tsai W. C., Lee W. C., Chiang S. C., Chen Y. C., and Chen T. J., “Factors of Missed Appointments at an Academic Medical Center in Taiwan,” Journal of the Chinese Medical Association 82, no. 5 (2019): 436–442, 10.1097/JCMA.0000000000000068. [DOI] [PubMed] [Google Scholar]
- 11. Rea C. J., Toomey S. L., Hauptman M., et al., “Predictors of Subspecialty Appointment Scheduling and Completion for Patients Referred From a Pediatric Primary Care Clinic,” Clinical Pediatrics (Phila) 63, no. 4 (2024): 512–521, 10.1177/00099228231179673. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Bahammam F. A., Tolley N., Kuchai R., Benjamin E., and Rollin M., “Analyzing Factors and Reasons Behind NonAttendance of ENT Patients: Clinical Audit,” Ear, Nose, & Throat Journal (2024): 01455613241283798, 10.1177/01455613241283798. [DOI] [PubMed] [Google Scholar]
- 13. Zhong X., Liang M., Sanchez R., et al., “On the Effect of Electronic Patient Portal on Primary Care Utilization and Appointment Adherence,” BMC Medical Informatics and Decision Making 18, no. 1 (2018): 84, 10.1186/s12911-018-0669-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Shekelle P. G., Begashaw M. M., Miake‐Lye I. M., Booth M., Myers B., and Renda A., “Effect of Interventions for Non‐Emergent Medical Transportation: A Systematic Review and Meta‐Analysis,” BMC Public Health 22, no. 1 (2022): 799, 10.1186/s12889-022-13149-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Licaj I., Haddak M., Pochet P., and Chiron M., “Individual and Contextual Socioeconomic Disadvantages and Car Driving Between 16 and 24 Years of Age: A Multilevel Study in the Rhône Département (France),” Journal of Transport Geography 22 (2012): 19–27, 10.1016/j.jtrangeo.2011.11.018. [DOI] [Google Scholar]
- 16. Engels B. and Liu G. J., “Social Exclusion, Location and Transport Disadvantage Amongst Non‐Driving Seniors in a Melbourne Municipality, Australia,” Journal of Transport Geography 19, no. 4 (2011): 984–996, 10.1016/j.jtrangeo.2011.03.007. [DOI] [Google Scholar]
- 17. “Effect of Standardized Reminder Calls on Trauma Patient No‐Show Rate,” accessed December 16, 2024, https://pubmed.ncbi.nlm.nih.gov/37861231/. [DOI] [PubMed]
- 18. Margham T., Williams C., Steadman J., and Hull S., “Reducing Missed Appointments in General Practice: Evaluation of a Quality Improvement Programme in East London,” British Journal of General Practice 71, no. 702 (2021): e31–e38, 10.3399/bjgp20X713909. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Nguyen D. L., DeJesus R. S., and Wieland M. L., “Missed Appointments in Resident Continuity Clinic: Patient Characteristics and Health Care Outcomes,” Journal of Graduate Medical Education 3, no. 3 (2011): 350–355, 10.4300/JGME-D-10-00199.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Schectman J. M., Schorling J. B., and Voss J. D., “Appointment Adherence and Disparities in Outcomes Among Patients With Diabetes,” Journal of General Internal Medicine 23, no. 10 (2008): 1685–1687, 10.1007/s11606-008-0747-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. McQueenie R., Ellis D. A., McConnachie A., Wilson P., and Williamson A. E., “Morbidity, Mortality and Missed Appointments in Healthcare: A National Retrospective Data Linkage Study,” BMC Medicine 17, no. 1 (2019): 2, 10.1186/s12916-018-1234-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Huang Z., Ashraf M., Gordish‐Dressman H., and Mudd P., “The Financial Impact of Clinic No‐Show Rates in an Academic Pediatric Otolaryngology Practice,” American Journal of Otolaryngology 38, no. 2 (2017): 127–129, 10.1016/j.amjoto.2016.11.004. [DOI] [PubMed] [Google Scholar]
- 23. Moore C. G., Wilson‐Witherspoon P., and Probst J. C., “Time and Money: Effects of No‐Shows at a Family Practice Residency Clinic,” Family Medicine 33, no. 7 (2001): 522–527. [PubMed] [Google Scholar]