Abstract
Background
Ambulatory care sensitive conditions (ACSCs) are acute or chronic health issues that lead to potentially preventable hospitalizations when not treated in the outpatient primary care setting.
Objective
To describe national hospitalization rates due to ACSCs among adult inpatients in the US.
Design
A retrospective cross-sectional analysis of the 2018 US National Inpatient Sample (NIS) dataset from the Healthcare Cost and Utilization Project at the Agency of Healthcare Research and Quality was completed in the year 2022.
Participants
Participants were adult inpatients from community hospitals in 48 states of the US and District of Columbia.
Main Measures
ACSC admission rates were calculated using ICD-10 codes and the Purdy ACSC definition. The admission rates were weighted to the US inpatient population and stratified by age, sex, and race.
Key Results
ACSC hospitalization rates varied considerably across age and average number of hospitalizations varied across sex and race. ACSC hospitalization rates increased with age, male sex, and Native American and Black race. The most common ACSCs were pneumonia, diabetes, and congestive heart failure.
Conclusions
Previous studies have emphasized the importance of preventable hospitalizations, however, the national rates for ACSC hospitalizations across all ages in the US have not been reported. The national rates presented will facilitate comparisons to identify hospitals and health care systems with higher-than-expected rates of ACSC admissions that may suggest a need for improved primary care services.
KEY WORDS: hospital admission, inpatient, acute care, adults, ACSC
INTRODUCTION
Potentially preventable hospitalizations contribute greatly to annual healthcare expenditures in the US.1,2 A major contributor to preventable hospitalizations is a set of health problems known as ambulatory care sensitive conditions (ACSCs).3 Health conditions that are categorized as ACSCs are wide-ranging and include congestive heart failure, diabetes, hypertension, gastroenteritis, and ulcers.4 ACSCs can be acute or chronic and are all considered to be preventable conditions that can be treated effectively through quality primary care health services.5,6 Yet, when primary care is not available or inaccessible, ACSCs can lead to increased healthcare costs due to unnecessary hospitalizations.7
Existing evidence of the impact of ACSCs on the US healthcare system indicates that in 2006 there were over 4 million ACSC hospitalizations in the US resulting in 1 in 10 inpatient care dollars being spent on preventable hospitalizations in a single year.1 An examination of hospitalization data from 2005–2012 revealed that $41 billion was spent on preventable hypertension hospitalizations over an 8-year period.8 Additionally, previous studies identified that 20% of hospitalizations that occurred among a sample of older patients with diabetes were preventable and cost more than $1 billion.9 Generally, hospitalization rates for ACSCs have focused on admissions due to specific conditions and have been limited to specific regions or states of the US.10–13 Thus, a report of national hospitalization rates for a variety of ACSCs across age as well as average ACSC hospitalizations stratified by sex and race is warranted. This information can assist in determining which US hospitals have higher than expected rates of ACSC hospitalizations, which ACSCs are major contributors to hospitalization rates, and which subpopulations may require additional services to reduce ACSC hospitalizations. The present study addresses this gap by describing national rates for ACSC hospitalizations among US adults using a 2018 national inpatient sample.
METHODS
Study Population
Study subjects included inpatient admissions to hospitals that submitted data in 2018 to the Agency of Healthcare Research and Quality (AHRQ) National Inpatient Sample (NIS)—the largest all-payer hospital care database in the US. Data originated from 4,584 non-federal hospitals from across the US, yielding 7,105,226 adult inpatients in the dataset, representing a stratified sample of approximately 20% of discharges from U.S. community hospitals.14 The NIS database provides hospital admission information for patients admitted from the community, from long-term/subacute care, and from rehabilitation facilities, including patient demographics and primary and secondary diagnoses using International Classification of Diseases– 10th revision (ICD10) codes.15 The sample was weighted for extrapolation to the national US inpatient population and data analysis occurred in calendar year 2022.16 This weighting accounts for the stratified sampling design of the NIS and adjusts estimates to reflect the entire US inpatient population, rather than the 20% of inpatients that are part of the NIS. This study was exempt from requiring IRB approval.
The National Inpatient Sample includes data from inpatients aged 18–90+ y. However, the focus of this study was on the older inpatients, ≥45y, because this population is more commonly impacted by chronic disease and hospitalization.17 Thus, to examine the changes in ACSC hospitalization rates by age, the older inpatients ≥45y were stratified into smaller groups using 5-year increments (e.g. 45–49, 50–54, etc.). Patients aged 90 years and older were analyzed as a single group to maintain privacy. The younger adult inpatients aged 18-45y were included in the analysis but were not stratified into smaller age groups. Data was further stratified by sex (male and female) and race (Black, White, Hispanic, Asian or Pacific Islander, and Native American) to assess the effect of these demographics on ACSC hospitalization rates among patients in each age group.
Measures
Ambulatory Care Sensitive Conditions (ACSC)
This investigation examined ACSC hospitalization rates using the Purdy definition and the ICD-10 codes from the NIS database.15 The Purdy definition for ACSCs includes a variety of conditions such as congestive heart failure, gastroenteritis, diabetes, anemia, pelvic inflammatory disease, and ulcer.18 Hospitalizations were classified as ACSC hospitalizations if the primary discharge diagnosis was one of the specified ICD-10 codes for ACSCs according to Purdy; hospitalizations for all other diagnoses were classified as non-ACSC hospitalizations.
Comorbidities
To examine whether rates of hospitalizations for ACSCs were driven primarily by increased comorbidity, comorbidity rates among patients were quantified using the Charlson Comorbidity Index.19 Comorbidities were categorized dichotomously (present or absent) and included myocardial infarction, congestive heart failure, peripheral vascular disease, cardiovascular disease, chronic pulmonary disease, rheumatic disease, mild liver disease, renal disease, moderate liver disease, and malignancies. These comorbidities were identified using primary and secondary ICD-10 discharge diagnosis codes. Scores were calculated using the algorithms of Quan and colleagues.20
Statistical Analyses
Study population characteristics are presented as mean with 95% CI for continuous variables and frequencies (proportion) with 95% CI for categorical variables. Pearson’s chi-square tests for categorical variables were used to examine statistical differences based on age group. Weighted rates of ACSC admissions were calculated as the number of admissions with a primary discharge diagnosis for an ACSC relative to the total number of admissions. Weights were inversely proportional to the US inpatient population density, stratified by Census division, urban/rural location, teaching status, bed size, and hospital ownership to account for the NIS sampling scheme. These weighted rates extrapolate the rates in the NIS sample to the national US inpatient population, accounting for the former not being a random sample from the latter. Weighted ACSC rates were stratified by 5-year age intervals, sex, and race to allow comparison across these variables. The summary data for each age group was calculated consisting of frequency (proportion of total hospitalizations for that group) and SD. Summary data for each age group stratified by sex and race were also calculated consisting of mean and SD. Calculations were completed using the survey package in version 3.5.1 of the R programming environment (R Core Development Team, Vienna, Austria).21 To examine whether ACSC admission rates were driven by comorbidity, unweighted concordance between ACSC admission and the presence of one or more comorbidities on the CCI was calculated using Gwet’s AC1 statistic due to the high level of ties.22
RESULTS
The demographic information of the study population (N = 7,105,226) is presented in Table 1. The average age was 49.9 years (95% CI 49.9, 49.9) and 61.4% of the inpatients were aged 45 years and older. The largest age groups represented in this dataset were 60–64 (7.76%), 65–69 (8.14%), and 70–74 (8.00%). Most inpatients were White (67.0%) and female (56.1%).
Table 1.
Study Population Characteristics
| All | N | |
|---|---|---|
| N = 7,105,226 | ||
| Age (years) | 49.9 [49.9;49.9] | 7,105,226 |
| Age group: | 7,105,226 | |
| < 45 | 38.6% [38.5%;38.6%] | |
| 45–49 | 4.14% [4.12%;4.15%] | |
| 50–54 | 5.33% [5.31%;5.34%] | |
| 55–59 | 6.94% [6.92%;6.96%] | |
| 60–64 | 7.76% [7.74%;7.78%] | |
| 65–69 | 8.14% [8.12%;8.16%] | |
| 70–74 | 8.00% [7.98%;8.02%] | |
| 75–79 | 7.12% [7.10%;7.14%] | |
| 80–84 | 5.92% [5.90%;5.94%] | |
| 85–89 | 4.60% [4.58%;4.61%] | |
| 90 and above | 3.50% [3.48%;3.51%] | |
| Gender: | 7,104,516 | |
| Male | 43.9% [43.9%;43.9%] | |
| Female | 56.1% [56.1%;56.1%] | |
| Race: | 6,621,686 | |
| White | 67.0% [67.0%;67.1%] | |
| Black | 15.7% [15.7%;15.8%] | |
| Hispanic | 13.3% [13.3%;13.3%] | |
| Asian or Pacific Islander | 3.24% [3.23%;3.26%] | |
| Native American | 0.70% [0.69%;0.70%] |
Data are presented as mean and 95% confidence interval or proportion of total hospitalizations and 95% confidence interval
ACSC hospitalization rates based on age for each condition are presented in Table 2. There was a significant association between age group and ACSC hospitalization rates across nearly all conditions except nutrition and gangrene (no admissions identified for this condition in the 2018 dataset). Across all ages, the most common ACSCs were pneumonia, diabetes, and congestive heart failure. The most common ACSC among <45y age group was diabetes (type I and II diabetes are not differentiated) which increased with age until 50-54y then decreased. ACSC hospitalization rates for diabetes were highest among inpatients 45–49y and lowest among those 85y+ . Rates of congestive heart failure, pneumonia, and chronic obstructive pulmonary disease increased with age and were highest in the ≥90y age group.
Table 2.
ACSC Hospitalization Rates (%, SD) Based on Age Using the Purdy Definition
| All ages (y) | < 45 | 45–49 | 50–54 | 55–59 | 60–64 | 65–69 | 70–74 | 75–79 | 80–84 | 85–89 | ≥90 | P Value | N | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N = 7,105,226 | N = 2,741,024 | N = 293,937 | N = 378,534 | N = 492,857 | N = 551,201 | N = 578,068 | N = 568,088 | N = 505,829 | N = 420,564 | N = 326,616 | N = 248,508 | |||
| angina | 0.08 (2.91) | 0.02 (1.41) | 0.16 (4.02) | 0.17 (4.14) | 0.17 (4.13) | 0.14 (3.76) | 0.12 (3.49) | 0.12 (3.45) | 0.10 (3.22) | 0.09 (2.94) | 0.08 (2.78) | 0.08 (2.79) | 0 | 7,105,226 |
| asthma | 0.50 (7.07) | 0.83 (9.06) | 0.72 (8.48) | 0.56 (7.46) | 0.41 (6.41) | 0.30 (5.51) | 0.25 (4.97) | 0.20 (4.51) | 0.18 (4.24) | 0.18 (4.18) | 0.15 (3.90) | 0.13 (3.60) | 0 | 7,105,226 |
| cellulitis | 1.17 (10.7) | 0.81 (8.94) | 1.94 (13.8) | 1.80 (13.3) | 1.58 (12.5) | 1.47 (12.0) | 1.25 (11.1) | 1.19 (10.8) | 1.14 (10.6) | 1.20 (10.9) | 1.24 (11.0) | 1.41 (11.8) | 0 | 7,105,226 |
| chf | 1.73 (13.0) | 0.28 (5.27) | 1.60 (12.5) | 1.86 (13.5) | 2.09 (14.3) | 2.19 (14.6) | 2.21 (14.7) | 2.42 (15.4) | 2.84 (16.6) | 3.33 (17.9) | 4.16 (20.0) | 5.09 (22.0) | 0 | 7,105,226 |
| epilepsy | 0.82 (9.03) | 1.06 (10.2) | 1.09 (10.4) | 1.00 (9.97) | 0.90 (9.42) | 0.77 (8.73) | 0.65 (8.07) | 0.54 (7.35) | 0.51 (7.13) | 0.48 (6.88) | 0.45 (6.70) | 0.36 (6.02) | 0 | 7,105,226 |
| copd | 0.24 (4.90) | 0.09 (2.92) | 0.22 (4.68) | 0.23 (4.79) | 0.27 (5.21) | 0.29 (5.40) | 0.32 (5.63) | 0.33 (5.78) | 0.39 (6.20) | 0.42 (6.45) | 0.47 (6.81) | 0.52 (7.18) | 0 | 7,105,226 |
| gastroenteritis | 0.60 (7.72) | 0.44 (6.62) | 0.64 (8.00) | 0.62 (7.87) | 0.62 (7.85) | 0.59 (7.65) | 0.60 (7.71) | 0.66 (8.10) | 0.74 (8.54) | 0.82 (9.00) | 0.92 (9.54) | 1.06 (10.3) | 0 | 7,105,226 |
| dental | 0.08 (2.84) | 0.11 (3.34) | 0.15 (3.88) | 0.12 (3.39) | 0.09 (3.00) | 0.06 (2.52) | 0.06 (2.38) | 0.04 (2.00) | 0.04 (1.94) | 0.03 (1.69) | 0.02 (1.47) | 0.02 (1.51) | <0.001 | 7,105,226 |
| diabetes | 1.91 (13.7) | 1.58 (12.5) | 3.59 (18.6) | 3.39 (18.1) | 2.94 (16.9) | 2.56 (15.8) | 2.10 (14.3) | 1.77 (13.2) | 1.51 (12.2) | 1.31 (11.4) | 0.97 (9.80) | 0.68 (8.21) | 0 | 7,105,226 |
| ent | 0.11 (3.30) | 0.19 (4.32) | 0.09 (2.92) | 0.07 (2.73) | 0.06 (2.53) | 0.05 (2.30) | 0.04 (2.11) | 0.05 (2.29) | 0.05 (2.22) | 0.06 (2.43) | 0.07 (2.72) | 0.08 (2.82) | 0 | 7,105,226 |
| gangrene | 0.00 (0.00) | 0.00 (0.00) | 0.00 (0.00) | 0.00 (0.00) | 0.00 (0.00) | 0.00 (0.00) | 0.00 (0.00) | 0.00 (0.00) | 0.00 (0.00) | 0.00 (0.00) | 0.00 (0.00) | 0.00 (0.00) | n/a | 7,105,226 |
| htn | 0.03 (1.86) | 0.01 (1.16) | 0.06 (2.38) | 0.06 (2.35) | 0.05 (2.23) | 0.05 (2.13) | 0.04 (2.09) | 0.04 (2.12) | 0.05 (2.12) | 0.05 (2.31) | 0.04 (2.06) | 0.05 (2.24) | <0.001 | 7,105,226 |
| pneumonia | 2.24 (14.8) | 0.94 (9.63) | 1.74 (13.1) | 2.05 (14.2) | 2.24 (14.8) | 2.47 (15.5) | 2.57 (15.8) | 3.04 (17.2) | 3.45 (18.3) | 4.07 (19.8) | 4.72 (21.2) | 5.71 (23.2) | 0 | 7,105,226 |
| anemia | 0.12 (3.44) | 0.06 (2.40) | 0.18 (4.23) | 0.12 (3.46) | 0.11 (3.31) | 0.14 (3.70) | 0.16 (3.97) | 0.15 (3.90) | 0.18 (4.21) | 0.18 (4.27) | 0.20 (4.49) | 0.19 (4.39) | 0 | 7,105,226 |
| nutrition | 0.02 (1.58) | 0.03 (1.60) | 0.02 (1.51) | 0.02 (1.47) | 0.02 (1.54) | 0.03 (1.59) | 0.02 (1.53) | 0.02 (1.58) | 0.03 (1.72) | 0.02 (1.58) | 0.02 (1.56) | 0.02 (1.42) | 0.406 | 7,105,226 |
| vaccine | 0.01 (1.04) | 0.02 (1.29) | 0.03 (1.73) | 0.02 (1.38) | 0.01 (1.02) | 0.01 (0.86) | 0.01 (0.72) | 0.00 (0.50) | 0.00 (0.47) | 0.00 (0.31) | 0.00 (0.49) | 0.00 (0.20) | <0.001 | 7,105,226 |
| pid | 0.05 (2.20) | 0.09 (3.04) | 0.12 (3.48) | 0.06 (2.35) | 0.02 (1.37) | 0.01 (1.12) | 0.01 (1.05) | 0.01 (0.93) | 0.00 (0.63) | 0.00 (0.71) | 0.00 (0.35) | 0.00 (0.40) | 0 | 7,105,226 |
| ulcer | 0.37 (6.05) | 0.10 (3.08) | 0.43 (6.54) | 0.43 (6.56) | 0.50 (7.04) | 0.52 (7.19) | 0.55 (7.37) | 0.55 (7.42) | 0.60 (7.75) | 0.64 (7.95) | 0.60 (7.74) | 0.53 (7.24) | 0 | 7,105,226 |
| pyelonephritis | 0.37 (6.11) | 0.37 (6.09) | 0.60 (7.70) | 0.49 (6.98) | 0.39 (6.27) | 0.37 (6.03) | 0.35 (5.90) | 0.35 (5.94) | 0.34 (5.83) | 0.32 (5.69) | 0.31 (5.54) | 0.27 (5.19) | <0.001 | 7,105,226 |
| Index | 10.5 (30.6) | 7.00 (25.5) | 13.4 (34.0) | 13.1 (33.7) | 12.5 (33.0) | 12.0 (32.5) | 11.3 (31.7) | 11.5 (31.9) | 12.1 (32.7) | 13.2 (33.8) | 14.4 (35.1) | 16.2 (36.9) | 0 | 7,105,226 |
Data are presented as percent of total hospitalizations that were categorized as each individual ACSC (ambulatory care sensitive condition) and standard deviation. chf, congestive heart failure; copd, chronic obstructive pulmonary disease; ent, ear nose and throat infections; htn, hypertension; pid, pelvic inflammatory disease. Index score represents the hospitalization rate due to any ACSC for each age group
Mean ACSC hospitalizations stratified by sex for each age group are presented in Table 3. Across all ages, males had higher average percentage of ACSC hospitalizations than females. Males had an average of 11.6 ACSC hospitalizations (SD = 32.0) and females had an average of 9.6 ACSC hospitalizations (SD = 29.5). Examining specific age groups identified that male inpatients in the <45y, 45-49y, 50-54y, and 55-59y age groups had higher percentage of ACSC hospitalizations compared to females of the same age. Yet, this trend reversed at age 65 with females having higher mean ACSC hospitalizations among older age groups. For both male and female inpatients, the oldest age group (≥90y) had the highest average percentage of ACSC hospitalizations compared to all other age groups of the same sex.
Table 3.
Mean Percentage of ACSC Hospitalizations Based on Sex Using the Purdy Definition
| Sex | Statistic | All ages | <45 | 45–49 | 50–54 | 55–59 | 60–64 | 65–69 | 70–74 | 75–79 | 80–84 | 85–89 | ≥90 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Male | Mean | 11.6 | 9.6 | 13.7 | 13.2 | 12.6 | 12 | 11.2 | 11.4 | 11.9 | 13 | 14.1 | 15.8 |
| SD | 32 | 29.5 | 34.4 | 33.8 | 33.2 | 32.5 | 31.6 | 31.8 | 32.4 | 33.6 | 34.8 | 36.4 | |
| Female | Mean | 9.6 | 5.5 | 13.1 | 12.9 | 12.4 | 12 | 11.4 | 11.6 | 12.3 | 13.4 | 14.7 | 16.4 |
| SD | 29.5 | 22.8 | 33.7 | 33.6 | 32.9 | 32.5 | 31.7 | 32 | 32.9 | 34.1 | 35.4 | 37.1 |
Data are presented as mean percentage with standard deviation of ACSC (ambulatory care sensitive conditions) hospitalizations based on biological sex using the Purdy definition of ACSC
Mean ACSC hospitalizations stratified by race for each age group is presented in Table 4. Among all ages, Black (12.3, SD = 32.8) and Native American (12.1, SD = 32.6) inpatients had higher average percentage of ACSC hospitalizations than White (10.5, SD = 30.7), Asian or Pacific Islander (7.4, SD = 26.2), and Hispanic (9.8, SD = 29.8) inpatients. This finding was consistent across most age groups with a few exceptions, including in the 65-69y age group with Black and Hispanic inpatients having the highest means, and in the 85-89y age group with Native American and Hispanic inpatients having the highest means. In the ≥90y age group, Native American patients had the highest average percentage of ACSC hospitalizations which was considerably higher than all other age and race groups. While there was not a clear trend for increasing average percentage of ACSC hospitalizations with age for all races, the highest mean among all race and age groups occurred among inpatients ≥90y.
Table 4.
Mean Percentage of ACSC Hospitalizations Based on Race Using the Purdy Definition
| Race | Statistic | All ages | <45 | 45–49 | 50–54 | 55–59 | 60–64 | 65–69 | 70–74 | 75–79 | 80–84 | 85–89 | ≥90 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| White | Mean | 10.5 | 6.8 | 12.5 | 12.2 | 11.6 | 11.3 | 10.7 | 11.1 | 11.9 | 13.1 | 14.5 | 16.4 |
| SD | 30.7 | 25.2 | 33 | 32.7 | 32.1 | 31.6 | 31 | 31.5 | 32.3 | 33.7 | 35.2 | 37 | |
| Black | Mean | 12.3 | 9.6 | 16 | 15.6 | 15.1 | 14.5 | 13.6 | 13.3 | 13.5 | 14 | 13.7 | 14.6 |
| SD | 32.8 | 29.4 | 36.7 | 36.3 | 35.8 | 35.2 | 34.2 | 33.9 | 34.2 | 34.7 | 34.4 | 35.3 | |
| Hispanic | Mean | 9.8 | 6.8 | 14.2 | 14.3 | 14 | 13.7 | 12.8 | 13 | 13.5 | 14 | 14.8 | 16.4 |
| SD | 29.8 | 25.1 | 35 | 35 | 34.7 | 34.3 | 33.4 | 33.6 | 34.2 | 34.7 | 35.5 | 37 | |
| Asian or Pacific Islander | Mean | 7.4 | 3.9 | 11 | 10.8 | 10.3 | 10.7 | 10.8 | 10.7 | 11.9 | 12.6 | 13.3 | 13.7 |
| SD | 26.2 | 19.5 | 31.3 | 31 | 30.4 | 30.9 | 31 | 30.9 | 32.4 | 33.1 | 34 | 34.4 | |
| Native American | Mean | 12.1 | 9.4 | 16 | 15.3 | 14.3 | 14.6 | 12.5 | 13.7 | 14.3 | 15.5 | 16.6 | 18.3 |
| SD | 32.6 | 29.2 | 36.7 | 36 | 35 | 35.3 | 33.1 | 34.4 | 35 | 36.2 | 37.2 | 38.7 |
Data are presented as mean percentage with standard deviation of ACSC (ambulatory care sensitive conditions) hospitalizations based on race using the Purdy definition of ACSC
The concordance between ACSC hospitalization rates using the Purdy definition and the presence of comorbidities using the Charlson Comorbidity Index was statistically significant but small with Gwet’s AC1 statistic of 0.094 (95% CI 0.093, 0.095), indicating only 9.4% agreement between admission for any ACSC and the presence of any comorbidity measured by the Charlson index. Of ACSC hospitalizations, 81% had at least one comorbidity, while 19% had no comorbidities. Of hospitalizations with at least one comorbidity, 15% were admitted for ACSCs while 85% were admitted for conditions other than ACSCs. However, the comorbidity scores were higher among inpatients admitted for an ACSC compared to those admitted for other reasons (1.82 versus 1.19, p < 0.001).
DISCUSSION
ACSCs are acute and chronic health conditions that, when not prevented or treated in the primary care setting, can lead to hospitalization.23 Thus, ACSC hospitalization rates are often used as indicators of the adequacy of primary care services.24–27 Previous research on ACSC hospitalizations in the US have examined rates of ACSC hospitalizations in Medicare relative to the total number of beneficiaries, rather than the number of hospitalized beneficiaries.28 Such information is useful for governmental and other public health agencies in efforts to increase the quality of primary care in regions with the highest rates of ACSCs and in the allocation of health care resources to regions with high rates of ACSCs to care for potential excess hospitalizations.24 However, data on the rates of ACSC hospitalizations relative to the total number of beneficiaries, rather than the number of hospitalized beneficiaries, does little to assist individual hospitals and their affiliated primary care practices in identifying rates of ACSC admissions that are higher than expected, indicating a potential need for change in healthcare delivery at the individual practice level. This study addresses this unmet need by providing US national rates of ACSC hospitalizations among adult inpatients stratified by age, sex, and race, which is the first study of its kind to our knowledge.
The need to lower the rates of ACSC hospitalizations generally has been expressed, and comparison to the national rates in this study allows providers and administrators to target reduction efforts to those healthcare systems with the greatest need. Furthermore, this study also reports the rates of ACSC hospitalizations by condition and age and average hospitalizations due to ACSC stratified by sex and race. The most common ACSCs contributing to hospitalization are pneumonia, congestive heart failure, and diabetes, with inpatients who are ≥90y, Black or Native American, and male having the highest mean ACSC hospitalizations. This level of detail will allow even more refined targeting of reduction efforts to the specific local healthcare environment. In addition, this study demonstrates that the overlap between ACSC hospitalizations and comorbidity is small, with only 9.4% agreement between the presence of comorbidities on the Charlson index and the presence of an ACSC admissions. This indicates not only that many individuals with comorbidities are admitted for non-ACSC conditions, but also that a substantial minority of ACSC admissions occur among individuals without commonly recognized comorbidities. Thus, there is a need to focus specifically on reductions in ACSCs and not just on reductions in comorbidity generally.
Hospitalizations for ACSCs are not benign events and warrant specific efforts to reduce their occurrence. For one, these conditions contribute greatly to healthcare expenditures annually. A study of AHRQ data from 2006 found that inpatient healthcare costs for potentially preventable conditions was more than $30 billion in one year, $15 billion of which was due solely to hospitalizations for pneumonia and congestive heart failure.1 Secondly, ACSC hospitalizations contribute to other adverse health issues including delirium and hospital-associated disability.29,30
It might also be necessary to target at risk populations to reduce ACSC admissions. Prior work has shown that risk for ACSC hospitalizations is increased among individuals who are older, female, non-white (non-Hispanic Black, more specifically), live in the rural setting, and do not have a primary care provider.31,32 These evidence are generally consistent with findings of the present investigation with a few exceptions. The findings herein support that index hospitalization rates for ACSCs increase with age except for inpatients aged 65-74y. A potential reason for the observed reduction is that new receipt of Medicare benefits among these age groups stimulates greater use of primary care health services for ACSCs, thus leading to fewer ACSC hospitalizations.5 Primary care access, quality and use could also be contributing to increased ACSC rates among Black and Native American inpatients in the present study. There are profound social and structural inequities experienced by Native and Black American communities that create barriers to receiving appropriate and necessary healthcare. Evidence supports that reduced access to healthcare insurance coverage, to culturally competent health workforce, and to health services in general along with high rates of poverty and suboptimal health literacy impacts access to preventative healthcare services, including primary care.33–36 Further, Black and Native American adults suffer disproportionately from multiple conditions that are common sources of ACSC hospitalizations,32,37 which could also contribute to increased mean ACSC hospitalizations observed in the present study. Assessment of the effect of sex on ACSC hospitalizations showed that males had a slightly higher average number of ACSC hospitalizations than females despite there being more females in the nation-wide sample overall. This is in contrast with another study showing the opposite trend, though this previous study did not include participants that were older than 65y and was also specific to one region of the US.31
A major strength of this investigation is the use of a national inpatient sample to calculate ACSC admission rates. This provides a more representative and generalizable sample than using data from a single state or region of the US or data that is specific to a particular ACSC. Further, this investigation focuses on hospitalization rates due to all ACSCs across the age continuum and mean ACSC hospitalizations based on sex and race for each age group, which, to the best of our knowledge has not been reported previously.
Despite the many strengths of this investigation, there are some limitations worth noting. For one, this analysis was completed with data collected before the era of COVID-19. The rapid transition to telemedicine at the start of the COVID-19 pandemic left many individuals living in the rural setting and in poverty without a method for accessing primary care services.38 Further, it became more difficult for communities with significant social and structural health inequities to receive preventative screenings and other services through primary care.39,40 As a result, the pandemic may have amplified the findings related to ACSC hospitalizations and Native American and Black adults in the US. As health services return to greater use of in-person visits as COVID-19 cases decline, we would expect hospitalizations for ACSCs to be similar to the rates reported here, but this needs to be confirmed in future research. Secondly, the primary discharge diagnoses were made using ICD-10 codes which will not effectively capture nonbillable diagnoses. Also, there is not a standardized process for provider diagnosis and billing practices at the hospitals represented in the NIS dataset. The NIS dataset captures multiple hospitalizations for the same subject in the same year as separate events, but due to deidentification of the data, the correlation among multiple admissions cannot be accounted for in the analysis. Further, this analysis examined a single year of inpatient data, thus does not provide information about changes in ACSC admission rates overtime. Finally, there is not a gold standard definition or classification method for ACSCs. Our team used the Purdy definition for ACSCs which is a very commonly used definition clinically but does not include some very specific conditions that are part of other definitions such as depression, dementia, or atrial fibrillation. Future research should aim to determine the gold standard classification for ACSC hospitalizations for better comparison across hospitals, healthcare systems, and research studies.
In conclusion, the national norms for ACSC hospitalizations in the US are provided using the 2018 NIS database. ACSC hospitalization rates across key demographics provide invaluable insight into the impact of individual ACSCs on the US inpatient healthcare system based on age, race, and sex. This knowledge can help direct the development of policies and interventions for reducing potentially preventable hospitalizations due to ACSCs among individual hospitals and associated primary care practices. Prioritizing primary care visits and improving early recognition of specific conditions will allow ACSCs to be treated before hospitalization is required, which could lead to increased hospital bed availability and reduced overall healthcare expenditures.
Author Contributions
Loyd – conceptualization, methodology, data curation, resources, supervision, visualization, writing – original draft, writing – review and editing, final draft approval.
Blue – conceptualization, investigation, writing- original draft, writing – review and editing, final draft approval
Turner – conceptualization, investigation, writing – original draft, writing – review and editing, final draft approval
Weber – conceptualization, investigation, writing – original draft, writing – review and editing, final draft approval
Guy – conceptualization, investigation, writing – original draft, writing – review and editing, final draft approval
Zhang – data curation, formal analysis, methodology, software, writing – original draft, final draft approval
Martin – conceptualization, investigation, methodology, validation, writing – original draft, final draft approval
Kennedy – conceptualization, data curation, formal analysis, investigation, methodology, software, resources, validation, visualization, supervision, writing- original draft, writing- review and editing, final draft approval.
Brown—conceptualization, methodology, validation, resources, writing- original draft, writing- review and editing, final draft approval.
Funding
This work was supported by funding provided by the UAB Clinical and Diagnostic Sciences Department (CL). Additional support was provided by the Center for Clinical and Translational Science at UAB (which is supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under award number UL1TR003096). Further, REK is supported by the NIA under award number AG060993. The authors verify that the content of this manuscript is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Data Availability
The data used in this study are publicly available through the Healthcare Cost and Utilization Project National Inpatient Sample database. https://hcup-us.ahrq.gov/nisoverview.jsp.
Declarations:
Conflict of Interest:
The authors declare that there are no conflicts of interest.
Prior Presentation or Publication:
The authors declare that this work is original research and has not been published in part or in whole in any language previously. This work has not been presented.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Jiang HJ, Russo CA, Barrett ML. Nationwide Frequency and Costs of Potentially Preventable Hospitalizations, 2006: Statistical Brief #72. Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. Agency for Healthcare Research and Quality (US); 2006. [PubMed]
- 2.Weissman JS, Gatsonis C, Epstein AM. Rates of avoidable hospitalization by insurance status in Massachusetts and Maryland. JAMA. 1992;268(17):2388-94. [PubMed]
- 3.Billings J, Zeitel L, Lukomnik J, et al. Impact of socioeconomic status on hospital use in New York City. Health Aff (Millwood). Spring 1993;12(1):162–73. 10.1377/hlthaff.12.1.162. [DOI] [PubMed]
- 4.Purdey S, Huntley A. Predicting and preventing avoidable hospital admissions: a review. J R Coll Physicians Edinb. 2013;43(4):340-4. 10.4997/jrcpe.2013.415. [DOI] [PubMed]
- 5.White BM, Ellis C, Jr., Simpson KN. Preventable hospital admissions among the homeless in California: a retrospective analysis of care for ambulatory care sensitive conditions. BMC Health Serv Res. 2014;14:511-511. 10.1186/s12913-014-0511-7. [DOI] [PMC free article] [PubMed]
- 6.Rosano A, Loha CA, Falvo R, et al. The relationship between avoidable hospitalization and accessibility to primary care: a systematic review. Eur J Public Health. 2012;23(3):356-360. 10.1093/eurpub/cks053. [DOI] [PubMed]
- 7.Galarraga JE, Mutter R, Pines JM. Costs associated with ambulatory care sensitive conditions across hospital-based settings. Acad Emerg Med. 2015;22(2):172-81. 10.1111/acem.12579. [DOI] [PubMed]
- 8.Will JC, Zhang Z, Ritchey MD, et al. Medication Adherence and Incident Preventable Hospitalizations for Hypertension. Am J Prev Med. 2016;50(4):489-499. 10.1016/j.amepre.2015.08.021. [DOI] [PubMed]
- 9.Kuo YF, Chen NW, Baillargeon J, et al. Potentially Preventable Hospitalizations in Medicare Patients With Diabetes: A Comparison of Primary Care Provided by Nurse Practitioners Versus Physicians. Med Care. 2015;53(9):776-83. 10.1097/mlr.0000000000000406. [DOI] [PMC free article] [PubMed]
- 10.Mazurenko O, Shen J, Shan G, et al. Nevada's Medicaid expansion and admissions for ambulatory care-sensitive conditions. Am J Manag Care. 2018;24(5):e157-e163. [PubMed]
- 11.Sentell TL, Seto TB, Quensell ML, et al. Insights in Public Health: Outpatient Care Gaps for Patients Hospitalized with Ambulatory Care Sensitive Conditions in Hawai'i: Beyond Access and Continuity of Care. Hawaii J Health Soc Welf. 2020;79(3):91-97. [PMC free article] [PubMed]
- 12.Hutchison J, Thompson ME, Troyer J, et al. The effect of North Carolina free clinics on hospitalizations for ambulatory care sensitive conditions among the uninsured. BMC Health Serv Res. 2018;18(1):280. 10.1186/s12913-018-3082-1. [DOI] [PMC free article] [PubMed]
- 13.Chang CF, Troyer JL. Trends in potentially avoidable hospitalizations among adults in Tennessee, 1998–2006. Tenn Med. 2011;104(10):35–8, 45. [PubMed]
- 14.US Department of Health and Human Services. Healthcare Cost and Utilization Project. https://www.hcup-us.ahrq.gov/nisoverview.jsp.Accessed 1 Jan 2021.
- 15.World Health Organization. ICD-10 : international statistical classification of diseases and related health problems : tenth revision, 2nd ed. https://apps.who.int/iris/handle/10665/42980.Accessed 22 Feb 2021.
- 16.Ciol MA, Hoffman JM, Dudgeon BJ, et al. Understanding the use of weights in the analysis of data from multistage surveys. Arch Phys Med Rehabil. 2006;87(2):299-303. 10.1016/j.apmr.2005.09.021. [DOI] [PubMed]
- 17.Barnett K, Mercer SW, Norbury M, et al. Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study. Lancet. 2012;380(9836):37-43. 10.1016/s0140-6736(12)60240-2. [DOI] [PubMed]
- 18.Purdy S, Griffin T, Salisbury C, et al. Ambulatory care sensitive conditions: terminology and disease coding need to be more specific to aid policy makers and clinicians. Public Health. 2009;123(2):169-73. 10.1016/j.puhe.2008.11.001. [DOI] [PubMed]
- 19.Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. Journal of Chronic Diseases. 1987;40(5):373–383. 10.1016/0021-9681(87)90171-8. [DOI] [PubMed]
- 20.Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130-9. 10.1097/01.mlr.0000182534.19832.83. [DOI] [PubMed]
- 21.Lumley T. Analysis of Complex Survey Samples. J Stat Softw. 2004;9(8):1–19. 10.18637/jss.v009.i08.
- 22.Gwet KL. Computing inter-rater reliability and its variance in the presence of high agreement. Br J Math Stat Psychol. 2008;61(Pt 1):29-48. 10.1348/000711006x126600. [DOI] [PubMed]
- 23.Hodgson K, Deeny SR, Steventon A. Ambulatory care-sensitive conditions: their potential uses and limitations. BMJ Qual Saf. 2019;28(6):429-433. 10.1136/bmjqs-2018-008820. [DOI] [PubMed]
- 24.Kim AM, Park JH, Yoon TH, et al. Hospitalizations for ambulatory care sensitive conditions as an indicator of access to primary care and excess of bed supply. BMC Health Serv Res. 2019;19(1):259. 10.1186/s12913-019-4098-x. [DOI] [PMC free article] [PubMed]
- 25.Lavoie JG, Philips-Beck W, Kinew KA, et al. The relationship between rates of hospitalization for ambulatory care sensitive conditions and local access to primary healthcare in Manitoba First Nations communities : Results from the Innovation in Community-based Primary Healthcare Supporting Transformation in the Health of First Nations in Manitoba (iPHIT) study. Can J Public Health. 2021;112(2):219-230. 10.17269/s41997-020-00421-3. [DOI] [PMC free article] [PubMed]
- 26.Gibson OR, Segal L, McDermott RA. A systematic review of evidence on the association between hospitalisation for chronic disease related ambulatory care sensitive conditions and primary health care resourcing. BMC Health Serv Res. 2013;13:336. 10.1186/1472-6963-13-336. [DOI] [PMC free article] [PubMed]
- 27.Caminal J, Starfield B, Sánchez E, et al. The role of primary care in preventing ambulatory care sensitive conditions. Eur J Public Health. 2004;14(3):246–251. 10.1093/eurpub/14.3.246. [DOI] [PubMed]
- 28.Weeks WB, Wang H, Smith J, et al. Ambulatory Care Sensitive Condition Admission Rates in Younger and Older Traditional Medicare and Medicare Advantage Populations, 2011-2019. J Gen Intern Med. 2022;37(7):1814-1817. 10.1007/s11606-021-06955-7. [DOI] [PMC free article] [PubMed]
- 29.Kennedy RE. Delirium due to avoidable hospitalizations among older adults. Alzheimer's Dement. 2021;17(S10):e056449. 10.1002/alz.056449.
- 30.Loyd C, Markland AD, Zhang Y, et al. Prevalence of Hospital-Associated Disability in Older Adults: A Meta-analysis. J Am Med Dir Assoc. 2020;21(4):455-461.e5. 10.1016/j.jamda.2019.09.015. [DOI] [PMC free article] [PubMed]
- 31.Shi L, Samuels ME, Pease M, et al. Patient characteristics associated with hospitalizations for ambulatory care sensitive conditions in South Carolina. South Med J. 1999;92(10):989-98. 10.1097/00007611-199910000-00009. [DOI] [PubMed]
- 32.Jørgensen TSH, Allore H, Elman MR, et al. The importance of chronic conditions for potentially avoidable hospitalizations among non-Hispanic Black and non-Hispanic White older adults in the US: a cross-sectional observational study. BMC Health Serv Res. 2022;22(1):468. 10.1186/s12913-022-07849-y. [DOI] [PMC free article] [PubMed]
- 33.Arnett MJ, Thorpe RJ, Jr., Gaskin DJ, et al. Race, Medical Mistrust, and Segregation in Primary Care as Usual Source of Care: Findings from the Exploring Health Disparities in Integrated Communities Study. J Urban Health. 2016;93(3):456-67. 10.1007/s11524-016-0054-9. [DOI] [PMC free article] [PubMed]
- 34.Artiga S, Arguello R, Duckett P. Health coverage and care for American Indians and Alaska Natives. In: Disparities Policy: The Kaiser Family Foundation. https://www.kff.org/disparities-policy/issue-brief/health-coverage-and-care-for-american-indians-and-alaska-natives/.Accessed 4 May 2022.
- 35.Willging CE, Sommerfeld DH, Jaramillo ET, et al. “Improving Native American elder access to and use of health care through effective health system navigation”. BMC Health Serv Res. 2018;18(1):464. 10.1186/s12913-018-3182-y. [DOI] [PMC free article] [PubMed]
- 36.Connell CL, Wang SC, Crook L, et al. Barriers to Healthcare Seeking and Provision Among African American Adults in the Rural Mississippi Delta Region: Community and Provider Perspectives. J Commun Health. 2019;44(4):636–645. 10.1007/s10900-019-00620-1. [DOI] [PMC free article] [PubMed]
- 37.Veazie M, Ayala C, Schieb L, et al. Trends and disparities in heart disease mortality among American Indians/Alaska Natives, 1990–2009. Am J Public Health. 2014;104 Suppl 3(Suppl 3):S359–67. 10.2105/ajph.2013.301715. [DOI] [PMC free article] [PubMed]
- 38.Cantor JH, McBain RK, Pera MF, et al. Who Is (and Is Not) Receiving Telemedicine Care During the COVID-19 Pandemic. Am J Prev Med. 2021;61(3):434-438. 10.1016/j.amepre.2021.01.030. [DOI] [PMC free article] [PubMed]
- 39.Satish T, Raghunathan R, Prigoff JG, et al. Care Delivery Impact of the COVID-19 Pandemic on Breast Cancer Care. JCO Oncol Pract. 2021;17(8):e1215-e1224. 10.1200/op.20.01062. [DOI] [PubMed]
- 40.Collins N, Crowder J, Ishcomer-Aazami J, et al. Perceptions and Experiences of Frontline Urban Indian Organization Healthcare Workers With Infection Prevention and Control During the COVID-19 Pandemic. Front Sociol. 2021;6:611961. 10.3389/fsoc.2021.611961. [DOI] [PMC free article] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data used in this study are publicly available through the Healthcare Cost and Utilization Project National Inpatient Sample database. https://hcup-us.ahrq.gov/nisoverview.jsp.