Rural-Urban Disparities in Quality of Inpatient Psychiatric Care
Abstract
Objective:
Rural residents have higher rates of serious mental illness than urban residents, but little is known about the quality of inpatient psychiatric care available to them locally or how quality may have changed in response to federal initiatives. This study aimed to examine differences and changes in the quality of inpatient psychiatric care in rural and urban hospitals.
Methods:
This national retrospective study of 1,644 facilities examined facility-level annual quality-of-care data from the Inpatient Psychiatric Facility Quality Reporting program, 2015–2019. Facility location was categorized as urban, large rural, or small or isolated rural on the basis of zip code–level rural-urban commuting area codes. Generalized regression models were used to assess rural-urban differences in care quality (five continuity-of-care and two patient experience measures) and changes over time.
Results:
Rural inpatient psychiatric units performed better than urban units in nearly all domains. Improvements in quality of care (excluding follow-up care) were similar in rural and urban units. Rates of 30- and 7-day postdischarge follow-up care decreased in all hospitals but faster in rural units. Timely transmission of transition records was more frequent in small or isolated rural versus urban units (mean marginal difference=22.5, 95% CI=6.3–38.8). Physical restraint or seclusion use was less likely in rural than in urban units (OR=0.6, 95% CI=0.5–0.8).
Conclusions:
Rural psychiatric units had better care quality at baseline (better follow-up care, better timely transmission of transition records, and lower rates of physical restraint use) than urban units, but during 2015–2019, follow-up care performance decreased overall and more in rural than urban units.
HIGHLIGHTS
•
Patients in rural inpatient psychiatric units had better continuity of care and better experiences with care than those in urban units.
•
Continuity of care—measured as follow-up care within 7 or 30 days of discharge and provision of appropriate justification for prescription of multiple antipsychotic medications at discharge—declined between 2015 and 2019, whereas management of transition records improved consistently across rural and urban facilities.
•
Rural facilities were less likely than urban facilities to use physical restraints and seclusion during the study period.
One in five adults experiences mental illness each year, and among these individuals, one-quarter experience serious mental illness (1). Serious mental illness has affected an increasing number of adults in the United States, from 9.8 million in 2014 to 14.2 million in 2020 (1). The overall prevalence of serious mental illness is estimated to be similar across rural and urban adult residents, with rates increasing from 3.8% in 2010 to 5.0% in 2020 in large metropolitan or urban counties and from 4.6% to 5.7% in nonmetropolitan or rural counties during the same period (2). Inpatient psychiatric facilities serve as essential sources of care for individuals with serious mental illness. Among all adults with serious mental illness, an estimated 929,000 (6.5%) received inpatient treatment in 2020 (3). Given that nearly 95% of rural counties have mental health professional shortage areas (MHPSAs) (4, 5), the quality of the care that is available in these regions is particularly important. However, little is known about the quality of inpatient psychiatric care in rural and urban hospitals in the United States.
Patients hospitalized in psychiatric units often have complex conditions that require continual engagement with health care providers (6). However, rural residents face provider shortages at a far greater rate than do urban residents. Across all rural counties, 65% lacked a psychiatrist in 2015, compared with 27% of urban counties (4). Assessing the total mental health workforce, the Health Resources and Services Administration has identified 3,916 (of 6,464; 60.6%) MHPSAs in rural America (5). As of September 2022, about 37 million rural residents lived in MHPSA communities, and 7,871 mental health practitioners are needed to remedy these shortages (5).
The limited mental health workforce may make it difficult for rural psychiatric units to meet many patient-related quality improvement metrics (7, 8). These measures include promoting individuals’ autonomy, independence, and safety by minimizing use of restraint and seclusion; providing appropriate justification for prescribing individuals multiple antipsychotic medications; and offering a transition record to patients and caregivers at discharge (7). For individuals discharged from inpatient psychiatric facilities, shortages of mental health providers might also hinder their ability to receive early follow-up care, potentially leading to higher rates of suicide attempt and death (9). Stigma surrounding mental health in small communities may pose an additional barrier, leading many rural patients to bypass their local rural hospital and instead seek care in urban hospitals (10). Conversely, rural hospitals may have some quality advantages. Urban inpatient psychiatric units may face difficulties providing evidence-based continuity of care to rural patients because such units would need to make additional efforts to connect rural patients to outpatient care, whereas rural units discharging patients to proximal rural settings may more readily identify and work directly with local providers for transition planning and record transmission (11). Indeed, one study found that rural residents were more likely to be admitted to high-quality facilities than were urban residents (12).
Recognizing differences in quality of care among inpatient psychiatric care providers, the Patient Protection and Affordable Care Act required the development of metrics for care quality in mental health services. These metrics were developed by the Centers for Medicare and Medicaid Services (CMS) and were implemented in 2012 (13). Data from the Inpatient Psychiatric Facility Quality Reporting (IPFQR) program assess quality of care within inpatient units across multiple dimensions; these data have been publicly available since 2015. Since the IPFQR implementation, studies have noted improvements in quality of care associated with IPFQR (14, 15). However, studies of rural facilities have been sparse. One study found variation in rates of 30-day readmission across hospitals in metropolitan (20.3%), rural micropolitan (19.7%), and other rural counties (20.4%) (16); however, that study did not focus on rurality and did not consider care coordination and patient safety performance across inpatient psychiatric facilities. In this study, we aimed to provide an overview of the quality of inpatient psychiatric care in rural and urban hospitals.
Methods
Data Sources
This retrospective study included both cross-sectional and longitudinal analyses of data from the publicly available 2015–2019 IPFQR program data sets (13, 17). Under this program, inpatient psychiatric facilities must report their performance on a set of structure- and process-related quality measures or face a two-percentage-point penalty to their Medicare standard federal rate for that year (13). All facilities with inpatient psychiatric services furnished by Medicare and classified as psychiatric hospitals or distinct psychiatric units in acute care or critical access hospitals are subject to the IPFQR program requirements (13). The CMS measures inpatient psychiatric quality across the domains of continuity of care, patient experience, readmission, and substance use screening and treatment as part of the National Quality Strategy (18). Values for measures are calculated by using Medicare claims and clinical notes data.
We linked IPFQR data to information from the 2015–2019 American Hospital Association Annual Surveys for hospital characteristics (19) and to the 2015–2019 American Community Survey zip code tabulation area (ZCTA)–level 5-year estimates of community characteristics for the community in which a facility was located (20).
Institutional review board approval was waived because the analyses were based on publicly available data. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (or STROBE) reporting guidelines.
Measures
Rurality was measured at the ZCTA level and was categorized into three groups on the basis of rural-urban commuting area (RUCA) codes: urban (primary RUCA codes, 1–3), large rural (4–6), and small or isolated rural communities (7–10). This study included 1,254 unique urban hospitals, 260 unique large rural hospitals, and 130 unique small or isolated rural hospitals, with variations in the number of hospitals across years (see Table S1 in the online supplement to this article).
Since 2015, CMS has published and updated annually the facility-specific quality performance metrics used in this study, except for two continuity-of-care measures related to transitions to outpatient care, which were first published in 2017. Continuity-of-care measures included follow-up care, appropriate justification for multiple antipsychotic medications at discharge, and transition record management. CMS quality metrics are calculated across varying populations, as appropriate for each metric. (Descriptions of eligible populations per measure are available in the online supplement .)
For the two patient experience measures, CMS used the number of minutes that psychiatric inpatients in a facility were kept in physical restraint or seclusion and converted this figure to hours. In this study, we categorized whether physical restraint or seclusion was used in each facility each year, because more than half of the facilities had no physical restraint or seclusion events and the variation in hours of physical restraint or seclusion use was driven by urban facilities (see Figure S1 in the online supplement ).
Research suggests that for-profit or nonprofit status is associated with changes in use of restraint or seclusion after the IPFQR implementation, and this characteristic was therefore included in our adjusted analyses (21). Other hospital characteristics included hospital primary services (psychiatric or general medicine and surgery), system affiliation, teaching status, accreditation by the Joint Commission or Det Norske Veritas, critical access hospital status, rural referral center status, number of psychiatric beds, and proportion of registered nurses among all hospitalwide nurses. ZCTA-level metrics included age groups, racial-ethnic composition, and sociodemographic factors (rates of unemployment, uninsurance, households with broadband access, and households <200% of the federal poverty level).
Statistical Analysis
We first used chi-square tests for frequency distributions and one-way analysis of variance tests to compare hospital and ZCTA-level characteristics of facilities located in urban, large rural, and small or isolated rural areas. Descriptive and multivariable regression analyses were used to identify differential trends in quality of inpatient psychiatric care across urban, large rural, and small or isolated rural facilities. In the unadjusted analysis, we used Mann-Kendall trend tests for the continuity-of-care measures, because these tests are operationalized as continuous variables, and used Cochran-Armitage tests to compare trends in proportions over the years of study for use of physical restraint and seclusion. In the multivariable regression analysis, we used generalized linear regressions for the continuity-of-care outcomes and logistic regressions for the patient experience outcomes, both with adjusted standard errors to account for state-level clustering and Bonferroni-corrected p values for multiple comparisons. Models controlled for the aforementioned covariates and included interaction terms between year and hospital rurality to evaluate the annual differential trends in outcomes across urban, large rural, and small or isolated rural facilities. Variance inflation factors (VIFs) were calculated, and no violations of multicollinearity between independent variables were observed (VIF=1.86).
Results
Description of Facilities
Of 1,644 hospitals with inpatient psychiatric capability in 2019, about 76% were in urban areas, 16% in large rural areas, and 8% in small or isolated rural areas (Table 1). Compared with the urban inpatient psychiatric facilities, rural facilities were significantly more likely to be principally general medicine and surgery hospitals (vs. specialized psychiatric hospitals) and to be publicly owned (vs. private facilities) (both p<0.001). The rural hospitals studied had fewer beds than their urban counterparts and were less likely to be accredited, to be teaching hospitals, to offer alcohol or drug dependency services, or to be system affiliated.
| Hospital characteristic | Urban (N=1,254) | Large rural (N=260) | Small or isolated rural (N=130) | |||
|---|---|---|---|---|---|---|
| N | % | N | % | N | % | |
| Primary servicesb | ||||||
| Psychiatric | 467 | 37 | 71 | 27 | 19 | 15 |
| General medicine and surgery | 787 | 63 | 189 | 73 | 111 | 85 |
| Ownershipb | ||||||
| Public | 237 | 19 | 77 | 30 | 47 | 36 |
| Private nonprofit | 595 | 48 | 121 | 47 | 57 | 44 |
| Private for-profit | 422 | 34 | 62 | 24 | 26 | 20 |
| System affiliationb | 868 | 69 | 141 | 54 | 57 | 44 |
| Teaching hospitalb | 601 | 48 | 38 | 15 | 6 | 5 |
| Accredited hospitalb | 1,111 | 89 | 196 | 75 | 69 | 53 |
| Alcoholism or chemical dependency outpatient servicesb | 374 | 30 | 35 | 14 | 5 | 4 |
| CMS hospital designation | ||||||
| Critical access hospitalb | 11 | <1 | 12 | 5 | 38 | 29 |
| Rural referral center | 34 | 3 | 34 | 13 | 3 | 2 |
| Sole community hospital | 22 | 2 | 46 | 18 | 24 | 19 |
| Facility-level psychiatric beds (M±SD)b | 57.7±60.3 | 32.1±56.1 | 25.1±65.1 | |||
| Facility-level % registered nurses of all nurses (M±SD)b | 91.2±9.4 | 86.8±10.9 | 82.8±11.4 | |||
| Community characteristic | M | SD | M | SD | M | SD |
| ZCTA-level age-group mixb | ||||||
| % residents ages <15 years | 11.4 | 3.4 | 12.3 | 2.3 | 12.4 | 2.7 |
| % residents ages 15–24 | 14.6 | 9.9 | 14.1 | 5.7 | 12.0 | 3.6 |
| % residents ages 25–44 | 27.8 | 7.1 | 24.1 | 3.5 | 23.2 | 3.8 |
| % residents ages 45–64 | 18.3 | 3.7 | 18.8 | 2.4 | 19.6 | 2.6 |
| % residents ages ≥65 | 15.6 | 6.2 | 18.2 | 6.3 | 19.9 | 4.6 |
| ZCTA-level sociodemographic mixb | ||||||
| % unemployed | 5.8 | 4.1 | 5.3 | 2.2 | 5.5 | 3.4 |
| % below federal poverty level | 16.4 | 10.2 | 17.5 | 6.7 | 19.3 | 9.0 |
| % uninsured | 8.7 | 5.6 | 9.2 | 4.7 | 9.6 | 5.4 |
| % with broadband access | 82.0 | 9.1 | 76.7 | 7.6 | 70.9 | 10.1 |
| % <200% federal poverty level | 34.2 | 15.4 | 38.6 | 9.5 | 42.3 | 10.8 |
| ZCTA-level race-ethnicity mixb | ||||||
| % non-Hispanic White | 68.8 | 21.7 | 81.4 | 16.9 | 80.7 | 20.2 |
| % non-Hispanic Black | 17.7 | 20.8 | 11.9 | 16.9 | 13.5 | 19.2 |
| % Hispanic | 12.2 | 10.8 | 7.2 | 7.9 | 3.8 | 4.4 |
| % non-Hispanic Asian | 5.8 | 8.2 | 1.3 | 2.2 | .7 | 2.5 |
| % non-Hispanic American Indian or Alaska Native | .6 | 1.2 | 1.2 | 2.8 | 2.0 | 6.4 |
a
CMS, Centers for Medicare and Medicaid Services; ZCTA, zip code tabulation area.
b
Differences among areas were statistically significant (p<0.001); p values were calculated from chi-square tests for frequency distributions and from one-way analysis of variance for ZCTA-level characteristics across urban, large rural, and small or isolated rural facilities.
The population in study hospitals’ local communities differed across the rural-urban continuum. Hospitals in both large and small or isolated rural ZCTAs had larger proportions of patients reporting their race-ethnicity as non-Hispanic White than did hospitals in urban ZCTAs. Rural populations were older than in urban areas, with the proportion of the population ages ≥65 years increasing from mean±SD=15.6%±6.2% in urban ZCTAs to 18.2%±6.3% in large rural ZCTAs and 19.9%±4.6% in small or isolated rural ZCTAs (means and SDs of percentages were for ZCTAs across all areas). Rural communities were more economically vulnerable, with higher levels of poverty, near poverty, and lack of health insurance, along with lower levels of broadband access.
Quality of Care
Rural facilities had better performance in 2019 compared with urban facilities on nearly all measures, with rural facilities having higher percentages in continuity-of-care measures and lower rates in use of physical restraint and seclusion (Table 2). Throughout the study years, rural facilities consistently performed better in transition record management for discharged patients, physical restraint use, and seclusion use, compared with urban facilities.
| Measure | Facility-level annual performanceb | % change (baseline to 2019) | p | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2015 | 2016 | 2017 | 2018 | 2019 | ||||||||
| M | SD | M | SD | M | SD | M | SD | M | SD | |||
| Follow-up care ≤30 days after hospitalization for mental illness (%) | <.001 | |||||||||||
| All | 55.6 | 13.2 | 55.5 | 13.5 | 52.5 | 15.2 | 50.1 | 14.6 | 49.6 | 14.0 | −10.7 | |
| Urban | 54.4 | 12.8 | 54.8 | 13.1 | 52.3 | 14.4 | 50.3 | 13.7 | 49.3 | 13.5 | −9.3 | |
| Large rural | 60.2 | 14.2 | 58.2 | 14.6 | 52.5 | 16.3 | 50.5 | 16.6 | 50.7 | 16.0 | −15.7 | |
| Small or isolated rural | 60.4 | 14.0 | 59.1 | 15.2 | 53.8 | 19.5 | 47.3 | 19.0 | 50.2 | 15.8 | −16.8 | |
| Follow-up care ≤7 days after hospitalization for mental illness (%) | .047 | |||||||||||
| All | 32.7 | 12.2 | 34.3 | 12.6 | 29.5 | 13.9 | 27.4 | 12.9 | 27.3 | 12.4 | −16.6 | |
| Urban | 32.1 | 11.7 | 33.8 | 12.2 | 30.1 | 13.3 | 27.8 | 12.2 | 27.6 | 11.9 | −14.0 | |
| Large rural | 34.7 | 13.3 | 35.0 | 13.4 | 26.8 | 14.8 | 26.3 | 14.2 | 26.6 | 14.6 | −23.2 | |
| Small or isolated rural | 36.8 | 16.3 | 41.6 | 17.6 | 28.1 | 17.4 | 24.9 | 16.7 | 24.6 | 13.4 | −33.2 | |
| Discharge with multiple antipsychotics with justification (%) | .034 | |||||||||||
| All | 72.2 | 28.4 | 72.1 | 27.2 | 63.0 | 30.5 | 62.5 | 31.2 | 62.7 | 31.3 | −13.2 | |
| Urban | 72.0 | 27.7 | 71.7 | 26.9 | 64.3 | 29.5 | 63.7 | 30.1 | 62.8 | 30.8 | −12.8 | |
| Large rural | 74.2 | 28.5 | 74.1 | 27.3 | 58.3 | 32.9 | 56.8 | 34.1 | 63.4 | 32.6 | −14.5 | |
| Small or isolated rural | 70.8 | 36.3 | 73.5 | 31.6 | 54.4 | 36.2 | 59.0 | 36.7 | 59.5 | 36.5 | −16.0 | |
| Discharge with transition records with specified elements (%) | <.001 | |||||||||||
| All | 50.7 | 37.2 | 63.1 | 36.0 | 68.3 | 34.0 | 34.5 | |||||
| Urban | 48.3 | 36.8 | 61.5 | 35.9 | 67.4 | 33.9 | 39.6 | |||||
| Large rural | 54.8 | 37.1 | 64.7 | 36.5 | 67.5 | 34.6 | 23.1 | |||||
| Small or isolated rural | 67.3 | 36.4 | 75.3 | 33.7 | 78.6 | 31.9 | 16.8 | |||||
| Timely transmission of transition records (%) | <.001 | |||||||||||
| All | 47.2 | 36.4 | 57.3 | 36.0 | 60.1 | 34.5 | 27.3 | |||||
| Urban | 44.4 | 35.8 | 55.4 | 35.6 | 58.8 | 34.3 | 32.4 | |||||
| Large rural | 52.4 | 36.4 | 59.6 | 37.1 | 61.0 | 34.9 | 16.5 | |||||
| Small or isolated rural | 65.5 | 36.9 | 73.0 | 33.7 | 72.1 | 33.9 | 10.1 | |||||
| N | % | N | % | N | % | N | % | N | % | % change (baseline to 2019) | p | |
| Any physical restraint use | ns | |||||||||||
| All | 1,189 | 74 | 1,182 | 73 | 1,193 | 74 | 1,164 | 75 | 1,116 | 72 | −2.7 | |
| Urban | 991 | 80 | 1,003 | 81 | 1,000 | 80 | 981 | 81 | 948 | 78 | −2.5 | |
| Large rural | 148 | 59 | 133 | 54 | 143 | 60 | 138 | 60 | 127 | 56 | −4.8 | |
| Small or isolated rural | 50 | 41 | 46 | 37 | 50 | 41 | 45 | 39 | 41 | 37 | −10.7 | |
| Any seclusion use | ns | |||||||||||
| All | 930 | 58 | 924 | 57 | 900 | 56 | 914 | 59 | 883 | 57 | −1.9 | |
| Urban | 777 | 63 | 776 | 63 | 761 | 61 | 769 | 64 | 736 | 61 | −4.0 | |
| Large rural | 112 | 44 | 109 | 44 | 105 | 44 | 105 | 45 | 113 | 50 | 12.5 | |
| Small or isolated rural | 41 | 34 | 39 | 32 | 34 | 28 | 40 | 34 | 34 | 30 | −9.5 | |
a
p values were calculated with analysis of variance tests for trends in mean percentages and Cochran-Armitage tests for differences in trends of proportions over the years.
b
See the online supplement for the number of facilities studied in each year, by rural-urban designation.
However, inpatient psychiatric care quality varied substantially across facilities. For example, across all settings, on average, the low-performing (i.e., the bottom quartile) facilities provided only 47% of their patients with detailed discharge information and transmitted only 31% of their patients’ records in time for the patient’s first postdischarge physician visit, compared with 97% and 92%, respectively, for the high-performing (i.e., top-quartile) facilities. These interquartile ranges were larger among urban facilities (46%–96% for transition records with specified elements and 29%–90% for timely transmission of transition records) than among small or isolated rural facilities (77%–100% and 53%–99%, respectively; data are available on request), suggesting greater variation among urban facilities.
Trends in Quality of Care
Between 2015 and 2019, significant rural-urban differences were found in trends concerning the quality of psychiatric care. During this period, facilities located in small or isolated rural ZCTAs had a greater decrease in mean rates of 30-day follow-up care (from 60.7% in 2015 to 48.8% in 2019) than did urban facilities (from 54.6% to 49.2%) (Figure 1), with similar trends for 7-day follow-up rates (see Figure S2 in the online supplement ). The proportion of patients discharged on multiple antipsychotic medications with appropriate justification decreased between 2016 and 2018 across all units (see Figure S3 in the online supplement ).
FIGURE 1. Follow-up care and transition records management in urban, large rural, and small or isolated rural areas, 2015–2019a
aA: mean percentages of inpatient discharges for a diagnosis of mental illness or intentional self-harm among patients ages ≥6 years that resulted in follow-up care with a mental health provider within 30 days. B: mean percentages of discharged inpatients or their caregivers who received a transition record. (Additional details are available in the online supplement to this article. Data points are presented in Table 2. See Figures S2–S5 in the online supplement for other quality measures.)
In multivariable analyses (Table 3), facilities in small or isolated rural ZCTAs had better baseline performance than urban facilities in the percentage of patients receiving 30-day (mean marginal difference in percentage points=6.2, p=0.006) and 7-day follow-up care (mean marginal difference=7.3, p=0.006), percentage of discharged patients receiving transition records with specified elements (mean marginal difference=21.3, p=0.066), and timely transmission of transition records (mean marginal difference=22.5, p=0.033). Physical restraint or seclusion use was less likely in large rural than urban units (OR=0.6, 95% CI=0.5–0.8). Between 2015 and 2019, the rates of 30- and 7-day postdischarge follow-up care decreased by 1.3 and 1.4 percentage points annually among patients discharged from urban facilities, respectively, and these rates decreased more in rural than in urban units (large rural, −1.4 and −1.0 percentage points; small or isolated rural, −2.1 and −2.4 percentage points, respectively). We observed no rural-urban differences in baseline performance or trends regarding appropriate justification for patients discharged with multiple antipsychotic medications, but this measure declined by 2.6 percentage points annually among hospitals in all areas. In contrast, transition care management improved across all areas, with the percentage of discharged patients receiving transition records with specified elements increasing by 9.3 percentage points and timely transmission of transition records increasing by 6.7 percentage points per year.
| Quality differences at baseline (reference: urban hospitals) | Differential annual trends (reference: annual trend for urban hospitals) | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Large rural | Small or isolated rural | Urban hospitals (annual linear trend) | Large rural × year | Small or isolated rural × year | |||||||||||
| Measure | Estb | 95% CI | pc | Estb | 95% CI | pc | Estb | 95% CI | pc | Estb | 95% CI | pc | Estb | 95% CI | pc |
| Follow-up care ≤30 days after hospitalization for mental illness | 4.2 | 2.4, 6.0 | .008 | 6.2 | 3.5, 8.9 | .006 | −1.3 | −1.6, −1.0 | <.001 | −1.4 | −2.1, −.7 | .002 | −2.1 | −3.1, −1.1 | .011 |
| Follow-up care ≤7 days after hospitalization for mental illness | 2.5 | .6, 4.4 | .026 | 7.3 | 4.5, 10.2 | .006 | −1.4 | −1.6, −1.1 | <.001 | −1.0 | −1.7, −.3 | .035 | −2.4 | −3.5, −1.4 | <.001 |
| Discharge with multiple antipsychotics with justification | 2.2 | −3.8, 8.2 | ns | −3.2 | −12.0, 5.5 | ns | −2.6 | −3.5, −1.8 | <.001 | −2.0 | −4.2, .1 | ns | .2 | −2.9, 3.3 | ns |
| Discharge with transition records with specified elements | 11.1 | −1.0, 23.2 | ns | 21.3 | 5.0, 37.5 | ns | 9.3 | 7.7, 10.9 | <.001 | −2.4 | −6.2, 1.5 | ns | −3.1 | −8.2, 2.1 | ns |
| Timely transmission of transition records | 9.7 | −2.4, 21.8 | ns | 22.5 | 6.3, 38.8 | .033 | 6.7 | 5.1, 8.4 | <.001 | −1.9 | −5.8, 1.9 | ns | −3.3 | −8.5, 1.8 | ns |
| OR | 95% CI | pc | OR | 95% CI | pc | OR | 95% CI | pc | OR | 95% CI | pc | OR | 95% CI | pc | |
| Physical restraint use | .62 | .46, .84 | .01 | .57 | .39, .85 | .025 | .98 | .92, 1.03 | ns | 1.00 | .89, 1.12 | ns | .92 | .79, 1.07 | ns |
| Seclusion | .61 | .47, .80 | .008 | .64 | .43, .94 | .045 | .99 | .95, 1.04 | ns | 1.05 | .95, 1.17 | ns | .92 | .79, 1.07 | ns |
a
Models were adjusted for hospital ownership, system affiliation, teaching status, accreditation by the Joint Commission or Det Norske Veritas, critical access hospital status, rural referral center status, number of psychiatric beds, proportion of registered nurses across hospitalwide nurses, and zip code tabulation area–level age and race-ethnicity composition.
b
Mean marginal difference in percentage points. Est, estimate.
c
Bonferroni-adjusted p values are reported for tests on individual comparisons and multiple outcomes.
Discussion
To our knowledge, this is the first study to examine differences in the quality of inpatient psychiatric care in rural and urban hospital settings and their changes over time. We found that, at baseline, patients served in rural hospitals had better continuity of care and care experiences than those served in urban hospitals. Continuity of care declined over the study period at all locations, with the exception of management and timely transmission of transition records. The proportion of patients receiving outpatient mental health follow-up care within 7 or 30 days of discharge also declined across all locations. Similarly, appropriate justifications in the discharge record for patients receiving multiple antipsychotic medications decreased across all facilities. Conversely, management of transition records consistently improved in rural and urban facilities over 2017–2019. Rural facilities were less likely to use physical restraints and seclusion than were urban facilities, with no significant annual changes in use.
The higher quality of inpatient psychiatric care in rural facilities is encouraging, given that rural residents had a higher prevalence than urban residents of suicide death after discharge (22), readmission (23), and emergency department (ED) visits for mental illness (24). Previous studies have found that patients who did not receive follow-up care after discharge were six times more likely to be readmitted than those who received follow-up care (25). Moreover, the risk for suicide has been found to be highest immediately after discharge from psychiatric care and to remain elevated for months thereafter (9, 26, 27). Follow-up after psychiatric hospitalization is therefore critical to ensure that patients’ transition to communities is supported by outpatient mental health providers to mitigate suicidal ideation, avoidable readmission, and ED visits.
Previous research in the United States has found that patients who were discharged from hospital inpatient psychiatric units made millions of ED visits within 30 days of discharge (9). A multistate study found that nearly 18% of ED visits or hospital acute care visits were for complications related to a recent psychiatric hospitalization (28, 29). Ensuring early follow-up care within at least 30 days of discharge is especially important in rural communities, which have shortages of mental health professionals and constraints on acute care capacity (5). However, in this study, even though rural facilities had a better baseline performance in follow-up care compared with urban facilities, the rural units had a steeper decrease in the proportion of patients receiving follow-up care over the study period. These results suggest a need for systematic quality improvement at the facility level to support all—and particularly rural—facilities’ efforts to ensure early follow-up after psychiatric hospitalization.
Detailed discharge information is necessary to maximize benefits from early follow-up care after hospitalization because it often serves as the primary mode of communication between psychiatric units and outpatient providers (30). Coordinated transitions from inpatient to outpatient settings, within the same institution or among different organizations, can prevent hospital readmissions, duplication of services, medical errors, and waste of resources. Previous literature has reported that patients participating in a hospital program of providing detailed, personalized instructions at discharge, including a review of medication routines and assistance with arranging follow-up appointments, had 30% fewer subsequent ED visits and hospital readmissions than patients who received usual care at discharge (31). Of note, this study found that provision and transmission of transition records has progressed across urban and rural facilities, with increases of about 15 percentage points annually, from 53% in 2017 to 83% in 2019. Yet, variations in transition record provision and timely transmission were still prominent in 2019, with the bottom quartile of facilities providing only 47% of their patients with detailed discharge information and transmitting only 31% of their patients’ records in time for a patient’s first postdischarge physician visit. These results raise concerns regarding the substantial deviations in adherence to evidence-based practices in some inpatient psychiatric units, despite nationwide improvements overall.
Rural facilities, especially more isolated facilities, outperformed urban facilities on coordinating postdischarge transition records for their patients. This finding might be due to the proximity of patient residence to the facility, the collaborative nature of small rural communities and organizations, or the data management requirements of the swing bed program in certain rural hospitals. First, compared with rural facilities, urban facilities might serve more nonlocal patients, including a higher proportion of nonlocal patients with serious mental illness, hindering the ability of urban units to provide adequate transition records or to connect rural patients with outpatient care (10). When discharging rural patients, urban inpatient psychiatric units often must collaborate with distant outpatient care settings, whereas rural units discharging patients to proximal rural settings can more readily identify and work directly with local providers for transition planning and record transmission (11). Second, rural communities are known for a strong spirit of collaboration—especially in towns where the population is smaller, residents know one another, and mental health care resources are limited—leading to closer networking and collaboration than in urban communities (32). These connections might facilitate the provision of detailed, personalized discharge records to local patients and families and timely transmission of transition records to other local outpatient providers. Third, since 2002, after passage of the Balanced Budget Act of 1997 (33), rural hospitals with ≤100 beds have been eligible to participate in the swing bed program—a reimbursement mechanism that covers posthospitalization extended care services. These swing beds are not used for postdischarge psychiatric patients. Still, small rural hospitals with swing beds must document detailed transition records, likely contributing to high performance in transition record coordination for discharged patients across all hospital settings.
Several limitations of this study are worth noting. First, our analysis was solely based on facility-level data and could not be adjusted for the complexity of cases. Important factors that may have affected receipt of follow-up care among patients include, for example, disease acuity, proximity to care, psychotropic medication use, and family functioning and social support. Future studies are warranted to examine patient-level quality of care, especially between rural residents who received and those who did not receive care in local psychiatric units. Second, the follow-up measure uses data only from Medicare fee-for-service claims, and results cannot be generalized to facility performance for patients with other payers, including Medicare Advantage. Third, our findings cannot be generalized to psychiatric units that are not eligible for IPFQR. Finally, using ZCTA-level sociodemographic characteristics and socioeconomic mix as proxies for potential patient mix in a facility might not reflect the actual composition of the patient population.
Conclusions
Since the CMS IPFQR program was implemented in 2014, quality of inpatient psychiatric care has improved, but follow-up care has not. Patients served at rural psychiatric units generally receive a higher quality of care, as indicated by better follow-up care, better timely transmission of transition records, and lower rates of physical restraint use in rural than in urban units. Understanding the reasons for these rural-urban differences in psychiatric care quality and the barriers that decrease follow-up care after discharge from urban and rural units is needed to improve mental health outcomes.
Supplementary Material
File (appi.ps.20220277.ds001.pdf)
- View/Download
- 239.59 KB
References
1.
Key Substance Use and Mental Health Indicators in the United States: Results From the 2020 National Survey on Drug Use and Health. Table 10.1A—Any Mental Illness in Past Year: Among People Aged 18 or Older; by Demographic Characteristics, Numbers in Thousands, 2008–2020. Pub no PEP21-07-01-003. Rockville, MD, Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality, 2020. https://www.samhsa.gov/data/sites/default/files/reports/rpt35323/NSDUHDetailedTabs2020/NSDUHDetailedTabs2020/NSDUHDetTabsSect10pe2020.htm. Accessed May 14, 2022
2.
Key Substance Use and Mental Health Indicators in the United States: Results From the 2020 National Survey on Drug Use and Health. Table 10.4B. Pub no PEP21-07-01-003. Rockville, MD, Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality, 2021. https://www.samhsa.gov/data/report/2020-nsduh-detailed-tables. Accessed Oct 7, 2022
3.
Key Substance Use and Mental Health Indicators in the United States: Results From the 2020 National Survey on Drug Use and Health. Table 10.17A. Pub no PEP21-07-01-003. Rockville, MD, Substance Abuse and Mental Health Services Administration, Center for Behavioral Health Statistics and Quality, 2021. https://www.samhsa.gov/data/report/2020-nsduh-detailed-tables. Accessed Oct 7, 2022
4.
Andrilla CHA, Patterson DG, Garberson LA, et al: Geographic variation in the supply of selected behavioral health providers. Am J Prev Med 2018; 54:S199–S207
5.
Designated Health Professional Shortage Areas Statistics: First Quarter of Fiscal Year 2022 Designated HPSA Quarterly Summary. Rockville, MD, US Department of Health and Human Services, Health Resources and Services Administration, Bureau of Health Workforce, 2022. https://data.hrsa.gov/Default/GenerateHPSAQuarterlyReport. Accessed Oct 5, 2022
6.
Blair R, Brown JD, Barry X, et al: Transitions in Care and Service Use Among Medicare Beneficiaries in Inpatient Psychiatric Facilities Issue Brief. Washington, DC, US Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation, 2019. https://aspe.hhs.gov/reports/transitions-care-service-use-among-medicare-beneficiaries-inpatient-psychiatric-facilities-issue-0. Accessed May 14, 2022
7.
Fisher WA: Elements of successful restraint and seclusion reduction programs and their application in a large, urban, state psychiatric hospital. J Psychiatr Pract 2003; 9:7–15
8.
Delaney KR, Vanderhoef D: The psychiatric mental health advanced practice registered nurse workforce: charting the future. J Am Psychiatr Nurses Assoc 2019; 25:11–18
9.
Fontanella CA, Warner LA, Steelesmith DL, et al: Association of timely outpatient mental health services for youths after psychiatric hospitalization with risk of death by suicide. JAMA Netw Open 2020; 3:e2012887
10.
Stewart EG: Mental Health in Rural America: A Field Guide. New York, Routledge, 2018
11.
Avey JP: Discharge Planning From Urban Psychiatric Facilities to Rural Communities Using Telehealth. Order no 3619191. Anchorage, University of Alaska Anchorage, 2014. https://www.proquest.com/dissertations-theses/discharge-planning-urban-psychiatric-facilities/docview/1536437152/se-2
12.
Shields MC: Patient characteristics associated with admission to low-safety inpatient psychiatric facilities: evidence for racial inequities. Psychiatr Serv 2021; 72:1151–1159
13.
Fiscal Year (FY) 2022 Inpatient Psychiatric Facility (IPF) Prospective Payment System (PPS) Final Rule. Pub no CMS-1750-F. Baltimore, Centers for Medicare and Medicaid Servies, 2021. https://www.cms.gov/newsroom/fact-sheets/fiscal-year-fy-2022-inpatient-psychiatric-facility-ipf-prospective-payment-system-pps-final-rule-cms. Accessed May 14, 2022
14.
Shields MC, Busch AB: The effect of Centers for Medicare and Medicaid’s Inpatient Psychiatric Facility Quality Reporting program on the use of restraint and seclusion. Med Care 2020; 58:889–894
15.
Carrillo S, Nazir N, Howser E, et al: Impact of the 2015 CMS Inpatient Psychiatric Facility Quality Reporting rule on tobacco treatment. Nicotine Tob Res 2017; 19:976–982
16.
Benjenk I, Shields M, Chen J: Measures of care coordination at inpatient psychiatric facilities and the Medicare 30-day all-cause readmission rate. Psychiatr Serv 2020; 71:1031–1038
17.
IPFQR Archived Resources—Claims-Based Measures. Baltimore, Centers for Medicare and Medicaid Services, Inpatient Psychiatric Facility Quality Reporting Program, 2022. https://qualitynet.cms.gov/ipf/ipfqr/resources#tab5. Accessed May 14, 2022
18.
Blair R, Liu J, Rosenau M, et al: Development of Quality Measures for Inpatient Psychiatric Facilities: Final Report. Washington, DC, US Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation, 2015. https://aspe.hhs.gov/reports/development-quality-measures-inpatient-psychiatric-facilities-final-report-0. Accessed May 14, 2022
19.
2019 AHA Annual Survey—American Hospital Association. Chicago, American Hospital Association, 2020. https://www.ahadata.com/system/files/media/file/2020/10/2019AHAAnnual.pdf
20.
American Community Survey 5-Year Data (2009–2020). Suitland, MD, United States Census Bureau, 2021. https://www.census.gov/data/developers/data-sets/acs-5year.html. Accessed May 14, 2022
21.
Shields MC: Effects of the CMS’ public reporting program for inpatient psychiatric facilities on targeted and nontargeted safety: differences between for-profits and nonprofits. Med Care Res Rev 2022; 79:233–243
22.
Ivey-Stephenson AZ, Crosby AE, Jack SPD, et al: Suicide trends among and within urbanization levels by sex, race/ethnicity, age group, and mechanism of death—United States, 2001–2015. MMWR Surveill Summ 2019; 66:1–16
23.
Heslin KC, Weiss AJ: Hospital Readmissions Involving Psychiatric Disorders, 2012. Statistical brief 189. Rockville, MD, Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project Statistical Briefs, 2015. https://www.hcup-us.ahrq.gov/reports/statbriefs/sb189-Hospital-Readmissions-Psychiatric-Disorders-2012.jsp. Accessed Oct 7, 2022
24.
Schroeder S, Leigh-Peterson M: Rural and Urban Utilization of the Emergency Department for Mental Health and Substance Abuse. Grand Forks, University of North Dakota Center for Rural Health, NORC Walsh Center for Rural Health Analysis, 2017. https://ruralhealth.und.edu/assets/355-922/rural-urban-utilization-emergency-department.pdf
25.
Moore C, McGinn T, Halm E: Tying up loose ends: discharging patients with unresolved medical issues. Arch Intern Med 2007; 167:1305–1311
26.
Richardson JS, Mark TL, McKeon R: The return on investment of post discharge follow-up calls for suicidal ideation or deliberate self-harm. Psychiatr Serv 2014; 65:1012–1019
27.
Chung D, Hadzi-Pavlovic D, Wang M, et al: Meta-analysis of suicide rates in the first week and the first month after psychiatric hospitalisation. BMJ Open 2019; 9:e023883
28.
Trends in the Utilization of Emergency Department Services, 2009–2018. Washington, DC, US Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation, 2021. https://aspe.hhs.gov/reports/trends-utilization-emergency-department-services-2009-2018. Accessed May 14, 2022
29.
Vashi AA, Fox JP, Carr BG, et al: Use of hospital-based acute care among patients recently discharged from the hospital. JAMA 2013; 309:364–371
30.
Smith TE, Haselden M, Corbeil T, et al: Relationship between continuity of care and discharge planning after hospital psychiatric admission. Psychiatr Serv 2020; 71:75–78
31.
Horwitz LI, Moriarty JP, Chen C, et al: Quality of discharge practices and patient understanding at an academic medical center. JAMA Intern Med 2013; 173:1715–1722
32.
Rural Action Plan. Washington, DC, US Department of Health and Human Services, 2020. https://www.hhs.gov/sites/default/files/hhs-rural-action-plan.pdf
33.
Swing Bed Providers. Baltimore, Centers for Medicare and Medicaid Services, 2022. https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/SNFPPS/SwingBed. Accessed May 14, 2022
Information & Authors
Information
Published In
History
Received: 25 May 2022
Revision received: 19 July 2022
Accepted: 23 August 2022
Published online: 2 November 2022
Published in print: May 01, 2023
Keywords
Authors
Competing Interests
Dr. Brown is the cofounder of Brown Research Consulting Inc. The other authors report no financial relationships with commercial interests.
Funding Information
This study was supported by the Health Resources and Services Administration (HRSA) of the U.S. Department of Health and Human Services (DHHS) (grant number U1C45498, Rural Health Research Grant Program Cooperative Agreement).The information or content and conclusions in this article are those of the authors and should not be construed as the official position or policy of, nor should any endorsements be inferred by, HRSA, DHHS, or the U.S. government.
Metrics & Citations
Metrics
Citations
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
For more information or tips please see 'Downloading to a citation manager' in the Help menu.
View Options
View options
PDF/EPUB
View PDF/EPUBLogin options
Already a subscriber? Access your subscription through your login credentials or your institution for full access to this article.
Personal login Institutional Login Open Athens loginNot a subscriber?
PsychiatryOnline subscription options offer access to the DSM-5-TR® library, books, journals, CME, and patient resources. This all-in-one virtual library provides psychiatrists and mental health professionals with key resources for diagnosis, treatment, research, and professional development.
Need more help? PsychiatryOnline Customer Service may be reached by emailing [email protected] or by calling 800-368-5777 (in the U.S.) or 703-907-7322 (outside the U.S.).