Scholarly article on topic 'Hospitalization and Mortality Rates in Long-Term Care Facilities: Does For-Profit Status Matter?'

Hospitalization and Mortality Rates in Long-Term Care Facilities: Does For-Profit Status Matter? Academic research paper on "Health sciences"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{"Nursing homes" / "long-term care" / "nonprofit organizations" / proprietary / mortality / hospitalization / "quality indicators"}

Abstract of research paper on Health sciences, author of scientific article — Peter Tanuseputro, Mathieu Chalifoux, Carol Bennett, Andrea Gruneir, Susan E. Bronskill, et al.

Abstract Objectives To establish if proprietary status (ie, for-profit or not-for-profit) is associated with mortality and hospitalizations among publicly funded long-term care (nursing) homes. Methods We conducted a retrospective cohort study of new admissions in 640 publicly funded long-term care facilities in Ontario, Canada (384 for-profit, 256 not-for-profit). A population-based cohort of 53,739 incident admissions into long-term care facilities between January 1, 2010, and March 1, 2012, was observed. We measured adjusted rates of hospital admissions and mortality, per 1000 person-years (PY) of follow-up, among for-profit and not-for-profit facilities at 3, 6, and 12 months postadmission. Rates were measured postadmission and until discharge or death, whichever came first. Results One year after admission and before discharge, 11.7% of residents died and 25.7% had at least one hospitalization. After 12 months of follow-up, residents in for-profit facilities had a hospitalization rate of 462 per 1000 PY versus 358 per 1000 PY in not-for-profit facilities. During this period, the crude mortality rate in for-profit facilities was 208 per 1000 PY versus 185 per 1000 PY in not-for-profit facilities. At 3, 6, and 1 year after admission, for-profit facilities had an adjusted hazard ratio of 1.36 (95% confidence interval [CI] 1.28–1.43), 1.33 (95% CI 1.27–1.39), and 1.25 (95% CI 1.21–1.30) for hospitalizations and hazards of 1.20 (95% CI 1.11–1.29), 1.16 (95% CI 1.09–1.24), and 1.10 (95% CI 1.05–1.16) for mortality, respectively. Conclusions Publicly funded for-profit facilities have significantly higher rates of both mortality and hospital admissions.

Academic research paper on topic "Hospitalization and Mortality Rates in Long-Term Care Facilities: Does For-Profit Status Matter?"

JAMDA

journal homepage: www.jamda.com

Original Study

Hospitalization and Mortality Rates in Long-Term Care Facilities: Does For-Profit Status Matter?

Peter Tanuseputro MD, MHSCa b *, Mathieu Chalifoux MScb, Carol Bennett MSc Andrea Gruneir PhDc, Susan E. Bronskill PhDc, Peter Walker MD; Douglas Manuel MD, MSc

a Bruyère Research Institute, Bruyère Centre of Learning, Research and Innovation in Long-Term Care, Ottawa Hospital, Ottawa, Ontario, Canada b Ottawa Hospital Research Institute, Clinical Epidemiology Program, Ottawa, Ontario, Canada c Institute for Clinical Evaluative Sciences, Toronto, Ontario, Canada

CrossMark

ABSTRACT

Keywords: Nursing homes long-term care nonprofit organizations proprietary mortality hospitalization quality indicators

Objectives: To establish if proprietary status (ie, for-profit or not-for-profit) is associated with mortality and hospitalizations among publicly funded long-term care (nursing) homes.

Methods: We conducted a retrospective cohort study of new admissions in 640 publicly funded long-term care facilities in Ontario, Canada (384 for-profit, 256 not-for-profit). A population-based cohort of 53,739 incident admissions into long-term care facilities between January 1, 2010, and March 1, 2012, was observed. We measured adjusted rates of hospital admissions and mortality, per 1000 person-years (PY) of follow-up, among for-profit and not-for-profit facilities at 3, 6, and 12 months postadmission. Rates were measured postadmission and until discharge or death, whichever came first. Results: One year after admission and before discharge, 11.7% of residents died and 25.7% had at least one hospitalization. After 12 months of follow-up, residents in for-profit facilities had a hospitalization rate of 462 per 1000 PY versus 358 per 1000 PY in not-for-profit facilities. During this period, the crude mortality rate in for-profit facilities was 208 per 1000 PY versus 185 per 1000 PY in not-for-profit facilities. At 3, 6, and 1 year after admission, for-profit facilities had an adjusted hazard ratio of 1.36 (95% confidence interval [CI] 1.28-1.43), 1.33 (95% CI 1.27-1.39), and 1.25 (95% CI 1.21-1.30) for hospitalizations and hazards of 1.20 (95% CI 1.11-1.29), 1.16 (95% CI 1.09-1.24), and 1.10 (95% CI 1.05-1.16) for mortality, respectively.

Conclusions: Publicly funded for-profit facilities have significantly higher rates of both mortality and hospital admissions.

© 2015 AMDA - The Society for Post-Acute and Long-Term Care Medicine. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

The authors declare no conflicts of interest.

This study was funded by a research grant from the Ontario Ministry of Health and Long Term Care (MOHLTC) to the Bruyère Centre of Learning Research and Innovation in Long Term Care (CLRI). This study was also supported by the Institute for Clinical Evaluative Sciences, which is funded by an annual grant from the Ontario MOHLTC. PT is supported by the Bruyère CLRI and the Population Health Intervention Research Network (TGF 96112). The opinions, results and conclusions reported in this paper are those of the authors and are independent from the funding sources. No endorsement by MOHLTC is intended or should be inferred. The funders had no influence in study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.

We have gone through and met the STROBE checklist for observational studies in our manuscript. More details of this process can be requested from the corresponding author.

* Address correspondence to Peter Tanuseputro, MD, MHSC, Bruyère Research Institute, Bruyère Centre of Learning, Research and Innovation in Long-Term Care, Ottawa Hospital - Civic Campus, 1053 Carling Ave, Box 693, 2-006 Admin Services Building, Ottawa ON K1Y 4E9, Canada.

E-mail address: peter.tanuseputro@mail.utoronto.ca (P. Tanuseputro).

Long-term care (LTC) facilities, also commonly known as nursing homes, deliver care to individuals unable to live in the community due to illness and/or disability. In addition to housing, LTC facilities provide personal and medical support, including 24-hour nursing care. The need for LTC facilities increases with age, and is expected to increase in aging populations.1,2 Discussions on quality of care in LTC facilities are widespread,3-6 and monitoring of performance indicators is becoming increasingly common.7-12

Internationally, there is significant heterogeneity in how LTC facilities are owned and operated; for example, more than half of facilities in Canada, the United States, and the United Kingdom are managed by for-profit institutions,13-15 and not-for-profit facilities can be managed by private (eg, religious or lay) or public (eg, municipal, provincial, or federal) corporations. Homes across jurisdictions have varying mixes of public/private funding, depending on factors such as legislation and level of government involvement in

http://dx.doi.org/10.1016/jjamda.2015.06.004

1525-8610/© 2015 AMDA - The Society for Post-Acute and Long-Term Care Medicine. This is an open access article under the CC BY-NC-ND license (http://creativecommons. org/licenses/by-nc-nd/4.0/).

care delivery. The effect of LTC facilities' proprietary status on quality of care has been studied, and one review in 2009 suggested that not-for-profit nursing homes provide superior care.15 There were, however, considerable inconsistencies across the studies and among outcomes. Only 40 of 82 studies reviewed showed statistically significant findings in favor of not-for-profit facilities for all outcomes, a few showed findings in favor of for-profit facilities, and many showed non-statistically significant findings.15 Several reasons have been proposed for the inconsistencies across studies, including underlying differences in the characteristics of residents that for-profit and not-for-profit homes attract and accept, differences in public and private funding levels both within and across profit status, and inadequate risk-adjustment.5 In addition, most studies rely on publicly reported quality indicators (eg, prevalence of pressure ulcers or use of physical restraints) that are often tied to remuneration, introducing potential reporting biases.16,17 It has also been suggested that some quality-of-care indicators may reflect clinical outcomes that are too insignificant to affect important health care events, such as hospitalizations.18,19

We examined mortality and hospitalizations in a large, population-based cohort of newly admitted residents in Ontario, Canada. We did so in an environment in which funding mechanisms and resident-placement schemes are structured uniformly across all facilities. First, all facilities receive standardized base public funding, with co-payments (within this funding) paid by the resident based on financial means. Funding is set relative to the needs of residents, measured using internationally validated interRAI Long-Term Care Facilities Assessment System tools.20 Unlike other jurisdictions, such as the United States, England, and other European countries, all Ontario homes are legislated against charging patients additional funds for core services. This largely addresses concerns of differential funding/payer levels and mixes observed in other jurisdictions.15 Despite the uniformity of public funding, for-profit facilities in Ontario are able to draw funds deemed as profit for the financial benefit of owners and/or shareholders. Not-for-profit facilities can also make profit, but such funds must be reserved and used solely for the purposes of facility matters.

Second, the resident-to-facility matching process in Ontario has features that remove biases in patient and facility preferences. Ontario's referral system is organized centrally; potential residents choose a number of facilities (currently 5), and are placed on the wait list of those facilities. Where applicants fall on each of the facilities' respective wait lists depends primarily on the relative needs of other applicants on the list. When a facility has an open bed, the first person on the list is offered the bed. Profit status is typically not a consideration when patients choose their list of facilities. Where patients fall on each list, and how quickly a list moves is out of the control the applicant. Significant disincentive is placed on applicants to avoid rejecting a matched home; these clients would automatically be put on the bottom of the wait list, and can be financially penalized if waiting from an acute care bed. Conversely, a recent report shows that facilities in Ontario rarely reject a matched resident.21 This is in contrast to most other jurisdictions, such as in the United States and England, where homes have greater control on who to admit and how many (if any) government-funded residents they accept. Finally, our study considers and adjusts for a large set of both individual- and facility-level covariates, largely not done in previous studies.15 We aimed to conclusively determine whether proprietary status (forprofit or not-for-profit) is associated with rates of hospitalizations and mortality in the LTC population.

Methods

We carried out a population-based retrospective cohort study to examine the differences between rates of hospitalization and

mortality in for-profit and not-for-profit LTC homes. We captured all incident admissions to LTC facilities between January 1, 2010, and March 31, 2012, in Ontario, Canada. To accomplish this, we applied several exclusions, including removing an admission if it was observed that the resident was transferred from another facility or had a previous admission in a LTC facility (Appendix Figure 1). Using encrypted health card numbers as unique but encoded identifiers, records of health care use were linked across various administrative databases. No written consent was obtained; all data were encrypted using health card numbers as unique identifiers. Thus all records used were de-identified and anonymized. All data were housed and analyzed at the Institute for Clinical Evaluative Sciences (ICES), a prescribed entity for the purposes of section 45 Ontario's Personal Health Information Privacy Act. Ethics approval was obtained from ICES at the Sunnybrook Health Sciences Centre Research Ethics Board in Toronto, Ontario, and from the Ottawa Hospital Research Institute Ethics Board in Ottawa, Ontario (Protocol 20130579-01H).

Data Sources and Definitions

Incident admissions to LTC facilities were identified using the Canadian Continuing Care Reporting System (CCRS). The CCRS collects information on all residents and facilities using the validated Resident Assessment Instrument Minimum Data Set (RAI-MDS).20,22 Assessments are done at entry, quarterly, and on any significant health status changes. We used the CCRS to capture information on patient demographics, clinical status, and functional status.22 We categorized facilities by for-profit and not-for-profit (religious, lay, or government) ownership, and by facility size. All homes fall under the same provincial LTC legislation; for-profit facilities typically distribute a portion of net revenues to investors, whereas not-forprofits do not. Information from the CCRS is used to support system planning and clinical practice, and directly affects the level of funding provided by the government, based on level of need for services for each resident.

Data on hospitalizations and mortality were obtained using linkage of individuals in the CCRS to the Discharge Abstract Database (DAD) and the Registered Persons Database (RPDB), respectively. The RPDB also was used to obtain patient age, sex, and postal code. Following well-established methods, both neighborhood income of client's last residence and rurality were captured by linking to Statistics Canada census data using postal codes.23

Outcomes

Outcomes were calculated at 3, 6, and 12 months after admission using rates per 1000 person-years (PY) of follow-up in LTC. Patients were followed from admission to any discharge, death, or for 365 days, whichever came first. We examined time to first hospital-ization and time to death for each individual, during the follow-up period, while residents were in their incident LTC facility. Hospitali-zations that occur before discharge from the LTC facility are captured by observing a record in the DAD with an admission date that overlapped the length of stay in the LTC facility.

Some hospitalizations and all deaths led to discharge from a LTC facility. The date of such events are captured in the DAD admission date or RPDB death date, which should be the same as the CCRS discharge date (end of follow-up period for each resident); to account for administrative discrepancies in these dates, hospitalizations (in DAD) and deaths (in RPDB) captured 3 days post-CCRS discharge date were included if it was noted in the CCRS that these discharges occurred to hospital or death, respectively.

Table 1

Number of Facilities by Facility Ownership and Facility Type

No. (%) of For-Profit Facilities No. (%) of Not-For-Profit Facilities All

Religious Lay Government Total Not-For-Profit

All LTC facilities, % total facilities 384 (60.0) 25 (3.9) 95 (14.8) 136 (21.3) 256 (40.0) 640 (100)

No. of beds

1-19 0(0) 1 (4) 0 (0) 12 (8.8) 13(5.1) 13 (2.03)

20-49 29 (7.6) 1 (4) 10(10.5) 15 (11.0) 26(10.2) 55 (8.6)

50-99 150 (39.1) 8 (32) 32 (33.7) 18 (13.2) 58 (22.7) 208 (32.5)

100-149 97 (25.3) 6 (24) 27 (28.4) 28 (20.6) 61 (23.8) 158 (24.7)

>150 108 (28.1) 9 (36) 26 (27.4) 63 (46.3) 98 (38.3) 206 (32.2)

Facility neighborhood quintile

1 (lowest) 85 (22.1) 3(12) 21 (22.1) 43 (31.6) 67 (26.2) 152 (23.8)

2 68 (17.7) 8 (32) 21 (22.1) 14(10.3) 43 (16.8) 111 (17.3)

3 86 (22.4) 6 (24) 14(14.7) 30 (22.1) 50(19.5) 136 (21.3)

4 76(19.8) 3(12) 18 (18.9) 29 (21.3) 50(19.5) 126(19.7)

5 (highest) 67 (17.5) 5 (20) 21 (22.1) 20 (14.7) 46 (18.0) 113 (17.7)

Missing 2 (0.5) 0 (0) 0 (0) 0(0) 0(0) 2 (0.3)

Facility urbanicity

Urban 299 (77.9) 23 (92) 76 (80) 89 (65.4) 188 (73.4) 487 (76.1)

Rural 85 (22.1) 2 (8) 19(20) 47 (34.6) 68 (26.6) 153 (23.9)

Statistical Analysis

Facility-level rates of hospitalization and mortality per 1000 PY were calculated. Facilities were grouped into 2 sets of quintiles, according to each facility's overall rate of hospitalization and mortality after 12 months of follow-up for all resident admissions. The facilities were then divided among for-profit and not-for-profit status. Hospitalization and mortality rates for all residents in all for-profit and not-for-profit facilities were also calculated at 3, 6, and 12 months.

Multivariable models

Using individual-level data, we created separate multivariable models for time to hospitalization and time to mortality, examining the effect of for-profit status while adjusting for patient demographics (age, sex, marital status, neighborhood income before entry), facility variables (facility urbanicity and facility size), and for patient illness and acuity (location before admission, and Changes in Health, End-stage disease, and Signs and Symptoms [CHESS] score).24 CHESS was chosen as a prevalidated, composite measure of illness and disability; it includes components of cognitive and physical disability (eg, change in decision-making and activities of daily living), and various signs and symptoms of illness and decline (eg, dyspnea, vomiting, decreased fluid intake, and weight loss). We examined differences in residents between profit status for various other potential confounding variables; we chose not to include these variables in our model (in addition to CHESS), for reasons of parsimony and colinearity between variables.

Cox-proportional hazards were used to model time to death. Competing risk models were used to explore the impact of covariates on hospitalizations. Competing risk models weight each individual and, if they experience a competing event (ie, mortality) that prevents the event of interest (ie, hospitalizations), the model calculates the probability of experiencing the event of interest had the competing event never occurred. The P values and 95% confidence intervals (CIs) of each hazard ratio were calculated. We used SAS 9.3 (SAS Institute, Inc, Cary, NC) for all analyses.

Sensitivity analyses

As part of our sensitivity analyses, to examine the consistency of our findings, we compared rates of hospitalizations and mortality in for-profit and not-for-profit homes across a large number of subgroups. In addition to our main outcomes, we examined rates of publicly reported quality-of-care indicators on admission and at 3 months after admission: use of physical restraints, pressure ulcers,

incontinence, and falls.12 We also examined a subset of hospitaliza-tions that have been previously determined to be care-sensitive and potentially preventable (anemia, dehydration, urinary tract infection, decubitus ulcers, and gangrene).25,26 Finally, to ensure that our selection criteria did not bias our results, we examined the effect of removing those classified as short stays (expected to stay fewer than 90 days) and, separately, the effect of including all admissions (ie, including nonincident admissions).

Results

We examined LTC admissions in 640 facilities, 384 (60.0%) of which were for-profit and 256 (40.0%) were not-for-profit (Table 1). More than half of not-for-profit facilities were governmentally (municipal, provincial, or federal) owned, with the rest being operated by lay private or religious organizations. About one-third of all facilities had 150 beds or more (28.1% among for-profits versus 38.3% among for-profits), and 89% (92.5% among for-profits versus 84.8% among not-for-profits) had more than 50 beds.

There were 53,739 new, distinct residents admitted (Table 2). There was a predominance of women (64.9%), and those older than 80 years (68.3%). Most were admitted from an inpatient acute care facility (38.6%) or from home (34.5%). Residents tended to live in poorer neighborhoods before entry. A higher proportion of residents in forprofit facilities were admitted from inpatient acute care (41.6% versus 33.5%), whereas a higher proportion of residents in not-for-profit facilities were admitted from home (38.7% versus 33.0%). There were otherwise minimal differences in residents by proprietary status.

Table 3 compares the clinical characteristics of residents by facility proprietary status. Overall, residents had significant limitations in cognition and in performing activities of daily living, regardless of facility ownership. Most residents fell into the clinically complex (25.8%) or reduced physical function (35.0%) resource utility groups (RUGs). RUGs are used in Ontario to determine the level of funding for each individual. Most residents had low instability, as measured by the CHESS score.24 More than half of residents (55.7%) were recorded as having dementia on admission. There were very small differences in the clinical characteristics of residents across proprietary status.

Rates of Mortality and Hospitalization

In the year of follow-up after admission, 11.7% of residents died and 25.7% had at least one hospitalization before discharge. It should

Characteristics of Residents in Facilities of For-Profit and Not-For-Profit Facilities, Obtained from the CCRS

Table 3

Number of Residents in For-Profit and Not-For-Profit Homes, by Clinical Variables, Obtained from the CCRS

Characteristics

For-Profit

Not-For-Profit Total

Characteristics

For-Profit

Not-For-Profit Total

Activities of daily living limitations

Female 21,886 (64.6) 12,993 65.5) 34,879 64.9) Independent 1992 5.9) 1573 7.9) 3565 6.6)

Male 11,991 ( 35.4) 6817 34.4) 18,808 35.0) Supervision 2955 8.7) 2003 10.1) 4958 9.2)

Unspecified 28 ( 0.1) 24 0.1) 52 0.1) Limited 6439 19.0) 3910 19.7) 10,349 19.3)

Age at entry Extensive 1 8833 26.1) 5196 26.2) 14,029 26.1)

18-49 414 1.2) 251 1.3) 665 1.2) Extensive 2 5241 15.5) 2619 13.2) 7860 14.6)

50-59 1011 3.0) 499 2.5) 1510 2.8) Dependent 6696 19.7) 3647 18.4) 10,343 19.3)

60-69 2520 7.4) 1260 6.4) 3780 7.0) Total Dependence 1749 5.2) 886 4.5) 2635 4.9)

70-79 7176 (21.2) 3895 19.6) 11,071 (20.6) Cognitive Performance Scale

80-89 15,901 46.9) 9641 48.6) 25,542 47.5) 0: Intact 6618 19.5) 4256 21.5) 10,874 20.2)

90+ 6883 20.3) 4288 21.6) 11,171 20.8) 1: Boderline Intact 4670 13.8) 2706 13.6) 7376 13.7)

Marital status 2: Mild impairment 7737 22.8) 4065 20.5) 11,802 22.0)

Never married 2586 7.6) 1456 7.3) 4042 7.5) 3: Moderate Impairment 9646 28.5) 5657 28.5) 15,303 28.5)

Married 10,187 30.1) 6315 31.8) 16,502 30.7) 4: Moderate Severe 2030 6.0) 1089 5.5) 3119 5.8)

Widowed 17,529 (51.7) 10,282 51.8) 27,811 51.8) Impairment

Separated 788 2.3) 374 1.9) 1162 ( 2.2) 5: Severe Impairment 2066 6.1) 1470 7.4) 3536 6.6)

Divorced 2197 6.5) 1021 5.2) 3218 6.0) 6: Very Severe Impairment 1138 3.4) 591 3.0) 1729 3.2)

Unknown 618 1.8) 386 2.0) 1004 (1.9) Resource Utility Group: III ( 44-group)

Lived alone before entry Behavioral problems 954 2.8) 529 2.7) 1483 2.8)

No 25,702 75.8) 15,030 75.8) 40,732 75.8) Clinically complex 8545 25.2) 5302 26.7) 13,847 25.8)

Yes 7081 20.9) 4214 21.3) 11,295 21.0) Impaired cognition 4028 11.9) 2873 14.5) 6901 12.8)

Unknown 1122 3.3) 590 3.0) 1712 3.2) Reduced physical function 12,186 35.9) 6597 33.3) 18,783 35.0)

Language spoken at home Special rehabilitation 4259 12.6) 2516 12.7) 6775 12.6)

Non-English 5806(17.1) 3776 19.0) 9582 17.8) Extensive care 912 2.7) 439 2.2) 1351 2.5)

English 28,099 82.9) 16,058 81.0) 44,157 (82.2) Special care 3021 8.9) 1578 8.0) 4599 8.6)

Last residence income quintile CHESS score

1 lowest) 7926 23.4) 4366 22.0) 12,292 22.9) 0 ( no instability) 16,738 49.4) 9861 49.7) 26,599 49.5)

2 6353 18.7) 3481 17.6) 9834 18.3) 1 10,542 31.1) 6109 30.8) 16,651 31.0)

3 6169 18.2) 3678 18.5) 9847 18.3) 2 4652 13.7) 2659 13.4) 7311 13.6)

4 5366 15.8) 3172 16.0) 8538 15.9) 3 1447 4.3) 850 4.3) 2297 4.3)

5 highest) 4925 (14.5) 3029 15.3) 7954 (14.8) 4 463 1.4) 306 1.5) 769 1.4)

Unknown 3166 ( 9.3) 2108 10.6) 5274 9.8) 5 ( highest level of instability) 63 0.2) 49 0.3) 112 0.2)

Previous stay in board and care, assisted living, or group home in past 5 years Chronic disease prevalence:

No 26,325 77.6) 14,919 75.2) 41,244(76.7) Alzheimer 4698 13.9) 3202 16.1) 7900 14.7)

Yes 6055 (17.9) 3703 18.7) 9758 18.2) Dementia other than 14,207 41.9) 7808 39.4) 22,015 41.0)

Unknown 1525 4.5) 1212 6.1) 2737 5.1) Alzheimer

Admitted From Cancer 3618 10.7) 2073 10.5) 5691 10.6)

Ambulatory health service 636 1.9) 163 0.8) 799 1.5) Stroke 7227 21.3) 3564 18.0) 10,791 20.1)

Inpatient acute care service 14,113 41.6) 6636 33.5) 20,749 38.6) Congestive heart failure 4911 14.5) 2525 12.7) 7436 16.9)

Inpatient rehabilitation 1177 3.4) 607 3.1) 1784 (3.3) Emphysema/COPD 5665 16.7) 2881 14.5) 8546 15.9)

service Depression 7504 22.1) 4321 21.8) 11,825 22.0)

Inpatient continuing 2035 6.0) 1473 7.4) 3508 6.5) Diabetes 9442 27.9) 4858 24.5) 14,300 26.6)

care service Arthritis 12,912 38.1) 7434 37.5) 20,346 37.9)

Inpatient psychiatry service 418 1.2) 242 1.2) 660 1.2) Number of Medications:

Home care service 2854 8.4) 2305 11.6) 5159 9.1) 0 242 0.7) 104 0.52) 346 0.6)

Residential care service 4262 12.6) 2867 14.5) 7129 13.3) 1-5 4577 13.5) 2645 13.3) 7222 13.4)

board and care) 6-10 13,923 41.1) 8031 40.5) 21,954 40.9)

Private home 8324 24.6) 5369 (27.1) 13,693 25.5) 11-15 11,113 32.8) 6517 32.9) 17,630 32.8)

no home care) 16-20 3366 9.9) 2097 10.6) 5463 10.2)

21+ 684 2.0) 440 2.2) 1124 (2.1)

be noted that residents who were discharged before the full year of follow-up ( eg, back home, or to another facility) were no longer followed for hospitalization or mortality and did not contribute to the follow-up ( PY) denominator. Overall, not accounting for such censoring, 24.7% of residents 1 year postadmission died and 31.8% were hospitalized ( data not shown). Not including death, 46.5% of the admission cohort was discharged before the full year of follow-up.

The rates of mortality per 1000 PYof follow-up were 291.0, 238.1, and 198.7 at 3, 6, and 12 months, respectively (Table 4). For hospitalizations, the rates were 608.3, 508.8, and 422.1 per 1000 PY, respectively. Compared with not-for-profit facilities, for-profit facilities had 18.4%, 16.7%, and 12.3% greater mortality rates per 1000 PY and 38.2%, 35.9%, and 29.2% greater hospitalization rates at 3, 6, and 12 months, respectively ( P < .01). For the 12-month follow-up period, for-profit facilities had 14.3% and 12.0% of its facilities in the lowest quintiles for mortality and hospitalizations ( as opposed to the expected 20%), whereas not-for-profit facilities had 28.4% and 32.4% in the same quintiles, respectively. Conversely, for-profit homes had

COPD, chronic obstructive pulmonary disease.

23.7% and 25.5% of its homes in the highest quintiles for mortality and hospitalizations ( at 12 months), whereas not-for-profit homes had 14.2% and 11.3% in the same quintiles, respectively.

Multivariable Models for Mortality and Hospitalizations

At 12 months after admission, for-profit homes had an adjusted hazard ratio of 1.10 ( 95% CI 1.05-1.16) for mortality ( Table 5) and 1.25 ( 95% CI 1.21-1.30) for hospitalization relative to not-for-profit homes Table 6). These models were adjusted for individual demographic variables ( age, sex, marital status, and resident neighborhood income quintile), facility variables urbanicity and size), and measures of illness and acuity ( location before admission and CHESS score). The adjusted hazards for hospitalizations and mortality generally trended higher when observing periods closer to admission models at 3 and 6 months) ( Tables 5 and 6). Increasing facility size was protective for mortality but was a risk for hospitalization.

Crude Mortality and Hospitalization Rates per PY in the 3, 6, and 12 Months After Admission to For-Profit and Not-For-Profit LTC Facilities

Characteristics

No. of Facilities (%)

For-Profit

Not-For-Profit

All Facilities

Mortality rate per 1000 PY

Q1*: 0-115.0 54(14.1) 73 (28.7) 127 (19.9)

Q2 115.0-168.9 81 (21.1) 47 (18.5) 128 (20.1)

Q3 168.9-216.0 78 (20.3) 50(19.7) 128 (20.1)

Q4 216.0-283.4 81 (21.1) 47 (18.5) 128 (20.1)

Q5 283.4-785.9 90 (23.4) 37 (14.6) 127 (19.9)

Overall rate, 12 mo 207.5 184.7y 198.7

Overall rate, 6 mo 251.8 215.7y 238.1

Overall rate, 3 mo 308.9 261.0y 291.0

Hospitalization rate, per 1000 PY

Q1*: 0-269.3 51 (13.3) 76 (29.9) 127 (19.9)

Q2 269.3-344.5 65 (16.9) 63 (24.8) 128 (20.1)

Q3 344.5-426.0 77 (20.1) 51 (20.1) 128 (20.1)

Q4 426.0-537.0 85 (22.1) 43 (16.9) 128 (20.1)

Q5 537.0-1229.8 106 (27.6) 21 (8.3) 127 (19.9)

Overall rate, 12 mo 462.4 358.0y 422.1

Overall rate, 6 mo 565.4 416.1 y 508.8

Overall rate, 3 mo 678.4 490.9y 608.3

*All facilities were sorted into quintiles by their 12-month mortality or hospitalization rate, where Q1 identifies the 20% of all facilities with the lowest rates of outcome and Q5 the facilities with the highest outcome rates. yP < .01 (compared with rates in for-profit facilities).

Sensitivity Analyses

Hospitalization and mortality rates were lower in not-for-profit facilities in 48 and 43 of the 48 subgroups examined, respectively (Appendix Tables 1 and 2). For publicly reported quality of care indicators, the prevalence of worsening/any incontinence, ulcers, and falls were similar between for-profit and not-for-profit facilities. The prevalence of physical restraints was lower in for-profit facilities on admission (4.7% versus 6.4%) and at 3 months (7.0% versus 8.9%) (Appendix Table 3). When looking at only care-sensitive conditions, the overall 12-month hospitalization rates were 87.7 per 1000 PY in for-profit facilities, and 78.4 per 1000 PY in not-for-profit facilities (data not shown).

Excluding all short stays, defined as stays of fewer than 90 days that led to discharge to the community, led to a very similar 12-month hazard ratio of 1.11 for mortality, and 1.20 for hospitalization (P < .01) (data not shown). Separately, including nonincident admissions, led to a 12-month hazard of 1.15 for mortality and 1.22 for hospitalization (P < .01) (data not shown).

Table 5

Hazard Ratios for Mortality (Cox-Proportional Hazard) Model at 3, 6, and 12 Months After Admission to For-Profit and Not-For-Profit LTC Facilities

Parameter Variable Hazard Ratio (95% Confidence Interval)

Ownership

Not-for-profit 1.00 1.00 1.00

For-profit 1.10(1.05- -1.16)y 1.16(1.09- -1.24)y 1.20 (1.11-1.29)y

Age, y

18-49 1.00 1.00 1.00

50-59 1.49 (0.83- 2.69) 1.96 (0.92- 4.20) 1.81 (0.75-4.37)

60-69 2.13 (1.24-3.67)y 2.63 (1.29-5.37)y 2.50 (1.10-5.70)*

70-79 2.85 (1.68- 4.86)y 3.48 (1.73- -7.02)y 3.16 (1.41-7.11^

80-89 4.04 (2.38- 6.86)y 4.94 (2.45- -9.93)y 4.43 (1.98-9.95)y

90+ 6.68 (3.93- -11.35)t 7.98 (3.96- -16.07)y 6.94 (3.09-15.59)y

Male 1.00 1.00 1.00

Female 0.68 (0.64-0.72)y 0.68 (0.63- -0.72)y 0.68 (0.63-0.73)y

Unspecified 0.95 (0.45- 2.00) 0.76 (0.28- 2.03) 0.78 (0.25- 2.44)

Marital Status

Never married 1.00 1.00 1.00

Divorced 1.05 (0.89- 1.23) 0.98 (0.81- 1.19)8 1.00 (0.79-1.26)

Married 1.24(1.10- -1.40)t 1.18 (1.03- 1.36)* 1.17 (0.99-1.38)

Separated 0.91 (0.72- 1.16) 0.82 (0.61- 1.10) 0.81 (0.57-1.16)

Unknown 1.13 (0.90- 1.42) 1.04 (0.79- 1.37) 1.07 (0.77-1.50)

Widowed 1.14(1.01- 1.28)* 1.08 (0.94- 1.24) 1.08 (0.91-1.28)

Resident income quintile

Q1 1.00 1.00 1.00

Q2 1.01 (0.93- -1.09) 1.02 (0.93- 1.12) 0.96 (0.85-1.07)

Q3 1.07 (0.99- 1.15) 1.08 (0.99- 1.19) 1.07 (0.96-1.19)

Q4 1.07 (0.99- 1.16) 1.04 (0.95- 1.15) 0.98 (0.87-1.11)

Q5 1.08 (0.99- 1.17) 1.09 (0.98- 1.20) 1.11 (0.99-1.25)

Missing 1.12 (1.02- 1.23)* 1.12 (1.00- 1.25) 1.06 (0.93-1.21)

Facility urbanicity

Urban 1.00 1.00 1.00

Rural 0.94 (0.88- 1.02) 0.91 (0.84- 1.00)* 0.89 (0.80-0.99)*

Admitted from

Home care 1.00 1.00 1.00

Ambulatory 1.47 (1.23- -1.76)y 1.53 (1.23- -1.90)y 1.62 (1.26-2.10)y

or other

services

Inpatient 1.32 (1.26- -1.39)y 1.43 (1.34- -1.52)y 1.49 (1.38-1.60)y

services

Facility size, no. beds

1 - 49 1.00 1.00 1.00

50-99 0.90 (0.79- 1.01) 0.93 (0.80- 1.07) 1.04 (0.87-1.24)

100-149 0.72 (0.63- -0.81)y 0.71 (0.61- -0.83)y 0.79 (0.65-0.94)y

150+ 0.70 (0.62- -0.79)y 0.70 (0.61- -0.81)y 0.77 (0.65-0.92)y

CHESS score, 1.68 (1.64- -1.72)y 1.83 (1.79- -1.88)y 1.99 (1.94-2.05^

continuous

*P < .05. yP < .01.

Discussion

In the 1 year of follow-up after admission to a LTC facility, residents living in for-profit homes had an adjusted 10% higher risk of mortality and 25% higher risk of hospitalization; in the 3 months immediately after admission, the hazards were higher at 20% and 36%, respectively. Unadjusted numbers consistently favor not-forprofit facilities for the vast majority of subgroups examined, including across 14 geographic regions and 10 chronic conditions. Facility size was also a predictor of outcome, with smaller facilities (ie, fewer than 50 beds) showing higher mortality rates and, conversely, lower hospitalization rates.

Five studies have examined the effect of profit status on mortality and/or hospitalizations, all using data before 2000. For each

outcome, only one study significantly and clearly favored not-for-profit facilities.25,28 Our study adds to previous literature by examining both outcomes in a large, more current, population-based setting, and by addressing major limitations of previous studies, including the control

for a large set of potential confounders.15 The one Canadian study, using unadjusted analyses, found lower hospitalization rates for not-forprofit facilities and no significant relationship for mortality.25

We have shown that residents in for-profit homes consistently and robustly experience higher mortality and hospitalization rates. This occurred in an environment with common funding mechanisms, and a centralized system that leads to largely similar residents being accepted in both types of homes. It has been hypothesized that differences in outcomes may be related to reinvestments that not-forprofit facilities make into patient care that otherwise would be consumed as profit in for-profit facilities.25,31 Unlike other jurisdictions, for-profit homes (along with not-for-profit homes) in Ontario are unable to bill patients for additional funds. Under legislation, Ontario LTC homes are required to have at least one registered nurse on duty at all times; however, staffing levels and mixes, including registered practical nurses, personal support workers, and therapists, are otherwise unregulated.6 For-profit facilities in Canada and elsewhere have been shown to have lower staffing levels.32,33

Hazard Ratios for Hospitalization (Competing Risk) Model at 3, 6, and 12 Months After Admission to For-Profit and Not-For-Profit LTC Facilities

Parameter Variable Hazard Ratio (95% Confidence Interval)

12 mo 6 mo 3 mo

Ownership

Not-for-profit 1.00 1.00 1.00

For-profit 1.25 (1.21- -1.30)y 1.33 (1.27- 1.39)y 1.36(1.28- -1.43)y

Age, y

18-49 1.00 1.00 1.00

50-59 1.10 (0.90- -1.36) 1.09 (0.86- 1.39) 1.22 (0.92- 1.62)

60-69 1.14 (0.94- -1.38) 1.11 (0.89- 1.39) 1.13 (0.87- 1.46)

70-79 1.16 (0.96- -1.39) 1.15 (0.92- -1.41) 1.15 (0.89- 1.49)

80-89 1.13 (0.94- -1.36) 1.11 (0.85- 1.38) 1.20 (0.87- 1.44)

90+ 1.07 (0.88- 1.29) 1.06 (0.75- -1.31) 1.08 (0.84- 1.40)

Male 1.00 1.00 1.00

Female 0.80 (0.77- 0.83)y 0.78 (0.75- 0.82)y 0.77 (0.73-0.81)y

Other 0.92 (0.50- 1.67) 1.20 (0.66- -2.18) 1.16 (0.59- 2.30)

Marital status

Never married 1.00 1.00 1.00

Divorced 1.14(1.04- 1.26)y 1.10 (0.98- 1.23) 1.10 (0.96- 1.26)

Married 1.23 (1.14- 1.33)y 1.20(1.10- 1.31)y 1.20(1.08- 1.34)y

Separated 1.22 (1.07- 1.38)y 1.24(1.07- 1.44)y 1.20 (1.00- 1.44)*

Unknown 1.26(1.10- -1.45)* 1.19(1.00- 1.41)* 1.33 (1.10- 1.62)y

Widowed 1.20(1.11- 1.29)y 1.17 (1.07- 1.28)y 1.16(1.04- 1.29)y

Resident income quintile

Q1 1.00 1.00 1.00

Q2 0.96 (0.91- 1.01)y 0.98 (0.92- 1.04) 0.97 (0.90- -1.04)

Q3 0.94 (0.89- -0.99)* 0.96 (0.90- 1.02) 0.97 (0.90- 1.05)

Q4 0.95 (0.90- 1.01) 0.96 (0.90- 1.02) 0.96 (0.88- -1.04)

Q5 0.93 (0.88- 0.98)y 0.95 (0.89- 1.02) 0.95 (0.87- 1.03)

Missing 0.94 (0.89- 1.01) 0.94 (0.88- 1.02) 0.95 (0.87- -1.04)

Facility urbanicity

Urban 1.00 1.00 1.00

Rural 0.91 (0.86- 0.97) 0.93 (0.86- 0.99)* 0.90 (0.83- 0.98)*

Admitted from

Home care 1.00 1.00 1.00

Ambulatory or 1.26(1.11- 1.43)y 1.35 (1.17- 1.56)y 1.48 (1.25- 1.74)y

other services

Inpatient services 1.29(1.25- -1.34)y 1.37 (1.32- 1.43)y 1.44 (1.36- 1.51)y

Facility size, no. of beds

1-49 1.00 1.00 1.00

50-99 1.41 (1.25- 1.58)y 1.39(1.21- 1.60)y 1.46 (1.23- 1.73)y

100-149 1.50(1.33- 1.68)y 1.50(1.30- 1.72)y 1.59 (1.34- 1.88)y

150+ 1.61 (1.44- -1.81)y 1.61 (1.40- 1.84)y 1.72 (1.46-2.03)y

CHESS score, 1.12 (1.10- 1.14)y 1.19(1.16- 1.21)y 1.26(1.23- 1.28)y

continuous

*P < .05. yP < .01.

Some not-for-profit facilities also have been hypothesized to be closer associated with acute care facilities and provide more specialized care.25,32 Other potential differences include level of ties to the community (including volunteer presence), being associated with a chain or multisite enterprise,25 differences in capital funding, and differences in fundraising. We do not provide data supporting these hypotheses, which provide potential explanations for our findings. We do observe robust differences in "hard" outcomes. These differences are interestingly largely not present when we examine regularly reported quality-of-care indicators.12 This discordance was observed in a previous study34 and could be related to inaccuracies in indicator data, or to observed performance for narrow indicators of care that do not affect broader outcomes.18,19

Strengths

First, we examined a large, population-based cohort with similar characteristics on admission (Tables 2 and 3), regardless of facility proprietary status. The similarity of the populations is likely related to Ontario's formalized placement process. This is unlike facilities in the

United States (where the majority of studies have been published), for example, that admit differing populations according to the focus of each home. This standardization across homes allowed us to explore the role of proprietary status independent of funding or selection biases.

Second, our cohort was assessed using validated interRAl tools that capture detailed clinical and demographic information. These data are tied directly to funding and take into account patient complexity, thus increasing the impetus to ensure data quality, and reducing the incentive to preferentially accept patients of certain complexity. The data enabled control for multiple confounders at the individual level, adding to the strength of the comparison.

Third, along with the comparability of patient populations, the comparability of funding payment levels across all facilities is an issue that has challenged other studies. Among those who enter LTC facilities in Ontario, funding is fully public, with patient copayments linked directly to income, and set according to patient need. This process leads to comparable base funding (although not necessarily spending) among facilities, as suggested by the similar proportions of residents across RUGs that affect funding (Table 3). Finally, we examine differences in "hard" outcomes that reflect meaningful clinical changes that are not subject to misclassification or selective (under or over) reporting.

Weakness

We are unable to take into account all factors previously hypothesized to be associated with quality of care, including facility staffing levels and mixes, whether or not the facility belongs to a corporate chain, other sources of income (eg, from donations), and how closely associated the facility is with specialized medical ser-

vices.

15,25,32

Many of these factors, however, can be thought of as explanatory factors to differences in outcomes, or quality indicators themselves. The comparability of residents and payment models, resulting from a matching process with randomlike elements (ie, from a centralized process with multiple wait lists for each resident), also likely reduces the effect of unmeasured confounders.

Our study was conducted on an incident admission cohort to improve the comparability of the cohorts created; it does not examine outcomes past 1 year after admission. Furthermore, although examining "hard" outcomes of hospitalizations and mortality adds an objective dimension to quality of care, we recognize that some hospitalizations may be appropriate and that death may not necessarily reflect a poor outcome (eg, in palliative care). Nevertheless, our sensitivity analyses show that the higher hospi-talization rate for for-profit facilities remained when looking at a subset of potentially avoidable causes (Appendix Table 3). We also showed that residents with similar health conditions and with similar levels of health (ie, CHESS score) died at a higher rate in for-profit facilities (Appendix Table 2).

Conclusions

The differences in outcomes among residents in for-profit and not-for-profit homes suggest differences in patient experience, and likely influences health care costs. Our results should be generalizable particularly to jurisdictions where LTC is publicly funded, and where funding and placement processes are standardized. In jurisdictions without such systems, we highlight the potential effect of profit status alone, adjusting for these differences. Improving care, avoiding hos-pitalizations, and appropriately caring for residents until death are meaningful goals, with increasing relevance as need for LTC rises with the aging population. Future work can elucidate the narratives behind differences in outcomes across ownership status and facility size, both

of which are amenable to policy change. Policy needs to accordingly evolve to ensure that all residents equally receive optimal care.

References

1. Population Projections for Canada, Provinces and Territories, 2009 to 2036. Ottawa, Canada: Statistics Canada, Divison D; 2010.

2. Oeppen J, Vaupel JW. Broken limits to life expectancy. Science 2002;296: 1029-1031.

3. Tolson D, Dewar B, Jackson G. Quality of life and care in the nursing home. J Am Med Dir Assoc 2014;15:154-157.

4. Information Canadian Institute for Health Information. When a nursing home is home: How do Canadian nursing homes measure up on quality? Ottawa: CIHI; 2013.

5. Konetzka R. Do not-for-profit nursing homes provide better quality? BMJ 2009; 339:b2683.

6. Konetzka R, Stearns S, Park J. The staffing-outcomes relationship in nursing homes. Health Serv Res 2008;43:1025-1042.

7. Health Care Use at the End of Life in British Columbia. Ottawa: Canadian Institute for Health Information; 2008.

8. Estabrooks C, Knopp-Sihota J, Norton P. Practice sensitive quality indicators in RAI-MDS 2.0 nursing home data. BMC Res Notes 2013;6:460.

9. Jones R, Hirdes J, Poss J, et al. Adjustment of nursing home quality indicators. BMC Health Serv Res 2010;10:96.

10. Nakrem S, Vinsnes A, Harkless G, et al. Nursing sensitive quality indicators for nursing home care: International review of literature, policy and practice. Int J Nurs Stud 2009;46:848-857.

11. Carter M, Porell F. Nursing home performance on select publicly reported quality indicators and resident risk of hospitalization: Grappling with policy implications. J Aging Soc Policy 2006;18:17-39.

12. Health Quality Ontario. Long-Term Care Benchmarking Resource Guide. Ontario: Health Quality Ontario; 2013.

13. Statistics Canada. Residential Care Facilities 2009/10. Ottawa: Statistics Canada; 2011. Division HS.

14. Kerrison S, Pollock A. Regulating nursing homes: Caring for older people in the private sector in England. BMJ 2001;323:566-569.

15. Comondore V, Devereaux P, Zhou Q, et al. Quality of care in for-profit and not-for-profit nursing homes: Systematic review and meta-analysis. BMJ 2009;339: b2732.

16. Konetzka R, Polsky D, Werner R. Shipping out instead of shaping up: Reho-spitalization from nursing homes as an unintended effect of public reporting. J Health Econ 2013;32:341-352.

17. Werner R, Konetzka R, Polsky D. The effect of pay-for-performance in nursing homes: Evidence from state Medicaid programs. Health Serv Res 2013;48: 1393-1414.

18. Bradley E, Herrin J, Elbel B, et al. Hospital quality for acute myocardial infarction: Correlation among process measures and relationship with short-term mortality. JAMA 2006;296:72-78.

19. Werner R, Bradlow E. Relationship between Medicare's hospital compare performance measures and mortality rates. JAMA 2006;296:2694-2702.

20. Kim H, Jung Y, Sung M, et al. Reliability of the interRAI Long Term Care Facilities (LTCF) and interRAI Home Care (HC). Geriatr Gerontol Int 2015;15: 220-228.

21. Office of the Auditor General of Ontario. 2012 Annual Report. Ontario, Canada: Queen's Printer for Ontario; 2012.

22. InterRAI. Long-term care facilities (LTCF). Available at: http://www.interrai.org/ long-term-care-facilities.html. Accessed October 31, 2014.

23. Stephenson A, Hux J, Tullis E, et al. Socioeconomic status and risk of hospi-talization among individuals with cystic fibrosis in Ontario, Canada. Pediatr Pulmonol 2011;46:376-384.

24. Hirdes J, Frijters D, Teare G. The MDS-CHESS scale: A new measure to predict mortality in institutionalized older people. J Am Geriatr Soc 2003;51: 96-100.

25. McGregor M, Tate R, McGrail K, et al. Care outcomes in long-term care facilities in British Columbia, Canada. Does ownership matter? Med Care 2006;44: 929-935.

26. Shapiro E, Tate R. Monitoring the outcomes of quality of care in nursing homes using administrative data. Can J Aging 1995;14:755-768.

27. Chou S. Asymmetric information, ownership and quality of care: An empirical analysis of nursing homes. J Health Econ 2002;21:293-311.

28. Konetzka R, Yi D, Norton E, Kilpatrick K. Effects of Medicare payment changes on nursing home staffing and deficiencies. Health Serv Res 2004;39: 463-488.

29. Zinn J, Aaronson W, Rosko M. Variations in the outcomes of care provided in Pennsylvania nursing homes. Facility and environmental correlates. Med Care 1993;31:475-487.

30. Spector W, Selden T, Cohen J. The impact of ownership type on nursing home outcomes. Health Econ 1998;7:639-653.

31. Nudelman P, Andrews L. The "value added" of not-for-profit health plans. N Engl J Med 1996;334:1057-1059.

32. McGregor M, Cohen M, McGrail K, et al. Staffing levels in not-for-profit and forprofit long-term care facilities: Does type of ownership matter? CMAJ 2005; 172:645-649.

33. Berta W, Laporte A, Valdmanis V. Observations on institutional long-term care in Ontario: 1996-2002. Can J Aging 2005;24:70-84.

34. Werner R, Konetzka R, Stuart E, et al. Impact of public reporting on quality of postacute care. Health Serv Res 2009;44:1169-1187.

Appendix

Final Cohort (n = 53,739)

Appendix Figure 1. Study cohort creation for all incident Ontario LTC admissions.

Appendix Table 1

Twelve-Month Hospitalization Rates by Proprietary Status of Facility, Across Patient and Facility Characteristics

Appendix Table 2

Twelve-month Mortality Rates by Proprietary Status of Facility, Across Patient and Facility Characteristics

For-Profit Not-For-Profit Total For-Profit Not-For-Profit Total

Hospitalizations Age, y, at entry

Age, y, at entry 18 - 49 48.2 23.6 39.4

18-49 377.1 251.2 332.0 50-59 63.0 63.1 63.0

50-59 433.6 304.2 392.2 60-69 96.7 107.4 100.2

60-69 454.4 375.8 428.7 70-79 144.1 131.5 139.5

70-79 471.5 372.3 435.9 80-89 206.8 175.5 194.2

80-89 475.2 348.1 424.1 90+ 340.9 288.3 319.7

90+ 435.6 374.2 410.8 Sex

Sex Female 190.5 167.8 181.6

Female 423.3 322.2 383.9 Male 239.5 218.5 231.6

Male 538.8 432.2 498.7 Other 322.6 222.8 286.0

Other 516.2 445.5 490.3 Admitted from

Admitted from Ambulatory and other services 249.5 202.9 231.2

Ambulatory and other services 562.9 325.6 469.9 Inpatient services 247.7 238.6 244.6

Inpatient services 548.0 422.5 505.2 Home services 166.7 146.8 158.3

Home services 376.6 315.1 350.6 No. of beds

No. of beds 1 - 49 298.7 282.9 289.8

1 - 49 380.5 216.7 289.0 50-99 262.1 177.6 243.1

50-99 428.3 344.1 409.3 100-149 196.7 176.3 189.5

100-149 446.8 357.8 415.2 >150 175.5 180.4 177.7

>150 495.7 372.8 439.8 Facility urbanicity

Facility urbanicity Urban 194.8 183.4 190.5

Urban 469.9 364.0 429.5 Rural 284.3 189.0 245.1

Rural 415.0 324.8 377.9 CHESS score

CHESS score: 0 (no instability) 118.6 102.6 112.5

0 (no instability) 384.5 297.1 350.8 1 228.8 202.8 218.8

1 500.2 388.3 457.3 2 355.7 311.2 338.6

2 632.3 486.2 576.1 3 570.6 511.6 546.9

3 657.7 502.7 595.4 4 1082.3 900.3 1003.8

4 787.1 610.4 710.9 5 (highest level of instability) 6572.2 5174.6 5886.1

5 (highest level of instability) 852.0 504.8 681.5 Chronic disease prevalence

Chronic disease prevalence Alzheimer 160.6 140.7 152.1

Alzheimer 341.1 252.7 303.3 Dementia other than Alzheimer 208.8 192.3 202.6

Dementia other than Alzheimer 407.5 308.4 370.4 Cancer 422.2 369.8 401.9

Cancer 551.7 443.8 510.0 Stroke 219.7 196.2 211.5

Stroke 498.5 404.6 465.9 Congestive heart failure 370.9 349.5 363.1

Congestive heart failure 738.0 566.7 675.8 Atherosclerotic heart disease 272.5 229.5 254.5

Atherosclerotic heart disease 559.8 447.2 512.7 Emphysema/chronic obstructive 279.0 271.9 276.5

Emphysema/chronic obstructive 662.2 522.3 612.3 pulmonary disease

pulmonary disease Depression 184.7 160.7 175.5

Depression 436.2 353.5 404.5 Diabetes 214.4 187.7 204.8

Diabetes 594.5 451.8 543.3 Arthritis (osteoarthritis, rheumatoid) 210.6 185.8 201.1

Arthritis (osteoarthritis, rheumatoid) 449.3 360.3 415.0 Local Health Integration Network (LHIN)

Local Health Integration Network (LHIN) Erie St. Clair 212.2 218.9 213.8

Erie St. Clair 500.0 322.7 457.1 South West 246.3 213.3 235.1

South West 454.1 363.1 423.3 Waterloo Wellington 280.2 212.8 259.1

Waterloo Wellington 324.5 293.5 314.8 Hamilton Niagara Haldimar Brant 227.7 208.4 220.1

Hamilton Niagara Haldimar Brant 390.5 282.8 348.2 Central West 170.7 135.5 159.2

Central West 595.1 381.5 525.2 Mississauga Halton 151.3 117.8 143.5

Mississauga Halton 521.8 433.7 501.2 Toronto Central 145.1 128.4 135.4

Toronto Central 553.0 420.5 475.8 Central 151.5 151.3 151.4

Central 559.9 446.6 520.0 Central East 185.6 183.3 184.9

Central East 476.3 335.1 430.6 South East 261.5 210.9 243.1

South East 377.9 319.3 356.5 Champlain 231.1 200.5 215.7

Champlain 402.4 373.8 388.0 North Simcoe Muskoka 217.4 285.0 238.2

North Simcoe Muskoka 474.5 214.4 394.5 North East 189.3 165.9 176.3

North East 475.5 399.8 433.6 North West 149.1 139.1 142.8

North West 555.4 408.6 462.7

Appendix Table 3

Prevalence of Quality Indicators on Admission and 3 Months After Admission, by Proprietary Status

Quality Indicators On Admission, % Residents 3-Months After Admission, % Residents

For-Profit Not-For-Profit Total For-Profit Not-For-Profit Total

Physical restraints 4.7 6.4 5.3 7.0 8.9 7.8

Any pressure ulcer 11.4 9.4 10.7 5.6 5.4 5.5

Worsening pressure ulcer — — — 4.1 4.2 4.2

Any incontinence 54.0 51.9 53.2 63.6 65.4 64.4

Worsening incontinence — — — 31.8 33.4 32.5

Falls 21.3 21.0 21.2 13.3 14.7 13.9

Note: The dashes signify that the values were not measured.