Table 2.
Included studies with identified impacts of private equity ownership. Studies are in alphabetical order (continued from table 1)
| Reference | Country | Participants | Comparisons | Primary outcomes | Study period | Study type | Findings related to primary outcomes |
|---|---|---|---|---|---|---|---|
| Cerullo et al 202243 | USA | Hospitals | Non-PE | Costs to operator, quality | 2002-17 | Longitudinal | PE hospitals were associated with a decrease in cost per adjusted discharge (−$431.6 ($−622.2 to $−241.0), P<0.001) PE hospitals were associated with a decrease in total staff FTEs per occupied bed (−0.50 (−0.71 to −0.30), P<0.001), as well as a decrease in total RN and LPN FTEs (−0.04 (−0.07 to −0.02), P<0.01), suggesting lower nurse staffing |
| Cerullo et al 2022*44 | USA | Hospitals | Non-PE | Health outcomes, costs to patients or payers, quality | 2001-18 | Longitudinal | PE hospitals showed a relative decrease in acute myocardial infarction in-hospital mortality compared with controls (−1.14 pp (−1.86 to −0.42 pp), P=0.03) from pre-acquisition to post-acquisition PE hospitals showed a relative decrease in acute myocardial infarction 30 day mortality compared with controls (−1.41 pp (−2.26 to −0.56 pp), P=0.03) from pre-acquisition to post-acquisition No differences in 30 day payments, or 30 day readmissions for acute myocardial infarction, acute stroke, chronic obstructive pulmonary disease, congestive heart failure, or pneumonia were identified |
| Cerullo et al 202145 | USA | Hospitals | Non-PE | Quality | 2004-18 | Longitudinal | Relative to controls, PE hospitals were more likely to offer certain profitable services lines after PE acquisition: robotic surgery (6.2%, P<0.001), digital mammography (4.1%, P=0.02), adult interventional cardiac catheterization (3.8%, P=0.01), in-hospital hemodialysis (3.6%, P=0.01), free-standing or satellite emergency department (2.5%, P=0.03), birthing room or labor and delivery (2.3%, P=0.01) Relative to controls, PE hospitals were less likely to offer inpatient orthopedic surgery after acquisition (profitable service line) (−2.6%, P=0.03) Relative to controls, PE hospitals were less likely to offer outpatient psychiatric care after acquisition (unprofitable service line) (−4.0%, P=0.001) Relative to controls, PE hospitals were more likely to offer psychiatric emergency services after acquisition (unprofitable service line) (4.0%, P=0.01) Relative to controls, PE hospitals were less likely to offer ambulance services after acquisition (miscellaneous) (4.9%, P<0.001) |
| Creadore et al 202146 | USA | Dermatology clinics | Non-PE | Quality | 2020 | Cross sectional | Appointment availability for patients with Blue Cross Blue Shield was higher at PE clinics than at control clinics (98.5% (96% to 99%) v 94.6% (92% to 96%), P=0.03) Appointment availability for patients with Medicare was higher at PE clinics than at control clinics (97.5% (94% to 99%) v 92.8% (90% to 95%), P=0.02) PE clinics were more likely than controls to have appointments available with NPCs (80% (75% to 84%) v 63% (59% to 67%), P=0.001) PE clinics were more likely to have a next day appointment with any clinician relative to control clinics (30% (26% to 35%) v 21% (19% to 21%), P=0.001) |
| Gandhi et al 202022 | USA | Nursing homes | Non-PE | Quality | 1993-2017 | Longitudinal | From pre-2006 to post-2009, PE facilities showed both higher and lower levels of CNA, LPN, and total nursing expenditures, performing better under high scrutiny and reporting conditions and worse under low scrutiny and reporting conditions After the implementation of the Five-Star rating system in 2009, PE facilities had fewer health inspection deficiencies relative to controls (−5.501, P<0.05) After the implementation of the Five-Star rating system in 2009, PE facilities had greater long-stay quality scores relative to controls (2.777, P<0.01) |
| Gandhi et al 202047 | USA | Nursing homes | Non-PE; previously PE | Health outcomes, quality | 2020 | Cross sectional | PE was associated with an average decrease in the probability of confirmed covid-19 outbreaks in residents relative to controls (−7.1 pp (−11.3 to −2.9 pp), P<0.001) PE was associated with an average decrease in the probability of suspected covid-19 outbreaks in residents relative to controls (−13 pp (−17.8 to −8.3 pp), P<0.001) PE was associated with a decrease in the probability of confirmed covid-19 outbreaks in staff relative to controls (−5.4 pp (−9.8 to −1.1 pp), P=0.014) PE was associated with a decrease in the probability of suspected covid-19 outbreaks in staff relative to controls (−8.7 pp (−13.4 to −4.0 pp), P<0.001) PE was associated with a decrease in the likelihood of having a shortage of N95 masks (−6.4, P<0.01), surgical masks (−7.6 pp, P<0.01), protective eyewear (−4.8 pp, P<0.01), gowns (−7.0 pp, P<0.01), and gloves (−3.3 pp, P=0.02) relative to controls |
| Gupta et al 2021†48 | USA | Nursing homes | Non-PE | Health outcomes, costs to patients or payers, costs to operators, quality | 2000-17 | Longitudinal | PE was associated with an increase in mortality during stay duration and 90 days after discharge (0.0169 (0.007), P<0.05) PE was associated with an increase in the log amount billed per patient stay (0.1777 (0.028), P<0.01) PE was associated with an increase in the log amount billed per patient stay and up to 90 days post-discharge (0.1054 (0.024), P<0.01) PE was associated with worse Five-Star overall quality scores (−0.082 (0.036), P<0.05) PE was associated with a lower level of all staffing per patient day (−0.048 (0.016), P<0.01) PE was associated with a lower level of CNA staffing per patient day (−0.066 (0.010), P<0.01) PE was associated with a lower level of LPN staffing per patient day (−0.019 (0.006), P<0.01) PE was associated with a greater level of RN staffing per patient day (0.037 (0.005), P<0.01) PE was associated with greater building lease costs for facilities (0.560 (0.061), P<0.01) PE was associated with worse quality scores pertaining to patient antipsychotic medication use, patient mobility scores, and deficiencies |
| Harrington et al 201249 | USA | Nursing homes | Non-PE (for profit, non-profit, government) | Quality | 2003-08 | Longitudinal | Relative to controls, PE homes showed an increase in total number of deficiencies in years 2006 and 2007 post-sale years relative to pre-sale years (0.197 (0.083), P<0.05; 0.205 (0.075), P<0.01, respectively) Relative to controls, PE homes showed an increase in number of severe deficiencies in 2006 and 2007 post-sale years relative to controls (0.450 (0.192), P<0.05; 0.393 (0.173), P<0.05, respectively) |
| Huang and Bowblis 2019‡50 | USA | Nursing homes | Non-PE (for profit) | Quality | 2005-10 | Longitudinal | PE was associated with mixed impacts on multiple quality indicators, including catheter use, bowel/bladder incontinence, physical mobility, pressure ulcers, contractures, anti-anxiety medication use, and antidepressant medication use |
$
1.00 (£0.79; €0.92).
CNA=certified nursing assistant; FTE=full time equivalent; LVN=licensed vocational nurse; NPC=non-physician clinician.
*
Mortality was categorized as a health outcome and readmission was categorized as a quality metric.
†
Impacts are classified as negative, in alignment with the authors’ reporting, as the increase in RN staffing was offset by larger decreases in CNAs and LPNs staffing.
‡
Reported results are from the 2SRI model without fixed effects in the referenced manuscript (table 4, column 3).