In the Journals

Medicare savings program not associated with improved quality, spending

Medicare Shared Savings Program participation was not tied to improvements in spending, quality of care or outcomes, according to study results published in the Annals of Internal Medicine.

“Evidence suggests that [Medicare Shared Savings Program accountable care organizations] are associated with modest improvements in spending and quality,” Adam A. Markovitz, BS, from the University of Michigan Medical School and School of Public Health, and colleagues wrote. “However, [Medicare Shared Savings Program] evaluations may be subject to confounding from nonrandom participation or attrition within [accountable care organizations].”

To evaluate the effects of the Medicare Shared Savings Program (MSSP) while accounting for nonrandom clinician exits, researchers compared MSSP accountable care organization (ACO) participants and control beneficiaries. Data for beneficiaries was based on national claims between 2008 to 2014 from a random 20% sample of Medicare fee-for-service beneficiaries who were aged 65 years and older. MSSP participants were identified with provider and beneficiary files from CMS.

Researchers used adjusted longitudinal models to compare changes in spending and quality between MSSP participants and local control beneficiaries, secular trends and market-level factors. The number of nearby clinicians in a MSSP was used as an instrumental variable to account for nonrandom entry and exit from the MSSP.

The effect of MSSP on hospitalization for hip fracture was used as a falsification test, as previous studies showed that the rates of hip fracture hospitalization are indicative of true population incidence rates.

The models showed that MSSP was associated with a –$118 spending reduction per beneficiary quarter (95% CI, –151 to –85) and improvements in all quality indicators.

Instrumental variable models found that MSSP was not tied to changes in quality of spending per beneficiary-quarter (change = $5; 95% CI, –51 to 62)

The falsification tests found that MSSP was tied to hip fracture in the adjusted model, with –0.24 hospitalizations for hip fracture per 1000 beneciary-quarters (95% CI, –0.32 to –0.16). In the instrumental variable model, MSSP was not associated with hip fracture, with 0.05 hospitalizations per 1000 beneciary-quarters (95% CI, –0.1 to 0.2).

Supplemental analyses found that high-cost clinicians had a 30.4% chance of leaving the MSSP compared with median-cost clinicians, who had a 13.8% chance of leaving. The increased rate of high-cost clinicians who left MSSPs suggested that improvements in MSSP spending and quality may have been driven by the nonrandom exit of high-cost clinicians rather than the effectiveness of MSSPs, according to the researchers.

“These findings suggest caution in extending ACOs to other settings and patients without stronger evidence that the program saves money or improves quality of care,” Markovitz and colleagues wrote. “Our study underscores the degree to which selection bias may affect evaluations of voluntary reforms and the challenges inherent in evaluating these programs.” – by Erin Michael

Disclosures: Markovitz reports no relevant financial disclosures. Please see study for all other authors’ relevant financial disclosures.

Medicare Shared Savings Program participation was not tied to improvements in spending, quality of care or outcomes, according to study results published in the Annals of Internal Medicine.

“Evidence suggests that [Medicare Shared Savings Program accountable care organizations] are associated with modest improvements in spending and quality,” Adam A. Markovitz, BS, from the University of Michigan Medical School and School of Public Health, and colleagues wrote. “However, [Medicare Shared Savings Program] evaluations may be subject to confounding from nonrandom participation or attrition within [accountable care organizations].”

To evaluate the effects of the Medicare Shared Savings Program (MSSP) while accounting for nonrandom clinician exits, researchers compared MSSP accountable care organization (ACO) participants and control beneficiaries. Data for beneficiaries was based on national claims between 2008 to 2014 from a random 20% sample of Medicare fee-for-service beneficiaries who were aged 65 years and older. MSSP participants were identified with provider and beneficiary files from CMS.

Researchers used adjusted longitudinal models to compare changes in spending and quality between MSSP participants and local control beneficiaries, secular trends and market-level factors. The number of nearby clinicians in a MSSP was used as an instrumental variable to account for nonrandom entry and exit from the MSSP.

The effect of MSSP on hospitalization for hip fracture was used as a falsification test, as previous studies showed that the rates of hip fracture hospitalization are indicative of true population incidence rates.

The models showed that MSSP was associated with a –$118 spending reduction per beneficiary quarter (95% CI, –151 to –85) and improvements in all quality indicators.

Instrumental variable models found that MSSP was not tied to changes in quality of spending per beneficiary-quarter (change = $5; 95% CI, –51 to 62)

The falsification tests found that MSSP was tied to hip fracture in the adjusted model, with –0.24 hospitalizations for hip fracture per 1000 beneciary-quarters (95% CI, –0.32 to –0.16). In the instrumental variable model, MSSP was not associated with hip fracture, with 0.05 hospitalizations per 1000 beneciary-quarters (95% CI, –0.1 to 0.2).

Supplemental analyses found that high-cost clinicians had a 30.4% chance of leaving the MSSP compared with median-cost clinicians, who had a 13.8% chance of leaving. The increased rate of high-cost clinicians who left MSSPs suggested that improvements in MSSP spending and quality may have been driven by the nonrandom exit of high-cost clinicians rather than the effectiveness of MSSPs, according to the researchers.

“These findings suggest caution in extending ACOs to other settings and patients without stronger evidence that the program saves money or improves quality of care,” Markovitz and colleagues wrote. “Our study underscores the degree to which selection bias may affect evaluations of voluntary reforms and the challenges inherent in evaluating these programs.” – by Erin Michael

Disclosures: Markovitz reports no relevant financial disclosures. Please see study for all other authors’ relevant financial disclosures.