CMS Responds to NAHC Letter on CAHPS

The Centers for Medicare & Medicaid Services (CMS) recently responded to a Fall 2012 letter from National Association for Home Care & Hospice (NAHC) in which NAHC posed a number of home health agency questions and concerns about the Home Health Consumer Assessment of Healthcare Providers and Systems (CAHPS). CMS responded as follows to NAHC’s letter, failing to agree to any shortcomings or need for changes to HH CAHPS.

  • Regarding home health agencies belief that patients that receive services from other organizations while under a home health plan of care (e.g. housekeeping, chore, etc.) are unable to differentiate between providers, often resulting in lower scores for home health agencies, CMS explained that since they do not survey these other provider types they do not have data to evaluate the potential impact of other services received concurrently with home health.
  • CMS also did not find evidence to support the contention that up to 5% of survey responses must be discarded because patients check multiple responses to questions, answering for each individual discipline that provided service. However, CMS agreed to continue to monitor the data for this issue and “to keep an eye out for this on site visits made to review vendor operations.”
  • In response to NAHC’s question as to whether CMS should consider including economic status in the risk adjustment model for those agencies that serve a significantly higher percentage of dually eligible patients, CMS stated that they evaluated the potential effect of having Medicaid-reimbursed home health care using data from the HHCAHPS Survey Mode Experiment and found that Medicaid status was not a statistically significant predictor, so Medicaid status is not used as an adjuster. However, CMS reminded NAHC that education is a patient mix adjuster, which may function as a proxy for Medicaid status.
  • NAHC asked about the impact of patterns of unanswered questions on agency scores, to which CMS answered that they do not analyze unanswered questions for patterns of missing data. CMS expressed the belief that agencies are not being penalized for patterns of missing data.
  • Regarding concerns about the impact of missing patient demographic information on mix adjusters CMS answered that “any missing data on demographic characteristics is imputed using a hot deck statistical procedure so that we can calculate the proportion of an agency’s patients having each of the demographic characteristics that are used as an adjuster.” As a result, CMS concluded that “under the assumption missing demographic data is random in nature, we believe any impacts on agency scores are negligible.”
  • NAHC also asked how the lack of weighting of State averages impacts the standing of very small agencies with small sample sizes. CMS explained that since State averages are designed to enable the consumer to compare one agency to another within a state, State averages are purposely not weighted to avoid skewing the average. In regard to risk adjustment of small agency results, CMS replied that all agency results are statistically adjusted for patient-mix, regardless of agency size.
  • NAHC asked CMS whether they would be willing to discuss the large number of questions in the survey, with attention to the six medication questions. CMS answered that the number of questions in the HHCAHPS survey was reduced during the field testing undertaken prior initiation of the surveys and that CMS and RTI project staff have not observed any issues with non-response related to the length of the survey.
  • Another area of concern expressed by NAHC was the finding that scores on medication questions are significantly lower for patients who reside in group homes and assisted living facilities when their medications are managed and administered by facility personnel. CMS addressed this concern by stating that “agencies are responsible for communicating regularly with residents in all living situations (home, assisted living, group homes, etc.) regarding information on their drug regimen. Even if the medication is being administered by the facility, patients should know about the home health agency’s role in this process.” This same position was taken by CMS related to congregate living environmental assessments, saying that “during assessment, agency staff should be assessing the patient’s environment for ambulation and/or fall risk in these congregate settings.”
  • As far as home health agencies perception that patients fail to understand the question of “How often did the provider from this agency seem informed and up-to-date about the care and treatment you got at home,” CMS supported the current phrasing of this item, saying it was based on extensive cognitive testing of the survey during the development phase.
  • Finally, in response to the request that CMS consider revising the wording of the agency rating question “…where 0 is the worst possible and 10 is the best possible…” since patients who have not had home health services in the past have no frame of reference for responding, CMS said that this is a standard CAHPS question and was extensively tested across provider settings. Therefore, there are no plans to amend the CAHPs question or scoring scale.
Regarding reports that wording of the CAHPS cover letter leads patients to call the vendor, rather than the agency about problems related to their condition or services, CMS offered that they will look at the model letter they provide to vendors, but that vendors (and or their HHA) can revise the letter as long as it contains certain required elements.

Despite CMS’ unwillingness to further analyze home health agency reports or adopt suggestions made, NAHC requests that home health agencies continue to share their concerns and questions, as well as recommendations to improve HH CAHPS so that we can pursue additional dialogue with CMS and lobby for necessary changes to improve the survey and the survey process.


From the NAHC Report article