What was the research about?
Nursing home residents and people with dementia can have symptoms such as restlessness or aggression. Some types of medicines, like antipsychotics, may help treat these symptoms. But these medicines have serious side effects, especially in the elderly, and raise the risk of falls, confusion, and even death.
In this study, the research team wanted to learn if training staff at nursing homes and doctors’ offices would change how often people use these medicines. The team offered training materials to nursing homes or doctors’ offices in 29 counties in rural Iowa. The training included information about the risks of these types of medicines and the opportunity to talk with specialists. The team compared symptoms and medicine use among people with dementia and nursing home residents in the 29 counties with those in 10 counties that weren’t offered the training.
What were the results?
Offering the training didn’t change how often people with dementia and nursing home residents used these types of medicines. It may have reduced symptoms for some nursing home residents.
Nursing home residents in counties that had the training were
- More likely to receive doses of antipsychotics that were too high
- Less likely to have symptoms such as restlessness and aggression
Who was in the study?
The study used Medicare data from 6,275 nursing home residents and 4,845 people with dementia who lived in the community. All lived in rural counties in Iowa.
What did the research team do?
The research team offered the training differently in the 29 counties. In 10 counties, county extension staff delivered the materials in person. In 19 counties, the team mailed it to nursing homes and doctors’ offices. In all 29 counties, nursing home staff and doctors could have calls or video meetings with a pharmacist or nurse who were experts in dementia. These specialists could answer questions about medicine and how to plan treatment for patients’ needs.
The research team looked at medicine use before and for up to 15 months after the training. They compared results in counties that had the training with those that didn’t.
Doctors, nursing home directors, quality improvement groups, Alzheimer’s groups, and a patient family member helped plan the study.
What were the limits of the study?
The research team first presented the training at conferences in 2011 and 2012. Clinicians and nursing home staff may have been familiar with the information. The study took place at the same time as a national project to improve dementia care in nursing homes. That project might have affected the study results. The team had data about symptoms for nursing home residents only. The study looked only at medicine paid for by insurance.
Future research could look at other ways to improve safe medicine use for the elderly.
How can people use the results?
Clinicians and nursing home directors can use the results when considering ways to improve safe use of medicines for nursing home residents and patients with dementia.
To determine whether an outreach program for providers in rural areas, about behavioral and psychological symptoms of dementia, reduces unnecessary medication use without worsening symptoms among outpatients with dementia and nursing home residents
|Population||6,275 Medicare-eligible adults ages 66 or older residing in a nursing home, and 4,845 outpatient Medicare-eligible adults ages 66 or older with a dementia diagnosis|
Primary: antipsychotic use, anticholinergic use
Secondary: benzodiazepine use, antipsychotic use at excessive doses, measures of behavioral or psychological symptoms
|Timeframe||15-month follow-up for primary outcomes|
This multi-phase study examined the effect of an outreach program for prescribing clinicians and nursing home staff to improve prescribing of antipsychotics, anticholinergics, and benzodiazepines among outpatients with dementia and nursing home residents with or without dementia in rural counties in Iowa. The program included distance learning materials, called Iowa Improving Antipsychotic Appropriateness in Dementia Patients (IA-ADAPT), and consultations with specialists.
In 29 intervention counties, 307 prescribing clinicians and 107 nursing homes received the IA-ADAPT materials. In 10 of these counties, county extension staff delivered IA-ADAPT materials to nursing home staff and physicians who treat patients with dementia, and in 19 counties, researchers mailed the IA-ADAPT materials. Across counties, 14 prescribing clinicians and 46 nursing home staff participated in specialist consultations.
Researchers compared the intervention counties with 10 rural counties where nursing home staff and physicians did not receive IA-ADAPT materials and were not offered consultations.
Researchers used Medicare Part D data to assess changes in antipsychotic or anticholinergic use among nursing home residents and outpatients with dementia before and up to 15 months after clinicians or nursing home staff in intervention counties received IA-ADAPT materials. They used Minimum Data Set information to measure changes in behavioral or psychological symptoms of dementia among nursing home residents.
Clinicians, nursing home administrators, quality improvement organizations, Alzheimer’s advocacy group representatives, and a patient-family representative helped plan the study.
Changes in antipsychotic, anticholinergic, and benzodiazepine use in nursing home residents or outpatients with dementia did not differ between the intervention and control counties.
After intervention exposure, nursing home residents from intervention counties were
- More likely to use antipsychotics at excessive doses (odds ratio [OR]=1.069; 95% confidence interval [CI]: 1.024, 1.115)
- Less likely to exhibit behavioral or psychological symptoms of dementia for which antipsychotics would be appropriate (OR=0.967; 95% CI: 0.954, 0.980)
Some analyses suggested reduced antipsychotic and anticholinergic use in certain subgroups, but these findings were not consistent across different analytic approaches.
The study took place while another national project to reduce antipsychotic use in nursing home residents was in progress, which may have created confounders that could not be controlled. Clinicians in control and intervention counties might have been exposed to IA-ADAPT materials in 2011 or 2012, when these materials were presented through a website and conferences. Researchers had data on behavioral and psychological symptoms for nursing home residents only, not outpatients with dementia. Prescription claims data only included medications covered by insurance.
Conclusions and Relevance
An outreach program for community prescribers and nursing home staff had little effect on patient medication use. The program was associated with reduced behavioral and psychological symptoms of dementia in nursing home residents.
Future Research Needs
Future research could focus on different ways to improve appropriate medication use in patients with dementia and nursing home residents.
Final Research Report
View this project's final research report.
Related Journal Citations
Peer review of PCORI-funded research helps make sure the report presents complete, balanced, and useful information about the research. It also assesses how the project addressed PCORI’s Methodology Standards. During peer review, experts read a draft report of the research and provide comments about the report. These experts may include a scientist focused on the research topic, a specialist in research methods, a patient or caregiver, and a healthcare professional. These reviewers cannot have conflicts of interest with the study.
The peer reviewers point out where the draft report may need revision. For example, they may suggest ways to improve descriptions of the conduct of the study or to clarify the connection between results and conclusions. Sometimes, awardees revise their draft reports twice or more to address all of the reviewers’ comments.
Peer reviewers commented, and the researchers made changes or provided responses. The comments and responses included the following:
- Reviewers wondered if greater stakeholder input into the study questions could have improved provider participation. The researchers said they had worked with stakeholders for many years before this study, and their stakeholders expressed as much surprise at the low participation rates as the researchers. Given space constraints, they added only a brief mention of all of their efforts to identify community priorities in dementia care.
- Reviewers noted that the initial study design became more confusing when the researchers increased the number of participating counties in the second phase, after low uptake in the first phase. In addition, by the time the second phase started, the control condition facilities were most likely affected by new mandates from the Centers for Medicare and Medicaid Services’ Partnership Initiative. Therefore, the control condition facilities were more similar to the intervention under study. The researchers acknowledged that these factors could have contributed to the lack of significant findings in the study, but they noted that by focusing on three-month outcomes they lost some sensitivity to treatment effects. After reanalyzing their data based on their original strategy using one-month outcomes, they found many more, albeit small, beneficial effects.
- Reviewers commented that although the report referenced using interrupted time series (ITS) analysis for evaluation of the standard intervention and the multi-component intervention, the results were not presented in a format typically seen with this type of analysis. The researchers confirmed that they did use ITS as the design and analysis framework. However, they explained that because of the multiple intervention time points, they used a more complicated type of ITS and were not able to provide the typical time series graphics seen in reports that use ITS.
- Reviewers asked the researchers to expand on the implications of paying providers to participate in aspects of the study, noting that this approach may limit the study’s generalizability. The researchers noted that participation was lower in the phase of the study where they offer payment to participants than in the phase where they did not offer payment. Since they therefore did not recommend payment as an incentive in this case, the researchers declined to expand on the topic other than reporting the higher participation rates in the phase of the study without provider payment.
Conflict of Interest Disclosures
Study Registration Information
- Has Results