Geriatric Nursing 36 (2015) 182e191
Contents lists available at ScienceDirect
Geriatric Nursing journal homepage: www.gnjournal.com
Feature Article
Caring for people with dementia and challenging behaviors in nursing homes: A needs assessment geriatric nursing Jeanette M. Daly, RN, PhD a, *, Camden P. Bay, MS b, Barcey T. Levy, PhD, MD a, b, Ryan M. Carnahan, MS, PharmD b a b
Department of Family Medicine, Carver College of Medicine, The University of Iowa, Iowa City, IA, USA Department of Epidemiology, College of Public Health, The University of Iowa, Iowa City, IA, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 5 November 2014 Received in revised form 29 December 2014 Accepted 5 January 2015 Available online 9 February 2015
An estimated 50% of nursing home residents have a dementia diagnosis. The purpose of this research was to conduct a needs assessment of directors of nursing (DON) in Iowa nursing homes in relation to caring for patients with Behavioral and Psychological Symptoms of Dementia. DON responses were linked to Online Survey Certification and Reporting/Certification and Survey Provider Enhanced Reporting (OSCAR/CASPER) data to examine how facility characteristics may be associated with use of and confidence in non-drug management strategies. From 431 questionnaires mailed to DONs, 160 (37%) were returned. Regression analysis showed that those who were more confident in managing challenging behavior were more likely to have satisfaction with current training on managing challenging behaviors and had a psychiatrist available to visit the facility. Facilities with a larger proportion of patients with challenging behaviors being treated with non-drug approaches instead of antipsychotics had DONs who were more likely to be confident in non-drug management strategies and have knowledge about the FDA antipsychotic medications risks. Ó 2015 Elsevier Inc. All rights reserved.
Keywords: Nursing home Director of nursing Dementia Challenging behaviors Antipsychotic medication
Currently, there are more than 15,000 Medicare and Medicaidcertified nursing homes in the United States1 with almost 3.3 million patients receiving care in those nursing homes.2 An estimated 50% of nursing home residents have a diagnosis of dementia3,4 with 13% having dementia or Alzheimer’s disease as their primary diagnosis.5 Neuropsychiatric disturbances occur in 60e98% of people with dementia, frequently triggering nursing home placement and antipsychotic treatment.6 Antipsychotics have been a mainstay of pharmacologic management. In 2006, 25.9% of nursing home residents in the United States received antipsychotics,7 compared to 16.4% just 10 years earlier. Recent estimates place the prevalence of nursing home antipsychotic use around 22%.8 Evidence suggests that antipsychotic use is often suboptimal, based on lack of appropriate indications, inappropriate dosing,9,10 and high variability in usage rates across facilities.10 This is of particular concern since antipsychotics are associated with serious side effects in people with dementia, including an increased risk of mortality
Funding: ARHQ R18 HS19355-01. * Corresponding author. 01290-F PFP, 200 Hawkins Drive, Iowa City, IA 52242, USA. Tel.: þ1 319 384 8995. E-mail address: [email protected] (J.M. Daly). 0197-4572/$ e see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.gerinurse.2015.01.001
which was the reason for a black box warning required by the Food and Drug Administration. In 2014 the Centers for Medicare and Medicaid Services (CMS) launched the National Partnership to Improve Dementia Care initiative to enhance dementia care quality and address antipsychotic overuse, with a goal of reducing antipsychotic use in nursing homes by 30% at the end of 2016.11 Regulatory requirements have facilitated a marked improvement in decreasing the use of antipsychotic medications in nursing homes from 24% in the fourth quarter of 2011 to 20% in the first quarter of 2014.11 The current focus of the initiative is to not only reduce antipsychotics but also to encourage person-centered care and non-drug management of behavioral symptoms of dementia in order to reduce unnecessary antipsychotic use.12 Challenging behaviors for persons with dementia are common and multi-faceted being attributable to the environment, chronic conditions, health, or medications. More than 80% of nursing home residents with dementia exhibit one or more forms of challenging behaviors.13 Examples of challenging behaviors include agitation, anger, depression, disrobing, eating abnormalities, hoarding, inappropriate sexual behavior, irritability, paranoia, physical and verbal aggression, repetition, swearing, and withdrawal. There are situations in which antipsychotic use in dementia patients is considered appropriate by most clinicians as well as CMS, particularly when a patient’s behaviors are dangerous to
J.M. Daly et al. / Geriatric Nursing 36 (2015) 182e191
themselves or others.14 Initial therapy for patients with dementia who have challenging behaviors is environmental modifications and non-pharmacologic approaches. However, if that is unsuccessful and antipsychotic use is warranted, an appropriate indication for use of the medication needs to be documented. Once antipsychotic medication is started ongoing monitoring to evaluate its effectiveness and observation for adverse effects is necessary. If the medication is ongoing, the lowest effective dose should be established and maintained. In 2013, the percentage of nursing home residents taking antipsychotic medication ranged from 12.3% in Hawaii to 33.1% in Louisiana with a mean of 24.2% in the U.S. 15 In Iowa the percentage of nursing home residents taking antipsychotic medication was 21.2% in 2012.15 The purpose of this research study was to conduct a needs assessment of directors of nursing (DON) in Iowa nursing homes in relation to caring for patients with dementia and those with challenging behaviors. That is a needs assessment to establish knowledge deficits in which to appropriately care for people with Behavioral and Psychological Symptoms of Dementia (BPSD) and then to best determine how they would like to receive additional training. This needs assessment would help to inform the development of educational materials and tools to help guide care for people with BPSD. In addition, the DON responses were linked to the Online Survey Certification and Reporting/Certification and Survey Provider Enhanced Reporting (“OSCAR/CASPER”) survey and inspection data to provider responses to examine how information on facility operations, patient census and regulatory compliance may be associated with use of and confidence in non-drug management strategies for behavioral symptoms of dementia. Comparison of both resident and facility data allows us to look at the variation and influence of these factors. In particular, we utilized regression modeling to aid in assessing the relationship between the proportion of dementia patients with challenging behavior treated using non-drug approaches and nurse’s knowledge, confidence, and education, family involvement, medical director education, availability of a psychiatrist or nurse practitioner with a dementia/ psychiatric specialty, facility operations, and patient census (see Table 1 for the entire list OSCAR/CASPER Date). Additionally, we wanted to look at nurse’s confidence in managing challenging patient behaviors associated with dementia and non-drug approaches. Methods After institutional review board approval, a mailed survey was sent to Iowa nursing homes. The Medicare nursing home compare website (medicare.gov/NursingHomeCompare) listed 443 nursing homes in Iowa. Of the 443, 12 were hospitals and eliminated from the list because the cover letter read director of nursing and would go to the hospital director and not the nursing home director, leaving 431 nursing homes. All nursing home DONs were invited to participate and questionnaires were mailed in January 2011. DONs were asked to complete the questionnaire while skipping any questions they did not want to answer and mail it back. If no response was given by a DON within three weeks, a second questionnaire was mailed. Following the two mailings, no further mailings or contacts were made. Instrument A 22-item questionnaire (see Appendix 1) entitled, “Needs Assessment: Caring for People with Dementia and Challenging Behaviors” developed by the research team of clinicians, researchers, and education specialists, was used to assess the needs of DONs in regard to caring for individuals with BPSD. To ensure
183
content validity, the questionnaire was critique by current or former nursing home employees that were on the research team. Questions included information pertaining to the treatment of dementia patients and the education of nursing home staff and family members about dementia. To better understand current dementia management approaches, DONs were asked about which antipsychotics were most often prescribed (a list of commonly used antipsychotics was provided), satisfaction with current training on managing dementia behaviors, if the nursing staff provides influence on the selection of the antipsychotic medication prescribed, their awareness of the FDA warning on antipsychotics in dementia, and their usage as well as confidence in non-drug treatment approaches for managing BPSD. To gather information related to the education of nursing home staff and family members of dementia patients, DONs were asked how they obtain information about dementia treatment, what dementia topics interest them the most, preferred methods of delivery (such as, CD-ROM, e-mail, handbook, mail, online resources, or pocket guides, poster) for dementia treatment information and what method would be best for communicating about the treatment of dementia symptoms with family members. Information about the medical director and availability of a psychiatrist or dementia specialist nurse consultant to consult or make a visit to the nursing home was collected. DON demographic information was collected and included age, gender, race, ethnicity, total years in practice, and total years employed at the current nursing home. OSCAR/CASPER data OSCAR is an administrative database of the Centers for Medicare and Medicaid Services (CMS). In July 2012, OSCAR data was replaced by the CASPER system. CASPER is a part of a large relational database operating within CMS Automated Survey Processing Environment (ASPEN). Every nursing home in the United States that is certified to provide services under either Medicare or Medicaid is listed in the survey data. The number of variables in a data file depends on the survey process and extensive data is collected for nursing homes. Data are available for download, but new survey data overwrites the previous data. Archived longitudinal data for this study were purchased from Cowles Research Group so the data was current with the survey.16 OSCAR/CASPER data were linked to each nursing home’s DON responding to the questionnaire through the nursing home’s respective National Provider Identifier (NPI). Thirty-two CMS OSCAR/CASPER variable descriptions were retrieved from their respective forms which included: 1) Long Term Care Facility Application for Medicare and Medicaid, 2) Resident Census and Conditions of Residents, 3) Statement of Deficiencies and Plan of Correction (see Table 1). The research team reviewed the forms and determined which variables to include in the data analysis. Data analysis Using the returned questionnaires, basic descriptive statistics for questions of interest were calculated. For any statistics derived from the questionnaires and reported as proportions, missing responses have been removed from the denominator for calculation; any missing data exceeding 10% for a question has been noted in the results. The analysis explored predictors for two of the questions asked on the questionnaire: 1) the proportion of dementia patients who had challenging behavior treated with non-drug approaches instead of medications (questionnaire takers were given five choices: “less than 1/4,” “less than 1/2 but more than 1/4,” “about 1/2,” “more than 1/2 but less than 3/4,” and “more than 3/4”), 2) the
184
J.M. Daly et al. / Geriatric Nursing 36 (2015) 182e191
Table 1 Potential predictors for models. Source
Variablea
Needs Assessment Survey
-
-
-
-
-
Censusb
-
-
Census e Staffing
b
-
Census e Residents
b
-
-
-
a b
Proportion of dementia patients with challenging behavior treated using non-drug approaches (only in confidence model) Confidence in managing challenging dementia patient behaviors with non-drug approaches (only in proportion of dementia patients model) How often family members are involved in decision to use medication for challenging behaviors in dementia patients Awareness of FDA warnings on antipsychotic risks for patients with dementia Medical director specialty/additional training in geriatrics (dichotomous) Board-certified psychiatrist available by phone for help with medication management (dichotomous) Board-certified psychiatrist available through visits for help with medication management (dichotomous) Nurse or nurse practitioner specializing in dementia/psychiatrics available by phone (dichotomous) Nurse or nurse practitioner specializing in dementia/psychiatrics available through visits (dichotomous) Highest nursing education (continuous and categorical) Medical director years in practice Medical director years employed at facility Consulting nurses are used for information or clinical tools regarding challenging behaviors in patients with dementia (dichotomous) Consulting pharmacists are used for information or clinical tools regarding challenging behaviors in patients with dementia (dichotomous) Consulting drug company representatives are used for information or clinical tools regarding challenging behaviors in patients with dementia (dichotomous) Satisfaction with training of yourself and staff regarding management of challenging behaviors in people with dementia Provider category code (skilled nursing facility (dually certified)/skilled nursing facility (distinct part)/skilled nursing facility/ nursing facility) (categorical) Special care unit with beds for Alzheimer’s disease available (dichotomous) Nurse aide training and competency evaluation program (dichotomous) Medical director FTE Other physicians FTE Physician extenders FTE RN director of nurses FTE Nurses with administrative duties FTE Registered Nurse FTE and HPRD (hours per resident per day) Licensed practical/licensed vocational nurses (LPN/LVN) FTE and HPRD Certified nurse aides FTE and HPRD Nurse aides in training FTE and HPRD Registered Nurses þ LPN/LVN þ certified/in training nurse aides FTE and HPRD Medication aides/technicians FTE Therapeutic recreation specialists FTE Social workers FTE Other social service staff FTE Mental health services FTE Proportion of residents with . - documented psychiatric diagnosis (excludes dementias and depression) - dementia or Alzheimer’s disease - behavioral health care needs - behavioral health care needs and an individualized health care plan Proportion of residents receiving . - psychoactive medication - antipsychotic medication - antianxiety medication - antidepressant medication - hypnotic medication Proportion of residents’ dependent or requiring assistance for . - bathing - dressing - transferring - toilet use - eating
Variables were treated as continuous unless otherwise noted in parentheses. Variables from OSCAR/CASPER.
confidence in managing patients with dementia who had challenging behavior with non-drug approaches (question responses were on a 5-point Likert-like scale from “not at all confident” to “totally confident”). These were referred to as the “Proportion of Dementia Patients” model and the “Confidence” model respectively. Separate models were used for the two response questions. Responses to both questions were treated as categorical and collapsed down to three categories in order to simplify interpretation, suggesting that ordinal logistic regression should be utilized. The Proportion of Dementia Patients outcome was collapsed into three levels: 1) Less than 1/2 (n ¼ 39), 2) Approximately 1/2 (n ¼ 31), and 3) More than 1/2 (n ¼ 83). The Confidence outcome
was collapsed into three levels: 1) Not confident (n ¼ 52), 2) Medium confidence (n ¼ 74), and 3) Confident (n ¼ 32). Univariable analysis was performed with predictors of interest from the questionnaire and OSCAR/CASPER data for each response. Predictors with p-values 0.10 were selected for further analysis in a multivariable model (see Table 2). If there was evidence that the proportional odds assumption of the ordinal logistic regression models was not met (score test), multinomial logistic regression was utilized. Multinomial logistic regression requires a different interpretation for odds ratios than ordinal, as it does not assume proportional odds, but still provided relevant information for the investigation.
J.M. Daly et al. / Geriatric Nursing 36 (2015) 182e191
185
Table 2 Candidate predictors selected through univariable analysis for proportion of dementia patients and confidence outcomes. Variablea Outcome: proportion of patients with dementia treated with non-drug approaches How often family members are involved in decision to use medication for challenging behaviors in dementia patients Confidence in managing challenging dementia patient behavior with non-drug approaches Awareness of FDA warnings on antipsychotic risks in dementia Satisfaction with training of yourself and staff regarding management of challenging behaviors in people with dementia Outcome: confidence in managing challenging behaviors of patients with dementia patient with non-drug approaches How often family members are involved in decision to use medication for challenging behaviors in dementia patients Awareness of FDA warnings on antipsychotic risks in dementia Visits from board-certified psychiatrist (dichotomous) Consulting pharmacists are used for information or clinical tools regarding challenging behaviors in patients with dementia (dichotomous) Satisfaction with current training for managing challenging behaviors in dementia patients Medical director years employed at facility Special care unit with beds for Alzheimer’s disease available (dichotomous) a
Odds ratios
90% CI
p-value
1.44 3.29 1.65 1.40
1.13e1.83 2.25e4.80 1.31e2.08 1.07e1.83
0.01
Contents lists available at ScienceDirect
Geriatric Nursing journal homepage: www.gnjournal.com
Feature Article
Caring for people with dementia and challenging behaviors in nursing homes: A needs assessment geriatric nursing Jeanette M. Daly, RN, PhD a, *, Camden P. Bay, MS b, Barcey T. Levy, PhD, MD a, b, Ryan M. Carnahan, MS, PharmD b a b
Department of Family Medicine, Carver College of Medicine, The University of Iowa, Iowa City, IA, USA Department of Epidemiology, College of Public Health, The University of Iowa, Iowa City, IA, USA
a r t i c l e i n f o
a b s t r a c t
Article history: Received 5 November 2014 Received in revised form 29 December 2014 Accepted 5 January 2015 Available online 9 February 2015
An estimated 50% of nursing home residents have a dementia diagnosis. The purpose of this research was to conduct a needs assessment of directors of nursing (DON) in Iowa nursing homes in relation to caring for patients with Behavioral and Psychological Symptoms of Dementia. DON responses were linked to Online Survey Certification and Reporting/Certification and Survey Provider Enhanced Reporting (OSCAR/CASPER) data to examine how facility characteristics may be associated with use of and confidence in non-drug management strategies. From 431 questionnaires mailed to DONs, 160 (37%) were returned. Regression analysis showed that those who were more confident in managing challenging behavior were more likely to have satisfaction with current training on managing challenging behaviors and had a psychiatrist available to visit the facility. Facilities with a larger proportion of patients with challenging behaviors being treated with non-drug approaches instead of antipsychotics had DONs who were more likely to be confident in non-drug management strategies and have knowledge about the FDA antipsychotic medications risks. Ó 2015 Elsevier Inc. All rights reserved.
Keywords: Nursing home Director of nursing Dementia Challenging behaviors Antipsychotic medication
Currently, there are more than 15,000 Medicare and Medicaidcertified nursing homes in the United States1 with almost 3.3 million patients receiving care in those nursing homes.2 An estimated 50% of nursing home residents have a diagnosis of dementia3,4 with 13% having dementia or Alzheimer’s disease as their primary diagnosis.5 Neuropsychiatric disturbances occur in 60e98% of people with dementia, frequently triggering nursing home placement and antipsychotic treatment.6 Antipsychotics have been a mainstay of pharmacologic management. In 2006, 25.9% of nursing home residents in the United States received antipsychotics,7 compared to 16.4% just 10 years earlier. Recent estimates place the prevalence of nursing home antipsychotic use around 22%.8 Evidence suggests that antipsychotic use is often suboptimal, based on lack of appropriate indications, inappropriate dosing,9,10 and high variability in usage rates across facilities.10 This is of particular concern since antipsychotics are associated with serious side effects in people with dementia, including an increased risk of mortality
Funding: ARHQ R18 HS19355-01. * Corresponding author. 01290-F PFP, 200 Hawkins Drive, Iowa City, IA 52242, USA. Tel.: þ1 319 384 8995. E-mail address: [email protected] (J.M. Daly). 0197-4572/$ e see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.gerinurse.2015.01.001
which was the reason for a black box warning required by the Food and Drug Administration. In 2014 the Centers for Medicare and Medicaid Services (CMS) launched the National Partnership to Improve Dementia Care initiative to enhance dementia care quality and address antipsychotic overuse, with a goal of reducing antipsychotic use in nursing homes by 30% at the end of 2016.11 Regulatory requirements have facilitated a marked improvement in decreasing the use of antipsychotic medications in nursing homes from 24% in the fourth quarter of 2011 to 20% in the first quarter of 2014.11 The current focus of the initiative is to not only reduce antipsychotics but also to encourage person-centered care and non-drug management of behavioral symptoms of dementia in order to reduce unnecessary antipsychotic use.12 Challenging behaviors for persons with dementia are common and multi-faceted being attributable to the environment, chronic conditions, health, or medications. More than 80% of nursing home residents with dementia exhibit one or more forms of challenging behaviors.13 Examples of challenging behaviors include agitation, anger, depression, disrobing, eating abnormalities, hoarding, inappropriate sexual behavior, irritability, paranoia, physical and verbal aggression, repetition, swearing, and withdrawal. There are situations in which antipsychotic use in dementia patients is considered appropriate by most clinicians as well as CMS, particularly when a patient’s behaviors are dangerous to
J.M. Daly et al. / Geriatric Nursing 36 (2015) 182e191
themselves or others.14 Initial therapy for patients with dementia who have challenging behaviors is environmental modifications and non-pharmacologic approaches. However, if that is unsuccessful and antipsychotic use is warranted, an appropriate indication for use of the medication needs to be documented. Once antipsychotic medication is started ongoing monitoring to evaluate its effectiveness and observation for adverse effects is necessary. If the medication is ongoing, the lowest effective dose should be established and maintained. In 2013, the percentage of nursing home residents taking antipsychotic medication ranged from 12.3% in Hawaii to 33.1% in Louisiana with a mean of 24.2% in the U.S. 15 In Iowa the percentage of nursing home residents taking antipsychotic medication was 21.2% in 2012.15 The purpose of this research study was to conduct a needs assessment of directors of nursing (DON) in Iowa nursing homes in relation to caring for patients with dementia and those with challenging behaviors. That is a needs assessment to establish knowledge deficits in which to appropriately care for people with Behavioral and Psychological Symptoms of Dementia (BPSD) and then to best determine how they would like to receive additional training. This needs assessment would help to inform the development of educational materials and tools to help guide care for people with BPSD. In addition, the DON responses were linked to the Online Survey Certification and Reporting/Certification and Survey Provider Enhanced Reporting (“OSCAR/CASPER”) survey and inspection data to provider responses to examine how information on facility operations, patient census and regulatory compliance may be associated with use of and confidence in non-drug management strategies for behavioral symptoms of dementia. Comparison of both resident and facility data allows us to look at the variation and influence of these factors. In particular, we utilized regression modeling to aid in assessing the relationship between the proportion of dementia patients with challenging behavior treated using non-drug approaches and nurse’s knowledge, confidence, and education, family involvement, medical director education, availability of a psychiatrist or nurse practitioner with a dementia/ psychiatric specialty, facility operations, and patient census (see Table 1 for the entire list OSCAR/CASPER Date). Additionally, we wanted to look at nurse’s confidence in managing challenging patient behaviors associated with dementia and non-drug approaches. Methods After institutional review board approval, a mailed survey was sent to Iowa nursing homes. The Medicare nursing home compare website (medicare.gov/NursingHomeCompare) listed 443 nursing homes in Iowa. Of the 443, 12 were hospitals and eliminated from the list because the cover letter read director of nursing and would go to the hospital director and not the nursing home director, leaving 431 nursing homes. All nursing home DONs were invited to participate and questionnaires were mailed in January 2011. DONs were asked to complete the questionnaire while skipping any questions they did not want to answer and mail it back. If no response was given by a DON within three weeks, a second questionnaire was mailed. Following the two mailings, no further mailings or contacts were made. Instrument A 22-item questionnaire (see Appendix 1) entitled, “Needs Assessment: Caring for People with Dementia and Challenging Behaviors” developed by the research team of clinicians, researchers, and education specialists, was used to assess the needs of DONs in regard to caring for individuals with BPSD. To ensure
183
content validity, the questionnaire was critique by current or former nursing home employees that were on the research team. Questions included information pertaining to the treatment of dementia patients and the education of nursing home staff and family members about dementia. To better understand current dementia management approaches, DONs were asked about which antipsychotics were most often prescribed (a list of commonly used antipsychotics was provided), satisfaction with current training on managing dementia behaviors, if the nursing staff provides influence on the selection of the antipsychotic medication prescribed, their awareness of the FDA warning on antipsychotics in dementia, and their usage as well as confidence in non-drug treatment approaches for managing BPSD. To gather information related to the education of nursing home staff and family members of dementia patients, DONs were asked how they obtain information about dementia treatment, what dementia topics interest them the most, preferred methods of delivery (such as, CD-ROM, e-mail, handbook, mail, online resources, or pocket guides, poster) for dementia treatment information and what method would be best for communicating about the treatment of dementia symptoms with family members. Information about the medical director and availability of a psychiatrist or dementia specialist nurse consultant to consult or make a visit to the nursing home was collected. DON demographic information was collected and included age, gender, race, ethnicity, total years in practice, and total years employed at the current nursing home. OSCAR/CASPER data OSCAR is an administrative database of the Centers for Medicare and Medicaid Services (CMS). In July 2012, OSCAR data was replaced by the CASPER system. CASPER is a part of a large relational database operating within CMS Automated Survey Processing Environment (ASPEN). Every nursing home in the United States that is certified to provide services under either Medicare or Medicaid is listed in the survey data. The number of variables in a data file depends on the survey process and extensive data is collected for nursing homes. Data are available for download, but new survey data overwrites the previous data. Archived longitudinal data for this study were purchased from Cowles Research Group so the data was current with the survey.16 OSCAR/CASPER data were linked to each nursing home’s DON responding to the questionnaire through the nursing home’s respective National Provider Identifier (NPI). Thirty-two CMS OSCAR/CASPER variable descriptions were retrieved from their respective forms which included: 1) Long Term Care Facility Application for Medicare and Medicaid, 2) Resident Census and Conditions of Residents, 3) Statement of Deficiencies and Plan of Correction (see Table 1). The research team reviewed the forms and determined which variables to include in the data analysis. Data analysis Using the returned questionnaires, basic descriptive statistics for questions of interest were calculated. For any statistics derived from the questionnaires and reported as proportions, missing responses have been removed from the denominator for calculation; any missing data exceeding 10% for a question has been noted in the results. The analysis explored predictors for two of the questions asked on the questionnaire: 1) the proportion of dementia patients who had challenging behavior treated with non-drug approaches instead of medications (questionnaire takers were given five choices: “less than 1/4,” “less than 1/2 but more than 1/4,” “about 1/2,” “more than 1/2 but less than 3/4,” and “more than 3/4”), 2) the
184
J.M. Daly et al. / Geriatric Nursing 36 (2015) 182e191
Table 1 Potential predictors for models. Source
Variablea
Needs Assessment Survey
-
-
-
-
-
Censusb
-
-
Census e Staffing
b
-
Census e Residents
b
-
-
-
a b
Proportion of dementia patients with challenging behavior treated using non-drug approaches (only in confidence model) Confidence in managing challenging dementia patient behaviors with non-drug approaches (only in proportion of dementia patients model) How often family members are involved in decision to use medication for challenging behaviors in dementia patients Awareness of FDA warnings on antipsychotic risks for patients with dementia Medical director specialty/additional training in geriatrics (dichotomous) Board-certified psychiatrist available by phone for help with medication management (dichotomous) Board-certified psychiatrist available through visits for help with medication management (dichotomous) Nurse or nurse practitioner specializing in dementia/psychiatrics available by phone (dichotomous) Nurse or nurse practitioner specializing in dementia/psychiatrics available through visits (dichotomous) Highest nursing education (continuous and categorical) Medical director years in practice Medical director years employed at facility Consulting nurses are used for information or clinical tools regarding challenging behaviors in patients with dementia (dichotomous) Consulting pharmacists are used for information or clinical tools regarding challenging behaviors in patients with dementia (dichotomous) Consulting drug company representatives are used for information or clinical tools regarding challenging behaviors in patients with dementia (dichotomous) Satisfaction with training of yourself and staff regarding management of challenging behaviors in people with dementia Provider category code (skilled nursing facility (dually certified)/skilled nursing facility (distinct part)/skilled nursing facility/ nursing facility) (categorical) Special care unit with beds for Alzheimer’s disease available (dichotomous) Nurse aide training and competency evaluation program (dichotomous) Medical director FTE Other physicians FTE Physician extenders FTE RN director of nurses FTE Nurses with administrative duties FTE Registered Nurse FTE and HPRD (hours per resident per day) Licensed practical/licensed vocational nurses (LPN/LVN) FTE and HPRD Certified nurse aides FTE and HPRD Nurse aides in training FTE and HPRD Registered Nurses þ LPN/LVN þ certified/in training nurse aides FTE and HPRD Medication aides/technicians FTE Therapeutic recreation specialists FTE Social workers FTE Other social service staff FTE Mental health services FTE Proportion of residents with . - documented psychiatric diagnosis (excludes dementias and depression) - dementia or Alzheimer’s disease - behavioral health care needs - behavioral health care needs and an individualized health care plan Proportion of residents receiving . - psychoactive medication - antipsychotic medication - antianxiety medication - antidepressant medication - hypnotic medication Proportion of residents’ dependent or requiring assistance for . - bathing - dressing - transferring - toilet use - eating
Variables were treated as continuous unless otherwise noted in parentheses. Variables from OSCAR/CASPER.
confidence in managing patients with dementia who had challenging behavior with non-drug approaches (question responses were on a 5-point Likert-like scale from “not at all confident” to “totally confident”). These were referred to as the “Proportion of Dementia Patients” model and the “Confidence” model respectively. Separate models were used for the two response questions. Responses to both questions were treated as categorical and collapsed down to three categories in order to simplify interpretation, suggesting that ordinal logistic regression should be utilized. The Proportion of Dementia Patients outcome was collapsed into three levels: 1) Less than 1/2 (n ¼ 39), 2) Approximately 1/2 (n ¼ 31), and 3) More than 1/2 (n ¼ 83). The Confidence outcome
was collapsed into three levels: 1) Not confident (n ¼ 52), 2) Medium confidence (n ¼ 74), and 3) Confident (n ¼ 32). Univariable analysis was performed with predictors of interest from the questionnaire and OSCAR/CASPER data for each response. Predictors with p-values 0.10 were selected for further analysis in a multivariable model (see Table 2). If there was evidence that the proportional odds assumption of the ordinal logistic regression models was not met (score test), multinomial logistic regression was utilized. Multinomial logistic regression requires a different interpretation for odds ratios than ordinal, as it does not assume proportional odds, but still provided relevant information for the investigation.
J.M. Daly et al. / Geriatric Nursing 36 (2015) 182e191
185
Table 2 Candidate predictors selected through univariable analysis for proportion of dementia patients and confidence outcomes. Variablea Outcome: proportion of patients with dementia treated with non-drug approaches How often family members are involved in decision to use medication for challenging behaviors in dementia patients Confidence in managing challenging dementia patient behavior with non-drug approaches Awareness of FDA warnings on antipsychotic risks in dementia Satisfaction with training of yourself and staff regarding management of challenging behaviors in people with dementia Outcome: confidence in managing challenging behaviors of patients with dementia patient with non-drug approaches How often family members are involved in decision to use medication for challenging behaviors in dementia patients Awareness of FDA warnings on antipsychotic risks in dementia Visits from board-certified psychiatrist (dichotomous) Consulting pharmacists are used for information or clinical tools regarding challenging behaviors in patients with dementia (dichotomous) Satisfaction with current training for managing challenging behaviors in dementia patients Medical director years employed at facility Special care unit with beds for Alzheimer’s disease available (dichotomous) a
Odds ratios
90% CI
p-value
1.44 3.29 1.65 1.40
1.13e1.83 2.25e4.80 1.31e2.08 1.07e1.83
0.01
