The number of people aged 65 or over, living in Counties Manukau District Health Board (DHB), is expected to increase by 132% from 2006 to 2026.1 Over the same time, across New Zealand, the number aged 85 or over is expected to treble,2 placing significant pressure on future health services.3 4 Only 15% of people over 85 in NZ remain independent of support services.5 6 Without any other changes, the number of older people in residential care would double by 2021.5 The trend is for older service users with higher needs having shorter stays.7Counties Manukau DHB has made a commitment to ensure that older people in ARRC facilities have the same access to geriatric services as those living in their own home. The DHB elderly service had noted considerable variability between rest homes in their catchment area in the number and reasons for hospital admissions. Anecdotally, they were aware that the ARRC system was under duress and it appeared that many nurses and doctors were working in the system with little support. The Health and Disability Commissioner has stated that rest home care has been a disproportionate source of complaints.8The Community Geriatric Service (CGS) is a new initiative to provide a consultative service for general practitioners (GPs) and nurses working in ARRC facilities. The service includes a community geriatrician, two Clinical Nurse Specialists (CNS) and a social worker. A multi-component support strategy was devised following discussions with stakeholders and a survey of relevant literature.9-11 The strategy included elements that have individually been shown to be successful, but we are not aware of such a combination being used previously either in New Zealand or internationally. One of the large rest homes agreed to support this development and trial the new system.Methods The settingThe facility comprised a hospital with 50 residents and a rest home with 46 residents. It is owned by a local subsidiary of a large international organisation which has a focus on aged residential care. The hospital was staffed by a unit coordinator who was a registered nurse, four registered nurses, two enrolled nurses and 28-30 caregivers. The rest home was staffed by a unit co-ordinator who was an enrolled nurse, another enrolled nurse and 12 caregivers. A nurse manager was responsible for the overall operation of the facility. One GP (the GP) provided primary care for 90% of the residents. Primary care for the remainder was shared across five GPs. The interventionThe intervention consisted of five main components delivered in an intensive phase from December 2007 to May 2008, followed by an ongoing maintenance phase. The specialist staff consisted of one geriatrician and one clinical nurse specialist (CNS) who did not consult directly with patientsthey worked with the front-line caregivers to discuss, train, mentor and support. Medication reviews were conducted for all residents age 85 or more, and for younger residents on 9 or more medications. The review team included the geriatrician, a CNS, the GP, a community pharmacist (who participated by teleconference), the facility clinical manager and the two unit coordinators. During the intensive phase the review was held weekly at the rest home and was led by the geriatrician. Since then meetings have continued monthly and are led by the GP. The geriatrician and CNS read the residents medical record prior to the review. At times this review naturally extended beyond medications to include a fuller clinical discussion. Medication changes were based on the Beers criteria.10 The geriatrician wrote detailed notes following the review, which were faxed to the facility within a week. Changes to medications and progress chart were made by the GP. Facility nurses and caregivers were informed of changes at each shift change. Changes were discussed with the family wherever possible. Two telephone hotlines were established. The first provided registered nurses with advice from a CNS, and the second gave GPs direct access to the geriatrician. The hours of service were 0830 to 1530. A template was provided to guide clinical information collection prior to using the hotline. The nursing hotline was not intended to address acute problemsthe advice given in such cases was to contact the GP or, if needed, arrange acute admission to hospital. Advanced Nursing Support was provided on-site by a CNS. This included review of residents who were complex and the direct care staff required advice. From these reviews, on-site education was provided by the CNS. The nurses were also offered a web and CD course Assessment Treatment and Rehabilitation Advanced Core Training (ATRACT), which was available to all nurses working in the DHB catchment area.12 Both nurses and GPs were encouraged to use a Counties Manukau DHB scheme called Primary Options for Acute Care (POAC) that gives access to additional resources short term if they could keep a patient out of Middlemore Hospital. This scheme funded a CNS to train all the registered nurses in intravenous fluid administration. Nurses and GPs were also encouraged and supported to enrol patients in the DHB Chronic Care Management (CCM) programme. This programme is intended to provide systematic case management free to patients with congestive heart failure, diabetes, chronic obstructive pulmonary disease, cardiovascular disease and depression. It has been largely unavailable to residents in ARRC facilities as it depends on specific information technologies and is relatively complex and time consuming. Nurses and GPs were offered training to initiate and support a formal process of Advance Care Planning (ACP) that was undertaken by a project manager who had a background as a social worker. ACP supports patients and their family/whanau to think ahead to the care and medical treatment one would desire to receive in the future. While not binding on health care providers, a Plan should be taken into account if later treatment decisions are made when the person is not competent to discuss and consent. The process used was adapted from an Australian and United States model.13 Data collection-direct observationTwo authors (AA, HC) directly observed the medication reviews and made field notes. Data collection-interviewsInterview schedules were constructed following a review of the stated programme objectives and discussion with the programme developers, senior managers at the facility and members of the evaluation team. Two interviewsone before and one after the intensive phase of the interventionwere held with each of the senior management team at the rest home and hospital, the geriatrician, the GP and the pharmacist. Interviews were also held with two CNSs and two DHB elderly services managers; the registered nurses who had been involved in medication reviews and with a convenience sample of enrolled nurses and caregivers at the facility. Interviews before the intensive phase were conducted in November 2007, and subsequent interviews were in June to August 2008. Interviews were conducted at a time and location of the interviewees choice. Interviews were recorded and transcribed. Transcripts were sent to interviewees for verification. All transcripts were independently coded for themes by three authors (AA, VA, HC) using a general inductive approach.14 The original proposal was to interview residents or their families. However, these interviews were not held as senior management of the facility decided that few residents aged over 85 years would be able to remember whether changes in their medication, made several months prior to an interview, had affected their health. It was also considered that the families of the residents would not have sufficient knowledge of the details of the programme to enable valid information to be collected. Data collection-admissions to hospital and hotlinesQuantitative data were collected on medication changes, hotline use and admissions to hospital. The Community Geriatric Service hold a weekly case conference during which they assess and record the appropriateness of each hospital admission of rest home residents. This is a clinical judgement which may differ from the formal discharge ICD-10 codes used to assess Ambulatory Sensitive Hospitalisation. The hospital database was queried for the time of admission (i.e. week day, night or weekend); if accompanied by a referral note; and length of stay. The geriatrician kept a record of hotline calls and recorded a judgement about whether an admission was avoided by a hotline call. EthicsPatients and their families were all given leaflets explaining the project prior to it starting and verbal consent was obtained in all instances from either the patient or their legal representative. Staff and DHB interviewees provided written consent. The research was approved by the Northern Regional Ethics Committee NTY/08/05/043. Results Four medication reviews were directly observed. Sessions lasted about one hour and each reviewed about five patients. Twenty two informants contributed 43 interviews. Medication reviewsSixty-four residents had their medications reviewed (56 aged over 85 and 8 aged 50\u201385). Across these residents 84 different medications were being prescribed at the start of the study, a total of 466 medications to the 64 residents, reduced to 366 after the reviews (a reduction of 21%). For 50 residents at least one medication was stopped, and for a further 8 at least one medication was reduced; overall 54 different medications were stopped. Table 1 shows the top ten prescriptions at the start of the intervention and after the medication reviews. Seventeen residents had at least one medication started, the most common being paracetamol (7), calcium carbonate (3) and cholecalciferol (3). Table 1. Medications prior to intervention, and those stopped or reduced after review Top 10 medications prescribed at start of intervention. (Number of residents) Top 10 individual drugs stopped or reduced. (Number of residents) Stopped medications that can lower blood pressure and that can impair cognition Cholecalciferol (44) Calcium carbonate (36) Furosemide (30) Omeprazole (27) Aspirin (29) Metoprolol (17) Ducosate (16) Simvastatin (14) Paracetamol (11) Alendronate (11) Calcium carbonate (15) Simvastatin (11) Omeprazole (10) Multi-vitamins (7) Metoprolol (6) Furosemide (6) Aspirin (6) Enalapril (5) Paracetamol-codeine (5) Paracetamol-dextropropoxyphene (5) Lower blood pressure (beta-blockers, calcium blockers, alpha-blockers, ACE inhibitors, diuretics) 32 medications 25 people Impair consciousness (tricyclics, neuroleptics, hypnotics, sodium valproate, codeine, dextropropoxyphene) 17 medications 16 people Medication delivery time was reduced. My drug round takes less time. I am not giving out as many meds. It used to take one and a half hours now takes one hour (CG #4) Reduced delivery time saved resources for the facility; however the pharmacy supplying the facility was concerned that reducing medication would also reduced their revenue. Medications stopped or reduced could be grouped into those that might lower blood pressure and those that might impair cognitionalso shown in Table 1. All staff at the facility reported physical and mental improvements in residents, which they attributed to reduced medication. Reducing over-medication has reduced the risk of falls. (CG #3) They used to be sleepy and confused and that doesnt happen so much now. (CG #5) The nurses and caregivers valued the timely and detailed notes both for individual patient care and more generic education. Having full notes in the patients files with explanations for medication and care changes mean we can understand the reasons for the change. (CG #5) One unanticipated change was observed by a senior caregiver. While we are getting the medications better reviewed, there is also more of a personal interest in the resident. The doctors have moved to talking with the residents rather than talking around them. (CG #3) However, time and availability of the GP remained an issue. ...there are usually patients that I have admitted to the rest home in the last month and I have not had time to review the case notes before they are presented to me at the case review (GP) Admission Rates from the facility to Middlemore HospitalKey informants suggested that, prior to the intervention, some inappropriate or ad hoc admissions to Middlemore Hospital were contributed to by limitations of knowledge, experience and continuity of nursing staff. Problems arise when the condition of the patient is outside the ability or experience of the nurse. This commonly occurs at nights or weekends where the nursing staff may have less experience or training than the nurses who are on duty during week days. There are a number of nurses who are foreign trained who have less experience. (KI #10) Managers, nurses and caregivers believed that, as a result of the intervention, residents were less likely to be referred to secondary care. panic send-offs have stopped (KI #10) we have probably had less necessity to make emergency calls after hours because the whole programme has made our residents more stable (KI #6) Actual admission numbers are shown in Table 2. Changes are not statistically significant. Table 2. Acute admissions to Middlemore hospital during 6 months periods before and during the intervention then during the maintenance phase Variables Jun-Nov 07 Dec 07\u2013May 08 Jun 08\u2013Nov 08 Admissions (patients) 34 (26) 25 (21) 33 (29) Formal referral 1 8 23 Unnecessary admission* (week, afterhours) 1, 3 0, 2 Not assessed Patient days in hospital; total, median, (inter-quartile range)** 227, 4 (1\u20139) 181, 3 (1\u20138) 301, 5 (2\u20139) Note: No attempt is made to adjust for seasonal variation. *Admission judged to be unnecessary by consensus during weekly community geriatric service case conference. Examples include uncomplicated falls. ** t-test before versus during p = 0.81; during versus after p = 0.61. Education and trainingWeekly in-house education sessions were offered, principally intended for the RNs. However, the sessions were offered at times when the RNs were unable to attend due to other work commitments including the medication reviews. The limited number of RNs meant that it was not generally possible to roster some for clinical duties while releasing others for education sessions. One of the things that has highlighted for us is that we shouldnt be trying to do the whole thing at the same time at the same place. (KI #9) Nevertheless, these sessions were highly valued by those who did attend, which was mainly the caregivers. As a result, these care givers felt more valued by the organisation and said that their behaviour had changed as they had more information and were now aware of the reasons for particular tasks. Care givers are more aware of necessity to weigh, take blood pressure regularly etc. (KI #3) The intended individually tailored formal education programmes (ATRACT) for the registered nurses did not materialise. The nurses had neither protected time nor access to a computer during working hours, and did not access it out of working hours. I have visited the [web] site and it seems very helpful. But it needs time to read it all. (RN #2) Primary Options for Acute care (POAC)All registered nurses at the facility were trained by a CNS to administer intravenous therapy, although the training was completed late in the intervention. By September 2008 only one patient had been so treated. In part this may be because cases were relatively infrequent, but also because the necessary decisions were outside the scope of nursing practice and at times they had difficulty getting a doctor to visit after hours. Chronic Care Management programme (CCM)The GP provided a dedicated laptop computer and one of his practice nurses commenced enrolling residents into CCM. Early indications are that both rest home staff and patients have found the process educational and likely to improve care. An action plan is left in the residents room for family and rest home staff with correct response for chest pains for a patient with heart disease. (KI #11) HotlinesThe unit coordinators and clinical manager used the hotlines two to three times a week during the intervention and in the following 6 months. They were enthusiastic about the process and wanted the hours extended. it is good to have that line of advice... Sometimes you are sort of just guessing yourself or trying to do the best you can...whereas you have back-up there now. (KI #7) None of the RNs used the nurse hotlinethey were encouraged to take advice from someone more senior within their facility. No. I have not got to the point of using it. I always go to a senior. (RN #3) No. Not me personally. If I have a problem I take it to the senior manager and then it goes to the unit co-ordinator then the clinical manager. (RN #1) The GP did not use the hotline. Advance Care Planning (ACP)All nurses at the facility, but no GPs, received training. During the 6 month intensive intervention, no ACPs were completed. The introduction of ACPs was delayed by the need for a legal review of the documents, which were developed and introduced relatively late in the intervention. By this time implementation was further challenged by outbreaks of illness in the residents and a building project at the hospital. Staff were concerned that encouraging residents to sign a document regarding future care might undermine their sense of security with the quality of care the staff were providing. The time taken to work with a resident to complete a plan also concerned staff. In addition, senior management suggested that the cognitive level of the residents was too low, a perception that differed from that of the social worker and the specialist geriatrician.\

The number of people aged 65 or over, living in Counties Manukau District Health Board (DHB), is expected to increase by 132% from 2006 to 2026.1 Over the same time, across New Zealand, the number aged 85 or over is expected to treble,2 placing significant pressure on future health services.3 4 Only 15% of people over 85 in NZ remain independent of support services.5 6 Without any other changes, the number of older people in residential care would double by 2021.5 The trend is for older service users with higher needs having shorter stays.7Counties Manukau DHB has made a commitment to ensure that older people in ARRC facilities have the same access to geriatric services as those living in their own home. The DHB elderly service had noted considerable variability between rest homes in their catchment area in the number and reasons for hospital admissions. Anecdotally, they were aware that the ARRC system was under duress and it appeared that many nurses and doctors were working in the system with little support. The Health and Disability Commissioner has stated that rest home care has been a disproportionate source of complaints.8The Community Geriatric Service (CGS) is a new initiative to provide a consultative service for general practitioners (GPs) and nurses working in ARRC facilities. The service includes a community geriatrician, two Clinical Nurse Specialists (CNS) and a social worker. A multi-component support strategy was devised following discussions with stakeholders and a survey of relevant literature.9-11 The strategy included elements that have individually been shown to be successful, but we are not aware of such a combination being used previously either in New Zealand or internationally. One of the large rest homes agreed to support this development and trial the new system.Methods The settingThe facility comprised a hospital with 50 residents and a rest home with 46 residents. It is owned by a local subsidiary of a large international organisation which has a focus on aged residential care. The hospital was staffed by a unit coordinator who was a registered nurse, four registered nurses, two enrolled nurses and 28-30 caregivers. The rest home was staffed by a unit co-ordinator who was an enrolled nurse, another enrolled nurse and 12 caregivers. A nurse manager was responsible for the overall operation of the facility. One GP (the GP) provided primary care for 90% of the residents. Primary care for the remainder was shared across five GPs. The interventionThe intervention consisted of five main components delivered in an intensive phase from December 2007 to May 2008, followed by an ongoing maintenance phase. The specialist staff consisted of one geriatrician and one clinical nurse specialist (CNS) who did not consult directly with patientsthey worked with the front-line caregivers to discuss, train, mentor and support. Medication reviews were conducted for all residents age 85 or more, and for younger residents on 9 or more medications. The review team included the geriatrician, a CNS, the GP, a community pharmacist (who participated by teleconference), the facility clinical manager and the two unit coordinators. During the intensive phase the review was held weekly at the rest home and was led by the geriatrician. Since then meetings have continued monthly and are led by the GP. The geriatrician and CNS read the residents medical record prior to the review. At times this review naturally extended beyond medications to include a fuller clinical discussion. Medication changes were based on the Beers criteria.10 The geriatrician wrote detailed notes following the review, which were faxed to the facility within a week. Changes to medications and progress chart were made by the GP. Facility nurses and caregivers were informed of changes at each shift change. Changes were discussed with the family wherever possible. Two telephone hotlines were established. The first provided registered nurses with advice from a CNS, and the second gave GPs direct access to the geriatrician. The hours of service were 0830 to 1530. A template was provided to guide clinical information collection prior to using the hotline. The nursing hotline was not intended to address acute problemsthe advice given in such cases was to contact the GP or, if needed, arrange acute admission to hospital. Advanced Nursing Support was provided on-site by a CNS. This included review of residents who were complex and the direct care staff required advice. From these reviews, on-site education was provided by the CNS. The nurses were also offered a web and CD course Assessment Treatment and Rehabilitation Advanced Core Training (ATRACT), which was available to all nurses working in the DHB catchment area.12 Both nurses and GPs were encouraged to use a Counties Manukau DHB scheme called Primary Options for Acute Care (POAC) that gives access to additional resources short term if they could keep a patient out of Middlemore Hospital. This scheme funded a CNS to train all the registered nurses in intravenous fluid administration. Nurses and GPs were also encouraged and supported to enrol patients in the DHB Chronic Care Management (CCM) programme. This programme is intended to provide systematic case management free to patients with congestive heart failure, diabetes, chronic obstructive pulmonary disease, cardiovascular disease and depression. It has been largely unavailable to residents in ARRC facilities as it depends on specific information technologies and is relatively complex and time consuming. Nurses and GPs were offered training to initiate and support a formal process of Advance Care Planning (ACP) that was undertaken by a project manager who had a background as a social worker. ACP supports patients and their family/whanau to think ahead to the care and medical treatment one would desire to receive in the future. While not binding on health care providers, a Plan should be taken into account if later treatment decisions are made when the person is not competent to discuss and consent. The process used was adapted from an Australian and United States model.13 Data collection-direct observationTwo authors (AA, HC) directly observed the medication reviews and made field notes. Data collection-interviewsInterview schedules were constructed following a review of the stated programme objectives and discussion with the programme developers, senior managers at the facility and members of the evaluation team. Two interviewsone before and one after the intensive phase of the interventionwere held with each of the senior management team at the rest home and hospital, the geriatrician, the GP and the pharmacist. Interviews were also held with two CNSs and two DHB elderly services managers; the registered nurses who had been involved in medication reviews and with a convenience sample of enrolled nurses and caregivers at the facility. Interviews before the intensive phase were conducted in November 2007, and subsequent interviews were in June to August 2008. Interviews were conducted at a time and location of the interviewees choice. Interviews were recorded and transcribed. Transcripts were sent to interviewees for verification. All transcripts were independently coded for themes by three authors (AA, VA, HC) using a general inductive approach.14 The original proposal was to interview residents or their families. However, these interviews were not held as senior management of the facility decided that few residents aged over 85 years would be able to remember whether changes in their medication, made several months prior to an interview, had affected their health. It was also considered that the families of the residents would not have sufficient knowledge of the details of the programme to enable valid information to be collected. Data collection-admissions to hospital and hotlinesQuantitative data were collected on medication changes, hotline use and admissions to hospital. The Community Geriatric Service hold a weekly case conference during which they assess and record the appropriateness of each hospital admission of rest home residents. This is a clinical judgement which may differ from the formal discharge ICD-10 codes used to assess Ambulatory Sensitive Hospitalisation. The hospital database was queried for the time of admission (i.e. week day, night or weekend); if accompanied by a referral note; and length of stay. The geriatrician kept a record of hotline calls and recorded a judgement about whether an admission was avoided by a hotline call. EthicsPatients and their families were all given leaflets explaining the project prior to it starting and verbal consent was obtained in all instances from either the patient or their legal representative. Staff and DHB interviewees provided written consent. The research was approved by the Northern Regional Ethics Committee NTY/08/05/043. Results Four medication reviews were directly observed. Sessions lasted about one hour and each reviewed about five patients. Twenty two informants contributed 43 interviews. Medication reviewsSixty-four residents had their medications reviewed (56 aged over 85 and 8 aged 50\u201385). Across these residents 84 different medications were being prescribed at the start of the study, a total of 466 medications to the 64 residents, reduced to 366 after the reviews (a reduction of 21%). For 50 residents at least one medication was stopped, and for a further 8 at least one medication was reduced; overall 54 different medications were stopped. Table 1 shows the top ten prescriptions at the start of the intervention and after the medication reviews. Seventeen residents had at least one medication started, the most common being paracetamol (7), calcium carbonate (3) and cholecalciferol (3). Table 1. Medications prior to intervention, and those stopped or reduced after review Top 10 medications prescribed at start of intervention. (Number of residents) Top 10 individual drugs stopped or reduced. (Number of residents) Stopped medications that can lower blood pressure and that can impair cognition Cholecalciferol (44) Calcium carbonate (36) Furosemide (30) Omeprazole (27) Aspirin (29) Metoprolol (17) Ducosate (16) Simvastatin (14) Paracetamol (11) Alendronate (11) Calcium carbonate (15) Simvastatin (11) Omeprazole (10) Multi-vitamins (7) Metoprolol (6) Furosemide (6) Aspirin (6) Enalapril (5) Paracetamol-codeine (5) Paracetamol-dextropropoxyphene (5) Lower blood pressure (beta-blockers, calcium blockers, alpha-blockers, ACE inhibitors, diuretics) 32 medications 25 people Impair consciousness (tricyclics, neuroleptics, hypnotics, sodium valproate, codeine, dextropropoxyphene) 17 medications 16 people Medication delivery time was reduced. My drug round takes less time. I am not giving out as many meds. It used to take one and a half hours now takes one hour (CG #4) Reduced delivery time saved resources for the facility; however the pharmacy supplying the facility was concerned that reducing medication would also reduced their revenue. Medications stopped or reduced could be grouped into those that might lower blood pressure and those that might impair cognitionalso shown in Table 1. All staff at the facility reported physical and mental improvements in residents, which they attributed to reduced medication. Reducing over-medication has reduced the risk of falls. (CG #3) They used to be sleepy and confused and that doesnt happen so much now. (CG #5) The nurses and caregivers valued the timely and detailed notes both for individual patient care and more generic education. Having full notes in the patients files with explanations for medication and care changes mean we can understand the reasons for the change. (CG #5) One unanticipated change was observed by a senior caregiver. While we are getting the medications better reviewed, there is also more of a personal interest in the resident. The doctors have moved to talking with the residents rather than talking around them. (CG #3) However, time and availability of the GP remained an issue. ...there are usually patients that I have admitted to the rest home in the last month and I have not had time to review the case notes before they are presented to me at the case review (GP) Admission Rates from the facility to Middlemore HospitalKey informants suggested that, prior to the intervention, some inappropriate or ad hoc admissions to Middlemore Hospital were contributed to by limitations of knowledge, experience and continuity of nursing staff. Problems arise when the condition of the patient is outside the ability or experience of the nurse. This commonly occurs at nights or weekends where the nursing staff may have less experience or training than the nurses who are on duty during week days. There are a number of nurses who are foreign trained who have less experience. (KI #10) Managers, nurses and caregivers believed that, as a result of the intervention, residents were less likely to be referred to secondary care. panic send-offs have stopped (KI #10) we have probably had less necessity to make emergency calls after hours because the whole programme has made our residents more stable (KI #6) Actual admission numbers are shown in Table 2. Changes are not statistically significant. Table 2. Acute admissions to Middlemore hospital during 6 months periods before and during the intervention then during the maintenance phase Variables Jun-Nov 07 Dec 07\u2013May 08 Jun 08\u2013Nov 08 Admissions (patients) 34 (26) 25 (21) 33 (29) Formal referral 1 8 23 Unnecessary admission* (week, afterhours) 1, 3 0, 2 Not assessed Patient days in hospital; total, median, (inter-quartile range)** 227, 4 (1\u20139) 181, 3 (1\u20138) 301, 5 (2\u20139) Note: No attempt is made to adjust for seasonal variation. *Admission judged to be unnecessary by consensus during weekly community geriatric service case conference. Examples include uncomplicated falls. ** t-test before versus during p = 0.81; during versus after p = 0.61. Education and trainingWeekly in-house education sessions were offered, principally intended for the RNs. However, the sessions were offered at times when the RNs were unable to attend due to other work commitments including the medication reviews. The limited number of RNs meant that it was not generally possible to roster some for clinical duties while releasing others for education sessions. One of the things that has highlighted for us is that we shouldnt be trying to do the whole thing at the same time at the same place. (KI #9) Nevertheless, these sessions were highly valued by those who did attend, which was mainly the caregivers. As a result, these care givers felt more valued by the organisation and said that their behaviour had changed as they had more information and were now aware of the reasons for particular tasks. Care givers are more aware of necessity to weigh, take blood pressure regularly etc. (KI #3) The intended individually tailored formal education programmes (ATRACT) for the registered nurses did not materialise. The nurses had neither protected time nor access to a computer during working hours, and did not access it out of working hours. I have visited the [web] site and it seems very helpful. But it needs time to read it all. (RN #2) Primary Options for Acute care (POAC)All registered nurses at the facility were trained by a CNS to administer intravenous therapy, although the training was completed late in the intervention. By September 2008 only one patient had been so treated. In part this may be because cases were relatively infrequent, but also because the necessary decisions were outside the scope of nursing practice and at times they had difficulty getting a doctor to visit after hours. Chronic Care Management programme (CCM)The GP provided a dedicated laptop computer and one of his practice nurses commenced enrolling residents into CCM. Early indications are that both rest home staff and patients have found the process educational and likely to improve care. An action plan is left in the residents room for family and rest home staff with correct response for chest pains for a patient with heart disease. (KI #11) HotlinesThe unit coordinators and clinical manager used the hotlines two to three times a week during the intervention and in the following 6 months. They were enthusiastic about the process and wanted the hours extended. it is good to have that line of advice... Sometimes you are sort of just guessing yourself or trying to do the best you can...whereas you have back-up there now. (KI #7) None of the RNs used the nurse hotlinethey were encouraged to take advice from someone more senior within their facility. No. I have not got to the point of using it. I always go to a senior. (RN #3) No. Not me personally. If I have a problem I take it to the senior manager and then it goes to the unit co-ordinator then the clinical manager. (RN #1) The GP did not use the hotline. Advance Care Planning (ACP)All nurses at the facility, but no GPs, received training. During the 6 month intensive intervention, no ACPs were completed. The introduction of ACPs was delayed by the need for a legal review of the documents, which were developed and introduced relatively late in the intervention. By this time implementation was further challenged by outbreaks of illness in the residents and a building project at the hospital. Staff were concerned that encouraging residents to sign a document regarding future care might undermine their sense of security with the quality of care the staff were providing. The time taken to work with a resident to complete a plan also concerned staff. In addition, senior management suggested that the cognitive level of the residents was too low, a perception that differed from that of the social worker and the specialist geriatrician.\