by Dr Virginia Painter
Patients with Parkinson’s disease (PD) are more likely to be hospitalised than the general population with a high prevalence of inpatient complications such as falls, delirium, infections, increased length of stay and worsening of motor symptoms1-4. Medication errors are common in these patients with implications for nutrition and physical rehabilitation and correlated with increased frequency of complications and length of stay and worsening of motor symptoms1,5-6.
There is limited literature on the specific management of hospital inpatients with PD for the reduction of these complications. The 2017 NICE Guidelines for Parkinson’s disease in adults provide best practice advice which includes strict adherence to dosage timing including allowing self-medication, adjustment of medications only after discussion with a PD specialist, and avoidance of sudden withdrawal or inefficacy of medication7. The Royal United Hospital, Bath has further recommendations written in consultation with the Movement Disorders Section, British Geriatrics Society (BGS MDS), relating to early referral for speech and language therapy (SLT), physiotherapy (PT) and PD specialist where appropriate.
The first aim of this project was to audit the adherence with the recommendations as mentioned above, for inpatient management of patients with PD during the first twenty-four hours of admission. The second aim was to develop an intervention designed to improve adherence with those recommendations.
Audit Study Population and Data Collection
A pre-intervention snapshot was collected prospectively over a five day period with repeated snapshots similarly collected every five weeks after the intervention was commenced. Criteria for inclusion were any hospital inpatient with a known diagnosis of PD, or currently being assessed for a movement disorder by a PD specialist. Patients were excluded if the current hospital admission had been included in the preceding snapshot. The pro-forma for data was created based on the BGS MDS recommendations and refined based on pilot data initially collected over a one day period. Data was collected from the paper medical record including progress notes, nursing and medical pro-formas, and medication charts.
Developing an Intervention
Following the pre-intervention snapshot, there was a strategic meeting between the co-investigators (myself, PD specialist, PD MNP) to brainstorm ideas, discuss cause and effect, and determine key stakeholders. Based on issues identified in the pre-intervention snapshot, possible solutions were considered using an impact matrix. Meetings were held with the ED/MAU Senior Nurse-in-charge, Frailty Flying Squad MNP (specialised ED/MAU elderly care team) and Physiotherapy department head. The intervention focused on increasing awareness and accessibility of an already available aide-memoire by placing it on the intranet and increasing signage and availability in the ED/MAU. Short 10-15 minute education sessions were commenced for medical and nursing staff with a focus on staff working in the ED, MAU and Elderly Care Wards.
There were twenty patients included in the pre-intervention snapshot which commenced in April 2017. These patients were mostly male (65%), from home (85% vs. 5% warden-controlled vs. 5% residential home vs. 5% nursing home) with an average age of 84 years. They were most commonly categorised as having complex stage PD (60% vs. 20% diagnostic vs. 15% maintenance vs. 5% palliative stage) and 85% were taking levodopa (10% DA, 15% MAOB-I, 10% amantadine, 15% COMT-I). Most patients spent the first twenty-four hours of their admission in the ED (95%) and MAU (90%), but also in the Acute Care of the Elderly (ACE) Unit (10%), Specialist PD ward (5%) and general surgical wards (10%).
In 80-85% of patients, usual drug information was collected correctly, prescribed correctly and the time of the last dose taken checked. The first dose was given within 30 minutes of the prescribed time in 55% of patients. Dopamine blockers were not prescribed in 100% of patients. Swallowing ability was considered in 80% of patients with SLT referral made in 100% of patients where it was deemed appropriate. An alternative route was prescribed in 100% of patients made nil by mouth. Referral to PT was made in 45% of patients where it was deemed appropriate with Day 4-5 being the average first day seen by a PT. Referral to a PD specialist was made in 67% of patients where it was indicated.
Ongoing Snapshots after commencing the Intervention
There have been fifty patients included since commencement of the intervention with data collected over four separate snapshot periods. These patients were mostly female (58%), from home (76% vs. 4% warden-controlled vs. 12% residential home vs. 8% nursing home) with an average age of 82 years. They were most commonly categorised as having complex stage PD (62% vs. 8% diagnostic vs. 30% maintenance) and 85% were taking levodopa (12% DA, 4% MAOB-I, 12% COMT-I).
Regarding medication prescription and administration, there has been minimal change with 74-78% of patients having usual drug information was collected correctly, prescribed correctly and the time of the last dose taken checked. The first dose was given within 30 minutes of the prescribed time in 58% of patients. Adherence with swallowing recommendations deteriorated with swallowing ability considered in 72% of patients, SLT referral made in only 60% of patients where it was deemed appropriate, and an alternative route prescribed in 77% of patients made nil by mouth. Referral to PT improved to 62% of patients where it was deemed appropriate with Day 2-3 being the average first day seen by a PT. Referral to a PD specialist was similar and made in 63% of patients where it was indicated.
This audit demonstrated varying levels of adherence to hospital recommendations for the inpatient management of patients with PD during the first 24 hours of admission with minimal change during the initial months of implementing an intervention. Key concerns identified by the pre-intervention snapshot included providing the first dose on time, and early referral to PT and PD specialist where appropriate, however early referral to SLT where appropriate was subsequently identified as an area of poor adherence in later snapshots.
Many key recommendations relating to medication prescription and adherence had better adherence compared with studies in the literature. Providing the first dose on time may be worse comparatively due to our study definition of a delay greater than thirty minutes which is strict. Referrals to PT and SLT may be underreported due to the common verbal or whiteboard meeting referral utilised in the ED/MAU/ACE which may not be reflected in the written clinical notes. More specific to the local system, PT is only available to geriatric patients in ED/MAU when planned for imminent discharge and bed block prevents these patients from being moved to a ward where PT is readily available, thus reducing the referral rate.
Overall, information from the audit is difficult to interpret in the context of small sample sizes. This may be limited by the prospective method of case finding based on a manual physical search of the hospital, leading to possible missing cases. The pre-intervention snapshot would benefit from further retrospective data collection to better inform areas of concern with poor adherence and develop more specific interventions. Subsequent sample sizes required data collation which makes analysis of the progress of the intervention difficult and ideally would not continue for future snapshot collections. There is a reduced ability to focus the intervention on different ward locations as data cannot be sub-divided.
Learning and reflection
This project has provided a wonderful opportunity to gain theoretical knowledge and practical skills for initiating and conducting a quality improvement project. Whilst it remains a work in progress with minimal change in outcomes thus far, it has been a useful exercise in fostering a project that makes one appreciate the difficulties associated with regular commitment, consistent implementation over the long term whilst balancing clinical responsibilities. It has demonstrated to me the difficulties of adherence in the context of systems issues such as the PT staff shortage and resource allocation that is difficult to modify within the scope of my project.
Regular checks of the ED/MAU for the signage and availability of pro-formas were important initially, however, they are difficult in the context of a haphazard and busy work schedule and certainly not sustainable in the long term. My nursing co-investigator has been paramount in showing me the importance of “buy-in” and commitment from key stakeholders to change practice through her continued attempts in education sessions to place greater initiative on the nursing staff in the ED/MAU to be vested in adherence with the recommendations. This is especially relevant in the context of form fatigue.
Going forward, this intervention would benefit from involvement of SLT stakeholders as well as the plan to continue the education sessions. Retrospective data collection would be useful in case the focus of the intervention has been misplaced. Furthermore, it has inspired me to examine our management of inpatients with PD when I return to my usual place of practice in Australia next year.
- Derry et al, Postgraduate Medical Journal (2010), Medication management in people with Parkinson’s disease during surgical admissions, 86, 334-337.
- Gerlach et al, BMC Neurology (2012), Deterioration of Parkinson’s disease during hospitalisation: Survey of 684 patients, 12:13.
- Gerlach et al, Parkinsonism and Related Disorders (2013), Motor outcomes during hospitalization in Parkinson’s disease patients: A prospective study, 19, 737-741.
- Mueller et al, Langenbeck’s Archives of Surgery (2009), Parkinson’s disease influences the perioperative risk profile in surgery, 394, 511-515.
- Magdalinou et al, Parkinsonism and Related Disorders (2007), Prescribing medications in Parkinson’s disease (PD) patients during acute admissions to a District General Hospital, 13, 539-540.
- Martinez-Ramirez et al, PLoS One (2015), Missing dosages and neuroleptic usage may prolong length of stay in hospitalised Parkinson’s disease patients, 10,
- NICE, Parkinson’s disease in adults (2017),