Article Text
PDF
Abstract
Objectives Difficulties accessing medicines in the terminal phase hamper the ability of patients to die at home. The aim of this study was to identify changes in community access to medicines for managing symptoms in the terminal phase throughout South Australia (SA), following the development of a ‘Core Medicines List’ (the List) while exploring factors predictive of pharmacies carrying a broad range of useful medicines.
Methods In 2015, SA community pharmacies were invited to participate in a repeat survey exploring the availability of specific medicines. Comparisons were made between 2012 and 2015. A ‘preparedness score’ was calculated for each pharmacy, scoring 1 point for each medicine held from the following 5 classes: opioid, benzodiazepine, antiemetic, anticholinergic and antipsychotic.
Results The proportion of pharmacies carrying all items from the List rose from 7% in 2012 to 18% in 2015 (p=0.01). Multiple linear regression demonstrated that a monthly online newsletter subscription (p=0.04) and provision of a clinical service to aged care facilities (p=0.02) were predictors of pharmacies carrying all items on the List. Furthermore, multiple linear regression demonstrated that the provision of an afterhours service (p=0.02) and clinical services to aged care facilities (p=0.04) were predictors of pharmacies with a high ‘preparedness score’. In responding to issues with supply of medicines at end of life, respondents were more likely to contact the prescriber if aware of palliative patients (p=0.03).
Conclusions These results suggest that there is value in developing and promoting a standardised list of medicines, ensuring that community palliative patients have timely access to medicines in the terminal phase.
- Formularies
- Quality Improvement
- Ambulatory Care
- Terminal care
- General Practitioners
- Pharmacists
Statistics from Altmetric.com
Background
Coordinated multidisciplinary community-based models of palliative care can support caregivers and significantly increase the proportion of people able to die at home.1 ,2 Enablers for home-based care at the end of life include support from their general practitioner (GP), training of caregivers and access to symptom control measures, including medications.3–6 However, research has shown access to medicines, through community pharmacies, can be difficult—hampering efforts to support patients to die at home.7 Furthermore, community palliative care teams may not recognise the important role of community pharmacists as part of the community multidisciplinary team.8 Yet involvement of community pharmacists, to facilitate access to medicines, makes good sense for three reasons.
First, South Australia (SA), with a population of ∼1.7 million,9 has around 450 community pharmacies. Geographically, community pharmacies are accessible and eager to assist palliative care patients and their caregivers.8
On a policy level, the National Medicines Policy and Palliative Care Strategy have recognised the need for greater access to medicines at the end of life and thus improved the community availability of medicines through the development of a palliative care section within the Australian Pharmaceutical Benefits Scheme (PBS).10
Finally, the likely symptoms encountered in the terminal phase (fatigue, pain, nausea, dyspnoea, noisy breathing and delirium) are well recognised, along with the range of pharmacological strategies to manage them, increasing the likelihood of pharmacies to be able to anticipate what stock to carry.11
Despite convenient access to community pharmacies, a strong policy platform and a good understanding of which medicines are required to manage symptoms, barriers still hamper timely access to medicines in the terminal phase.
Australia is not alone with this approach. In Scotland, work has been conducted to better engage community pharmacies in providing stock for community-based patients at the end of life.12 Their Community Pharmacy Palliative Care Network promotes a select group of pharmacies—via the internet—that stock a broad list of medicines useful in managing the terminal phase. There are concerns, however, about maintaining adequate turnover of a broad list and the ethics of a public health service promoting one pharmacy over another; community pharmacies are, after all, entities.
In 2012, we designed a ‘Community Pharmacy Survey’ to provide objective baseline data to describe the community pharmacy involvement in care for patients dying at home.13 We showed:
Good awareness, by community pharmacists, of palliative patients using their business;
Willingness of pharmacists to contact nearby pharmacies to access urgent medicines, in the event that the prescribed medicine was unavailable;
Many community pharmacists continue to struggle to anticipate which medicines to stock; and
Many formulations on the pharmacy shelf were imminently expiring, questioning the sustainability of ongoing supply of this range of products.
Our baseline study resulted in two significant interventions:
First, a monthly educational newsletter was developed to provide ongoing education, for community pharmacists, relating to medicines management in the palliative setting. The newsletter is distributed electronically and archives are available at: http://www.caresearch.com.au/caresearch/tabid/2429/Default.aspx. There are currently 263 subscribers to this email distribution list.
Second, a ‘Core Medicines List’ (the List) was developed (see table 1). This was based on the understanding that having a wide range of formulations to manage a small range of symptoms in the terminal phase may:
Be unnecessary in the management of uncomplicated community-based palliative care patients; and
Make it difficult for pharmacists to anticipate which medicines to stock.
Relationship between ‘Community Pharmacy Survey’ results, ‘Core Medicines List’ and ‘Preparedness Score’
The development of the List, under the auspices of the SA Health Palliative Care Clinical Network, has been described elsewhere.14 It aimed to ensure that the particular medicines that are prescribed align with community pharmacy stocks of medicines. A range of strategies were used to disseminate the List with a multidisciplinary primary care audience (see table 2).
Strategies used to promote the List with the multidisciplinary primary care audience
In 2015, the ‘Community Pharmacy Survey’ was reissued. This paper identifies changes in access to medicines in SA community pharmacies for the management of terminal phase symptoms, as a result of the introduction of the List. In addition, this paper explores factors predictive of pharmacies carrying all five core medicines, as well as those carrying a broader range of medicines useful in terminal care.
Methods
The study population comprised all South Australian community pharmacies, as identified by the Australian Health Practitioner Regulation Agency (AHPRA) database. Pharmacies were excluded if they would not be expected to carry end-of-life medicines (eg, the airport pharmacy).
The development of the ‘Community Pharmacy Survey’ and the reporting of baseline data have been described elsewhere.11 It contains four sections, relating to:
Demographic information and range of services provided by the pharmacy;
Awareness of palliative patients using the pharmacy;
The range of medicines stocked and expiry dates of these medicines, using a predetermined list of 13 oral (liquid) and injectable formulations likely to be prescribed in the last days of life (see table 1);
How pharmacy staff respond when unable to immediately supply end-of-life medicines.
Since 7 out of the 13 formulations had multiple strengths available, an additional question was included in part 3, asking each respondent to indicate if they kept an alternate strength of these medicines.
All eligible registered community pharmacies in SA were mailed a package containing a covering letter, the survey and a return envelope, as suggested by the Dillman Total Design Survey Method.15 All letters were addressed to the proprietor or manager where this information was known. Respondents could return the survey using the prepaid envelope or facsimile.
Since some of the pharmacies were subscribing to the newsletter, each survey was allocated an individual code. This allowed anonymous identification of surveys being returned from pharmacies subscribing to the newsletter, and identification of non-responders. Once the survey was returned, the data were entered into a secure results database.
All responding surveys were allocated to Socio-Economic Indexes for Areas (SEIFA) deciles based on the postcode identified on the survey and socioeconomic data from the 2011 Australian census.
Finally, a ‘preparedness score’ (ranging from 0 to 5) was calculated for each pharmacy, using the data provided, by adding one point if they held a medicine for each of the following pharmacological classifications: opioid, benzodiazepine, antiemetic, anticholinergic and antipsychotic. A ‘preparedness score’ of 0 indicated a poor likelihood of finding medicines useful to manage terminal phase symptoms.
Statistics
All data were analysed using Stata V.14.0 (Statacorp).
Comparisons between two groups were made using Student's t-test, Wilcoxon's rank-sum test or a χ2 test, as appropriate. Statistical significance was defined as a p value ≤0.05.
Multiple linear regression was used to identify predictors of the number of core medicines stocked, and multiple logistic regression was used to identify predictors of being ‘fully prepared’.
Results
Four hundred and forty-seven surveys were mailed out and six were returned without opening. Of the remaining 441 pharmacies, 141 (32%) responded, compared with 24% in 2012 (p=0.01). Since surveys were received from a broad range of socioeconomic areas, the data were considered to be sufficiently geographically representative of all pharmacies in SA and no follow-up letters were issued.
Respondents
The proprietor or manager accounted for 104/141 (74%) responses with pharmacists, pharmacy interns and pharmacy technicians making up the remainder. This was similar to the breakdown observed in 2012.
Demographics
Forty-eight (34%) responses came from pharmacies with a rural postcode and 93 (66%) from metropolitan Adelaide, with a response rate of 33% and 31%, respectively.
All demographic information (including distribution of SEIFA deciles) and range of services offered by community pharmacies were statistically similar to those obtained in the 2012 survey (see table 3).
Characteristics of responders and community pharmacies
Awareness of palliative patients
One hundred and sixteen (82%) respondents stated that they were aware of at least one palliative patient using their pharmacy over the preceding 12 months, compared with 87% in 2012. Furthermore, the means by which respondents learnt about the palliative status of a patient including (1) the patient or caregiver, (2) the information provided on the prescription, or (3) by discussion with another healthcare professional were similar to those identified in 2012.
Respondents who indicated that they had been informed by another healthcare service of at least one palliative care patient using their services held a significantly higher number of the core medicines (z=−2.4, p=0.02) and had a larger ‘preparedness score’ (z=−2.3, p=0.02).
Access to stock
Each pharmacy stocked a median of 3 medicines (IQR 1–6) listed within the survey, the same number reported at baseline. The availability of most formulations remained similar between 2012 and 2015. The exception was Clonazepam 1 mg/mL injection being more likely to be found in community pharmacies in 2015 (25% of pharmacies) than in 2012 (13% of pharmacies; p=0.02).
Core medicines
The proportion of SA community pharmacies carrying all five core medicines increased from 7% in 2012 to 18% (p=0.01; see figure 1).
Proportion of community pharmacies carrying all five core medicines.
Predictors of community pharmacies carrying all five core medicines include those that subscribe to the electronic newsletter (p=0.04) and those that provide a clinical service to an aged care facility (p=0.02).
The pharmacies that provided data in the 2015 survey represented over 95 distinct postcodes across SA. Twenty (21%) of the postcodes had at least one pharmacy that held all five core medicines, compared with 7/73 (10%) postcodes in 2012 (p=0.04).
Preparedness
The median ‘preparedness score’ was 2, indicating that half of the time community pharmacies will be able to provide at least two out of a possible five pharmacological classifications suitable for management of terminal phase symptoms. Forty-one (29%) pharmacies had a ‘preparedness score’ of 5. This variable was new and could not be compared with the 2012 data.
Predictors of community pharmacies with a ‘preparedness score’ of 5 include those with an afterhours or on call service (p=0.02) and those that offer a clinical service to aged care facilities (p=0.04).
Number of months to expiry
The median expiry dates improved for two of the five core medicines in 2015, compared with the baseline (see figure 2). The medicines where the largest increases in duration to expiry were reported were Clonazepam (from a median of 12 to 21 months) and Haloperidol (from a median of 21 to 30 months).
Comparison of minimum, maximum and median months to expiry.
There was no decline in median months to expiry for the other three core medicines.
Response if unable to immediately supply medicines
As in 2012, respondents acknowledged that they would employ multiple strategies to source any medicines they did not stock. The most common strategy (75%) continued to be contacting another pharmacy to source any medicines they were unable to immediately supply. Twenty-seven respondents (19%) indicated that they would contact the doctor on the patient's behalf to either recommend a change in medicine, or a change in strength of the medicine.
Importantly, the strongest predictor as to whether a responder would contact the prescriber was awareness of palliative patients using their business (p=0.03). This is a new finding and was not identified in 2012.
Discussion
This study demonstrates that significantly more SA community pharmacies carried all five core medicines following the delivery of a range of multidisciplinary education strategies, almost tripling the 2012 figures. This indicates that the likelihood of South Australians being able to access items from the List through community pharmacies in 2015 has significantly improved.
In practical terms, almost one in five community pharmacies in SA carry the full List. With an average of 3.1 pharmacies per metropolitan postcode, there is a high likelihood of a caregiver accessing these medicines through their usual pharmacy or one close by. Clearly, this still disadvantages those without a car or access to good transport links.
Furthermore, in the rural setting, there are 1.5 community pharmacies per rural postcode. Caregivers in country areas of SA continue to face significant challenges in accessing appropriate medicines to manage terminal phase symptoms through a community pharmacy.
When a pharmacy has to order the medicine from their supplier, there are extra delays. Most caregivers faced with the prospect of waiting a day for the medicines to arrive are most likely to approach another pharmacy or refer the issue back to the palliative care nurse in order to expedite symptom control. Clearly, further strategies need to be developed if these results are to be further improved on.
The strongest predictors of community pharmacies stocking all five core medicines were subscription to the monthly electronic newsletter and providing clinical services to an aged care facility. While these aspects are largely invisible to patients, caregivers and other healthcare professionals, it suggests that engagement with community pharmacists on a clinical level can facilitate standardisation of practice. Encouragingly, respondents indicating that they had been informed by another healthcare service of at least one palliative care patient using their services held a significantly higher number of the core medicines and had a larger ‘preparedness score’. This is significant as it demonstrates that simple communication, in advance of patient deterioration, can guide what medicines pharmacies have on hand. Factors such as geography and number of pharmacists were not significant predictors, suggesting that standardisation of practice is largely independent of size and location of the pharmacy.
The increase in the number of months to expiry for two of the five core medicines (clonazepam and haloperidol) is meaningful for two reasons. First, this increase in the number of months to expiry suggests an increase in turnover of these medicines, which is most likely due to a shift in prescribing practices. Second, community pharmacies carrying clonazepam or haloperidol ampoules were less likely to have stock expiring in 2015 compared with 2012. Since the costs associated with medicines expiring on the dispensary shelf are carried by the pharmacy business, this has practical implications towards the sustainability of businesses continuing to carry these medicines, as there is less financial risk for pharmacies carrying stock. Interventions aiming to improve the timely access to medicines must factor in the sustainability for all stakeholders, including community pharmacies. This may include financial subsidy programmes, to support community pharmacies with poor turnover of stock.
In the terminal phase, a small range of symptoms can present suddenly or be exacerbated. Importantly, five pharmacological classifications of medicines can manage these anticipated symptoms: opioid, benzodiazepine, antiemetic, anticholinergic and antipsychotic. The ‘preparedness score’ indicates the likelihood of a pharmacy having this range of pharmacological classes in stock to manage these symptoms. It is promising that nearly a third of the pharmacies recorded a ‘preparedness score’ of 5. Factors that are predictive of pharmacies with a higher ‘preparedness score’ include awareness of palliative patients using their business and subscribing to monthly newsletters. In promoting the List, it was expected that some prescribers, through personal preferences, will prescribe different medicines and formulations. With nearly one in three responding pharmacies with a ‘preparedness score’ of 5, the data support this theory. The authors note that the number of respondents indicating that they would contact the prescriber in the event of being unable to immediately provide stock remains low. One of the predictors of a respondent discussing the situation with the prescriber was their awareness of at least one palliative care patient using their business within the previous 12 months. If the pharmacist is prepared to contact the prescriber, the pharmacist may facilitate faster access to medicines, including potentially changing to a complementary medicine or substituting an alternate strength of medicine. This provides weight to strategies that proactively engage community pharmacists within the palliative care multidisciplinary team, early within the patient's trajectory. Examples of how pharmacists can engage with palliative care populations have been well documented and include medication reviews, case conferences, drug information and proactive discussions about which medicines are likely to be prescribed.16
About three-quarters of respondents indicated that they would contact nearby pharmacies to either borrow stock or identify another access point for medicines. These data confirm that this is a suitable strategy for metropolitan areas. Other approaches to support access to these medicines, such as the use of just-in-case kits and using government-run country hospitals, need to be explored for patients living in rural areas of the state.
The development of the List is simply one strategy to improve access to medicines. It is a safeguard for patients who deteriorate suddenly at the end of life and is no replacement for good advance planning. Proactive engagement by prescribers with community pharmacists should be standard when preparing for deterioration of palliative care patients. It also allows for multidisciplinary discussions that support tailored prescribing, adapting to the patient's needs and taking into account established medicines for symptom control and significant comorbidities. Strategies must target all stakeholders engaged with the medicines management cycle, including doctors, community nurses, pharmacists, consumers and policymakers, if we are to ensure that the medicines stocked are also prescribed. The findings of this study suggest that the multidisciplinary approach to promoting this List has improved proactive engagement with pharmacists. We suggest that further research is required to identify the mechanisms. This is important with the recent endorsement of a national community prescribing list by the Australian and New Zealand Society of Palliative Medicine (ANZSPM).17
Strengths and limitations
The survey response rate was greater in 2015 compared with 2012, indicating a strong interest in palliative care. Furthermore, there is no difference in the proportion of pharmacies in rural postcodes that did not respond, indicating that this interest extends into rural areas of the state.
All five formulations included within the List were contained within the original 2012 ‘Community Pharmacy Survey’. This allowed evaluation of the uptake of this list in SA community pharmacies since the baseline survey.
The List was not mandated, meaning that prescribers, through personal preferences, were open to use different medicines and formulations beyond those recommended, if they wished. These personal preferences could also influence which medicines the pharmacies stocked. The ‘preparedness score’ was calculated on the information provided within the survey to identify pharmacies that could still hold formulations from five pharmacological classifications. Yet the list of medicines included within the survey is simply an overview of the range of medicines that are likely to be prescribed, based on prescribing patterns from an Australian metropolitan inpatient palliative care unit. The 2015 survey was expanded to enquire about alternative formulations stocked, but fell short of creating a comprehensive list of medicines used to manage symptoms anticipated in the terminal phase. There are other medicines (eg, atropine) that may be appropriate in managing symptoms that were excluded.
Despite a strong response from across SA, it is most likely that the encouraging results are from respondents with an interest and passion for supporting palliative care patients. Further work is required to continue to engage with GPs, community nurses and pharmacists to ensure that good access to medicines is sustainable.
While increases in proportions of pharmacies carrying all items from the List and improvements in months-to-expiry data suggest a shift in prescribing, further research is needed to explore the impact of this list on prescribers.
Recommendations
With the List in mind, prescribers should be encouraged to liaise with the patient's usual pharmacy prior to prescribing, to ensure that their preferred medicines are available.
Palliative care organisations should raise the issue of access to medicines with community-based prescribers in advance, to ensure that their patients are adequately prepared for symptom management, where a home death is favoured.
Policymakers need to create incentives for community pharmacies to hold a small range of medicines that may be required urgently, but so infrequently, that loss of stock due to expiry is otherwise a financial burden.
Conclusion
This study reveals that a multidisciplinary educational intervention significantly improved the availability of medicines from a List in SA community pharmacies. There are strong indicators that the promotion of this list is guiding management of patients in the terminal phase. The results of this study suggest that there is value in developing and promoting a List, facilitating timely access for community palliative patients to appropriate medicines in the terminal phase. While the development and promotion of the List has resulted in significant uptake in metropolitan areas, there is still substantial work to be done in rural areas.
Acknowledgments
The authors would like to thank Debra Rowett and John Gray for their valuable assistance in developing the original ‘Community Pharmacy Survey’, and the staff of the responding community pharmacies for their cooperation.
References
Footnotes
Contributors PT contributed to the planning, conduct and reporting of the work described in the article. CH and PH contributed to the statistical analysis of the data and reporting of the work described in the article. TT contributed to the planning, conduct and reporting of the work described in the article.
Competing interests None declared.
Ethics approval Ethics approval for the study was obtained from the Southern Adelaide Clinical Human Research Ethics Committee (SAC HREC).
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement The authors are prepared to make unit record data available to share with anyone who is interested in comparing regional differences.