What Are the Factors Identifying Caregivers Who Need Help in Managing Medications for Palliative Care Patients at Home? A Population Survey

Abstract

Background:

For most people, the last 12 months of life are spent living in the community, with the support of family and friends for a number of caregiving functions. Previous research has found that managing medicines is challenging for caregivers. Currently there is little information describing which caregivers may struggle with tasks associated with managing a loved one's medicines.

Aim:

The aim of this study was to identify factors that flag caregivers who are likely to experience problems when managing someone else's medications.

Setting/Participants:

The annual South Australian Health Omnibus Survey provides a face-to-face, cross-sectional, whole-of-population view of health care. Structured interviews, including questions covering palliative care and end-of-life care, were conducted with 14,625 residents in their own homes.

Results:

Of the 1068 respondents who had provided care for someone who died of a terminal illness in the last five years, 7.4% identified that additional support with medicine management would have been beneficial. In addition, three factors were predictive of the need for additional support in managing medicines: aged <65 years; lower household income; and living in a metropolitan region.

Conclusion:

The findings of this study provide insights to inform the development of palliative care service models to support informal caregivers in the management of medications for people with a life-limiting illness.

Background

Good symptom control at the end of life is valued by patients, family, and friends providing care, and health professionals above all else.^1,2 For most people, this will include prescribed medications, which generally increase in number as death approaches.³ This includes a combination of medications for long-term comorbidities and medications for symptom control.^4,5 The proportion of people with palliative care needs who are on 10 or more medications is very high,^4,5 putting these patients at substantial risk of drug–drug and drug–host harms.⁶

More than 95% of care in the last year of life occurs in the community^7,8 and relies heavily on the presence and active engagement of families and friends to provide that care.^9,10 This results in a large number of relatively untrained individuals who administer medications to people with life-limiting illnesses. At a community level, it is really important that we support caregivers as they take on this difficult and challenging role.¹¹ This is particularly important as caregivers may fulfill this role several times in their life despite little preparation and often with poor evaluation of their ability or willingness to provide this care.^9,10

Over 80% of caregivers of hospice patients report managing and administering medications to relieve patients' pain and other distressing symptoms.¹² Yet, many caregivers have unmet needs and require preparation for the role, with ongoing information, education, and personal support to assist them.¹³

Medication management relates to all practices used to manage the provision of medicines and can involve their purchase, administration, and monitoring.¹⁴ For patients, medication management can be confusing and complicated; medications are often changed and doses altered, especially if the dosage is based on fluctuating severity of symptoms or other clinical findings.^12,15 For caregivers, managing the administration of medications for another person has additional challenges.^16,17 Medication management creates intense and ongoing responsibilities for caregivers that increase as the patient progresses through the illness and loses the capacity to contribute to his or her own medication management. Furthermore, the complexity of care provided in the community continues to increase with much more expected of caregivers in medication management than ever before, including the administration of subcutaneous medications and the need to make judgments about the use of “as-needed” (pro re nata) medications.^18,19 The medications used for symptom control have a range of potentially serious side effects, especially if given differently than prescribed.

Currently, there is little information about how caregivers perceive the task of managing medications for a palliative care patient.²⁰ Consequently, it is important to identify possible factors that flag caregivers who require additional support in managing medications.

Aim

The aim of this study was to determine whether there are any clinico-demographic factors that identify caregivers who perceive they need additional support in managing medicines. The null hypothesis was that there would be no characteristics that could help identify this group of caregivers. If any characteristics were identified, these findings could be used to develop and implement resources about medication management, focused on the unmet needs of these caregivers.

Methods

Survey design

The South Australian Health Omnibus Survey (HOS) is an annual, face-to-face, cross-sectional, whole-of-population, multistage, systematic clustered area sampling survey. It is run by a commercial research organization that provides deidentified data for researchers on a user-pays basis. Approximately 200 questions (including those covering palliative and end-of-life care) are asked annually, of which only the demographic questions remain constant.

Interviews are conducted by trained interviewers in participants' homes and last between 60 and 90 minutes. This approach provides a way of assessing the population rather than people identified only by health services with whom they have had contact.

Setting and subjects

The Australian state of South Australia has a population of ∼1.74 million people (6.9% of the Australian population), the majority of whom live in the state capital.²¹ The HOS is carried out between September and December each year. The HOS is piloted with 50 respondents from the general community annually. The methodology has been described in detail elsewhere.²²

A stratified sampling method by the size of a nonmetropolitan town identifies a representative cohort from metropolitan and country towns (with populations of more than 1000) with more than 5000 properties approached annually.

The units of randomization are (1) census collection districts (CCDs) and (2) the starting point within the CCDs. From the starting point in each CCD, there is a skip pattern of every 10th property. Properties approached may be houses, businesses, properties with other uses, or vacant land. Hotels, hospitals, caravan parks, and aged care facilities are excluded from the sample.

For dwellings, interviewers return up to six times if no one responded before acknowledging that they were unable to make contact. Participation rates are calculated on the number of potential participants with whom contact is made, and therefore, the denominator is fewer than the total number of properties approached. For example, if the property identified was a commercial property, or a vacant block of land, these were not included in the denominator of the participation rate.

One interview was conducted in each household with the person aged ≥15 years who most recently had a birthday. If that person was unwilling to participate, and another member of the household was unable to replace them, then that household was classified as declining participation.

For this particular study, having established that the person had provided care for someone who died of a terminal illness in the last five years, respondents were asked “…if additional support would have been helpful.” One potential response was a perceived need for help with medication management.

Data quality

All data were double entered with a supervisor, following up missing responses by telephone. In addition, 10% of each interviewer's respondents were recontacted to confirm their eligibility and ensure consistency of responses by reanswering a selected number of questions. Data were anonymized and then released to researchers.

Statistical analysis

Data were weighted to the 2010 Estimated Residential Population for South Australia by five-year age groups, sex, rurality, and household size.

Data analyses were conducted using SPSS for Windows Version 19.0. (2011; SPSS Chicago, IL). A p-value of 0.05 was accepted as the threshold for statistical significance, and adjusted by the Bonferroni correction for multiple comparisons (p ≤ 0.004). Simple descriptive statistics were presented together with a multivariable regression seeking to identify any distinguishing characteristics for the subgroup of caregivers who identified that they perceived the need for more help with medication management. Included in the regression model was a combination of factors that were significant in univariable outcomes and those that were biologically plausible, including factors relating socioeconomic status and age (given the relationship to health literacy) and rurality (given differing levels of community clinical support).

Sensitivity analysis

To confirm the magnitude and direction of key findings, analyses were rerun with unweighted data.

Results

Among the population of 14,625 interviewed (2002, 2003, 2005–2007), 1085 (7.3%) respondents identified that they had provided day-to-day hands-on or intermittent hands-on care for “someone close to them” with a life-limiting illness during the previous five years (Table 1). Seventy-nine (7.4%) of these respondents identified that additional support with medication management would have been helpful. Key parameters that sought to identify characteristics of caregivers who perceived that they required additional help managing medications included the following: demographic characteristics; caregiving profile; characteristics of the deceased care recipient and palliative care service use (Table 1).

Table 1.

Characteristics of Caregivers and Need for Additional Support with Medications from a Survey of 3000 People Randomly Selected Each Year in South Australia

Characteristic	Additional support needed with medications		Total	p Value
Characteristic	No (n = 1006)	Yes (n = 79)	Total	p Value
Age group (years)
<65	798 (92.5)	65 (7.5)	863	0.531
65+	208 (93.7)	14 (6.3)	222	0.531
Gender
Male	373 (94.0)	24 (6.0)	397	0.265
Female	633 (92.1)	54 (7.9)	687	0.265
Country of birth
ESB	909 (92.9)	69 (7.1)	978	0.554^d
NESB	97 (91.5)	9 (8.5)	106	0.554^d
Work status
Work full-time or part-time	439 (91.6)	40 (8.4)	479	0.403
Unemployed, home duties, student, and other	355 (93.2)	26 (6.8)	381	0.403
Educational attainment
Did not complete school, high school only, or still studying	456 (92.1)	39 (7.9)	495	0.429
Trade qualification, certificate, diploma, or higher	549 (93.4)	39 (6.6)	588	0.429
Household income (per year)^a
Up to $60,000	556 (91.7)	50 (8.3)	606	0.093
$60,001 or more	321 (94.7)	18 (5.3)	339	0.093
Rurality
Metropolitan	678 (91.4)	64 (8.6)	742	0.007
Nonmetropolitan	328 (95.9)	14 (4.1)	342	0.007
SEIFA
Low SEIFA score	429 (92.5)	35 (7.5)	464	0.702
High SEIFA score	577 (93.1)	43 (6.9)	620	0.702
Most involved level of care
Daily “hands-on” care	423 (92.8)	33 (7.2)	456	0.964
Intermittent “hands-on” care	583 (92.8)	45 (7.2)	628	0.964
Place of death^b
Home	80 (94.1)	5 (5.9)	85	0.584^d
Elsewhere	326 (95.3)	16 (4.7)	342	0.584^d
Palliative care services accessed
Yes	615 (93.3)	44 (6.7)	659	0.410
No/don't know	391 (92.0)	34 (8.0)	425	0.410
Care again^c
Yes	282 (94.3)	17 (5.7)	299	0.596^d
Maybe/no	73 (92.4)	6 (7.6)	79	0.596^d

Five years of data.

Values are numbers (percentages).

A large proportion of respondents declined to provide this answer.

Only asked in years 2005–2006.

Only asked in years 2006–2007.

Fisher's exact test.

NESB, non-English-speaking background; ESB, English-speaking background; SEIFA, Socioeconomic Index for Areas—an index of relative socioeconomic disadvantage and is based on the postcode provided by recipients.

In univariate analyses, caregivers who lived in metropolitan areas were more likely to identify unmet needs in medication management compared with caregivers who were living in rural areas (8.6% and 4.1%, respectively; p = 0.007). This was the only difference identified between the groups. It is relevant to note that accessing a palliative care service made no difference between groups, and death at home similarly did not differ between groups.

In multivariate analyses, factors included in the model comprised the following: caregivers' ages; caregivers' sex; country of birth; the caregiver's (estimated) household income (noting that this may change as the result of the death of someone else in the household); rurality; most involved level of care (daily or intermittent hands-on care); and whether the caregiver identified that a palliative care service was used (Table 2). The fit of the model was adequate (Hosmer and Lemeshow p = 0.114) with the Nagelkerke R² being 4.3%.

Table 2.

Regression Model for Predictors of Perceived Need for Additional Support in Medication Management for Caregivers Who Had Provided Care for Someone at the End of Life

Factors	Odds ratio	95% Confidence interval		p Value
Factors	Odds ratio	Lower	Upper	p Value
Caregivers' age	0.982	0.967	0.998	0.031
Caregivers' sex
Female	1.000
Male	0.842	0.492	1.441	0.530
Country of birth
NESB	1.000
ESB	1.028	0.446	2.367	0.949
Estimated household income
>$60,000	1.000
≤$60,000	1.820	1.017	3.260	0.044
Rurality
Nonmetropolitan	1.000
Metropolitan	2.425	1.263	4.656	0.008
Most intense level of care
Intermittent hands-on care	1.000
Day-to-day hands-on care	1.089	0.649	1.826	0.747
Use of a palliative care service
No (or unknown)	1.000
Yes	0.954	0.572	1.593	0.858

Three factors were predictive of the need for additional support in medication management: a person younger than 65 years (odds ratio [OR] 0.982 [95% confidence intervals, CIs] 0.967–0.998); lower household income (OR 1.820 [95%CI 1.017–3.260]); and living in the metropolitan region (OR 2.425 [95%CI 1.263–4.656]).

Discussion

This study identifies a subgroup of caregivers supporting people at the end of life who subsequently indicated that would have liked more help managing medications. Three factors create a composite picture of a caregiver more likely to identify the need for assistance: a person younger than 65 years, from a lower income household, and living in a metropolitan region. This is an important finding that should be considered from the time a patient is referred to palliative care, and their caregiver(s) are identified. Consideration needs to be given to the systematic training and ongoing support that need to be in place for such caregivers.

People aged <65 years in general are likely to be working and have competing commitments such as caring for children.^23,24 They may also be living separately to the patient, particularly if they are caring for a parent. Financial stress for caregivers has been discussed in the literature.²⁵ In palliative care, the frequent changing of medications, formulations, and routes of administration creates a significant financial burden for the families of people being cared for in the community.²⁵ Being in a lower income household may potentially mean that difficult choices relating to household costs—including the purchase of medicines—are required if there is insufficient money.

People living in metropolitan areas are less likely to have strong communities supporting them when compared with nonmetropolitan areas.²⁶

Palliative care services have a responsibility to assess caregivers' level of competence and confidence in managing medications for the patient from the time of initiating referral right through to the time of death. All caregivers need careful assessment, with particular emphasis on caregivers whose characteristics are defined by this study. Complexity and responsibility of caregivers' medication management increase as death approaches often with increasing numbers of medications,³ new routes of medication administration, changing doses, and the requirement for “as-needed” medications.^18,19 The latter requires caregivers to make complex clinical decisions for which they are often ill-prepared.

Palliative care is often provided to people who frequently have a higher risk of medication misadventure due to the following:

Polypharmacy associated with multimorbidity.

Frequent transitions between health care settings.

Multiple prescribers engaged with their care.

Extensive changes to their medications (e.g., dose changes, switching formulations, ceasing of medications as clinical conditions change).

Presence of worsening end-organ failure.

This creates an environment that is frequently changing for the person managing someone else's medications.

Assuming responsibility for the management of medications for another person is stressful, even when caregivers are supported in taking on the role. Given the large number of medications that people take at the end of life and the fact that the overall number increases as death approaches, and is associated with frequent changes,³ it would be understandable if some caregivers found the process of managing someone else's medications stressful.

Strengths of the study

Given that respondents are contacted independently of involvement with a health service, it is likely that this is a relatively representative population, reflecting the experiences of a wide range of caregivers supporting people at the end of life.

Limitations of the study

Limitations of this study include that no one younger than 15 years was able to be interviewed; it is known that people younger than 15 years do provide care to people at the end of life.²⁷ This is especially important, given this study identified that caregivers younger than 65 years were more likely to have unmet needs in managing medicines.

People living in towns with a population <1000 were excluded as the HOS was not conducted in these areas. This excluded people living in remote parts of the state.

The study excludes (and is therefore blind to) the increasing proportion of the population living in caravan parks, including the so-called gray nomads who, because of age, are more likely to need palliative care yet are less likely to have local support networks.

The data were collected 15 years ago, but given the absence of any data on this topic, it was felt important to create an understanding of some robust data to inform further work in this important area of hospice/palliative clinical care.

Implications for clinical practice

In practice, this study shows that when considering medication management, clinicians need to consider caregivers and assess their level of confidence in managing medications. This is particularly important given the wide range of people called upon to provide care at the end of life. Each caregiver's perspective on medication management should be reviewed regularly, especially as new medications are added, routes of administration change, the intensity of care increases, and clinical assessments are required if ‘as-needed’ medications are introduced.¹⁸

Targeting specific subpopulations of caregivers (such as those with a lower income), for medication management services, may provide a pragmatic approach in a resource-limited environment.

A key health professional in the clinical care team for providing education and ongoing support for caregivers providing medication management is the pharmacist. In Australia, Home Medication Reviews (HMRs) provide one model for engaging with patients and their caregivers within the home environment: reviewing medications; providing education; and creating a mechanism for ongoing support.²⁸ As HMR programs are evaluated and their evidence base strengthened, understanding the perspective of caregivers for people at the end of life should be an area of particular focus.

Implications for future research

Having established the baseline rate of 1 in 13 people identifying that additional support would have been useful, future surveys can explore the level of medication decision making expected, especially for “as-needed” medications. This includes clinical assessment skills to inform “as-needed” medications or where the dose of regular medications needs to be adjusted. Understanding this by classes of medication and routes of administration will be important in future qualitative and quantitative work. Another question of interest will be whether caregivers who themselves are on regular medications are more likely to be able to manage someone else's medications with more confidence. Finally, an objective prospective evaluation of the competence of caregivers to manage medications would build logically on the foundation created by this study, especially for people who perceived no need for additional support.

This initial survey was not designed to identify the consequences of feeling unsupported in managing medications. Such work now needs to be undertaken. Does this contribute to avoidable hospital admissions? Is symptom control poorer when a caregiver is less confident? Are compliance rates for medication administration affected when caregivers feel unsupported? Such work will need to involve longitudinal data collection in future research.

Generalizability

This study was conducted in one jurisdiction in a resource-rich country that has universal health care and a health system in the community built strongly around general practitioners. Community pharmacists in Australia are trained to deliver advice on medications also and their support of caregivers may differ in other parts of the world.

Ethics and Consent

The HOS receives annual South Australian Department of Health Research approval. Verbal consent was obtained from all participants before commencing an interview and continued participation was accepted as ongoing consent.

Data Sharing

Data are available on request from bona fide researchers.

Footnotes

Acknowledgments

The authors acknowledge the generosity of the community members who gave their time to answer questions. This is especially generous of people who are grieving the death of someone close to them, and sharing their experience is hoped to improve the experience for other caregivers in the future.

Funding Information

This study was funded from discretionary funds held by the Academic Unit of Palliative and Supportive Services, Flinders University, Australia.

Author Disclosure Statement

No competing financial interests exist.

References

Steinhauser

, Christakis

, Clipp

, et al.: Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA, 2000; 284:2476–2482.

Singer

, Martin

, Kelner

: Quality end-of-life care. Patients' perspectives. JAMA, 1999; 281:163–168.

Currow

, Stevenson

, Abernethy

, et al.: Prescribing in palliative care as death approaches. J Am Geriatr Soc, 2007; 55:590–595.

Lunney

, Lynn

, Foley

, et al: Patterns of functional decline at the end of life. JAMA, 2003; 289:2387–2392.

Morin

, Vetrano

, Rizzuto

, et al.: Choosing wisely? Measuring the burden of medications in older adults near the end of Life: Nationwide, longitudinal cohort study. Am J Med, 2017; 130:927–936.e9.

Thompson

: Deprescribing in palliative care. Clin Med, 2019; 19:311–314.

Lau

, Joyce

, Clayman

, et al.: Hospice providers' key approaches to support informal caregivers in managing medications for patients in private residences. J Pain Symptom Manage, 2012; 43:1060–1071.

McNamara

, Rosenwax

, Murray

, Currow

: Early admission to community-based palliative care reduces use of emergency departments in the ninety days before death. J Palliat Med, 2013; 16:774–779.

Higginson

, Sen-Gupta

: Place of care in advanced cancer: A qualitative systematic literature review of patient preferences. J Palliat Med, 2000; 3:287–300.

10.

Waller

, Girgis

, Currow

, Lecathelinais

: Development of a palliative care needs assessment tool (PC-NAT) for use by multi-disciplinary health professionals. Palliat Med, 2008; 22:956–964.

11.

Joyce

, Lau

: Hospice experiences and approaches to support and assess family caregivers in managing medications for home hospice patients: A providers survey. Palliat Med, 2013; 27:329–338.

12.

Lau

, Berman

, Halpern

, et al.: Exploring factors that influence informal caregiving in medication management for home hospice patients. J Palliat Med, 2010; 13:1085–1090.

13.

Paice

, Muir

, Shott

: Palliative care at the end of life: Comparing quality in diverse settings. Am J Hosp Palliat Care, 2004; 21:19–27.

14.

Australian Commission on Safety and Quality in Health Care:. National Safety and Quality Health Service Standards, 2nd ed. Sydney, Australia: ACSQHC, 2017.

15.

Orlu-Gul

, Raimi-Abraham

, Jamieson

, et al.: Public engagement workshop: How to improve medicines for older people?. Int J Pharm, 2014; 459:65–69.

16.

Smith

, Francis

, Gray

, et al.: A multi-centre survey among informal carers who manage medication for older care recipients: Problems experienced and development of services. Health Soc Care Community, 2003; 11:138–145.

17.

Tait

, Cheung

, Wiese

, Staff

: Improving community access to terminal phase medicines in Australia: Identification of the key considerations for the implementation of a ‘core medicines list’. Aust J Prim Health, 2017; 23:373–378.

18.

Russell

, Rowett

, Currow

: Pro re nata prescribing in a population receiving palliative care: A prospective consecutive case note review. JAMA, 2014; 62:1736–1740.

19.

Israel

, Reymond

, Slade

, et al.: Lay caregivers' perspectives on injecting subcutaneous medications at home. Int J Palliat Nurs, 2008; 14:390–395.

20.

Joyce

, Berman

, Lau

: Formal and informal support of family caregivers managing medications for patients who receive end-of-life care at home: A cross-sectional survey of caregivers. Palliat Med, 2014; 28:1146–1155.

21.

Australian Bureau of Statistics: https://abs.gov.au/ausstats/abs@.nsf/mf/1306.4 (Last accessed July 11, 2019).

22.

Currow

, Burns

, Agar

, et al.: Palliative caregivers who would not take on the caring role again. J Pain Symptom Manage, 2011; 41:661–672.

23.

Briggs

, Fisher

: Warning: Caring Is a Health Hazard. Canberra, Australia: Carers Association of Australia, 1999.

24.

Hudson

: Home-based support for palliative care families: Challenges and recommendations. Med J Aust, 2003; 179:S35–S37.

25.

Choi

, Seo

: Analysis of caregiver burden in palliative care: An integrated review. Nurs Forum, 2019; 54:280–290.

26.

Marsh

, Thompson

, Mond

: Living, loving, dying: Insights into rural compassion. Aust J Rural Health, 2019; 27:328–335.

27.

Syse

, Aas

, Loge

: Children and young adults with parents with cancer: A population-based study. Clin Epidemiol, 2012; 4:41–52.

28.

The Pharmacy Guild of Australia. 6th Community Pharmacy Agreement: Home Medicines Review (HMR)—Programme specific guidelines. Australia: The Pharmacy Guild of Australia and the Australian Government Department of Health, ACT, 2015. http://6cpa. com. au/medication-management-programs/home- medicines-review/ (Last accessed July 28, 2019).