Sage Journals: Discover world-class research

Abstract

Background

Understanding relationships between polypharmacy, treatment burden and other factors (e.g. age, knowledge of medicines), could inform interventions to reduce treatment burden.

Aim

To explore sociodemographic, clinical and medication factors associated with high treatment burden.

Design and setting

Secondary analysis of multi-centre UK primary care baseline clinical trial data.

Methods

Participants were ≥18 years, prescribed ≥5 repeat medicines with ≥1 indicator of potentially inappropriate prescribing (PIP). Survey data captured treatment burden (13-item Multimorbidity Treatment Burden Questionnaire; dichotomised with high burden ≥22), sociodemographic characteristics, quality of life, self-reported knowledge of medicines, and medication adherence. Electronic health records provided data on age, gender, long-term conditions, PIP, medicines use (last 3 months), and consultations (last 12 months). Associations between treatment burden and other factors were modelled using multivariable logistic regression.

Results

1711 adults from 37 general practices were included (mean age 72 years, 51% male). 381 (23%) reported high treatment burden. Multivariable analysis found high treatment burden was associated with younger age, being unemployed or in paid work, university or high education, multimorbidity, anxiety/depression, polypharmacy (≥8 medicines), lower medication adherence, knowledge of medicines, and quality of life. There was strong evidence (p<0.005) that the association between high treatment burden and polypharmacy was greater in people aged <60 years, those with poor knowledge of medicines, more multimorbidity, and those with low/moderate medication adherence.

Conclusion

High treatment burden was more common in adults with polypharmacy who were younger, had multimorbidity, poor medication knowledge, or low adherence. This should inform design of interventions addressing treatment burden.

Keywords

treatment burden polypharmacy multimorbidity medicines

Introduction

Treatment burden is the effort required for patients to look after their health conditions and the impact this has on their everyday life.^1,2 This includes collecting and taking medication, self-monitoring health conditions, attending and co-ordinating appointments, actioning lifestyle advice and seeing different healthcare professionals. Treatment burden is particularly relevant to the growing number of people experiencing multimorbidity (multiple long-term conditions in the same individual),³ which affects half of adults in the UK and a third of adults globally.^4,5 Healthcare systems tend to be disease-centred rather than patient-centred, and patients with multimorbidity are often required to attend separate appointments with different healthcare professionals, and to take different treatments for each of their long-term conditions. Several studies have reported an association between high treatment burden and multimorbidity.^6–8 High treatment burden is associated with poor health outcomes, including reduced quality of life.^6,7

The cumulative complexity model conceptualises treatment burden as a balance between patient workload (healthcare demands) and capacity to perform the work.⁹ Patients who have high workload (e.g. lots of appointments to co-ordinate and attend) but good capacity to manage the work (e.g. strong support from family or friends, no caring or work commitments etc) may report low treatment burden. Conversely, patients who have low workload, but poor capacity may report high burden. Treatment burden acts as a feedback loop between workload and capacity. Patients may become more complex over time as they accumulate long-term conditions and treatments. Some patients demonstrate relative resilience towards treatment burden and patients may have sufficient capacity to resist overburden through normalisation of health-related workload.¹⁰ Accordingly, while it is possible to identify factors generally associated with higher treatment burden, there is significant individual variation.

Polypharmacy (prescribing of multiple medicines to one individual) is common in individuals with multimorbidity.^11,12 For many patients, polypharmacy is appropriate - multiple medications are necessary to improve quality of life or increase longevity.¹³ Problematic polypharmacy occurs when the risk of taking multiple medications outweighs the potential benefit.^13,14 The impact of problematic polypharmacy includes an increased risk of adverse drug reactions, poorer health related quality of life and increased likelihood of emergency department attendance.¹⁵ Polypharmacy, poor adherence to medicines and low health literacy are associated with increased treatment burden.^8,16,17 However, the extent to which the association between polypharmacy and treatment burden might vary with other factors remains unclear.

Understanding the degree to which other factors (age, medication adherence and levels of multimorbidity) modify the association between treatment burden and levels of polypharmacy may provide a more nuanced understanding of these relationships. For example, groups of patients who would benefit most from interventions to reduce treatment burden. We hypothesised that greater levels of polypharmacy would be associated with higher treatment burden in younger individuals (for whom polypharmacy might be considered less typical or more difficult to manage), individuals with poor knowledge of their medicines, and individuals who find adherence to medication difficult.

The aims of this study were to examine the associations of high treatment burden with sociodemographic, clinical and prescribing characteristics, and to examine how the association between high treatment burden and greater levels of polypharmacy varied with age, medication literacy, multimorbidity, and medication adherence.

Methods

Study setting

This study is a secondary analysis of data collected in the multi-centre cluster-randomised trial, “Improving Medicines Use in People with Polypharmacy in Primary Care” (IMPPP; ISRCTN 90146150).¹⁸ The IMPPP trial assessed whether a complex intervention to optimise medicine use reduced potentially inappropriate prescribing (PIP) in adults with polypharmacy (prescribed ≥5 repeat medicines). General practices were recruited from two regions in England – the South-West and West Midlands. The current analysis uses baseline pre-randomisation data from the trial. The trial aimed to recruit 1,850 participants from 37 GP practices, powered to detect a clinically meaningful reduction of 0.5 in PIP indicators.¹⁸

Study population, eligibility criteria and recruitment

Patients were eligible if they were aged ≥18 years, prescribed ≥5 current long-term (i.e. “repeat”) medicines, and exposed to at least one indicator of PIP.¹⁸ Potentially eligible patients were identified pre-randomisation via an automated search of the electronic health records (EHR). A clinical member of the practice screened search results against exclusion criteria, excluding those patients (1) receiving end-of-life care; (2) deemed inappropriate to contact by practice staff; (3) unable to complete study questionnaires or medication review appointments; or (4) planning on moving general practice within the six-month follow-up period.

Patients were invited by post prior to randomisation. Those who wished to participate completed a consent form and baseline survey and returned both via post to the research team. Only participants who consented to their data being used for further research were included in the current analysis.

Survey data

The baseline survey captured socio-demographic characteristics, including ethnicity (seven categories based on census classification), English Index of Multiple Deprivation (IMD)¹⁹; determined from the participant’s postcode), education level, employment status and whether participants lived alone.

Treatment burden was measured in the baseline survey using the 13-item Multimorbidity Treatment Burden Questionnaire (MTBQ), a simply worded measure that has been validated in two studies of adults with multimorbidity in the UK.^6,20 Each question was scored as follows: 0 (not difficult, does not apply), 1 (a little difficult), 2 (quite difficult), 3 (very difficult) and 4 (extremely difficult). A global MTBQ score was calculated as the average score from the questions answered, multiplied by 25 to give a score from 0 to 100. Participants were excluded if more than 50% of responses to the MTBQ were missing. Treatment burden was categorised as no (0), low (0-10), medium (10-21) and high burden (≥22), as recommended by the validation study.⁶

Health-related quality of life (HRQoL) was measured in the baseline survey using the 12-item Short Form Survey (SF-12),²¹ a validated 12-item measure that assesses mental health status and physical functioning. The SF-12 was scored by calculating two summary scores: the Physical Component Summary (PCS) and Mental Component Summary (MCS). Higher scores indicated better physical functioning and mental health. HRQoL was also measured using a visual analogue scale adapted from the EQ-5D measure, asking participants to rate the question “Please tell us how good or bad your health is TODAY using the scale below” on a scale of 0 (worst health) to 100 (best health).²²

To assess knowledge of medicines, as part of the survey, participants were asked “Do you know what each of your prescription medicines is for?”(yes/no). Those answering “no” were asked to choose one of the following (i) “I am unsure what one or two of my prescription medicines is for”; (ii) “I am unsure what some of my prescription medicines are for”; or (iii) “I am unsure what most or all of my prescription medicines are for”. These questions were developed by the study team with input from the patient and public involvement (PPI) advisory group. For the logistic regression analysis, knowledge of medicines was dichotomised into “yes” (I know what each of my prescription medicines is for) and “no” (I do not know what each of my prescription medicines is for).

Adherence to medicines was measured in the baseline survey using the 10-item Medication Adherence Rating Scale (MARS-10).^23,24 Each question was scored 1 if good adherence was indicated, with the sum of all questions providing a total adherence score, categorized as low adherence (0-4), moderate adherence (5-7) and high adherence (8-10).

Data from electronic GP records

The following anonymised data were collected from each participant’s electronic health record (EHR): age, gender, long-term conditions, PIP indicators, medicines prescribed in the preceding 3 months, and in-hours general practice consultations (phone and in-person) recorded in the preceding 12-months. Current medication count was the number of unique chemical substances issued in the 3-month period, irrespective of strength or dose. Long-term conditions were the list of 20 clinically important conditions as defined by the validated Cambridge Multimorbidity Score (CMS).²⁵ These conditions were the highest-ranking with regards to prevalence and effect size. PIP indicators were based on a series of clinical code-based rules, combining prescribing, diagnostic and test codes as listed in the IMPPP trial protocol.¹⁸ Multimorbidity was measured using the 20-condition general outcome version of the CMS.²⁵

Analysis

The analysis was conducted using Stata (version 18.0).²⁶ Descriptive statistics were used to summarise the characteristics of trial participants. To explore associations between high treatment burden (outcome) and various exposure variables, treatment burden was dichotomised into no/low/medium burden (MTBQ score <22) and high burden (MTBQ score ≥22). Association with high treatment burden was modelled using univariable and multivariable logistic regression. The multivariable models additionally included the covariates age, gender, socioeconomic deprivation (IMD quintile), ethnicity, education, employment status, living alone, geographical location of GP practice (South-West England or Midlands), CMS, number of current medicines, adherence to medicines (dichotomised into low/moderate adherence (0-7) and high adherence (8-10)), and GP practice (random effect). These were chosen as they were considered to be clinically relevant with a biologically plausible association, as judged by clinical and topic experts on the research team. GP practice was included as a variable as it was a design characteristic of the IMPPP trial.

Individual models were used to examine the association between individual long-term conditions and treatment burden, adjusting for other overall morbidity burden. Regression models thus included a binary variable for the specific individual long-term condition, and a modified CMS which excluded the corresponding individual condition. For example, to explore an association between diabetes and treatment burden, the weighting for diabetes was removed from the CMS for individuals who had diabetes.

Interaction between polypharmacy and age, multimorbidity, knowledge of medicines and adherence

Although the association between polypharmacy and treatment burden was expected, we hypothesised that this may nevertheless vary with other factors. For example, it’s widely accepted that younger people describe higher treatment burden but whether this modifies the association between polypharmacy and treatment burden has been hitherto unknown. We pre-specified four key effect modifiers, which we considered relevant based on the existing literature: age, level of multimorbidity, knowledge of medicines and adherence to treatment. We hypothesized that younger people were more likely to report high treatment burden due to work and childcare commitments. The 60-year age cut off was chosen rather than a 65-year cut off because a reasonable proportion (23%) of participants aged 60-64 years were retired, compared to 7% of participants aged 55-59 years.

To examine the interaction between levels of polypharmacy and age in relation to treatment burden, an interaction term was created between number of medicines (dichotomised using median; <8 medicines, ≥8 medicines) and age (<60 years and ≥60 years), resulting in four categories. A logistic regression model was then applied comparing non-high treatment burden (MTBQ score 0-21) and high treatment burden (MTBQ score ≥22). The regression models were adjusted for age, gender, individual deprivation, ethnicity, education, employment status, living alone, geographical location, CMS, and clustering by GP practice (random effect). This process was repeated to examine the interaction between levels of polypharmacy and other factors, including multimorbidity (dichotomised using median CMS; i.e. low multimorbidity <2.2, high multimorbidity ≥2.2), knowledge of medicines (knowledge of all medicines, unsure of medicines) and adherence to treatment (low/moderate adherence, high adherence), in relation to treatment burden.

Missing data

Multivariate analyses were conducted on complete cases with no imputation of missing data.

Results

A total of 1727 participants were enrolled in the IMPPP trial (Figure 1). Of these, 1711 consented for their data to be used for further research and were included in the analyses. Global MTBQ scores were positively skewed: 14% reported no burden, 38% low burden, 25% medium burden and 23% high burden (Table 1, Figure 2).

Figure 1.

Participant flow diagram. * i.e. excluded due to practices already meeting the target of 50 eligible participants during the 6-month intervention delivery period.

Table 1.

Participant characteristics and outcome scores (N=1711).

	n/mean	%/(SD)
Socio-demographic characteristics
Mean age (SD), years (missing n=0)	71.7	(10.9)
Age group, years (missing n=0)
18-39	12	<1
40-59	207	12
60-79	1089	64
≥80	403	24
Gender (missing n=0)
Male	873	51
Female	838	49
Ethnicity (missing n=60, 4%)
White	1608	97
Asian or Asian British	19	1
Black/African/Caribbean	13	<1
Mixed/multiple ethnicity	6	<1
Other ethnicity	5	<1
Index of multiple deprivation quintile (missing n=0)
1 (most deprived)	174	10
2	293	17
3	279	16
4	474	28
5 (least deprived)	491	29
Geographical location (missing n=0)
West Midlands	995	58
South West	716	42
Education (missing n=155, 9%)
No educational qualifications	400	26
School age qualifications	722	46
University degree or higher	302	19
Other	155	8
Employment (missing n=148, 9%)
Retired	1111	71
Unemployed^a	175	11
Paid work (full or part-time)	277	18
Lives alone (missing n=76, 4%)
Yes	431	26
No	1204	74
Medical history
Number of long-term conditions^b (missing n=0)
0-1	93	5
2-4	973	57
≥5	645	38
Cambridge multimorbidity score^c (missing n=11, <1%)
Median (IQR)	2.2	(1.3, 3.2)
Long-term conditions^b (missing n=0)
Hypertension	1210	71
Anxiety/depression	823	48
Painful condition	756	44
Hearing loss	479	28
Irritable bowel syndrome	227	13
Current asthma	302	18
Diabetes	629	37
Coronary heart disease	480	28
Chronic kidney disease	436	26
Atrial fibrillation	296	17
Constipation	170	10
Stroke or TIA	241	14
COPD	191	11
Connective tissue disorder	199	12
Cancer	188	11
Alcohol problems	127	7
Heart failure	154	9
Dementia	11	<1
Psychosis or bipolar disorder	16	<1
Epilepsy	0	0
Prescribing and health utility
Number of current medicines^d (missing n=0)
<8	835	49
≥8	876	51
Do you know what each of your prescription medicines is for?^e (missing n=68, 4%)
Yes	1489	91
No, unsure about some medicines	131	8
No, unsure about most or all medicines	23	1
Adherence to medicines^f (missing n=27, 2%)
Low adherence	85	5
Moderate adherence	462	27
High adherence	1137	68
Potentially inappropriate prescribing indicators (missing n=11, <1%)
0-1	665	39
≥2	1035	61
Number of primary care consultations in the last 12 months^g (missing n=11, <1%)
0	122	7
1-4	601	35
5-9	578	34
10-19	337	20
≥20	62	4
Treatment burden and health related quality of life
MTBQ global score (missing n=40, 2%)
No burden (0)	239	14
Low burden (<10)	638	38
Medium burden (10-21)	413	25
High burden (≥22)	381	23
SF-12 score (missing n=37, 2%)
Physical health component score (mean, SD), n= 1684	35.8	(11.3)
Mental health component score (mean, SD), n= 1674	47.0	(11.7)
EQ-5D visual analogue scale (missing n=48, 3%)
Mean (SD)	61	(22)

^aUnemployed includes participants who selected: looking after the home; unable to work due to long-term illness or disability; and unable to work due to carer responsibilities.

^b20 clinically important long-term conditions included in the Cambridge Multimorbidity Score.

^cCambridge multimorbidity score calculated using weighted scores for each of the long-term conditions.

^dNumber of current medicines prescribed in the preceding 3 months recorded on the electronic health records.

^e“Do you know what each of your prescription medicines is for?” (yes/no). Those answering “no” were asked to choose one of the following: (i) unsure about one or two medicines; (ii) unsure about some medicines; (iii) unsure about most or all medicines. Responses (i) and (ii) were combined.

^fAdherence to medicines was measured using the 10-item Medication Adherence Rating Scale (MARS-10)24 and categorized as low adherence (0-4), moderate adherence (5-7) and high adherence (8-10).

^gNumber of primary care consultations include GP and nurse consultations recorded on the electronic health records.

Abbreviations: standard deviation (SD), transient ischemic attack (TIA), chronic obstructive pulmonary disease (COPD).

Figure 2.

Distribution of global treatment burden scores among participants (N=1,671). A global Multimorbidity Treatment Burden Questionnaire (MTBQ) score was calculated as the average score from the questions answered, multiplied by 25 to give a score from 0 to 100.⁶ The dotted vertical lines represent the cut off values between no treatment burden (0), low treatment burden (<10), medium burden (10-21) and high burden (≥22). MTBQ global score was missing for 40 (2%) of participants (1,671 included in the analysis for Figure 2).

Participant characteristics are shown in Table 1. The mean age was 72 years, 51% were male, and 97% were of white ethnicity. Most participants were retired (71%); had no educational qualifications (26%) or school age qualifications (46%); and lived with others (74%). Almost all participants had multimorbidity (≥2 long-term conditions; 95%). The most common long-term conditions were hypertension (71%), anxiety/depression (48%) and painful condition (44%).

Half the participants were prescribed eight or more medicines. 8% of participants were unsure about what some of their prescription medicines were for and 1% were unsure about what all their medicines were for. Most (68%) had high adherence to medications and only 5% had low adherence. Almost two thirds (61%) had two or more PIPs. The number of consultations in the last 12 months ranged from 0 (7%), to 20 or more (4%), with most having 1-9 consultations (69%). Just over a half of participants had no treatment burden (14%) or low burden (38%) and 23% reported high treatment burden.

Sociodemographic factors associated with high treatment burden on univariable analyses (p<0.05) were younger age, having higher educational qualifications, being employed or in paid work compared to retired, and living alone (Table 2). Individuals living in less deprived areas (IMD quintiles 3, 4 or 5) were less likely to report high treatment burden than those living in the most deprived areas (IMD quintile 1. Greater multimorbidity, and specific clinical conditions (anxiety/depression, painful condition, irritable bowel syndrome, current asthma, and diabetes) were also associated with high treatment burden. Aspects of prescribing associated with high treatment burden on univariable analyses included being prescribed ≥8 medicines, poor knowledge of medicines, and low/moderate adherence to medicines. Higher use of primary care service use (≥20 consultations in the preceding 12 months) and higher physical, mental and global HRQoL were also associated with a reduction in odds of high treatment burden (Table 2). There was no association (p=0.267) between treatment burden and potentially inappropriate prescribing.

Table 2.

Factors associated with high treatment burden (N=1671).

	n	Non-high burden (0-21)	High burden (≥22)	Unadjusted odds ratio^a (95%CI)	Adjusted odds ratio^b (95%CI)	p value for overall test^c
Socio-demographic characteristics
Age group (years)	1671
18-39	11	3 (27%)	8 (73%)	12.08 (3.12, 46.67)	7.06 (1.55, 32.2)
40-59	203	122 (60%)	81 (40%)	3.01 (2.05, 4.41)	1.40 (0.79, 2.50)
60-79	1070	848 (79%)	222 (21%)	1.19 (0.88, 1.60)	1.11 (0.76, 1.64)	0.0738
≥80 [ref]	387	317 (82%)	70 (18%)
Gender	1671
Female [ref]	818	623 (76%)	195 (24%)
Male	853	667 (78%)	186 (22%)	0.89 (0.71, 1.12)	1.02 (0.77, 1.35)	0.913
IMD quintile	1671
1 (most deprived) [ref]	173	117 (68%)	56 (32%)
2	288	212 (73%)	76 (26%)	0.75 (0.50,1.13)	0.71 (0.43, 1.17)
3	277	219 (79%)	58 (21%)	0.55 (0.36, 0.85)	0.60 (0.35, 1.00)
4	458	355 (78%)	103 (22%)	0.61 (0.41, 0.89)	0.78 (0.48, 1.26)
5 (least deprived)	475	387 (78%)	103 (22%)	0.48 (0.32, 0.70)	0.62 (0.38, 1.01)	0.2604
Ethnicity	1636
White [ref]	1593	1234 (77%)	359 (23%)
Other ethnicity	43	30 (70%)	13 (30%)	1.49 (0.77, 2.89)	‘1.10 (0.47, 2.58)	0.820
Education	1542
No educational qualifications [ref]	395	314 (79%)	81 (21%)
School age qualifications	718	560 (78%)	158 (22%)	1.09 (0.81, 1.47)	0.96 (0.67, 1.38)
University degree or higher	299	208 (70%)	91 (30%)	1.70 (1.20, 2.40)	1.88 (1.23, 2.86)
Other	130	107 (82%)	23 (18%)	0.83 (0.50, 1.39)	0.95 (0.53, 1.69)	0.0018
Employment	1547
Retired [ref]	1098	916 (83%)	182 (17%)
Unemployed^d	174	101 (58%)	73 (42%)	3.64 (2.59, 5.11)	3.10 (2.01, 4.76)
Paid work	275	185 (67%)	90 (33%)	2.45 (1.82, 3.30)	2.62 (1.74, 3.94)	<0.0001
Lives alone	1621
No [ref]	1199	941 (78%)	258 (22%)
Yes	422	311 (74%)	111 (26%)	1.30 (1.01, 1.68)	1.30 (0.95-1.79)	0.104
Geographical area	1671
West Midlands [ref]	977	752 (77%)	225 (23%)
South West	694	538 (78%)	156 (22%)	0.97 (0.77, 1.22)	1.14 (0.85, 1.53)	0.389
Medical history
Number of long-term conditions^e
Mean (SD; n)	1671	3.92 (1.77; 1290)	4.43 (1.90; 381)	1.16 (1.09, 1.24)	1.22 (1.04, 1.43)	0.014
Cambridge multimorbidity score
Mean (SD; n)	1661	2.28 (1.40; 1283)	2.60 (1.49; 378)	1.16 (1.08, 1.26)	1.17 (1.05, 1.30)	0.004
Long-term conditions
Hypertension	1182	925 (78%)	257 (22%)	0.82 (0.64, 1.05)	0.92 (0.67, 1.26)	0.603
Anxiety/depression	802	562 (70%)	240 (30%)	2.23 (1.76, 2.83)	1.67 (1.25, 2.24)	0.001
Painful condition	731	532 (73%)	199 (27%)	1.57 (1.25, 1.98)	1.29 (0.96, 1.73)	0.091
Hearing loss	468	362 (77%)	106 (23%)	0.99 (0.77, 1.28)	1.20 (0.87, 1.64)	0.266
Irritable bowel syndrome	222	158 (71%)	65 (29%)	1.45 (1.061.99)	1.23 (0.84, 1.82)	0.290
Current asthma	295	208 (77%)	87 (29%)	1.54 (1.17, 2.05)	1.35 (0.95, 1.93)	0.093
Diabetes	615	457 (74%)	158 (26%)	1.30 (1.03, 1.64)	1.08 (0.80, 1.45)	0.621
Coronary heart disease	468	364 (78%)	104 (22%)	0.96 (0.74, 1.24)	0.91 (0.66, 1.25)	0.558
Chronic kidney disease	421	316 (75%)	105 (25%)	1.18 (0.91, 1.52)	1.42 (1.02, 1.97)	0.039
Atrial fibrillation	289	224 (78%)	65 (22%)	0.98 (0.72, 1.33)	1.10 (0.75, 1.61)	0.631
Constipation	161	119 (74%)	42 (26%)	1.22 (0.84, 1.77)	1.02 (0.64, 1.62)	0.934
Stroke or transient ischaemic attack (TIA)	235	181 (77%)	54 (23%)	1.01 (0.731.41)	1.20 (0.81, 1.79)	0.369
Chronic obstructive pulmonary disease (COPD)	186	140 (75%)	46 (24%)	1.13 (0.79 to 1.61)	0.96 (0.60, 1.52)	0.858
Connective tissue disorder	195	147 (75%)	48 (25%)	1.12 (0.79, 1.59)	1.28 (0.84, 1.96)	0.255
Cancer	185	150 (81%)	35 (19%)	0.77 (0.52, 1.14)	0.91 (0.57, 1.43)	0.676
Alcohol problems	126	95 (75%)	31 (24%)	1.12 (0.73, 1.70)	1.09 (0.64, 1.85)	0.763
Heart failure	146	106 (73%)	40 (27%)	1.31 (0.90, 1.93)	1.22 (0.76, 1.96)	0.402
Dementia	10	7 (70%)	3 (30%)	1.46 (0.38, 5.67)	3.61 (0.63, 20.74)	0.150
Psychosis or bipolar disorder	16	12 (75%)	4 (25%)	1.13 (0.36, 3.53)	1.25 (0.31, 5.08)	0.754
Epilepsy	0	0	0
Prescribing history and health utility
Number of current medicines^f	1671
<8 [ref]	819	680 (83%)	139 (17%)
≥8	852	610 (72%)	242 (28%)	1.94 (1.53, 2.46)	1.64 (1.21, 2.21)	0.001
Do you know what each of your prescription medicines is for?	1627
Yes [ref]	1475	1156 (78%)	319 (22%)
No	152	101 (66%)	51 (34%)	1.82 (1.28, 2.62)	1.56 (0.99, 2.46)	0.054
Adherence, medicines^g	1667
High adherence [ref]	1128	944 (84%)	184 (16%)
Low or moderate adherence	539	342 (63%)	197 (37%)	2.96 (2.33, 3.74)	2.80 (2.11, 3.31)	<0.0001
Potentially inappropriate prescribing indicators	1661
0-1 [ref]	646	513 (79%)	133 (21%)
≥2	1015	770 (76%)	245 (24%)	1.23 (0.97, 1.56)	1.19 (0.88, 1.61)	0.267
Number of primary care consultations^h	1661
0 [ref]	121	98 (81%)	23 (19%)
1-4	590	480 (81%)	110 (19%)	0.98 (0.59, 1.61)	1.30 (0.69, 2.45)
5-9	563	420 (75%)	143 (25%)	1.45 (0.89, 2.37)	1.58 (0.84, 2.97)
10-19	327	245 (75%)	82 (25%)	1.43 (0.85, 2.39)	1.73 (0.89, 3.37)
≥20	60	40 (67%)	20 (33%)	2.13 (1.05, 4.30)	2.22 (0.91, 5.42)	0.2616
Health-related quality of life
SF-12
Physical health component score (mean, SD, n)	1658	37.21 (12.52, 1279)	30.92 (11.29, 379)	0.96 (0.96, 0.97)	0.96 (0.94, 0.97)	<0.0001
Mental health component score (mean, SD, n)	1658	49.15 (10.75, 1279)	39.83 (11.80, 379)	0.93 (0.92, 0.94)	0.94 (0.93, 0.96)	<0.0001
EQ-5D visual analogue scale (VAS)²²
Mean (SD, n)	1648	65.02 (20.65, 1274)	47.65 (21.52, 374)	0.96 (0.96, 0.97)	0.97 (0.96, 0.97)	<0.0001

MTBQ global score was missing for 40 (2%) of participants (1,671 included in the analysis for table 2).

^aLogistic regression comparing non-high treatment burden (MTBQ score 0-21) and high treatment burden (MTBQ score ≥22).

^bLogistic regression comparing non-high treatment burden (MTBQ score 0-21) and high treatment burden (MTBQ score ≥22), adjusted for age, gender, individual deprivation, ethnicity, education, employment status, living alone, geographical location, Cambridge Multimorbidity Score and number of current medicines. Individual models were used to examine the association between individual long-term conditions and treatment burden, adjusting for age, gender, individual deprivation, ethnicity, education, employment status, living alone, geographical location, number of current medicines and other overall morbidity burden (removing the weighting on the individual condition from the CMS score). The multivariate analyses were conducted on complete cases with no imputation of missing data.

^cP-value for the overall test of association for a multi-category variable in the adjusted analysis.

^dUnemployed includes participants who selected: looking after the home; unable to work due to long-term illness or disability; and unable to work due to carer responsibilities.

^e20 clinically important long-term conditions included in the Cambridge Multimorbidity Score.

^fNumber of current medicines prescribed in the preceding 3 months recorded on the electronic health records.

^gAdherence to medicines was measured using the 10-item Medication Adherence Rating Scale (MARS-10)²⁴ and categorized as low to moderate adherence (0-7) and high adherence (8-10).

^hNumber of primary care consultations include GP and nurse consultations recorded on the electronic health records.

Abbreviations: 95% confidence interval (95% CI), Index of multiple deprivation (IMD).

Following adjustment using multivariable analysis, the magnitude and direction of most associations was generally maintained. Of note, there was particularly strong evidence (p<0.0001) for a positive association between high treatment burden and being unemployed or in paid work compared to being retired (adjusted odds ratio (aOR) 3.10, 95% confidence interval (CI) 2.01 to 4.76; aOR 2.62, 95%CI 1.74 to 3.94 respectively); low/moderate adherence to medication (aOR 2.80, 95%CI 2.11 to 3.31); and poor HRQoL (aOR 0.97, 95%CI 0.96 to 0.97 for a one point increase in VAS; aOR 0.96, 95%CI 0.96 for a one point increase in PCS; aOR 0.94, 95% CI 0.93-0.96 for a one point increase in MCS). Positive associations were also found between high treatment burden and university or higher education (aOR 1.88, 95% CI 1.23 to 2.86, p=0.018); greater levels of polypharmacy (≥8 medicines; aOR 1.64; 95%CI 1.21-2.21, p=0.001); multimorbidity (mean CMS; aOR 1.17; 95% CI 1.05 to 1.30, p value=0.004); anxiety/depression (aOR 1.67; 95% CI 1.25-2.24, p=0.001); and chronic kidney disease (aOR 1.42, 95% CI 1.02 to 1.97, p=0.039). Individuals aged 18-39 years were more likely to report high treatment burden compared to those aged ≥80 years (aOR 7.06, 95% CI 1.55 to 32.3), although overall there was no evidence (p=0.0738) for an association between age and high treatment burden.

The interactions between treatment burden, higher levels of polypharmacy and four key factors (age, multimorbidity, knowledge of medicines, medication adherence) are shown in Table 3. For this analysis, we adjusted for the same factors as the final multivariable model. The findings suggest a clear positive gradient between the interaction term and treatment burden, for all four factors studied. Participants who were younger (<60 years) and had higher levels of polypharmacy (≥8 medicines) had an increased odds of high treatment burden compared to older participants prescribed <8 medicines (aOR 2.01; 95% CI 1.18 to 3.43; p value = 0.0055). There was strong evidence that participants who had higher levels of multimorbidity (CMS ≥median) and greater levels of polypharmacy (≥8 medicines) had greater odds of high treatment burden compared to those with lower levels of multimorbidity and polypharmacy (aOR 2.11, 95% CI 1.49 to 2.97, p=0.0001). Individuals who had poor knowledge of what their medicines were for, and higher levels of polypharmacy had greater odds of high treatment burden compared to individuals who had good knowledge of their medicines and lower levels of polypharmacy (aOR 2.92; 95% CI 1.72 to 4.94; p=0.0007). Participants who had higher levels of polypharmacy and reported low/moderate adherence to treatment, had 4.5-times greater odds of high treatment burden compared to those who reported high adherence and had lower levels of polypharmacy (aOR 4.59; 95% CI 3.01 to 7.01; p value<0.0001).

Table 3.

Analyses of high treatment burden in relation to the interaction between polypharmacy^a and age, multimorbidity, knowledge of medicines and adherence.

	n	Adjusted odds ratio (95%CI)^b	P Value for the overall test
Interaction between polypharmacy and age	1,402
No polypharmacy – ≥60 years [ref]	592
No polypharmacy – <60 years	96	1.63 (0.94, 2.85)
Polypharmacy – ≥60 years	619	1.67 (1.21, 2.31)
Polypharmacy – <60 years	95	2.01 (1.18, 3.43)	0.0055
Interaction between polypharmacy and multimorbidity^c	1,402
No polypharmacy – below median CMS score [ref]	471
No polypharmacy – at/above median CMS score	217	1.33 (0.86, 2.08)
Polypharmacy – below median CMS score	247	1.93 (1.31, 2.85)
Polypharmacy – at/above median CMS score	467	2.11 (1.49, 2.97)	0.0001
Interaction between polypharmacy and knowledge of medicines^c	1367
No polypharmacy – knows what each medicine is for [ref]	636
No polypharmacy – unsure about medicines	37	1.03 (0.40, 2.61)
Polypharmacy – knows what each medicine is for	606	1.43 (1.05, 1.95)
Polypharmacy – unsure about medicines	88	2.92 (1.72, 4.94)	0.0007
Interaction between polypharmacy and adherence to treatment^d	1,398
No polypharmacy – high adherence [ref]	481
No polypharmacy – low or moderate adherence	205	2.82 (1.84, 4.32)
Polypharmacy – high adherence	481	1.65 (1.12, 2.42)
Polypharmacy – low or moderate adherence	231	4.59 (3.01, 7.01)	<0.0001

^aPolypharmacy defined as ≥8 prescribed medicines in the last 3 months.

^cCambridge multimorbidity score calculated using weighted scores for each of the long-term conditions.²⁵.

^dAdherence to medicines was measured using the 10-item Medication Adherence Rating Scale (MARS-10)²⁴ and categorized as low to moderate adherence (0-7) and high adherence (8-10).

The multivariate analyses were conducted on complete cases with no imputation of missing data.

AbbreviationsConfidence interval (CI), Cambridge Multimorbidity Score (CMS).

Discussion

This study of mostly older adults experiencing potentially problematic polypharmacy found that almost a quarter of participants reported high treatment burden. Several sociodemographic factors (e.g. being unemployed or in paid work compared to being retired, and university education), the presence of multimorbidity and several specific health problems (e.g. anxiety/depression), and various aspects of medication use (e.g. increasing numbers of medications, poor knowledge of medicines, low/moderate medication adherence) were independently associated with high treatment burden. The strength of association between medication count and high treatment burden was greatest in younger individuals, and those with multimorbidity, poor medication literacy and poor medication adherence.

Comparison with existing literature

Our findings align well with the wider literature. We have expanded on the current body of knowledge by further exploring the association between treatment burden and knowledge of medicines; the association between treatment burden and multimorbidity; and the interaction between higher levels of polypharmacy, treatment burden and other factors. To interpret the study findings, we have drawn on the cumulative complexity model.⁹ This conceptualises treatment burden as the balance between workload and capacity.

In line with other studies, we found an association between younger age and high treatment burden.^16,27 Furthermore, the association between medication count and high treatment burden was most marked in younger individuals. One explanation for this is that younger people may be less willing to accept the need for multiple medicines and may expect to maintain their health with less effort. As a result, they may have different expectations of how much work should be required to manage their well-being compared to older people, who may be more accustomed to ongoing medical treatments. A further explanation may be younger patients having less capacity to manage the work of looking after their health due to competing priorities, such as work and caring for children and parents. This hypothesis is supported by our finding that people who were retired were less likely to report high burden of treatment.

To our knowledge, this is the first study to investigate the relationship between treatment burden and knowledge of medicines. We found that individuals who were uncertain about their medicines reported higher treatment burden. This aligns with findings elsewhere in the literature of an association between high treatment burden and low health literacy more generally.^8,17 Furthermore, as hypothesised, we found that the strength of association between medication count and treatment burden was greatest in those individuals with poor medication literacy. This may reflect individuals with poor knowledge about their medicines being less able to manage the work of being a patient. This is also supported by the finding that poor medication adherence also modifies the association between treatment burden and medication count, with poor adherence being indicative of excessive patient workload.

We found, as reported elsewhere, a positive association between number of long-term conditions and high treatment burden.^6–8,17,28 However, simple counts fail to capture differences between specific long-term conditions. Our finding that a multimorbidity score (CMS) weighted according to its different component conditions is strongly associated with high treatment burden is therefore an important, albeit expected, new observation, with multiple health problems both increasing the individual’s workload and impairing their physical or mental capacity to cope.²⁵ As hypothesised, we found that the strength of association between medication count and treatment burden was greatest in individuals with high multimorbidity scores.

We also found strong evidence of an association between high treatment burden and anxiety/depression. This has been reported in other studies and may reflect that having a mental health condition impacts on an individual’s capacity to manage the work of looking after their health.^6,7,29,30 Related to this, high treatment burden was also associated with low physical and mental health related quality of life, in keeping with existing literature.^6,7,28 This highlights the negative impact high treatment burden can have on an individual’s wellbeing and the importance of novel interventions to assess, monitor and reduce treatment burden.

In this study, we found an association between being highly educated (University degree or higher compared with no educational qualifications) and high treatment burden. This aligns with the findings of a population health survey in Denmark³¹ but differs from studies in the United States.^29,32 One possible explanation is that highly educated patients may have higher expectations of themselves and the healthcare system.

Strengths and limitations

High quality data, collected using robust processes as part of a large multi-centre trial, are an important strength of this study. We were able to combine survey data with data from the EHR and there were minimal missing data. Rather than relying on self-report of long-term conditions, we captured these data from the EHR. Long-term conditions were based on a list validated for primary care research purposes (CMS), and multimorbidity was quantified using a validated measure.²⁵ Underpinning our work is the choice of a validated patient-reported measure of treatment burden (MTBQ).^6,20 The MTBQ has been translated, culturally adapted and validated in multiple languages, and has been shown to be valid in different cultural contexts.^7,33–35

A limitation of the study is that 97% of participants reported their ethnicity as white, with low representation from global majority ethnicities may limit the generalisability of the findings. Non-English speaking patients were excluded from the trial. This group would be an interesting group to investigate as it is likely they experience high treatment burden, not least due to difficulties navigating UK healthcare services. In keeping with other studies, few participants lived in more deprived areas, further limiting generalisability. These issues highlight the importance of striving for more inclusive research and capturing a wide range of patient voices. Adherence was assessed self-reported without objective validation (e.g. pill counts), possibly leading to an overestimation. Despite adjusting for many covariates, residual confounding cannot be excluded. Findings must also be interpreted in the context of a population recruited as part of a clinical trial and, as this was an observational study, causality cannot be inferred.

Implications for research, practice and policy

Treatment burden is not currently considered by policy makers when reviewing and developing models of care for adults with multimorbidity. We believe our work adds to a growing body of literature calling for recognition and reduction of treatment burden in patients with multiple long-term conditions and polypharmacy. Policies and interventions aimed at optimising medication use should include assessment and management of treatment burden. Our work points to certain groups of patients to target for support with high burden of treatment, including those who are younger, prescribed greater numbers of medicines, have poor knowledge of their medicines, have poor medication adherence, and are diagnosed with anxiety/depression. The findings also support the need to educate patients about their medicines, which may not only improve adherence but also potentially reduce treatment burden. Finally, our findings support the use of the Cambridge Multimorbidity Score to identify patients at risk of high burden of treatment.

Conclusion

This study provides clear evidence for the association between high treatment burden and a number of important sociodemographic and clinical factors, including several aspects of medication use. It is apparent that the strength of association between treatment burden and number of medications is modified by various factors related to patients’ capacity to manage their health. Interventions to reduce treatment burden should target individuals with polypharmacy, multimorbidity, mental health conditions and work and caring commitments. Supporting patients to understand what their medicines are for may reduce treatment burden.

Supplemental Material

Supplemental Material - Development of the Associations between treatment burden, socio-demographics, clinical factors and use of medicines in adults with polypharmacy: A cross-sectional study

Supplemental Material for Associations between treatment burden, socio-demographics, clinical factors and use of medicines in adults with polypharmacy: A cross-sectional study by Polly Duncan, Chloe Gamlin, Deborah McCahon, Daisy Gaunt, Beenish Ashfaq, Rupert A. Payne in Journal of Multimorbidity and Comorbidity.

Footnotes

Acknowledgements

The study was hosted by NHS Bristol, North Somerset and South Gloucestershire Integrated Care Board. Trial delivery was facilitated by the NIHR Clinical Research Network and the Bristol Trials Centre. We would like to thank the patients and practices who participated in the study, members of the study patient and public involvement advisory group, and members of the independent trial steering committee and data monitoring committee.

ORCID iDs

Polly Duncan

Chloe Gamlin

Deborah McCahon

Daisy Gaunt

Beenish Ashfaq

Rupert A. Payne

Ethical considerations

The IMPPP trial protocol and associated study materials were approved by the Wales REC 6, NHS Research Ethics Committee (REC reference 19/WA/0090), and Health Research Authority. Written informed consent was obtained from all participants in the study. The informed consent process included the option for sharing anonymous data for subsequent research. Favourable opinion for the current study was received from the University of Bristol, Health Science Student Research Ethics Committee (Ref: 18975).

Consent to participate

Patients were invited by post. Those who wished to participate completed a consent form and baseline survey and returned both via post to the research team. Only participants who consented to their data being used for further research were included in the current analysis.

Consent for publication

Participants consented to study findings being published.

Funding

The IMPPP trial was funded by the NIHR Health and Social Care Delivery Research (HSDR) programme (ref. 16/118/14). The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care.

Declaration of conflicting interests

PD developed and validated the original Multimorbidity Treatment Burden Questionnaire (MTBQ). RP developed and validated the Cambridge Multimorbidity Scale.

Data Availability Statement

Access to anonymised data may be granted, where appropriate governance approvals are in place, following review via the data.bris University of Bristol Research Data repository ().

STROBE statement

We have included a completed STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) checklist as a supplementary file to ensure transparency and adherence to established reporting guidelines for observational research ().

Supplemental Material

Supplemental material for this article is available online.

References

Eton

Ramalho de Oliveira

Egginton

, et al. Building a measurement framework of burden of treatment in complex patients with chronic conditions: a qualitative study. Patient related outcome measures 2012;3:39-49. doi: 10.2147/prom.s34681 [published Online First: 2012/11/28].

May

Eton

Boehmer

, et al. Rethinking the patient: using Burden of Treatment Theory to understand the changing dynamics of illness. BMC health services research 2014;14:281. doi: 10.1186/1472-6963-14-281 [published Online First: 2014/06/28].

Barnett

Mercer

Norbury

, et al. Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study. Lancet. England: 2012

Elsevier Ltd

2012:37-43.

Salisbury

Johnson

Purdy

, et al. Epidemiology and impact of multimorbidity in primary care: a retrospective cohort study. The British journal of general practice : the journal of the Royal College of General Practitioners 2011;61(582):e12-21.

Chowdhury

Chandra Das

Sunna

, et al. Global and regional prevalence of multimorbidity in the adult population in community settings: a systematic review and meta-analysis. EClinicalMedicine 2023;57:101860. doi: 10.1016/j.eclinm.2023.101860 [published Online First: 20230216].

Duncan

Murphy

Man

, et al. Development and validation of the Multimorbidity Treatment Burden Questionnaire (MTBQ). BMJ Open 2018;8(4):e019413. doi: 10.1136/bmjopen-2017-019413 [published Online First: 2018/04/12].

Marie Hauge

Polly

Mathias

, et al. Danish validation of the Multimorbidity Treatment Burden Questionnaire (MTBQ) and findings from a population health survey: a mixed-methods study. BMJ Open 2022;12(1):e055276. doi: 10.1136/bmjopen-2021-055276.

Morris

Roderick

Harris

, et al. Treatment burden for patients with multimorbidity: cross-sectional study with exploration of a single-item measure. The British journal of general practice : the journal of the Royal College of General Practitioners 2021;71(706):e381-e90. doi: 10.3399/bjgp.2020.0883 [published Online First: 20210429].

Shippee

Shah

May

, et al. Cumulative complexity: a functional, patient-centered model of patient complexity can improve research and practice. Journal of clinical epidemiology 2012;65(10):1041-51. doi: 10.1016/j.jclinepi.2012.05.005.

10.

Gallacher

May

Montori

, et al. Understanding patients' experiences of treatment burden in chronic heart failure using normalization process theory. Annals of family medicine 2011;9(3):235-43. doi: 10.1370/afm.1249.

11.

Payne

Avery

Duerden

, et al. Prevalence of polypharmacy in a Scottish primary care population. Eur J Clin Pharmacol 2014;70(5):575-81. doi: 10.1007/s00228-013-1639-9.

12.

Guthrie

Makubate

Hernandez-Santiago

, et al. The rising tide of polypharmacy and drug-drug interactions: population database analysis 1995-2010. BMC Med 2015;13:74. doi: 10.1186/s12916-015-0322-7.

13.

Duerden

Avery

Payne

. Polypharmacy and medicines optimisation: making it safe and sound: London: The King's Fund 2013.

14.

Duncan

Duerden

Payne

. Deprescribing: a primary care perspective. Eur J Hosp Pharm 2017;24(1):37-42. doi: 10.1136/ejhpharm-2016-000967.

15.

Wallace

McDowell

Bennett

, et al. Impact of Potentially Inappropriate Prescribing on Adverse Drug Events, Health Related Quality of Life and Emergency Hospital Attendance in Older People Attending General Practice: A Prospective Cohort Study. J Gerontol A Biol Sci Med Sci 2017;72(2):271-77. doi: 10.1093/gerona/glw140 [published Online First: 20160727].

16.

Selvakumar

Sivanandy

Ingle

, et al. Relationship between Treatment Burden, Health Literacy, and Medication Adherence in Older Adults Coping with Multiple Chronic Conditions. Medicina (Kaunas) 2023;59(8) doi: 10.3390/medicina59081401 [published Online First: 20230731].

17.

Friis

Lasgaard

Pedersen

, et al. Health literacy, multimorbidity, and patient-perceived treatment burden in individuals with cardiovascular disease. A Danish population-based study. Patient Educ Couns 2019;102(10):1932-38. doi: 10.1016/j.pec.2019.05.013. Epub 19 May 15.

18.

Payne

Blair

Caddick

, et al. Improving Medicines use in People with Polypharmacy in Primary Care (IMPPP): Protocol for a multicentre cluster randomised trial comparing a complex intervention for medication optimization against usual care [version 1; peer review: 2 approved]. NIHR Open Research 2022;2(54) doi: 10.3310/nihropenres.13285.1.

19.

Abel

Barclay

Payne

. Adjusted indices of multiple deprivation to enable comparisons within and between constituent countries of the UK including an illustration using mortality rates. BMJ Open 2016;6(11):e012750. doi: 10.1136/bmjopen-2016-012750 [published Online First: 20161115].

20.

Duncan

Scott

Dawson

, et al. Further development and validation of the Multimorbidity Treatment Burden Questionnaire (MTBQ). BMJ Open 2024;14(4):e080096. doi: 10.1136/bmjopen-2023-080096 [published Online First: 20240410].

21.

Brazier

Roberts

. The estimation of a preference-based measure of health from the SF-12. Medical care 2004;42(9):851-9. doi: 10.1097/01.mlr.0000135827.18610.0d.

22.

EuroQol--a new facility for the measurement of health-related quality of life. Health Policy 1990;16(3):199-208. [published Online First: 1990/11/05].

23.

Thompson

Kulkarni

Sergejew

. Reliability and validity of a new Medication Adherence Rating Scale (MARS) for the psychoses. Schizophr Res 2000;42(3):241-7.

24.

Fialko

Garety

Kuipers

, et al. A large-scale validation study of the Medication Adherence Rating Scale (MARS). Schizophr Res 2008;100(1-3):53-9. doi: 10.1016/j.schres.2007.10.029 [published Online First: 20071220].

25.

Payne

Mendonca

Elliott

, et al. Development and validation of the Cambridge Multimorbidity Score. CMAJ 2020;192(5):E107-E14. doi: 10.1503/cmaj.190757.

26.

StataCorp. Stata Statistical Software: Release 18. College Station, TX: StataCorp LLC. , 2023.

27.

Rogers

Yost

Rosedahl

, et al. Validating the Patient Experience with Treatment and Self-Management (PETS), a patient-reported measure of treatment burden, in people with diabetes. Patient related outcome measures 2017;8:143-56. doi: 10.2147/prom.S140851 [published Online First: 20171109].

28.

Eton

Anderson

Cohn

, et al. Risk factors for poor health-related quality of life in cancer survivors with multiple chronic conditions: exploring the role of treatment burden as a mediator. Patient related outcome measures 2019;10:89-99. doi: 10.2147/prom.S191480 [published Online First: 20190321].

29.

Boyd

Wolff

Giovannetti

, et al. Healthcare task difficulty among older adults with multimorbidity. Medical care 2014;52 Suppl 3:S118-25. doi: 10.1097/MLR.0b013e3182a977da [published Online First: 2014/02/25].

30.

Gibbons

Kenning

Coventry

, et al. Development of a multimorbidity illness perceptions scale (MULTIPleS). PLoS ONE 2013;8(12):e81852. doi: https://doi.org/10.1371/journal.pone.0081852

31.

Pedersen

Duncan

Lasgaard

32.

Eton

Linzer

Boehm

, et al. Deriving and validating a brief measure of treatment burden to assess person-centered healthcare quality in primary care: a multi-method study. BMC Fam Pract 2020;21(1):221. doi: 10.1186/s12875-020-01291-x [published Online First: 20201028].

33.

Dou

Huang

Duncan

, et al. Translation, cultural adaptation and validation of the Chinese Multimorbidity Treatment Burden Questionnaire(C-MTBQ): a study of older hospital patients. Health and quality of life outcomes 2020;18(1):194. doi: 10.1186/s12955-020-01395-z.

34.

Schulze

Breckner

Duncan

, et al. Adaptation and validation of a German version of the Multimorbidity Treatment Burden Questionnaire. Health and quality of life outcomes 2022;20(1):90. doi: 10.1186/s12955-022-01993-z.

35.

Samorinha

Saidawi

Duncan

, et al. Translation, cross-cultural adaptation and validation of the Arabic multimorbidity treatment burden questionnaire (MTBQ-A): A study of adults with multimorbidity. Research in Social and Administrative Pharmacy 2024;20(4):411-18. doi: https://doi.org/10.1016/j.sapharm.2024.01.001

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.19 MB

Associations between treatment burden,socio-demographics,clinical factors and use of medicines in adults with polypharmacy: A cross-sectional study

Abstract

Background

Aim

Design and setting

Methods

Results

Conclusion

Keywords

Introduction

Methods

Study setting

Study population, eligibility criteria and recruitment

Survey data

Data from electronic GP records

Analysis

Interaction between polypharmacy and age, multimorbidity, knowledge of medicines and adherence

Missing data

Results

Discussion

Comparison with existing literature

Strengths and limitations

Implications for research, practice and policy

Conclusion

Supplemental Material

Supplemental Material - Development of the Associations between treatment burden, socio-demographics, clinical factors and use of medicines in adults with polypharmacy: A cross-sectional study

Footnotes

Acknowledgements

ORCID iDs

Ethical considerations

Consent to participate

Consent for publication

Funding

Declaration of conflicting interests

Data Availability Statement

STROBE statement

Supplemental Material

References

Supplementary Material