Core outcome set for trials aimed at improving appropriate polypharmacy

in older people in primary care

Running title: COS for trials aimed at improving polypharmacy

Audrey Rankin, PhD,1 Cathal A. Cadogan, PhD,2 Cristín Ryan, PhD,2 Barbara Clyne, PhD,3

Susan M. Smith, PhD,3 Carmel M. Hughes, PhD1

Corresponding Author:

Professor Carmel M. Hughes, School of Pharmacy, Queen's University Belfast, 97 Lisburn

Road, Belfast BT9 7BL, UK.

Tel./Fax: +44 28 90 972147 / +44 28 9097 2155 E-mail: c.hughes@qub.ac.uk

Abstract word count: 272

Main text word count: 2595

Number of figures: 1

Number of tables: 2

Number of references: 29

Funding sources: HRB Centre for Primary Care Research under grant number HRC/2014/1,

Royal College of Surgeons in Ireland (RCSI), Dublin, Ireland.

1 School of Pharmacy, Queen’s University Belfast, UK.

2 School of Pharmacy, Royal College of Surgeons in Ireland, Ireland.

3 Department of General Practice, Royal College of Surgeons in Ireland, Ireland.

1

1

2

3

4

5

67

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

ABSTRACT

BACKGROUND/OBJECTIVES: Intervention studies have sought to improve appropriate

polypharmacy in older people (≥65 years), yet heterogeneity in outcome measurements

across studies has hindered efforts to synthesise results. One proposed solution is the

development and use of a core outcome set (COS). The aim of this study was to develop a

COS for use in effectiveness trials of interventions aiming to improve appropriate

polypharmacy in older people in primary care.

DESIGN: Standard COS development methodology was followed, comprising: identification

of outcomes in studies from an update of a Cochrane systematic review and previously

collected qualitative data; and an online Delphi consensus exercise involving three rounds.

PARTICIPANTS: An international panel of 160 stakeholders comprising 120 healthcare

experts and a public participant panel of 40 older people.

MEASUREMENTS: Outcomes identified from studies included in the Cochrane review and

secondary analysis of previously collected qualitative data were scored on a 9-point Likert

scale using the GRADE scoring system anchored between 1 (not important) and 9 (critical).

Consensus criteria for the COS were defined as ≥70% of participants scoring the outcome as

‘critical’ and ≤15% scoring the outcome as ‘not important.’

RESULTS: 29 outcomes identified from the Cochrane review and existing qualitative data

were included in the Delphi exercise. The final COS comprised 16 outcomes. The seven

highest ranked outcomes were serious adverse drug reactions, medication appropriateness,

falls, medication regimen complexity, quality of life, mortality and medication side effects.

CONCLUSION: A COS for interventions aiming to improve appropriate polypharmacy for

older people in primary care has been developed. Future work will focus on identifying

appropriate tools to measure these outcomes and testing the implementation of the COS.

2

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

Key words: Core outcome set, Polypharmacy, Older people, Primary care, Consensus

exercise

3

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

INTRODUCTION

Globally, the number of older people (≥65 years) is growing and is predicted to reach nearly

1.5 billion by 2050, representing almost 25% and 15% of the population in the United

Kingdom (UK) and United States, respectively.1,2 The use of multiple medicines

(polypharmacy) in older populations is commonplace.3 Obtaining a balance between the

prescribing of ‘too many’ medicines (inappropriate polypharmacy) and ‘many’ medicines

(appropriate polypharmacy) is challenging, particularly when prescribing for older people

with multimorbidity.4,5 Polypharmacy is associated with an increased risk of potentially

inappropriate prescribing, whereby the negative effects associated with medication

prescribing outweigh the potential benefits.6

Studies to improve appropriate polypharmacy in older people often differ in the

outcomes reported, resulting in an inability to form conclusions about intervention

effectiveness for specific outcomes (e.g. hospitalisations).7 The Core Outcome Measures for

Effectiveness Trials (COMET) initiative has proposed the development and reporting of a

core outcome set (COS) as one method for addressing this problem.8,9 A COS is an agreed

standardised set of outcomes which should be measured and reported as a minimum in all

trials in a specific clinical area.8

The aim of this study was to develop a COS for use in trials investigating the

effectiveness of interventions aimed at improving appropriate polypharmacy in older people

in primary care.

METHODS

Development of the COS followed the COMET initiative methodology,8 and has been

reported according to the Core Outcome Set-STAndards for Reporting (COS-STAR)

4

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

guidelines.10 The study involved two phases: (1) Generating and refining a long-list of

outcomes and (2) Delphi consensus exercise. A Project Steering Group (PSG) comprising

staff members [e.g. academic general practitioners (GPs), pharmacists, research staff] from

Queen’s University Belfast and the Royal College of Surgeons in Ireland oversaw the COS

development process.

Phase 1: Generating and refining a long-list of outcomes (July 2016-Sept 2016)

This phase involved three steps: (1) identifying outcomes used in previous studies by

updating a Cochrane systematic review7 (2) identifying outcomes from previously collected

qualitative data11,12 and (3) initial screening of outcomes from steps 1 and 2, by the PSG.

1. Identification of outcomes used in previous intervention studies

Outcomes were extracted from 20 studies included in an ongoing update of a

Cochrane review.7 Studies were included if they aimed to improve appropriate

polypharmacy in older people, in any healthcare setting (e.g. hospital, community,

nursing homes) and included a validated measure of potentially inappropriate

prescribing (e.g. Beers criteria13).

2. Secondary analysis of qualitative data

Secondary analysis was conducted on an existing qualitative dataset involving semi-

structured interviews (15 GPs, 15 community pharmacists), and focus groups (50

older patients receiving polypharmacy).11,12 Data were extracted on outcomes deemed

of importance to participants for future polypharmacy-focussed intervention.

3. Initial screening of the long-list of outcomes

The outcomes identified from steps 1 and 2 were reviewed by the PSG to produce a

long-list of outcomes for the Delphi consensus exercise (Phase 2). Refinement

involved identifying any process measures (i.e. outcomes relating to intervention

5

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

implementation), duplicate outcomes, and outcomes outside the scope of the COS (i.e.

irrelevant to the specific interventions and study population). Outcomes were then

categorised into eight ‘outcome themes’ by the PSG including: ‘medication-related

outcomes’, ‘healthcare utilisation’, ‘patient-related outcomes’, ‘satisfaction’, ‘adverse

effects or harms’, ‘clinical outcomes’, ‘knowledge’ and ‘resource use’.

Phase 2: Delphi consensus exercise (Nov 2016–May 2017)

Phase 2 involved conducting a Delphi consensus exercise with key stakeholders, to reach

consensus regarding the outcomes to include in the COS.8 This involved: (1) identification of

panel members, (2) development of the Delphi questionnaire and (3) the main consensus

exercise.

1. Identification of panel members

No formal guidelines exist as to how many participants should form a Delphi panel in

respect of COS development;8,14 previous studies have included between 10 and 300

participants.14 To ensure elicitation of a range of opinions, a target of 160 participants

(40 public participants, 120 international experts) were recruited for the Delphi panel.

Selection of expert panel members involved discussions amongst the PSG and

compilation of lists of experts with knowledge relevant to the scope of the COS

(Table S1). Individuals were selected if they were researchers and/or clinicians with

expertise relating to the care and/or prescribing for older people (based on the PSG’s

knowledge of their research profiles and publication records), were an editor of a key

journal in geriatric medicine or were individuals representing the interests of older

people (i.e. individuals from support groups or charities). Selection of the public

participant panel was facilitated through publicity (e.g. newsletters, social media)

within charities and organisations in Northern Ireland. Public participants aged ≥65

6

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

years and resident in the community were eligible for inclusion. Potential participants

were emailed invitations, which included participant information sheets and consent

forms which were signed and returned. Experts were also asked to suggest colleagues

who could be included as Delphi panel members. Recruitment was conducted in

staggered batches until the required sample size was achieved.

2. Development of Delphi questionnaire

Outcomes identified in Phase 1 were listed alphabetically in the questionnaire under

each ‘outcome theme’ to avoid potential weighting effects. Detailed instructions on

questionnaire completion were included, with a plain English definition of each

outcome. Participants were asked to rate each outcome on how important it was to

measure in studies aimed at improving appropriate polypharmacy using a 9-point

Likert scale anchored between 1 (‘limited importance’) and 9 (‘critical importance’),

based on the Grading of Recommendations Assessment, Development and Evaluation

(GRADE) scoring system,15 as recommended by COMET.10 There were three scoring

categories: a score of 1-3 represented an outcome of limited importance, 4-6

represented an outcome that was important but not critical and 7-9 represented an

outcome that was critical. An ‘unable to score’ option was included for those unsure

how to rate an outcome. At the end of Round 1, participants were also asked to

suggest additional outcomes for inclusion in the COS. The questionnaire was piloted

with a convenience sample (n=5) of researchers based at the School of Pharmacy,

Queen’s University Belfast to check for face and content validity. Responses from the

pilot questionnaire were positive and, hence, no changes were deemed necessary.

3. Delphi consensus exercise

The consensus exercise encompassed three rounds of Delphi questionnaires as

recommended by the COMET initiative,8 distributed using a web-based survey tool

7

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

(Survey-Gizmo®) (Supplementary Figure 1). All consenting participants were emailed

a web-link to access the questionnaire. Only respondents who completed the previous

round were invited to complete the next round. Round 2 consisted of all outcomes

from Round 1, along with any additional outcomes suggested by the Delphi panel.

The third round consisted of outcomes for which no consensus had been reached in

the second round. In Rounds 2 and 3, participants were emailed their individual

scores, together with group feedback (i.e. the number of participants scoring within

each GRADE category) (Supplementary Figure 1). Participants were asked to

reconsider their own scores in light of the group response when scoring outcomes in

Rounds 2 and 3.

Data analysis

All statistical analysis was performed using SPSS 22.0. Consensus criteria were specified a

priori; any outcome with a rating of 7–9 by ≥70% and 1–3 by ≤15% of the panel was to be

included in the COS, while any outcome with a rating of 1-3 by ≥70% and 7-9 by ≤15% of

the panel was to be excluded.8 All other combinations indicated that no consensus had been

achieved for the outcome. A higher threshold of ≥75% of the panel rating 7–9 and ≤25%

rating 1–3 was applied if a higher proportion of outcomes than expected were rated critical

(based on the PSG’s judgement, and giving due consideration to current COMET and

Cochrane Collaboration recommendations regarding outcomes).7,16 Round 1 responses were

analysed according to the number of participants scoring each outcome within the GRADE

criteria (i.e. 1-3, 4-6 or 7-9) for the purpose of group feedback in Round 2. Responses from

Round 2 were analysed using the same consensus criteria as Round 1. Items for which

consensus was not achieved were carried forward to Round 3, where responses were analysed

with the more stringent criteria. The final COS was then further categorised into overarching

8

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

‘outcome themes’ of health-related areas (e.g. knowledge) by the PSG.17 The seven highest

ranking outcomes (from those included in the final COS from Rounds 2 and 3) were also

determined based on the percentage of individuals scoring an outcome as ‘critical’ (7-9). The

rationale behind this deviation from the protocol was that the median number of outcomes

reported within Cochrane reviews was seven.16 The completed COS-STAR checklist has been

included in the supplementary tables (Table S2).10

RESULTS

In Phase 1, review of all data sources identified 54 potential outcomes: 32 outcomes from the

Cochrane review7 and 22 outcomes from the qualitative data.11,12 Initial screening of these

outcomes by the PSG resulted in a final long-list of 29 outcomes (Figure 1).

In Phase 2, 152 (41 public, 111 expert) of 163 invited participants (93.3%) completed

Round 1 (Table 1 and S3). Demographic details of the Delphi panel members are provided in

Table 1. The expert panel consisted of pharmacists, researchers and doctors, residing in

Australia, North America and Europe, with the majority of participants located within the UK

(40.5%), Ireland (9.0%), Canada (8.1%) and Spain (8.1%). An additional 29 outcomes were

suggested, after review by the PSG most were considered to overlap with existing outcomes

or to be outside the scope of the COS (Table S4). Four new outcomes were included in the

second round of the Delphi questionnaire [‘medication regimen complexity’, ‘patient

perception of treatment (or medication) burden’, ‘caregiver burden’ and ‘carers’ satisfaction

with the prescribing of many medicines (polypharmacy)’], resulting in 33 outcomes going

forward into Round 2.

Round 2 questionnaires were completed by 140 (38 public, 102 expert) of 152

participants (92.1%). Analysis of Round 2 data (using the a priori consensus criteria) resulted

in 12 outcomes being included in the final COS, and the remaining 21 outcomes for which

9

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

consensus was not reached going forward into Round 3 (Table S5). Round 3 questionnaires

were completed by 127 (35 public, 92 expert) of 140 participants (90.7%). Application of the

more stringent consensus criteria resulted in four additional outcomes being included in the

final COS; the remaining 17 outcomes were excluded (Table S6), resulting in a final COS

consisting of 16 outcomes, across six overarching themes (Table 2). The seven highest

ranking outcomes were: serious adverse drug reactions, medication appropriateness, falls,

medication regimen complexity, quality of life, mortality and medication side effects. There

were some differences in scoring between the public and expert panels. For example, in

Round 3, the outcomes ‘pain’ and ‘patients’ satisfaction with care provided’, were all scored

highly (i.e. these outcomes reached the a priori consensus criteria for inclusion in the COS)

by the public participants but not by the expert panel (S4-6).

DISCUSSION

This study has followed formal methodological guidance, involving key stakeholders to

develop a COS for use in effectiveness studies aiming to improve appropriate polypharmacy

in older people in primary care.8 The 16 outcomes identified, with priority given to the seven

highest ranking outcomes, can be used in future studies within the scope of the COS. The

adoption of this COS will streamline the outcomes routinely measured in trials investigating

appropriate polypharmacy in older people in primary care.

Polypharmacy management in older adults is an active area of research, with

numerous calls for trials to consider the relevance of outcomes selected to assess

interventions targeting appropriate polypharmacy in older people.18,19 Indeed, alongside the

development of the current COS for use in polypharmacy interventions, a number of other

COSs are being developed in the area of pharmaceutical care (e.g. optimisation of prescribing

in older adults in care homes,20 medication reviews in older people in any healthcare

10

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

setting21). Each COS has a unique scope, but collectively, their implementation will add

rigour to effectiveness studies in pharmaceutical care. This will ultimately facilitate

comparison and synthesis of outcome data across studies, thereby helping to determine which

interventions work and inform both clinical decision-making and health policy.22

This study has several strengths. Firstly, we followed COS development guidelines, as

outlined by the COMET initiative,8 alongside the COS-STAR guidelines,10 which detail the

items that COS developers should report (Table S2). Secondly, during Phase 1, the

identification of outcomes was not restricted to the results from the Cochrane review.

Outcomes were extracted from existing qualitative data (involving GPs, pharmacists,

patients), which elicited views on potential outcomes,11,12 resulting in the opinions of a broad

range of stakeholders being included in the development process.23,24 Thirdly, the Delphi

panel convened for this study included experts with a diverse range of nationalities, and

public participants. In-line with the scope of this COS, it was important to involve older

people to facilitate a move away from researcher-only selected outcomes.25 The value of

public participants’ involvement is evidenced within the final round of the consensus exercise

whereby the outcome ‘patients’ knowledge’ would not have been included in the final COS if

only experts had been included. Finally, a high response rate across the three Delphi rounds

(93.3%, 92.1% and 90.7%, respectively), was achieved and is similar to that reported in other

COSs.20

Some limitations must be noted. Firstly, participants in the Delphi panel were

confined to English speakers, while public participants were only sampled from Northern

Ireland. It is, however, unclear whether including panel members outside of this remit would

have resulted in different outcomes being selected for inclusion in the COS. Secondly, due to

financial reasons we were unable to supplement the Delphi rounds with a face-to-face

consensus meeting, with other COS studies not conducting meetings for similar reasons.20,21

11

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

Finally, the number of outcomes included in the COS could be considered relatively large;

however, this may be related to the scope of the COS, since a wide range of multi-faceted

approaches to interventions could be provided to improve appropriate polypharmacy,7 which

affect resource use, medications and the patient.

We have endeavoured to generate a concise set of outcomes that would be usable to

researchers by implementing more stringent consensus criteria (≥75% scoring 7-9) in Round

3. Furthermore, we recognise that having many outcomes may be impractical and highlight

the seven highest ranking outcomes, (i.e. in-line with current COMET and Cochrane

Collaboration recommendations),7,16 suggesting that these should be priority outcomes; with

the remaining outcomes included depending on the specific interventions or underlying

theoretical frameworks.

This work has identified which outcomes should be considered for inclusion in trials

focusing on improving appropriate polypharmacy in older people in primary care. The next

crucial step will be to determine how these outcomes should be measured. For example,

within the updated Cochrane review, three different measurement tools were used [the 15D26,

the 12-Item Short-Form Health Survey (SF-12)27, 36-item Short-Form Health Survey (SF-

36)28] to measure quality of life. Application of the COnsensus-based Standards for the

selection of health Measurement INstruments (COSMIN) checklist,29 will be used to inform

the selection of the most appropriate instrument where more than one instrument is available.

In cases where an appropriate tool does not exist for use in older people, existing tools may

need to be adapted and validated within this population or new tools may be need to be

developed.

CONCLUSION

12

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

301

302

303

304

305

306

This work has identified a list of 16 outcomes which should be considered for inclusion in

effectiveness studies aimed at improving appropriate polypharmacy in older people in

primary care. The implementation of this COS may benefit patients and healthcare providers

by enabling the evaluation of current practice and facilitating evidence synthesis across

studies. Future work should determine the most appropriate way to measure each outcome

included in this COS.

ACKNOWLEDGMENTS

We wish to thank all the participants in this study. We would also like to acknowledge

AgeNI, Patient and Client Council, the Commissioner for Older People for Northern Ireland

(COPNI), Volunteer Now NI and The University of the Third Age (U3A) for their assistance

in the recruitment of public participants.

Funding Source: This work is part of a wider research programme funded by the HRB

Centre for Primary Care Research under grant number HRC/2014/1, Royal College of

Surgeons in Ireland (RCSI), Dublin, Ireland.

Conflict of Interest: None of the authors have any conflicts to release.

Author Contributions: AR: Study concept and design, acquisition of subjects and data,

analysis and interpretation of data, preparation of manuscript and final approval of version to

be published. CC, CR, BC, SS and CH: Study concept and design, member of PSG, critically

revised the manuscript and final approval of version to be published.

Sponsor’s Role: The HRB Centre for Primary Care Research provided funds for this

research but had no role in the conduct of the study, analyses, or production of the

manuscript.

13

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

Study Registration: The study was prospectively registered with the Core Outcome

Measures in Effectiveness Trials (COMET) initiative (registration number 933 available at

www.comet-initiative.org/studies/details/933).

Ethics approval and consent to participate: The study has been approved by the School of

Pharmacy Ethics Committee (Queen’s University Belfast) reference: 021PMY2016. All

interview and focus group participants gave written informed consent.

REFERENCES

1. World Health Organisation. Global health and ageing. Geneva: 2011. Available at

http://www.who.int/ageing/publications/global_health.pdf. Accessed June 06, 2017.

2. Office for National Statistics. Statistical Bulletin: Older people’s day 2010. Available at

http://webarchive.nationalarchives.gov.uk/20160105160709/http://ons.gov.uk/ons/rel/

mortality-ageing/focus-on-older-people/older-people-s-day-2010/index.html. Accessed

April 10, 2017.

3. Barnett K, Mercer SW, Norbury M. et al. Epidemiology of multimorbidity and

implications for health care, research, and medical education: A cross-sectional study.

The Lancet 2012;380(9836):37-43.

4. Kantor ED, Rehm CD, Haas JS et al. Trends in Prescription Drug Use Among Adults in

the United States From 1999-2012. JAMA 2015;314(17):1818-1830.

5. Moriarty F, Hardy C, Bennett K et al. Trends and interaction of polypharmacy and

potentially inappropriate prescribing in primary care over 15 years in Ireland: a repeated

cross-sectional study. BMJ open 2015;5(9):e008656.

6. Cahir C, Fahey T, Teeling M et al. Potentially inappropriate prescribing and cost

outcomes for older people: A national population study. Br J Clin Pharmacol

2010;69(5):543-552.

14

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

350

351

352

353

354

7. Patterson SM, Cadogan CA, Kerse N et al. Interventions to improve the appropriate use

of polypharmacy for older people. Cochrane Database Syst Rev 2014;10:CD008165.

8. Williamson PR, Altman DG, Blazeby JM et al. Developing core outcome sets for clinical

trials: Issues to consider. Trials 2012;13(1):132.

9. Prinsen CA, Vohra S, Rose MR et al. Core outcome measures in effectiveness trials

(COMET) initiative: Protocol for an international Delphi study to achieve consensus on

how to select outcome measurement instruments for outcomes included in a 'core

outcome set'. Trials 2014;15(1):247.

10. Kirkham JJ, Gorst S, Altman DG et al. Core Outcome Set–STAndards for Reporting: The

COS-STAR Statement. PLoS Med 2016;13(10):e1002148.

11. Cadogan CA, Ryan C, Francis JJ et al. Improving appropriate polypharmacy for older

people in primary care: Selecting components of an evidence-based intervention to target

prescribing and dispensing. Implement Sci 2015;10(1):161.

12. Patton DE, Cadogan CA, Ryan C et al. Improving adherence to multiple medications in

older people in primary care: selecting intervention components to address patient-

reported barriers and facilitators. Health Expectations 2017;000:1-11.

13. American Geriatrics Society 2015 Beers Criteria Update Expert Panel. American

Geriatrics Society 2015 updated Beers criteria for potentially inappropriate medication

use in older adults. J Am Geriatr Soc 2015;63(11):2227–46.

14. Sinha IP, Smyth RL, Williamson PR. Using the Delphi technique to determine which

outcomes to measure in clinical trials: Recommendations for the future based on a

systematic review of existing studies. PLoS Med 2011;8(1):e1000393.

15. Guyatt GH, Oxman AD, Kunz R et al. GRADE guidelines: 2. framing the question and

deciding on important outcomes. J Clin Epidemiol 2011;64(4):395-400.

15

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

16. Williamson PR, Altman DG, Bagley H et al. The COMET Handbook: version 1.0. Trials

2017;18(3):280.

17. Macefield RC, Jacobs M, Korfage IJ et al. Developing core outcomes sets: methods for

identifying and including patient-reported outcomes (PROs). Trials 2014:15(1):49.

18. Cadogan CA, Ryan C, Hughes CM. Appropriate polypharmacy and medicine safety:

when many is not too many. Drug safety 2016;39(2):109-16.

19. Kjeldsen LJ, Olesen C, Hansen MK et al. Clinical Outcomes Used in Clinical Pharmacy

Intervention Studies in Secondary Care. Pharmacy 2017;5(2):28.

20. Millar AN, Daffu-O’Reilly A, Hughes CM et al. Development of a core outcome set for

effectiveness trials aimed at optimising prescribing in older adults in care homes. Trials

2017;18(1):175.

21. Spinewine A, Beuscart JB, Dalleur O. Development of a core outcome set for medication

review in older people. Available at http://www.comet-initiative.org/studies/details/806.

Accessed October 05, 2016.

22. Singal AG, Higgins PDR,Waljee AK. A primer on effectiveness and efficacy trials. Clin

Trans Gastroenterol 2014;5:e45.

23. Young B, Bagley H. Including patients in core outcome set development: issues to

consider based on three workshops with around 100 international delegates. Research

Involvement and Engagement 2016;2(1):25.

24. Keeley T, Williamson P, Callery P et al. The use of qualitative methods to inform Delphi

surveys in core outcome set development. Trials 2016;17(1):230.

25. Gorst SL, Gargon E, Clarke M et al. Choosing Important Health Outcomes for

Comparative Effectiveness Research: An Updated Review and Identification of Gaps.

PloS one 2016;11(12):e0168403.

16

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

401

402

26. Sintonen H. The 15D instrument of health-related quality of life: properties and

applications. Ann Med 2001;33:328–36.

27. Ware J, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of

scales and preliminary tests of reliability and validity. Med Care 1996;34:220–33.

28. Ware JE, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). Med Care

1992;30:473-483.

29. Mokkink LB, Terwee CB, Patrick DL et al. The COSMIN checklist for assessing the

methodological quality of studies on measurement properties of health status

measurement instruments: an international Delphi study. Quality of Life Research

2010;19(4):539-549.

LEGENDS

Figure 1 Core Outcome Set (COS) development process

Table 1 Demographic profile of expert and public participants

Table 2 Final Core Outcome Set (COS) for effectiveness trials aimed at improving

appropriate polypharmacy in older people in primary care

SUPPORTING INFORMATION

Additional Supporting Information may be found in the online version of this article:

Supplementary Figure 1 Example of the Delphi questionnaire from Round 2

Supplementary Table 1 Expert panel members

Supplementary Table 2 Core Outcome Set-STandards for Reporting (COS-STAR) checklist

Supplementary Table 3 Delphi questionnaire Round 1 results (N=152)

Supplementary Table 4 Screening of additional outcomes suggested by Delphi panel in

Round 1

17

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

427

Supplementary Table 5 Delphi questionnaire Round 2 results (N=140)

Supplementary Table 6 Delphi questionnaire Round 3 results (N=127)

Hearing Impairment and

18

428

429

430

431

432

433

434

435

436

437

438

Incident Dementia: Findings

19

439

440

441

442

from theEnglish Longitudinal

20

443

444

445

446

447

Study of Age

21

448

449

22

Table 1 Demographic profile of expert and public participantsRound 1 Round 2 Round 3

Expert participants n=111 n=102 n=92GenderMale [n (%)] 48 (43.2) 43 (42.2) 40 (43.5)Female [n (%)] 63 (56.8) 59 (57.8) 52 (56.5)

Continent of residence [n (%)]Australia 9 (8.1) 8 (7.8) 7 (7.6)Europe 91 (82.0) 84 (82.4) 76 (82.6)North America 11 (9.9) 10 (9.8) 9 (9.8)

Professional area [n (%)]*

Doctor 24 (21.6) 22 (21.6) 20 (21.7)Nurse 2 (1.8) 2 (2.0) 1 (1.1)Pharmacist 71 (64.0) 66 (64.7) 61 (66.3)Researcher 51 (45.9) 50 (49.0) 47 (51.1)Journal editor 8 (7.2) 8 (7.8) 8 (8.7)Other† 8 (7.2) 6 (5.9) 6 (6.5)

Years of experience (Mean ± SD) 19.44 ± 9.7 18.96 ± 9.6 18.81 ± 9.8

Public participants n=41 n=38 n=35GenderMale [n (%)] 20 (48.8) 19 (50.0) 18 (51.4)Female [n (%)] 21 (51.2) 19 (50.0) 17 (48.6)

Age (Mean ± SD) 70.34 ± 4.3 70.63 ± 4.3 70.14 ± 3.9

County of residenceNorthern Ireland [n (%)] 41 (100) 38 (100) 35 (100)* Percentages do not add up to 100% as some experts’ selected more than one professional area† Other professional areas included policy makers, healthcare / pharmacy consultants and representatives from older persons’ support groups and charities

451

452

453

454

455

456

457

23

Table 2 Final Core Outcome Set (COS) for effectiveness trials aimed at improving appropriate polypharmacy in older people in primary careThemes Outcomes† DefinitionAdverse effects or harm

Falls* An unexpected event in which the patient comes to rest on the ground, floor or lower level

Serious adverse drug reactions*

Any unexpected effect of treatment that results in death, or is life-threatening, requires admissions to hospital or a longer than expected hospital stay, or results in disability

Clinical outcomes

Mortality (all cause)* The death of a patient for any reason

Healthcare utilisation

Hospitalisations Admission or re-admission to hospital for treatment or monitoring

Knowledge Patients’ knowledge Patients’ knowledge of their medication and/or conditionMedication-related outcomes

Adherence The extent to which a patients’ medication-taking corresponds to agreed recommendations from a healthcare provider

Medication appropriateness*

Where medicines have been prescribed in accordance to the best available evidence and are suitable for a patient taking their medical history and co-morbidities (i.e. the presence of one or more medical condition) into consideration. As measured by a validated assessment of prescribing appropriateness (e.g. STOPP/START criteria or Beers criteria)

- Clinically significant drug interactions

The effect of a drug can be changed by another drug, herbal medicines, food or drink and this can lead to a change or complication in the patients’ condition

- The number of ‘regular’ medicines prescribed

The total number of ‘regular’ medications that a patient has been prescribed (i.e. a prescribed medication that is scheduled or part of a repeat prescription), which would not include over-the-counter and herbal products if used regularly)

- Therapeutic duplication

Therapeutic duplication describes a situation where a patient is prescribed two (or more) medicines of the same drug-class / pharmacological class, (e.g. a patient is prescribed two beta-blockers at the same time) which may increase the risk of adverse events

Medication regimen complexity*

A measure of how complicated patients find taking their medicines such as the number of medicines, how often they need to be taken and how they should be taken

Medication side effects*

A side effect can be described as an undesirable effect of a medication, either when taken or when it is withdrawn (stopped)

Prescribing errors Errors made in the prescribing of patients’ regular or ‘when required’ medication

Patient-related outcomes

Cognitive functioning

Patients' cognitive function (e.g. memory, attention, language, confusion, reasoning, orientation to time and place)

Quality of life* The standard of health, comfort and happiness experienced by an individual, including quality of life relating to medication use

- Patient perception of treatment (or medication) burden

An individual's perception of the effect on functioning and wellbeing that may be caused by exposure to treatment including the use of many medicines (polypharmacy)

† Seven highest ranking outcomes across all identified themes based on the % of panel members rating the outcome as ‘critical’ in Rounds 2 and 3* Indicates the seven highest ranking outcomes across all identified themes- Indicates where closely related outcomes were grouped together, with the main outcome underlined