BowEbrah606pb.qxd 1/30/2007 2:23 PM Page HANDBOOK OF HEALTH RESEARCH METHODS Investigation, Measurement and Analysis This handbook helps researchers to plan, carry out, and analyse health research, and evaluate the quality of research studies The book takes a multidisciplinary approach to enable researchers from different disciplines to work side-by-side in the investigation of population health, the evaluation of health care, and in health care delivery Handbook of Health Research Methods is an essential tool for researchers and postgraduate students taking masters courses, or undertaking doctoral programmes, in health services evaluation, health sciences, health management, public health, nursing, sociology, socio-biology, medicine and epidemiology However, the book also appeals to health professionals who wish to broaden their knowledge of research methods in order to make effective policy and practice decisions Ann Bowling is Professor of Health Services Research in the Department of Primary Care and Population Sciences at University College London, and has a part secondment to the MRC Health Services Research Collaboration, University of Bristol Her other publications with Open University Press include: Measuring Disease (2001), Research Methods in Health (2002) and Measuring Health (2004) Shah Ebrahim is Professor of Epidemiology of Ageing and Head of the Department of Social Medicine, University of Bristol, and also Honorary Professor of Epidemiology at University College London He is coordinating editor of the Cochrane Heart Group and co-editor of the International Journal of Epidemiology He is widely published in the fields of geriatric medicine, stroke and cardiovascular epidemiology Cover design: Kate Prentice ISBN 0-335-21460-6 Bowling and Ebrahim Contributors: Joy Adamson, Geraldine Barrett, Jane P Biddulph, Ann Bowling, Sara Brookes, Jackie Brown, Simon Carter, Michel P Coleman, Paul Cullinan, George Davey Smith, Paul Dieppe, Jenny Donovan, Craig Duncan, Shah Ebrahim, Vikki Entwistle, Clare Harries, Lesley Henderson, Kelvyn Jones, Olga Kostopoulou, Sarah J Lewis, Richard Martin, Martin McKee, Graham Moon, Ellen Nolte, Alan O’Rourke, Ann Oakley, Tim Peters, Tina Ramkalawan, Caroline Sanders, Mary Shaw, Andrew Steptoe, Jonathan Sterne, Anne Stiggelbout, S.V Subramanian, Kate Tilling, Liz Twigg, Suzanne Wait Handbook of Health research methods • Which research method should I use to evaluate services? • How I design a questionnaire? • How I conduct a systematic review of research? Handbook of Health research methods Investigation, measurement and analysis Edited by Ann Bowling and Shah Ebrahim 780335 214600 HANDBOOK OF HEALTH RESEARCH METHODS HANDBOOK OF HEALTH RESEARCH METHODS Investigation, measurement and analysis Edited by Ann Bowling and Shah Ebrahim Open University Press Open University Press McGraw-Hill Education McGraw-Hill House Shoppenhangers Road Maidenhead Berkshire England SL6 2QL email: enquiries@openup.co.uk world wide web: www.openup.co.uk and Two Penn Plaza, New York, NY 10121-2289, USA First published 2005 Copyright © Ann Bowling and Shah Ebrahim 2005 Individual chapters © The Contributors 2005 All rights reserved Except for the quotation of short passages for the purpose of criticism and review, no part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of the publisher or a licence from the Copyright Licensing Agency Limited Details of such licences (for reprographic reproduction) may be obtained from the Copyright Licensing Agency Ltd of 90 Tottenham Court Road, London, W1T 4LP A catalogue record of this book is available from the British Library ISBN–10: 335 21460 (pb) 335 21461 (hb) ISBN–13: 978 0335 214600 (pb) 978 0335 2146 (hb) Library of Congress Cataloging-in-Publication Data CIP data applied for Typeset by RefineCatch Limited, Bungay, Suffolk Printed in the UK by Bell and Bain Ltd, Glasgow Contents List of contributors Preface Part Introduction Part Research on health and health care Paul Dieppe Describing and evaluating health systems Ellen Nolte, Martin McKee and Suzanne Wait Multidisciplinary methods of investigation ix xi 12 45 Evidence-based health care: systematic reviews Paul Cullinan 47 Critical appraisal Alan O’Rourke 62 Features and designs of randomized controlled trials and non-randomized experimental designs Kate Tilling, Jonathan Sterne, Sara Brookes and Tim Peters 85 Epidemiological study designs for health care research and evaluation Richard M Martin 98 Finding and using secondary data on the health and health care of populations Mary Shaw 164 Quantitative social science: the survey Ann Bowling 190 vi Contents Approaches to qualitative data collection in social science Simon Carter and Lesley Henderson 215 10 Combined qualitative and quantitative designs Joy Adamson 230 11 Design and analysis of social intervention studies in health research Ann Oakley 12 Area-based studies and the evaluation of multilevel influences on health outcomes Graham Moon, S.V Subramanian, Kelvyn Jones, Craig Duncan and Liz Twigg 246 266 13 293 14 Part Mathematical models in health care Jane P Biddulph Economic evaluation of health care Jackie Brown 314 Multidisciplinary research measurement 15 Psychological approaches to measuring and modelling clinical decision-making Clare Harries and Olga Kostopoulou 329 331 16 362 17 Techniques of questionnaire design Ann Bowling 394 18 Measuring health outcomes from the patient’s perspective Ann Bowling 428 19 Genetics, health and population genetics research Sarah J Lewis, George Davey Smith and Shah Ebrahim 445 20 Part Approaches to measuring patients’ decision-making Clare Harries and Anne Stiggelbout Tools of psychosocial biology in health care research Andrew Steptoe 471 Data analysis 21 Key issues in the statistical analysis of quantitative data in research on health and health services Kate Tilling, Tim Peters and Jonathan Sterne 495 497 Contents 22 Part Key issues in the analysis of qualitative data in health services research Jenny Donovan and Caroline Sanders Essential issues to consider when conducting research vii 515 533 23 Involving service users in health services research Vikki Entwistle 535 24 Ethical and political issues in the conduct of research Geraldine Barrett and Michel P Coleman 555 25 Training for research Tina Ramkalawan 584 General glossary Index 601 611 List of contributors Dr Joy Adamson, Department of Health Sciences, University of York, UK Dr Geraldine Barrett, Department of Health and Social Care, Brunel University, UK Dr Jane P Biddulph, Department of Primary Care and Population Sciences, University College London, University of London, UK Professor Ann Bowling, Department of Primary Care and Population Sciences, University College London, University of London, UK Ms Sara Brookes, Department of Social Medicine, University of Bristol, UK Dr Jackie Brown, MRC Health Services Research Collaboration, Department of Social Medicine, University of Bristol, UK Dr Simon Carter, Sociology Group, Department of Public Health and Policy, London School of Hygiene and Tropical Medicine, University of London, UK Professor Michel P Coleman, Non-Communicable Disease Epidemiology Unit, London School of Hygiene and Tropical Medicine, University of London, UK Dr Paul Cullinan, Department of Occupational and Environmental Medicine, National Heart and Lung Institute, Imperial College London, University of London, UK Professor George Davey Smith, Department of Social Medicine, University of Bristol, UK Professor Paul Dieppe, MRC Health Services Research Collaboration, Department of Social Medicine, University of Bristol, UK Professor Jenny Donovan, Department of Social Medicine, University of Bristol, UK Dr Craig Duncan, Institute for the Geography of Health, University of Portsmouth, UK Professor Shah Ebrahim, Department of Social Medicine, University of Bristol, UK Dr Vikki Entwistle, Health Services Research Unit, Department of Public Health, University of Aberdeen, UK Dr Clare Harries, Department of Psychology, University College London, University of London, UK Dr Lesley Henderson, Department of Human Sciences, Brunel University, UK Professor Kelvyn Jones, School of Geographical Sciences, University of Bristol, UK Dr Olga Kostopoulou, Department of Primary Care and General Practice, University of Birmingham, UK Dr Sarah J Lewis, Department of Social Medicine, University of Bristol, UK Dr Richard Martin, Department of Social Medicine, University of Bristol, UK Professor Martin McKee, European Centre on Health of Societies in Transition, London School of Hygiene and Tropical Medicine, University of London, UK Professor Graham Moon, Institute for the Geography of Health, University of Portsmouth, UK Index Note on using this index: (i) common terms are indexed in alphabetical order; (ii) specialist terms, with specialized and relevant uses of common terms, are indexed within specialist subject headings Analytic study, 102, 191, 201–203 Atomistic fallacy, 271, 278 Action research, 518 Audiotaped data analysis, 523–525 Area based studies aggregate data, 270, 271, 278 aggregate area effect, 270 aggregative fallacy, 271 area, 270, 271, 273, 274–283 area typology, 269, 274 atomistic fallacy, 271, 278 collective effect, 271–272 confidentiality restrictions on data, 277 composition effect, 271–285 contextual effect, 271–285 deprivation, 267, 269, 273, 274, 275 ecological effects, 270 ecological fallacy, 270, 271, 278 ecological study, 266, 267–270 hierarchical sampling, 277 hierarchy, 276–285 inequality, 267, 269 individual, 267, 275–284 mechanisms for place differences, 273 morbidity, 266, 267, 269 mortality, 267, 268, 274, 285 multilevel modelling, 276–286 units, definition, 277 units, number of, 269, 286 operationalization, 274, 286–287 outcomes, 269, 283–284, 286 spatial models, 281–283 time, 283–284, 285 Attitude scales/questions, 408–417 Audio computer assisted questionnaires, 205 Bayesian methods, 55, 195, 502, 503 Before–after comparison, 86 Behaviour questions, 420–421 Bias, 87, 98, 99, 100, 107, 110, 113, 118, 124–125, 129, 131, 132, 133, 134, 135, 139, 142, 146, 147–150, 195, 201, 202, 204, 205, 207, 209, 210, 249, 401–402, 409–410, 416, 417–422 Biological measures, 471 Biological variables, 472, 473, 477 Blind, 65, 71, 99, 156, 256 Blood pressure, 481–484 Bootstrapping, 511, 512 Cardiovascular measures, 481–484 Cases, 126–129 Case-control study, 102, 126, 127–135, 455–456, 203–204 Case only study, 456 Case parent trio study, 456 Case-sibling study, 456 Case series study, 102, 113 Categorical scale, 407 Causality, 140, 151, 191, 201 Chromosomes, 446 Clinical decision making algorithms, 334 analytical, 332 automatization, 332 bias, 347 612 Index Brunswik, 332, 342 choice of method, 350–352 cognitive continuum, 333 conjoint analysis, 342, 345–346 consistency, 344 critical incident approach, 335 decision tree, 334 discrete choice analysis, 346 environment, 332, 343 episodic memories, 334 experiments, 346–348, 350 experts, 331–334, 350 fast and frugal models, 345 focus groups, 334 framing, 347 generalizability, 350, 351 heuristics, 332, 334, 345, 347 hypothetical cases, 335, 337, 339, 344 hypothetico-deductive, 334 ideographic analysis, 342, 351 illness scripts, 334 interviews, 334–335, 349, 351 intuitive, 332 judgement, 332, 350, 351 judgment see judgement judgement analysis, 342–346, 351 judgment analysis see judgement analysis laboratory methods, 349, 350 lens model, 343 naturalistic decision making, 335, 348–350, 351 observation, 338, 349 patient confidentiality, 350 pattern recognition, 332, 333, 334 paper patients, 335, 337 problem solving, 332 processes, 335, 338 process tracing, 338–341, 342, 351 protocols, 334 qualitative methods, 349, 350 questionnaires, 349 reactivity, 341, 346, 350 recall, 334 recordings, 349, 350 recognition-primed decisionmaking, 333, 349 regression models, 343–345, 346 rules of thumb, 332, 334 self-insight, 335–336 self-reports, 335–336 script activation, 334 semi-structured interviews, 334 simulation, 348, 349, 350 structural approaches, 341–346, 351 task analysis, 351 think-aloud protocols, 335, 338, 349, 351 uses of, 331 verbal methods, 349 validity, 339 veracity, 351 Clinical significance, 80, 194, 501 Clinical trials, 50, 65, 69–73, 86 Cluster randomization/trial, 86, 93, 253–254 Clustered data, 510 Cohort studies, 101, 134, 135–142, 455 Combined qualitative and quantitative designs analysis, 235–6 interpretation, 236–237 complementarity, 232, 233 definition, 230 distinctions between approaches, 231 health care use, 239 health services research, 237–238, 240 integrationalist approach, 232 quality, 237 philosophical perspectives, 231–232 randomized controlled trials, 239 sampling, 237 triangulation, 232, 233 typologies, 234–235 validity, 236, 237 Competencies, 592–593, 594 Computerised/electronic questionnaires, 204, 205, 209, 403–404 Cronbach’s alpha, 397 Concurrent validity, 396, 398 Confidence intervals, 79–80, 498–499 Confidentiality issues, 277, 350 see Ethics Confounding, 98, 100, 118, 131, 133, 138, 141, 146, 150–152, 502–504 Conjoint analysis, 342, 345–346 Consent, 254–256, 566 Consistency, 396 Construct validity, 396, 398 Content validity, 396, 398 Convergent validity, 396, 398 Criterion validity, 396, 398 Consumers see Service users Controls, 129–132, Index Controlled trial, 86 Consent, 566 Content analysis, 517, 521 Convenience sample, 226 Conversation analysis, 518, 522 Cortisol, 477–479 Cost-benefit, 316–317, 372 Cost-consequence, 315, 316 Cost-effectiveness, 303–306, 315, 317–318, 363, 372, 376 Cost-minimization, 315, 316 Cost-utility, 303–306, 315, 318–319 Critical appraisal blind, 65, 71 clinical significance, 80 clinical trials, 69–73 comparison, 64, 65 confidence intervals, 79–80 definition, 63–64 effect, 66, 71, 72 intervention, 64, 65 forest plot, 56, 80 generalizability, 63, 64, 67, 68, 72 homogeneity, 68 meta-analysis, 48, 54–55, 68, 80, 81 outcome, 64, 65, 66, 72, 75 presentation, 79 problem formulation, 64–65 qualitative research, 68, 75–78 quantitative research, 65 questions, 65 randomization, 66, 70 randomized controlled trial, 69, 70 reviews, 67–68 sample, 65, 66, 70–71, 76 statistical significance, 80 sub-group analysis, 81 systematic review, 73–75 trials, 65 validity, 63, 67, 69 Crossover trial, 92 Cross-sectional survey, 102, 120–125, 191, 200–201, 203 Cultural equivalence, 395–396, 419 Data protection, 167, 561–562, 568, 571, 574–575 Declaration of Helsinki, 555–556, 566, 567, 577 see Ethics Decomposed techniques, 363–364, 373–376, 377, 379 Decision analysis, 301–303, 324, 363, 365, 373, 384 613 Decision making, 362, 363, 379, 380–385 Decision tree, 301–3, 334 Descriptive study/survey, 102, 107–114, 190, 191, 200–201, 247–248 DHEA and DHEA sulphate, 479–480 Disability-adjusted life years, 372 Discounting, 325 Discourse analysis, 517, 521–522 Discriminative ability, 399 Discriminant validity, 396, 398 Disease, definition, Disease specific quality of life, 429–433 Discrete choice analysis/experiments, 346, 376, 377 DNA, 446 Drop outs/loss to follow-up/ withdrawal, 103, 148, 139, 141, 142, 195, 201 Ecological fallacy, 118, 119, 270, 271, 278 Ecological study, 102, 114–120, 266, 267–270 Effectiveness, 320 Efficiency, 314, 315, 320 Electronic questionnaires, 204, 205, 209, 403–404 Economics cost-benefit, 316–317 cost-consequence, 315, 316 cost-effectiveness, 315, 317–318 cost-minimization, 315, 316 cost-utility, 315, 318–319 costing interventions, 321–323 decision analysis, 324 see Mathematical models discounting, 325 efficiency, 314, 315, 320 effectiveness, 320 see Cost effectiveness eliciting utility values, 319–320 EuroQol EQ-5D, 320, 323 modelling, 324 Monte Carlo simulation method, 324 see Mathematical models outcomes, 315, 323 person trade-off, 372, 383 principles of, 315 probability trade-off, 372–373 randomized controlled trials, 323, 324 614 Index resource costs, 321–323, 324 QALY, 315, 318–319, 365 sensitivity analysis, 324 SF-6D health classification system, 376 simulation, 324 standard gamble, 319 time preference, 324 time trade-off, 319, 320, 368, 369, 371, 375, 383–384 treatment trade-off, 372–373 types of economic evaluation, 315 uncertainty in economic evaluations, 324 utility, 319–320, 363, 364–378, 379, 382, 384, 385, 373–376 values, 363–379, 383 visual analogue scale, 319, 320, 368–369, 383–384 willingness to pay, 372, 383–384 Embarrassing questions, 420 Epidemiology analytic study, 102, 191, 201–203 bias, 87, 98, 99, 100, 107, 110, 113, 118, 124–125, 129, 131, 132, 133, 134, 135, 139, 142, 146, 147–150 blind, 65, 71, 99, 156, 256 cases, 126–129 case-control study, 102, 126, 127–135, 203–204 case series study, 102, 113 causality, 140, 151, 191, 201 cohort studies, 101, 134, 135–142, 191, 201, 202, 203 cohort effects, 110 confounding, 98, 100, 118, 131, 133, 138, 141, 146, 150–152, 502–504 controls, 129–132 cross-sectional survey, 102, 120–125, 191, 200–201, 203 crude rates, 106 death, 103, 104, 105, 108, 114, 127, 131–132, 136, 137 definition, fundamental tasks, 98, 102 design, 152–156 descriptive study/survey, 102, 107– 114, 190, 191, 200–201, 247–248 drop-out/loss to follow-up/ withdrawal, 103, 148, 139, 141, 142, 195, 201 ecological study, 102, 114–120 ecological fallacy, 118, 119 epidemiological modelling, 114 event see outcome exposure, 114–115, 116, 118, 128, 132, 137–139, 142–146 geographic study, 114 incidence, 102, 103–106, 127, 136, 141 incident cases127, 128 interventional study, 101 see Statistics longitudinal survey, 135–142, 191, 201–203 matching, 133–134 mortality see death multistage sampling, 123 observational study, 98, 99, 100–102 odds ratio, 144–146 outcomes, 99, 102, 139–141 over-matching, 133 patient record data, 107 see Secondary data period effect, 110 person analyses, 111 person time at risk, 103, 104 place analyses, 111 planning epidemiological fieldwork, 152–157 period prevalence, 103 point prevalence, 103 power, statistical, 87, 137, 193 prevalence, 102–103, 105–106, 122, 123–124, 127 prevalent cases, 127, 128 primary care data, 108 see Secondary data psychosocial biology, 475 quality control, 155 questionnaires, 153–154 random sample, 122 random sampling, 123 randomized controlled trial, 52, 53, 57, 69, 70, 85–87, 98–99, 101, 239, 250, 323, 324, 509–510, 518, 528 rates, 103–5, 106, 142–144 ratios, 142–146 record linkage, 114 see Secondary data response rate, 122, 125, 127, 147, 152, 153, 207, 208, 209, 210, 211, 212, 403 reverse causality, 98, 100, 124–5 routine statistics, 106–9, 115 see Secondary data, Health care systems Index sample, 123–124, 137 sampling, 122, 123–124 secondary care data, 107 see Secondary data self-reports, 153 simple random sampling, 123 spurious, 148 standardized rates, 106 stratified, 123 survey data, 108 see Secondary data survey see cohort, longitudinal, cross-sectional surveys, Surveys systematic error, 118 systematic sampling, 123 time analyses, 106, 107 trials, 100 uses, 99 Error measurement, 208, 401 non-measurement, 208 sampling, 191–192 survey, 402 Ethics autonomy, 565, 566 beneficence, 565 electronic health records, 572, 575, 576 ethics committees, ethical review, 563–567 clinical trials, harmonization, 559–560, 566 consent, recording of, 566 consent and randomization, 254–256 Council of Europe draft protocol, 558 Council for International Organizations of Medical Sciences guidelines, 556 Data confidentiality restrictions, 277 data protection, 561–562 Data Protection Act, 561–562, 568, 571, 574–575 Declaration of Helsinki, 555–556, 566, 567, 577 Human Tissue Bill, 574 identifiable data, 567–577 international codes and regulations, 555–563 justice, 565 national laws, codes, regulations, 567–577 non-maleficence, 565 615 Patient Information Advisory Group, 571–572 principles guiding ethics committees, 564–565 process tracing research, 381 public’s view, 577 research governance, 567, 569 social science research, 566 Ethnography, 215–216, 517, 518–519, 523 Ethnomethodology, 215–216 Experiments analysis, 88–90 before–after comparison, 86 blind, 65, 71, 99, 156, 256 block randomization, 88 clinical trial, 86 cluster randomised trial, 86, 93 concealment, 87 control, 91 controlled trial, 86 crossover trial, 92 decision making, 346–348, 350, 383 equipoise about impact of intervention, 91 historical controlled study, 86 intention to treat, 87, 98 intervention, 91 minimization, 88 n-of-one trials, 93–94 non-randomised experiments, 86 parallel group trial, 92 patient preference trial, 94–95 placebo, 91 power calculation, 87 see Statistics protocol, 89 randomization, 87 randomized controlled trial, 52, 53, 57, 69, 70, 85–87, 98–99, 101, 239, 250, 518, 528, 323, 324, 509–510 reports, 91 randomized cross-over trial, 86, simple randomization, 88, stratified randomization, 88, types of, 85–86 EuroQoL EQ-5D, 320, 323, 376, 411, Factor analysis, 395, 400, 411 Factor structure, 396, 398, 400 Fast and frugal models, 345 Filter questions, 404 616 Index Focus groups, 220–221, 515 Focused analytical studies in health care systems, 30 Forrest plot, 56, 80 Framework approach, 522, 523 Framing, 347, 420 Frequentist theory, 195 Funnel plot, 51, Funnelling, 417 Generalizability, 63, 64, 67, 68, 72, 350, 351, 527 Genetics age of onset, 451 allele, 450 candidate genes, 455 case control study, 455–456 case only, 456 case-sibling, 456 case parent trios, 456 cell division, 447–449 chromosomes, 446 cohort study, 455 cosanguinity, 461–462 dominance, 452 DNA, 446 familial aggregation of disease, 450 founder effects, 462 gene expression, 446–447 genetic components of disease, 450–455 genome, 446–450 genotype, 450 heterozygosity, 450 heritability, 453 homozygosity, 450 individualized medicine, 467 linkage analysis, 454 meiosis, 448 Mendelian inheritance, 451–543 Mendelian randomization, 460–461 mitosis, 448 models of disease, 461–463 mutations, 449–450 phenotype, 450 polymorphism, 449–450, 456–457, 467 population admixture, 459 population-based studies, 455–461 recessive, 452 recombination, 449 role in disease, 445 screening, 463–467 segregation analysis, 451–453 twin studies, 454 Genome, 446–450 Grounded theory, 519–520, 523 Guttman scale, 408, 412 Health definition, 3, 4, 14, 430, 437, 585 index, 375, 376 outcome, 428 status/measurement, 376, 408, 410–411, 412, 413–417, 418, 429–437, 441 state classification, 375, 376, 383 utilities see Economics, Utilities Health belief model, 379, 380, 383 Health related quality of life (HRQoL), 395, 406, 412, 414–417, 429 definition, 430–432 measurement, 432–434, 437–441 Health care systems assessment, 20–24 classification, 15–20 definition, 4–6, 13–15, descriptive studies, 25–26 economic typologies, 15–16 effectiveness, 24, 32, 34 efficiency, 32 equity, 32 framework for assessment, 20–24 financing context, 20–23 focused analytical studies, 30 indicator, 32 international comparison, 24–29 league tables, 37–38 levels, 20–24 outcome, 20, 24, 26, 27, 28, 30, 31, 35, 36, 37, 38 quantitative studies, 26–30 performance assessment, 14, 30–38 performance indicator, 32 process, 23, 24, 26, 31, 256–258, 259 quality, 32 role of state, welfare typologies, 15–19 routine statistics, 28 see Epidemiology, Secondary data systems theory, 19 Health service acceptability, 8, appropriateness, 9, 10 audit, 10 effectiveness, 8, 9, 24, 32, 34, 320 efficiency, 7, 8, 32, 314, 315, 320 Index equity, 8, 9, 10, 32 implementation, 8–9 outcomes see Outcomes quality, quality assurance, 9, 10 research, 7, 9, 10–11, 237–238, 240, 384, 510, 515–516, 517, 518, 521, 522, 523, 524, 525, 526, 527, 528, 536, 585 resources, 11, 13 Health Utilities Index, 373, 376 Heart rate variability, 484 Heuristics, 332, 334, 345, 347, 382 Historical controlled study, 86 Hierarchical sample, 277 Holistic methods, 362, 363–373, 382, 383, 384 Hypothesis, 200 see Null hypothesis Hypothesis testing, 499–501 Hypothetico-deductive, 334 Ideographic, 342, 351 Illness, definition of, 3,4 Immunological measures, 485–486 Incidence, 102, 103–106, 127, 136, 141 In-depth interview, 217–20, 515, 516, 518 Individual levels, 267, 275–284 Individualized measurement, 411 Inflammatory markers, 484–485 Information acquisition patterns, 381 integration theory, 375 presentation effects, 370 processing, 371, 381 Intention to treat, 509 Internal consistency, 396, 397, 400 Internal reliability, 396 Inter-rater reliability, 396, 397 Interval scale data/measurement, 395, 408 Intervention studies, 64, 66, 98, 256–258, 477 Interviews bias, 204, 205, 207, 209, 210 critical incident, 335 in-depth, 217–220, 335, 380, 383, 515, 516, 518 focus groups, 220–221, 334 structured, 204, 205, 209, 210, 218, 335 semi-structured, 204, 209, 218–220, 334 617 think aloud, 335, 338, 349, 351, 381 telephone, 205, 206, 209–210 unstructured, 218, 517, 520 Interviewer, 206, 208, 209, interviewer bias, 401 Item response, 208, 395 -item reliability, 397 -total reliability, 397 Jacknife, 511, 512 Judgement, 332, 350, 351, 370–372, 376, 377, 378, 382, 413 Judgment see Judgement Judgement analysis, 342–346, 351, 376 Judgment analysis – see Judgement analysis Knowledge questions, 420 League tables, 37–38 Lens model, 343 Level of data/measurement, 395, 407–408 Likert scale, 406, 407, 408, 412 Linear regression, 504–506, 507 Literature, electronic searches, 50, 51–52 Logistic regression, 506, 507, 511 Longitudinal survey, 135–142, 191, 201–203 Matching, 133–134 Mathematical models analytic, 297, 298 building a model, 306–308 conceptual models, 293 continuous time, 297–299 cost-effectiveness analysis, 303–306 cost-utility analysis, 303–306 decision analysis, 301–3 decision tree, 301–3 definition, 293–294 deterministic, 297, 298–299 differential and difference equations, 301 discrete event, 296, 297, 298–299 health care evaluation model, 294 health care, 296, 300, 302–3 Markov, 300–301, 303 Monte Carlo simulation method, 309 physical, 293 618 Index QALY, 304–305 sensitivity analysis, 309 simulation, 296, 297–298 stochastic, 297, 298–299 types, 297 uses of, 294–295, 308 validation, 307–308 verification, 307 Maximum variation sampling, 519 Measurement error, 208, 401 Memory questions, 422 Mendelian inheritance, 451–543 Mendelian randomization, 460–461 Mental stress testing, 476–477 Meta-analysis, 48, 54–55, 68, 80, 81 Missing data, 53–54, 510–511 Mixed methods see Combined qualitative and quantitative designs Mode of questionnaire administration, 206–208, 403 Modelling, 324 see Area based studies, Mathematical models Markov, 300–301 Monte Carlo simulation method, 309, 324 multilevel, 276–286 economic, 324 epidemiological, 114 Morbidity, 266, 267, 269 see Health Mortality, 267, 268, 274, 285 see Epidemiology, death Multi-attribute utility theory, 373–376 Multidimensional measurement, 416 Multiple form reliability, 397 Multiple regression, 507 Multistage sampling, 123, 198, 199 Musculoskeletal measures, 486–487 N-of one trials, 93–94 Narrative analysis, 520–521 Narrative reviews, 47 Naturalistic approaches, 335, 348–350, 351, 381 Naturalistic monitoring studies, 475–476 Naturalistic paradigm, 216–217 Neuroendocrine factors, 477–481 Nominal scale data, 395, 407, 408 Normal distribution, 194, 498 Non-randomised experiments, 86 Null hypothesis, 192, 193–194, 499, 500 Numeric scale, 407 Observation, 338, 349, 381, 240, 515, 517, 518–519 Odds ratio, 48, 54, 144–146 Operationalization, 274, 286-287 Option, 363, 370, 371,373, 376, 377, 378, 379, 378, 379 Ordinal logistic regression, 506, 507 Ordinal scale data, 395, 407–408 Outcome, 20, 24, 26, 27, 28, 30, 31, 35, 36, 37, 38, 49, 54, 64, 65, 66, 72, 75, 99, 102, 139–141, 258–259, 315, 323, 364, 365, 366, 367, 376, 382, 497 P values, 193–194, 499–502 Panel survey, 202–203 Parallel group trial, 92 Patient based measure, 429 Patients’ decision making attitudes and beliefs, 362, 379–380 Brunswik, 377 choice, 365, 371, 378, 379 conjoint analysis, 373, 376–377, 383 cost-benefit, 372 cost-effectiveness, 363, 372, 376 decision aids, 385 decision analysis, 363, 365, 373, 384 decision making, 362, 363, 379, 380–385 decomposed techniques, 363–364, 373–376, 377, 379 disability-adjusted life years, 372 discrete choice experiments, 376, 377 ethical issues, 381 experiments, 383 EuroQol EQ-5D, 376 focus groups, 380 framing, 364, 370 functional measurement, 373 health belief model, 379, 380, 383 health index, 375, 376 health services research, 384 health status, 376 health state classification, 375, 376, 383 Health Utilities Index, 373, 376 Heuristic, 382 holistic methods, 362, 363–373, 382, 383, 384 hypothetical cases, 376 interviews, in depth, 380, 383 information acquisition patterns, 381 Index information integration theory, 375 information presentation effects, 370 information processing, 371, 381 judgement, 370–372, 376, 377, 378, 382 judgment see judgement judgement analysis, 376 judgment analysis see judgment analysis magnitude estimation, 369, 383–384 multi-attribute utility theory, 373–376 naturalistic decision making, 381, 383 naturalistic techniques, 381 observation, 381 option, 363, 370, 371, 373, 376, 377, 378, 379, 378, 379 option effects, 370–372 outcome, 364, 365, 366, 367, 376, 382 person trade-off, 372, 383 preferences, 362, 363–379, 382 probability, 364, 379 probability trade-off, 372–373 procedural effects, 371 process, 362, 363, 383 process tracing techniques, 376, 380–382, 383 protection motivation theory, 379, 380, 383 quality of life, 364 QALY, 365 questionnaires, 381 regression models, 375, 378 reliability, 363, 378 repertory grid technique, 376, 378–379, 383 risk, 364, 365, 366, 370, 373 Short Form-36 (SF-36), 376 standard gamble, 365–367, 383–384 theory of planned behaviour, 379, 380, 383 theory of reasoned action, 379–380, 383 think-aloud interviews, 381 time, 366–367, 372 time trade-off, 368, 369, 371, 375, 383–384 treatment trade-off, 372–373 uses of, 363–364, 382–3 utility, 363, 364–378, 379, 382, 384, 385 619 values, 363–379, 383 visual analogue scale, 368–369, 383–384 willingness to pay, 372, 383–384 Patient preference trial, 94–95 Pattern recognition, 332, 333, 334 Performance assessment, 14, 30–38 Person time at risk, 103, 104 Person trade-off, 372, 383 Pilot study, 402–403 Positivism, 7, 231, 232 Postal questionnaire, 204, 208, 210 Postgraduate/postdoctoral training, 588–599, 593–597 Power, statistical, 87, 137, 193 Pre-coded response formats, 394 Predictive validity, 396, 398 Preferences, 94–95, 362, 363–379, 382, 362, 363–379, 382, 412 Principal components analysis, 398 Probability, 193, 194–195, 198–199, 364, 379 Probability trade-off, 372–373 Processes, health systems, 23, 24, 26, 31, 256–258, 259 Process evaluation, 256–258 Process tracing techniques, 376, 380–382, 383 Prospective survey, 191, 201–203 Protection motivation theory, 379, 380, 383 Psychosocial biology allostatic load, 473–474 biological variables, 472, 473, 477 biological measures, 471 blood pressure, 481–484 cardiovascular measures, 481–484 catecholamines, 480–481 cortisol, 477–479 DHEA and DHEA sulphate, 479–480 epidemiological studies, 475 heart rate variability, 484 immunological measures, 485–486 inflammatory markers, 484–485 intervention studies, 477 psychosocial factors, 471 psychobiological function, 473–474 mental stress testing, 476–477 musculoskeletal measures, 486–487 naturalistic monitoring studies, 475–476 neuroendocrine factors, 477–481 620 Index research methods, 475–477 stress, 472–473 theoretical foundations, 472–474 Prevalence, 102–103, 105–106, 122, 123–124, 127 Psychobiological function, 473–474 Psychometrics, 394–395, 396–402 Psychosocial factors, 471 Process tracing, 338–341, 342, 351 Publication bias, 50, 51 Purposive sampling, 226, 519 Qualitative research see Combined qualitative and quantitative designs action research, 518 audiotaped data analysis, 523–525 coding, 524, 527 constant comparison, 519, 520, 521, 523, 524 content analysis, 517, 521 convenience sample, 226 conversation analysis, 518, 522 critical appraisal, 68, 75–78 data analysis, 523 data comparison, 524 data reduction, 523, 524 data saturation, 525 discourse analysis, 517, 521–522 emic, 215–216 ethnography, 215–216, 517, 518–519, 523 ethnomethodology, 215–216 framework approach, 522, 523 focus groups, 220–221, 515 generalizability, 527 grounded theory, 519–520, 523 health services research, 240, 515–516, 517, 518, 521, 522, 523, 524, 525, 526, 527, 528 in-depth interviews, 217–20, 515, 516, 518 interpretive, 216–217 iterative approach, 525 maximum variation sampling, 519 narrative analysis, 520–521 naturalistic paradigm, 216–217 naturalistic decision making, 335, 348–350, 351 observation, 240, 515, 517, 518–519 perspectives, 215–216, 232 purposive sampling, 76, 226, 519 questerviews, 518 quota sample, 199, 226 randomized controlled trial, 239, 518, 528 reflexivity, 226–227, 528 reliability, 525, 526–527 rigour, 525, 526–527 sampling, 225–226 semi-structured interview, 218–220 snowball sample, 226 story-telling, 520 systematic reviews, 48 thematic analysis, 520, 521 thematic description, 524 theoretical sampling, 519, 225–226 theoretical sensitivity, 519 theory, 516 transcription of recordings, 523–524 triangulation, 52 types of methods, 7, 48, 215–217 unstructured interviews, 218, 517, 520 vignettes, 518 Quality of life (QoL), 364, 394, 400, 408, 414–417, 428, 429, 431–432, 437–441 Quality of Life in Depression Scale, 433 QALY, 304–305, 315, 318–319, 365 Quantitative research critical appraisal, 65 range of methods, 190 see Epidemiology, Surveys, Questionnaires, Combined qualitative and quantitative designs health systems research, 26–30 scientific method, 6–7, 190 Questerviews, 518 Question form, 404, 405–417 order, 205, 208, 403, 404, 417–419 wording, 205, 404, 410, 419–421 Questionnaires acceptability, 395, 396 accuracy of responses, 207 alternate forms, 395 attitudes, 408–417 audio computer-assisted, 205 behaviour questions, 420–421 bias, 204, 205, 207, 209, 210, 400–402, 409–410, 416, 417–422 burden, 395, 206–207, 406, 410 categorical scale, 407 completeness of response, 207 comprehensiveness, 398, 400 Index computerised/electronic questionnaires, 204, 205, 209, 403–404 concurrent validity, 396, 398 consistency, 396 construct validity, 396, 398 content validity, 396, 398 convergent validity, 396, 398 covering letter, 403 criterion validity, 396., 398 Cronbach’s alpha, 397 cultural equivalence, 395–396, 419 data quality, 207 design, 153, 394 discriminative ability, 399 discriminant validity, 396, 398 electronic questionnaires see computerised questionnaires embarrassing questions, 420 error, 400–402, 404 equivalence of meaning, 395, 396, 400 face validity, 396 factor analysis, 395, 400, 411 factor structure, 396, 398, 400 filter questions, 404 follow-up reminders, 404 funnelling, 417 Guttman scale, 408, 412 homogeneity, 396, 398 item response, 208 item-item reliability, 397 item-total reliability, 397 individualized measurement, 411 internal consistency, 396, 397, 400 inter-rater reliability, 396, 397 interview, 209 see Interviews interval scale data, 395, 408 internal reliability, 396 item response and non-response, 395 judgement, 413 judgment see judgement knowledge questions, 420 layout, 403–404, 409 level of data/measurement, 395, 407–408 Likert scale, 406, 407, 408, 412 measurement error, 401 memory questions, 421, 422 mode of administration, 206 –208, 403 multiple form reliability, 397 multidimensional measures, 416 621 nominal scale data, 395, 407, 408 numeric scale, 407 open-ended questions, 394 operationalise, 402 ordinal scale data, 395, 407–408 paper and pencil, 205 piloting, 402–403 postal, 204, 208, 210 precision, 399 pre-coded response formats, 394 predictive validity, 396, 398 preferences, 412 principal components analysis, 398 psychometrics, 394–395, 396–402 question form, 404, 405–417 question order, 205, 208, 403, 404, 417–419 question wording, 205, 404, 410, 419–421 random error, 396 ratio scale data, 395, 408 reactive effects, 402 recall, 153–154, 422 receiver operating characteristic, 399 reliability, 208, 395, 396–402, 414–416 repertory grid techniques, 412–413 reproducibility, 396, 397 response see item response, Response/rate, Surveys response format/scales, 205, 405–410 responsiveness/sensitivity to change, 395, 396, 399, 416 scales and scaling methods, 394, 408–413 scores, 408, 410, 411 self-administered, 204, 206–207, 208 self-completion, 204 self-report, 153 semantic-differential scale, 412 semi-structured, 394 sensitive questions, 417–418, 420–421 sensitivity, 396, 399 single item, 408, 413–417 specificity, 396, 399 sensitivity analysis, 399 spilt half reliability, 397 stability, 397 standardized, 204, 395 statistics for assessing reliability, 397 structured, 394, 395 622 Index systematic error, 402 test-retest reliability, 396, 397 threatening questions, 420 three-dimensional scale, 412 thought-listing, 412 Threatening questions, 417–418, 420–421 Thurstone scale, 412 Time frames, 422 total survey error, 402 translation, 395, 419 see cultural equivalence unidimensional scale, 412 validity, 153–154, 395, 396–402, 414–416 vignettes, 413 visual analogue scales, 406, 407, 415–416 weighting, 408, 410, 411, 417 wording, 395 Quota sample, 199, 226 Random error, 396 Random sampling, 122, 123, 195–197, 198, 199, 202 Randomization, 66, 70 Randomized controlled trial, 52, 53, 57, 69, 70, 85–87, 98–99, 101, 239,250, 518, 528, 323, 324, 509–510 Randomized cross-over trial, 86, Ratio scale data, 395, 408 Reactive effects, 341, 346, 350, 402 Recall, 334, 153–154, 422 Receiver operating characteristic, 399 Recognition-primed decision-making, 333, 349 Record linkage, 114, 178–181 Reflexivity, 26–227, 528 Reliability, 208, 363, 378, 395, 396–402, 414–416, 525, 526–527 Regression models, 343–345, 346, 504–508 see Patients’ decision making, Statistics Regression to the mean, 202 Repertory grid technique, 376, 378–379, 383, 412–413 Reproducability of data, 396, 397 Research councils, 586 Resource costs, 321–323, 324 Response format/scales, 205, 405–410 Response and non-response, 195 see Response rate Response rate, 122, 125, 127, 147, 152, 153, 207, 208, 209, 210, 211, 212, 403 see Item response Response shift, 432 Responsiveness to change, 395, 396, 399, 416 Retrospective survey, 191, 200–220 Reverse causality, 98 Reviews, 67–68 see Systematic reviews Rigour, 525, 526–527 Risk, 364, 365, 366, 370, 373 Risk ratio, 54 Routine statistics, 28, 106–109, 115, 182–184 Rules of thumb, 332, 334 Sample size, 193–194 Sampling, 191–199 cluster, 197, 199 convenience, 226 frame, 192–193, 403 hierarchical, 277 mixed methods, 237 maximum variation, 519 multilevel, 193 see Area based studies multistage, 123, 198, 199 non-random, 199 probability proportional to size, 198, 199 purposive, 76, 226, 519 qualitative research, 225–226 quota, 199, 226 random, 122, 123, 195–197, 198, 199, 202 response rate, 122 snowball, 226 simple, 123, 196, 199 size, 65, 66 see Statistics, power snowball, 226 stratified, 123, 197, 199 systematic, 123, 196, 199 telephone, 198, 199 theoretical, 225–226, 519 unrestricted, 196, 199 error, 191–192 theory unit, 193 Scales and scaling methods, 394, 408–413 Scientific method, 4–7 Scored data, 408, 410, 411 Secondary data cancer registries, 171–172 Data Protection Act, 167 see Ethics Index deprivation, 184 disease registers, 171–173 electronic records, 166–167 General Practice Research Database, 169–170 hospital activity statistics, 176 hospital episodes statistics, 176–178 patient record data, 165–181 NHS Central Register, 179–181 population data, 182–184 prescribing data, 174–176 primary care data, 166–176 public health observatories, 177 record linkage, 178–181 see Epidemiology, Experiments routine statistics, 182–184 see Epidemiology, Health care systems secondary care, 176–178 see Epidemiology surveys, 170–171, 181–182 see Epidemiology types of secondary data165 Self-administered questionnaire, 204, 206–207, 208 Self-report, 153, 335–336 Semantic-differential scale, 412 Semi-structured interview, 204, 209, 218–220, 334, 394 Sensitive questions, 417–418, 420–421 Sensitivity, 396, 399 Sensitivity analysis, 57, 324, 399 Service users acceptability of health services, 539 concerns and views, 535, 538, 539 consumer advocacy groups, 535 dissemination of research results, 549–550 empowerment, 540 formulating research questions, 545–546 growth of interest, 537–538 health services research, 535 interpretation of results, 548–549 quality of health care, 537 quality of research, 539, 540 planning involvement, 550 profile of research, 540 rationales, 538–541 relevance of health services, 538 recruitment and data collection, 548 research, 535, 536–538, 541–550 research design, 546–547 research funding bodies, 543–544 623 research networks, 544–545 research priorities, 538–539 research projects, 545 theory, 541 Short Form-36 conceptual base, 434 content, 434–435 examples of items, 413, health classification system SF-6D, 376 predictive and discriminative ability, 433, 436 question order effects, 418 reliability of SF-36 437 scoring, 435–436 shorter versions, 417, 418, 434 validity, 436 Significance/testing, 80, 193–195, 500–501 Simple random sampling, 123, 196, 199 Single item question, 408, 413–417 Simulation, 296, 297–298, 324 see Mathematical models, Monte Carlo simulation method Situated learning Social intervention studies bias, 249 blinding, 256 cluster randomization, 253–254 consent and randomization, 254–256 descriptive studies, 247–248 equipoise about impact of intervention, 253–254 hallmarks of randomized controlled trial, 250 objections to randomization, 253–254 outcome, 258–259 process, 256–258, 259 placebo, 246 qualitative, 257, 259 quantitative, 257, 258, randomization, 249 randomized controlled trial, 248–253 social component, 246 subjective measures, 259 standardization, 256 validity, 248, 250 Social theory, 216–217, 232, 272 Snowball sample, 226 Specificity, 396, 399 Split-half reliability, 397 624 Index Standard gamble, 319, 365–367, 383–384 Statistics analysis plan, 508 Bayesian methods, 195, 502, 503 bootstrap, 511, 512 clustered data, 510 confidence intervals, 79–80, 194, 498–499, 501 confounding, 502–504 covariate adjusted analyses, 509 exposure, 497 Factor analysis, 395, 400, 411 Factor structure, 396, 398, 400 Frequentist methods, 195 health services research, 510 hypothesis testing, 499–501 intention to treat, 509 interactions, 508 jacknife, 511, 512 linear regression, 504–506, 507 logistic regression, 506, 507, 511 missing data, 510–511 multiple regression, 507 normal distribution, 194, 498 null hypothesis, 192, 193–194, 195, 499, 500 odds ratio, 48, 54, 144–146 one- or two-sided tests, 193–194 ordinal logistic, 506, 507 outcome, 497 P values, 193–195, 499–502 see significance testing population inferences, 497–498 power, 87, 137, 193 probability, 193, 194–195, 198–199, 364, 379 Principal components analysis, 398 randomized controlled trials, 509–510 regression models, 504–508, 511–512 robust standard errors, 512 risk factors, 497 risk ratio, 54 robust standard errors, 511, 512 standard deviation, 194 test assumptions, 194 type I and II errors, 192, 193, 194, 195 significance/testing, 80, 193–195, 500–501 see P values standard deviation, 194 unit of analysis, 190 Story-telling, 520 Stratified sampling, 123, 97, 199 Stress, 472–473 Stability of questions, 397 Standard deviation, 194 Standardization, 256 Standardized questions, 204, 395 Structured questions, 394, 395 Structured interview, 204, 205, 209, 210, 218m, 335 Surveys analytic, 102, 191, 201–203 bias, 195, 201, 202 case control, 102, 126, 127–135, 203–204 causality, 140, 151, 191, 201 cohort, 191, 201, 202, 203 cohort sequential, 203 cross-sectional, 102, 120–125, 191, 200–201, 203 definition, 190–191 descriptive study, 102, 107–114, 190, 191, 200–201 drop outs/loss to follow-up/ withdrawal, 103, 148, 139, 141, 142, 195, 201 external validity, 191 follow-up reminders, 404 hypotheses, 200 incidence, 201 interview see Interviews longitudinal, 135–142, 191, 201–203 measurement error, 208, 401 natural experiments, 201 non-measurement error, 208 null hypothesis, 192, 193–194, 499, 500 panel, 202–203 probability theory, 194 prospective, 191, 201–203 regression to mean, 202 response and non-response, 195 see response rate questionnaires see Questionnaires response rate, 122, 125, 127, 147, 152, 153, 208, 209, 210, 211, 212, 403 retrospective, 191, 200–20 sampling see Sampling sampling frame see Sampling trend, 202 Systematic error, 118, 402 Index Systematic review analysis, 54–57 Bayesian methods, 55 definition, 48 electronic searches, 50, 51–52 expression of effects, 54 extracting data, 53 fixed effects model, 55 forest plot, 56, 80 funnel plot, 51 inclusion criteria, 52–53 interpretation, 57 heterogeneity, 55–57 homogeneity, 55 means, 54 meta-analysis, 48, 54–55, 68, 80, 81 methodological quality, 53 missing data, 53–54 odds ratio, 48, 54, 144–146 random effects model, 55 randomized controlled trials, 52, 53, 57 risk ratio, 54 outcome, 49, 54 process, 49 publication bias, 50, 51 qualitative data, 48 questions, 49 Systematic sampling, 123, 196, 199 Telephone interview, 205, 206, 209–210 sampling, 198, 199 Test-retest reliability, 396, 397 Thematic analysis, 520, 521 description, 524 Theoretical sampling, 519, 225–226 Theory of planned behaviour, 379, 380, 383 Theory of reasoned action, 379–380, 383 Think aloud interview techniques, 335, 338, 349, 351, 381 Thought listing, 412 Threatening questions, 417–418, 420–421 Thurstone scale, 412 Time-frames in questions, 422 Time preference, 324 Time trade-off, 319, 320, 368, 369, 371, 375, 383–384 625 Total survey error, 402 Training for health and health services research competencies, 592–593, 594 contract research staff, 595–596 funding, 587 interdisciplinarity, 591, 592 mixed methods training, 590–591 multidisciplinary, 585–586, 591, 592 research councils, 586 situated learning, 590 postgraduate, 588–599 postdoctoral, 593, 595 research careers, 593–596 Transcripts of recordings, 523–524 Translation of questionnaires, 395, 419 see Cultural equivalence Treatment trade-off, 372–373 Trend survey, 202 Triangulation, 52, 232, 233 Twin studies, 454 Unidimensional scale, 412 Unit of analysis, 190 multilevel, 193, 277 sampling, 193 Unstructured interviews, 218, 517, 520 Utility, 319–320, 363, 364–378, 379, 382, 384, 385, 373–376 Validity, 63, 67, 69, 153–154, 191, 236, 237, 248, 250, 339, 395, 396–402, 414–416 Values, 363–379, 383 Vignettes, 413, 518 Visual analogue scale, 368–369, 383– 384, 406, 407, 415–416 Weighting questions, 408, 410, 411, 417 Willingness to pay, 372, 383–384 World Health Organization classification of health condition, definition, 3, 14, 430, 437, 585 World Health Organization Quality of Life questionnaire (WHOQOL) conceptual base, 437 content, 438–439 reliability, 440–441 scoring, 440 short version, 438 validity, 440 .. .HANDBOOK OF HEALTH RESEARCH METHODS HANDBOOK OF HEALTH RESEARCH METHODS Investigation, measurement and analysis Edited by Ann Bowling and Shah Ebrahim Open University Press Open University Press. .. is interdisciplinary, i.e that each discipline is of equal importance to the project Further reading Bowling, A (1997) Research Methods in Health: Investigating Health and Health Services Buckingham:... of government, or one of the many charitable foundations working in the field of health Similarly, does one include those involved in training health professionals? This role has often been linked