Table 1

Characteristics of included studies

StudyAuthorsCountryTime frameParticipants, settingStudy designInteractionOutcomes
1Steinman et al 1 USASpring 1999Surveys about attitudes and behaviours towards industry gifts in 105 residents at a university-based internal medicine residency programmeCross-sectionalPSR interactions, giftsMost participants (61%) hold positive attitudes towards gifts from industry and PSR interactions and believe they do not influence their own prescribing, but only 16% believed other physicians were similarly unaffected (p<0.0001)
11De Ferrari et al 9 PeruMarch 2013Questionnaire in 155 faculty and trainee physicians of five different clinical departments working in a public general hospitalCross-sectionalPSR interactions, medical samples, promotional material, dinnersPositive attitude towards representatives (88.5% of participants). Faculty physicians received a larger amount of medical samples and promotional material and were more prone to believe that gifts and lunches do not influence their prescribing behaviour (42.2% vs 23.6%; p=0.036)
12Thomson et al 31 New Zealand1991Questionnaire survey of 67 general practitionersCross-sectionalInteractions with PSRMost general practitioners (67%) had a negative attitude towards PSR interaction
13Kamal et al 32 EgyptJuly and August 2013Interviews with 18 physiciansCross-sectionalInteraction with PSRPositive attitude towards PSR interaction
14Hodges10 CanadaOctober 1993–February 1994Survey in 105 residents of psychiatryCross-sectionalInteraction with PSR, drug samples, lunchesPositive attitude towards PSR interaction (56.5% of participants). The more money and promotional items a participant had received, the more likely he or she was to believe that discussions with representatives did not affect prescribing (p<0.05)
15Gibbons et al 33 USANot reportedSurvey of 392 physicians in two tertiary care medical centresCross-sectionalPSR interactions, gifts, samples, travel, lunchesPositive attitude towards PSR interactions, gifts, samples and lunches
16Spingarn et al 56 USAFebruary 199075 internal medicine physicians in university medical centreRetrospective cohortPSR interaction (teaching)Attendees inappropriately prescribed PSR speakers drug compared with non-attendees (p=0.029)
17Zaki58 Saudi ArabiaSeptember–November 2013Survey of 250 physiciansRandomised, cross-sectional surveyConferences, drug samplesFavourable towards promotion
18Orlowski et al 199457 USA1987–198910 physicians that were invited for a symposium and tracking the pharmacy inventory usage reports for these drugs before and after the symposiaCohortConference travelSignificant increase in the prescribing pattern of drugs occurred following the symposia (p<0.001)
19Scheffer et al 34 Brazil2007–2009Survey of 300 physicians prescribing antiretroviral drugsCross-sectionalInteraction with representative, drug samples, journalsFrequency of interaction; the majority of (64%) of the physicians had multiple forms of interactions with PSR
20Brett et al 35 USANot reportedQuestionnaire of 93 physicians in a medical schoolCross-sectionalInteraction with PSRImpact on attitudes; most physicians believed that most of PSR activities do not pose major ethical problems
21Gupta et al 36 IndiaJune–September 2014Survey of 81 physicians in single hospitalCross-sectionalInteraction with representative, drug samples, journalsImpact on prescribing; 61.7% of participants think that PSR has an impact on their prescribing (p=0.0001)
22Morgan et al 37 USAMarch 2003Survey of 397 obstetrician-gynaecologistsCross-sectionalDrug samples, promotional material, lunchImpact on prescribing, positive attitudes; most respondents thought it is proper to accept drug samples (92%), lunch (77%), an anatomical model (75%) or a well-paid consultantship (53%) from PSR
23Alosaimi et al 19 Saudi Arabia2012Survey of 659 physiciansCross-sectionalInteraction with PSRPositive attitude towards PSR interaction
24Chren and Landefeld61 USA1989–199040 case physicians and 80 control physiciansCase–controlPSR interactions, honoraria, researchIncreased prescription of company’s drug after PSR interaction, honoraria and research (p<0.001, all)
25Randall et al 59 USAOctober 2001Intervention group of physicians (n=18) that received education about PSR interaction and control group (n=14)Controlled trialInteraction with PSRThe majority of residents found the interactions and gifts useful. Compared with the comparison group, the intervention group significantly decreased the reported number of office supplies and non-educational gifts (p<0.05)
26Caudill et al 38 USANot reportedSurvey of 446 primary care physiciansCross-sectionalInteraction with PSRSignificant positive correlation between physician cost of prescribing and perceived credibility, availability, applicability and use of information provided by PSR (p<0.01)
27Andaleeb and Tallman20 USANot reported223 physicians in northwestern PennsylvaniaCross-sectionalInteraction with PSRPositive attitude towards PSR interaction
28Reeder et al 39 USA1991–199287 residents of emergency medicineCross-sectionalInteraction with PSR, giftsMost participants believed that PSR interaction had no impact on their prescribing
29Lichstein et al 40 USAJanuary–March 1990272 directors of internal medicine residency programmesCross-sectionalInteraction with PSRMost participants had a positive attitude towards PSR interactions
30Brotzman et al 41 USANot reportedDirectors of 386 family practice residency programmeCross-sectionalInteraction with PSRMajority of programmes do not have guidelines for interaction with PSR
31Alssageer and Kowalski42 LibyaAugust–October 2010Survey of 608 physicians in public and private practice settingsCross-sectionalInteraction with PSR, drug samples, printed materialsPositive attitude towards PSR interactions
32Lieb and Brandtonies, 201021 Germany2007Survey of 208 physicians (neurology, cardiology and general medicine)Cross-sectionalInteraction with PSR, drug samples, printed materials, lunchesFrequency and impact on attitudes
33Lieb and Scheurich22 Germany2010–2011Survey of 160 physicians in private and public practicesCross-sectionalInteraction with representative, drug samples, printed materials, CMEHigh expenditure prescribing; avoidance of industry-sponsored CME is associated with more rational prescribing habits
34Lieb and Koch,43 GermanyMay–July 2012Survey of 1038 medical students at eight universitiesCross-sectionalInteraction with representative, drug samples, printed materials, lunchesMost participants have contact with the pharmaceutical company; 24.6% of the participants thought gifts would influence their future prescribing behaviour, while 45.1% thought gifts would influence their classmates’ future prescribing behaviour (p<0.001)
35Brown et al 44 USA2008 and 2013251 directors of family medicine residency programmesCross-sectionalInteraction with PSR, gifts, lunchesNegative attitude towards PSR interactions
37Rahman et al 45 BangladeshDecember 2008–January 2009Survey of 83 village physiciansCross-sectionalInteraction with PSRImpact on their prescribing
38Lee and Begley,12 USA2008Nationally representative survey of 4720 physiciansCross-sectionalGiftsGifts were associated with lower perceived quality of patient care; an inverse relationship between the frequency of received gifts and the perceived quality of care was observed
39Montastruc et al 13 FranceAugust–October 2011Survey among 631 medical residentsCross-sectionalInteraction with representativeMost participants believed that PSR interaction had no impact on their prescribing; participants who had a more positive opinion were more frequently exposed to PSR (p<0.001)
40Klemenc-Ketis and Kersnik46 SloveniaOctober 2011895 family physicians at the primary level of careCross-sectionalInteraction with PSRPositive effect on knowledge; participants value PSRs’ selling and communication skills and trustworthiness highly
41Hurley et al 47 USA20103500 dermatologistsCross-sectionalFree drug samplesImpact on their prescribing; the provision of samples with a prescription by dermatologists has been increasing over time, and this increase is correlated (r=0.92) with the use of the branded generic drugs promoted by these sample
42Makowska48 PolandNovember–December 2008Survey of 382 physiciansCross-sectionalGiftsPositive attitude towards PSR interactions
43Siddiqui et al 49 PakistanNot reportedQuestionnaires of 352 medical studentsCross-sectionalInteraction with representativePositive attitude towards PSR interaction
55Workneh et al 50 EthiopiaFebruary–March 2015Survey of 90 physicians from public and private health facilitiesCross-sectionalInteraction with representative, giftsPositive attitude towards industry, impact on prescribing behaviour; nearly half of the physicians reported that their prescribing decisions were influenced by PSR
57Khan et al 51 PakistanNot reportedQuestionnaires in 472 physiciansCross-sectionalInteraction with representative, giftsPositive attitude towards PSR interaction
58Saito et al 67 JapanJanuary–March 20081417 physicians working in internal medicine, general surgery, orthopaedic surgery, paediatrics, obstetrics-gynaecology, psychiatry and ophthalmologyNational surveyInteraction with industry, receipt of gifts, funds, CME, samplesPositive attitude towards PSR and gifts, value information from PSR, interactions higher with physicians who prefer to prescribe brand names
59Ziegler18 USA199327 physicians working in public and private hospitalsSurveyAccuracy of information provided by PSRs about drugsIncorrect information often provided by speakers goes unnoticed by physicians
60Lurie et al 68 USANot reported240 internal medicine faculty physicians in academic medical centresSurveyEffect of interaction with PSR, free meals, honoraria and research supportImpact on prescribing behaviour and formulary change requests
62DeJong et al 52 USAAugust–September 2013279 669 physicians who wrote Medicare prescriptions in any of four drug classes: statins, cardioselective β-blockers, ACE inhibitors and angiotensin-receptor blockers, and selective serotonin and serotonin-norepinephrine reuptake inhibitors PhysiciansCross-sectionalIndustry-sponsored mealsReceipt of industry-sponsored meals was associated with an increased rate of brand name prescription.
63Yeh et al 53 USA2011All licensed Massachusetts physicians who wrote prescriptions for statins paid for under the Medicare drug benefit in 2011 (n=2444)Cross-sectionalEffect of industry payment on prescription of branded drugs for cholesterol controlPayment for meals and educational programmes increased prescription of brand name statins.
65Bowman and Pearle et al 69 USANot reported121 physician attendeesSelf-report surveyEffect of CME on prescribing behaviourSponsoring company’s drugs were favoured during prescription
66Fischer et al 65 USANovember 2006–March 2007Multidisciplinary focus groups with 61 physiciansSurveyEffect of industry marketing strategies on prescription and cognitive dissonance of physiciansMost participants reported no PSR impact on their prescribing, value to have ability to evaluate information of PSRs
67Chimonas et al 66 USAJune 2004Six focus groups in 32 academic and community physiciansSurveyPSR interactionsPositive attitude towards PSR interaction
72Yeh et al 54 USANot reported1610 US medical studentsCross-sectionalInteraction with representative, gifts, lunchesPolicies separating students from representatives reduced number of interactions
73Larkin et al 73 USAJanuary 2006–June 2009Paediatricians, child and adolescent psychiatrists in five medical centresSurveyInteraction with PSRAntidetailing policies reduced the prescription of off-label antidepressants and antipsychotics for children
74Esmaily et al 60 IranNot reported112 general physicians were randomised in two groups: (1) outcome-based educational intervention for rational prescribing and (2) concurrent CME programme in the field of rational prescribingRandomised trialEffect of outcome and retinal prescribingRational prescribing improved in some of the important outcome-based indicators. No difference between two arms of the study
76Parikh et al 55 USA2014Descriptive, cross-sectional analysis of Open Payments data and 9 638 825 payments to physicians and paediatricians from 1 January to 31 December 2014Cross-sectionalComparison of PSR interactions between paediatricians and other specialists; among subspecialties of paediatrics.Paediatricians get fewer gifts from PSR than internists. There is variation among subspecialties for extent of interaction.
78Chressanthis et al 74 USANot reportedClinical decisions of 72 114 physicians were statistically analysed using prescription dataSurveyEffect of restricting PSRs on clinical practice and knowledgeRestricting PSRs affected information flow about drugs, both negative and positive.

  • We excluded 2000 records as they were not relevant (n=1641), not original research (n=269), about medical students (n=4) and non-medical (eg, ecological and econometric; n=86).

  • PSRs, pharmaceutical sales representatives.