Ricci-Cabello et al. BMC Health Services Research (2020) 20:920 https://doi.org/10.1186/s12913-020-05753-x RESEARCH ARTICLE Open Access Adherence to breast cancer guidelines is associated with better survival outcomes: a systematic review and meta-analysis of observational studies in EU countries Ignacio Ricci-Cabello1,2,3, Adrián Vásquez-Mejía4, Carlos Canelo-Aybar3,5, Ena Niño de Guzman5*, Javier Pérez-Bracchiglione6, Montserrat Rabassa5, David Rigau5, Ivan Solà3,5, Yang Song5, Luciana Neamtiu7, Elena Parmelli7, Zuleika Saz-Parkinson7* and Pablo Alonso-Coello3,5 Abstract Background: Breast cancer (BC) clinical guidelines offer evidence-based recommendations to improve quality of healthcare for patients with or at risk of BC. Suboptimal adherence to recommendations has the potential to negatively affect population health. However, no study has systematically reviewed the impact of BC guideline adherence -as prognosis factor- on BC healthcare processes and health outcomes. The objectives are to analyse the impact of guideline adherence on health outcomes and on healthcare costs. Methods: We searched systematic reviews and primary studies in MEDLINE and Embase, conducted in European Union (EU) countries (inception to May 2019). Eligibility assessment, data extraction, and risk of bias assessment were conducted by one author and crosschecked by a second. We used random-effects meta-analyses to examine the impact of guideline adherence on overall survival and disease-free survival, and assessed certainty of evidence using GRADE. Results: We included 21 primary studies. Most were published during the last decade (90%), followed a retrospective cohort design (86%), focused on treatment guideline adherence (95%), and were at low (80%) or moderate (20%) risk of bias. Nineteen studies (95%) examined the impact of guideline adherence on health outcomes, while two (10%) on healthcare cost. Adherence to guidelines was associated with increased overall survival (HR = 0.67, 95%CI 0.59–0.76) and disease-free survival (HR = 0.35, 95%CI 0.15–0.82), representing 138 more survivors (96 more to 178 more) and 336 patients free of recurrence (73 more to 491 more) for every 1000 women receiving adherent CG treatment compared to those receiving non-adherent treatment at 5 years follow-up (moderate certainty). Adherence to treatment guidelines was associated with higher costs, but adherence to follow-up guidelines was associated with lower costs (low certainty). (Continued on next page) * Correspondence: enino@santpau.cat; zuleika.saz-parkinson@ec.europa.eu 5 Iberoamerican Cochrane Centre - Department of Clinical Epidemiology and Public Health, Biomedical Research Institute Sant Pau (IIB Sant Pau), Sant Antonio María Claret 167, 08025 Barcelona, Spain 7 European Commission, Joint Research Centre (JRC), Ispra, Italy Full list of author information is available at the end of the article © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Ricci-Cabello et al. BMC Health Services Research (2020) 20:920 Page 2 of 12 (Continued from previous page) Conclusions: Our review of EU studies suggests that there is moderate certainty that adherence to BC guidelines is associated with an improved survival. BC guidelines should be rigorously implemented in the clinical setting. Trial registration: PROSPERO (CRD42018092884). Keywords: Breast cancer, Clinical guidelines, Adherence, Survival, Systematic review Background deviations would contribute to increased survival of Breast cancer is the world’s most common cancer in patients with breast cancer (proposing that observed im- women with an estimated 2.08 million new cancer cases provements in health outcomes during the last two de- diagnosed in 2018, accounting for 24.2% of all cancers cades may be more due to higher resource availability [1]. The illness is diagnosed more frequently in devel- rather than adherence to guidelines) [21, 22]. To the best oped countries. According to the European Cancer In- of our knowledge, no previous systematic review has ex- formation System, more than 400,000 incident female amined the impact of breast cancer CGs adherence on breast cancer cases and 98,000 deaths were estimated in health outcomes. the European Union (EU28) for 2018 [2] (Additional file 1). Adherence to breast cancer CGs might behave as a Preventing, diagnosing and treating breast cancer is, there- prognostic factor for health outcomes. A prognosis fac- fore, an important priority for health policymakers. Treat- tor is any measure that, among people with a given ment procedures have changed dramatically over recent health condition, is associated with a subsequent clinical years. As new and more precise diagnostic strategies have outcome [23]. Even though most researched prognostic shown, treatment for early and metastatic breast cancer factors are biomarkers, or patient’s characteristics, these has improved [3–7]. Similarly, advances in breast cancer may also be measured outside the individual, at an eco- screening and treatment have reduced breast cancer mor- logical level, such as health care access, quality of care, tality across the age spectrum in the past decade [8–10]. and breast cancer guidelines adherence in which the ex- Clinical guidelines (CGs) are statements that include posure of individuals is inferred [23]. The objective of recommendations intended to optimise patient care [11]. this systematic review is two-fold: i) to identify if adher- Evidence-based high-quality CGs are becoming increas- ence (compared to non-adherence) to CGs impacts on ingly available as a result of recent advances in health patient-related outcomes and ii) to identify if adherence services research [12]. However, the effectiveness of the (compared to non-adherence) to CGs impacts on health- CGs heavily depends not only on their quality but also care costs. on how they are implemented, and embedded in clini- cians´ routine clinical practice [13]. Methods Since CGs recommendations should be based on best We conducted a systematic review following the standard available evidence, adherence to CGs is expected to re- Cochrane Collaboration methods [24] and adhering to the sult in better patient outcomes. However, most CGs rely PRISMA statement reporting guidelines [25]. We registered on evidence from clinical trials, which are usually per- the research protocol in PROSPERO (CRD42018092884). formed with relatively small samples at sites with experi- enced investigators and highly selected participants, and Data sources and searches therefore are at risk of overestimating benefits and We designed search strategies (May 2019) to search underestimating harms [14]. Moreover, they are gener- Embase (accessed through Ovid) and MEDLINE (PubMed). ally conducted in high-income countries, and, as a con- In a first step we focused the search to identify systematic sequence, tend to be very resource-intensive [15]. reviews as a source of primary studies, and then we Therefore, the external validity and clinical utility of searched primary studies. Additional file 2 shows the search their recommendations must be confirmed in real clin- algorithms. ical practice. Moreover, non-adherence can be due to valid reasons (mainly related to contraindications and Eligibility of studies patient preferences), and therefore many CGs deviations We used the PICO (Population, Intervention, Compari- are intentional and do not necessarily impact negatively son, Outcome) framework to guide our eligibility criteria on the quality of care [16]. (Table 1). We included observational studies (before- Some studies have shown that CGs adherence can after, cohort, case-control, and cross-sectional studies) have a positive impact on patient outcomes in the field examining the impact of CGs adherence on: 1) breast can- of breast cancer [17–20]. On the other hand, some au- cer patient-related outcomes (overall survival, disease-free thors have suggested a paradox by which guideline survival, quality of life, incidence-based mortality, harm), Ricci-Cabello et al. BMC Health Services Research (2020) 20:920 Page 3 of 12 Table 1 Structured clinical question Population Intervention Comparison Outcomesa Healthcare professionals involved Adherence to breast cancer Non-adherence to breast cancer 1) Impact on patient-related outcomes in breast cancer care (all processes) CGs in the process of care CGs in the process of care. •Overall survival •Disease-free survival •Quality of life •Incidence based-mortality •Harm 2) Impact on health care costs a List of prioritised outcomes produced as a result of agreement between JRC experts and CCIB members and 2) healthcare costs. We included studies conducted Assessment Tool for Before-After (Pre-Post) Studies only in the 27 countries members of the European Union With No Control Group [27]; this tool includes 12 cri- at the time the review was conducted (as this systematic teria, each of them rated as presence or absence of risk review was conducted within the European Commission of bias. One author (IRC) carried out the risk of bias as- Initiative on Breast Cancer). Multicounty studies con- sessment, and a second author (AVM) cross-checked ducted in EU and non-EU countries were excluded. The this assessment. Disagreements were solved by discus- list of the 27 countries included is available in Add- sion or with the help of a third author. itional file 3. We only included studies published in Eng- lish. First, for calibration purposes two reviewers (IRC and AVM) independently screened 20% of the search results Data synthesis and analysis based on title and abstract and selected the references to Survival measures (overall survival and disease-free sur- be assessed based on full text. Then, the remaining 80% of vival) and impact on health cost were compared between the references were screened by a senior systematic re- adherent and non-adherent groups of patients. For sur- viewer (IRC). All references selected for full-text assess- vival outcomes, only studies reporting HRs using Cox ment were double-checked by a second author (AVM). adjusted models with at least 5 years of follow up were The level of agreement between both reviewers was 85%. pooled, using the Der Simonian and Laird random ef- Disagreements were solved by discussion or with the help fects model [28]. Statistical heterogeneity was assessed of a third author (ENDG). using Cochran Q and I2 measure. An I2 value above 50 and 75% was predefined as moderate and high hetero- Data extraction and risk of bias assessment geneity, respectively [29]. If selected studies had a similar We extracted and described in tables, the main charac- source of data, (i.e. the BRENDA cohort [30]) to avoid teristics of included studies (e.g. country, publication the risk of double counting populations or events, we se- year, guideline scope, study design, aim, and year of lected the cohort values with the longest follow-up and study, number of patients, patient’s characteristics, and with more complete data reporting. Results were adherence definition), and of the outcomes of interest. expressed as HR and the related 95% CI, a HR < 1 de- Data extraction was conducted in pairs, one author (IR) note advantage for adherence and HR > 1 denotes advan- conducted the first extraction, subsequently, a second tage for non-adherence to CGs. The anticipated absolute author (JPB) cross-checked the extracted data. If neces- effects estimates were measured as risk differences. We sary, the corresponding authors of the studies included collected the mortality and recurrence rates for a follow- were contacted to retrieve further information. up period of 60 months in both the intervention (i.e. ad- For risk of bias assessment, we used specific tools de- herent to CGs) and control (i.e. non-adherent to CGs) pending on the study design. For longitudinal studies groups. The estimation of risk differences was performed (including both prospective and retrospective follow-up indirectly from the pooled hazard ratios and the estima- designs) we used the Newcastle - Ottawa Quality Assess- tion of the baseline risk. For the latter, data was obtained ment Scale [26], which contains nine items classified in from Kaplan Meier survival curves from the control three main domains: selection, comparability, and out- groups [31]. We used Review Manager v.5.3 [32] for the come. The studies are graded according to a score from estimation of pooled hazard ratios, and we used GRAD 0 (highest risk of bias) to 9 (lowest risk of bias). The Epro [33] for the estimation of the absolute difference of following cut-off points were applied: < 5, high risk of events. We narratively synthesised findings for impact on bias; 5–7, moderate risk of bias; > 7, low risk of bias. For healthcare costs, and survival outcomes for the remaining non-controlled before-after studies, we used the Quality studies not included in the meta-analysis. Ricci-Cabello et al. BMC Health Services Research (2020) 20:920 Page 4 of 12 To assess the certainty of evidence we applied the total of 8137 unique citations from database searches, Grading of Recommendations Assessment, Development which were reviewed along with another 193 references and Evaluation (GRADE) approach [12]. We downgraded identified from ten identified systematic literature re- the certainty of the evidence based on the assessment views. We selected 112 references for review at the full- of: study limitations, inconsistency of results, indirect- text level. Of these, we excluded 91 publications (reasons ness of evidence, imprecision, and reporting bias. for exclusion available in Additional file 3), and finally GRADE principles apply also to prognostic questions included 21 studies [17–22, 30, 36–49]. [34]. In contrast to treatment effects questions, in prognosis questions, cohorts with a prospective de- sign, initially provide high certainty as they enable op- Characteristics of the included studies timal measurement of predictors and outcomes [35]. Most of included studies were conducted in Germany (15/ Quality of evidence is classified in four categories: 21; 71%) and were published during the last decade (19/21, high (which indicates we are very confident that the 90%). Most were retrospective cohort studies (18/21, 86%); true prognosis lies close to that of the estimate); mod- of these, 13 were single cohorts (BRENDA study [30]) but erate (we are moderately confident that the true prog- included different periods and/or addressed different clin- nosis is likely to be close to the estimate, but there is a ical questions; three were non-controlled before-after stud- possibility that it is substantially different); low (our ies. Two studies examined the impact of adherence to CGs confidence in the estimate is limited: the true progno- on healthcare costs, whereas 20 studies examined the im- sis may be substantially different), and; very low qual- pact on health outcomes (most frequently survival related ity (any estimate of effect is very uncertain) [34]. We measures). (Tables 2a and 2b, Additional file 4). No studies report the main findings both narratively and as tabu- were identified examining the rest of the outcomes consid- lated summaries. ered in this review (quality of life, incidence-based mortal- ity, and harm). Results Almost all studies (n = 20) examined the impact of Search results adherence to treatment guidelines, two also addressed The eligibility process of the original studies is sum- adherence to diagnostic guidelines and one addressed marised in a PRISMA flowchart (Fig. 1). We retrieved a adherence to guidelines for follow-up (Table 3). Fig. 1 PRISMA flowchart describing selection of included systematic reviews and original studies Ricci-Cabello et al. BMC Health Services Research (2020) 20:920 Page 5 of 12 Risk of bias assessment cancer CGs was associated with better overall survival Out of the 21 studies included, 17 presented low risk of rates (HR = 0.67 (95% CI 0.59 to 0.76), I2: 0%). This bias [17, 20–22, 36, 37, 39–44, 46–49] and four pre- means that for every 1000 patients, 129 more patients sented moderate risk of bias [18, 19, 38, 45]. The most (from 91 more to 165 more) would potentially survive in common reasons for risk of bias were related to the the group of patients managed in compliance with breast source of the non-exposed cohort, outcome assessment, cancer CGs, compared to those who were not (non-ad- and adequacy of follow-up (Tables 2a and 2b, Additional herent group). The certainty of evidence for adherence file 4). to CGs as a prognosis factor for overall survival was moderate due to risk of bias (studies relied only on the Impact on patient outcomes accuracy of medical records). Overall survival Additionally, another 18 studies not meeting the cri- The random effects meta-analysis of four studies [17, 40, teria for performing a meta-analysis as specified in the 49, 50] with 15,974 patients followed up for a median of methods section, assessed this outcome. Their character- 5 years (Fig. 2, Table 3), showed that adherence to breast istics and main results are available in Additional files 4 and 5. A direct association between adherence to breast Table 2a Characteristics of the included studies (n = 21) cancer CGs and better overall survival was more fre- Study characteristics n (%) References quently observed regardless of the study period [22], the Country use of different cut-offs or number of deviations to CGs France 2 (10%) [38, 39] recommendations used to determine adherence [17, 42, 48] and patient’s age [17, 18, 37, 41, 44, 45]. Although Germany 15 (71%) [18–22, 30, 37, 41–43, 45–49] non-adherence was associated with lower survival rate Italy 2 (10%) [17, 40] both in triple negative breast cancer (TNBC) and non- Netherlands 2 (10%) [36, 44] TNBC patients, this association was stronger in TNBC Publication year than in non-TBNC patients [19, 43]. Seven studies ex- 2015–2019 8 (35%) [17–22, 45, 49] plored adherence by treatment modality [18, 19, 30, 37, 2009–2014 11 (52%) [30, 40–44] [37, 39, 46–48] 43, 47, 48], revealing that there was better overall sur- vival in patients receiving CG-compliant treatments for ≤ 2008 2 (10%) [36, 38] breast-conserving therapy, chemotherapy, endocrine Study design therapy, and radiotherapy. One before after study did Non-controlled 3 (14%) [21, 38, 40] not find significant differences in survival rates compar- before-after study ing adherent versus non-adherent groups [21]. Retrospective 18 (86%) [17–20, 22, 30, 36, 37, 39, 41–49] cohort study Disease-free survival Guideline scope The random effects meta-analysis of three studies Diagnosis 2 (10%) [17, 40] followed up a median of 5 years with 9224 patients [20, Follow-up 1 (5%) [38] 36, 49] (Fig. 3, Table 3) showed better disease-free Treatment 20 (95%) [17–22, 30, 36, 37, 39–49] survival rates in patients receiving CG recommended Outcomes‡ treatment compared to those not receiving CG recom- mended treatment (HR = 0.35 (95% CI from 0.15 to Overall survival 18 (86%) [12–17, 24–27, 29, 31–37, 49] 0.82); I2 = 96%). This means that for every 1000 women Disease free survival 16 (76%) [13–15, 17, 24, 26, 28, 29, 31–37, 49] there were 336 more women free of recurrence (73 more Costs associated to 2 (10%) [38, 39] to 491 more) in patients managed in compliance with CGs adherence breast CGs compared to those who were not. The cer- Quality of life 0 (0%) – tainty of evidence for adherence to breast cancer CGs as Incidence-based 0 (0%) – a prognosis factor for disease-free survival was moderate. mortality As in overall survival, we considered rating down our Harm 0 (0%) – confidence by one level for risk of bias for the same rea- Risk of bias son (Table 3). Low 17 (81%) [17, 20–22, 36, 37, 39–44, 46–49] Additionally, this outcome was reported in another 15 Moderate 4 (20%) [18, 19, 38, 45] studies not meeting the criteria for performing a meta- analysis as specified in the methods section. An associ- High 0 (0%) – ation between adherence to breast cancer CGs and bet- ‡ Percentages exceed 100% because the categories are not mutually exclusive (i.e. some studies involved more than one type of guideline and more than ter disease-free survival was observed consistently one type of outcome) regardless of the study period [22, 47], use of different Ricci-Cabello et al. BMC Health Services Research (2020) 20:920 Page 6 of 12 Table 2b Characteristics of the clinical practice guidelines examined by the included studies Author/Year/Reference Scope of the Type of health care recommendation Guidelines studied guideline(s) adherence was studied Andreano 2017 [17] National (Italian) guideline Diagnosis and treatment (generic) NICE guideline a; ESMO guideline b and European guidelines* de Roos 2005 [36] National (Dutch) Treatment (treatment of patients Otter 2003c with DCIS) Ebner, Hancke 2015 [18] National (German) Treatment (generic) S3 guideline d Ebner 2015 [19] National (German) Treatment (generic) S3 guideline d Hancke 2010 [37] National (German) and Treatment (generic) 2005 St Gallen consensus e; S3 guideline d international guidelines* Jacke 2015 [21] National (German) Treatment (generic) S3 guideline d Mille 2000 [38] Regional (French) Follow-up (post therapeutic follow-up) Centre Régional Léon Bérard guidelines f Poncet 2009 [39] National and regional Breast cancer treatment (specific for French post licensing guidelines (2001), (French) guidelines trastuzumab treatment) regional clinical guidelines published by the regional oncology care network called “Convergence” in the French Rhone-Alpes area (no additional information provided). Sacerdote 2013 [40] Regional (Italy) guidelines Treatment (generic) Piedmont Clinical Practice Guideline g Schwentner 2012 [43] National (German) Treatment (adjuvant) S3 guideline d Schwentner 2012 [42] National (German) Treatment (generic) S3 guideline d Schwentner 2013 [41] National (German) Treatment (generic) S3 guideline d Van de Water 2012 [44] National (Dutch) Treatment (guidelines for breast and Dutch guideline h axillary surgery, radiotherapy, chemotherapy and endocrine therapy) Van Ewijk 2015 [45] National (German) Treatment (adjuvant treatment) S3 guideline d Varga 2010 [46] National (German) Treatment (treatment of early-onset S3 guideline d breast cancer) Wollschlager 2017 [20] National (German) Treatment (guidelines for adjuvant treatment) S3 guideline d Wockel, Kurzeder et al. 2010 [48] National (German) and Treatment (generic) 2005 St Gallen consensus e; S3 guideline d international guidelines* Wockel, Varga et al. 2010 [47] National (German) and Treatment (generic) 2005 St Gallen consensus e; S3 guideline d international guidelines* Wolters 2015 [22] National (German) Treatment (adjuvant treatment) S3 guideline d Wockel 2014 [30] National (German) Treatment (generic) S3 guideline d Wimmer 2019 [49] National (German) Treatment (radiotherapy) S3 guideline d * adherence to recommendations from two different guidelines examined in this study a, Early and locally advanced breast cancer diagnosis and treatment: full guideline. National Collaborating Centre for Cancer, Cardiff https://www.nice. org.uk/guidance/cg80/resources/early-and-locallyadvanced- breast-cancer-diagnosis-and- reatment-975,682,170,565 b, Senkus E, Kyriakides S, Ohno S et al. (2015) Primary breast cancer: ESMO Clinical Practice Guidelines for diagnosis, treatmentand follow-up. Ann Oncol 26:v8–v30. doi:https://doi.org/10.1093/annonc/ mdv298 c, Otter R (ed) (2003) Richtlijnen voor diagnostiek en behandeling van premaligne en maligne aandoeningen in de IKN-regio 2003 pp. 338–339. Groningen: IKN; ISBN 90–74,114–25-3 d, Kreienberg K, Kopp I, Lorenz et al. Interdisciplinary S3 Guidelines for the diagnosis and treatment of breast cancer in women. German Cancer Society. 2004 e, Goldhirsch A, Glick JH, Gelber RD et al. Meeting highlights: international expert consensus on the primary therapy of early breast cancer 2005. Ann Oncol 2005; 16: 1569–1583. 9 f, Centre Régional Léon Bérard, Réseau Oncora: The’saurus ONCORA en Cancérologie. Paris, France, Arnette Blackwell, 1997, p 374 g, Regione Piemonte Assessorato Sanità, Commissione Oncologica Regionale, Centro di Riferimento per l’Epidemiologia e la Prevenzione Oncologica in Piemonte: Tumore della mammella - linee guida clinico organizzative per la Regione Piemonte. 2002 h, Oncoline. Dutch National Breast Cancer Guidelines. http://www.oncoline.nl/mammacarcinoom cut-offs to determine adherence [43, 47, 48], patient’s received recommended breast-conserving therapy (70– age [18, 41] and subtype of tumour [42, 43]. Subgroup 79 years: 74–83%; > 79 years: 54%) than women aged analyses showed that adherence to recommendations in ≤69 years (93%) [37] (Additional file 5). The included CGs was directly associated with better (p < 0.05) disease studies did not report relevant data for the rest of the free-survival for most modalities of therapy [18, 19, 30, patient-important outcomes of interest (quality of life, 47, 48]. Women over 70 years of age less often (p < 0.05) incidence based-mortality, and harm). Table 3 Summary of findings Outcomes № of participants Certainty of the Relative effect (95% CI) Anticipated absolute effects (studies) Follow-up evidence (GRADE) Risk with Non-adherence Risk difference with adherence to CG to CG* Overall survival assessed with: 15,974 (4 non-randomised ⨁⨁⨁◯ MODERATE a,b HR 0.67 (0.59 to 0.76) [death] 431 per 1.000 c 138 more per 1.000 (96 more to 178) adjusted hazard ratio follow up: studies) [17, 40, 49, 50] Ricci-Cabello et al. BMC Health Services Research median 60 monthsa a,b,d Disease free survival assessed 9224 (3 non-randomised ⨁⨁◯◯ LOW HR 0.35 (0.15 to 0.82) [recurrence] 370 per 1.000 e 336 more per 1.000 (73 moreto 491 more) with: adjusted hazard ratio follow studies) [36, 49, 50] up: median 60 monthsa f, Treatment costsf (1 observational study) [39] ⨁⨁◯◯ LOW Higher costs for treatment concordant to CG. (2020) 20:920 f f, Follow-up costs (1 observational study) [38] ⨁⨁◯◯ LOW Non-CGs adherent follow-up was 2.2 to 3.6 times more expensive than the compliant one [5, 7]. *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI) CI Confidence interval, HR Hazard Ratio a) To assess this outcome we considered adherence to clinical guidelines as a prognosis factor for survival related outcomes. According to GRADE guidance for the assessment of prognosis clinical questions, observational studies are the main source of evidence, and have to potential to provide a high level of certainty. However, the optimal study design for prognosis questions are high quality prospective cohort studies. Although all the included studies used regression analyses to adjust for most important confounding clinical variables, we rated down our confidence in effects due concerns of serious risk of bias related with the study design (all selected studies were retrospective cohorts based on medical records or hospital database registries) b) Despite the existence of important differences across studies both in terms of the context of application, and of the conceptualisation of adherence applied, we did not rate down for indirectness, as results were highly consistent across different contexts and definitions c) The mortality rate in the group not receiving CG adherent treatment was 43.14% (1477/3424) d) We rated down one level our confidence in effects due to the high level of unexplained heterogeneity (I2: 96%). However, it should be noted that in the context of guideline development, it may not be appropriate to rate down for heterogeneity, since the desirable effects are highly consistent across studies e) Th recurrence rate in the group not receiving CG adherent treatment was 37% (578 / 1553) f) To assess the impact of adherence on costs we used an intervention (rather than a prognosis) approach, since cost is not a clinical outcome and the objective of this assessment was to examine a potential causal relationship. For this reason, the certainty of evidence was rated as “low”, as only observational studies were available GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect Page 7 of 12 Ricci-Cabello et al. BMC Health Services Research (2020) 20:920 Page 8 of 12 Fig. 2 Random effect meta-analysis of the association between adherence to breast cancer clinical guidelines with overall survival rate Impact on health care cost survival. We observed that for every 1000 women man- Poncet et al. (2009) observed that treatment costs with aged according to CGs compared to those that were not trastuzumab per patient and per year were higher for pa- (non-adherent), there were 138 more survivors and 336 tients receiving CGs-concordant treatment than for pa- more free of recurrence patients over 5 years of follow-up. tients receiving CGs discordant treatment (EUR 54975 vs. EUR 44186, respectively). The main reason for this Our results in the context of previous research difference was the treatment cost since the type of mole- Previous systematic reviews have shown that adherence cules for combination with trastuzumab recommended to breast cancer CGs [51] and other types of cancer CGs in the CGs were more expensive than other alternatives [52, 53] remains suboptimal. Sustainable use of CGs is not stated in the CGs [39]. Mille et al. (2000) observed also notably suboptimal, with studies showing up to a 50% that, in patients with localised breast cancer, the expend- decrease in adherence after 1 year of implementation [54]. iture associated to CGs concordant healthcare (consulta- Suboptimal adherence to CGs could increase healthcare tions and examinations) was consistently lower (from 2.2 costs if healthcare resources are overused (e.g. overtreat- to 3.6 times lower) compared to CGs discordant health- ment, overuse of diagnostic or screening techniques), but care. Unjustified examinations mainly explained this dif- also if they are underused (i.e. increased costs to cover ference [38]. The certainty of the evidence for the impact the additional healthcare needs that people may face on health care costs was low both for adherence to treat- with worsening conditions due to provision of inad- ment guidelines and for adherence to follow-up guidelines equate care). In our review, we identified no solid evi- because in both cases only observational studies were dence about the economic impact of non-adherence to available (Table 3, Additional file 4). breast cancer CGs. This finding supports the results from a previous systematic review, which noted an ab- Discussion sence of robust economic evaluations of the impact of Main findings CGs implementation [55]. In this systematic review and meta-analysis, we observed As far as we know, no previous systematic review has that there is moderate certainty that adherence to breast examined the impact of breast cancer CGs adherence as cancer CGs by healthcare providers is associated with a a prognosis factor of health outcomes. Our findings, substantial increase in overall survival and disease-free however, resonate with those from a previous systematic Fig. 3 Random effect meta-analysis of the association between adherence to breast cancer clinical guidelines with disease-free survival Ricci-Cabello et al. BMC Health Services Research (2020) 20:920 Page 9 of 12 review of adherence to gynaecologic cancer surgery CGs, greatly depending on the time chosen; in this work we which suggested that adoption of CGs was an effective selected the estimates at 5 years of follow up. The es- tool for disease control; noting that CGs adherence timation of absolute effect estimates was calculated should be considered as a process measure of quality indirectly using the pooled hazard ratios adjusted for cancer care [56]. A number of non-condition specific prognostic factors and the number of events in the systematic reviews have evaluated the effectiveness of control groups. We are aware that selecting different CGs [57–61]. These reviews (which differed in terms of baseline risks according to different settings or types population, interventions, and outcomes considered) of population would vary substantially the number of show a trend that supports that CGs improve patient events. outcomes. However, they consistently noted that the amount and quality of evidence upon which to de- Implications for practice and research velop their conclusions were very poor and that more The evidence identified in our review supports, with methodologically robust studies are needed to build a moderate certainty, that adherence to breast cancer CGs stronger evidence base around the impact of CGs is associated with greater survival and disease-free sur- adherence. vival. In light of our findings, the use of CGs should be more widely implemented. However, the way in which Strengths and limitations guideline adherence and implementation should be ef- Our review has several strengths. We followed a ro- fectively enhanced remains uncertain: although several bust methodology to produce quantitative estimates strategies have been proposed, and educational interven- of the impact of CGs adherence, conducting meta- tions are promising [64, 65], the evidence base about the analyses, evaluating for the first time, the impact of efficacy of the proposed interventions is still limited adherence to breast cancer CGs on patient-important [66]. Addressing this knowledge gap should be a priority health outcomes (i.e., outcomes that patients value for future research [67–69]. directly, instead of surrogate, outcomes that clinicians Our study provides valuable information, especially for may consider important) [62]. We conducted detailed health policymakers, to support the development of strat- data extraction from selected studies and used specific egies to implement CGs and/or reinforce CGs-adherence tools for risk of bias assessment. We followed best in breast cancer care. Greater use of explicit theory to practice for evidence assessment, using summary of understand barriers, design interventions, and explore me- finding tables and GRADE to assess the certainty of diating pathways and moderators is also much needed to effects [63]. advance in this area [70, 71]. In addition, the certainty of Our study has some limitations. First, we searched two evidence in relation to the impact of adherence to CGs on bibliographic databases and only one author conducted health care cost was low, and the available information is the initial screening based on titles and abstracts (i.e., based on studies published more than a decade ago, when only 20% of the references were cross-checked by a sec- the cost of Trastuzumab was considerably more expensive ond reviewer). Although data extraction and risk of bias than nowadays. Robust economic studies evaluating the assessment process was conducted in pairs using a impact of adherence to CGs on health care costs are cross-checking strategy rather than conducting them in- needed to progress in this area. dependently. by; these aspects could have limited our ability to identify additional studies. However, we in- Conclusion cluded studies from four countries in the EU, with con- Our systematic review of studies conducted in the EU sistent findings. It is therefore unlikely that additional shows that there is moderate certainty that adherence to studies would have substantially changed the results. breast cancer guidelines is associated with improved Second, we restricted the selection of studies to those overall and disease-free survival. The impact on cost is published in English and in EU countries (due to the inconclusive (low certainty). Breast cancer guidelines scope of the study), and most (71%) of the studies identi- should be rigorously implemented in the clinical setting. fied were conducted in a single country (Germany). This may limit the generalisability of our findings, which may Supplementary information not represent other contexts with different CGs develop- Supplementary information accompanies this paper at https://doi.org/10. ment methods or implementation strategies. And third, 1186/s12913-020-05753-x. since we framed our question as a prognosis question, our conclusions can only suggest an association rather Additional file 1. Figure 1A. Breast cancer mortality in Europe 2018 - estimates. Figure 1B. Breast cancer incidence in Europe 2018 -estimates than imply causality. Moreover, the absolute effect size Additional file 2. Search strategy of the risk of death or recurrence expressed should be Additional file 3. Excluded studies with reasons for exclusion interpreted with caution. The number of events will vary Ricci-Cabello et al. BMC Health Services Research (2020) 20:920 Page 10 of 12 Additional file 4. Characteristics of the included studies 2. ECIS - European Cancer Information System From https://ecis.jrc.ec.europa.eu, accessed on 19/08/2019. © European Union, 2019. Additional file 5. Adherence definitions and main findings 3. Early Breast Cancer Trialists' Collaborative Group (EBCTCG). Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet. 2005; Abbreviations 365(9472):1687-717. CGs: Clinical guidelines; CI: Confidence interval; EU: European Union; HR: Hazard ratio; PRISMA: Preferred Reporting Items for Systematic Reviews 4. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF. and Meta-analysis; BRENDA: “Breast Cancer Care Under Evidence Based Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Guidelines” retrospective cohort project; TNBC: Triple Negative Breast Cancer Sci U S A. 2003;100(7):3983–8. 5. Slamon DJ, Leyland-Jones B, Shak S, Fuchs H, Paton V, Bajamonde A, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic Acknowledgments breast cancer that overexpresses HER2. N Engl J Med. 2001;344(11):783–92. Not applicable. 6. van de Vijver MJ, He YD, van't Veer LJ, Dai H, Hart AA, Voskuil DW, et al. A gene-expression signature as a predictor of survival in breast cancer. N Engl Authors’ contributions J Med. 2002;347(25):1999–2009. IRC, AVM, ENDG, JPB, YS, MR, IS, DR, CC, PAC, ZSP, LN, EP were responsible 7. Van’t Veer LJ, Dai H, van de Vijver MJ, He YD, Hart AA, Mao M, et al. Gene for conducting the systematic review. IRC, DR, IS, CC, LN, EP, PAC contributed expression profiling predicts clinical outcome of breast cancer. Nature. 2002; to the definition of the research protocol. IS conducted the search. IRC, AVM, 415(6871):530–6. ENDG, JPB, YS, MR conducted data extraction and quality appraisal of 8. Berry DA, Cronin KA, Plevritis SK, Fryback DG, Clarke L, Zelen M, et al. Effect included studies. ENDG, AVM, JPB conducted the statistical analysis. IRC, of screening and adjuvant therapy on mortality from breast cancer. N Engl J ENDG, AVM, JPB, YS, MR, IS, DR, CC, PAC contributed to the interpretation Med. 2005;353(17):1784–92. and reporting of the results. IRC and ENDG drafted the first version of the 9. Gangnon RE, Stout NK, Alagoz O, Hampton JM, Sprague BL, Trentham-Dietz article. All authors reviewed critically and provided comments on subsequent A. Contribution of Breast Cancer to Overall Mortality for US Women. Med versions of the article. All authors read and approved the final manuscript Decis Making. 2018;38(1_suppl):24s–31s. prior to submission. 10. Kalager M, Zelen M, Langmark F, Adami HO. Effect of screening mammography on breast-cancer mortality in Norway. N Engl J Med. 2010; Funding 363(13):1203–10. The systematic review was carried out by the Iberoamerican Cochrane 11. Clinical Practice Guidelines We Can Trust. Washington DC: National Center under Framework contract 443094 for procurement of services Academy Press (US); 2011. Available from: https://www.ncbi.nlm.nih.gov/ between European Commission Joint Research Centre and Asociación books/NBK209546/. Colaboración Cochrane Iberoamericana. 12. Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, et al. AVM received a training grant D43 TW007393 Fogarty International Centre of GRADE: an emerging consensus on rating quality of evidence and strength the US National Institutes of Health for the Emerging Diseases and Climate of recommendations. BMJ. 2008;336(7650):924–6. Change Research Unit of the School and Public Health Administration at 13. Davis DA, Taylor-Vaisey A. Translating guidelines into practice: a systematic Universidad Peruana Cayetano Heredia. review of theoretic concepts, practical experience and research evidence in the adoption of clinical practice guidelines. Can Med Assoc J. 1997;157(4): Availability of data and materials 408–16. The datasets used and/or analysed during the current study are available 14. Ford I, Norrie J. Pragmatic trials. N Engl J Med. 2016;375(5):454–63. from the corresponding author on reasonable request. 15. Gandhi S, Verma S, Ethier JL, Simmons C, Burnett H, Alibhai SM. A systematic review and quality appraisal of international guidelines for early Ethics approval and consent to participate breast cancer systemic therapy: Are recommendations sensitive to different Not applicable. global resources? Breast (Edinburgh, Scotland). 2015;24(4):309–17. 16. Arts DL, Voncken AG, Medlock S, Abu-Hanna A, van Weert HC. Reasons for Consent for publication intentional guideline non-adherence: a systematic review. Int J Med Inform. Not applicable. 2016;89:55–62. 17. Andreano A, Rebora P, Valsecchi MG, Russo AG. Adherence to guidelines Competing interests and breast cancer patients survival: a population-based cohort study ZP, LN, and EP are current employees of the Joint Research Centre, European analyzed with a causal inference approach. Breast Cancer Res Treat. 2017; Commission. 164(1):119–31. IRC, ENDG, JPB, YS, MR, IS, DR, CC, PAC are employees of the Iberoamerican 18. Ebner F, Hancke K, Blettner M, Schwentner L, Wockel A, Kreienberg R, et al. Cochrane Center. Aggressive intrinsic subtypes in breast Cancer: a predictor of guideline adherence in older patients with breast Cancer? Clin Breast Cancer. 2015; Author details 15(4):e189–95. 1 Balearic Islands Health Research Institute (IdISBa), Palma, Spain. 2Primary 19. Ebner F, van Ewijk R, Wockel A, Hancke K, Schwentner L, Fink V, et al. Tumor Care Research Unit of Mallorca, Balearic Islands Health Service, Palma, Spain. biology in older breast cancer patients--what is the impact on survival 3 CIBER de Epidemiología y Salud Pública (CIBERESP), Madrid, Spain. 4Facultad stratified for guideline adherence? A retrospective multi-centre cohort study de Medicina Humana, Universidad Nacional Mayor de San Marcos, Lima, of 5378 patients. Breast (Edinburgh, Scotland). 2015;24(3):256–62. Peru. 5Iberoamerican Cochrane Centre - Department of Clinical Epidemiology 20. Wollschlager D, Meng X, Wockel A, Janni W, Kreienberg R, Blettner M, et al. and Public Health, Biomedical Research Institute Sant Pau (IIB Sant Pau), Sant Comorbidity-dependent adherence to guidelines and survival in breast Antonio María Claret 167, 08025 Barcelona, Spain. 6Interdisciplinary Centre for cancer-Is there a role for guideline adherence in comorbid breast cancer Health Studies (CIESAL), Universidad de Valparaíso, Valparaíso, Chile. patients? A retrospective cohort study with 2137 patients. Breast J. 2018; 7 European Commission, Joint Research Centre (JRC), Ispra, Italy. 24(2):120-7. 21. Jacke CO, Albert US, Kalder M. The adherence paradox: guideline deviations Received: 7 May 2020 Accepted: 22 September 2020 contribute to the increased 5-year survival of breast cancer patients. BMC Cancer. 2015;15:734. 22. Wolters R, Wischhusen J, Stuber T, Weiss CR, Krockberger M, Bartmann C, et al. References Guidelines are advantageous, though not essential for improved survival 1. Ferlay J EM, Lam F, Colombet M, Mery L, Piñeros M, Znaor A, among breast cancer patients. Breast Cancer Res Treat. 2015;152(2):357–66. Soerjomataram I, Bray F. Global Cancer Observatory: Cancer Today Lyon, 23. Riley RD, Hayden JA, Steyerberg EW, Moons KG, Abrams K, Kyzas PA, et al. France International Agency for Research on Cancer; 2018 [Available from: Prognosis research strategy (PROGRESS) 2: prognostic factor research. PLoS https://gco.iarc.fr/today. Med. 2013;10(2):e1001380. Ricci-Cabello et al. BMC Health Services Research (2020) 20:920 Page 11 of 12 24. Higgins JP, Altman DG, Gotzsche PC, Juni P, Moher D, Oxman AD, et al. The aged >65 who cannot participate in adjuvant clinical breast cancer trials? A Cochrane Collaboration's tool for assessing risk of bias in randomised trials. retrospective multi-center cohort study of 4,142 patients. Arch Gynecol BMJ. 2011;343:d5928. Obstet. 2015;291(3):631–40. 25. Moher D, Liberati A, Tetzlaff J, Altman DG, Group P. Preferred reporting 46. Varga D, Wischnewsky M, Atassi Z, Wolters R, Geyer V, Strunz K, et al. Does items for systematic reviews and meta-analyses: the PRISMA statement. guideline-adherent therapy improve the outcome for early-onset breast PLoS Med. 2009;6(7):e1000097. cancer patients? Oncology. 2010;78(3–4):189–95. 26. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the 47. Wockel A, Varga D, Atassi Z, Kurzeder C, Wolters R, Wischnewsky M, et al. assessment of the quality of nonrandomized studies in meta-analyses. Eur J Impact of guideline conformity on breast cancer therapy: results of a 13- Epidemiol. 2010;25(9):603–5. year retrospective cohort study. Onkologie. 2010;33(1–2):21–8. 27. NIH. Quality Assessment Tool for Before-After (Pre-Post) Studies With No 48. Wockel A, Kurzeder C, Geyer V, Novasphenny I, Wolters R, Wischnewsky M, Control Group 2014 [Available from: https://www.nhlbi.nih.gov/health-pro/ et al. Effects of guideline adherence in primary breast cancer--a 5-year guidelines/in-develop/cardiovascular-risk-reduction/tools/before-after. multi-center cohort study of 3976 patients. Breast (Edinburgh, Scotland). 28. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials. 2010;19(2):120–7. 1986;7(3):177–88. 49. Wimmer T, Ortmann O, Gerken M, Klinkhammer-Schalke M, Koelbl O, Inwald 29. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. EC. Adherence to guidelines and benefit of adjuvant radiotherapy in Stat Med. 2002;21(11):1539–58. patients with invasive breast cancer: results from a large population-based 30. Wockel A, Wolters R, Wiegel T, Novopashenny I, Janni W, Kreienberg R, et al. cohort study of a cancer registry. Arch Gynecol Obstet. 2019;299(4):1131-40. The impact of adjuvant radiotherapy on the survival of primary breast 50. Wollschlager D, Meng X, Wockel A, Janni W, Kreienberg R, Blettner M, et al. cancer patients: a retrospective multicenter cohort study of 8935 subjects. Comorbidity-dependent adherence to guidelines and survival in breast cancer- Ann Oncol. 2014;25(3):628–32. is there a role for guideline adherence in comorbid breast cancer patients? A 31. Skoetz N, Goldkuhle M, van Dalen EC, Akl EA, Trivella M, Mustafa RA, et al. retrospective cohort study with 2137 patients. Breast J. 2018;24(2):120–7. GRADE guidelines 27: how to calculate absolute effects for time-to-event 51. Henry NL, Hayes DF, Ramsey SD, Hortobagyi GN, Barlow WE, Gralow JR. outcomes in summary of findings tables and evidence profiles. J Clin Promoting quality and evidence-based care in early-stage breast cancer Epidemiol. 2020;118:124–31. follow-up. J Natl Cancer Inst. 2014;106(4):dju034. 32. Review Manager (RevMan) Version 5.3. ed. Copenhagen: The Nordic 52. Subramanian S, Klosterman M, Amonkar MM, Hunt TL. Adherence with Cochrane Centre, The Cochrane Collaboration; 2014. colorectal cancer screening guidelines: a review. Prev Med. 2004;38(5):536–50. 33. GRADEpro GDT: GRADEpro Guideline Development Tool [Software]. 53. Carpentier MY, Vernon SW, Bartholomew LK, Murphy CC, Bluethmann SM. McMaster University, 2020 (developed by Evidence Prime, Inc.). Available Receipt of recommended surveillance among colorectal cancer survivors: a from gradepro.org. systematic review. J Cancer Surviv. 2013;7(3):464–83. 34. Iorio A, Spencer FA, Falavigna M, Alba C, Lang E, Burnand B, et al. Use of 54. Ament SM, de Groot JJ, Maessen JM, Dirksen CD, van der Weijden T, Kleijnen J. GRADE for assessment of evidence about prognosis: rating confidence in Sustainability of professionals' adherence to clinical practice guidelines in estimates of event rates in broad categories of patients. BMJ. 2015;350:h870. medical care: a systematic review. BMJ Open. 2015;5(12):e008073. 35. Moons KGM, Royston P, Vergouwe Y, Grobbee DE, Altman DG. Prognosis 55. Hoomans T, Evers SM, Ament AJ, Hubben MW, van der Weijden T, and prognostic research: what, why, and how? BMJ. 2009;338:b375. Grimshaw JM, et al. The methodological quality of economic evaluations of 36. de Roos MA, de Bock GH, Baas PC, de Munck L, Wiggers T, de Vries J. guideline implementation into clinical practice: a systematic review of Compliance with guidelines is related to better local recurrence-free survival empiric studies. Value Health. 2007;10(4):305–16. in ductal carcinoma in situ. Br J Cancer. 2005;93(10):1122–7. 56. Ferron G, Martinez A, Gladieff L, Mery E, David I, Delannes M, et al. 37. Hancke K, Denkinger MD, Konig J, Kurzeder C, Wockel A, Herr D, et al. Adherence to guidelines in gynecologic cancer surgery. Int J Gynecol Standard treatment of female patients with breast cancer decreases Cancer. 2014;24(9):1675–8. substantially for women aged 70 years and older: a German clinical cohort 57. Bahtsevani C, Uden G, Willman A. Outcomes of evidence-based clinical study. Ann Oncol. 2010;21(4):748–53. practice guidelines: a systematic review. Int J Technol Assess Health Care. 38. Mille D, Roy T, Carrere MO, Ray I, Ferdjaoui N, Spath HM, et al. Economic 2004;20(4):427–33. impact of harmonizing medical practices: compliance with clinical practice 58. Grimshaw JM, Russell IT. Effect of clinical guidelines on medical practice: a guidelines in the follow-up of breast cancer in a French Comprehensive systematic review of rigorous evaluations. Lancet (London, England). 1993; Cancer Center. J Clin Oncol. 2000;18(8):1718–24. 342(8883):1317–22. 39. Poncet B, Colin C, Bachelot T, Jaisson-Hot I, Derain L, Magaud L, et al. 59. Lugtenberg M, Burgers JS, Westert GP. Effects of evidence-based clinical Treatment of metastatic breast cancer: a large observational study on practice guidelines on quality of care: a systematic review. Qual Saf Health adherence to French prescribing guidelines and financial cost of the anti- Care. 2009;18(5):385–92. HER2 antibody trastuzumab. Am J Clin Oncol. 2009;32(4):369–74. 60. Thomas LH, Cullum NA, McColl E, Rousseau N, Soutter J, Steen N. Guidelines 40. Sacerdote C, Bordon R, Pitarella S, Mano MP, Baldi I, Casella D, et al. in professions allied to medicine. Cochrane Database Syst Rev. 1999;(1). Art. Compliance with clinical practice guidelines for breast cancer treatment: a No.: CD000349. https://doi.org/10.1002/14651858.CD000349.. Available at: population-based study of quality-of-care indicators in Italy. BMC Health https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD000349/full. Serv Res. 2013;13:28. 61. Worrall G, Chaulk P, Freake D. The effects of clinical practice guidelines on 41. Schwentner L, Wockel A, Konig J, Janni W, Ebner F, Blettner M, et al. patient outcomes in primary care: a systematic review. CMAJ : Canadian Adherence to treatment guidelines and survival in triple-negative breast Medical Association journal = journal de l'Association medicale canadienne. cancer: a retrospective multi-center cohort study with 9,156 patients. BMC 1997;156(12):1705–12. Cancer. 2013;13:487. 62. Guyatt GH, Oxman AD, Kunz R, Atkins D, Brozek J, Vist G, et al. GRADE 42. Schwentner L, Wolters R, Wischnewsky M, Kreienberg R, Wockel A. Survival guidelines: 2. Framing the question and deciding on important outcomes. J of patients with bilateral versus unilateral breast cancer and impact of Clin Epidemiol. 2011;64(4):395–400. guideline adherent adjuvant treatment: a multi-centre cohort study of 5292 63. Schünemann HJ, Lerda D, Dimitrova N, Alonso-Coello P, Gräwingholt A, patients. Breast (Edinburgh, Scotland). 2012;21(2):171–7. Quinn C, et al. Methods for development of the European Commission 43. Schwentner L, Wolters R, Koretz K, Wischnewsky MB, Kreienberg R, initiative on breast Cancer guidelines: recommendations in the era of Rottscholl R, et al. Triple-negative breast cancer: the impact of guideline- guideline TransparencyEuropean commission initiative on breast Cancer adherent adjuvant treatment on survival--a retrospective multi-Centre guidelines. Ann Intern Med. 2019;171(4):273-80. cohort study. Breast Cancer Res Treat. 2012;132(3):1073–80. 64. Haggman-Laitila A, Mattila LR, Melender HL. A systematic review of the 44. van de Water W, Bastiaannet E, Dekkers OM, de Craen AJ, Westendorp RG, outcomes of educational interventions relevant to nurses with simultaneous Voogd AC, et al. Adherence to treatment guidelines and survival in patients strategies for guideline implementation. J Clin Nurs. 2017;26(3–4):320–40. with early-stage breast cancer by age at diagnosis. Br J Surg. 2012;99(6): 65. Gagliardi AR, Alhabib S. Trends in guideline implementation: a scoping 813–20. systematic review. Implement Sci. 2015;10:54. 45. Van Ewijk R, Wockel A, Gundelach T, Hancke K, Janni W, Singer S, et al. Is 66. Flodgren G, Hall AM, Goulding L, Eccles MP, Grimshaw JM, Leng GC, guideline-adherent adjuvant treatment an equal alternative for patients Shepperd S. Tools developed and disseminated by guideline producers to Ricci-Cabello et al. BMC Health Services Research (2020) 20:920 Page 12 of 12 promote the uptake of their guidelines. Cochrane Database Syst Rev. 2016; (8). Art. No.: CD010669. https://doi.org/10.1002/14651858.CD010669.pub2. Available at: https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858. CD010669.pub2/full. 67. Shekelle P, Woolf S, Grimshaw JM, Schunemann HJ, Eccles MP. Developing clinical practice guidelines: reviewing, reporting, and publishing guidelines; updating guidelines; and the emerging issues of enhancing guideline implementability and accounting for comorbid conditions in guideline development. Implement Sci. 2012;7:62. 68. Gagliardi AR, Brouwers MC. Do guidelines offer implementation advice to target users? A systematic review of guideline applicability. BMJ Open. 2015; 5(2):e007047. 69. Grimshaw JM, Thomas RE, MacLennan G, Fraser C, Ramsay CR, Vale L, et al. Effectiveness and efficiency of guideline dissemination and implementation strategies. Health Technol assess (Winchester, England). 2004;8(6):iii-iv, 1-72. 70. Davies P, Walker AE, Grimshaw JM. A systematic review of the use of theory in the design of guideline dissemination and implementation strategies and interpretation of the results of rigorous evaluations. Implement Sci. 2010;5:14. 71. Liang L, Bernhardsson S, Vernooij RW, Armstrong MJ, Bussieres A, Brouwers MC, et al. Use of theory to plan or evaluate guideline implementation among physicians: a scoping review. Implement Sci. 2017;12(1):26. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.