Adherence to oral anticancer chemotherapy: What influences patients’ over or non-adherence? Analysis of the OCTO study through quantitative–qualitative methods
- Aurélie Bourmaud1, 2Email author,
- Emilie Henin1,
- Fabien Tinquaut2,
- Véronique Regnier2,
- Chloé Hamant2,
- Olivier Colomban1,
- Benoit You1, 3,
- Florence Ranchon1, 4,
- Jérôme Guitton1, 5,
- Pascal Girard1,
- Gilles Freyer1, 3,
- Michel Tod1,
- Catherine Rioufol1, 4,
- Véronique Trillet-Lenoir1, 3 and
- Franck Chauvin1, 2, 6
© Bourmaud et al. 2015
Received: 30 December 2014
Accepted: 15 June 2015
Published: 4 July 2015
Numerous oral anticancer chemotherapies are available. Non-adherence or over-adherence to these chemotherapies can lead to lowered efficacy and increased risk of adverse events. The objective of this study was to identify patients’ adherence profiles using a qualitative–quantitative method.
A capecitabine treatment was initiated for 38 patients with advanced breast or colorectal cancer. At inclusion, information on patients’ beliefs was reported using a questionnaire. Later, Information on patients’ relation to treatment was obtained from a sub-group during an interview with a sociologist. Questionnaires were analyzed using Multiple Classification Analysis to cluster patients. Treatment adherence was evaluated by an electronic medication event monitoring systems (MEMS caps) and then correlated with patient clusters. Interviews were analyzed to complete and explain results.
38 patients were enrolled between 2008 and 2011 and completed the questionnaire. Twenty had adherence measured with MEMS caps all along treatment. Between 4 and 6 months after inclusion, 16 patients were interviewed. Patient profile B (retired, with a regular life, surrounded by a relative’s attention to drug adherence, with a low educational level) was statistically associated with adequate adherence (p = 0.049). A tendency for lower adherence was observed among more highly educated patients with an irregular, active life (NS). All patients taking capecitabine demonstrated a risk of over-adherence, potentiating side effects.
These encouraging primary results suggest that further studies should be undertaken and that educational programs tailored to patient profiles should be evaluated to enhance adherence for those who need it and to empower all patients to manage treatment side effects.
Oral anticancer therapy is being increasingly used every year, in particular cytotoxic agents, and more recent targeted therapies, are often administered orally. The main advantages of oral administration are the fewer hospitalisations and the autonomy gained by patients. However, these advantages can be compromised since the patient has to be responsible for the ambulatory administration and monitoring of these drugs. This raises two areas of potential concern: the patient’s adherence and the occurrence of unknown side effects. Capecitabine is a fluorouracil (5 FU) prodrug and is prescribed orally for the treatment of advanced and metastatic breast, colorectal and gastric cancers. Several studies have reported patient adherence rates for capecitabine that vary from 58 to 100%, mainly dependant on the measurement method [1–11]. In most studies, non-adherence rate remains at about 20–25% [2, 3, 5, 7, 11]. The reasons for non-adherence can be grouped under three main themes : Personal factors (belief in the treatment, emotional state); treatment factors (complexity of treatment, side effects, costs) and healthcare provider-related factors (relationship with healthcare professionals, prescribing practices). Few factors have been identified to explain non-adherence with capecitabine: the number of co-medications, and the number of side effects [3, 9, 10, 13]. Adherence with capecitabine is relatively high compared with other oral cancer chemotherapies . It has been suggested that there is only one specific group of patients who are non-adherent and that adherence-enhancing intervention should be targeted to this group only [7, 9, 13]. The difficulty is to be able to identify them early and propose the interventions to them.
The second area of concern for oral chemotherapy is the monitoring of side effects. Capecitabine is responsible for gastrointestinal and dermatologic side effects that can lead to life-threatening toxicity, which can be resolve by temporarily interrupting treatment or modifying the dose. However, serious side effects still occur with capecitabine  and these are often related to over-adherence since, when side effects occur, there is no medical surveillance or management of doses, the patients continue to take the drug [16, 17]. This over-adherence has been observed, particularly, with capecitabine [2, 3, 6, 8, 18]. There is, therefore a need to develop interventions that can enable quick recognition of these side effects and their timely, adequate management.
The objective of the OCTO study (Observance à une ChimioThérapie Orale—Adherence to oral chemotherapy) was to model adherence of ambulatory patients treated with capecitabine. An ancillary study (OCTOquali) within the main OCTO study was conducted assuming that there were profiles for patients to determine their non-adherence or over-adherence. The identification of these profiles would allow early tailored, patient education programmes for the adequate administration and monitoring of their care to be proposed to the patients.
The main objective of the OCTOquali study was to define patient profiles that will predict and prevent poor adherent behaviour.
The OCTOquali study was a prospective monocenter cohort study that recruited patients initiating an oral capecitabine treatment between November 2008 and September 2011. The patients answered a questionnaire at inclusion and a sub-group of patients was interviewed 4–6 months after the inclusion in the study to explore their beliefs and behavior related to their treatment. These results were analyzed to identify sociological profiles among the patients and compare these profiles with the patients’ adherence data.
Patients were recruited in the medical oncology department at the Lyon-Sud university hospital of Lyon (France). Consecutive adults with colorectal cancer at a resected or metastatic stage or breast cancer at a locally advanced or metastatic stage, with an indication for an ambulatory capecitabine treatment, were screened by the oncologists and recruited for the study. Written informed consent was obtained from each participant. The study was approved by Lyon’s South-East ethical committee number 4 (approval number 2008-004097-41).
Patients were planned to be followed during six treatment cycles. The treatment dose was based on their body mass index (BMI), and prescribed as a combination of 150 and 500 mg tablets. Treatment was taken as morning and evening doses, at exactly 12 h intervals, within 30 min after a meal. Tablets could not be split to help swallowing. A treatment cycle was 14 days of treatment followed by a ‘rest’ week. Half the patients (N = 20) were given a bottle of capecitabine with a MEMS cap (Medication Electronic Monitoring Systems, Aardex), to measure their adherence electronically. At inclusion, the patients completed an initial questionnaire exploring socio-economic characteristics and beliefs about their new oral chemotherapy, including self-predicted adherence. 4 to 6 month after inclusion, the patients were proposed a semi-structured interview, conducted by a sociologist to evaluate their representations and behavioral adaptations made since initiating capecitabine treatment.
Measure of adherence
Adherence was measured during all treatment intake (at most 6 cycles) using the timing of each opening of the bottle with MEMS caps for 20 patients: This was used to define three types of adherence. Patients with an opening time variability of >1 h standard deviation were considered to have poor adherence, those with an opening time variability of <15 min were considered to have high adherence and patients with a variability between were considered to have adequately adherence. Those cut-offs were defined as clinical relevant by a multidisciplinary team (two oncologists and one sociologist). Missed doses were very rare in this cohort so we did not measured adherence this way.
The questionnaire was constructed by a sociologist, an oncologist and a methodologist. The questionnaire was tested on five patients and modified to improve its clarity. The questionnaire, which contained 32 closed questions answered directly by the patients, focussed on their perceptions and expectations (Additional file 1). It examined socio-economics conditions, relationship with their oncologist, beliefs in treatment efficacy and toxicity, expected adherence, expected changes in behaviour related to taking oral chemotherapy.
Face-to-face interviews were proposed to patients after the end of the adherence measurement period (6 cycles at most). Patients were randomly recruited for the interviews, sampling continuing until data saturation was achieved. Interviews were conducted face-to-face with a sociologist, lasted at least 1 h and investigated those topics : (i) the declared adherence behavior and perception (ii) the patient understanding of his treatment and its management (iii) the place of relatives in treatment management (iv) patient’s adverse event declaration and management.
Patients’ answers were first analysed descriptively by frequencies (percentages) and medians (interquartiles). Then they were analysed with a multiple classification analysis (MCA) followed by an ascending hierarchical classification (AHC), which is an agglomerative hierarchical clustering procedure, to statistically and graphically identify patient clusters.
Adherence categories identified with MEMS caps were projected onto a MCA cluster map to graphically identify adherence clusters. Patients without adherence measured with MEMS caps were considered as having missing data. Then categories of adherence were compared with the whole clusters tendency using univariate analyses. All analyses were performed using SAS (version 9.1) with a significance threshold of p < 0.05.
The interviews were analysed with socio-analytical methods: the different elements of the discourse during each interview were connected and then underwent content analysis.
Clusters identified with the MCA methods were linked with the results of the interview analyses, to evaluate the concordance of common topics, to explain the MCA findings and to describe profiles of interest.
Characteristics, N = 38
Patients, n (%) or mean (±sd)
As a couple without children
As a couple with children
On sick leave
Actual or former profession
Artisan or laborer
Does not apply
Disease related characteristics
Type of cancer
Length of oral chemotherapy
Reason for discontinuation
Responses to the questionnaire
Responders, N = 38
Already had taken oral chemotherapy
Knew about the choice between oral/intravenous
Knew the prescribed dose
Had another treatment for their cancer
Takes a treatment for side effects related to previous cancer treatment
Takes a treatment for another condition
Feelings related to capecitabine
Advantages of the oral route, according to the patient
Fewer side effects
Disadvantages of the oral route, according to the patient
The lack of rigour in drug administration
Relationship with the oncologist
Explanations given by the oncologist about:
The treatment (organisation, administration) were insufficient
The treatment side effects occurrence were insufficient
Management of the treatment side effects were insufficient
Questions asked to the oncologist about the first prescription capecitabine
Loss of hair
Intention to obtain more information elsewhere
Expected changes in daily life due to capecitabine
Will have to be organised around the treatment administration
If side effects appear:
I will stop the treatment
I will consult a clinician
I will continue the treatment whatever
Opinions about adherence
I think that it is alright to miss a dose
I think that it is alright to stop voluntarily
I think that missing a dose is dangerous
Did not want to answer the question
Clusters identified by the MCA and AHC using questionnaires’ responses
description of the three clusters identified by CMA and AHC
Clusters, N = 38
Cluster A, N = 16
Educational level: college
Actual of former profession: executive
Reason for discontinuation: disease progression
Length of oral chemotherapy ≤3 cycles
Intention to get more information elsewhere
Thought a dose could be missed during treatment
In couple with children
Cluster B, N = 9
Educational level: primary school
Thought that missing a dose was serious
Cluster C, N = 13
Educational level: secondary school
No discontinuation of protocol
Length of oral chemotherapy ≥3 cycles
Did not know the prescribed dose well
Occupation: sick leave
Did not think that a dose could be missed during treatment
Already had had oral chemotherapy
Did not know there was a choice between oral/IV treatment
Adherence results with MEMS records
Among the 20 patients whose adherence was measured electronically, adherence was globally estimated as concordant with the prescription since only 23 missed doses were recorded for 2,272 predicted doses (99% adherence). Two patients were classified as being less adherent (10%), 4 as being highly adherent (20%) and 14 (70%) as being adequately adherent, based on the mean inter-dose interval, which was estimated to be 12.15 h (Additional file 1: Figure S2). The standard deviation for the morning dose was estimated to be 1.003 h and 1.06 for the evening doses.
Cluster profile and adherence behaviour association
Comparison of interview results and MCA results: confirmation of profiles by a mixed method
Detailed results from the interviews are reported in the Additional file 1.
Interviews identified two different profiles based on feelings and behaviour related to risk of non-adherence to capecitabine (Additional file 1: Table S1). MCA showed that one profile was similar to cluster A and the other was similar to both clusters B and C. Patients who had active lives, with an active partner and children still at home, said that was difficult and constraining to adjust to the capecitabine administration. They said their relatives were too busy and not available to share their adherence issue, leaving the patient alone with this responsibility. This loneliness generated anxiety among those patients, who were worried about forget to take a dose. These patients liked the idea of supervised IV administration in a hospital setting, in order to decrease their loneliness by being about to talk with healthcare providers. In contrast, patients with a regular life, and supportive and available relatives said that capecitabine administration was easy to organize and they tolerated it well. They shared the responsibility of adherence with their partner who helped them with the timing of the doses. They said they were satisfied and were not anxious about taking oral chemotherapy at home.
We were unable to identify profiles of patients for their feelings and behaviour related to side effects and over-adherence. They all had a great expectation in the efficacy of this treatment. This hope surpassed any criticism towards capecitabine intake and any side effects occurrence. What patients declared to fear at most was dose reduction. So face to side effects occurrence, all patients developed their own management strategy, in order to continue to take the treatment whatever the cost.
This study combined complex descriptive statistical methods and a sociological approach to identify, explain and confirm two adherence profiles: adequately adherent (Cluster B) and less adherent (Cluster A). Patients with no professional or family obligations, retired, and leading a calm and scheduled life, were found to be adequately adherent. Adherence was enhanced when the patient had a relative who helped them manage their treatment with them. These patients had a lower educational level, and followed the prescription strictly, for fear they would forget a dose. In contrast, patients with an irregular, active life, with family and professional obligations were more likely to be non-adherent. To keep a schedule when unexpected and external events can occur is more difficult. These patients experienced loneliness and thought they were benefit from having contact with a health professional, to share the burden of treatment, even that meant replacing the oral treatment with intravenous treatment. These patients had a higher educational level and were more aware that they could miss a dose of their treatment. This highlights that they are more in need of special support to help them manage their treatment.
The results showed that patients under oral capecitabine generally adhere to their treatment, with those in Cluster A being occasionally non-adherent (Cluster A). The occurrence side effects and their management was the main issue for all the patients, as they all reported having experienced side effects, often quite severe (62.5%). The interviews with the sociologist brought out the fact that some of these side effects were probably due to over-adherence: All the patients had high expectations about the efficacy of the capecitabine treatment, and were all ready to suffer from any kind of side effect, as long as the treatment was not stopped. Thus, they developed their own to strategies to support the side effects and were reluctant to tell their oncologist about them out of fear that the treatment would be stopped. This behaviour could lead to the side effects becoming more severe suggesting that there is a need for tailored support to teach patients how to manage the side effects.
Our adherence results are coherent with previously published results, i.e. 100% if adherence is defined as the number of forgotten tablets [1, 9]. However, when adherence is measured with more specific tools, some patients are less adherent than the others (from 40 to 9%, according to the studies [1, 2, 9, 10]), so our estimation of 10% of non-adherent patients is at the low end of this range, but remains consistent.
Some studies that measured over-adherence reported this to be between 4 and 44% [2, 6, 8]. In this study we did not specifically measure over-adherence, but rather investigated patients’ beliefs and behaviour related to over-adherence. All patients said they could be over-adherent during the interviews, giving a risk of over-adherence of 100%. This was supported by their misconception about efficacy and side effects reported on the questionnaire they completed when they received the prescription. Patients systematically overestimated the balance between the advantages and disadvantages of the oral treatment. They underestimated the risk of side effects. Even before starting the treatment, 10.5% of the patients said they would not stop treatment if side effects occurred. This misgiving increased over time, as was observed in the interviews.
Our results showed that there was a good relationship between the oncologist and patient; only three patients thought the explanations about side effects were insufficient and only one thought explanations about the treatment were insufficient). In addition, the results showed that the patients were motivated to be an active stakeholder in their care: 34.2% said they were going to find more information about the treatment, and 68.4% said they were going to organise their daily life around the treatment administration. These two positive attitudes should be used by healthcare providers to enhance effective management of treatment administration and monitoring by the patients. Patients appeared to be eager to become independent about their treatment, so a solution could be that their oncologist, that they trust could propose to them to participate in a programme to help them to manage their treatment. A therapeutic education programme would seem to be the most relevant intervention in this context. Therapeutic education programmes have been shown to be effective in improving treatment adherence in patients with non-oncological pathologies [19, 20]. It is recommended in the setting of oral capecitabine [2, 21] and would empower the patients.
The main limitation of this study is that only 38 patients were included, with only 20 of them having adherence measured with MEMS. This led to a lack of power, and to non-significant results for 2 of the 3 profiles identified. However, the AHC and MCA analyses enabled us to identify an adequate adherence profile. Also, graphically there seemed to be a close relationship between low adherence and Cluster A, even if this was not statistically significant. What’s more, the content analyses of the 16 sociologist’s interviews provided a wealth of information which was consistent with the quantitative results, and confirmed findings. Another limitation is the little number of patients with poor and high adherence. Thus it is difficult to say if the lack of association between those profiles and adherence behavior is due to lack of statistical power or if there is actually no association. The last limitation is that there was no objective measurement of over adherence and assessment of its relationship to the occurrence and severity of side effects, although it seems that this assumption is valid based on the patients’ interviews. The results from this study have now to be confirmed in a larger study. If confirmed, those profiles could be used to develop targeted educational interventions, with the objective of enhancing adherence and side effects management through empowerment of patients.
In conclusion, this study, with a mixed quantitative–qualitative method, identified an adequate adherence profile for oral capecitabine. There also seems to be a specific low-adherence profile. The results show that over adherence and inadequate management of side effects are major issues with oral capecitabine treatment and provide some explanation about the patients’ behavior. Future studies should confirm those results and evaluate the role of tailored therapeutic educational interventions. Patients are looking forward for such interventions, driven by their healthcare providers.
Availability of supporting data
The data sets supporting the results of this article are included within the article (and its additional file).
VTL, PG, MT and FC conceived the project. EH, OC, VR, BY, FR, JG, GF, CR collected all the data. EH, AB and FT did the statistical analyses. CH did the sociologic analyses. AB wrote the first draft and revised drafts of the manuscript. VTL, PG, MT, FC, EH, OC, VR, BY, FR, JG, GF, CR, FT and CH critically revised the manuscript for important intellectual content. All authors had full access to all of the data in the study and can take responsibility for the integrity of the data. AB and EH take responsibility for the integrity of the article. AB is the guarantor for the study. All authors read and approved the final manuscript.
The authors would like to thank Margaret Haugh (MediCom Consult) for translation and editing services funded by the Hygée Centre.
The OCTO study was funded by Roche Laboratory.
Compliance with ethical guidelines
Competing interests Emilie Hénin received funding from ‘Fondation Synergie Lyon Cancer’ and the French ‘La Ligue Nationale contre le Cancer’. The other authors declare that they have no competing interests.
Ethical approval The Lyon’s South-East ethical committee approved the protocol.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. 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.
- Walter T, Wang L, Chuk K, Ng P, Tannock IF, Krzyzanowska MK (2013) Assessing adherence to oral chemotherapy using different measurement methods: Lessons learned from capecitabine. J Oncol Pharm Pract 20:249–256PubMedView ArticleGoogle Scholar
- Bhattacharya D, Easthall C, Willoughby KA, Small M, Watson S (2012) Capecitabine non-adherence: exploration of magnitude, nature and contributing factors. J Oncol Pharm Pract 18:333–342PubMedView ArticleGoogle Scholar
- Partridge AH, Archer L, Kornblith AB, Gralow J, Grenier D, Perez E et al (2010) Adherence and persistence with oral adjuvant chemotherapy in older women with early-stage breast cancer in CALGB 49907: adherence companion study 60104. J Clin Oncol 28:2418–2422PubMed CentralPubMedView ArticleGoogle Scholar
- Uematsu T, Nakashima M, Fujii M, Hamano K, Yasutomi M, Kodaira S et al (1996) Measurement of 5-fluorouracil in scalp hair: a possible index of patient compliance with oral adjuvant chemotherapy. Eur J Clin Pharmacol 50:109–113PubMedView ArticleGoogle Scholar
- Macintosh PW, Pond GR, Pond BJ, Leung V, Siu LL (2007) A comparison of patient adherence and preference of packaging method for oral anticancer agents using conventional pill bottles versus daily pill boxes. Eur J Cancer Care (Engl) 16:380–386View ArticleGoogle Scholar
- Mayer EL, Partridge AH, Harris LN, Gelman RS, Schumer ST, Burstein HJ et al (2009) Tolerability of and adherence to combination oral therapy with gefitinib and capecitabine in metastatic breast cancer. Breast Cancer Res Treat 117:615–623PubMedView ArticleGoogle Scholar
- Simons S, Ringsdorf S, Braun M, Mey UJ, Schwindt PF, Ko YD et al (2011) Enhancing adherence to capecitabine chemotherapy by means of multidisciplinary pharmaceutical care. Support Care Cancer 19:1009–1018PubMed CentralPubMedView ArticleGoogle Scholar
- Spoelstra SL, Given BA, Given CW, Grant M, Sikorskii A, You M et al (2013) An intervention to improve adherence and management of symptoms for patients prescribed oral chemotherapy agents: an exploratory study. Cancer Nurs 36:18–28PubMedView ArticleGoogle Scholar
- Krolop L, Ko Y-D, Schwindt PF, Schumacher C, Fimmers R, Jaehde U (2013) Adherence management for patients with cancer taking capecitabine: a prospective two-arm cohort study. BMJ Open 3:7View ArticleGoogle Scholar
- Winterhalder R, Hoesli P, Delmore G, Pederiva S, Bressoud A, Hermann F et al (2011) Self-reported compliance with capecitabine: findings from a prospective cohort analysis. Oncology 80:29–33PubMedView ArticleGoogle Scholar
- Decker V, Spoelstra S, Miezo E, Bremer R, You M, Given C et al (2009) A pilot study of an automated voice response system and nursing intervention to monitor adherence to oral chemotherapy agents. Cancer Nurs 32:E20–E29PubMedView ArticleGoogle Scholar
- Ruddy K, Mayer E, Partridge A (2009) Patient adherence and persistence with oral anticancer treatment. CA Cancer J Clin 59:56–66PubMedView ArticleGoogle Scholar
- Thivat E, Van Praagh I, Belliere A, Mouret-Reynier MA, Kwiatkowski F, Durando X et al (2013) Adherence with oral oncologic treatment in cancer patients: interest of an adherence score of all dosing errors. Oncology 84:67–74PubMedView ArticleGoogle Scholar
- Bassan F, Peter F, Houbre B, Brennstuhl MJ, Costantini M, Speyer E et al (2014) Adherence to oral antineoplastic agents by cancer patients: definition and literature review. Eur J Cancer Care (Engl) 23:22–35View ArticleGoogle Scholar
- Weingart SN, Toro J, Spencer J, Duncombe D, Gross A, Bartel S et al (2010) Medication errors involving oral chemotherapy. Cancer 116:2455–2464PubMedGoogle Scholar
- Regnier Denois V, Poirson J, Nourissat A, Jacquin J-P, Guastalla JP, Chauvin F (2011) Adherence with oral chemotherapy: results from a qualitative study of the behaviour and representations of patients and oncologists. Eur J Cancer Care 20:520–527View ArticleGoogle Scholar
- Patel K, Foster NR, Farrell A, Le-Lindqwister NA, Mathew J, Costello B et al (2013) Oral cancer chemotherapy adherence and adherence assessment tools: a report from North Central Cancer Group Trial N0747 and a systematic review of the literature. J Cancer Educ 28:770–776PubMedView ArticleGoogle Scholar
- Spoelstra SL, Given BA, Given CW, Grant M, Sikorskii A, You M et al (2013) Issues related to overadherence to oral chemotherapy or targeted agents. Clin J Oncol Nurs 17:604–609PubMedView ArticleGoogle Scholar
- Who Working Group (1988) Therapeutic Patient Education Continuing Education Programmes for Health Care Providers in the Field of Prevention of Chronic Diseases. WHO Regional Office for Europe, CopenhagueGoogle Scholar
- Hartigan K (2003) Patient education: the cornerstone of successful oral chemotherapy treatment. Clin J Oncol Nurs 7(6 Suppl):21–24PubMedView ArticleGoogle Scholar
- Chau I, Legge S, Fumoleau P (2004) The vital role of education and information in patients receiving capecitabine (Xeloda). Eur J Oncol Nurs 8(Suppl 1):S41–S53PubMedView ArticleGoogle Scholar