Screenplays & Play

A study of a multi-level intervention to improve non-adherence in difficult to control asthma

Description
A study of a multi-level intervention to improve non-adherence in difficult to control asthma
Published
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  A study of a multi-level intervention to improvenon-adherence in difficult to control asthma Jacqueline Gamble a,b , Michael Stevenson c , Liam G. Heaney a,b, * a Centre for Infection and Immunity, Queen’s University of Belfast, 97 Lisburn Road, Belfast BT9 7BL, Northern Ireland,United Kingdom b Regional Respiratory Centre, Belfast City Hospital, Lisburn Road, Belfast BT9 7AB, Northern Ireland, United Kingdom c Department of Epidemiology and Public Health, Queen’s University of Belfast, Royal Victoria Hospital, Grosvenor Road,Belfast BT12 6BJ, Northern Ireland, United Kingdom Received 8 November 2010; accepted 24 March 2011Available online 20 April 2011 KEYWORDS Difficult asthma;Medication;Adherence;Intervention;Concordance Summary Background:  Difficult to control asthma accounts for significant morbidity and healthcare cost,and non-adherence to medication is a common cause. It remains unclear if targeting non-adherence in this population improves healthcare outcomes. Methods:  All subjects were referred to a Specialist Difficult Asthma Service (60% from Respira-tory physicians); poor adherence was identified using prescription refill records for inhaledcombination therapy. A sequential 2 phase study examined the effect of identifying and tar-geting non-adherence to inhaled long-acting  b -agonist/inhaled steroid combination therapy;phase 1  e  an observational study utilising objective measures of non-adherence to facilitatea medical concordance discussion followed by phase 2, a 12 month prospective single blindrandomised controlled trial where subjects with persistent poor adherence were randomisedto a nurse-led menu driven intervention. Results:  A total of 239 patients were assessed; 31 of 83 subjects (37%) who were initially non-adherent, significantly improved adherence after concordance interview, with reducedprescribed daily dose of ICS (data  p < 0.001), rescue prednisolone courses (data,  p < 0.001)and hospital admissions (data,  p Z 0.006). With the menu driven intervention, adherence alsoimproved (intervention 37.6% to 61.9%, control group 31.7% to 28.8%) with reduced mainte-nance oral steroid dose in subjects on maintenance steroids. Conclusion:  Poor adherence in difficult-to control asthma is common, but when identified and tar-geted can be improved and this is associated with large improvements in important healthcareoutcomes.Previousnihilismtowardsnon-adherenceinthispopulationisnotsupportedbythisstudy. ª 2011 Elsevier Ltd. All rights reserved. * Corresponding author at: Regional Respiratory Centre, Belfast City Hospital, Lisburn Road, Belfast BT9 7AB, Northern Ireland, UK.Tel.:  þ 44 28 90 263821; fax:  þ 44 28 90 263879. E-mail address:  l.heaney@qub.ac.uk (L.G. Heaney). available at www.sciencedirect.comjournal homepage: www.elsevier.com/locate/rmed Respiratory Medicine (2011)  105 , 1308 e 13150954-6111/$ - see front matter   ª  2011 Elsevier Ltd. All rights reserved.doi:10.1016/j.rmed.2011.03.019  Introduction Asthma is a leading, preventable cause of morbidity,mortality and economic burden, estimated to affect 300million people worldwide. 1,2 Despite treatment with highdose therapy approximately 5% of adult patients remaindifficult to control with persisting symptoms and frequentexacerbations. 3 This sub-group of patients accounts for a disproportionate amount of asthma morbidity and healthcare costs and are a particular cause for concern because ofthe potential consequences of uncontrolled diseaseincluding fatal or near fatal asthma. 4,5 We have previously reported a high prevalence of non-adherence to inhaled and oral steroid medication, in adultsubjects referred to a Regional Difficult Asthma Service(difficult asthma was defined as persistent symptoms despiteprescribed treatment at Step 4/5 of the British ThoracicSociety/Scottish Intercollegiate Guideline Network (BTS/SIGN) treatment steps). 6 We also demonstrated that non-adherencewasassociatedwithincreasedhospitaladmissions,impaired quality of life and was more common in women. 6 A key question remains whether poor adherence can beaddressed in this difficult to control population. There isdata on adherence interventions in mild/moderate asthma,where a recent Cochrane concluded that the evidence for any intervention improving adherence was ‘surprisinglyweak’, with little suggestion that medication adherencecan be improved consistently and on a long term basis. Thisreview also concluded that increasing effectiveness ofadherence interventions might have a greater impact onthe health of the population than any improvement inspecific medical treatments. 7 However, there is no data on interventions in adultsubjects with difficult asthma. We have previously reportedthat patients with difficult asthma have individually specificreasons for non-adherence, many of which (e.g. lack ofknowledge, drug misconceptions and steroid phobia) couldbe addr essed using an individualised menu driven inter-vention. 8 Techniques such as audio-visual reminders,education and diary keeping and psychological or behav-ioural based interventions, which have been positivelyutilised used in mental health, 9 e 14 also seemed to beappropriate to utilise in subjects with difficult asthma.The aim of this study was to determine if non-adherenceto inhaled combination LABA/steroid medication whenidentified in difficult asthma, could be improved using (a)a simple concordance interview and (b) a menu drivenpsycho-educational intervention strategy, with better asthma outcome. Methods Participants All subjects were attending the Northern Ireland RegionalDifficult Asthma Service, which assesses 70 e 80 newpatients a year, of which 60% are tertiary referrals fromother Respiratory Specialists; at the time of the study, 239patients were attending the Service. The Service has a  ca 65% ‘discharge rate’ after assessment and stabilisationthrough our multi-disciplinary assessment protocol, back toprimary care or to the referring Respiratory Specialist. 15 None of the subjects had non-adherence suspected asa major clinical issue at the time of referral and all deniednon-adherence at the time of first clinical assessment atthe clinic. Difficult asthma was defined as persistentsymptoms, despite treatment at BTS/SIGN step 4/5. 16 Measuring adherence Adherence was measured as previously described andvalidated, using prescription refill records. 6 In NorthernIreland, all prescribed medication is obtained viaprescription from a single General Practitioner (GP) andprescription refills for inhaled combination therapy (ICT)were obtained for the previous 6 months and expressed asa percentage of prescribed medication. Non-adherence wasdefined as prescription filling of  50%, which again we havevalidated as a meaningful ‘cut-off’ value for identifyingexcessive healthcare utilisation and poor asthma outcome. 6 Patient demographics, hospital admissions, lung func-tion, oral prednisolone courses, and quality of life datawere obtained from medical records.The study consisted of 2 phases and ethical approval for the study was obtained from the Office for Research EthicsCommittees Northern Ireland (ORECNI). In Phase 1, whennon-adherence was identified, a patient concordancediscussion was arranged, which communicated the fact thatpatients were non-adherent with medication and a treat-ment plan was agreed to address poor adherence. The aimof phase 1 was to examine the effect of both the commu-nication to the patient, that the treating healthcare teamwas aware of their non-adherence and the effect of theconcordance discussion. Patients who became adherentafter initial concordance discussion during this phase(defined as adherent at Screen 2  e  see below), were fol-lowed up 6 months after this point to monitor their adherence and asthma outcome.All patients had prescription records rescreened after the initial concordance discussion (Screen 2, time of screenmedian 9 months [range 6 e 12]) to identify if their adher-ence had altered and to determine if they were suitable for the menu driven intervention phase of the study (Phase 2).Subjects who remained non-adherent were approached toenter phase 2. Inclusion criteria were   18 years of age,attending the Difficult Asthma Service, a Juniper AsthmaControl Score   3 17 and remained non-adherent despiteparticipating in Phase 1. Exclusion criteria were currenttobacco smoking or significant other co-morbidity whichcontributed to persisting respiratory symptoms.The Phase 2 study was a 12 month prospective singleblind randomised controlled trial of an individualisedpsycho-educational nurse-led intervention to improveadherence to medication. Patients were randomly allo-cated to either the intervention or control group. Allparticipants received standard asthma care at the difficultasthma service but the control group did not receive anyfurther intervention. The intervention group subjects wereoffered up to 8 individual intervention visits within a 12week period. Subsequent follow-up visits for control andIntervention group were conducted at 6 months andImproving medication adherence in difficult asthma 1309  12 months post-recruitment. Fig. 1 demonstrates patientflow through the study.The menu driven intervention was designed and deliv-ered by an experienced respiratory nurse with basic levelpsycho-therapy training. Although the proposed interven-tion was unique in its design, to increase the internal val-idity of the study, the Compliance Therapy Model was usedto provide the underpinning theoretical framework. 18 Thismodel encompassed the Transtheoretical Model of Change,Motivational Interviewing and Cognitive BehaviouralTherapy principles, providing a flexible short termintervention using patient’s individual reasons for non-adherence as a guide to plan intervention content. Themodel used a non-confrontational technique which elicitedself-motivation and provided a process to resolve ambiva-lence towards medication taking. Outcome variables The primary outcome for both study phases was change inadherence to ICT. Two indicators of change in adherencewere measured; change in % of ICT prescriptions refilled Figure 1  Patient flow through the study. 1310 J. Gamble et al.  and change in number of participant’s filling   50% ofprescription refills for ICT in each group. Secondaryoutcomes included daily prescribed dose of inhaled corti-costeroid (ICS), courses of rescue oral corticosteroids, totalinhaled and nebulised beta-agonist doses, hospital admis-sions and lung function (spirometry) as well as AsthmaControl Questionnaire (ACQ), 17 Asthma Quality of Life Score(AQLQ) 19 and Hospital Anxiety and Depression Scale/StateTrait Anxiety Scale (HADS). 20 Data analysis Data were coded and analysed using a Statistical Packagefor Social Sciences (SPSS), Version 16. 21 Descriptive statis-tics are shown as mean  SD, or median [range] for nor-mally and non-normally distributed data as appropriate.Between group testing was performed using t tests or MannWhitney test for continuous variables and Chi squared for dichotomous variables (in cases where the expectedfrequency fell below 5 the Fisher’s exact test was used).For Phase 1 outcomes, a one-way person fixed, timefactor ANOVA was used to determine differences acrossBaseline (Screen 1), Screen 2 and Screen 3. When normalityassumptions for ANOVA were not met, a non-parametricFriedman Test was used.For Phase 2 outcomes, a one-way analysis of co-variance(ANCOVA), controlling for pre-test scores was used tocompare the differences between the groups. Whenassumptions were not met, a non-parametric Man-n e Whitney U test for independent samples was used. A pre-planned analysis using student  t  paired  t -test (primaryoutcome) and a one-way person fixed, time factor analysisof variance (ANOVA) (secondary outcomes) to examineintra-group changes was used as the ANCOVA testingdemonstrated an intervention effect. When assumptionswere not met a non-parametric Friedman Test was used.Data are presented as mean  SD or median, range.Significance was taken as  p < 0.05. Results Phase 1 During Phase 1, 31 of 83 (37%) subjects became adherentafter concordance discussion. They were predominantlyfemale ( n Z 21 [68%], age 45  14.6 y); 12 (39%) had beenreferred from another Respiratory physician with apparentsevere refractory asthma.Outcome data for adherence screen 2 (median 9 months,range 6 e 12 after concordance discussion) and adherencescreen 3 (6 months after screen 2), for these 31 subjects areshown in Table 1. At screen 3, 27 of the 31 subjects (90%)remained adherent (still filling  > 50% prescription for ICT)and overall % adherence was also significantly better for thegroup.The prescribed daily dose of ICS (  p < 0.001) and thenumber of rescue courses of oral prednisolone required(  p < 0.001) were significantly improved and there wasa significant decrease in the number of patients requiring atleast one course of rescue prednisolone (  p < 0.001). Intotal, 16 (52%) patients did not require any rescue coursesof prednisolone for at least 12 months post discussion.Improvements were seen in FEV 1  (  p Z 0.01) and FEV 1 %predicted (  p Z 0.05) and there was a statistically signifi-cant reduction in the number of admissions to hospital withan exacerbation of asthma across the 3 time periods(  p Z 0.006) (Fig. 2).Of the remaining 83 subjects identified as non-adherentat Screen 1, 10 failed to re-attend the service after non-adherence was identified and discussed. At Screen 2, 8subjects had low symptom scores despite poor medicationadherence (asthma control score < 3), 13 subjects metexclusion criteria (6 current smokers, 4 significant co-morbidity precluding definite improvement with medicationadherence, severe allergic bronchopulmonary aspergillosis,severe personality disorder, prominent co-existentdysfunctional breathlessness and severe bronchiectasis) Table 1  Adherence and clinical outcomes in the 31 subjects who became adherent after initial concordance discussion.AdherenceScreen 1Adherence Screen 2[median 9 months,range 6 e 12 postinitial screen]AdherenceScreen 3[6 monthsafter Screen 2] F p  Value CI% Adherence to ICT 37.3  14.0 88.5  15.7 82.3  24.3 59.58  < 0.001 16.9 to 28.7ICS daily dose ( m g) b 1616.1  490.0 1293.6  560.5 1280.7  557.6 17.61  < 0.001 a  247.7 to   87.8No. courses rescueprednisolone2.2  1.5 0.6  1.1 0.7  1.0 33.76  < 0.001 a  1.0 to   0.5Maintenance prednisolonedose(mg), ( n Z 34)14.1  6.1 13.1  5.9 11.5  4.7 1.66 0.22   2.2 to 0.5Total  b 2-agonist doses 1705.2  2523.9 1481.9  2072.9 1500.0  2658.8 1.16 0.29   267.7 to 80.6Total  b 2-agonist nebules 221.5  327.9 93.3  140.1 169.3  218.4 1.05 0.32   102.3 to 34.2FEV 1  (l) 2.1   0.7 2.2  0.7 2.2  0.6 6.50 0.01 a 0.0 to 0.2FEV 1  % 74.4   20.5 77.0  21.4 75.5  17.2 4.09 0.05 a 0.0 to 6.8FEV 1  /FVC 68.3  13.1 67.2  15.1 63.2  12.4 0.71 0.40   2.6 to 1.1 Data are presented as mean  SD. a Significance was taken at  p < 0.05. b Bioequivalence of inhaled steroids was fluticasone Z 2  beclomethasone Z 2  budesonide. Improving medication adherence in difficult asthma 1311  and 3 subjects were excluded due to age less than18 years. Phase 2 Of the initial 239 patients at Screen 1, 207 were excluded(Fig. 1), and  n Z 12 refused to take part in the study,resulting in 20 eligible subjects; 9 were randomised to theintervention group and 11 in the control group. Demo-graphic details are shown in Table 2.Two participants randomised to the intervention group,after initial consent to participation, did not attend for the intervention visits and were not included in the followup. All other participants remained in the study for thefull duration. Effects of the intervention at 12 months The primary outcome (%ICT inhalers filled) was significantlyimproved (intervention increased from 37.6% to 61.9%,control group 31.7% to 28.8%,  p Z 0.01, Fig. 3). There wasalso an increase in the number of participants in theintervention group filling > 50% of prescription refills for ICT(intervention group, 4 of 7 filling > 50% v 1 of 10 filling > 50%in the control group Z 0.05). Based on these changes, thenumber needed to treat (NNT) was calculated as follows;90.9% of the control group had no change in adherence at12 months, compared to 42.9% of the intervention group.The difference, the absolute risk ratio (AAR) was 48.5% (95%CI 7.65% to 88.46%) estimating the NNT to be 3 (95% CI 1.1to 13.1).There was a significant difference in daily prescribeddose of maintenance prednisolone in subjects on oralsteroids (intervention group, 4 subjects, mean 9.4 mgs vcontrol group, 3 subjects 20.0 mgs,  p Z 0.001). There wereno other between group differences identified (Table 3).There was a trend towards a decrease in the number ofrescue courses of oral prednisolone required in the inter-vention group, though this did not reach statistical signifi-cance (mean 3.2 mgs to 1.7 mgs,  p Z 0.09). Discussion We have previously reported a high level of poor adherencewith ICT in subjects referred to our Regional DifficultAsthma Service, many of whom had been referred fromother respiratory specialists with apparent ‘severe refrac-tory asthma’ where adherence to medication was notobjectively assessed or addressed. 6 This study examinedwhether identification and addressing poor ICT adherenceusing an objective measure in this population can alter clinical outcome. All patients in this study had received Figure 2  Hospital admissions for the preceding 6 monthperiods at adherence screen 1 (pre-concordance discussion),and at adherence screen 2 (median 9 months, range 6 e 12) andadherence screen 3 (6 months post screen 2) in the 31 subjectswho became adherent after initial concordance discussion. Table 2  Baseline characteristics of the Phase 2 study participants.Demographics GroupsIntervention ( n Z 9) Control ( n Z 11)  p  ValueFemale (%) 7 (78%) 10 (91%) 0.57Age (SD) 50.0 (9.1) 45.2 (10.0) 0.28Duration of asthma (years) (median, range) 38 (9 e 52) 30 (14 e 50) 0.55 Marital status (%) Married 5 (56%) 8 (73%) 0.71Single 1 (11%) 1 (9%)Other 3 (33%) 2 (18%) Employment status (%) Working 3 (33%) 4 (36%) 0.53Retired/pension 1 (11%) 0 (0%)Disability/benefits 5 (56%) 7 (64%) Source of referral (%) Tertiary 5 (56%) 5 (45%) 1.00GP 4 (44%) 6 (55%) Smoking history  Ex (%) 2 (22%) 3 (27%) 1.00Pack year history (Median, range) 27.5 (1 e 35) 20 (1 e 20) 0.26 1312 J. Gamble et al.
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks