Nurse-led medicines’ monitoring in care homes, implementing the Adverse Drug Reaction (ADRe) Profile improvement initiative for mental health medicines: An observational and interview study
aff001; Timothy Banner
aff001; Marie Gabe-Walters
aff003; Jane M. Mikhail
aff001; Gerwyn Panes
aff001; Jeff Round
aff004; Sherrill Snelgrove
aff001; Mel Storey
aff001; David Hughes
Authors place of work:
College of Human and Health Sciences, Swansea University, Swansea, Wales, United Kingdom
aff001; Cardiff and Vale University Health Board, Wales, United Kingdom
aff002; Swansea Bay University Health Board, Wales, United Kingdom
aff003; Institute of Health Economics, Edmonton, Alberta, Canada
aff004; Hywel Dda University Health Board, Wales, United Kingdom
Published in the journal:
PLoS ONE 14(9)
Preventable adverse effects of medicines often pass unnoticed, but lead to real harm.
Nurse-led monitoring using the structured Adverse Drug Reaction (ADRe) Profile identifies and addresses adverse effects of mental health medicines.
This study investigated the implementation and clinical impact of ADRe, and barriers to and facilitators of sustained utilisation in routine practice.
Administration of ADRe was observed for 30 residents prescribed mental health medicines in ten care homes. The study pharmacist reviewed completed ADRes against medication records. Policy context was explored in 30 interviews with service users, nurse managers and strategic leads in Wales.
Residents were aged 60–95, and prescribed 1–17 (median 9 [interquartile range (IQR) 7–13]) medicines. ADRe identified a median of 18 [IQR 11.5–23] problems per resident and nurses made 2 [1–2] changes to care per resident. For example: falls were reported for 9 residents, and care was modified for 5; pain was identified in 8 residents, and alleviated for 7; all 6 residents recognised as dyspnoeic were referred to prescribers. Nurses referred 17 of 30 residents to prescribers. Pharmacists recommended review for all 30. Doubts about administering ADRe, sometimes expressed by people who had not yet used it, diminished as it became familiar. ADRe was needed to bridge communication between resident, nurses and prescribers. When barriers of time, complacency, and doctors’ non-availability were overcome, reporting with ADRe made prescribers more likely to heed nurses’ concerns regarding residents’ welfare. Clinical gains were facilitated by one-to-one time, staff-resident relationships, and unification of documentation.
To our knowledge, ADRe is the only instrument that brings a full account of patients’ problems to medication reviews. This juxtaposition of signs and symptoms against prescriptions facilitates dose adjustments and de-prescribing and leads to: reduced pain and sedation; early identification of problems linked to ADRs, such as falls; and timely medication reviews e.g. for dyspnoea.
People and places – Population groupings – Professions – Medical personnel – Nurses – Pharmacists – Medical doctors – Medicine and health sciences – Health care – Health care providers – Nurses – Medical doctors – Mental health and psychiatry – Pharmacology – Drugs – Antipsychotics – Adverse reactions – Neurology – Sleep disorders – Dyssomnias – Insomnia
Preventable adverse drug reactions and events (ADRs/ ADEs) have proved an intractable problem over the last decade, causing 5–8% of unplanned hospital admissions in the UK [1, 2] rising to ~10–15% amongst older adults [3–5], costing the UK NHS £1bn-£2.5bn each year . ADRs linked to avoidable errors are responsible for 712–22,303 UK deaths each year, costing £98.5 m–£1.6bn . However, higher prevalence in larger prospective studies , and widespread non-recognition  suggest that these figures may be an underestimate. Admissions related to ADRs are associated with 28–32% longer hospital stays  and ADRs/ADEs frequently lead to re-hospitalisation . All-cause mortality is higher amongst older adults prescribed mental health medicines . Demographic change and increased prescribing are likely to exacerbate ADR-related problems, and there is uncertainty over the best strategy for medicines optimisation .
Poor medicines’ management, including errors by patients and professionals, is largely preventable [5, 13] particularly with additional enhanced monitoring [14–19]. The Adverse Drug Reaction (ADRe) Profile  merged from earlier research on ADRs and nurse-led medication checking, and was designed to improve practice in this area, initially in relation to mental health medicines prescribed for care home residents [21, 22].
The ADRe Profile represents a unique approach to collating patient information and testimony to minimise ADRs, optimise prescribing, and prevent medicines-related harm and admissions. It asks nurses or carers to record the signs and symptoms of the undesirable effects of mental health medicines, as listed in manufacturers’ summaries of product characteristics (SmPCs). The supporting information suggests putative aetiologies for each sign or symptom . ADRe is then shared with pharmacists and doctors reviewing medicines administration record (MAR) charts. This paper reports on implementation and operation of ADRe in a sample of Welsh care homes.
The aims of this study were to investigate: 1) implementation of ADRe in care homes for older people, 2) the clinical impact of ADRe, including integration with pharmacists’ medication reviews, and 3) the barriers to and facilitators of sustained utilisation in routine practice.
Ethical approval was conferred on 17th February 2017 from NHS (National Health Service) Wales Research Ethics Committee (REC) 6 (reference no. 16/WA/0358, IRAS ID 213050). The study design and methods were described in the study protocol  and are only summarised here, in accordance with SQUIRE 2.0 and COREQ standards (S1 File).
We estimated that 30 interviews and 30 observations would allow for 4–5 themes at 50% prevalence with 10 instances with 90% power, and would achieve data saturation.
We included residents who were: expected to remain in the home for one year; currently taking any of antipsychotics, antidepressants, anti-epileptics/ mood stabilisers, anxiolytics or hypnotics (benzodiazepines or Z drugs); willing and able to give signed informed consent themselves, or where capacity was lacking, having a consultee willing to give advice. We excluded those not well enough to participate, as screened by their nurses, aged <18 or receiving active palliative care .
Stakeholder interviews were conducted with study participants and key stakeholders able to give a strategic overview of the problems, including strategic leads, and the care home inspectorate .
Volunteer care homes were sought at dissemination events and by emails sent to all 53 eligible care homes in one Welsh health board area. After initial discussion, each care home manager was sent detailed information. Residents were recruited by their nurses based on their prescriptions . Three homes that had participated in the feasibility study , and linked with a participating home, were excluded. Professionals and service users interviewed were nurses, and where possible, residents or service users from the 10 care homes, and stakeholders identified by snowball sampling (Fig 1).
Administration of ADRe was observed in ten homes caring for people prescribed mental health medicines: five from our earlier randomised controlled trial , and five newly recruited. The wider policy context was explored in 30 interviews with service users, nurse managers and strategic leads in Wales.
This mixed-method evaluation integrated data from non-participant observation, nurses’ logs and interviews to explore how, and why ADRe works, and in what contexts [25, 26]. Debriefing and stakeholder interviews explored the problems uncovered, the role of ADRe in identifying them, changes needed to optimise clinical gain and communication between professionals.
Implementation experiences: discrepancies between “as intended” and “as delivered”.
The clinical impact of ADRe was measured via a) problems found b) changes in processes and outcomes of care: additional care delivered, as indicated by nurses, and number and nature of clinical gains, and c) the potential for ADRe to enhance pharmacists’ medication reviews. All signs and symptoms were recorded to avoid outcome selection bias.
The barriers to, and facilitators of utilisation in routine practice were investigated using both observation and interview data. Contextual elements contributing to clinical gain were noted.
Between September 2017 and March 2018, researchers observed administration of ADRe to 30 residents in 10 care homes (SJ, MGW, SS, MS, GP) and interviewed 30 stakeholders (SJ, MGW, DH, JM, SS, MS). Data collection templates and interview questions were based on earlier work  and a pilot feasibility study  and finalised at project meetings (Tables A1 and A2 in S3 File, S2 File). Free-text fieldnotes were recorded. Interviewers were researchers with extensive experience of qualitative interviewing, with one exception, who observed earlier interviews.
We observed nurses’ implementation of the ADRe Profile with 30 residents, alongside standard care, within existing resources, and checked for evidence of adverse drug reactions (listed in the BNF and manufacturers’ literature). The study consultant pharmacist (TB) later reviewed the ADRe profiles alongside a copy of the corresponding MAR charts.
Thirty semi-structured qualitative interviews were conducted with the main stakeholders involved in the administration of medicines in care homes: nurses and care home staff from the 10 homes (12), residents and relatives (5), prescribers (3), dispensing pharmacists (3), independent service users (3), strategic leads: Welsh policy makers (3) and a senior official from the overseeing inspectorate (1). Pharmacists and prescribers were recruited via PRIME centre Wales. Policy makers were recruited from the Welsh Government Health and Social Services Department. Respondents opted to be interviewed in their workplaces, homes or cafés. Interviews were audio-taped, transcribed verbatim and anonymised . The interview schedule is appended (S2 File).
Demographic details, time for implementation, instances of problems identified on ADRe Profiles, and numbers and reasons for referrals were described using measures of central tendency for both normally and non-normally distributed data to facilitate consistency in reporting. Interviews and observation fieldwork were coded, categorised, analysed and closely interpreted by all authors . Compliance and clinical gains were described. Interview analysis was based on the constant comparative method . The two data sets were triangulated to identify common themes as described in the published protocol  (Fig 1). The previous themes of clinical gains, barriers, facilitators and proposals for change formed a provisional template [20–24]. Final codes and themes, as well as data saturation, reflected collective decisions.
The local NHS Research Ethics Committee approved the study on 17th February 2017. Written and verbal information was offered and potential participants. Written informed consent was obtained for all interviews and observations. Where nurses judged that residents lacked capacity to consent, consultees signed on their behalf; consultees were relatives or professionals not involved in the study, who were in regular contact with the participants .
Recruitment and retention
The research design involved data collection in ten care homes, comprising five from the previous trial  (100% retention) and the first five of seven ‘new’ care homes to respond positively to our invitation to 50 eligible care homes (14% response rate) (Fig 1). However, three of these latter five subsequently withdrew. These were replaced with two who had volunteered after we reached our target, plus a home approached on our behalf by a participating manager. Of the three withdrawals, one home closed, and two reported that staff found ADRe too ‘difficult’. This prompted us to interview independent service users .
Nurses were interviewed from all ten homes, and eight nurses returned logs of patient responses. Residents were difficult to recruit to interviews because relatively few were capable of participating, and resident or family interviews were only completed in five homes (six in all). In nine homes, all residents initially approached by their nurses consented to being observed; in the tenth, one resident declined and was replaced by an additional resident from another home. In all homes, we witnessed only care, kindness and compassion. The 30 residents observed are described in Table 1.
1. Implementing ADRe
ADRe administration took 10–50 minutes (Table 2). Nine administrations were interrupted and distractions occurred in a further five, extending the time needed for completion. Recording vital signs took ~7 minutes. Equipment was missing for 20 observations, increasing time taken as working sphygmomanometers or thermometers were sought, sometimes in vain. Two sphygmomanometers failed to give readings from a standing position, and one nurse using previous BP recordings found only sitting BPs recorded. Given the associations between mental health medicines and hypotension, omissions of vital signs detracted from assessments of ADRs, and clinical impact. No laboratory or ECG results were available in homes (Table 3a and 3b); typically, results were logged in GPs’ surgeries. This was concerning for a man prescribed cyproterone (8.2) and more so for a woman prescribed lithium (9.1, age 93), as several problems identified might have been related to lithium (tremor, confusion, ataxia, falls and xerostomia), and perindopril, quetiapine and venlafaxine were co-prescribed.
While most questions were completed in the order on ADRe, vital signs were sometimes postponed. Where these had been recorded in the resident’s notes within the last week, they were often transferred; where vital signs were stable, this was considered acceptable by care staff. Some residents had limited comprehension and speech, and entries were sometimes based on nurses’ interpretations of their health and records, for example, of bowel movements.
2. Clinical impact
a) Problems found
Care staff identified around 17 potential problems per resident when completing ADRe (Table 2), ranging from the need for oral care to extrapyramidal signs to chest pain and oxygen saturation of 89% (Tables A1 and A2 in S3 File). They made around two changes to care per resident, with changes recorded for 27 of 30 residents. In 17 cases prescribers were contacted, mainly for ‘medication review’ to address problems identified, but usually with a specific focus, such as reducing specified mental health medicines, suspected UTI or respiratory problems. Signs and symptoms possibly related to mental health medicines were almost universal. Nearly all residents had been reviewed by dentists (23 of 30) and opticians (29 of 30) in the last year. Problems found and addressed are listed in Tables A1 and A2 in S3 File.
Possible iatrogenic aetiologies of the problems recorded on ADRe were identified by the study pharmacist and checked (Tables 2 and 3). These included up to eight medicines predisposing to falls, seizures likely exacerbated by drug interactions or irregular administration of medicines, diarrhoea where laxatives had been prescribed (for earlier constipation), insomnia where olanzapine was administered too early, at 6.00pm, under-treatment of respiratory conditions, and pruritus due to suboptimal application of prescribed emollients. Continuation of some therapies was queried, for example, statins, bisphosphonates in non-ambulant residents, and medicines for prostatic hypertrophy with catheters in situ. Alcohol use was identified in two residents, possibly accounting for behaviour problems that might otherwise have been attributed to carbamazepine (home 2.1).
b) Processes and outcomes of care
Not all problems identified on ADRe were addressed (Tables A1 and A2 in S3 File), and some were not amenable to medical intervention. Debility, insomnia, cognitive decline and confusion were related to underlying disease, but were often exacerbated by prescribed medicines. Clinical gains ranged from effective analgesia to amelioration of sedation, confusion or hallucinations on discontinuation of antipsychotics (described in Table 4). In six of eight logs returned, nurses reported contacting prescribers to seek reduction in doses of antipsychotics or sedatives. Residents were reported to be ‘brighter’ where the prescriber acquiesced (five homes).
Some homes adopted ADRe more enthusiastically than others. Implementation appeared most successful where benefits were recognised (despite initial scepticism in some instances) and prescribers responded to contacts. Certain barriers to introduction of ADRe were evident (Table 4). Triangulation of cases and nurse feedback via in-depth interviews validated these themes (Table 5; Table B in S3 File).
Vital signs. ADRe introduced recording of oxygen saturation, standing BP, and girth. Hypoxia was recognised as a problem in three of nine residents with hypoxia (Table 4, homes 1,5); otherwise symptoms were attributed to asthma or former smoking or marked as “no problem”. Thirteen residents were assessed for possible postural hypotension: two had a SBP fall of ≥20mmHg, and two of ≥10mmHg (3 of the 4 were prescribed antipsychotics, three antidepressants and two both). Nurses noted the need to recheck, assess risk of falls or advise residents to stand slowly. ADRe identified mental health medicines, anti-hypertensives, medicines for prostatic enlargement, and drug combinations as potential aetiologies and advised on falls (Table A1 in S3 File, Tables 3 and 4, homes 1 & 7).
Falls. ADRe recorded falls for nine of 30 residents (6 of whom were prescribed antipsychotics), ataxia for 14 (nine prescribed antipsychotics), dizziness for seven (five prescribed antipsychotics), and postural hypotension for four (above). Nurses amended care plans for five residents in relation to dizziness, ataxia, postural hypotension or falls (Table 4, homes 1,7), without mentioning any relationship to medicines. The pharmacist recommended medication review to address risks of falling due to factors such as ataxia, dizziness or extra-pyramidal symptoms (EPS) for 14 residents, attributable to antipsychotics (8), AEDs (5), antidepressants (6) anti-hypertensives (3), benzodiazepines or zopiclone (6), and combinations of medicines. Ten of these 14 residents were referred to prescribers by their nurses (Table 4).
Weight. Action regarding weight change was taken for 11 residents, and for current weight for a further two, including introduction of diet diaries for nine (Table 4, homes 3,9). The pharmacist suggested that mirtazapine and valproate might be associated with weight gain, but this was not actioned.
Pain. ADRe identified pain for eight residents, one worsening: seven via the “pain question” and one when checking for non-verbal cues. Nurses acted on seven reports, by administering analgesia, monitoring pain and seeking medical advice (Table 4, homes 3,5,8,9). One resident’s message to prescriber was “pain”.
Dyspnoea. ADRe identified six residents as short of breath, four of whom were prescribed respiratory medicines. The pharmacist recommended review for three of four residents treated for respiratory conditions: two had oxygen saturations ≤96%. Nurses referred all six participants to prescribers after administering ADRe (Tables A1 and A2 in S3 File, homes 1,4,5,6).
Feeling cold. 14 residents either said they were cold or were cold to the touch (some could not speak). Extra blankets or clothes were offered. For two residents, the pharmacist recommended review of beta blockers, but nurses had not made this connection.
Extra-pyramidal symptoms (EPS). ADRe identified problems associated with EPS: 15 instances of restlessness (11 prescribed antipsychotics), 12 of feet shuffling (eight prescribed antipsychotics), nine of abnormal posture (seven prescribed antipsychotics), eight of abnormal gait (six prescribed antipsychotics), five of abnormal tongue movements (all prescribed antipsychotics). Nurses rarely acted on hand tremor (one of nine was referred to the CMHT), and there were no documented responses to tongue movements, abnormal posture or gait. Pharmacist review cited EPS as a reason to review doses of antipsychotics for nine residents, AEDs for another, and drug combinations for another.
c) Pharmacist reviews
By juxtaposing the problems listed on ADRe with the prescriptions issued, around three items per resident were identified for prescription review or additional monitoring. Examples appear in Tables 4 and 5. Reviews were recommended for most mental health medicines (Table 6), for example antipsychotic review was recommended for 17 of 18 residents. Potential problems included safety concerns e.g. falls or risk of falls due to ataxia, abnormal gait or EPS (8), and quality of life issues e.g. insomnia when risperidone was administered nocte rather than mane (3). Nurses referred eight of the 18, and four regimens (all from 1 home) had been changed on initial use of ADRe, before its observed use. AED review was recommended for 11 of 13 residents, to prevent falls (5), or address residents’ concerns e.g. incontinence (1): nurses had referred six. ADRe identified both worsening (resident 6.3) and breakthrough seizures (resident 2.1). Most residents prescribed antidepressants were recommended for review, sometimes to mitigate ADRs from other medicines e.g. tremor, EPS, CNS depression (3) or reduce risks of bleeding and prolonged QTc intervals (2). Benzodiazepines were contributing to several problems. No harm was identified or observed as a result of administering ADRe.
Although several homes acknowledged delays in medication reviews, particularly for GP prescriptions, only one (N10) of the 10 care home managers consulted pharmacists. Some mentioned annual pharmacist reviews of MAR charts. Pharmacist interviewees explained that pharmacists supplying medicines to care homes do not normally conduct medication reviews unless specifically contracted to do so by the health board.
3. ADRe in practice
We identified several barriers to implementation and associated clinical gains, some more surmountable than others. Barriers included time, integration with existing records and information systems, complacency, prescribers’ attitudes, access to doctors, staff education and turnover. Clinical gains from ADRe were facilitated by: allocation of one-to-one time, good staff-resident relationships, unifying pertinent supporting information into a single document, and making staff feel valued.
Time. Time and conflicting work demands were the most frequently cited barriers to regular use of ADRe. “The biggest barrier is time, but having said that, once you’ve used the tool and you’ve used it regularly, it becomes second nature, so it’s not that time-consuming (N10). It’s like everything else really, slow at first and use it regularly and get quicker” (N1). However, where attention to symptoms or de-prescribing made residents more comfortable, the residents were “more settled” (3.2, 4.3), less aggressive and with reduced incontinence (10.2), less confused / sedated (10.1, 10.2, 10.4), warmer (2.3), pain-free (8.1, 9.2, 5.1, 5.2, 5.3), less constipated (6.1) and accordingly easier and more pleasant to work with (Table 4). Time taken to complete ADRe was extended by (unsuccessful) attempts to locate laboratory test results and equipment. Prescribers’ time was also identified as a problem, and ADRe might lead to more efficient use of this scarce resource (SU2).
Integration with existing records. Each care home had its own record system. ADRe unified existing data collection. As one informant said, more work is needed to “integrate into routines, looking at whatever else they are monitoring with a view to keeping it manageable and not distracting them from interaction with residents” (Su2).
Prescribers. Where prescribers engaged well with care home staff and residents, important clinical gains were made (Table 4). In some instances, nurses needed to advocate for residents to effect prescription changes. One nurse who successfully advocated for a reduction in antipsychotic prescribing said: “Pharmacists are keen to come on board, the consultants too. GPs are a difficult group. (…) they always think they know best. (…) we all know that Joe Bloggs doesn’t need the risperidone, but the GP thinks he does, so it’s getting round that sort of thinking. [… There is] resistance from the GPs because they might feel threatened, they think that nurses shouldn’t be having any input–they know best” (N10). Sometimes, prescribers refused to action problems identified: “ADRe picked up issues. Like I say, I spoke to the GP, they weren’t forthcoming in helping: they shut me down” (N5). As a result, chest pain, dyspnoea and insomnia were not addressed, despite requests for contact from residents.
Resignation. While some nurses effectively challenged prescribers, others had become resigned to inertia: “Unlikely the medicines will be changed. I am sure they won’t be changed” (N7). Some were disillusioned after repeated attempts to get medication reviewed: “You speak to the receptionist that you want to speak with the GP with reference bla, bla, bla, you could be waiting two- three days. Then you get frustrated” (N5). Higher-level stakeholders acknowledged that “a lot of people are on a lot of drugs that aren’t reviewed”I1.
Education and staff turnover. Nurses and care staff who participated soon became adept in applying ADRe, and reported that ADRe was easy to use for 27 of 30 residents. Problems with staff shortages and turnover militated against education initiatives, and managers in two homes spoke of problems in ensuring that staff had sufficient information about medicines. ADRe and supporting information attempted to fill this “education gap”, and interviewees described it as an aide memoir that helped collect specific information in a systematic way.
Interpersonal relationships. ADRe depended on positive relationships between residents, carers, nurses and prescribers. Both carers and residents appreciated the one- to-one care, and its contribution to patient-centred care in which residents or families could participate.
All information in one place. Offering prescribers a single document with problems highlighted facilitated inter-professional communication, so that: “everyone is aware of what issues are coming up with the patient, and the decisions can be made between all the healthcare professionals” (P2). One nurse said: “They [doctors, pharmacists] don’t look at the assessments and care plans and things that we do because it’s just mainly for the local health board and for us to have, in order for us to handle the residents better” (N8). ADRe offered a quick and accessible alternative to up to 20 detailed care plans, promoting clinical gains. This was a view echoed by several respondents:
“All the information in one place allows you to focus on the person. (…) I did the profile myself, (…), so I did the vital signs, wrote them all down and I looked through all this to see any of the symptoms, ticked them off. (…) it was their agitation. I discussed with the GP then” (N2).
“All the information on one piece of paper, (…), rather than going through separate pieces in separate places. (…) If the doctor comes in and says “Have you done the observations?” it’s all together in one place and you can give the details to the GP” (N9).
Recognising problems. Some nurses acknowledged that their familiarity with residents might blind them to problems: “We get used to the person. We might omit some signs (…) when you look over the profile it will ask you different questions that maybe you never thought of: someone being on a different kind of tablet might give them tremor. You might see that but might classify that as age. You might not actually put it as medication. So this is why the profile is good” (N8). ADRe also crystallised nurses’ clinical suspicions that medicines were not optimised, despite correct doses: “If we have got the evidence, they [doctors] are much more inclined to take our word for it, listen to our point of view” (N3). However, ADRe was seen as complex, and in the view of a few, over-complex. The comprehensiveness of the instrument was both a weakness, in terms of being too difficult for some staff, and a strength, in that “[It] puts little things together that may be missed” (N3). ADRe might allow insidious, subtle ADRs to be detected and managed, and certain hard-to-diagnose conditions (hypothyroidism, pernicious anaemia) to be recognised and treated. When compared with easier tools, ADRe “covers everything, but becomes lengthier. You won’t miss anything here, but administering this would take more time and resources” (GP2).
Valuing staff. ADRe works by harnessing care staff’s familiarity with patients to inform doctors of ongoing undesirable effects of medicines prescribed. The supporting information provided with the ADRe Profile suggests which medicines might be responsible, and this guidance can be used to support advocacy for review and revision: “I made notes on the profile, to lead me in the right direction. Rather than just trying to put my opinion across, I had the evidence in front of me: look this is her BP, her falls have increased–four times more. Having the supporting information to say ‘it might be this medication, so can you come and review?’” (N6.2). “What it [ADRe] tends to do, is give value doesn’t it? The staff tend to feel the information that they are gathering is of value. It improves morale. The care staff–the staff feel that what they are doing is worthwhile and they are not just ticking forms, ticking boxes and just getting filed away” (N3).
Changing priorities: Shifting the zeitgeist
The prevalence, severity and reversibility of problems identified by ADRe indicate that change is both necessary and feasible. Respondents suggested that reducing the variation between homes, services and prescribers will reduce ADRs and ADEs, and the burdens these place on service users, staff and the NHS (Table C in S3 File). The service users were the group most dissatisfied with current practice, and the most enthusiastic advocates for ADRe, based on their personal experiences and perceived need for change in mental health medication management. Their perceptions are supported by the study pharmacist’s reviews (Table 3) and the changes brought about by implementation of ADRe (Table 4).
Changes that would help protect patients from harm associated with ADRs/ADEs were identified. These included shared access to essential information, improved staff education, adequate staffing, communication with GPs, consultants and pharmacists, and prescription reviews before repeats are issued (Table C in S3 File). ADRe made a contribution in several of these areas, providing standardised patient data, informal, experiential education, a mechanism for inter-professional communication and observations that feed into prescription reviews.
All nurses and stakeholders agreed that the paper version of ADRe needs to be developed to become compatible with digital information systems: “People won’t write it twice. (…) The questions that you have here, they are good (…) the only problem is we already have a system in place.” N8 (Table C in S3 File). Integrating ADRe with electronic patient records would allow a number of tangible improvements, such as cross-population of the ADRe Profile from other records (obviating the need for any duplication of recording), selective retrieval of relevant information (e.g. positive responses only), examination of trends over time, and an online link to interaction checkers. Developing an electronic version of ADRe appears a more useful path for development than any effort to further shorten the instrument. The latter would risk detracting from “the little things” that made ADRe comprehensive: “I don’t really know if it could be condensed” (N3).
Patient-centred medication review is essential. ADRe’s evolving evidence base indicates it can be effective and acceptable in supporting reviews before repeat prescribing [22–24]. This study indicates that ADRe brings several benefits: it relieves residents’ ADR burden of confusion, sedation and EPS; alleviates falls, pain, and dyspnoea; optimises seizure control; and bridges communication between residents, nurses and prescribers. However, to overcome the barriers of time, education and staff turnover (only some homes), prescribers’ availability and complacency, introducing ADRe into routine practice throughout the sector will need support from regulators. Systematic, routine ADR monitoring needs to be integrated into existing care and documentation. Where prescribers visited regularly, this was relatively straightforward; however, where doctors were uncontactable, it was difficult.
ADRs: An everyday problem in need of policy intervention
Meta-analyses identified little evidence for clinical gains from interventions to optimise prescribing [31–33] or polypharmacy , or physical health, (except individualised exercise interventions) . While many interventions [36–40] improve medication appropriateness, links to patient outcomes are unclear [12, 31, 41, 42] ADRe is different - it addresses the “little” things affecting patients, as a preventive strategy, for example, by addressing dizziness, balance and eyesight before falls happen (homes 1,2,6,7, Table 4) [43, 44].
ADRs and ADEs are common, though often unrecognised , and systematic checking with ADRe increased recognition, reporting and actions [20, 24] Preventable problems were almost universal, but often mundane, commonplace, un-recorded, and, consequently, overlooked. ADRe helped to break the Inverse Interest Law: the commoner the problem, the less the professional interest . Without ADRe, clinically uninteresting, yet debilitating, conditions, such as incontinence, confusion, xerostomia, hyper-salivation, and insomnia were often under-investigated (all homes Table 4).
The differences in medical care available to residents in different homes, particularly access to consultants , reflected the association between socio-economic deprivation and GP scarcity , and the Inverse Care Law . Without change in policy, homes in the poorest areas will be the least able to adopt ADRe, because they cannot contact prescribers to address problems identified (homes 5 & 7, Table 5), and the most economically-deprived will continue to be disproportionately represented amongst preventable unplanned hospital admissions .
Economics of ADRe
Many ADRs and ADEs are preventable, and there can be significant resource and cost implications for health systems in failing to tackle the problem. Patients experiencing avoidable admissions and extended hospitalisation owing to avoidable ADR/ADE incur excess treatment costs of between €2,851 (£2,463) and €9,015 (£7,787) per patient . The application of simple and inexpensive interventions such as ADRe within routine care has the potential to minimise preventable ADRs/ADEs. While the cost-effectiveness of any intervention must be demonstrated based on robust evidence of clinical efficacy and effectiveness, the median time required to complete ADRe was under 30 minutes, and in no cases exceeded an hour, suggesting that staff costs will be minimal. While it is difficult to accurately estimate wages for care home staff, a recently advertised post on www.carehome.co.uk, for a Care Home Clinical Lead Registered Nurse paid £17.50 per hour. With salary on-costs (in the UK, typically for pension contributions and employer national insurance contributions) of 15%, the cost per hour of labour is roughly £20.00 per hour. A crude calculation suggests that, at the lower bound of the cost of an ADR/ADE of about £2,400, one admission must be prevented for every 120 administrations of ADRe. Given that from 30 administrations, there were 17 referrrals, 5 interventions for falls, 1 for chest pain, and 6 for dyspnoea, it is likely that this threshold was achieved. This is a crude estimate based on the limited data available on both the costs of administering ADRe and the costs of ADRs for people living in care homes. However, it usefully illustrates that a low-tech intervention has the potential to provide benefit at health system level as well as patient level. The two other interventions identified in a systematic review of trials of medicines management interventions for people with dementia either depended on consultant input (increasing costs) or focused on medication adherence .
From paper to electronic health records (EHRs)
Both time and EHRs were identified as barriers to implementation of ADRe, and these are inter-related. EHRs can identify severe events , but have increased NHS workloads , with electronic medicines optimisation a particular challenge . Where clinical decision support is poor, electronic prescribing increases errors , and ADRe’s supporting information helps here. EHRs consume ~6 hours /day of responding doctors’ time and are associated with staff burnout in primary and secondary care . Electronic patient self-reporting of adverse events is effective in oncology  but most residents were too poorly to self-report. Although integration into EHRs overcomes some barriers (i.e. fitting into existing documentation, forcing responses to all items), it will not address others i.e. time, prescriber contact, staff turnover and overwhelming systems (Table 5). Introducing ADRe against this background of perceived stress and overwork is challenging. However, the initial time investment to complete ADRe in face-to-face interviews is repaid when residents feel they have been heard and are less confused and pain-free (Fig 2).
ADRe focuses multidisciplinary team communication
When medicines reviews are scheduled, giving pharmacists and prescribers a copy of ADRe alongside the MAR chart links residents’ problems to the clinical goal of freedom from ADRs (Fig 3) [58,59]. When reviewing medicines, pharmacists and doctors rarely have time to interview patients about every possible adverse effect of all their medicines, many residents are non-verbal, and the resident’s key carer is often off-duty, due to 12 hour shift patterns. For example, laxatives are commonly prescribed, but ADRe was needed to tell the pharmacist that the resident had diarrhoea, prompting adjusted prescribing (4 participants, Table 3). Some ~50% of ADRe’s signs and symptoms can be retrieved from patients’ notes, but this takes 1–1.5 hours [23, 24]. Medication reviews are labour-intensive , and under-resourced ; ADRe reduces overall costs by optimising pharmacists’ and prescribers’ time.
Strengths and limitations
Linking problems with prescriptions via systematic nurse-led monitoring is a unique approach to a large problem that is threatening to overwhelm care delivery [21, 22], including hospital admissions [1–6]. Juxtaposition of prescriptions and problems clarified mechanisms for clinical improvements. Without this detail, the complexities of the cases would have been overlooked, problems would have remained unattended, and residents left in pain or falling. However, some ADRe items were omitted, particularly time-consuming vital signs, or unaddressed.
We acknowledge that this study is limited by selection bias, selection by prescriptions rather than diagnostic codes, size, and self-reporting in interviews. Although we focussed on objective clinical findings, no researchers can discount entrapment by prior expectation  or subconscious outcome reporting bias favouring benefits and discounting harm caused by the intervention. There is a risk that researcher expectancy bias will lead the study to exaggerate ADRe’s ability to detect problems and underplay implementation difficulties. However, we sought to ensure that problems identified were real by observation and using care homes’ own logs as evidence of adverse reactions or risks identified, and getting these verified by an independent pharmacist recruited to the team. We assessed ADRe’s “usability” from observations and staff reports in interviews, which were coded by multiple team members who were well aware that it would be counterproductive to exaggerate acceptability. Moreover, reporting was unaffected by the more pervasive and serious “reciprocation” or commercial interest biases .
Concerns over potential volunteer bias in care home selection emerged when 3 homes withdrew, citing difficulties in maintaining effective service delivery affecting their ability to participate: others note that only interested, “good” homes, with open-door policies welcome researchers [24, 63]. The observation that some of our homes were in economically deprived areas, with poor GP contact , increases the transferability of the findings, but we acknowledge generalisation of findings is based on logical, rather than statistical, inference  and can best be confirmed by replication studies of additional settings . However, these findings show how residents can be helped, without sedation or exposure to risk from prescribed medicines: the challenges are to persuade everyone to emulate the best, to show that ADRe is easy to use and, with experience, the initial effort is more than repaid by clinical gain (Fig 2).
None of the 34 methods for ADR causality assessment is universally accepted  or easy to apply in practice . This study identified putative ADRs, which were sometimes confirmed by de-prescribing. It was beyond the study’s scope to identify errors or to systematically de-challenge and re-challenge participants with suspected medicines to determine which, if any, of the prescribed medicines was responsible in each case. Re-challenge is rarely practical, and formal assessments of ADR causality cannot confer certainty, prove a connection or quantify a contribution ; they rarely achieve consensus .
The study’s limited duration, resources, and size precluded detailed follow up of all participants, and uncommon events. The 30 detailed cases indicated data saturation, and triangulation of stakeholders’ views with cases (Table 5) demonstrated how ADRe can and should work.
Implications for practice: The case for change
ADRe is needed to minimise pain, dyspnoea, sedation, confusion, insomnia, EPS, falls, incontinence, and aggression. The problems addressed here permeate health and social care services [2, 8, 12]. The change in thinking to reflectively link problems with known ADRs is a barrier to be negotiated, but the pervasiveness of unmonitored harm and entrapment by prior expectation are profound challenges [2, 8, 12, 69]. Adjusting healthcare systems to check systematically for iatrogenic harm may be difficult until the benefits of minimising medicines-related harm are understood, the process is normalised  and gatekeepers accept that patients need strategies to communicate their signs, symptoms and concerns .
Demographic and prescribing changes are increasing the need for a comprehensive, systematic multi-professional approach to medicines optimisation. In the UK in 2008–11, 49% (of 7359) of people aged 65 and over took more than four medicines, a fourfold increase since 1991–5 , whilst emergency admissions for ADRs rose . Life expectancy is no longer rising .
The WHO asks health ministries to programme changes in professionals’ behaviour, systems and practices of medication management to reduce ADRs, polypharmacy and miscommunication [75, 76]. Workforce constraints suggest this may be best achieved by extending and expanding nurses’ roles, building on nurses’ knowledge and communication skills. ADRe can do this and address long-standing problems. Presenting evidence to gatekeepers is insufficient  change is mediated through social networks, power structures, and professional politics. The challenge is to persuade policy makers that a person-centred , clinically and cost-effective tool, such as ADRe,  deserves support.
Note: In the Welsh language “Adre” means homewards. We hope the ADRe profile will bring a sense of greater security for those prescribed mental health medicines.
S1 File [docx]
Revised Standards for Quality Improvement Reporting Excellence SQUIRE2.0 plus Consolidated criteria for reporting qualitative studies (COREQ).
1. Pirmohamed M, James S, Meakin S, Green C, Scott AK, Walley TJ, et al. Adverse drug reactions as cause of admission to hospital: prospective analysis of 18 820 patients. BMJ. 2004;329(7456):15–9. Epub 2004/07/03. doi: 10.1136/bmj.329.7456.15 15231615.
2. National Institute for Health and Care Excellence. Medicines optimisation:the safe and effective use of medicines to enable the best possible outcomes. NICE guideline 5 2015 [Accessed 07 April 2019]. https://www.nice.org.uk/guidance/ng5/evidence/full-guideline-6775454.
3. Kongkaew C, Noyce PR, Ashcroft DM. Hospital admissions associated with adverse drug reactions: a systematic review of prospective observational studies. Ann Pharmacother. 2008;42(7):1017–25. Epub 2008/07/03. doi: 10.1345/aph.1L037 18594048.
4. Oscanoa TJ, Lizaraso F, Carvajal A. Hospital admissions due to adverse drug reactions in the elderly. A meta-analysis. Eur J Clin Pharmacol. 2017;73(6):759–70. Epub 2017/03/03. doi: 10.1007/s00228-017-2225-3 28251277.
5. Parameswaran Nair N, Chalmers L, Connolly M, Bereznicki BJ, Peterson GM, Curtain C, et al. Prediction of Hospitalization due to Adverse Drug Reactions in Elderly Community-Dwelling Patients (The PADR-EC Score). PLoS One. 2016;11(10):e0165757. Epub 2016/11/01. doi: 10.1371/journal.pone.0165757 27798708.
6. Frontier Economic. Exploring the costs of unsafe care in the NHS: a report prepared for The Department of Health 2014 [Accessed 07 April 2019]. https://www.frontier-economics.com/media/2459/exploring-the-costs-of-unsafe-care-in-the-nhs-frontier-report-2-2-2-2.pdf.
7. Alhawassi TM, Krass I, Bajorek BV, Pont LG. A systematic review of the prevalence and risk factors for adverse drug reactions in the elderly in the acute care setting. Clin Interv Aging. 2014;9:2079–86. Epub 2014/12/10. doi: 10.2147/CIA.S71178 25489239.
8. Roulet L, Ballereau F, Hardouin JB, Chiffoleau A, Potel G, Asseray N. Adverse drug event nonrecognition in emergency departments: an exploratory study on factors related to patients and drugs. J Emerg Med. 2014;46(6):857–64. Epub 2014/02/26. doi: 10.1016/j.jemermed.2013.11.124 24565882.
9. Walter SR, Day RO, Gallego B, Westbrook JI. The impact of serious adverse drug reactions: a population-based study of a decade of hospital admissions in New South Wales, Australia. British journal of clinical pharmacology. 2017;83(2):416–26. Epub 2016/09/11. doi: 10.1111/bcp.13124 27614089.
10. El Morabet N, Uitvlugt EB, van den Bemt BJF, van den Bemt PMLA, Janssen MJA, Karapinar-Çarkit F. Prevalence and Preventability of Drug-Related Hospital Readmissions: A Systematic Review. Journal of the American Geriatrics Society. 2018;66(3):602–8. doi: 10.1111/jgs.15244 29468640
11. Du Y, Wolf IK, Busch MA, Knopf H. Associations between the use of specific psychotropic drugs and all-cause mortality among older adults in Germany: Results of the mortality follow-up of the German National Health Interview and Examination Survey 1998. PLoS One. 2019;14(1):e0210695. Epub 2019/01/15. doi: 10.1371/journal.pone.0210695 30640945.
12. Avery AJ, Bell BG. Rationalising medications through deprescribing. BMJ. 2019;364:l570. Epub 2019/02/09. doi: 10.1136/bmj.l570 30733211.
13. Hakkarainen KM, Andersson Sundell K, Petzold M, Hagg S. Prevalence and perceived preventability of self-reported adverse drug events—a population-based survey of 7099 adults. PLoS One. 2013;8(9):e73166. Epub 2013/09/12. doi: 10.1371/journal.pone.0073166 24023828.
14. Gabe ME, Davies GA, Murphy F, Davies M, Johnstone L, Jordan S. Adverse drug reactions: treatment burdens and nurse-led medication monitoring. Journal of Nursing Management. 2011;19(3):377–92. doi: 10.1111/j.1365-2834.2011.01204.x 21507109
15. Brenner S, Detz A, Lopez A, Horton C, Sarkar U. Signal and noise: applying a laboratory trigger tool to identify adverse drug events among primary care patients. BMJ Qual Saf. 2012;21(8):670–5. Epub 2012/05/26. doi: 10.1136/bmjqs-2011-000643 22626736.
16. Forster AJ, Murff HJ, Peterson JF, Gandhi TK, Bates DW. Adverse drug events occurring following hospital discharge. J Gen Intern Med. 2005;20(4):317–23. Epub 2005/04/29. doi: 10.1111/j.1525-1497.2005.30390.x 15857487.
17. Gurwitz JH, Field TS, Judge J, Rochon P, Harrold LR, Cadoret C, et al. The incidence of adverse drug events in two large academic long-term care facilities. Am J Med. 2005;118(3):251–8. Epub 2005/03/05. doi: 10.1016/j.amjmed.2004.09.018 15745723.
18. Steinman MA, Handler SM, Gurwitz JH, Schiff GD, Covinsky KE. Beyond the prescription: medication monitoring and adverse drug events in older adults. J Am Geriatr Soc. 2011;59(8):1513–20. Epub 2011/07/30. doi: 10.1111/j.1532-5415.2011.03500.x 21797831.
19. Gandhi TK, Seger AC, Overhage JM, Murray MD, Hope C, Fiskio J, et al. Outpatient adverse drug events identified by screening electronic health records. J Patient Saf. 2010;6(2):91–6. Epub 2010/06/01. doi: 10.1097/PTS.0b013e3181dcae06 22130350.
20. Medicines Management G. ADRE—THE ADVERSE DRUG REACTION PROFILE Swansea: Swansea University; [Accessed 7 April 2019]. http://www.swansea.ac.uk/adre/
21. Jordan S, Banner T, Gabe-Walters M, Mikhail JM, Round J, Snelgrove S, et al. Nurse-led medicines’ monitoring in care homes study protocol: a process evaluation of the impact and sustainability of the adverse drug reaction (ADRe) profile for mental health medicines. BMJ Open. 2018;8(9):e023377. Epub 2018/10/01. doi: 10.1136/bmjopen-2018-023377 30269073.
22. Jordan S, Logan PA, Panes G, Vaismoradi M, Hughes D. Adverse Drug Reactions, Power, Harm Reduction, Regulation and the ADRe Profiles. Pharmacy (Basel). 2018;6(3). Epub 2018/09/21. doi: 10.3390/pharmacy6030102 30231573.
23. Jordan S, Gabe M, Newson L, Snelgrove S, Panes G, Picek A, et al. Medication monitoring for people with dementia in care homes: the feasibility and clinical impact of nurse-led monitoring. Scientific World Journal. 2014:843621. Epub 2014/04/08. doi: 10.1155/2014/843621 24707218.
24. Jordan S, Gabe-Walters ME, Watkins A, Humphreys I, Newson L, Snelgrove S, et al. Nurse-Led Medicines’ Monitoring for Patients with Dementia in Care Homes: A Pragmatic Cohort Stepped Wedge Cluster Randomised Trial. PLoS One. 2015;10(10):e0140203. Epub 2015/10/16. doi: 10.1371/journal.pone.0140203 26461064.
25. Moore GF, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, et al. Process evaluation of complex interventions: Medical Research Council guidance. Bmj. 2015;350:h1258. Epub 2015/03/21. doi: 10.1136/bmj.h1258 25791983.
26. Willis CD, Saul J, Bevan H, Scheirer MA, Best A, Greenhalgh T, et al. Sustaining organizational culture change in health systems. J Health Organ Manag. 2016;30(1):2–30. Epub 2016/03/12. doi: 10.1108/JHOM-07-2014-0117 26964847.
27. Information Commissioners Office. Anonymisation: managing data protection risk code of practice Cheshire: ICO; 2012 [Accessed 07 April 2019]. https://ico.org.uk/media/1061/anonymisation-code.pdf.
28. Glaser B, Strauss A. The Discovery of Grounded Theory. Chicago: Aldine; 1967.
29. Siafis S, Papazisis G. Detecting a potential safety signal of antidepressants and type 2 diabetes: a pharmacovigilance-pharmacodynamic study. British journal of clinical pharmacology. 2018;84(10):2405–14. Epub 2018/06/29. doi: 10.1111/bcp.13699 29953643.
30. Ble A, Hughes PM, Delgado J, Masoli JA, Bowman K, Zirk-Sadowski J, et al. Safety and Effectiveness of Statins for Prevention of Recurrent Myocardial Infarction in 12 156 Typical Older Patients: A Quasi-Experimental Study. J Gerontol A Biol Sci Med Sci. 2017;72(2):243–50. Epub 2016/05/06. doi: 10.1093/gerona/glw082 27146371.
31. Alldred DP, Kennedy MC, Hughes C, Chen TF, Miller P. Interventions to optimise prescribing for older people in care homes. The Cochrane database of systematic reviews. 2016;2:CD009095. Epub 2016/02/13. doi: 10.1002/14651858.CD009095.pub3 26866421.
32. Page AT, Clifford RM, Potter K, Schwartz D, Etherton-Beer CD. The feasibility and effect of deprescribing in older adults on mortality and health: a systematic review and meta-analysis. British journal of clinical pharmacology. 2016;82(3):583–623. Epub 2016/04/15. doi: 10.1111/bcp.12975 27077231.
33. Johansson T, Abuzahra ME, Keller S, Mann E, Faller B, Sommerauer C, et al. Impact of strategies to reduce polypharmacy on clinically relevant endpoints: a systematic review and meta-analysis. British journal of clinical pharmacology. 2016;82(2):532–48. Epub 2016/04/10. doi: 10.1111/bcp.12959 27059768.
34. Rankin A, Cadogan CA, Patterson SM, Kerse N, Cardwell CR, Bradley MC, et al. Interventions to improve the appropriate use of polypharmacy for older people. The Cochrane database of systematic reviews. 2018;9:CD008165. Epub 2018/09/04. doi: 10.1002/14651858.CD008165.pub4 30175841.
35. Vancampfort D, Firth J, Correll CU, Solmi M, Siskind D, De Hert M, et al. The impact of pharmacological and non-pharmacological interventions to improve physical health outcomes in people with schizophrenia: a meta-review of meta-analyses of randomized controlled trials. World Psychiatry. 2019;18(1):53–66. Epub 2019/01/03. doi: 10.1002/wps.20614 30600626.
36. Ballard C, Orrell M, Sun Y, Moniz-Cook E, Stafford J, Whitaker R, et al. Impact of antipsychotic review and non-pharmacological intervention on health-related quality of life in people with dementia living in care homes: WHELD-a factorial cluster randomised controlled trial. Int J Geriatr Psychiatry. 2016. Epub 2016/09/20. doi: 10.1002/gps.4572 27640872.
37. Desveaux L, Saragosa M, Rogers J, Bevan L, Loshak H, Moser A, et al. Improving the appropriateness of antipsychotic prescribing in nursing homes: a mixed-methods process evaluation of an academic detailing intervention. Implement Sci. 2017;12(1):71. Epub 2017/05/28. doi: 10.1186/s13012-017-0602-z 28549480.
38. Mestres Gonzalvo C, Milosevic V, van Oijen BPC, de Wit H, Hurkens K, Mulder WJ, et al. The use of an electronic clinical rule to discontinue chronically used benzodiazepines and related Z drugs. Eur J Clin Pharmacol. 2018;74(2):227–31. Epub 2017/11/12. doi: 10.1007/s00228-017-2369-1 29127459.
39. Westbury JL, Gee P, Ling T, Brown DT, Franks KH, Bindoff I, et al. RedUSe: reducing antipsychotic and benzodiazepine prescribing in residential aged care facilities. Med J Aust. 2018;208(9):398–403. Epub 2018/05/12. 29747564.
40. Kua CH, Mak VSL, Huey Lee SW. Health Outcomes of Deprescribing Interventions Among Older Residents in Nursing Homes: A Systematic Review and Meta-analysis. J Am Med Dir Assoc. 2019;20(3):362–72 e11. Epub 2018/12/26. doi: 10.1016/j.jamda.2018.10.026 30581126.
41. Sino CG, Bouvy ML, Jansen PA, Schop IM, Egberts TC, Schuurmans MJ. Signs and symptoms indicative of potential adverse drug reactions in homecare patients. J Am Med Dir Assoc. 2013;14(12):920–5. Epub 2013/11/30. doi: 10.1016/j.jamda.2013.09.014 24286711.
42. Shafiee Hanjani L, Long D, Peel NM, Peeters G, Freeman CR, Hubbard RE. Interventions to Optimise Prescribing in Older People with Dementia: A Systematic Review. Drugs Aging. 2019;36(3):247–67. Epub 2018/12/20. doi: 10.1007/s40266-018-0620-9 30565157.
43. Pevnick JM, Shane R, Schnipper JL. The problem with medication reconciliation. BMJ Qual Saf. 2016;25(9):726–30. Epub 2016/01/23. doi: 10.1136/bmjqs-2015-004734 26795914.
44. Jennings E, Gallagher P, O’Mahony D. Detection and prevention of adverse drug reactions in multi-morbid older patients. Age Ageing. 2019;48(1):10–3. Epub 2018/10/10. doi: 10.1093/ageing/afy157 30299453.
45. Lewis G, Wessely S. The epidemiology of fatigue: more questions than answers. Journal of epidemiology and community health. 1992;46(2):92–7. doi: 10.1136/jech.46.2.92 1583440.
46. Goodman C, Davies SL, Gordon AL, Dening T, Gage H, Meyer J, et al. Optimal NHS service delivery to care homes: a realist evaluation of the features and mechanisms that support effective working for the continuing care of older people in residential settings. Health Services and Delivery Research. Southampton (UK) 2017.
47. Drennan VM, Grant RL, Harris R. Trends over time in prescribing by English primary care nurses: a secondary analysis of a national prescription database. BMC Health Serv Res. 2014;14:54. Epub 2014/02/07. doi: 10.1186/1472-6963-14-54 24499423.
48. Hart JT. The inverse care law. Lancet (London, England). 1971;1(7696):405–12. Epub 1971/02/27. doi: 10.1016/s0140-6736(71)92410-x 4100731.
49. Payne RA, Abel GA, Guthrie B, Mercer SW. The effect of physical multimorbidity, mental health conditions and socioeconomic deprivation on unplanned admissions to hospital: a retrospective cohort study. CMAJ. 2013;185(5):E221–8. Epub 2013/02/21. doi: 10.1503/cmaj.121349 23422444.
50. Formica D, Sultana J, Cutroneo PM, Lucchesi S, Angelica R, Crisafulli S, et al. The economic burden of preventable adverse drug reactions: a systematic review of observational studies. Expert Opin Drug Saf. 2018;17(7):681–95. Epub 2018/06/29. doi: 10.1080/14740338.2018.1491547 29952667.
51. McGrattan M, Ryan C, Barry HE, Hughes CM. Interventions to Improve Medicines Management for People with Dementia: A Systematic Review. Drugs Aging. 2017;34(12):907–16. Epub 2017/11/28. doi: 10.1007/s40266-017-0505-3 29177770.
52. Sheikh A, Cornford T, Barber N, Avery A, Takian A, Lichtner V, et al. Implementation and adoption of nationwide electronic health records in secondary care in England: final qualitative results from prospective national evaluation in "early adopter" hospitals. BMJ. 2011;343:d6054. Epub 2011/10/19. doi: 10.1136/bmj.d6054 22006942.
53. Jeffries M, Phipps D, Howard RL, Avery A, Rodgers S, Ashcroft D. Understanding the implementation and adoption of an information technology intervention to support medicine optimisation in primary care: qualitative study using strong structuration theory. BMJ Open. 2017;7(5):e014810. Epub 2017/05/13. doi: 10.1136/bmjopen-2016-014810 28495815.
54. Lyell D, Magrabi F, Raban MZ, Pont LG, Baysari MT, Day RO, et al. Automation bias in electronic prescribing. BMC Med Inform Decis Mak. 2017;17(1):28. Epub 2017/03/18. doi: 10.1186/s12911-017-0425-5 28302112.
55. Arndt BG, Beasley JW, Watkinson MD, Temte JL, Tuan WJ, Sinsky CA, et al. Tethered to the EHR: Primary Care Physician Workload Assessment Using EHR Event Log Data and Time-Motion Observations. Ann Fam Med. 2017;15(5):419–26. Epub 2017/09/13. doi: 10.1370/afm.2121 28893811.
56. Shanafelt TD, Dyrbye LN, Sinsky C, Hasan O, Satele D, Sloan J, et al. Relationship Between Clerical Burden and Characteristics of the Electronic Environment With Physician Burnout and Professional Satisfaction. Mayo Clin Proc. 2016;91(7):836–48. Epub 2016/06/18. doi: 10.1016/j.mayocp.2016.05.007 27313121.
57. Warrington L, Absolom K, Conner M, Kellar I, Clayton B, Ayres M, et al. Electronic Systems for Patients to Report and Manage Side Effects of Cancer Treatment: Systematic Review. J Med Internet Res. 2019;21(1):e10875. Epub 2019/01/27. doi: 10.2196/10875 30679145.
58. Jordan S, Kyriacos U. Medicines’ management: a public health problem on nursing’s agenda. J Nurs Manag. 2014;22(3):271–5. Epub 2014/04/24. doi: 10.1111/jonm.12238 24754749.
59. Keijsers CJ, Leendertse AJ, Faber A, Brouwers JR, de Wildt DJ, Jansen PA. Pharmacists’ and general practitioners’ pharmacology knowledge and pharmacotherapy skills. J Clin Pharmacol. 2015;55(8):936–43. Epub 2015/03/27. doi: 10.1002/jcph.500 25810359.
60. Schnipper JL, Mixon A, Stein J, Wetterneck TB, Kaboli PJ, Mueller S, et al. Effects of a multifaceted medication reconciliation quality improvement intervention on patient safety: final results of the MARQUIS study. BMJ Qual Saf. 2018;27(12):954–64. Epub 2018/08/22. doi: 10.1136/bmjqs-2018-008233 30126891.
61. Sackett DL HR, Guyatt GH, Tugwell P. Clinical Epidemiology: A Basic Science for Clinical Medicine, Second edition. Boston MA: Little, Brown & Co; 1991.
62. Vaganay A. Outcome Reporting Bias in Government-Sponsored Policy Evaluations: A Qualitative Content Analysis of 13 Studies. PLoS One. 2016;11(9):e0163702. Epub 2016/10/01. doi: 10.1371/journal.pone.0163702 27690131.
63. Ellwood A, Airlie J, Cicero R, Cundill B, Ellard DR, Farrin A, et al. Recruiting care homes to a randomised controlled trial. Trials. 2018;19(1):535. Epub 2018/10/05. doi: 10.1186/s13063-018-2915-x 30285850.
64. Mitchell JC. Case and Situation Analysis. The Sociological Review. 1983;31(2):187–211. doi: 10.1111/j.1467-954X.1983.tb00387.x
65. Agbabiaka TB, Savovic J, Ernst E. Methods for causality assessment of adverse drug reactions: a systematic review. Drug Saf. 2008;31(1):21–37. Epub 2007/12/22. doi: 10.2165/00002018-200831010-00003 18095744.
66. Lavan AH, Gallagher P. Predicting risk of adverse drug reactions in older adults. Ther Adv Drug Saf. 2016;7(1):11–22. Epub 2016/02/03. doi: 10.1177/2042098615615472 26834959.
67. Centre UM. The use of the WHO_UMC system for standardised case causality assessment Sweden: UMC; 2018 [Accessed 5 March 2019]. https://www.who.int/medicines/areas/quality_safety/safety_efficacy/WHOcausality_assessment.pdf.
68. Mouton JP, Mehta U, Rossiter DP, Maartens G, Cohen K. Interrater agreement of two adverse drug reaction causality assessment methods: A randomised comparison of the Liverpool Adverse Drug Reaction Causality Assessment Tool and the World Health Organization-Uppsala Monitoring Centre system. PLoS One. 2017;12(2):e0172830. Epub 2017/02/25. doi: 10.1371/journal.pone.0172830 28235001.
69. Willson A. The problem with eliminating ‘low-value care’. BMJ Qual Saf. 2015;24(10):611–4. Epub 2015/07/15. doi: 10.1136/bmjqs-2015-004518 26152137.
70. May CR, Mair F, Finch T, MacFarlane A, Dowrick C, Treweek S, et al. Development of a theory of implementation and integration: Normalization Process Theory. Implement Sci. 2009;4:29. Epub 2009/05/23. doi: 10.1186/1748-5908-4-29 19460163.
71. Robinson J. Research for whom? The politics of research dissemination and application. In: Buckeldee J and McMahon R, editor. The Research Experience in Nursing London: Chapman Hall; 1994. p. 165–88.
72. Gao L, Maidment I, Matthews FE, Robinson L, Brayne C, Medical Research Council Cognitive F, et al. Medication usage change in older people (65+) in England over 20 years: findings from CFAS I and CFAS II. Age Ageing. 2018;47(2):220–5. Epub 2017/10/17. doi: 10.1093/ageing/afx158 29036509.
73. Wu TY, Jen MH, Bottle A, Molokhia M, Aylin P, Bell D, et al. Ten-year trends in hospital admissions for adverse drug reactions in England 1999–2009. Journal of the Royal Society of Medicine. 2010;103(6):239–50. Epub 2010/06/02. doi: 10.1258/jrsm.2010.100113 20513902.
74. Office of National Statistics. Health state life expectancies by national deprivation deciles, England and Wales: 2014 to 2017. 2018 [Accessed 07 April 2019]. https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/healthinequalities/bulletins/healthstatelifeexpectanciesbyindexofmultipledeprivationimd/2015to2017
75. World Health Organisation. Medication without harm 2017 [Accessed 07 April 2019]. http://apps.who.int/iris/bitstream/10665/255263/1/WHO-HIS-SDS-2017.6-eng.pdf?ua=1&ua=1.
76. Donaldson LJ, Kelley ET, Dhingra-Kumar N, Kieny MP, Sheikh A. Medication Without Harm: WHO’s Third Global Patient Safety Challenge. Lancet (London, England). 2017;389(10080):1680–1. Epub 2017/05/04. doi: 10.1016/s0140-6736(17)31047-4 28463129.
77. Curtis K, Fulton E, Brown K. Factors influencing application of behavioural science evidence by public health decision-makers and practitioners, and implications for practice. Prev Med Rep. 2018;12:106–15. Epub 2018/09/21. doi: 10.1016/j.pmedr.2018.08.012 30233998.
78. National Assembly for Wales (NAfW) H, Social Care and Sport Committee,. Use of Antipsychotic medication in care homes. Cardiff Bay: National Assembly, Wales; 2018 [Accessed 21 July 2018]. http://www.assembly.wales/laid%20documents/cr-ld11556/cr-ld11556-e.pdf.