Methodological considerations in clinical outcomes assessment of pharmacy-based minor ailments management: A systematic review

Background The accessibility of services within community pharmacies provides an ideal opportunity to manage minor ailments, yet over £1.1 billion is spent by the National Health Service (NHS) in the United Kingdom (UK) in managing minor ailments in high cost settings. There is a need to review the evidence base around clinical effectiveness of pharmacy-based management of minor ailments since the absence of such may lead to under-utilisation of pharmacy services and non-implementation of available pharmacy service models. This study aimed to systematically review the methodological approaches used to assess clinical outcomes of pharmacy-based management of minor ailments in the research literature. Methods A systematic review was conducted to identify relevant literature using the following databases: Medline, EMBASE, CINAHL, IPA, CRD, CDSR, and Google Scholar from publication year 2000 onwards. Studies were included if they evaluated clinical outcomes of pharmacy-based management of any minor ailments, with or without a comparator setting such as Emergency Departments (EDs) or general practices. Screening and selection of titles, abstracts and full texts followed by data extraction and quality assessment (QA) was conducted. Paired researchers, from the team, reviewed papers using a protocol based on the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P). QA was undertaken using the Critical Appraisal Skills Programme (CASP). Reporting was conducted in accordance with PRISMA checklist and statements. Results A total of 19 studies were included. The majority of studies were observational, conducted in community pharmacies, and did not use a comparator participant group nor a comparator setting. Interventions included counselling, medicines supply and provision of advice on the management of minor ailments. One study used the randomised controlled trial (RCT) design with majority of the study utilising observational design. A range of clinical outcomes including symptom severity, pattern, resolution, and quality of life were reported. Methods used for the assessment of clinical outcomes were, overall, poorly reported. This included a lack of information on the development and validation of the data collection tools and the timing of baseline and follow-up data collection. Adverse clinical outcomes data were collected by only seven studies. Conclusions Currently, there are methodological limitations in the studies that have sought to assess clinical outcomes of pharmacy-based management of minor ailments. Such lack of high quality evidence may contribute to failings to shift care from high cost settings, such as EDs and general practices. Generation of high quality evidence is likely to influence public choices when seeking care for minor ailments. There is scope for development of a core outcomes set specific to minor ailments management and development of a validated methodology for measuring such outcomes in a research study.

Background Minor ailments are defined as 'common or self-limiting or uncomplicated conditions which can be diagnosed and managed with minimum or no professional support' [1]. For example, cough, cold, hay fever, red eye, minor sprains and pains. Despite the widely acknowledged expertise of community pharmacists, and pharmacy support staff, and their contribution to the management of minor ailments, internationally, there is an under-utilisation of pharmacy. For example, a recent analysis of routinely collected data in hospital emergency departments (ED) and in general practices in the United Kingdom (UK) demonstrated that between 5% and 13% of consultations respectively were for minor ailments equating to a cost of £1.1 billion [2]. Up to 20% of all general practitioner (GP) consultations are known to be for minor ailments consultations alone excluding the data where minor ailments forms a part of a consultation for a more serious condition [3]. The services of pharmacists in ED and general practices in managing minor ailments, inclusive of the recruitment of pharmacist independent prescribers, has recently been introduced in the UK [4].
There is overwhelming support for enhanced pharmacy-based management of minor ailments from pharmacists and associated professional and regulatory bodies across Europe [5][6][7][8]. For example, the Royal Pharmaceutical Society (RPS), the pharmacy regulator in the UK, supports the enhanced management of minor ailments through community pharmacy. It is proposed that enhanced management of minor ailments from pharmacy can improve patient choices in their use of health services, reduce the burden on general practice and ED, increase patient access to local services for minor ailment management and permit professional development opportunities for pharmacists. The government policies also envisage that enhanced pharmacy assisted self-care would contribute to significant patient health benefits in the longer term [6].
The existing burden of patients presenting with minor ailments to general practice and ED highlights the need for further research to identify any gaps between policy and practice. Lack of adequate evidence may negatively impact on the promotion of relevant practice and policy implementation [9][10][11]. Pharmacy-based services such as the Minor Ailment Scheme (MAS), which allows free access to the management of minor ailments, have been introduced through all community pharmacy across Scotland. Patients may present with a minor ailment in a community pharmacy and expect to receive a structured, formulary-based approach to treatment from a trained, pharmacy-based team. Those who fulfil eligibility criteria such as those 60 years or older, pregnant women and those on income support and allowance, are able to obtain over-the-counter medicines at no cost, thereby increasing access to treatments for patients where affordability may be a barrier [12]. In England, however, this particular service has not been offered at a national level, and Primary Care Trusts in England are yet to implement a publicly funded minor ailments scheme in their regions. Similar models of Government funded pharmacy-based management of minor ailments are uncommon in European countries [13].
The development of a robust evidence base for pharmacy-based management of minor ailments necessitates that the strengths and limitations of the current evidence is reviewed. Types of clinical outcomes and their methods of assessment relating to pharmacy minor ailments management have not previously been systematically reviewed. Measuring clinical outcomes of minor ailment management in research evaluation presents challenges for data collection, analysis and interpretation. For example, typically short and self-limiting episodes of illness, diversity in the range of conditions considered to be minor ailments and lack of appropriate follow-up methods are inherent issues in minor ailments research. Therefore, it is often difficult in research studies, including comparative evaluations, to synthesise clinical outcomes across different illnesses, treatments and settings. Currently, there is no gold standard with regard to the type of clinical outcome data that should be assessed and the methods of assessment that should be deployed as a component of an intervention study.
This study aimed to systematically review the methods and types of clinical outcomes assessment used in the evaluation of pharmacy-based minor ailment management.

Methods
A protocol was prepared using Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P, CRD42016050847) (S1 Appendix 1) [14]. An electronic search of MEDLINE, EMBASE, CINAHL, Centre for Review and Dissemination database (CRD), Cochrane Database of Systematic Reviews (CDSR), and Google Scholar were undertaken using Medical Subject Headings (MeSH) and natural language key words, Boolean and proximity (e.g. adj2] operators, truncations ( Ã ) and wild cards ($). The draft search strategy was utilised to perform a scoping search and moderated by the research team before a final version was agreed for each database. The reference lists of included studies were hand searched to identify any additional references for inclusion. Authors of conference abstracts were contacted to identify any related full text publications. Referencing software was used to manage included and excluded references and remove duplicate search results. An example search strategy is provided in S2 Appendix.

Screening and selection
Screening was conducted by one pair of researchers in three consecutive stages: screening of titles; screening of abstracts; screening of full text against the inclusion and exclusion criteria listed in the protocol (S1 Appendix). Disagreements regarding inclusion were resolved via discussion within the pairs.

Data extraction and quality assessment
A data extraction tool was developed based on the review aims and objectives, refined, reviewed and approved by the research team. The tool was piloted using two studies which were subsequently included in the review. The Critical Appraisal Skills Programme (CASP) quality assessment tools (for randomised controlled trials (RCT) and cohort studies) were used to assess study quality [15]. Studies were not excluded based on quality since the quality assessment of existing literature pertained to the study objectives. Data extraction and quality assessment were allocated to one of the three pairs of researchers from the team for independent review. The research team as whole met and discussed after a pilot data extraction exercise using the draft form. The data extraction form was then finalised and the team met on a regular basis to ensure that data extraction and quality assessment across the pairs were consistent. Any disagreements were resolved through discussion.

Synthesis
A narrative synthesis of the results was conducted. Due to the methodological nature of the systematic review, a meta-analysis was not considered appropriate.

Outcome measures
The following outcomes were evaluated in the context of evaluation of minor ailment management: a) types of clinical outcomes used and b) methods of clinical outcome measurement.

Guidelines considered for the review
Reporting of this review was based on the PRISMA statement and checklist [16] (S3 Appendix).

Quality assessment
Among the 18 non-RCT studies assessed for quality using the CASP quality assessment tool, 17 studies presented clear study aims (Tables 1 and 2). The majority of studies did not provide adequate information on the recruitment strategy (n = 12). These included missing information on how the participants were identified and followed up. The majority of the included studies did not indicate whether the follow-up duration was validated in the context of clinical area being evaluated (n = 13). Confounding factors were only rarely considered when measuring the follow-up outcomes (n = 2).
The only RCT study (18) included in the study was single blinded at participant level. Information on the randomisation process was presented clearly. Comparative data on the baseline characteristics across the control and intervention groups however was not made available ( Table 2). Complete outcome data were reported alongside intention to treat analysis.

Participants, sample size and recruitment
Sample size calculations or sampling strategies were rarely justified [1,18,20] (Table 4). Inclusions were typically stipulated with regard to patient age (e.g. over 17/18 years) and presenting with the minor ailment subject to study or receiving treatment for a specific minor ailment.

Intervention
The majority of studies stated that pharmacists or members of pharmacy staff delivered the intervention [17,18,20,21,24,26,29,[32][33][34]; although there was no clear distinction of their remit. One study used NHS Direct telephone staff to refer potential participants to out-ofhours pharmacy settings including referral to other settings such as the general practice [25]. Interventions were also delivered by general practitioners (GP) or nurses in a general practice [1,18], and doctors or nurses in ED [1] with the general practice or ED used as a comparator setting against community pharmacy in the two studies [1,25]. Nine studies reported that specific training had been provided to staff as either a component of the intervention or the research [19, 20, 24, 26-29, 31, 32]. For the majority of studies (n = 13), it was not specified who paid for the service at the point of care [1, 17-21, 24-28, 32, 34]. The remaining studies reported that minor ailment treatment was either paid for by patients [22,23,[29][30][31]33] or free at the point of care [33].

Types of clinical outcomes used
Symptoms status. Clinical outcomes were evaluated in relation to symptom severity, pattern (i.e. frequency of specific symptoms) and resolution of symptoms (Table 5). Four studies evaluated disease-specific symptoms (Table 5). These included evaluation of associated symptoms of cough [18], influenza-like illness or ear, nose and throat disorders [21], cold [23] and gastrointestinal issues [27]. In contrast to the aforementioned, one study utilised a generic tool to gather data on the symptom status across four minor ailments at baseline and follow-up [1].
Various methods were used to gather clinical data on symptom severity. These included a Visual Analogue Scale (VAS) [27] and Likert scales ranging from four [21,22] to 11 points [23]. One study reported using body mass index (BMI) measures, as a means of detecting symptoms which were associated with more serious conditions, in addition to measuring minor ailment specific symptoms [27].
Re-consultation. Re-consultation was assessed in four studies and was intended as a surrogate follow-up measure of clinical outcome assessment [1,30,32,33]. Patients in these studies were asked whether they had subsequently consulted with a GP/physician for the minor 2009 UK To compare health-related and cost-related outcomes of consultations for symptoms suggestive of minor ailments in EDs, GPs and community pharmacies.
Observational Community Pharmacies (10) General Practices (6) EDs (2) (Continued) ailment after their index consultation. In addition, one study also evaluated re-consultation within the same setting for the minor ailment presented during the index consultation [1]. Adverse clinical outcomes. Only seven studies included in this review evaluated adverse clinical outcomes (Table 5). A lack of standardised terminology and data collection tools was observed in relation to referrals and reporting adverse clinical outcomes. Terminologies used to refer to adverse clinical outcomes included: side effects, tolerability, drug-drug interactions, drug related problems (DRPs), overuse and misuse. For example, tolerability was a follow-up outcome in two studies evaluating the clinical outcomes of pharmacy management of cough [30] and cold [23]. Tolerability for the study purpose was not defined in either study. In general, there was a lack of information provided on the validation of tools utilised in the pharmacy setting.
Quality of life and health status. Four studies evaluated quality of life outcomes, two of which used disease-specific tools [18,24]. Birring [18] used the Leicester Cough Questionnaire for acute cough at baseline and follow-up. Krishnan [24] used Gastro-intestinal Quality of Life Index in the assessment of outcomes of patients recommended treatment for dyspepsia at baseline and follow-up. Neither study provided information on the validation of the tools or whether they were fit for use in the pharmacy setting. Two studies used more generic tools, for example, the Generic EuroQoL was used to evaluate quality of life in two studies [1,25] at both baseline and follow-up. These studies did not justify whether the tools used were sensitive or had been validated for use in the context of minor ailments. Health status was evaluated using global measures in four studies [1,22,25,26].

Methods of clinical outcomes assessment
Data collection tools and their administration. Scant methodological details were reported around how baseline and/or follow-up clinical data were collected from study participants (Table 5). Five studies described that baseline clinical data were recorded by the pharmacist or member of pharmacy staff in the pharmacy setting, as a component of the clinical consultation using structured questionnaires [17,[27][28][29]33]. Four studies collected baseline data via self-administered or researcher-administered questionnaires in pharmacy or comparator settings [1,17,23,33].
A limited number of studies attempted to provide information about the validation of the methods used to collect either baseline or follow-up clinical data (Table 5). Where information was available on the development of the data collection tools, these were mostly limited to testing of face and content validity. The utility of generic validated quality of life instruments within the context of the clinical areas evaluated was not clear [1,25].
Baseline and follow-up data. Five studies did not collect any baseline data [19,24,28,31,34] (Table 5). Two studies were explicit in reporting the timing of administration of data collection tools: either prior to [1] or after the index consultations [26]. Various timelines were Clinical outcomes assessment in minor ailments management: A methodological systematic review Clinical outcomes assessment in minor ailments management: A methodological systematic review Clinical outcomes assessment in minor ailments management: A methodological systematic review  [18,26,31]. Follow-up timelines were specific to each minor ailment in one study [26]. To illustrate, timelines were either 7 days for cold sores or insect bites, and 30 days from the index consultation for seasonal allergies based on the length of the symptoms [26]. Symptom or medication diaries were used by patients in two studies [18,27]. Missing follow-up data. Statistical or methodological techniques used to extrapolate follow-up missing data were applied by one study [22]. This study used the 'last observation carried forward' technique as imputation method for missing data over time.
Patient satisfaction measures. This systematic review only sought to review the methods used to measure patient satisfaction if they were assessed alongside clinical outcomes. Ten Re-consultation with GP following pharmacist consultation.
Unclear Unclear studies did not measure patient satisfaction [17-20, 24, 25, 27, 30, 31, 33]. Studies which utilised patient satisfaction measures varied in their approach, either using categorical questions [23,26,32], Likert scales [26,29,34], the Treatment Satisfaction Questionnaire for Medication [22] or the Medical Interview Satisfaction Scale [1]. One study did not clarify how patient satisfaction was measured [28]. Two studies reported using validated tools for measuring patient satisfaction [1,22]. Results related to clinical outcomes. Improvement of clinical outcomes was reported across a number of studies [1, 17-28, 30, 32]. Two studies demonstrated that the clinical effectiveness of pharmacy-based minor ailment management was equivalent to, or improved, when compared with management by other health care professionals in settings such as general practice or EDs [1,25].

Discussion
This study aimed to systematically review the methods and types of clinical outcomes in the evaluation of pharmacy-based management of minor ailments. The results have demonstrated a lack of high quality, adequately powered studies used in the evaluation of pharmacy-based minor ailments management. Amongst the studies included, only one study used an RCT design and there was a lack of adequately powered longitudinal follow-up studies. Within the included studies, explicit adherence to best practice guidelines [35][36][37][38] relating to study methodology was notably lacking. The Medical Research Council Framework signifies the importance of developing and validating the methodological tools used in the developing and evaluation of complex interventions [39].
The included studies evaluated a range of clinical outcomes which included symptom status (inclusive of symptom severity, pattern, and resolution); re-consultation; adverse clinical outcomes; and quality of life. However, key methodological information was missing around the choice of clinical outcomes used, the development and validation of data collection tools used and the timelines of baseline and follow-up outcome data. Given that some of the minor ailments may be self-limiting in nature, lack of validation of the follow-up timeline may affect estimates of the impact of the interventions. A number of studies evaluated the clinical outcome data only at the follow-up stages.
A number of studies did not utilise disease-specific clinical outcome assessments. Given the diverse clinical areas being evaluated, there is scope for the development of core outcome sets that could be applied to the evaluation of a range of minor ailments. Such core outcome sets have been developed through consensus around reporting of clinical outcomes in clinical trials [40]. Consistent outcome sets would also facilitate the ability to conduct meta-analyses of clinical outcomes and direct comparisons between studies.
A minority of the included studies considered evaluation of adverse clinical outcomes data at follow-up and these were often poorly defined in study reports. Best practice guidelines suggest that study reports include objective information on the incidence and type of clinically relevant adverse events (including any adverse events requiring discontinuation of therapy) as opposed to the use of general statements such as "well tolerated". Information on the severity, frequency, and timing of adverse events are of additional value [40]. Various types of symptom severity scales were used in the included studies, for both positive clinical outcomes and the adverse clinical outcomes. In particular, there was a lack of information provided on the validation of tools in the context of their use in minor ailments and pharmacy as a study setting. Study settings and disease conditions can have important influence in the validity and reliability of symptoms and quality of life assessment tools [41]. Future research should focus on developing standard practice to address the variance in their use.

Strengths and limitations
This is the first methodological systematic review of studies evaluating the type of clinical outcomes and methods used for their assessment in the evaluation of pharmacy-based management of minor ailments. Standard guidelines [14,15] were used to inform the review process. Being a methodological systematic review, this study did not seek to review in-depth the impact of pharmacy management of minor ailments on clinical outcomes. This systematic review applied wider inclusion criteria when considering the selection of studies. Thereby, the studies included in the review evaluated a broad range of minor ailments. Hence, in-depth consideration of methodological aspects of research relevant to individual clinical areas were not possible.

Impact on practice
The lack of high quality research, in terms of methodological rigour, as identified by this systematic review, is a barrier to the promotion of pharmacy-based services aimed at minor ailments management. In the UK, despite the country leading the reclassification of prescription only medicines for over-the-counter supply (including pharmacy only status), the burden of minor ailments in high cost settings still remains a key issue. This review demonstrates a lack of high quality evidence in relation to the clinical outcomes of pharmacy-based management of minor ailments; a factor which may contribute to failings to shift care from high cost settings, such as EDs and general practices. Reclassification decisions are often based on the 'do no harm' principle and on the experiences of their use as a prescription medicine and assumption of reduced costs to the health services. The evidence base for pharmacy-based provision of services and medicines are often ignored in such decisions.

Recommendations for research
The results of this study suggest that future research on pharmacy-based management of minor ailments should adhere to the following recommendations: • Utilisation of high quality RCT or longitudinal observational studies informed by best practice guidelines.
• Use of validated clinical outcome measures to generate high quality evidence. These include development and use of disease-specific outcomes assessment tools and core-outcome sets for a range of minor ailments.
• Use of out-of-hours services, online healthcare services, self-care without professional support, and over-the-counter management in non-pharmacy setting, in addition to pharmacist-led services in general practices and ED comparators to community pharmacy-based models of minor ailments management. Comparative evaluations will be of value to practitioners, policy makers and researchers in terms of identifying service improvement and cost effectiveness.
• Research studies should distinguish between interventions delivered by pharmacists and those delivered by support staff under pharmacist supervision in a pharmacy.
• Inclusion of adverse clinical outcomes since they were rarely considered by the included studies and are particularly important where clinical areas are novel such as in the context of the evaluation of newly reclassified medicines aimed for pharmacy supply.