Study characteristics
Qualitative studies were published between 1999 and 2019, and were undertaken in Australia (n = 1),22 South Africa (n = 1),20 and the UK (n = 10),6,9–11,21,23–27 in primary care practices and community health centres. The quantitative studies were published between 1999 and 2018, and were undertaken in Canada (n = 1);28 Kenya (n = 1);29 South Africa (n = 2);20,30 the UK (n = 2);21,31 the US (n = 8),32–39 in primary care practices, paediatric hospitals, community health centres, ambulatory care clinics, and mobile clinics. Multiple long-term conditions were included in the studies, commonly asthma, diabetes, and hypertension. Of the 24 unique studies, nine evaluated existing templates already in use in clinical practice,9–11,22–23,25–27,31 eight studies developed templates in a programme of research with the primary intention of embedding in routine practice,20,24,28,30,32,37–39 and seven studies had developed templates for research purposes that were subsequently embedded in clinical practice.6,21,29,33–36 Detailed study characteristics can be found for the qualitative and quantitative studies in Supplementary Table S2 and Supplementary Table S3 respectively.
Qualitative synthesis
Template design and data collection
Health professionals found templates acted as a reminder tool during consultations.6,10,20–24 As one nurse reported:
‘I think they’re absolutely spot on, the templates. They’re just like reminders to make sure you don’t miss anything and they just make life a lot easier, basically. ’23
Templates established structure and made priorities clear, resulting in more efficient reviews.20,23,25–26 Conversely, rigid template design could be restrictive11,21 if structure was followed so closely that questions appeared out of context.25 Furthermore, over-reliance on structure reduced the health professionals’ opportunities to use their own medical knowledge and skills.27 Although some nurses expressed that templates ‘make life a lot easier’,23 they also commented that templates mean ‘you don’t really have to think a lot for yourself. ’ 10 Templates were viewed as inflexible if they did not provide space to record important additional comments.20–21 Additionally, a ‘tick-box’ design, as opposed to free-text comments, forced health professionals to categorise patients’ status, overriding nuances.11,23
One GP stated:
‘I don’t want a load of prompts and a load of forms to fill in and click and buttons.’6 [GP]
Competing agendas
Templates encouraged health professionals to prioritise their agenda over that of the patients.10,11,25 Patients had to work hard to integrate their own concerns into discussions and, even when successful, health professionals used the template to steer patients back to tasks.11 One template, the first question on which enquired about the patient’s agenda — namely, ‘What is the most important health problem that you would like us to work on over the next few months?’ — was valued by health professionals and patients.6
In some contexts, completing templates was an essential task, as it was how a practice secured its income;9 as such, health professionals felt under pressure to complete tasks and ‘tick the boxes’ that were related to evidence-based quality indicators.10,23,26 One nurse stated:
‘That becomes number crunching, ticking boxes and that’s the bit I don’t like. ’23
Shaping patient–practitioner interactions
Template use could reduce eye contact and disrupt dialogue.6 When patients ‘digressed’ from the template tasks to talk about their concerns, some nurses used a shift in gaze towards the computer template to disturb the patients’ narrative and turn the conversation back to the next task.25,27 Templates caused less disjunction when screen positioning did not require clinicians to turn away from the patient.6 Nurses also used body positioning to indicate that the template had their full attention, by turning their whole body towards the screen, signalling lack of interest, and limiting the patients’ narrative.25,27 More positively, patients became familiar with the health professionals’ priorities imposed by the template and knew what to expect of the review process and understood what was deemed acceptable during the review.25
GPs interviewed felt that:
‘[templates were too] business focused and took away from real doctoring.’22 [GP]
Impact on patient centred-care
Health professionals acknowledged that template use could turn reviews into a tick-box exercise, which inhibited patient-centred care,10,23–24,26 with review appointments becoming focused on collecting data rather than being an opportunity for patients to discuss treatment options for managing their condition.23 As one nurse commented:
‘You spend more time looking at the screen and ticking boxes than actually looking at the person who’s come to see you, which is not very nice for the patient.’26
There was a risk of health professionals avoiding discussing patients’ concerns if they were not related to the condition under review,11,25 with patients expressing dissatisfaction if their problems were not addressed.9 One patient noted that:
‘This gives me that kind of overview where you think “well, I’m the person that’s getting attended here, it’s not what this GP wants or thinks it’s what … my needs are”.’6
Conversely, patients who were asked about their concerns responded positively and felt heard.6
GPs suggested that templates could be improved by enabling them to cater for patients with multiple conditions.22 Some health professionals adapted their templates and practice to facilitate patient-centred care — for example, by extending appointment times, adding free-text comment boxes, employing strategies to involve patients in the review, or by hand writing notes and completing the template when the patient had left.9,11,21–23
Template impact on treatment options, self-management, and health promotion
Some health professionals considered that templates encouraged a pharmacological approach to management, despite patients often preferring non-pharmacological options.10,23 Using the template, GPs shifted topics away from the patient-initiated self-management topics — for example, reducing medication need — to a discussion of options around the need for medication,10 which might deter patients from attending reviews.23
Health professionals felt that following the template and raising multiple health-promotion topics — for example, smoking, diet, alcohol — could cause upset and lead to the patient feeling criticised.10 One nurse noted an occasion when this had happened:
‘I mean she was feeling a bit sort of got at, the fact that I’d already had the diet and the alcohol. And then smoking was the last straw really.’10
As a result, nurses tended to avoid these lifestyle topics to preserve the patient relationship.10 Conversely, however, some nurses used the template as an excuse for asking self-management questions.10
Health professional differences in template use
Nurses, and staff with less training such as healthcare assistants, felt constrained to ‘obey’ templates, whereas GPs were happier to override template requirements.23 GPs often considered templates as too detailed, whereas nurses felt the detail was necessary.21 GPs who were provided with a short template were more able to integrate it into their consultations than nurses using relatively long templates in LTC reviews,6 although they did not always explore the patient’s agenda if they lacked the required expertise.6
Although nurses engaged conversationally with patients’ social circumstances, most GPs referred to biopsychosocial circumstances as context for patients’ health.6 Staff with less training, such as healthcare assistants, felt less equipped.23
It was noted that nurses initiated self-management dialogue more frequently than GPs.10
Some example quotes illustrating nurse and GP approaches to template use were:
‘Yeah, you’ve got an agenda. They [the patient] may well have an agenda. And I tend to, rightly or wrongly, get my agenda first. You know, make sure my agenda’s done.’10
[Nurse]
‘There will be another agenda I’ll be running side by side … I’ve been able to cope OK with that.’23
[GP]
Quantitative synthesis
Use of templates
Overall, the majority of studies reported a rapid uptake or increase in the use of templates over the study period;21,29–32,35–36 however, one study (in which some concerns about risk of bias were highlighted) reported that <60% of patients’ folders contained the template being studied.20
Impact on documentation
Of the 14 included studies, 11 (all at moderate risk of bias or with some concerns) reported that review templates statistically significantly improved the documentation of key measures for their respective LTC.28–30,32–39 The studies that did not improve documentation reported lack of engagement with the research process and excessive workload undermining the ability to complete the template.20–21
Across the included studies, templates were reported to have the greatest effect on the process of disease management, including improved documentation of unscheduled care32,37,39 and symptoms.34 Templates were associated with a statistically significant improvement in the recording of condition severity’,29,32,35,37,39 with a change in documentation between 20% (P = 0.0013)35 and 73% (P<0.001).39 Statistically significant improvement in documentation was also noted for environmental exposure (for example, mould, occupational hazards).29,32
One asthma template study reported a statistically significant increase in the documentation of changes in care plans, including social work referral, subspecialty consultation, and medicine change (from 49% to 63%, P = 0.0006).32
The results were mixed with regards to complications or comorbidities, and changes in care plans that were documented. With regard to asthma, for example, documentation of an asthma action plan provision was mixed: one study found that documentation increased (from 10% to 74%, P = 0.001),37 whereas one found no statistically significant difference.34
Specifically, the impact on the documentation of prescribed controller medication was mixed, with some studies indicating a statistically significant improvement in documentation,36,39 while another study did not.37 One study observed increased documentation of inhaled corticosteroid use before and after template implementation (from 39.4% to 51.1%, P = 0.0170).35
Studies involving patients with hypertension and diabetes found mixed results in documentation changes regarding complications, with one study finding a statistically significant improvement in documentation,29 and another study finding no difference.20
The only study to report on family history (for example, history of smokers in family) did not report a statistically significant change in documentation.34
Impact on health outcomes
The only study to report on health outcomes (for which there was some concern about risk of bias) showed no statistically significant effect on glycaemic control for patients with diabetes or blood pressure control for those with hypertension following a template being introduced 1 year before the study was undertaken.20