Chronic obstructive pulmonary disease (COPD) is a progressive respiratory condition, characterised by breathlessness and excess sputum production (Wedzicha et al, 2003). Managing COPD places great demands on the affected individual, due to fluctuating progressive respiratory symptoms, overlaid with complex comorbidities (Vogelmeier et al, 2017; Welch et al, 2021) and the reliance on self-management.
Effective self-management requires a range of skills that are vital for an individual to support the day-to-day responsibility of managing their condition, as well as responding to changes or exacerbations. Supporting self-management is a process that should be patient focused, with shared decision-making between patient and clinician (Welch et al, 2021). For some individuals the complexity of accessing, understanding and applying new health self-management skills, and changing their behaviours, constitutes a considerable new workload (May et al, 2014), and individuals' health literacy skills will impact on their ability to self-manage their long-term condition (Adams et al, 2009).
Health literacy is considered to be an individual's ability to access, understand and use information in order to make informed decisions about their health (Batterham et al, 2016; Nutbeam and Muscat, 2021). However, there is a lack of consistency in the definition of health literacy in the literature (Larsen et al, 2022). Health literacy is perceived as a skill (which can be improved), a risk factor for poor outcomes, or it can be taken into consideration by health professionals when providing care (Nutbeam, 2008).
Health literacy in people with COPD incorporates:
These skills empower individuals to exert greater control over their own health outcomes (Nutbeam, 2008).
Health literacy is a major factor in determining how successful patients are at using the medical treatments and support available to them (Kondilis et al, 2008), and is thought to be key in decreasing the disease burden of chronic conditions (Heine et al, 2021). Low health literacy levels are correlated with higher levels of morbidity and mortality (Adams et al, 2009) and, in COPD, Omachi et al (2013) report that those with the lowest levels of health literacy may have a five-times increased risk of visiting the emergency department, and a seven-times increased risk of hospitalisation. In contrast, higher levels of health literacy are thought to act as a precursor to successful self-management by equipping the individual with the knowledge and skills to manage their condition and improve their health outcomes (Kondilis et al, 2008).
Literacy is impacted by variables such as education level and psychological wellbeing (Wahl et al, 2021). However, health literacy is also recognised not to be just an individual process, because there can also be influences from a patient's environment, including their family and support networks, known as collective efficacy (Vassilev et al, 2019). In addition, external factors such as healthcare providers and the organisational healthcare structure influence an individual's health literacy (Batterham et al, 2016; Larsen et al, 2022). Health inequalities influenced by socioeconomic status also play a role in determining health outcomes (Gershon et al, 2012; Li et al, 2022). As a result, understanding health literacy levels in various populations may be key to developing supportive strategies (Friis et al, 2016).
For COPD populations, health literacy is particularly relevant. Studies that have explored the issue in participants with COPD suggest that it may be lower in this group compared with other non-chronic disease populations, such as people with diabetes or heart failure. For example, in Puente-Maestu et al's research (2016) 59% of their COPD sample had inadequate health literacy levels and a reduced understanding of their disease. Inadequate health literacy in COPD is associated with poor health outcomes, including respiratory-specific health outcomes and quality-of-life outcomes (Stellefson et al, 2019). Furthermore, patients report increased difficulty in completing their activities of daily living, along with increased levels of dependency and failure to respond to exacerbations, resulting in more hospital admissions.
Although the studies cited above make important contributions to understanding the impact of health literacy for people living with COPD, several gaps remain. First, the outcome measures used to measure health literacy in COPD (Omachi et al, 2013; Puente-Maestu et al, 2016) are predominately concerned with reading and comprehension; these are functional levels of health literacy and, consequently, may not adequately measure the breadth of the skills that contribute towards health literacy (Nutbeam, 2008). Second, research in a local UK sample is needed due to differences in healthcare systems and populations between countries; it therefore cannot be assumed that previous findings are generalisable to the UK population. As a result, the aim of this research was to investigate health literacy levels across one UK-based COPD population to address the health literacy gap in the research. This is key: with more self-management tools becoming digitalised, understanding health literacy before digital literacy enables informed development of digital media in people with COPD.
Methods
This study aimed to describe the health literacy levels of people with COPD in a UK population in the Southampton/Hampshire area and explore the relationship between health literacy levels and COPD severity. The study objectives were to:
| Grade | Description |
|---|---|
| 1 | Not troubled by breathlessness except on strenuous exercise |
| 2 | Troubled by shortness of breath when hurrying on the level or walking up a slight hill |
| 3 | Walks slower than people of the same age on the level because of breathlessness or has to stop when walking at own pace on the level |
| 4 | Stops for breath after walking about 100yards (90m) or after a few minutes on the level |
| 5 | Too breathless to leave the house or breathless when dressing or undressing |
Adapted from: Medical Research Council, 1959. Used with the permission of the Medical Research Council
Design and sampling
A cross-sectional, observational design was used to measure health literacy using the HLQ and COPD severity using the MRC Breathlessness Scale. This study design was selected because it allowed for data collection at a one-off time point, across a large sample. This is appropriate as health literacy is deemed to be a relatively stable skill (Nutbeam, 2008), and collecting data at one-off time points reduced the risk of loss to follow-up (Mann, 2003). The HLQ was selected because it considers the breadth of skills that make up health literacy (Osborne et al, 2013), rather than focusing solely on reading and comprehension.
Ethical approval for the study was granted by the North West–Liverpool East Research Ethics Committee. Participants were eligible to take part in the study if they were 18 years old or older and reported to have a diagnosis of COPD. There were no restrictions on participants' disease status or level of health. Participants were excluded if they were unable to read and understand English (no funds for a translator were available for this study), or identified as unsuitable due to factors such as a cognitive impairment, which would restrict their ability to give informed consent, as decided by one of the patient's direct care team.
The desired sample size for this study was calculated by using the results of Friis et al's (2016) article; using their estimates of outcome variability, otherwise known as the standard deviation, for two of the domains of the HLQ used in this article. These estimates have been used to ensure that the desired level of precision, 95% confidence intervals, is achieved, to ensure the sample is a good reflection of the population. Based on these calculations, a sample of 80 allowed for this level of accuracy.
Study setting and recruitment strategy
Recruitment ran from December 2017 to June 2018 via one of three routes: NHS outpatient clinics, NHS pulmonary rehabilitation (PR) or PR maintenance and Breathe Easy groups, in a localised region of the UK. This approach was chosen to maximise the ability to meet the sample size requirement.
When recruiting in NHS outpatient clinics the direct care team screened and invited eligible patients to the study. For PR and Breathe Easy groups the researcher attended a group session and presented a summary of the study. Interested patients were given a research pack, which included a stamped addressed envelope in which to return their completed questionnaire. Return of the questionnaire was taken as implied consent.
Data collection materials
Data were collected using three questionnaires:
| Low level of domain | High level of domain |
|---|---|
| Feeling understood and supported by healthcare providers (HCP) | |
| Unable to engage with doctors and other healthcare providers. Do not have a regular healthcare provider and/or have difficulty trusting healthcare providers as a source of information and/or advice | Has an established relationship with at least one healthcare provider who knows them well and whom they trust to provide useful advice and information and to assist them to understand information and make decisions about their health |
| Having sufficient information to manage my health (HSI) | |
| Feels that there are many gaps in their knowledge and that they do not have the information they need to live with and manage their health concerns | Feels confident that they have all the information that they need to live with and manage their condition and to make decisions |
| Actively managing my health (AMH) | |
| Do not see their health as their responsibility, they are not engaged in their health care and regard health care as something that is done to them | Recognise the importance of and are able to take responsibility for their own health. They proactively engage in their own care and make their own decisions about their health. They make health a priority |
| Social support for health (SS) | |
| Completely alone and unsupported for health | A person's social system provides them with all the support they want or need for health |
| Appraisal of Health information (CA) | |
| No matter how hard they try, they cannot understand most health information and get confused when there is conflicting information | Able to identify good information and reliable sources of information. They can resolve conflicting information by themselves or with help from others |
| Ability to actively engage with healthcare providers (AE) | |
| Are passive in their approach to health care. They accept information without question. Unable to ask questions to get information or to clarify what they do not understand. They accept what is offered without seeking to ensure that it meets their needs. Feel unable to share concerns. They do not have a sense of agency in interactions with providers | Is proactive about their health and feels in control in relationships with healthcare providers. Is able to seek advice from additional healthcare providers when necessary. They keep going until they get what they want. Empowered |
| Navigating the healthcare system (NHS) | |
| Unable to advocate on their own behalf and unable to find someone who can help them use the healthcare system to address their health needs. Do not look beyond obvious resources and have a limited understanding of what is available and what they are entitled to | Able to find out about services and support so they get all their needs met. Able to advocate on their own behalf at the system and service level |
| Ability to find good health information (FHI) | |
| Cannot access health information when required. Is dependent on others to offer information | Is an ‘information explorer’. Actively uses a diverse range of sources to find information and is up to date |
| Understand health information enough to know what to do (UHI) | |
| Has problems understanding any written health information or instructions about treatments or medications. Unable to read or write well enough to complete medical forms | Is able to understand all written information (including numerical information) in relation to their health and able to write appropriately on forms where required |
Adapted from Osborne et al, 2013
The HLQ uses a psychometric scale to underpin the domains, with each question increasing in difficulty within the same domain, allowing the authors to measure the breadth and depth of each area of health literacy. This ensures the scale is sensitive enough to detect small changes, regardless of the participant's level of health literacy (Bloom, 1956). Additionally, the scale was shown to have good reliability and validity (Raykov, 1997; Osborne, 2013).
Analysis
Following manual data entry onto Excel, the team statistician checked 33% of the participants' data for potential errors. Data were analysed using the statistical package SPSSv24. Histograms were used to guide choice of summary statistics. Mean and standard deviation were used for continuous demographics. The number of comorbidities is presented as median and quartiles. Categorical data were displayed using the number and percentage of the total sample for each category. HLQ results are presented as means, with corresponding 95% confidence intervals, and standard deviations for each of the nine domains.
Following visual assessment of the relationship between the MRC and HLQ domains, Spearman's correlations were calculated between each HLQ domain and the MRC score to measure the strength and direction of the relationship, which are presented with 95% confidence intervals.
In instances where participants ticked multiple boxes on the MRC scale in response to which statement best described how breathlessness affected their activities, the score recorded was taken from the statement with the highest score, indicating the individual's worst degree of breathlessness. This decision is supported by the inherent progressive nature of the scale, requiring the participant to experience symptoms from all the previous boxes. Missing data on the HLQ was managed by following the HLQ scoring advice provided by the developers of the HLQ (Osborne et al, 2013).
Results
Of the 130 questionnaire packs distributed, 61 participants returned the questionnaires, a return rate of 47%: 29 from clinics (48%), 21 from Breathe Easy support groups (34%), and 11 from PR classes/maintenance groups (18%). This study recruited roughly two-thirds of the participants from the NHS clinics and PR/maintenance groups in order to reduce a potential for bias of attending a support group, which has the potential to favour those of higher health literacy (Raykov, 1997).
Participant demographics
Participant demographics can be found in Table 3.
| Demographic | N (%) |
|---|---|
| Sex (F/M) | 27 (44%)/34 (56%) |
| Ethnicity | |
| White British | 60 (98%) |
| Missing | 1 (2%) |
| Mean (SD) | |
| Age (years) | 70 (7.9) |
| Time since diagnosis (COPD; years) | 8.1 (6.2) |
| MRC Breathlessness scale | 3.2 (1.1) |
| Median (quartiles) | |
| Number of comorbidities | 2.0 (1, 3) |
Health Literacy Questionnaire
Mean scores and confidence intervals for each domain of the HLQ are shown in Table 4. The means scores for each domain indicate a patient's self-reported level of confidence in each skill, with higher scores indicating higher levels of confidence. The results for each domain are also displayed in two box and whisker plots, which are a way of representing the distribution of the data, including the median, mean and interquartile ranges. The whiskers demonstrate the outliers. Part one of the HLQ is shown in Figure 1(a) and part 2 is shown in Figure 1(b).
| HLQ scale | Number of people | Mean score (SD) | 95% CI fo mean |
|---|---|---|---|
| Part 1 (scale 1−4) | |||
| Feeling understood and supported by healthcare providers | 61 | 2.8 (0.7) | 2.7 to 3.0 |
| Having sufficient information to manage my health | 61 | 2.9 (0.5) | 2.8 to 3.1 |
| Actively managing my health | 61 | 2.9 (0.5) | 2.8 to 3.1 |
| Social support for health | 61 | 2.9 (0.6) | 2.7 to 3.0 |
| Appraisal of health information | 61 | 2.6 (0.5) | 2.5 to 2.7 |
| Part 2 (scale 1−5) | |||
| Ability to actively engage with healthcare providers | 61 | 3.4 (1.0) | 3.2 to 3.6 |
| Navigating the healthcare system | 61 | 3.2 (0.8) | 3.0 to 3.4 |
| Ability to find good health information | 61 | 3.4 (0.8) | 3.2 to 3.6 |
| Understand health information well enough to know what to do | 61 | 3.8 (0.7) | 3.6 to 4.0 |
Scores were similar across all of the domains in part one of the questionnaire, with the exception of ‘appraisal of health information’ (CA) (mean: 2.6, 95% confidence interval (CI): 2.5, 2.7), which was lower. In part two, the ability to ‘understand health information well enough’ (UHI) scored noticeably higher (mean 3.8, 95% CI: 3.6, 4.0) and, ‘navigating the health care system’ (NHS) scored the lowest (mean 3.2, 95% CI: 3.0, 3.4).
HLQ and COPD severity
Table 5 displays the results of a Spearman's bivariate analysis. This is designed to measure the strength of a relationship between two variables and the direction of the relationship. Table 5 displays the relationship and direction of the relationship and the mean of each of nine HLQ domains. Higher scores indicate greater confidence in that skill. Table 5 also displays correlations between HLQ domains and the MRC Breathlessness Scale, of which higher scores indicate greater breathlessness. Two of the nine domains displayed a statistically significant correlation with MRC. The first shows a negative correlation for ‘having sufficient information to manage my health’ and MRC score (r=-0.27, 95% CI [-0.50 to -0.01], P=0.04), demonstrating that people with more severe breathlessness reported not having enough health information available to them. Second, a statistically significant positive correlation was seen between MRC score and ‘Appraisal of health information’ (r=0.26, [0.01 to 0.48], P=0.04), highlighting that people with more severe breathlessness report greater skill in appraising health information. The other domains in the HLQ did not show statistically significant correlations with MRC.
| Domain | Spearman's correlation | 95% Confidence interval | P value |
|---|---|---|---|
| Feeling understood and supported by healthcare providers | -0.11 | -0.36 to 0.15 | 0.39 |
| Having sufficient information to manage my health | -0.27 | -0.50 to -0.01 | 0.04 |
| Actively managing my health | 0.08 | -0.17 to 0.31 | 0.55 |
| Social support for health | -0.04 | -0.30 to 0.21 | 0.77 |
| Appraisal of health information | 0.26 | 0.01 to 0.48 | 0.04 |
| Ability to actively engage with healthcare providers | -0.07 | -0.31 to 0.16 | 0.57 |
| Navigating the healthcare system | -0.05 | -0.29 to 0.19 | 0.68 |
| Ability to find good health information | -0.08 | -0.31 to 0.14 | 0.57 |
| Understand health information well enough to know what to do | 0.00 | -0.25 to 0.25 | 0.98 |
Discussion
This is the first study to assess health literacy in a UK-based COPD population using the HLQ. The higher percentage of males in the sample reflects that of the wider UK COPD population (Snell et al, 2016). COPD is a condition strongly linked to increased smoking prevalence, and occupational exposure (Hegewald and Crapo, 2007), which are both more prevalent in demographics with lower socio-economic status (Lindström et al, 2014). Throughout the literature there is strong evidence to support that low socio-economic status, including low educational attainment, is linked with poor health outcomes (Eisner et al, 2009; Gershon et al, 2012). As a result, we need to consider that the increased prevalence of low socio-economic status in a COPD population might explain some of the differences seen in health literacy within the general public. This highlights a potential area for consideration for those designing or delivering healthcare services.
The aim of the study was to explore health literacy levels in one UK-based COPD population area, and by looking at the individual domains of the HLQ the authors were able to identify areas in which the population felt they were more and less confident in their ability, which included domains describing patients' own actions or abilities. These findings contrast with research in the general population, suggesting that these groups score highest on social interaction and communication-type domains (Jessup et al, 2017; Hawkins et al, 2017). This suggests that health literacy levels across the HLQ domains may differ between people with and without COPD (Chew et al, 2004; Beauchamp et al, 2015).
This is supported by Friis et al's (2016) data, which indicated that people with chronic conditions might score lower for the ‘engaging with HCPs’ domain compared to those in the general public. This could be a reflection on the wider health service, and organisations' health information not being designed to support patients with chronic conditions, with many services being reactive rather than proactive to changes in a patient's condition. This has been recognised in the NHS Long Term Plan (NHS England, 2023), including financial commitments to enable the realisation of proactive care. However, in terms of health literacy, clinicians also report that patients with COPD have difficulties engaging with them (Hawkins et al, 2017).
There are several explanations for why people with COPD may score lower in these domains. One proposition that has been discussed in the literature (Hawkins et al, 2017) may be a lack of skill in communicating with health professionals, and this might reflect difficulties in communication and confidence when having conversations on complex subjects. A difficult or misinterpreted interaction with a health professional is not only unhelpful, it might also be damaging to future care or treatment decisions – research has suggested that rushed interactions can lead to people with COPD being less invested in their own self-care (Welch et al, 2021).
Hawkins et al (2017) found that clinicians in their study reported thinking that patients do not ask the ‘right’ questions in consultations, because they thought they were not always able to identify gaps in their knowledge. However, this suggests that the ‘right’ questions are those considered to be so by the clinician – it privileges health professional concerns over those of patients. In addition, the lack of social support reportedly experienced by patients with COPD (Stellefson et al, 2018) could influence how they communicate and engage with health professionals and this may subsequently be reflected in their HLQ scores for social/interactive domains. Welch et al (2021) support an approach focused more towards personal and peeracquired knowledge when considering effective COPD self-management: research has shown that patients' self-management strategies form more spontaneously in response to symptoms (Apps et al, 2014) than as a result of formal education from their health professional.
The present authors' study also highlights areas of the HLQ that scored relatively lower among the COPD participants in this survey, including ‘appraisal of health information’ and ‘navigating the health care system’. The low-scoring ‘appraisal of health information’ domain was in keeping with research in a non-COPD-specific population (Beauchamp et al, 2015; Jessup et al, 2017). However, the authors' study results differed with respect to the ‘ability to navigate the health care system’, which was a low-scoring area for this COPD population. This may be an important finding, because an inability to navigate the healthcare system may be associated with suboptimal use of care and/or self-management, especially as healthcare services become increasingly virtual in a post-pandemic NHS. Further research to understand how people with COPD navigate the healthcare system would be valuable, and including patients in co-design of future interventions would be key to improve services (Roberts et al, 2008).
The secondary findings of this study show a relationship between two of the HLQ domains and COPD severity. The first relationship highlights that participants with more severe COPD reported lower scores for ‘having sufficient information to manage health’. This may be due to patients with less severe COPD requiring less information to manage their condition. However, it may also be the case that low health literacy has led to a lower ability to manage health, consequently leading to poorer health outcomes.
The second relationship indicated that appraisal skills also improved with increased COPD severity. These findings resonate with Stellefson et al's (2018) study into e-health literacy; they also found that people with more severe COPD scored higher on a scale on appraising information than those with less severe COPD. Worsening COPD status may drive the need to improve appraisal skills. Alternatively, COPD status is also likely to be an indicator of length of time since diagnosis; as a result, any improvements in appraisal could be the result of having had longer to develop the skills. This should also be considered in the context of patient-centred strategies to improve and promote independence (Welch et al, 2021).
Limitations
Although this study provides insight into an under-researched group, it should be noted that it has several limitations. First, although the HLQ provides a more comprehensive assessment of health literacy than measures of functional health literacy alone, the clinically significant ranges have not yet been defined. As such, it is not possible to distinguish between those with high, adequate or low levels of health literacy, unlike some other albeit simpler health literacy measures (Murphy et al, 1993; Chew et al, 2004). To improve this in future studies, a summative measure that categorises health literacy in these terms could be used alongside the HLQ.
Research by Wahl et al (2021) shows the impact of psychological wellbeing, and the need to consider this when determining the health literacy needs of a COPD population. This was not measured in this study. Another limitation was the failure to reach the target sample size of n=80. However, the authors were still able to achieve a level of precision very close to the planned figure.
Another limitation is the self-reported nature of the MRC, which although it is a refection of patient-reported breathlessness, does not necessarily correlate with disease severity. Finally, although the sample provided a fair reflection of the UK COPD population in terms of age and sex, it was homogenous in relation to ethnicity, which may reduce the generalisability of the study. Due to a lack of translation services, the need for patients taking part in the study to be able to read and understand written English might have contributed towards this.
Conclusion
This study suggests that, on a population level, people with COPD may find appraising health information and navigating the healthcare system to be the most challenging domains. Further research is needed to understand which aspects specifically present challenges in these areas; this will inform the design of future interventions and/or services. The finding that, as disease severity increases, perception of having sufficient information decreases, perhaps highlights the need for research to understand the changing informational needs of people with COPD throughout the course of their condition. These findings are specifically relevant for health professionals who work with patients with COPD and should be considered in relation to information provision, patient engagement and service design.