Healthcare professionals and health service researchers increasingly understand the importance of providing information to patients for informed decision-making, to ensure they are more involved in care provision. The process whereby patients and healthcare professionals join the process of sharing ownership of a decision made is thought to improve therapeutic and treatment outcomes.1 There is an increasing appreciation that, by meeting a patient’s needs with respect to their involvement in the management of their disease, the risks associated with poor medicine management are minimised. Despite the moves towards greater patient involvement, a gap still exists between how much a doctor tells a patient and how much the patient wants to know. Much work has been done to assess the information needs of patients with cancer and has found that almost all patients wanted to know their diagnosis, and most wanted to know about prognosis, treatments options and side effects.2 However, while all patients want basic information on diagnosis and treatment, not all wanted further information at all stages of their illness,3 and some indeed preferred personalised clinical information.4 Studies have also identified that, despite the current trend for shared decision-making, information materials for patients often omit relevant data, often adopt a patronising tone and often do not meet the needs of patients.5 In addition, greater involvement in decision-making does not come about by simply providing more information; the desire for information is not the same as a desire to participate in decision-making.6

We have, over recent years, tried to assess the information needs of patients with respect to their prescribed drugs, and preliminary research found that the degree of patient empowerment is related to their intrinsic desire for information and their worries about changes to medicines. We found that patients who expressed a low degree of worry about changes and a high desire for information about their drugs seemed less worried and more empowered when given additional information. Conversely, those who expressed worries about changes in their medicines (supplies, doses, formulations, drugs) and did not want information about their medicines (were happy knowing little) seemed more worried and less empowered when given additional information about those changes.7 So, it seems that, not only is information “good” for some patients—ie, it increases empowerment in line with the shared decision-making trends—but information is “bad” for other patients—ie, it increases anxieties and distrust in therapies. At this stage, we concluded that further exploration of influences between information desires and needs and provision was needed to develop information strategies for patients.

Subsequent studies identified age as a predominant factor associated with patient desire for information, together with educational and socio-economic status.89 We hypothesised that this EID scale would have value in targeting receptive patients, or in identifying those who may be refractory to excessive drug information, and could be used to effectively target information provision based on evidence, rather than supposition Although, the EID scale has seemed a valid way of identifying information needs of patients, however, we wanted to explore if medicine information desires differed between patients with different diagnoses, before developing interventions and strategies to meet the information needs of patients.

AIMS AND OBJECTIVES

The overall aim was to explore differences in information needs of patients between different disease groups. The specific objectives were:

• to identify information needs of patients using validated measures;

• to explore differences in information needs between different disease groups of patients.

METHODS

We collected data over a 2-year period, using standardised data-collection tools and trained researchers. The tool used to identify the extent of information required by patients is a six-item scale, the EID scale, which had been previously developed and validated as described.7^–9 Patient characteristics and demographics were collected through structured and open questions. We conducted interviews with patients, either as inpatients during a hospital stay or as outpatients attending a specific clinic at the hospital. Patients were selected using convenience sampling: that is, patients who fulfilled the inclusion criteria were selected from the wards and clinics during the recruitment periods, as opposed to randomly or purposively selected. The criteria for eligible patients were that they spoke English, were not confused or with a diagnosis of mental illness, and were over 16 years of age. Data were coded and subjected to quality-assurance checks prior to analysis using SPSS (version 12.0). In addition, to ensure consistency in the data, a sample of data from each researcher was recoded and entered and tested for reliability.

The EID questions asked during the interviews are as follows:

1. I need as much information about my medicines as possible.

2. Too much knowledge is a bad thing.

3. You can never know enough about these things.

4. I don’t need any more knowledge.

5. I read about my medicines/illness as much as possible.

6. What you don’t know doesn’t hurt you.

Tests and measures of association included Kruskal–Wallis chi squared and ANOVA. The scores to the scale were correlated with continuous variables such as age (Pearson’s correlation coefficient). The scores were divided around the mean score into two broad categories: high scorers (those scoring higher than the mean EID score, who tended to want more information) and low scorers (those scoring lower than the mean, who tended to want less information). It is widely accepted that dividing scores around the mean or median provides merely a crude indication of trends but was regarded as a suitable method for such a descriptive analysis in this study.

RESULTS

Data from interviews with 1717 patients were included in the analysis. The mean age of the sample was 60 (SD 16.4) years. Demographic data, such as marital status, occupation, gender and ethnicity are summarised in table 1. Other variables such as school-leaving age and level of qualification were recorded as secondary indicators of socio-economic status. Occupation was either difficult to obtain from the patient or relative themselves, or difficult to categorise using the NSSC Classification, which goes part way to explain the high number of “not classifieds.”

The primary diagnosis for each patient was recoded into a broad disease category using the same coding process throughout to ensure consistency and allow for comparison (table 2). As expected from a predominantly general medical population, the largest disease category was cardiovascular disease, followed by gastrointestinal and respiratory. Other specialist categories reflected the diverse nature of the other recruiting centres. There were differences in mean ages between the disease categories, F_6,1707 = 27.7, p<0.001 (table 3). Patients diagnosed as having a CNS tumour tended to be the youngest recruits, and patients with cardiovascular and respiratory diagnoses tended to be the oldest recruits.

The median number of prescribed medicines was 5.0 for the entire sample (range 0 to 21). The median duration of illness was 1.75 years (range 0 to 85 years), and the median number of times in hospital was 2 (range 1 to 25). There was a significant difference in the number of drugs prescribed between the disease categories (Kruskal–Wallis χ² 225.75, p<0.001). Oncology patients tended to take fewer drugs than the other patient groups (table 3).

The six-item EID scale was assessed as a single primary construct using principal-components analysis (PCA) and oblique rotation, and proved consistent with our previous findings.89 Each item on the EID scale was scored on a Likert scale (from 1 to 5): the scale range was 6 to 30, with a midpoint of 18. The scale score distribution for the entire sample was normally distributed. The mean score for the entire sample was 19.3, slightly above the scale midpoint (SD 5.41), which may imply that the sample had a higher-than-average desire for information. The internal consistency of the scale in this sample was good (coefficient alpha = 0.78).

Scores to the EID scale correlated with age and socio-demographic variables. The extent of information desired correlated negatively with the age of the patient (Pearson’s r = −0.32, p<0.001), implying that medicine-information desires decreases with age. We found that the extent of information desired correlates positively with socio-economic status—for example occupation (Pearson’s r = 0.29, p<0.001); so, a higher socio-economic status is associated with a higher desire for drug information. This could also be linked to access to such information being higher with higher socio-economic status. These correlations were consistent within all disease categories, with the exception of oncology patients, where age was not a correlate with EID scores, with a comparatively different regression coefficient close to zero. It should be noted that the oncology sample was smaller compared with the other disease groups, and the possibility of sample variance contributing to this observation should not be discounted.

The mean scores to the EID scale were compared across disease categories (fig 1). Significant differences were found in the extent of information desired between these patient groups. For example, endocrine and diabetic patients expressed high desires for information, whereas cardiovascular and respiratory patients expressed low desires for information. The effect of disease category on EID scores remained significant when controlling for age (ANCOVA, F_6,1703 = 26.04, p<0.001, partial η2 0.084 (ie, 8.4% “effect size”), which implies that disease state is a principal influence on medicine information desires, in addition to the strong correlate for age of the patient.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Error bar chart displaying the differences in scores to the EID scale by disease category.

DISCUSSION

The main aim of this research was to explore differences in medicine information needs of different disease groups of patients, and we found significant differences in the extent of information desired between patient groups (independent of age), implying that the diagnosis is a major factor in affecting the desire for information. While we found that scores to the EID scale correlated negatively with age (medicine information desires decreases with age) and positively with socio-demographic variables (higher socio-economic status is associated with a higher desire for drug information), these correlations were consistent within all disease categories, with the exception of oncology patients, which could be in part due to the narrower age range and smaller sample. Other studies have also found that age is a predominant factor associated with patient desire for information; however, this could be confounded by length of diagnosis.89 For example, a patient who has been diagnosed as having an illness for a significant period may require less information about their condition or prescribed drugs than someone who has been newly diagnosed.

In this study, patients were recruited using convenience sampling, which does not allow for robust hypothesis testing but was appropriate for this descriptive study, to describe the emergent relationships between demographics and desires for medicines information and to generate hypotheses. It was thought that this EID scale (extent of information desired) may have value in targeting receptive patients, or in identifying those who may be refractory to drug information and could be used to effectively target information provision based on evidence, rather than by supposition.89 There were, however, difficulties associated with classifying the socio-economic data: the entire sample could not be analysed, which could have underestimated the effects of social class and status on the information desired. In addition, resource constraints mean that the sample was limited to English-speaking patients only. We aim to conduct further work with non-English-speaking patients, as we hypothesise that their medicine-information desires may additionally be affected by language difficulties.

Because of the cross-sectional nature of the study, the stability or indeed sensitivity of the EID scores could not be tested over time. It could be that the desires for medicines information vary over time and may require different information provision during the progression of the disease, which requires further study to explore the effect of time and changing severity of symptoms on desires for information. There was heterogeneity within the disease categories: not only were the categories broad, but there were many disease states with varying levels of severity in each category, often dependent upon the duration of the disease. The results provide broad findings linked to the main diagnostic group, but further work should be done to evaluate the different desires for information within each disease group to uncover the potential influence of more than one diagnosis. It may be that more than one tool will be required in the future, to explore information desires of those with a single diagnosis, multiple diagnoses and chronic progressive disease. At present, the scale does not differentiate patients who have more than one diagnosis which may well exist. The influence of such factors on patients’ desires for information about medicines requires further study.

Other research in this area has found that patients may vary in their desire for involvement in decision-making, recommending that doctors need the skills, knowledge of their patients and the time to determine which patients, with which illnesses, wish to be involved.10 Over recent years, there has been a growing interest in providing information to support patients’ participation in choosing treatments and deciding on strategies for managing health problems. Concern has been expressed that failure to pay attention to the quality of information obtained by patients has serious consequences, leading to a national public health information strategy which recognises the advantages of raising standards of information provision and risks of not doing so.11

It is important for healthcare professionals to identify and understand that patients with different disease have different desires for information about their disease and their drugs which may influence the way they take their medicines and subsequently the ways we manage their long-term disease. The key implications of our findings are congruent with other research evaluating the information patients desire; that we should no longer assume all patients want the same information or indeed, that same amount of medicine information, because there is variation both within (eg, age) and between disease categories for medicine-information desires. This has important implications for professionals who may fear the move towards increased information provision for all which could result in simply dispensing the same level or quantity of information to patients. Evidence now suggests that we must take into account the patient’s age, socio-economic status and their disease, but how can we identify these medicine information needs?

We suggest that the diagnosis and disease have a significant bearing on the psychology of the patient with regard to their medicine-information desires and seek to assess how healthcare professionals can view patients as individuals when providing medicine information that meets their needs within the time and financial constraints they face. While the conclusion that people with different conditions will have different informational needs concurs with previous research, the identification of medicines information needs and subsequently addressing those needs to improve medicines taking behaviours is an appropriate way forward. We need further research to determine if the use of the EID scale is an efficient and effective way to identify patients’ medicine information desires and provides a useful tool for practitioners to effectively target interventions in healthcare provision.