Is anxiety a suitable measure of decision aid effectiveness: a systematic review?
Introduction
Decision aid interventions aim to encourage patients to engage in the decision making process when making treatment or screening choices by helping them to evaluate accurate information about all the decision options and their consequences in accord with personal values [1], [2]. Measures employed to evaluate the effectiveness of decision aids are varied and focus on assessments of behaviour, cognition and affect [3]. Although measures have been developed with reference to decision making theories that explicitly assess changes in health care decision processes [4], [5], [6], [7], most trials have employed outcome measures such as quality of life, anxiety and satisfaction when assessing decision aid effectiveness [1], [8]. Although these established outcome measures are suitable to assess health care interventions after patients have made a treatment choice, there is little evidence that such measures are associated with changes in patients’ effectiveness when making a choice. In consequence, their appropriateness to evaluate decision aid interventions is under debate [1], [8], [9].
One such established outcome measure currently employed to evaluate decision aid interventions is anxiety [1], [2]. Anxiety is defined as “an unpleasant emotional state or condition that is characterized by subjective feelings of tension, apprehension, and worry, and by activation or arousal of the autonomous system” [10]. Theoretically there is still some debate as to whether anxiety responses are normal reactions to stressful events or abnormal experience like an irrational fear [11]. Despite this lack of consensus, validated measures of patient anxiety have been used in clinical practice for a considerable time [12], [13], [14]. The main reason to assess anxiety has been to evaluate whether or not these levels are reduced in patients receiving treatment for anxiety disorders [15]. However, over the last 20 years, research into the impact of receiving treatment for physical illness suggests that many procedures are stressful to patients, impairing patient recovery and adherence to treatment regimens [16], [17]. Anxiety scores, then, have been used to assess the iatrogenic consequences associated with health care interventions and the effectiveness of information aids to ameliorate these effects. A third reason for assessing anxiety levels is in situations where patient’s level of arousal or emotion state impacts on the effectiveness of making treatment choices; some increase in arousal is associated with better recall and systematic evaluation of information whereas very low or high levels are related to less optimum processing strategies, i.e. the inverted U-shaped relationship between arousal and information processing ability [18], [19], [20], [21]. As yet, it is unclear what level of arousal is associated with effective decision making but it is possible that high anxiety levels are indicators of good decision making strategies.
It is feasible that a decision aid could be developed to facilitate the treatment of anxiety disorders; in which case, employing a measure of anxiety as the main outcome variable would be an appropriate assessment of the decision aid’s effectiveness. However, it is more likely that a measure of anxiety is employed in studies to assess the iatrogenic consequences, if any, of the decision aid in a non-psychiatric healthcare setting and/or to explore the relationship between arousal and effective decision making. This systematic review aims to integrate empirical studies employing a measure of anxiety in evaluations that assess decision aid effectiveness. The systematic review’s objectives are:
- •
To identify studies employing an anxiety measure in evaluations of decision aid effectiveness.
- •
To describe the measures employed and their application.
- •
To assess the impact of decision aids on patient anxiety.
- •
To explore the relationship between patient anxiety and decision making processes and outcome.
- •
To identify the most useful role for anxiety measures in future healthcare decision aid research.
Section snippets
Methods
This systematic review is a sub analysis of The Cochrane Systematic Review and An Inventory of Decision Aids for People Facing Health Treatment or Screening Decisions [22]. Therefore, the methods for this review are only described in brief here. Experts in the field of health care decision making and information management designed the search strategy. Medical and social science electronic databases were searched, behavioural decision making journals hand-searched, reference sections of
Results
From searching the electronic databases, 11,361 unique citations were identified and a further 26 from personal files and hand-searching. Of the 600 studies exploring healthcare decision making, 10 trials met the inclusion criteria of which eight were published (study numbers (SNs) 1–5, 7–9) and two unpublished (SNs 6, 10) (Table 1). Decisions were made in the following health contexts (Table 1): prenatal screening (n=2) and diagnosis (n=1); benign prostatic hypertrophy (n=1); hormone
Discussion
Both the STAI and HADS have been employed to assess the iatrogenic consequences of decision aids across a number of health contexts. The review found little consistency in the methods of assessing anxiety across studies with different time-points being used to assess affect. In addition, large attrition rates were observed at follow-up and appropriate analyses were not always employed. These methodological issues and varied decisions across several health contexts meant the data were too
Acknowledgements
We thank the authors of the Cochrane Review of Decision Aids for people facing health screening and treatment decisions: Annette O’Connor, David and Margaret Rovner, J. Tetroe, Hilary Llewellyn-Thomas, Vicky Entwistle, A. Rostom, V. Fiset, M. Barry, J. Jones.
References (36)
- et al.
Measuring the involvement of patients in shared decision-making: a systematic review of instruments
Patient Educ. Couns.
(2001) - et al.
Assessing the quality of randomized controlled trials: is blinding necessary
Control Clin. Trials
(1996) - et al.
Information and patient participation in screening for prostate cancer
Patient Educ. Couns.
(1999) - Bekker H, Thornton JG, Airey M, Connelly J, Hewison J, Lilleyman J, et al. Informed patient decision making: an...
- et al.
Decision aids for patients facing health treatment or screening decisions: a systematic review
Br. Med. J.
(1999) Patients’ health-care decision making: a framework for descriptive and experimental investigations
Med. Decis. Making
(1995)Validation of a decision conflict scale
Med. Decis. Making
(1995)- et al.
Patient satisfaction with health care decisions: the satisfaction with decision scale
Med. Decis. Making
(1996) - et al.
Patient decision making: an evaluation of two different methods of presenting information about a screening test
Br. J. Health Psychol.
(1997) - Bekker HL, Hewison J, Thornton JGT. Understanding why decision aids work: linking process with outcome. Patient Educ...