Abstract
Introduction
The McCabe Cold Sensitivity Severity scale (CSS) is one of the most used questionnaires for measuring cold sensitivity in patients with hand injuries. The Potential Work Exposure scale (PWES) measures cold exposure at work. Good reliability and validity have been reported for the Swedish translations. No version in the Norwegian language has been available.
Methods
The CSS and PWES were translated forward and backward, and then tested for content validity in patients with hand injuries (n = 23 and 12, respectively). We categorised the CSS scores (0–400) into five severity levels, from none to extreme cold hypersensitivity, and asked each patient how the categories matched their perceived cold hypersensitivity levels.
Results
Good content validity was confirmed for the Norwegian version of the CSS to assess cold sensitivity in patients with hand injuries. Preliminary findings support the categorisation of scores into five severity levels.
Conclusions
The Norwegian CSS is an appropriate instrument to assess cold sensitivity after hand injury.
Keywords
Introduction
The McCabe Cold Sensitivity Severity scale (CSS) and the Potential Work Exposure scale (PWES) were developed in Canada by McCabe and colleagues, 1 to measure degree of cold sensitivity and cold exposure at work in patients with hand injuries. The situations described in the questionnaires were chosen based on interviews with patients who were hypersensitive to cold after hand injuries. There has not been any Norwegian translation of the CSS, and only the Swedish translation has been systematically validated. 2 The aim of the present study was to translate the CSS and PWES into Norwegian and pre-test the instruments in patients with hand injuries to examine the face and content validity of the questionnaires. We will also describe our experiences from using the questionnaires.
The questionnaires
The CSS consists of four everyday situations representing different intensities of cold stimuli (holding a frozen package from the freezer, holding a glass of ice water, wash in cold water, and get out of a hot shower or bathtub with the air at room temperature). The responses are given on 100-mm visual analogue scales (VAS) by placing an x somewhere along the lines, which are supplied with five adjectival descriptions: not at all (0 mm), mild (25 mm), moderate (50 mm), severe (75 mm) and extreme (100 mm). The summed score may vary between 0 and 400; higher scores indicate more severe cold sensitivity. The PWES consists of three situations related to work in cold environments (manipulation of objects near or below freezing, the amount of time working in cold environments and performance of work that cannot be done with gloves or mittens). The format is as for the CSS, giving total scores between 0 and 300 for the three items, The descriptions along the VAS are never, rarely, occasionally, usually and all the time. 1
Methods
Translation procedure
Face validity has been defined as ‘the degree to which a measurement instrument, indeed, looks as though it is an adequate reflection of the construct to be measured.’ 3 Face validity concerns an overall view, often based on the first impression. 3 Content validity concerns whether the measurement instrument adequately represents the construct under study and concerns both face validity and more detailed properties of the instrument. 3 The items should be relevant for the target population and the target cultural context. Thus, culturally adaptations of items may be necessary. 4 The developers of the CSS did not define the construct under measure. Because the CSS items represent different intensity of cold exposure, and high scores imply bother also at mild cold exposure, the CSS score may be understood as reflecting a cold threshold to elicit bother.
In 2004, the questionnaires were translated into Norwegian in four steps. 1) Two native Norwegian speaking persons independently translated the four CSS items and three PWES items from English to Norwegian and agreed upon a common version. Only one of the translators worked with patients with hand injuries and was familiar with the concept of cold hypersensitivity. 2) Two native Norwegians, who had been living in England and were in their final year of professional Norwegian-English translator-studies, independently translated the questionnaire back to English; none of them had knowledge of the target patient group or the concept. 3) A committee including forward translators, hand therapists and two bilingual persons with English as their mother tongue discussed the wording of the drafts. Content in the backward translations was found to be coherent with the original version. Some wordings in the Norwegian version were slightly adjusted to be more precise. 4) The Norwegian pre-final version was tested in patients with hand injuries and adjusted one final time.
Pre-test
Twenty-three patients treated for hand injuries gave informed written consent to fill in the translated questionnaires and be asked about their understanding of the questions and scoring. Immediately after answering the questionnaires, they underwent cognitive interviews, including how they interpreted the questions, the relevance of the questions and whether they had any difficulties when scoring the different items.3,5 The summed scores were calculated, categorised into five groups and labelled according to different levels of cold hypersensitivity. Scores below 50 were labelled as none; 50–149 = mild; 150–249 = moderate; 250–350 = severe; and 350–400 = extreme hypersensitivity.
In the Swedish version, the item ‘wash in cold water’ was translated into ‘contact with cold water.’ We chose to keep the wording ‘wash in cold water.’ In the pre-test we added the Swedish translation ‘contact with cold water’ to compare the CSS versions including ‘contact with’ and ‘wash in,’ respectively.
Results
Nineteen male and four female patients with a mean age of 40 (range 19–65) years, participated in the pre-test of the CSS. Thirteen had injured their dominant and 10 patients their non-dominant hand. Injuries included surgically treated finger fractures (n = 5), nerve lacerations in fingers (n = 8) or forearm (n = 4) and amputated fingers which had been replanted or revascularised (n = 6). Median time since injury was 15 months (range, 4–240 months).
The median CSS score for the group was 123 (interquartile range, 55–201; min–max, 0–350). When replacing ‘washing in cold water’ with ‘contact with cold water,’ the total score was somewhat lower (median = 100) and the dispersion was somewhat wider (interquartile range = 46–221).
The patients took less than 5 min to complete the questionnaires. The major concern for the patients were to decide how strong the symptoms should be to justify the term “bother”, and that the situations did not include a time factor. The “bother” would depend on how long they were in the situation, for example holding a glass of ice water. The patients chose to score the items according to the time they normally spent on the actual activity. Most patients referenced short-term contact with cold objects, but one patient used a reference of bringing the glass from the kitchen to the dining room, and another moving frozen food from one floor to another. Contrary to what we expected, the patients did not have problems in identifying activities where they washed their hands in cold water. Examples are ‘wash the car,’ ‘wash hands at the cabin (no access to warm water),’ ‘wash hands in the morning, when it takes some time for the warm water to reach the tap,’ ‘rinse hands in cold water to remove fibreglass from the skin after work including building insulation,’ ‘wash fruit and vegetables, because that is a part of my work,’ ‘rinse clothes’ and ‘wash the animals’ food trays at the farm.’ In general, they found that ‘contact with cold water’ represented more diverse situations than ‘wash in cold water’ according to thermal intensity and exposure time, which made it more difficult to think of a single situation. Some patients scored the item higher if the term ‘contact’ was used, but most of them gave ‘wash’ a higher score. Two of the patients commented that they might have scored the items involving water somewhat differently if there was only one of them.
Not all patients had performed all the activities in the CSS questionnaire after their injury, and some said that not all items were relevant, because they had injured their non-dominant hand or they were too bothered by cold to include their injured hand in the requested activities. However, they had no difficulties to imagine the discomfort in the described situations. One patient said that if he had received the questionnaires at home, he would have tried out the activities before responding.
In general, the patients felt that the items would better reflect their condition if addressing more severe cold exposure. However, when presented for their categorized severity score, only one of the patients disagreed with the interpretation. He was considerably cold-exposed in leisure activities where he had to work bare-handed and felt that he was more bothered than the actual description indicated. Twelve of the patients also completed the PWES and only one had questions about the items, asking, ‘How often is often?’ Median score for the PWES was 187 (interquartile range = 29–220).
Clinical experiences
In Norway, both the CSS and the PWES questionnaires have been used both in clinical practice and research since the pre-test. When used in clinical practice, some patients, especially those with mild cold sensitivity, feel that the situations in the CSS do not properly reflect their amount of bother. They are most bothered in humid or rainy weather or if they have to stay out for a longer time in temperatures around and below 15℃. To make the questions relevant to them, it may thus be wise to inform them that the questionnaires measure their cold threshold and not how bothered they are by activities in their everyday lives.
It is our impression that most of the patients could easily visualize being in the four CSS situations. However, some patients want to perform the actual tasks before responding. When we have used the CSS in research, some patients have commented that they actually have tried out the activities before scoring. Very few patients have left items blank, but we have experienced blank items where patients comment that they have not been exposed to that particular situation or are not able to perform the activity (e.g. grip a glass). Most patients have scored the shower item very low, holding a glass of ice water higher and either to wash in cold water or holding a frozen package from the freezer highest.
Discussion
The translation of the two questionnaires was performed systematically based on the recommendations of Guillemin et al., 4 but we used bilingual professional back translators with Norwegian, and not the original language, as their mother tongue. Nearly all the patients found the questions easy to understand and answer and based their responses on specific situations from their everyday lives. Results from the cognitive interviews support the categorisation of CSS scores into the five severity groups, which relates to the cut-offs of the modified CSS scores (0–4) as suggested by Koman et al. 6
The CSS may not be suitable for patients with injuries who have not been exposed to cold after the injury yet or patients with very poor or no grip function in the injured hand, such as patients with a tight thumb web-space or an amputated hand.
The Norwegian version has not been further validated, but in a 6 - to 10-year follow-up of patients with severe hand injuries, CSS showed the ability to discriminate between patients with replanted versus revascularised fingers and between employed versus unemployed patients.7,8 The Swedish translation has shown good test–retest reproducibility (ICC = 0.85; 95% CI 0.79, 0.90; n = 100), internal consistency (Cronbach’s alpha = 0.88; range of item-total correlation, 0.59–0.83; n = 122), and face validity. 2 The CSS also showed high correlations with a single cold-sensitivity question (r = 0.73), the DASH questionnaire (r = 0.67), and the bodily-pain subscale of the SF-36 (r = 0.59); n = 122. 2 Similarly, good reliability and internal consistency were found for the PWES. Other studies have demonstrated stable CSS scores 3–12 months after severe hand injuries, 9 or change in CSS score or in all of the single CSS items over a longer time period,10,11 or after treatment.12,13 Some of the researchers have modified the scores to give a total score from 0 to 4, 0 to 16 or 0 to 40.6,11,13
Currently, the CSS and the Blond McIndoe Cold Intolerance Symptom Severity (CISS) questionnaire2,14–16 are the most commonly used questionnaires to measure characteristics of cold sensitivity in patients with hand injuries. Whereas the CSS score may be interpreted as a score of cold threshold, the CISS score (score 4–100) includes both a cold threshold (the four CSS items), use of strategies to ease or prevent symptoms, consequences for activity and other features associated with cooling of the hand. We find the CSS score to be informative and feasible for screening and evaluating cold hypersensitivity. However, the single items of the CISS questionnaire give a broader picture of the patients’ condition and may add important information for individualized tailored interventions in the clinical setting,
Footnotes
Conclusions
Good content validity was confirmed for the Norwegian version of the CSS to assess cold sensitivity in patients with hand injuries. Preliminary findings support the categorisation of scores into five severity levels.
Acknowledgements
I wish to thank Eli Saastad for contributing to the forward translation and discussion of the translations, and Inger Holm for help with editing the manuscript.
Conflict of interests
None declared.
Funding
The Norwegian Society for Hand Therapy gave financial support to the backward translation.