Questions about prognosis: examples of appraisals from different health professions

CHAPTER 9 Questions about prognosis

examples of appraisals from different health professions

Tammy Hoffmann, John W. Bennett, Marilyn Baird, Mark R. Elkins, Craig Lockwood, Sheena Reilly, Sharon Sanders

This chapter is an accompaniment to the previous chapter (Chapter 8) where the steps involved in answering a clinical question about prognosis were explained. In order to further help you learn how to deal with prognostic clinical questions when they arise and appraise the evidence, this chapter contains a number of worked examples of questions about prognosis from a range of health professions. The worked examples in this chapter follow the same format as the examples that are in Chapter 5. In addition, as with the worked examples that were written for Chapter 5, the authors of the worked examples in this chapter were asked not to choose a systematic review (for the reason that was explained in Chapter 5), but to instead find the next best available level of evidence to answer the prognostic question that was generated from the clinical scenario.

Occupational therapy example

Clinical scenario

You are an occupational therapist who has recently rotated into the outpatient department of a brain injury rehabilitation unit. After clients are discharged as inpatients and return home to live, they return to the hospital to receive outpatient therapy. One of your clients is Mary, a 20-year-old who sustained a mild traumatic brain injury in a motor vehicle accident 8 weeks ago. Her Glasgow Coma Scale was 13 on admission to hospital. Prior to the accident Mary worked full-time as an apprentice chef. She has been working in this job for 2 years and started after finishing high school. After discharge from hospital, Mary returned home to live with her parents and has been able to do simple tasks around the home and shopping with assistance from her mother. Your initial outpatient assessment of Mary revealed that her current difficulties are poor endurance, poor standing balance and memory and planning problems. Mary is already 2 months post-injury and her employer has just advised her that he is only able to keep her position in the apprenticeship scheme reserved for her for another 4 months. Mary has asked you how likely it is that she will be able to return to her job within that time frame. As you are new to working in this area of clinical practice, you decide to search the literature to help you answer this question.

Clinical question

In adults with mild traumatic brain injury, what is the likelihood of returning to work within 6 months of the injury occurring?

Search terms and databases used to find the evidence

Database: PubMed—Clinical Queries (with ‘prognosis category’ and ‘narrow search’ selected)

Search terms: (mild traumatic brain injury) AND (return to work^∗)

This search retrieves eight results. After reading through the titles of these articles, only one article appears to be relevant to your clinical question. After reading its abstract, you confirm that it is relevant and you obtain the full text of it so that you can appraise it.

Article chosen

Ruffolo C, Friendland J, Dawson D et al. Mild traumatic brain injury from motor vehicle accidents: factors associated with return to work. Arch Phys Med Rehabil 1999; 80:392–398.

Structured abstract

Study design: Inception cohort study (part of a larger case-control prospective study).

Setting: Tertiary care centre, Toronto, Canada.

Participants: 63 patients with mild traumatic brain injury who were admitted to hospital as a result of a motor vehicle accident; aged between 19 and 65 years (mean age 31.3 years, 62% male). Other eligibility criteria were: working before the accident in paid or unpaid employment; no history of head injury, neurological disease or hospitalisation for psychiatric illness; no severe disfigurement, amputation or spinal cord injury from the accident; and English-speaking.

Outcome: Return to work (at premorbid or modified level).

Prognostic factors studied: Injury severity, cognitive functioning at initial assessment and follow-up, social interaction at follow-up, discharge disposition and sociodemographic factors (age, gender, marital status, education and occupation).

Follow-up period: 6–9 months (mean = 7.4 months).

Main results: 42% of the participants who were followed up returned to work (12% to their premorbid employment and 30% to modified employment). Between the participants who returned to work and those who did not, there were statistically significant differences in levels of social interaction, premorbid occupation and discharge disposition.

Conclusion: Prognostic factors that were found to be positively related to return to work in people with mild traumatic brain injury were social interaction, jobs with greater decision-making freedom and discharge home. Cognitive impairment within the first month of the injury was not found to be a reliable indicator of likelihood of return to work.

Is the evidence likely to be biased?

• Was there a representative and well-defined sample?

Yes (well-defined population); cannot tell (representative sampling). The article does not specifically state that ‘all’ or ‘consecutive’ eligible participants who were admitted to the hospital were recruited into the study. The article mentions that this study was part of a larger case-control prospective study but no further details about it or its methodology are provided. However, the target population is well-defined as clear inclusion and exclusion criteria are provided.

• Was there an inception cohort?

Yes. Participants were recruited at a similar well-described point, namely after they were admitted to hospital after the traumatic brain injury occurred.

• Was exposure determined accurately?

Yes. In this study, it would have been very unlikely that an error was made in recruiting participants who had suffered a traumatic brain injury as a result of a motor vehicle accident as this exposure (or in this case, eligibility criteria) can be easily determined. The authors of the article also explain how the severity of participants’ traumatic brain injury was determined (mild traumatic brain injury was defined as having a Glascow Coma Scale admission score of 13 or more and/or a loss of consciousness of less than 1 hour).

• Were the outcomes measured accurately?

Yes. The main outcome was clearly defined as return to work and this was further categorised into return to premorbid employment or premorbid employment under modified conditions (which were also defined). The information about this outcome measure was gathered from components of three different questionnaires. It was not possible to measure this outcome in a blind fashion as it was based on participant self-report. It was not possible to blind assessors about participants’ status on each of these prognostic factors as these factors were either difficult to conceal from an assessor (for example, age, gender) or were based on participant self-report where the participant was technically the assessor.

• Were important prognostic factors considered?

Yes. The authors appear to have identified the major factors that could influence the outcome of return to work. The study examined five main categories of possible prognostic factors (refer to the structured abstract).

• Was the follow-up of participants sufficiently long and complete?

No, not sufficiently complete. The study recruited 63 participants and was able to follow up 50 of them (79% follow-up rate). Reasons were provided for why participants were lost to follow-up and participants who left the study were compared to those who remained in the study. Compared with participants in the group that were followed up, more participants in the group that were lost to follow-up experienced loss of consciousness at the time of injury and had jobs that had little independence and opportunity for decision making. With regards to length, the mean length of follow-up time was 6 to 9 months (mean of 7.4 months), which matches our clinical question. However, many studies that look at return to work after head injury follow participants for 12 months. Although only participants with mild traumatic brain injury were included in this study, they may have experienced impairments (either physical and/or cognitive impairments) that could continue to impact on their ability to return to work beyond 6 to 9 months post-injury.

What are the main results?

Return to work: Of the 50 participants who were able to be followed up, 21 (42%) had returned to work by the time of the follow-up assessment (12% to their premorbid employment and 30% to modified employment). You calculate the 95% confidence interval to be 32% to 52% (using the formula that was provided in Chapter 8, where 95% confidence interval = risk ± [1 ÷ √ 2n]). In other words, the likelihood of having returned to work by approximately 7 months post-traumatic brain injury could be as low as 32% or as high as 52%.

Prognostic factors related to return to work: Social interaction scores were significantly higher (p values 0.01 and 0.02) in the return to work group of participants compared to the group of participants who did not return to work. Significantly more (p <0.05) of the participants who were discharged to a rehabilitation facility, as opposed to being discharged home, had not returned to work by the time of the follow-up assessment. Participants were also significantly more likely to return to work if their job involved higher levels of independence/decision-making latitude. There were no significant differences between the two groups for any of the other factors that were analysed as potential prognostic factors.

How might we use this evidence to inform practice?

As you have determined the internal validity of this study to be reasonably strong (although you keep in mind the potential bias from the incomplete follow-up) and the results useful, you proceed to assessing the applicability of the evidence by comparing your client with the participants in the study, before deciding if you can use the evidence to help inform your practice. Mary’s mechanism and severity of injury is the same as participants in the study. She is younger than the mean age of study participants but meets the eligibility criterion for age and all of the other eligibility criteria of the study. In terms of the prognostic factors that were identified in this study, Mary was discharged home which, in the study, was found to be positively related to return to work, so this may increase the likelihood of her returning to work. However, you do not have much information about the other prognostic factors that were identified as being related to return to work, so you decide that you will assess her social interaction (using the measures used in the study) and find out about the extent of decision-making latitude/independence of her job.

After you obtain this information, you will use the results of this study to inform your discussion with Mary about the likelihood of her being able to return to work within the next 4 months. As a large proportion of the study participants who returned to work returned to modified work, there is a possibility that this may also be the case for Mary. As part of your treatment planning, you will arrange a time for Mary and yourself to meet with her employer to discuss the option of her returning to work in a modified capacity (such as shorter hours, different duties, graded return to work etc) if this is necessary. During this meeting, you also plan to obtain more detailed information about Mary’s duties at work and then use this, in conjunction with Mary, to set her rehabilitation program and goals.

Physiotherapy example

Clinical scenario

While working in your private physiotherapy practice, a 32-year-old baker presents 2 days after the onset of his first episode of acute low back pain. His pain is localised to his lumbar spine and is not accompanied by signs of more serious pathology such as weight loss, weakness or paraesthesia. However, it is currently preventing him from working and he urgently needs to arrange a replacement until he is able to work again. He is therefore primarily interested in any advice you can give him with regard to when he can expect to improve enough to return to work. He is also interested to know how quickly his pain is likely to improve with usual intervention.

Clinical question

In adults with acute low back pain, what is the average time to return to work and to resolution of pain?

Search terms and databases used to find the evidence

Database: PubMed—Clinical Queries (with ‘prognosis category’ and ‘narrow search’ selected)

Search terms: (acute OR recent) AND (low back pain) AND (return to work^∗)

The search returns 31 records. You scan the titles of the articles and read the abstracts of three studies that seem like they could be relevant. Of these, one is highly relevant as it is an inception cohort study of acute low back pain, it was conducted in the same healthcare setting as your work, it followed a large cohort and it measured the outcomes that are of primary concern to your client.

Article chosen

Henschke N, Maher C, Refshauge K et al. Prognosis in patients with recent onset low back pain in Australian primary care: inception cohort study. BMJ 2008; 377:a171. doi: 10.1136/bmj.a171.

Structured abstract

Study design: Inception cohort study.

Setting: Medical, physiotherapy and chiropractic practices in Sydney, Australia.

Participants: 973 participants (mean age 43.3 years; 54.8% male) aged at least 14 years with non-specific low back pain of less than 2 weeks’ duration. Exclusion criteria included radiculopathy, cancer, spinal fracture, spinal infection and inflammatory arthritis.

Outcomes: Time to return to work (determined by self-report of returning to previous work status), time to complete resolution of pain and return to function.

Prognostic factors studied: Age, gender, intensity of low back pain and level of interference with function at baseline plus individual variables grouped into seven factors—current history, past history, features of serious spinal pathology, sociodemographics, general health, psychological health and work.

Follow-up period: 1 year (with assessments also at baseline, 6 weeks and 3 months).

Main results: Median time to return to work was 14 days (95% confidence interval [CI] 11 to 17 days). Resolution of pain was much slower, at a median of 58 days (95% CI 52 to 63 days). A reasonable proportion of patients still had unresolved problems at 1 year. The cumulative probability of having returned to previous work status at 1 year was 89% and the cumulative probability of being pain-free at 1 year was 72%.

Conclusion: Prognosis of participants was not as favourable as is claimed in clinical guidelines. Most participants experienced slow recovery and almost one-third had not recovered from the presenting episode by the 12-month follow-up.

Is the evidence likely to be biased?

• Was there a representative and well-defined sample?

Yes. The article states that consecutive participants who presented to primary care with recent onset low back pain were invited to participate. Also, the target population is well-defined as clear inclusion and exclusion criteria are provided.

• Was there an inception cohort?

Yes. Participants were recruited within 2 weeks of onset of their condition, which minimises possible bias due to variable disease duration within the cohort.

• Was exposure determined accurately?

Yes. Eligibility was determined by doctors, physiotherapists and chiropractors, so misclassification is unlikely unless participants misreported their symptoms.

• Were the outcomes measured accurately?

Cannot tell. The outcomes (pain intensity, disability and work status) were based on participant self-report. Therefore, it was not possible to measure the outcomes in a blind fashion. Also, there is little indication of psychometric validation of any of the outcome measures, which may be one source of potential bias.

• Were important prognostic factors considered?

Yes. The researchers identified an extensive range of prognostic factors (refer to the structured abstract).

• Was the follow-up of participants sufficiently long and complete?

Yes. The follow-up rate was excellent (97%) and the length of follow-up appears to have been sufficient for the outcomes that were measured.

What are the results?

Return to work: The median time to return to work (to previous work hours and duties) was 14 days (95% CI 11 to 17 days). The cumulative probability of having returned to previous work status was 74.6% at 6 weeks, 83.2% at 12 weeks and 89.5% at 1 year.

Resolution of pain: The median time to resolution of pain was 58 days (95% CI 53 to 63 days). The cumulative probability of being pain-free was 39.9% at 6 weeks, 58.2% at 12 weeks and 72.5% at 1 year.

Complete recovery: This was measured by recovery on all three dimensions (return to work, no disability and no pain). The cumulative probability of being fully recovered was 39.0% at 6 weeks, 57.4% at 12 weeks and 71.8% at 1 year.

Prognostic factors: Factors associated with a longer time to recovery included older age, compensation cases, higher pain intensity, longer duration of low back pain before consultation, more days of reduced activity because of lower back pain before consultation, feelings of depression and a perceived risk of persistence.

How might we use this evidence to inform practice?

You compare your client’s characteristics to those of the participants in the sample and decide that he is similar enough that you can apply this evidence to him and his situation. As this prognostic evidence has little risk of potential bias and the confidence interval extends only a few days either side of the estimate of median time to return to work, you can confidently reassure your client that about 50% of people with acute low back pain return to work at about 2 weeks. Resolution of his pain is likely to take substantially longer, with 50% of people being pain-free by about 8 weeks. However, you also explain to your client that there is a small risk that the condition will not have fully resolved by 1 year, with 28% of the study participants not considered as fully recovered by 1 year. Because he is young, self-employed and has only had the pain for 2 days, however, he does not have several of the prognostic factors that were found to predict a longer time to recovery. Therefore, it is likely that he may have a better than average prognosis for recovery. You clearly explain all of these findings to your client.

Podiatry example

Clinical scenario

You are a podiatry student on clinical placement in a hospital podiatry department. You have just seen a 62-year-old man who attended with his wife. He has been referred by the diabetes educator for foot screening, education and management. He has had type 2 diabetes for approximately 10 years and, according to the client, his control has been good. Your examination indicates that neuropathy is present, vascular assessment is satisfactory, there is some foot deformity and there are active pressure lesions present plantar to the first metatarsal head bilaterally. Debridement of the callosity on the right foot reveals an ulcer. Together with your supervisor, you conduct a detailed ulcer assessment. The ulceration has white, macerated margins, the base is clean and pink to red and there is no exudate. The ulcer is round, 6mm across and 4mm deep. There do not appear to be any sinuses and you cannot probe to bone. While discussing an intervention plan with your client, his wife, who appears visibly worried, asks if he will end up losing his leg. You know that this is a possibility but are not sure of the actual risk, so your supervisor suggests you conduct a search to see if you can find some evidence to guide your answer to the question. You tell your client and his wife that you will endeavour to answer her question more accurately when you see them again in a few days time.

Clinical question

In a person with a neuropathic foot ulcer, what is the risk of amputation?

Search terms and databases used to find the evidence

Database: PubMed—Clinical Queries (with ‘prognosis category’ selected)

Search terms: ulcer^∗ AND (diabetic OR neuropathic) AND amputat^∗

This search returns 90 results and you quickly scan the titles. There are several articles which, on the basis of the title and abstract, look as though they will answer your question. They are all cohort studies that followed up participants with neuropathic ulcers. It is difficult to compare the studies in terms of the methods used by reading only the abstract, but one article stands out because it appears to include participants very similar to your client. You obtain the full text of the following cohort study of clients with diabetes who have their first foot ulcer.

Only gold members can continue reading. Log In or Register to continue

Related

Stay updated, free articles. Join our Telegram channel

Tags: Evidence-Based Practice Across the Health Professions

Mar 21, 2017 | Posted by admin in MEDICAL ASSISSTANT | Comments Off

Full access? Get Clinical Tree

Get Clinical Tree app for offline access

Get Clinical Tree app for offline access