I read with great interest the article “The use of pretest probability increases the value of high-resolution CT in diagnosing usual interstitial pneumonia”, by Brownell and colleagues [1]. The study has great methodological strength, and I applaud the authors for such an elegant work. But what really caught my attention was the clear use of pre-test probability and likelihood ratio to establish the diagnosis of usual interstitial pneumonia (UIP) in patients with suspected UIP. I believe this article should change the way we care for those patients.

The study included patients with “possible UIP" and “inconsistent with UIP” patterns on high-resolution computed tomography (HRCT) of the chest. Those patients represent a diagnostic dilemma we commonly face in interstitial lung diseases clinical practice. Three different radiologists (two in the derivation and one in the validation cohort) reviewed the HRCT scans, and most importantly: they were blinded to clinical information and pathology results. All patients had the reference standard surgical lung biopsy, which were prospectively evaluated by expert pathologists.

The likelihood ratio for male patients, with ≥ 60 years-old, and possible UIP with traction bronchiectasis score ≥ 4 was as high as 47 in the derivation cohort. Since likelihood ratios are a ratio of two likelihoods (the likelihood of a test results in disease / the likelihood of the same test result in no disease [2]), the further away from one the better. So, what a likelihood of 47 means is that male patients with ≥ 60 years-old and possible UIP with traction bronchiectasis score ≥ 4 are 47 times more likely to have UIP versus not have UIP. This would move us from a pretest probability of IPF of 60% to a post-test probability of IPF of 98% in the derivation cohort, for example.

How to translate that into clinical practice? We have to start with an estimation of the pre-test probability of UIP in our patients. For that, it would have been helpful if the authors had used the variables gender and age to stratify the pre-test probability, similarly to the pre-test probability stratification we do in the evaluation for pulmonary embolism (with the Well’s criteria [3]) and lung cancer (with the Mayo Clinic prediction rule [4]). Then, we could apply the likelihood ratios of the HRCT findings to reach a post-test probability. It would be useful to know the likelihood ratio of a possible UIP with traction bronchiectasis score ≥ 4 encountered in the present study.

In conclusion, we will probably continue to struggle to make an accurate diagnosis of UIP in our patients with suspected UIP. But at least, this study allows us to struggle with some good data in our hands.

References
[1] Brownell R, Moua T, Henry TS, et al. The use of pretest probability increases the value of high-resolution CT in diagnosing usual interstitial pneumonia. Thorax. 2017;72:424–429.
[2] Richardson WS, Wilson MC, Keitz SA, et al. Tips for Teachers of Evidence-based Medicine : Making Sense of Diagnostic Test Results Using Likelihood Ratios When to Use This Tip When to Use This Tip. J Gen Intern Med. 2007;23:87–92.
[3] Chunilal SD, Simel DL. Does This Patient Have Pulmonary Embolism ? JAMA J. Am. Med. Assoc. 2014;290:2849–2858.
[4] Swensen SJ, Silverstein MD, Edell ES, et al. Solitary pulmonary nodules : Clinical prediction model versus physicians. Mayo Cllinic Proc. 1999;74:319–329.

The paper by Yoon et al [1] addressees an important subject - diabetes mellitus (DM) probably increases the risk of TB by a factor of three [2]. The authors present data showing an association of poorer diabetes control status with both the characteristics of pulmonary TB at presentation, and the response to treatment. Compared to patients with no or controlled DM, those with uncontrolled DM reported worse symptoms at presentation, were more likely to be sputum smear positive, and had more substantial radiographic changes. Patients with uncontrolled DM were also more likely to remain sputum culture positive at two months, and either fail treatment or die.

Although these observations are entirely consistent with a biologically plausible explanation that hyperglycaemia itself influences the development of TB and its response to treatment, there is an important confounding factor which may not have been fully accounted for: treatment adherence, and the wider general use of health care.

Patients with uncontrolled diabetes, by definition, are less well treated than those with controlled diabetes. Part of the reason for this will be treatment adherence. Such patients may also be less well engaged with health services. Hence a reason for more advanced TB disease at diagnosis in those with uncontrolled DM compared to controlled or no DM might be due to later presentation to health services. Indeed, a recent study in China reported that patients with hyperglycaemia are more likely to delay presenting to health services with symptoms of pulmonary TB [3].

And at least part of the reason for delayed sputum culture conversion in uncontrolled DM in the current study could be poorer adherence to TB treatment. Although this was not entirely borne out by the data, there was a suggestion of poorer treatment compliance in patients who remained culture positive at two months compared to those who did not (Table 3; p=0.22). It is not clear from the paper how TB treatment compliance was associated with DM control status. Neither is it clear precisely what methods were employed by the TB nurse in the study to assess treatment adherence.

So although the findings of Yoon et al support a direct effect of diabetes control status on TB presentation and treatment response, further work is required to exclude the potential confounding effects of delayed presentation of TB and adherence to treatment. Health records could be used to assess level of engagement with health services prior to the TB diagnosis, and directly-observed TB treatment in all study subjects could address potential compliance issues.

References:

1. Yoon YS, Jung JW, Jeon EJ et al. The effect of diabetes control status on treatment response in pulmonary tuberculosis: A prospective study. Thorax. 2017; 72: 263–70.

2. Jeon CY, Murray MB. Diabetes mellitus increases the risk of active tuberculosis: A systematic review of 13 observational studies. PLoS Med. 2008; 5: e152.

3. Wang Q, Ma A, Han X et al. Hyperglycemia is associated with increased risk of patient delay in pulmonary tuberculosis in rural areas. J Diabetes. 2017; 9: 648–655.

I read with interest the article published by Arbillaga-Etxarri et al.[1] titled “Socioenvironmental correlates of physical activity in patients with chronic obstructive pulmonary disease (COPD)”. In the introduction section, the authors stated that the current interventions (e.g., pharmacological treatment, rehabilitation, self-management) aiming to change physical activity behavior in COPD patients lack effectiveness, particularly in the long-term. The authors argue that this absence of effectiveness could be due to a lack of knowledge of physical activity determinants in this population. To address this issue, Arbillaga-Etxarri et al.[1] examined the socio-ecological determinants of active behaviours in 400 COPD patients and found that, after controlling potential confounders, having a dog and grandparenting were positively associated with physical activity; effects sizes were small, β = .19 and very small β = .08 for dog walking and grandparenting, respectively. The authors concluded that these two socio-environmental characteristics should be considered to promote physical activity both at the clinical level and in future research. This study is important because there is a lack of knowledge regarding the determinants of physical activity in this population.[2]

Nonetheless, there are some reservations regarding the clinical utility of physical activity socio-environmental correlates to design physical activity programs. Socioenvironmental correlates of physical activity are often largely beyond participants’ control and thus non-modifiable during an intervention for a clinician, such as the ‘walkability’ of the environment. For example, in the current study, only 18 % of the sample had a dog and 12% of the participants reported walking it. The authors explained that patients could have concerns about dogs such as difficulties with controlling the animal, which could explain the lack of dog owners in the study. It could also be hypothesized that patients have a dog because they are already active and feel capable of owning a dog. Nevertheless, it could be difficult to use this leverage during a physical activity counseling intervention.
Physical activity is a complex behavior that may depend from many different factors, such as socio-environmental variables (e.g. age, family support), biological functions (e.g. functional capacity), or environmental factors (e.g. climatic conditions; ‘walkability’ of the environment). Among these potential predictors of physical activity there is an urgent need to identify factors that could be modifiable during an intervention. In this regard, the concept of motivation is an important target.
Motivation could be operationalized through different theoretical constructs, which have received different levels of empirical support regarding their associations with physical activity behaviour depending on the context [3]. These constructs could be clustered into (1) intentional processes (the development of objectives and intentions to be active), (2) affective judgments (feelings about physical activity) (3) self-perception of capability and opportunity (confidences in one’s capacity to be active) (4) self-regulation processes ( strategies used to maintain motivation and dealing with barriers) and (5) automatic or unconscious processes (physical activity habits driven by feelings about physical activity). Methodologically, these processes are often measured through either self-reported questionnaires [4] or computerized reaction-time tests.[5] To date, research dealing with motivational processes toward physical activity for patients with chronic respiratory disease are scarce. Selzler et al [6] found that stronger physical activity-specific self-efficacy was positively associated with exercise attendance, as well as 6-minute walk test improvement during pulmonary rehabilitation. Chevance et al [7] highlighted that unconscious feelings (measuring with computerized test) about physical activity prospectively predict self-reported physical activity at 6 months after pulmonary rehabilitation. These preliminary results are important because identifying modifiable determinants of physical activity could help (i) to motivate patient to integrate a program, or identify patient at risk to failing physical activity after an intervention and (ii) to design more effective evidence-based interventions regarding physical activity in COPD patients. In conclusion, future studies should consider the motivational determinants of physical activity as well as interventions to specifically enhance motivation.

Acknowledgments : I sincerely thank Anne-Marie Selzler for their comments

References

1. Arbillaga-Etxarri A, Gimeno-Santos E, Barberan-Garcia A, et al. Socio-environmental correlates of physical activity in patients with chronic obstructive pulmonary disease (COPD). Thorax, forthcoming. doi: 10.1136/thoraxjnl-2016-209209.
2. Gimeno-Santos E, Frei A, Steurer-Stey C, et al. Determinants and outcomes of physical activity in patients with COPD: a systematic review. Thorax, 2014;69: 731-39. doi: 10.1136/thoraxjnl-2013-204763.
3. Rhodes R. The Evolving Understanding of Physical Activity Behavior: A Multiprocess Action Control Approach. Advances in Motivation Science, 2017;4:171-205. doi: 10.1016/bs.adms.2016.11.001.
4. Rodgers W, Wilson P, Hall C, et al. Evidence for a Multidimensional Self-Efficacy for Exercise Scale. Res Q Exerc Sport, 2008;79:222-34. doi:10.1080/02701367.2008.10599485.
5. Chevance G, Heraud N, Guerrieri A, et al. Measuring implicit attitudes toward physical activity and sedentary behavior: Test-retest reliability of three scoring algorithms of the Implicit Association Test and Single Category-Implicit Association Test. Psychology of Sport and Exercise, 2017;31:70-8. doi: 10.1016/j.psychsport.2017.04.007.
6. Selzler A, Rodgers W, Berry T, et al. The importance of exercise selfefficacy for clinical outcomes in pulmonary rehabilitation. Rehabil Psychol, 2016;61:380-8. doi: 10.1037/rep0000106
7. Chevance G, Héraud N, Varray A, et al. Change in Explicit and Implicit Motivation toward Physical Activity and Sedentary Behavior in Pulmonary Rehabilitation and Associations with Postrehabilitation Behaviors. Rehabil Psychol, forthcoming. doi: 10.1037/rep0000137.

We are grateful to the authors for their comments on the PEARL paper, especially those supporting our decision to assess outcome over 90 days. In regard to CODEX, most, but not all, patients had been hospitalised and, more importantly, death or readmission was not the primary outcome.1 Developed tools tend to be optimal for their primary outcome; a tool specifically designed to predict readmission/ death without readmission is likely to be a better predictor of this outcome than one that was not developed primarily for this purpose. This may, at least in part, explain the observed difference in performance. Prognostic tools should also undergo external validation. However, we acknowledge that the brevity of the abstract makes this unclear. At the editor’s discretion, we suggest the abstract could be amended to state: “no tool has been developed and externally validated…”

We agree that data about mortality alone is relevant, and highlight that this is included in table E3 in the online supplement. The optimal predictors of death and readmission are not identical, although there is overlap. The reasons for including readmission or death without readmission as a combined outcome are: 1) they are competing risks, and assessing readmission alone would mean that death without readmission would be categorised as a favourable outcome; 2) a patient who would otherwise have died at home may be readmitted if they are identified as high risk and appropriate services are put in place; and 3) hospital admission may prevent death. One way to analyse readmissions alone without including death as a favourable outcome would be to exclude those that died. However, this would bias the population by excluding those at higher risk of readmission. We plan to separately publish data on long-term predictors of death.

We acknowledge the difficulties diagnosing heart failure with preserved ejection fraction, previously termed diastolic heart failure, and its prevalence in this population. However, this does not carry the same mortality risk as left ventricular dysfunction. We separately assessed heart failure as a clinical diagnosis alone (without the need for evidence of reduced left ventricular function on echocardiography); this did not have the same predictive power as left ventricular failure. Consequently, left ventricular failure based on echocardiogram results was appropriately selected. We also highlight the European Society of Cardiology position: “Echocardiography is the most useful, widely available test in patients with suspected HF to establish the diagnosis.”2

The Charlson Index comprises of 19 indices and is a component of CODEX and LACE. The PEARL index includes only two measures of co-morbidity (individually shown to be strong predictors of outcome), and was superior to LACE in all three cohorts, and to CODEX in two of three cohorts. It is clear that the PEARL score is more parsimonious than scores containing the Charlson index, and therefore it is easier to score. Furthermore, it can be recalled and calculated at the bedside. Whilst we agree that ideally a full medical history should be performed in all patients, the more indices that appear in a score, the more likely that there will be missing data leading to biased estimates.

1. Almagro P, Soriano JB, Cabrera FJ, et al. Short- and medium-term prognosis in patients hospitalized for COPD exacerbation: the CODEX index. Chest 2014; 145(5): 972-80.
2. Ponikowski P, Voors AA, Anker SD, et al. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: The Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC. Eur J Heart Fail 2016; 18(8): 891-975.