Key points

Assessment and outcomes

Historically, dermatologists and others looking after patients with psoriasis have tended to record response to treatment, if at all, with rather imprecise phrases such as “nearly clear,” “a bit better,” “slightly improved,” “worse,” or “flared up.” This probably still holds true for the majority of consultations between psoriasis patients and health care professionals. If they have instituted a new therapy, there is almost certainly a tendency for them to write “slightly better” rather than “no change,” even if there is no clear evidence of meaningful benefit: Such wishful thinking is understandable, but can lead to long delays in changing to more appropriate therapy. Furthermore, the views of the patient may either not be sought or alternatively be dismissed as insignificant. Until recently, it has been rare for formal assessments of severity to be undertaken outside the setting of clinical trials. Doctors managing hypertension would expect to get their patients’ blood pressure checked on a regular basis. It should be a routine for at least some form of formal assessment of psoriasis severity and impact to be recorded on a regular basis for all patients receiving active treatment for psoriasis. The situation is slowly changing for the better, largely as a result of the cost implications of instituting expensive new agents for psoriasis and the need to demonstrate that they are producing benefit. Guidance is available from a range of specialist societies, patient organizations, and national health care bodies.

Psoriasis is a disease with multiple dimensions, each of which can contribute in a range of different ways to its overall impact on the individual. To be able to demonstrate objectively that any intervention for psoriasis can successfully modify that impact, it is necessary firstly to have tools for capturing and measuring that impact meaningfully and reliably (severity assessment) and second to understand what any given changes in such assessments actually mean in terms of modifying that impact. Only then can a meaningful assessment of the outcome of that intervention be derived.

Measuring change without reference to baseline severity (eg, “worse,” “no change,” “better”) is little different from the traditional approach used by doctors in routine practice. In a chronic condition such as psoriasis, such assessments are of limited value for charting an individual patient’s long-term disease behavior, because recall of fluctuations in disease severity over time is unlikely to be reliable. Neither are they useful for evaluating outcomes across a cohort of patients with unknown and potentially widely varying initial disease severity, as in a clinical trial comparing different interventions. Severity assessment at least 2 time points is a prerequisite for adequate documentation of change and thus for assessing outcomes.

The difference between severity assessment and outcome assessment can be illustrated clearly using the best known instrument for assessing psoriasis, the Psoriasis Area and Severity Index (PASI) : The PASI assesses severity whereas a 75% reduction in PASI score (PASI-75) assesses outcome. Unfortunately, the term “outcome” is all too often used indiscriminately to describe both types, particularly in relation to so-called patient-reported outcome measures (PROMs). For instance, NHS England (The UK National Health Service as it applies to England) states: “PROMs measure a patient’s health status or health-related quality of life at a single point in time.” In similar vein, the US Food and Drug Administration states: “A PRO (patient-reported outcome) is any report of the status of a patient’s health condition that comes directly from the patient, without interpretation of the patient’s response by a clinician or anyone else. The outcome can be measured in absolute terms (eg, severity of a symptom, sign, or state of a disease) or as a change from a previous measure.” Outcome can be assessed only by examining change, whether the desired outcome be change, as in interventions to treat disease, or no change, as in interventions intended to halt disease progression.

In many fields of medicine, outcome is straightforward to assess. Where there are well understood and easily measurable risk factors for adverse health outcomes, such as hypertension or hyperglycemia, it is straightforward to define a successful outcome as a change of the parameter in question from abnormal/unacceptable to normal/acceptable. Thus, the outcomes of interventions to reduce risk of developing overt type II diabetes in individuals found to have high glycosylated hemoglobin levels (hemoglobin A1 _c ≥6.5%) can be assessed by measuring whether the intervention has resulted in change to levels (eg, hemoglobin A1 _c ≤6.0%) known to confer a lower risk.

With many inflammatory or mental health conditions, however, it is not possible to assess change with such simple means. In disorders such as psoriasis and arthritis, there is a complex interplay between the externally apparent manifestations of the disease, the symptoms experienced by the patient, and the gamut of possible further physical, social, and psychological consequences of them. Furthermore, the latter are not necessarily directly related to the objective severity of the condition. The medical profession has been rather slow to recognize this complexity, but over the past 20 years significant progress has been made. In fact, the new discipline of clinimetrics has grown up around developing and validating disease severity assessments and outcome measures. This topic is well reviewed by Fava and colleagues.

Assessment and outcomes

Historically, dermatologists and others looking after patients with psoriasis have tended to record response to treatment, if at all, with rather imprecise phrases such as “nearly clear,” “a bit better,” “slightly improved,” “worse,” or “flared up.” This probably still holds true for the majority of consultations between psoriasis patients and health care professionals. If they have instituted a new therapy, there is almost certainly a tendency for them to write “slightly better” rather than “no change,” even if there is no clear evidence of meaningful benefit: Such wishful thinking is understandable, but can lead to long delays in changing to more appropriate therapy. Furthermore, the views of the patient may either not be sought or alternatively be dismissed as insignificant. Until recently, it has been rare for formal assessments of severity to be undertaken outside the setting of clinical trials. Doctors managing hypertension would expect to get their patients’ blood pressure checked on a regular basis. It should be a routine for at least some form of formal assessment of psoriasis severity and impact to be recorded on a regular basis for all patients receiving active treatment for psoriasis. The situation is slowly changing for the better, largely as a result of the cost implications of instituting expensive new agents for psoriasis and the need to demonstrate that they are producing benefit. Guidance is available from a range of specialist societies, patient organizations, and national health care bodies.

Psoriasis is a disease with multiple dimensions, each of which can contribute in a range of different ways to its overall impact on the individual. To be able to demonstrate objectively that any intervention for psoriasis can successfully modify that impact, it is necessary firstly to have tools for capturing and measuring that impact meaningfully and reliably (severity assessment) and second to understand what any given changes in such assessments actually mean in terms of modifying that impact. Only then can a meaningful assessment of the outcome of that intervention be derived.

Measuring change without reference to baseline severity (eg, “worse,” “no change,” “better”) is little different from the traditional approach used by doctors in routine practice. In a chronic condition such as psoriasis, such assessments are of limited value for charting an individual patient’s long-term disease behavior, because recall of fluctuations in disease severity over time is unlikely to be reliable. Neither are they useful for evaluating outcomes across a cohort of patients with unknown and potentially widely varying initial disease severity, as in a clinical trial comparing different interventions. Severity assessment at least 2 time points is a prerequisite for adequate documentation of change and thus for assessing outcomes.

The difference between severity assessment and outcome assessment can be illustrated clearly using the best known instrument for assessing psoriasis, the Psoriasis Area and Severity Index (PASI) : The PASI assesses severity whereas a 75% reduction in PASI score (PASI-75) assesses outcome. Unfortunately, the term “outcome” is all too often used indiscriminately to describe both types, particularly in relation to so-called patient-reported outcome measures (PROMs). For instance, NHS England (The UK National Health Service as it applies to England) states: “PROMs measure a patient’s health status or health-related quality of life at a single point in time.” In similar vein, the US Food and Drug Administration states: “A PRO (patient-reported outcome) is any report of the status of a patient’s health condition that comes directly from the patient, without interpretation of the patient’s response by a clinician or anyone else. The outcome can be measured in absolute terms (eg, severity of a symptom, sign, or state of a disease) or as a change from a previous measure.” Outcome can be assessed only by examining change, whether the desired outcome be change, as in interventions to treat disease, or no change, as in interventions intended to halt disease progression.

In many fields of medicine, outcome is straightforward to assess. Where there are well understood and easily measurable risk factors for adverse health outcomes, such as hypertension or hyperglycemia, it is straightforward to define a successful outcome as a change of the parameter in question from abnormal/unacceptable to normal/acceptable. Thus, the outcomes of interventions to reduce risk of developing overt type II diabetes in individuals found to have high glycosylated hemoglobin levels (hemoglobin A1 _c ≥6.5%) can be assessed by measuring whether the intervention has resulted in change to levels (eg, hemoglobin A1 _c ≤6.0%) known to confer a lower risk.

With many inflammatory or mental health conditions, however, it is not possible to assess change with such simple means. In disorders such as psoriasis and arthritis, there is a complex interplay between the externally apparent manifestations of the disease, the symptoms experienced by the patient, and the gamut of possible further physical, social, and psychological consequences of them. Furthermore, the latter are not necessarily directly related to the objective severity of the condition. The medical profession has been rather slow to recognize this complexity, but over the past 20 years significant progress has been made. In fact, the new discipline of clinimetrics has grown up around developing and validating disease severity assessments and outcome measures. This topic is well reviewed by Fava and colleagues.

Psoriasis assessment tools: a historical perspective

For the current generation of dermatologists brought up to consider randomized, controlled trials as the norm for investigating new therapies for skin disease, it is instructive to look back a few decades. Until the advent of potent topical corticosteroids in the late 1950s, very few comparative trials in the field of psoriasis were conducted. The mainstays of treatment up until then had been tar, anthralin (dithranol), and broadband UVB phototherapy. At that time, there was no accepted methodology for performing clinical trials in inflammatory skin disease. Systemic therapy was largely limited to arsenic: Methotrexate was first investigated for treating psoriasis in the 1950s, but it was not until 2003 that this use of the drug was subjected to a randomized, controlled trial.

A study selected at random from among the small number of formal psoriasis trials conducted in the 1960s exemplifies how much has changed. It is clear that the investigators thought carefully how to design their study comparing 2 topical corticosteroid preparations with topical tar. Looked at from our perspective, however, it seems crude, with small patient numbers entered into an unblinded, unrandomized within-patient, left-right comparison of the 2 corticosteroids, which are then also compared with the patient’s historical response to tar. What the investigators did do, however, was make an attempt to define what they meant by each of their outcome categories ( Box 1 ) and this is by no means always the case, even in much more recent psoriasis trials.

In studies of psoralen photochemotherapy, which began to appear in the 1970s and where treatment could be expected to produce, at least in the short term, total or near total remission, the concept of “clearance” was introduced as the outcome success criterion. In a large, randomized study from 1979 with well over 100 patients in each treatment arm, clearance was defined explicitly as “when all the lesions were flat and free of scales”; this was achieved in 91% and 82% of those who received psoralen photochemotherapy and dithranol, respectively. More recently, “clearance” has tended to be qualified with phrases such as “clear or nearly clear” and “clear or minimal residual activity” to cater for those patients who may be left with a small residuum of psoriasis. The UK National Institute of Health and Care Excellence (NICE) has further defined “minimal residual activity” as equivalent to greater than 90% reduction in PASI score.

With the development of new systemic agents such as the oral retinoids, which were rarely capable of producing such clear-cut outcomes, the need for alternative outcome criteria became clear. It was for a dose-ranging study of etretinate published in 1978 that the PASI was created. The PASI was, however, just one of the proliferation of scoring systems for psoriasis that Naldi and colleagues found among psoriasis trials published in major journals between 1977 and 2000: They identified 44 separate measures among the 249 trials they examined. Nevertheless, a systematic review of treatments for severe psoriasis published in 2000 showed that PASI had become as widely used as “clearance” for defining outcomes, with each being utilized to define outcomes in about one quarter of all randomized controlled trials judged to be of sufficient quality for inclusion in the review ( Table 1 ). In the absence of a suitable alternative instrument, PASI was promoted for clinical trials of cyclosporine for psoriasis in the 1980s. There are many problems with PASI (which are discussed elsewhere), but a global consensus on how best to assess psoriasis and its response to treatment remains elusive. Many of the regulatory bodies continue to demand some form of PASI-based response criterion but supplemented by a simpler measure such as a change in global assessment score. The need for both has recently been questioned.

It is only much more recently that it has been recognized that there may be much to be gained by getting patients to assess their own disease severity either to complement or to replace an “objective” observer-rated assessment by a health professional. In diseases in which symptoms rather than signs predominate (eg, chronic pain), it is only the patient who can judge severity. In such a visible disease as psoriasis, however, rather little thought has been given to the potential advantages of asking patients to score their own disease. Not only may it be time saving for health professionals, but it can also give patients greater ownership of and participation in the management of their psoriasis. An adaptation of the PASI for completion by patients was introduced in the mid 1990s as the Self-Administered PASI. It replicates the drawbacks of PASI itself and requires a professional to convert the patient’s shading on line drawings to percentage body surface area (BSA) involvement. More recently the self-assessment Simplified Psoriasis Index (saSPI) has been proposed for patient self-reporting of psoriasis severity.

It has been recognized for at least 70 years that there is an intimate relationship between psoriasis and psychological distress. It has now become standard practice to include some measure of psychosocial impact in clinical trials of interventions for psoriasis. Such measures may be generic (applicable to any disease state), skin disease specific, or psoriasis specific.

The advantage of using a generic measure is that it may then be possible to compare the impact of a skin disease such as psoriasis with that of completely different disorders, such as musculoskeletal or respiratory disease. The disadvantage is that such generic measures may not capture the particular elements of skin disease satisfactorily, because they were not designed with skin disease in mind. Study of the impact of chronic disease (eg, chronic pain, arthritis, stroke) on psychological well-being and functioning, together with developing methods of quantifying it, has spawned its own society (International Society for Quality of Life Research), the mission of which is “to advance the scientific study of health-related quality of life and other patient-centered outcomes to identify effective interventions, enhance the quality of health care and promote the health of populations.” One of the first and still most widely used generic quality of life instruments is the Medical Outcomes Study 36-Item Short Form Survey (SF-36). The SF-36 has been used widely for psoriasis trials, but its components do not capture well much of the psychosocial impact that derives from having psoriasis.

The SF-36 was preceded by some 7 years by the Psoriasis Disability Index (PDI), which was pioneering when it was published in 1985. This has been used widely in clinical studies ; in more recent times, however, the PDI has been largely supplanted by the more generic and simpler Dermatology Life Quality Index (DLQI; Box 2 ) that, like the PDI, was developed by Professor Andrew Finlay. There is a range of other skin disease-specific, quality-of-life instruments that have been promoted, the most important of which are discussed elsewhere in this article.

The dimensions of psoriasis

Psoriasis is a complex disease, is often lifelong, and, in severe cases, life ruining. There is much more to it than its outwardly visible manifestations. This has recently been neatly summarized in the UK’s NICE Guidance on the assessment and management of psoriasis :

Death directly due to psoriasis is rare, but the chronic, incurable nature of psoriasis means that associated morbidity is significant. People with psoriasis, like those with other major medical disorders, have reduced levels of employment and income as well as a decreased quality of life. The impact of psoriasis encompasses functional, psychological, and social dimensions. Factors that contribute to this include symptoms specifically related to the skin (eg, chronic itch, bleeding, scaling and nail involvement), problems related to treatments (mess, odor, inconvenience and time), psoriatic arthritis, and the effect of living with a highly visible, disfiguring skin disease (difficulties with relationships, difficulties with securing employment and poor self esteem). Even people with minimal involvement state that psoriasis has a major effect on their life. The combined costs of long-term therapy and social costs of the disease have a major impact on healthcare systems and on society in general. About a third of people with psoriasis experience major psychological distress, and the extent to which they feel socially stigmatised and excluded is substantial. Healthcare professionals, including dermatologists, often fail to appreciate the extent of this disability and even when it is correctly identified, some estimates suggest that less than a third of people with psoriasis receive appropriate psychological interventions.

In attempting to capture these different dimensions, it is important to recognize that limiting assessment of psoriasis to its outward signs may grossly misrepresent the impact of that disease on the individual with psoriasis. Dermatologists have worked hard to ensure that the major regulatory and health care funding authorities incorporate some measure of quality of life impact into their appraisal of treatments for psoriasis. The principle of the “Rule of Tens,” in which scores greater than 10 of any 1 of 3 measures, percent BSA involvement, PASI, or DLQI, would constitute evidence of current severe psoriasis, has been adopted with minor modifications in a number of guidelines for the management of psoriasis. There are, however, other dimensions of psoriasis that may be equally relevant in helping patients and clinicians to make informed decisions about management but that may not be captured by these tools. These omissions may in part be attributed to the deficiencies of PASI itself as an assessment tool or to the absence of a number of important items from DLQI, such as sleep deprivation or the economic cost of having psoriasis. Neither the presence nor absence of psoriatic arthritis, nor information about the historical behavior of an individual’s psoriasis feature in any of the 3 measures, although they may be very important in reaching appropriate management decisions. Nevertheless, the acceptance that both “objective” severity and quality of life must be considered when assessing overall psoriasis severity has been a major advance.