Abstract
In recent years, the cost of health care around the world has risen at a rate that is deemed unsustainable. It has been estimated that 20% of this could be saved by rationalising laboratory investigations and reducing inappropriate requisitioning of the investigations. There are several reasons for the excessive, redundant, inappropriate or unnecessary investigations and procedures, which in some instances are unethical practices. The impact in financial terms is more in developing countries such as India with <5% of the population having medical insurance and hardly any other third-party payer system. The ‘Choosing Wisely’ campaign of the American Board of Internal Medicine, Canadian Rheumatology Association’s Choosing Wisely Committee and the ‘Society for Less Investigative Medicine’ (SLIM) initiative of the doctors of All-India Institute of Medical Sciences (AIIMS), New Delhi, all have provided recommendations to reduce unnecessary investigations, and these are among some of the efforts to reduce the cost of investigations without compromising the quality of care.
Introduction
In recent years, the cost of health care around the world has been rising at a rate that is unsustainable . For example, in the US, wasteful spending likely accounts for between one-third and one-half of all health-care spending. Pricewaterhouse-Coopers calculated that up to US$ 1.2 trillion, or half of all health-care spending, is the result of waste . An Institute of Medicine (IOM) report estimated that unnecessary health-care spending totalled US$ 750 billion in 2009 alone . Similarly, ‘The Carter Review’, a UK Department of Health-commissioned review of pathology services in England, estimated that 20% of this could be saved by rationalising laboratory investigations and reducing inappropriate requisitioning of the investigations . This UK ‘Review’ estimated that 25% of laboratory tests were unnecessary, representing a huge potential waste. Another example is from the Netherlands where expenditure on diagnostic tests grew at the rate of 7% a year without parallel improvement in the health status, suggesting that investigations were being overused . A recent Canadian ‘Choosing Wisely Committee’ report has discussed the problem of wasteful laboratory investigations and pointed out five investigation items with questionable significance in daily-routine rheumatology practice . According to these and several other reports, redundant, inappropriate or unnecessary tests and procedures are sadly becoming rampant . Several factors may be responsible for the increasing use of investigations. These could be as simple as ‘the tests are available’. Another factor could be the urge to make use of new technology. The fear of litigation may also be driving doctors to practice defensive medicine without realising that on average 5% of test results are outside their reference ranges. Therefore, once an abnormal test result is found, it may result in a cascading effect and doctors may order several other investigations, not realising that 5% of them would again be abnormal by chance . More worryingly, in some recent issues of the British Medical Journal , it has been discussed that in India some clinical laboratories and hospitals may be involved in the unethical practice of offering a certain percentage of the cost of the investigation as incentives to the referring doctors for ordering more tests and procedures . In developing countries such as India, with <5% of the population having medical insurance coverage, hardly any third-party payer system, and a per-capita income of only US$ 1570 in 2013, ranked 120th out of 164 countries by the World Bank , the crippling cost of health care is nothing short of a crisis. Weaker sections of the society are the worst affected . It could therefore be argued that the importance of the cost-effective use of investigations in developing countries is much more than in developed countries such as USA where the per-capita income was US$ 42,693 in 2012 . These trends have forced clinicians to start campaigns for reducing the overuse of tests and procedures, and to support patients in their efforts to make smart and effective care choices. Some of the efforts to reduce the cost of investigations without compromising the quality of care include the ‘Choosing Wisely’ campaign of the American Board of Internal Medicine and the American College of Rheumatology (ACR) , Canadian Rheumatology Association’s Choosing Wisely Committee’s recommendations and the ‘Society for Less Investigative Medicine’ (SLIM) initiative of doctors of All-India Institute of Medical Sciences (AIIMS), New Delhi, to reduce unnecessary investigations . In this chapter, we discuss the cost-effective approach to investigations with some examples without attempting to be all-inclusive. To allow the reader a comparison of cost, we have illustrated the cost (in US$) of investigations as they currently exist in India.
Tools for making a diagnosis
Traditionally, the tools for making a diagnosis include clinical history, physical examination and laboratory investigations (including imaging and histopathology). Several workers have reported the relative contributions of these ‘tools’ or ‘components’ towards diagnosis. Several studies have proven it beyond doubt that clinical history is absolutely critical to making a diagnosis . Once a provisional diagnosis is made, a short list of differential diagnosis can also be prepared without trying to be comprehensive. Then, a list of wisely chosen focussed laboratory investigations for confirming the provisional diagnosis and refuting a few of the main differential diagnoses can be drawn. Further, if some unexpected results are shown, a few additional investigations related to that diagnosis may be ordered as a ‘second round’ of investigations. This would be the most cost-effective use of investigations towards making a diagnosis.
Tools for making a diagnosis
Traditionally, the tools for making a diagnosis include clinical history, physical examination and laboratory investigations (including imaging and histopathology). Several workers have reported the relative contributions of these ‘tools’ or ‘components’ towards diagnosis. Several studies have proven it beyond doubt that clinical history is absolutely critical to making a diagnosis . Once a provisional diagnosis is made, a short list of differential diagnosis can also be prepared without trying to be comprehensive. Then, a list of wisely chosen focussed laboratory investigations for confirming the provisional diagnosis and refuting a few of the main differential diagnoses can be drawn. Further, if some unexpected results are shown, a few additional investigations related to that diagnosis may be ordered as a ‘second round’ of investigations. This would be the most cost-effective use of investigations towards making a diagnosis.
Categories of investigations and improving their cost-efficacy
Investigations may be used for different clinical purposes, as follows: (i) for confirming/supporting a provisional diagnosis and refute differential diagnoses, (ii) for establishing the baseline status of disease activity and the status of different organ systems, (iii) for conducting a follow-up of the disease activity status, (iv) for monitoring drug toxicities and (v) for evaluating the complications of the disease. The cost-effectiveness may be appraised in the context of any or all of the above purposes. The following examples of some of the common tests conducted in rheumatology practice will help clarify the issue:
- 1.
Acute-phase reactants (APRs)
- 2.
Haematology
- 3.
Biochemistry
- 4.
Immunology
- 5.
Histopathology
- 6.
Microbiology
- 7.
Imaging
Acute-phase reactants
Erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) are the most widely measured ARPs in clinical practice. Measurement of APRs at baseline and at follow-up serves as a surrogate for the degree of inflammation, and it helps in monitoring the response to drug treatment . ESR conducted by the standard Westergren (recommended) technique costs about US$ 2. For disease-modifying anti-rheumatic drug (DMARD) monitoring, the required ‘complete blood count (CBC)’ along with ESR costs only about US$ 7. Similarly, the CRP test alone costs US$ 12. In addition, APRs need to be repeated at each follow-up visit, thus further adding to the cost. Under these price constraints, among very poor patients, only ESR may be performed without CRP. It may not be ideal, but it would generally serve the purpose of assessing the degree of inflammation at baseline and at follow-up. However, in exceptional situations, for example, for distinguishing disease activity in systemic lupus erythematosus (SLE) from sepsis, CRP would be useful although expensive.
Ferritin is also one of the APRs. It costs US$ 12. Therefore, it would be considered an expensive test for the diagnosis of iron deficiency anaemia as there are cheaper options (described under ‘ anaemia ’ below). However, used as an APR, it has relevance in the diagnosis of a relatively uncommon condition, namely adult-onset Still’s disease (AOSD) . Despite being expensive, it would be justified in this uncommon disease.
Haematology
Anaemia
Anaemia is common in many rheumatological diseases. The reasons are multifactorial, and in practice cost can be kept to a bare minimum if common causes are investigated first. Iron deficiency is the most common cause of anaemia in the population of developing countries and also in rheumatological diseases. Therefore, a good history (blood loss due to haemorrhoids, menorrhagia, gastrointestinal bleeding causing black stools and poor eating habits) would help in making a provisional diagnosis of iron deficiency anaemia. Then, a look at the red cell morphology would be very informative yet cheap; microcytic hypochromic red blood cells (RBCs) would support the diagnosis of iron deficiency. The only other important cause for microcytic hypochromic anaemia is thalassaemia or thalassaemia trait, a common problem in northern India. A peripheral blood smear that shows an abnormally high count of ‘target cells’ would be a clue, but it may be necessary to use the Mentzer index to further support this diagnosis . Nutritional B12 and folate deficiency is also widespread in India , and, therefore, it would be expected in rheumatological diseases as well. Macrocytic RBC in the peripheral blood and mild pancytopenia are features of B12 or folate deficiency, and they would aid in diagnosis without adding to the cost of B12 and folate estimations. Anaemia of chronic disease is also a common cause. Persistent inflammation not responding to treatment in the absence of features of iron deficiency or folate B12 deficiency would be clinically sufficient to diagnose anaemia of chronic disease. A high serum ferritin level would be confirmatory but expensive. In SLE, haemolytic anaemia may be revealed by persistent reticulocytosis (not an expensive test). The cause may be either autoimmune haemolytic anaemia or microangiopathic haemolytic anaemia. Coombs’ test would be confirmatory but expensive. The presence of schistiocytes and fragmented RBCs would be strongly suggestive of the latter.
Leucocytes
As leucocyte count is part of the CBC panel, no additional cost is incurred. Borderline leucopenia is common in connective tissue diseases especially SLE in which it is mainly due to lymphopenia. Significant and steadily decreasing leucocyte count requires attention. Drug overdose or wrong dose schedule must be carefully excluded. Further investigations would require expensive and invasive (bone marrow study) steps. Leucocytosis, which is usually mild, is common in Systemic Immunoinflammatory Rheumatic Diseases (SIRD), but it could be due to infection. Very high white blood cell (WBC) count, however, is common in adult Still’s disease and severe systemic vasculitides as part of the inflammation. It does not require any special investigations.
Platelets
Platelet count is also a part of the CBC panel, so no extra cost is incurred. In inflammatory rheumatological diseases, platelet count is often high as part of the acute-phase reaction and it needs no special investigations. In SLE and anti-phospholipid syndrome (APS), however, a low platelet count is often seen but it is very rare for the count to reach bleeding range. Other causes of thrombocytopenia must be looked into, for example, a viral infection (dengue is a typical example) or drug toxicity (cytotoxic drugs), and they should be managed accordingly.
Biochemistry
There are a large number of tests listed under this category. One common mistake is to routinely request both alanine aminotransferase (ALT) and aspartate transaminase (AST). In the majority of situations in rheumatology, only the state of liver health and any drug toxicity is of interest, and thus ordering only ALT would suffice in most cases. It is worth noting that AST (more specifically, AST/ALT ratio) is mainly used as a non-invasive method of monitoring chronic liver diseases such as hepatitis C or non-alcoholic steatohepatitis (NASH) or for monitoring fatty liver progression to NASH. The cost of ALT and AST is US$ 2.5. Therefore, requesting only ALT would save costs by 50%. Similarly, the serum uric acid (SUA) test could be of no particular use when gout is not considered as a differential diagnosis. Similarly, requesting the SUA test in the paediatric age group or in women of reproductive age would be unwise; gout is a rarity among them. Estimating blood glucose and serum creatinine at baseline would be very useful. Along with family history and baseline blood glucose values, the rheumatologist would feel confident that diabetes would not become a major issue during the management of the patient being treated with glucocorticoids. The cost of a blood glucose test is US$ 1.5, which is cost-effective. Serum creatinine costs US$ 3, and it is very cost-effective. Serum creatinine also helps in the calculation of e-GFR (estimated glomerular filtration rate), which is an important measure for estimating renal status in the follow-up of patients treated with a variety of nephrotoxic drugs. Furthermore, this does not add to the cost as e-GFR calculator is freely available online. Also, the extra cost of serum urea test of US$ 2 does not give more information related to renal status, and it could be avoided. In a patient with proximal muscle weakness but with clinical features of osteomalacia, serum alkaline phosphatise (US$ 3) would be helpful as it would be markedly elevated. Vitamin D3 deficiency is worldwide, and more so in developing countries. Testing this deficiency costs US$ 35 and serves no purpose; routine supplementation of D3 would be wiser and cost-effective. Urinalysis is an inexpensive but highly useful investigation. As renal involvement is common in SIRD, testing urine for proteinuria and sediment abnormalities (RBCs, casts and dysmorphic RBCs) is mandatory as a ‘routine’ investigation. A fresh urine sample for sediment abnormalities is recommended. For quantifying proteinuria, a test for protein/creatinine ratio in spot urine is more accurate, sensitive and cost-effective than actual 24-h urine collection. Routine urinalysis costs US$ 1.5, and testing of the protein/creatinine ratio costs US $3.
Synovial fluid examination
In a polyarthritis with typical clinical pattern of any of the common inflammatory arthritis (RA psoriatic arthritis (PsA) and spondyloarthritis (SpA)), joint aspiration is not necessary as it neither aids in diagnosis nor helps the treatment. However, if intra-articular depot-glucocorticoid instillation is considered necessary as part of the treatment, joint aspiration must be carried out to relieve the intra-articular pressure and to obtain space for administering the drug. A synovial fluid total white cell count may be carried out to further confirm the inflammatory nature of the synovitis (WBC count usually >2000/cmm). Differential WBC count does not add to the diagnosis, prognosis or treatment. Therefore, this can be omitted, further reducing the cost. Crystal examination is usually advocated in situations where crystal arthritis is clinically strongly suspected. It would be unnecessary to carry out crystal examination in paediatric-age patients and in women of reproductive age.
Immunology
The topic has been reviewed in three seminal papers; the reader may like to refer to these articles . Immunological investigations are the most valued tools in the workup of SIRDs. They can be used for screening (subclinical disease), for supporting a suspected diagnosis, for the serial assessment of disease activity over time, for prognosis, for detecting flares, and for providing clues to complications. Some of the investigations are considered important enough to be included in the classification criteria for different diseases such as rheumatoid factor (RF) and anti-cyclic citrullinated peptide antibodies (ACPAs), criteria of RA, anti-nuclear antibody (ANA) and antibodies against other nuclear/cellular antigens in the classification criteria of SLE, anti-U1-RNP antibody in mixed connective tissue disease, anti-SSA/SSB in Sjögren’s syndrome, ANA/antibodies against extractable nuclear antigen (anti-ENA) in systemic sclerosis and anti-phospholipid antibodies (aPLa’s) in anti-phospholipid syndrome (APS).
ANA
The conditions where ANA may be detectable along with its specificity and sensitivity are given in Table 1 . Significant levels of ANA are observed in SLE, Sharp syndrome (mixed connective tissue disease), drug-induced lupus, scleroderma (systemic sclerosis – SSc), Sjögren’s syndrome and inflammatory myopathies. ANA is present in other diseases as well, for example, RA and RA vasculitis, discoid lupus and in pauciarticular juvenile chronic arthritis (juvenile idiopathic arthritis, JIA) with positivity reported with varying frequencies. ANA may also be positive in non-rheumatological patients and in the healthy population in about 3–15% , and, in more recent reports, using advanced and sensitive techniques, in up to 20% of normal individuals . People on some medications and blood relatives of those with CTDS may also have ANA. Therefore, ordering ANA routinely should be avoided.
| Autoantibodies | Associated CTD | Sensitivity | Specificity |
|---|---|---|---|
| ANA | SLE | 93 | 57 |
| Sjögren’s syndrome | 48 | 52 | |
| SS | 85 | 54 | |
| PM/dermatomyositis | 61 | 63 | |
| Raynaud’s phenomenon | 64 | 41 | |
| Specific ANA | |||
| Anti-dsDNA | SLE | 57 | 97 |
| Anti-Sm | SLE | 25–30 | High b |
| Anti-SSA/Ro | Sjögren’s syndrome, subacute cutaneous SLE, neonatal lupus syndrome | 8–70 | 87 |
| Anti-SSB/La | Sjögren’s syndrome, subacute cutaneous SLE, neonatal lupus syndrome | 16–40 | 94 |
| Anti-U3-RNP | SS | 12 | 96 |
| Anti-centromere | Limited cutaneous SS | 65 | 99.9 |
| Scl-70 | SS | 20 | 100 |
| Jo-1 | PM | 30 | 95 |
In a patient provisionally diagnosed as having SLE (or related disease), a positive ANA test, albeit in significant titres, would help establish the diagnosis. However, more importantly, a negative ANA result in a patient with suspected SLE would make this diagnosis almost untenable. An evidence-based guideline of by an ACR-appointed committee concluded that a positive ANA result was very useful for the diagnosis of SLE and systemic sclerosis, and somewhat useful for the diagnosis of Sjögren’s syndrome and polymyositis/dermatomyositis. ANA testing is also useful for identifying patients with JIA at a risk of asymptomatic uveitis and for distinguishing patients with primary Raynaud’s disease from those with Raynaud’s phenomenon associated with an underlying systemic autoimmune disease. On the other hand, ANA has no role in the serial assessment of disease activity, detection of flares or prediction of disease complications. More technical details of ANA testing can be found in chapter 6. High-throughput HANA is an expensive test costing approximately US$ 12 for the enzyme-linked immunosorbent assay (ELISA) assay and US$ 26 for the indirect immunofluorescence technique (IIFT) in different laboratories. In India, the use of poor-quality and unstandardised ELISA kits often leads to false-positive ANA, and therefore IIFT assay is used more often, adding to the cost.
Anti-ENA antibodies
The clinical utility of the anti-ENA test is unfortunately low, the reason being that some of the anti-ENAs have high disease specificity but rather low sensitivity ( Table 1 ). As can be seen, the sensitivity and specificity also varies depending upon the type of underlying CTD. The second reason is the significant overlap that exists among them. Last but not the least, ENA antibody testing is among the most expensive tests in rheumatology costing US$ 50; therefore, but for some exceptional situations, ENA antibody testing is not cost-effective.
Anti-DNA antibodies
This category includes antibodies against single- and double-stranded DNA (dsDNA) and anti-histone antibodies; the latter is usually indicative of drug-induced SLE. As shown in Table 1 , significant levels of anti-dsDNA antibodies (anti-dsDNAab) has 97% specificity for SLE, but unfortunately the sensitivity is rather low, only 57%. Therefore, a significant number of active SLE patients would be negative for anti-dsDNAab at any given time, a perfect example of the diagnostic inaccuracy of the test. Its relatively high specificity, its fluctuations with disease activity and its association with active glomerulonephritis make anti-dsDNAab testing a useful investigation in the follow-up evaluation and management of SLE patients with renal involvement. Another situation where anti-dsDNAab testing could be useful is in distinguishing a lupus flare from an infection. High levels of anti-dsDNAab (along with high procalcitonin level without a parallel rise in CRP) would favour lupus flare. If used for these clinical situations, the use of the anti-dsDNAab test would be cost-effective.
Rheumatoid factor
This test costs approximately US$ 6. Testing for RF is primarily used for the diagnosis of RA. Taken together with the added clinical features of an inflammatory polyarthritis, RF strengthens the possible diagnosis of RA. However, RF is not specific for RA. It is also present in a large number of conditions that include several other rheumatic diseases and several other chronic conditions including infections and malignancies as well as in normal individuals ( Table 2 ) . Conversely, approximately 30% of patients with RA are negative for RF. Thus, the sensitivity of RF is only 60–70% and the specificity is only 78% for RA . Therefore, the presence or absence of a positive RF must be interpreted in the clinical context. Like ANA, RF does not correlate with disease activity over time. Therefore, it has no role in monitoring the disease activity in the follow-up of patients with RA.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
