Currently, at least 15% of the general population with diabetes, hypertension, or dyslipidemia are undiagnosed.1 However, the increase of the prevalence of dyslipidemia as a risk factor for cardiovascular disease emphasizes the need for a proper screening. This is done through the measurement of serum cholesterol, triglycerides and high-density cholesterol (HDL-cholesterol) levels in fasting samples.2 Thereafter, abnormal screening results should be confirmed by a repeated sample on another occasion, and the average of both results should be used for risk assessment.3

An active search for patients with hypercholesterolemia is adviced, through cholesterol measurement in the general population before age 35 in men and 45 in women. Then every 5 years up to 75 years. Above that age cholesterol should be measured once, if not performed before. Hypertriglyceridemia detection is recommended only in certain cases. However, primary screening for the detection of hypercholesterolemia and hypertriglyceridemia at any age is required if the patient is diabetic, or there is family history of premature cardiovascular disease, dyslipidemia, or other risk factors.3,4 Regarding HDL-cholesterol, its measurement is also recommended in cases of hypercholesterolemia. As a general rule, an increase of 1mg/dL in is associated with a 3% lower risk of coronary heart disease in women, and a 2% in men.5

Nowadays, the detection of lipid disorders in adults lies in primary care setting. Previous studies showed the existence of inequalities in preventive practice among practitioners.6,7 A research in a country with two scenarios of primary care showed that general practitioners (GPs) requested more cholesterol and triglycerides than family medicine doctors and referred that from an economic perspective this difference was not justified.8 Other authors stated a high variability in the number of HDL-cholesterol requested between general practices expressed per 1000 practice patients a year, and also when expressed as a percentage of the requests for cholesterol.9 These facts emphasize the need for interventions to reduce the large differences in the tests requested by GPs. A first step will be to investigate the actual behaviour and variability, trying to identify the root causes of this variability, if it exists, and hence of inappropriateness, if discovered.

The aim of this research was, to study the inter-practice variability in lipid laboratory tests requested by GPs and to compare and analyze the variability according to the different characteristics of the geographical regions.

We obtained production statistics from 76 laboratories at different hospitals from diverse regions across Spain (Valencian Community, 23 laboratories and 4,848,900 patients; Andalucía, 12 laboratories and 3,849,485 patients; Castilla y León, 9 laboratories and 1,695,916 patients, and remaining regions 32 laboratories, 7,284,894 patients). 17,679,195 patients were included in the study, from 13 Communities throughout Spain. 5,823,053 cholesterol, 4,544,663 HDL-cholesterol and 5,599,358 triglycerides tests were ordered. Table 1 displays a summary of the organizational data of the different laboratories that participated in the study.

The descriptive statistical analysis and the variability index for the number of requests per 1000 inhabitants and ratios of related tests, is shown in Table 2. The variability of ratios of related tests was also very significant.

In rural hospitals, cholesterol, HDL-cholesterol and triglycerides tests per 1000 inhabitant's indicators results were higher than in urban or rural-urban locations. However, no significant differences in ratio of related tests requests were detected (Table 3).

Neither there was no statistically significant difference between geographical regions in appropriateness indicators (Figures 1 and 2).

Figure 1.

Distribution of every test requests per 1000 inhabitants ratios in the geographical regions of Spain studied.

Region: Valencian comunity (V); Andalucia (A); Castilla y León (CL); Other region (OR);

Test: cholesterol (Chol); HDL-cholesterol (HDL-c); triglycerides (Tg).

(0,11MB).

Figure 2.

Distribution of index indicator in the geographical regions of Spain studied.

Region: Valencian community (V); Andalucia (A); Castilla y León (CL); Other region (OR);

Index Test: HDL-cholesterol / cholesterol (HDL-c/Chol); triglycerides / Cholesterol (Tg/ Chol).

(0,09MB).

The study results highlight the variability in the request of lipid metabolism tests in Spain. The triglycerides/cholesterol and HDL-cholesterol/cholesterol ratios, even close to one in the first case, were high in all the studied laboratories, in spite of the current recommendations that do not recommend triglycerides and HDL-cholesterol measurement in the general population. The observed higher request of cholesterol, HDL-cholesterol and triglycerides in rural areas, inhabited by older population,13 is difficult to explain as cholesterol screening is not recommended over 75 years of age.

It is very difficult to know, in view of the obtained results, if cholesterol was appropriately requested to detect hypercholesterolemia in the general population. In the areas with the lowest requests, 106 tests per 1000 inhabitants, the test was probably not enough ordered to cover population-based cholesterol screening: every 5 years starting at 35 years in male, and 45 years in female patients. Furthermore, the high test request variability observed in our study - from 106 to 550 per 1000 inhabitants- is difficult to be exclusively explained by the population or patient case mix differences. Probably additional reasons are the root causes, as for instance different requesting habits of the GPs in the different studied areas.

Less than 10% of the population is at high risk of cardiovascular disease. However the low to intermediate risk group is so large that most cardiovascular events will occur in that group.14,15 This emphasizes the need of a proper dyslipidemia screening in general population. It is necessary to standardize the request of such laboratory tests to be sure, as a primordial need, that cholesterol is appropriate requested to screen hypercholesterolemia in the general population. An effort must be done in the less demanding areas to stimulate the cholesterol request, in view our study results that showed that cholesterol was probably under requested in certain areas.

Besides promoting the request of cholesterol in some regions, another step to achieve a correct request in the screening of lipid metabolism disorders would be to decrease the requests of triglycerides and HDL-cholesterol, as proposed earlier by some authors.8 As in a prior study in the United Kingdom,8 our data also suggested that HDL-cholesterol and triglycerides were probably over requested in many areas in Spain. More studies are needed to investigate the specific triglycerides/cholesterol and HDL-cholesterol/cholesterol indicators targets to reach in primary care setting in order to establish strategies for a proper lipid tests requesting.

The study had certain limitations. The differences in lipid tests request between health care regions in Spain could be partly explained by the possible variability in the patient population, or more or less often visit their specialist instead of the GP depending on the area.

There was a high variability in cholesterol, triglycerides and HDL-cholesterol requesting in primary care in Spain. Cholesterol was probably inappropriately under requested to screen for hypercholesterolemia in certain areas that would suggests a potential risk for inappropriate dyslipidemia screening in general population. However triglycerides request, at a one to one ratio when compared to cholesterol in most areas, seemed to be inappropriately over requested. Those results emphasize the need to establish interventions to achieve an appropriate and homogenized lipid tests request. More studies are needed to establish indicators targets for a correct and maintained evaluation of appropriate lipid tests request.

This research received specific grant (Ignacio H. de Larramendi Aid to research) from Fundación Mapfre.

None declared.