Counting without the Numbers: Venezuelan Attempts at Prevalence Studies

Background: Diagnostic studies involving the Spanish-speaking population are scarce.  Although there is no evidence that the clinical presentation of ASD varies across cultures, ethnicities, or races, there are disparities in prevalence and age of diagnosis for the different countries and ethnic groups.  Globally, ASD prevalence has increased and the age of diagnosis has decreased, but not for the Latino population.  Prevalence studies in LAMI countries serve many purposes: They offer data for health and educational planning, allow international comparisons for the study of risk factors, protective factors, and cultural influences, and, finally, permit the cultural adaptation of assessment tools.

Objectives: The aim of this manuscript is: (1) to discuss the results of epidemiological research in Venezuela in light of methodological features and  (2)  to offer conclusions about the best up-to-date methods for data collection.

Methods: We present the results of three different studies that focused on ASD prevalence in Venezuelan children.  We considered and analyzed sampling processes, screening and diagnostic procedures, and response rates. Then we reviewed and compared adjustments from standard prevalence estimation approaches used in other countries.

Results: The first report was designed as a population study However, birth records, visits to pediatricians, and vaccinations among others were not always registered.  Other logistic difficulties included the absence of public services (electricity, sewage, telephone) for many houses, which made them nonexistent in public records.  In addition, some urban areas were dangerous, which prevented researchers from going into the houses to assess children.  Obstacles to implementing this method hindered data collection, the determination of research findings, and, hence, the proper calculation of prevalence rates for ASD.  The second study aimed to establish prevalence through the review of records in public and private institutions for special education and in specialized clinics.  The overall prevalence of ASD was 1.7 per 1,000 (95% CI: 1.3-2.0).  For  this approach, we found that some of the children’s records were missing.  Moreover, despite legal requirements, not every child with disability was included in the school system or the health system.  The third study used school records. Thus, it was possible to assess the children in their schools and, consequently, to avoid security issues.  For this study, the estimated prevalence was 7.18 per 1,000 (95% CI:  5.3-9.6).  We faced the same difficulties with records because not all children were attending school at the time of the study.  The last approach was the best one we identified for the development of data in a culturally appropriate way. 

Conclusions: In Venezuela, as in most Latin American countries, the difficulties experienced while conducting prevalence studies were not related to translation and the adaptation of diagnostic instruments.  The main problems were related to sampling procedures and the quality of public records.  As we develop a stronger, valid, cross-cultural process for the identification of ASD, our findings could be compared to those of other countries and larger databases.