OBJECTIVES: 1) To estimate the prevalence of hypercholesterolemia and other risk factors for coronary artery disease among employees of the Philippine General Hospital, and 2) to determine the effect of sampling method on these estimated prevalence rates


METHODS: Empployees were surveyed using two sampling techniques: 1) simple random sampling, or 2) convenience sampling group using volunteers. The randomly sampled group was made up of four subgroups, namely: 1) doctors and dentists, 2) nurses, 3) other paramedical employees, and 4) non medical employees. Variables measured included serum total cholesterol, blood pressure, smoking habits and actual body weight expressed as percent of ideal body weight.


RESULTS: Compared to the volunteer group, the randomly sampled group had significantly lower mean values for age, total cholesterol, systolic and diastolic blood pressures and percent of ideal body weight. The prevalence of risk factors was also significantly lower in the randomly sampled group, in terms of hypercholesterolemia (21.14% vs. 30.88%, p.001), hypertension (15.88% vs. 20.35%, p=0.019), isolated systolic hypertension (0.79% vs 1.75% p=0.034), smoking (24.91% vs 29.47%, p=0.037), and obesity (18.51% vs 28.07% p<.001). Significant differences were also noted between the different subgroups sampled.


CONCLUSIONS: These findings demonstrate that the prevalence rate of risk factors for coronary artery disease are overestimated when convenience samples (e.g. volunteers) are used instead of randomly selected subjects. Convenience sampling should therefore be avoided, especially in third world countries where they are still commonly used.

Acta Medica Philippina,1994 30 ( 1 ) pages 35-41