SIR–There are a significant number of factors that can contribute to developmental delay/mental retardation* (DD/MR),1,2 and determining the cause in any individual child can be difficult.2–4 Nonetheless, this search for an etiology is important for providing explanation, anticipatory guidance, appropriate management, and accurate recurrence risks to the patient’s family.3–13 Several guidelines to aid in the diagnostic process have already been published,3–7,9,10,12–14 and while these strategies vary based on specialty, the core elements and recommendations are generally consistent. We initiated our studies to investigate the impact of implementing diagnostic guidelines into practice, in terms of the financial costs of various diagnostic procedures.

We conducted a retrospective chart review of new patients referred to our hospital’s Division of Genetic and Metabolic Disorders or Division of Neurology for evaluation of DD/MR during 2002, to explore if comprehensive evaluations (following the consensus of published guidelines) lead to increased etiological determination at a decreased financial cost. Charts were identified based on one of the following ICD-9 codes: 783.4, 784.5, 759.7, 319, 318.0–318.2, or 315.2. Charts were included for review if they met the following criteria: the child had not been previously evaluated by either service, was under 18 years of age, confirmed to have DD/MR, the testing ordered was notated, and test results were available or the patient had returned for follow-up. One hundred and twenty-two charts met the criteria for inclusion (73 males and 49 females; mean age 4y 8mo [SD 3y 9mo], range 2mo–17y 11mo).

The initial clinical evaluation was scored on an 11-point scale, which was based on the consensus of what published guidelines3,6,9,12,15 recommend be included in a diagnostic evaluation for children with DD/MR. One point was awarded for documentation of each of the following 11 components: prenatal, birth, developmental, and family histories; the presence/absence of focal neurological signs, dysmorphology, major birth defects, micro/macrocephaly, and growth abnormalities; specific differential diagnoses; and recommendations for sequential testing. Cases were divided into two groups based on score, with a higher score indicating a more comprehensive evaluation consistent with recommended guidelines, and a lower score indicating a less thorough and limited evaluation. Cases with a score of 10 or 11 were assigned to the comprehensive group (n=59). Cases scoring 4 to 9 were placed in the limited group (n=63). Age, sex, and test recommendations were recorded, and list price of testing totaled. Etiology was noted if one was determined.

There was no difference between groups (comprehensive vs limited) in mean age, sex, or inclusion of two evaluation components – major birth defects and growth abnormalities. Using Fisher’s exact Chi-squared test, there were significant differences between groups (p<0.05) in the frequency with which the remaining nine components were included. The greatest differences were seen in the inclusion of a prenatal history (94.9% vs 50.8%), documentation of dysmorphology (96.6% vs 57.1%), specific differential diagnoses (100% vs 33.3%), and recommendation for sequential test ordering (76.3% vs 15.9%). Mean total cost of investigations within the comprehensive group was (US) $1159 (SD 1347), and within the limited group was $3239 (SD 1777). There was a statistically significant difference in mean total cost between the groups with a mean cost difference of $2080 (95% CI $1511–$2648; SE $287; p<0.001). Etiologies were determined in 35.6% of the comprehensive group and 28.6% of the limited group; this difference was not statistically significant.

We subsequently designed a pilot prospective study based on our retrospective study, to further investigate the financial cost versus etiological determination of testing. Children eligible for enrollment included new patients referred to the same departments for the same indications that were utilized in the retrospective study. All other eligibility requirements were also the same, with the addition that no previous investigation had been performed prior to referral. For every child enrolled we ensured that a comprehensive evaluation was performed (i.e. a total of 10 or 11 points on previously described scale). Eleven children met inclusion criteria (5 males, 6 females; mean age 6y 2mo [SD 5y 6mo], range 7mo–16y). For each child, investigations were based on an algorithm designed by our team, modified from Battaglia and Carey;6 test recommendations and results were recorded, and financial costs (using prices utilized in the retrospective study to allow comparison) were totaled. Etiology was noted if one was determined.

The mean total cost of investigations within the prospective (comprehensive) population was $793 (SD $323). Mean cost difference between prospective group and retrospective limited group was $2446 (95% CI $1368–$3523; SE $540; p<0.001). Mean cost difference between prospective group and retrospective comprehensive group was $366 (95% CI $453–$1185; SE $411; p<0.05). Etiologies were determined in 36.4% of the prospective group, which was not a statistically significant difference from the retrospective comprehensive and limited groups.

We set out to study the rates of etiological determination and associated financial costs of DD/MR investigations based on published guidelines. In our retrospective study, comprehensive evaluations did not change the success rate in etiological detection but did significantly decrease the money spent in etiological testing. Comprehensive evaluations in the prospective study were also associated with a decrease in cost, without compromising the rate of etiological determination. Though our prospective numbers were small, these findings support our conclusion that if guidelines were to be used consistently, there could be a substantial decrease in the cost of diagnostic evaluations of children with DD/MR, without jeopardizing the rate of arriving at a diagnosis. As previous studies have suggested, a careful comprehensive medical history and a detailed physical examination (not forgetting the important dysmorphology exam) can also contribute significantly to an improved diagnostic yield,15 and may, therefore, filter out unnecessary tests.

Some limitations of our studies include the fact that the first was a retrospective study based on documented findings, therefore our results may not include the cases where facets of the comprehensive evaluation were part of the thought process of the physician, but were not documented. To allow for comparison with the retrospective study, common diagnoses such as Down syndrome were excluded from the prospective study. Thus our rate of finding an etiology of children with DD/MR was artificially decreased due to study design. Also, our prospective study was halted after 1 year due to low enrollment numbers; this time-frame did not allow ample time for serial follow-up, and also most likely decreased our etiological rate. Additionally, enrollment of patients into our prospective study without any prior evaluations was needed to accurately assess the cost of testing, but proved more difficult than anticipated, and only a small percentage of the children with DD/MR referred to our clinics were eligible to be enrolled. This could suggest that the primary care physicians or the developmental pediatricians in our area commonly do first tier investigations prior to obtaining a consultation from specialists in our institution; however, this is beyond the scope of this study.

Recent publications8,11,15 and our study point to a single ‘mantra’ for the evaluation of children with DD/MR, namely a thorough comprehensive clinical evaluation (not forgetting dysmorphology, prenatal and birth history, and family history) used as the basis for prioritization of step-wise investigations. As we all know, every child is unique, and the evaluation for the etiology of their features should be a contemplative one, and not a ‘one-size-fits-all’. Technology has improved tremendously, and most likely will continue to do so; with this will come increased costs. Given this, we may find ourselves soon forced to do a better job of prioritizing testing, so we might as well stop to ponder it now.

  • *

    UK usage: learning disability.


  1. Top of page
  2. Acknowledgements
  3. References

This study was supported by Project No. 1 H46 MC00215 from the Maternal and Child Health Bureau (Title V, Social Security Act), Health Resources and Services Administration, Department of Health and Human Services (ALD, LNT).


  1. Top of page
  2. Acknowledgements
  3. References