The risk of death from cancer increases sharply after 50 years of age and most cancers occur in middle-aged and older persons. The incidence of head and neck squamous cell carcinoma and thyroid cancer are increasing along with the nation’s population of elderly Americans. Consequently, more hospitalization will be required for this increasing number of elderly patients undergoing surgery and other treatments for head and neck cancer.
On average, elderly patients have more concurrent illnesses than younger persons, thus one might anticipate that the risk of major medical complications (such as heart failure, pneumonia, or acute renal failure) would be greater in elderly patients hospitalized for equivalent surgeries. Patient length of stay (LOS) is influenced directly by admission diagnosis, treatment, patient age, underlying medical disorders, and occurrence of complications.
In addition to age, other demographic variables that influence length of stay and the risk of medical complications are sex and race. Three otolaryngologist—head and neck surgeons hypothesized that elderly (older than 74 years) patients undergoing surgery for head and neck malignancies would experience a higher rate of medical complications and death, and longer average length of stay than younger patients. They also examined the influence of other variables, such as race, sex, and surgical site on these outcomes and on sources of payment for services, to estimate their impact on utilization of hospital resources.
The authors of “The Impact of Race and Age on Major Medical Complications and Resource Utilization Among Head and Neck Cancer Patients,” are Andrew G. Sikora MD PhD, Mark D. DeLacurel MD, both at the Department of Otolaryngology, New York University Medical School; and Arnold Komisar MD DDS, with the Department of Otolaryngology, Lenox Hill Hospital, all in New York City, NY. Their findings are being presented at the 107th Annual Meeting of The Triological Society, being held April 30-May 3, 2004, at the J.W. Marriott Desert Ridge Resort & Spa, Phoenix, AZ.
The present study consists of a cross-sectional assessment of an aggregate dataset created from the NHDS National Hospital Discharge Survey for 2001 and 2002. This national survey of inpatient hospital discharge data is generated by sampling medical records from 500 private and public short-stay hospitals in the US, approximately eight percent of the total. The NHDS database abstracts standardized information from patient discharge summaries, including demographic (race, age, sex), medical (medical diagnoses by ICD-9 codes; procedures performed by CPT codes), and administrative and resource utilization information (length of stay, hospital type and location, discharge status, insurance status).
From this aggregate database a separate dataset was created consisting of patients with malignant tumors of the head and neck. A database search yielded 682 records for patients who had undergone head and neck surgical procedures, specifically all ablative, reconstructive, and other surgical procedures of the head and neck (diagnostic procedures, i.e. laryngoscopy, esophagoscopy, were excluded). Patients whose only head and neck surgical procedure was tracheotomy were also excluded.
Cases were analyzed for the following variables: age, sex, race, medical diagnoses (by ICD-9 codes), procedures performed during the admission (by CPT code), length of stay, sources of payment, and discharge status (alive or dead). Head and neck surgical procedures and medical complications were abstracted from the relevant CPT and ICD-9 codes respectively. Outcome measures included: occurrence of any medical complication; death; LOS; and payment source.
Patient age was recoded into two groups: elderly (older than 74 years of age) and non-elderly (74 years of age or less). Patient race was recoded into four groups: white, black, other (primarily Asians), and unknown. Surgical procedures were assigned by type of surgery into two groups: patients undergoing surgery of the thyroid, parathyroid, or salivary glands; and patients undergoing any other head and neck surgeries. When more than one surgical procedure was performed for a given admission, the rank order of CPT codes was used to determine which procedure was given priority.
The most frequent sites for surgical procedures were the thyroid/parathyroid, neck, oral cavity, and larynx. Forty-one patients (six percent) experienced a medical complication during hospitalization, the most common being pneumonia. There were 14 deaths, of which two were associated with other medical complications.
The risk of major medical complications differed significantly according to age, sex, and surgery type. While the incidence of complications in non-elderly patients was only 4.5 percent, the corresponding incidence in patients older than 74 old was 13.1 percent. The incidence also differed according to sex and surgery type. The only variables found to be significantly associated with increased complications were age (older than 74) and surgical site other than thyroid/parathyroid/salivary.
Increased length of stay was associated with both demographic (age older than 74 years, male sex, and black race) and other variables (occurrence of medical complication, surgery other than thyroid/parathyroid/salivary). Age, sex, and surgery type were associated with a length of stay longer than seven days.
When the source of payment was examined according to patient race, blacks were found to be more likely than whites to list Medicaid (11.3 percent vs. 5.4 percent) or self pay/no charge (8.1 percent vs. 2.2 percent) as the primary payer. The difference in the proportion of patients who listed Medicaid or self pay/no charge as primary payer differed significantly between whites and blacks, but not between whites and persons of other or unknown race.
These findings indicate that elderly patients have a higher rate of complications and in-hospital deaths, and a longer average length of stay than younger patients undergoing surgery for head and neck malignancies. Blacks and whites have a similar incidence of complications, but blacks have a longer average LOS, and are less likely to have sources of health insurance other than Medicaid. The site of surgery significantly influences the incidence of major medical complications and average LOS, and the occurrence of a major medical complication significantly increases LOS.