Current pregnancy among women with spinal cord injury: findings from the US national spinal cord injury database.

STUDY DESIGN: Cross-sectional study. OBJECTIVES: To examine the prevalence of pregnancy and associations with sociodemographic and clinical factors among women with spinal cord injury (SCI). SETTING: US National Spinal Cord Injury Database, an SCI registry that interviews participants 1, 5 and then every 5 years post injury. Data include SCI clinical details, functional impairments, participation measures, depressive symptoms and life satisfaction. Women aged 18-49 are asked about hospitalizations in the last year relating to pregnancy or its complications. Data represent 1907 women, who completed 3054 interviews. METHODS: We used generalized estimating equations to examine bivariable associations between pregnancy and clinical and psychosocial variables and to perform multivariable regressions predicting pregnancy. RESULTS: Across all women, 2.0% reported pregnancy during the prior 12 months. This annual prevalence differed significantly by the years elapsed since injury; the highest rate occurred 15 years post injury (3.7%). Bivariable analyses found that younger age at injury was significantly associated with current pregnancy (P<0.0001). Compared with nonpregnant women, those reporting current pregnancy were significantly more likely to be married or partnered, have sport-related SCI, have higher motor scores and have more positive psychosocial status scores. Multivariable analyses found significant associations between current pregnancy and age, marital status, motor score and mobility and occupation scale scores. CONCLUSION: Current pregnancy rates among reproductive-aged women with SCI are similar to rates of other US women with chronic mobility impairments. More information is needed about pregnancy experiences and outcomes to inform both women with SCI seeking childbearing and clinicians providing their care.

Patient-physician communication at hospital discharge and patients' understanding of the postdischarge treatment plan.

BACKGROUND: The quality of discharge planning is an important determinant of patient outcomes following hospital discharge. Patients often report inadequate discussion prior to discharge regarding major elements of the postdischarge treatment plan, including medication and daily activities. OBJECTIVE: To determine whether this apparent lack of communication might be the result of differing perceptions on the part of patients and physicians regarding the patients' understanding of the treatment plan. METHODS: We surveyed 99 patients and their attending physicians. All patients had been discharged recently from an academic medical center with the diagnosis of acute myocardial infarction or pneumonia. We asked both patients and physicians about time spent prior to discharge discussing the postdischarge treatment plan and the patients' understanding of this plan. McNemar test was used to determine whether responses of patients and physicians differed. RESULTS: Physicians reported spending more time discussing postdischarge care than did patients (P = .10). Physicians believed that 89% of patients understood the potential side effects of their medications, but only 57% of patients reported that they understood (P < .001). Similarly, physicians believed that 95% of patients understood when to resume normal activities, while only 58% of patients reported that they understood (P < .001). CONCLUSIONS: Physicians overestimate patients' understanding of the postdischarge treatment plan. Steps should be taken to improve communication about postdischarge treatment.

In-hospital mortality for surgical repair of congenital heart defects: preliminary observations of variation by hospital caseload.

OBJECTIVE: To examine the impact of hospital caseload on in-hospital mortality for pediatric congenital heart surgery. DESIGN: Population-based, retrospective cohort study. SETTING: Acute care hospitals in California and Massachusetts. PATIENTS: Children undergoing surgery for congenital heart disease, identified by the presence of procedure codes indicating surgical repair of a congenital heart defect in computerized statewide hospital discharge abstract databases. Cases were grouped into four categories based on the complexity of the procedure. MAIN OUTCOME MEASURES: Adjusted odds ratios (OR) for in-hospital death were estimated using generalized estimating equations that account for the intra-institutional correlation among patients. RESULTS: A total of 2833 cases at 37 centers were identified. Compared with centers performing > 300 cases per year, after controlling for patient characteristics, centers performing < 10 cases per year had an OR for in-hospital death of 7.7 (95% confidence interval (CI) [1.6-37.8]); 10 to 100 cases, OR = 2.9 (95% CI [1.6-5.3]); 101 to 300 cases, OR = 3.0 (95% CI [1.8-4.9]). Independent risk factors for mortality included procedure complexity category (P < .0001), use of cardiopulmonary bypass (P < .0001), young age at surgery (P = .001), and transfer from another acute care hospital (P < .0001). Few differences were found by hospital caseload in length of stay or total hospital charges. CONCLUSIONS: For children with a congenital heart defect who underwent surgery in California in 1988 or Massachusetts in 1989, the risk of dying in-hospital was much lower if the surgery was performed at an institution performing > 300 cases annually. This study was limited by the absence of clinical detail in discharge abstract databases. If these findings are corroborated by other studies, health care delivery strategies that direct children requiring surgical correction of congenital heart defects to high-volume centers may substantially reduce overall mortality.

Using risk-adjusted outcomes to assess clinical practice: an overview of issues pertaining to risk adjustment.

Increasingly, health care providers are being evaluated and held accountable for their patients' outcomes, ranging from the costs to service consumption to death. To be meaningful, the outcomes under scrutiny must be important to patients or to the health care system as a whole, relatively common, and linked temporally and causally to the care provided. In addition, outcomes findings should be adjusted for patient risk factors, with the goal of accounting for pertinent clinical characteristics before drawing inferences about the effectiveness or quality of care. Risk adjustment "levels the playing field" in comparing outcomes across providers. Although this concept is straightforward, performing clinically credible risk adjustment is difficult, especially given the widespread data constraints. In this article, I review the major issues involved in performing risk adjustment for health care outcomes studies.

Validation of relative value scale for congenital heart operations.

BACKGROUND: To determine the validity of the newly assigned work relative value unit (RVU) scale for surgical procedures for congenital heart disease, we measured its relationship to length of hospital stay, total hospital charges, and mortality. METHODS: We identified cases by the presence of ICD-9-CM codes in nine statewide, administrative hospital discharge abstract databases for 1992. Computer algorithms were generated to assign RVUs to individual cases. Spearman correlation coefficients between work and practice expense RVUs and median length of hospital stay, total hospital charges, and in-hospital mortality were determined, as well as parameter estimates from linear and logistic regression. RESULTS: Using data from 5,192 cases involving 34 surgical procedures for congenital heart disease, higher work RVUs were associated with longer lengths of hospital stay (rs = 0.72, p < 0.0001), higher total hospital charges (rs = 0.81, p < 0.0001), and higher in-hospital mortality (rs = 0.45, p = 0.01). A 5-point increase in the relative value scale was associated with an increase in the length of stay by a multiplicative factor of 1.3 (p < 0.0001); total hospital charges by 1.5 (p < 0.0001); and the odds of in-hospital death by 1.9 (p < 0.0001). Findings were similar for practice expense RVUs, as work and practice expense RVUs were highly correlated (rs = 0.93, p < 0.0001). CONCLUSIONS: The group of work RVUs for surgical procedures for congenital heart defects are reasonable relative measures, on average, of physician work for these procedures, thus supporting the use of this scale to determine physician reimbursement. Practice expense RVUs may not be an independent measure for these procedures.

Changes in payment policies: impact on physicians' office testing.

Recent changes in payment policies include powerful pecuniary incentives to move care from expensive hospital settings to cheaper outpatient sites. Physicians face competition from a growing number of alternative providers in the diagnostic testing marketplace. Given that a concurrent trend involves aggressive utilization review with stiff penalties for noncompliance, physicians are challenged to practice appropriate restraint in ordering and performing tests.

Do severity measures explain differences in length of hospital stay? The case of hip fracture.

OBJECTIVE: To examine whether judgments about hospital length of stay (LOS) vary depending on the measure used to adjust for severity differences. DATA SOURCES/STUDY SETTING: Data on admissions to 80 hospitals nationwide in the 1992 MedisGroups Comparative Database. STUDY DESIGN: For each of 14 severity measures, LOS was regressed on patient age/sex, DRG, and severity score. Regressions were performed on trimmed and untrimmed data. R-squared was used to evaluate model performance. For each severity measure for each hospital, we calculated the expected LOS and the z-score, a measure of the deviation of observed from expected LOS. We ranked hospitals by z-scores. DATA EXTRACTION: All patients admitted for initial surgical repair of a hip fracture, defined by DRG, diagnosis, and procedure codes. PRINCIPAL FINDINGS: The 5,664 patients had a mean (s.d.) LOS of 11.9 (8.9) days. Cross-validated R-squared values from the multivariable regressions (trimmed data) ranged from 0.041 (Comorbidity Index) to 0.165 (APR-DRGs). Using untrimmed data, observed average LOS for hospitals ranged from 7.6 to 23.9 days. The 14 severity measures showed excellent agreement in ranking hospitals based on z-scores. No severity measure explained the differences between hospitals with the shortest and longest LOS. CONCLUSIONS: Hospitals differed widely in their mean LOS for hip fracture patients, and severity adjustment did little to explain these differences.

Chronic conditions and risk of in-hospital death.

OBJECTIVE: This study examined the relationship of in-hospital death and 13 conditions likely to have been present prior to the patient's admission to the hospital, defined using secondary discharge diagnosis codes. DATA SOURCES AND STUDY SETTING: 1988 California computerized hospital discharge abstract data, including 24 secondary diagnosis coding slots, from all general, acute care hospitals. STUDY DESIGN: The odds ratio for in-hospital death associated with each of 13 chronic conditions was computed from a multivariable logistic regression using patient age and all chronic conditions to predict in-hospital death. DATA EXTRACTION: All 1,949,276 general medical and surgical admissions of persons over 17 years of age were included. Patients were assigned to four groups according to the mortality rate of their reason for admission; some analyses separated medical and surgical hospitalizations. PRINCIPAL FINDINGS: Overall mortality was 4.4 percent. For all cases, mortality varied by chronic condition, ranging from 5.3 percent for coronary artery disease to 18.6 percent for nutritional deficiencies. The odds ratios associated with the presence of a chronic condition were generally highest for patients in the rare mortality group. Although chronic conditions were more commonly listed for medical patients, the associated odds ratios were generally higher for surgical patients, particularly in lower mortality groups. CONCLUSIONS: Studies examining death rates need to consider the influence of chronic conditions. Chronic conditions had a particularly significant association with the likelihood of death for admission types generally associated with low mortality rates and for surgical hospitalizations. The accuracy and completeness of discharge diagnoses require further study, especially relating to chronic illnesses.

An integer programming model to limit hospital selection in studies with repeated sampling.

OBJECTIVE: We describe an integer programming model that, for studies requiring repeated sampling from hospitals, can aid in selecting a limited set of hospitals from which medical records are reviewed. STUDY SETTING: The model is illustrated in the context of two studies: (1) an analysis of the relationship between variations in hospital admission rates across geographic areas and rates of inappropriate admissions; and (2) a validation of computerized algorithms that screen for complications of hospital care. STUDY DESIGN: Common characteristics of the two studies: (1) hospitals are classified into categories, e.g., high, medium, and low; (2) the classification process is repeated several times, e.g., for different medical conditions; (3) medical records are selected separately for each iteration of the classification; and (4) for budgetary and logistical reasons, reviews must be concentrated in a relatively small subset of hospitals. DATA COLLECTION/EXTRACTION METHODS. In each study, hospitals are ranked based on analysis of hospital discharge abstract data. CONCLUSIONS: The model is useful for identifying a subset of hospitals at which more intensive reviews will be conducted.

Purpose of admission and resource use during cancer hospitalizations.

This study examined the role of purpose of admission (POA) in hospitalizations for lung, colon, and breast cancers, using the 1985 20-percent Medicare provider analysis and review file. Six POA categories were created from discharge abstract data. Average hospitalization charges, per diem charges, length of stay, and rates of death varied significantly by POA (p < .001). Rural and small hospitals were more likely to admit patients for palliation, while urban and large hospitals admitted relatively more patients for active interventions (p < .0001). POA and indicators of case complexity added only modestly to the ability of diagnosis-related groups to predict hospitalization charges.

Using diagnoses to describe populations and predict costs.

The Diagnostic Cost Group Hierarchical Condition Category (DCG/HCC) payment models summarize the health care problems and predict the future health care costs of populations. These models use the diagnoses generated during patient encounters with the medical delivery system to infer which medical problems are present. Patient demographics and diagnostic profiles are, in turn, used to predict costs. We describe the logic, structure, coefficients and performance of DCG/HCC models, as developed and validated on three important data bases (privately insured, Medicaid, and Medicare) with more than 1 million people each.

Principal inpatient diagnostic cost group model for Medicare risk adjustment.

The Balanced Budget Act (BBA) of 1997 required HCFA to implement health-status-based risk adjustment for Medicare capitation payments for managed care plans by January 1, 2000. In support of this mandate, HCFA has been collecting inpatient encounter data from health plans since 1997. These data include diagnoses and other information that can be used to identify chronic medical problems that contribute to higher costs, so that health plans can be paid more when they care for sicker patients. In this article, the authors describe the risk-adjustment model HCFA is implementing in the year 2000, known as the Principal Inpatient Diagnostic Cost Group (PIPDCG) model.

The impact of reimbursement changes on doctors' office testing.

Most of the recent changes that have revolutionized the way health care is financed include powerful pecuniary incentives to move care out of the hospital and into such outpatient settings as doctors' offices. However, many of these changes also include aggressive utilization reviews with stiff penalties for noncompliance. Thus, although doctors' office testing will continue to expand, this growth may bring more external oversight into doctors' offices.

Predicting in-hospital mortality for stroke patients: results differ across severity-measurement methods.

OBJECTIVE: To see whether severity-adjusted predictions of likelihoods of in-hospital death for stroke patients differed among severity measures. METHODS: The study sample was 9,407 stroke patients from 94 hospitals, with 916 (9.7%) in-hospital deaths. Probability of death was calculated for each patient using logistic regression with age-sex and each of five severity measures as the independent variables: admission MedisGroups probability-of-death scores; scores based on 17 physiologic variables on admission; Disease Staging's probability-of-mortality model; the Seventy Score of Patient Management Categories (PMCs); and the All Patient-Refined Diagnosis Groups (APR-DRGs). For each patient, the odds of death predicted by the severity measures were compared. The frequencies of seven clinical indicators of poor prognosis in stroke were examined for patients with very different odds of death predicted by different severity measures. Odds ratios were considered very different when the odds of death predicted by one severity measure was less than 0.5 or greater than 2.0 of that predicted by a second measure. RESULTS: MedisGroups and the physiology scores predicted similar odds of death for 82.2% of the patients. MedisGroups and PMCs disagreed the most, with very different odds predicted for 61.6% of patients. Patients viewed as more severely III by MedisGroups and the physiology score were more likely to have the clinical stroke findings than were patients seen as sicker by the other severity measures. This suggests that MedisGroups and the physiology score are more clinically credible. CONCLUSIONS: Some pairs of severity measures ranked over 60% of patients very differently by predicted probability of death. Studies of severity-adjusted stroke outcomes may produce different results depending on which severity measure is used for risk adjustment.

Does severity explain differences in hospital length of stay for pneumonia patients?

OBJECTIVES: In the USA, the role of patient severity in determining hospital resource use has been questioned since Medicare adopted prospective hospital payment based on diagnosis-related groups (DRGs). Exactly how to measure severity, however, remains unclear. We examined whether assessments of severity-adjusted hospital lengths of stay (LOS) varied when different measures were used for severity adjustment. METHODS: The complete study sample included 18,016 patients receiving medical treatment for pneumonia at 105 acute care hospitals. We studied 11 severity measures, nine based on patient demographic and diagnosis and procedure code information and two derived from clinical findings from the medical record. For each severity measure, LOS was regressed on patient age, sex, DRG, and severity score. Analyses were performed on trimmed and untrimmed data. Trimming eliminated cases with LOS more than three standard deviations from the mean on a log scale. RESULTS: The trimmed data set contained 17,976 admissions with a mean (S.D.) LOS of 8.9 (6.1) days. Average LOS ranged from 5.0-11.8 days among the 105 hospitals. Using trimmed data, the 11 severity measures produced R-squared values ranging from 0.098-0.169 for explaining LOS for individual patients. Across all severity measures, predicted average hospital LOS varied much less than the observed LOS, with predicted mean hospital LOS ranging from about 8.4-9.8 days. DISCUSSION: No severity measure explained the two-fold differences among hospitals in average LOS. Other patient characteristics, practice patterns, or institutional factors may cause the wide differences across hospitals in LOS.

Use of follow-up services by patients referred from a walk-in unit: how can patient compliance be improved?

Timely use of necessary follow-up services is an important dimension of ambulatory care quality. Using a hospital-based walk-in center, this study identified patients who were referred for follow-up care and examined factors related to compliance with these referrals. The participants were 696 adults seen in a hospital-based walk-in unit between June 1, 1992, and December 1, 1992. Patients completed a self-administered questionnaire including questions about sociodemographic characteristics, prior use of health services, and the Medical Outcomes Study (MOS) 36-Item Health Survey. Medical findings, follow-up recommendations, insurance status, and compliance with follow-up referrals were ascertained using chart review, the hospital's computing system, and clinic records. Fifty percent of the patients were referred for follow-up medical care; 55% of these complied with follow-up referrals. Factors associated with referral for follow-up care included older age, inability to afford a physician, longer duration of chief complaint, the patient's belief that follow-up care would be needed, and worse MOS pain score. The most important factor associated with compliance with follow-up referral was scheduling appointments while patients were still in the walk-in unit. Patients with such scheduled appointments were almost 10 times more likely than others to receive follow-up (adjusted odds ratio = 9.6, 95% confidence interval = 4.4-21.2). The most important step a provider can take to improve compliance with follow-up referral is to schedule appointments before patients are sent home. This should presumably improve quality of ambulatory care.

Risk adjustment methods can affect perceptions of outcomes.

When comparing outcomes of medical care, it is essential to adjust for patient risk, including severity of illness. A variety of severity measures exist, but perceptions of outcomes may differ depending on how severity is defined. We used two severity-adjustment approaches to demonstrate that comparisons of outcomes across subgroups of patients can vary dramatically depending on how severity is assessed. We studied two approaches: model 1 was the admission MedisGroups score; model 2 was computed from age and 12 chronic conditions defined by diagnosis codes. Although common summary measures of model performance (R-squared and C) both suggested that model 1 is a better predictor of in-hospital death than model 2, the weaker model consistently produced more accurate expectations by payer class and age group. Using model 1 for severity adjustment suggested that Medicare patients did substantially worse than expected and Medicaid patients substantially better. In contrast, use of model 2 found Medicare patients doing as expected, but Medicaid patients faring poorly.

Use of screening and preventive services among women with disabilities.

Roughly 54 million Americans have some disability; at older ages, women are more likely to be disabled than men. Many people with disabilities today live virtually normal life spans, and therefore routine screening and preventive services are essential to their overall quality of care. We used the 1994-1995 National Health Interview Survey (NHIS), with Disability, Family Resources, and Healthy People 2000 supplements, to examine screening and preventive service use for adult women with disabilities living in the community--about 18.4% of women (estimated 18.28 million). Disability was associated with higher age-adjusted rates of: poverty; living alone; low education; inability to work; obesity; and being frequently depressed or anxious. Disabled women generally reported screening and preventive services at rates comparable to all women. Women with major lower extremity mobility difficulties had much lower adjusted odds of Papanicolaou smears (odds ratio, 0.6; 95% confidence interval, 0.4-0.9), mammograms (odds ratio, 0.7; 95% confidence interval, 0.5-0.9), and smoking queries (odds ratio, 0.6; 95% confidence interval, 0.5-0.8). Various approaches exist to improve access for disabled women to health care services.

Costs of potential complications of care for major surgery patients.

We examined computerized hospital discharge abstract data from 372,680 major surgery patients admitted to 404 California acute care hospitals in 1988 to identify potential complications of care. At least one potential in-hospital complication occurred for 10.8% of patients. Patients with complications were older and more likely to die in-hospital (9.4% compared to 1.0%, P < 0.0001). On average, patients with complications had longer stays (13.5 versus 5.4 days, p < 0.0001) and higher total charges ($30,896 versus $9,239, p < 0.0001). After adjusting for demographic, clinical, and hospital factors, patients with potential complications averaged $16,023 higher total hospital charges than uncomplicated patients. Complications were associated with 96.6% (95% confidence interval = 95.2%, 98.0%) higher hospital charges after adjusting for these factors. Across all patients, complications were related to over $647 million in additional total hospital charges for these major surgery patients.

Patient-centered quality measures for Asian Americans: research in progress.

We aim to develop and validate a questionnaire that examines quality of care from the patient's perspective for limited-English-proficient Asian Americans (AA) of Chinese and Vietnamese descent. We will conduct focus groups of patients to identify issues important to them, with an emphasis on communication and access to care. We will then draft a questionnaire and test its validity using standard survey research methods and direct observation of patient-provider encounters. Subsequent field testing will involve face-to-face patient interviews 1 month after an outpatient visit. We will evaluate alternate modes of administration to test feasibility and to maximize response. The result of our study will be a validated, culturally sensitive, patient-centered instrument that measures health care quality for limited-English-proficient AA patients. Our research will provide a template for developing future quality measures for other vulnerable populations.