Hospital Based Inpatient Psychiatric Services (HBIPS)

---

Hospital-Based Inpatient Psychiatric Services (HBIPS) Measure Set Initial Patient Population

The HBIPS measure set is unique in that there are two distinct Initial Patient Populations within the measure set, one for the discharge measures (HBIPS-1, HBIPS-4, HBIPS-5, HBIPS-6, and HBIPS-7) and the other for event measures (HBIPS-2 and HBIPS-3).

Initial Patient Population for Discharge Measures (HBIPS-1, HBIPS-4, HBIPS-5, HBIPS-6, and HBIPS-7)

The general population of the HBIPS discharge measures can be identified by using three data elements that are common to the discharge performance measures in the HBIPS set:

• ICD-9-CM Principal Diagnosis Code or ICD-9-CM Other Diagnosis Codes
• Discharge Date
• Birthdate
• Psychiatric Care Setting

The HBIPS Discharge Topic Population is defined as patients discharged from the Psychiatric Care Setting with an ICD-9-CM Principal or Other Diagnosis Code for Mental Disorders as defined in Appendix A, Table 10.01 and a Patient Age at Discharge (Discharge Date -– Birthdate) >= 1 year. The HBIPS Discharge population is known as HBIPS-DSC. When pharmacy logs are used to determine the HBIPS Discharge population, HBIPS-DSC is broken into 2 sub-populations HBIPS-NORX and HBIPS-RX. HBIPS-DSC or HBIPS-NORX/HBIPS-RX are further stratified by age into 4 strata.

Two approaches are available to identify the HBIPS cases for the discharge measures:

• The first approach uses the general HBIPS populating criteria of ICD-9-CM Principal or Other Diagnosis Codes on Table 10.01, Patient Age and Psychiatric Care Setting without using pharmacy logs. All cases are a part of the HBIPS-DSC population and are reported to HCD and PaS using ‘HBIPS-DSC’ as the measure set, sampled using the HBIPS-DSC sample size table, and processed through the 5 discharge measures (HBIPS-1, 4, 5, 6, and 7).

• The second approach uses pharmacy logs to identify cases discharged on at least one antipsychotic medication for HBIPS-4 and 5. When using this approach to identify HIBPS cases, the discharge group is further divided into two sub-populations:
  • The HBIPS-NORX sub-population identifies cases for the discharge measures which do not evaluate patients being discharged on antipsychotic medications (HBIPS-1, HBIPS-6, and HBIPS-7). The criteria for identifying cases for this sub-population are similar to HBIPS-DSC in the first approach because only ICD-9-CM Principal or Other Diagnosis Codes, Patient Age and Psychiatric Care Setting are used to identify the sub-population and identical sample size table is used. All cases reported for the HBIPS-NORX sub-population are reported to HCD and PaS using ‘HBIPS-NORX’ as the measure set, sampled using the HBIPS-NORX sample size table, and processed through HBIPS-1, 6, and 7. The only difference between HBIPS-DSC and HBIPS-NORX is when submitting HBIPS-DSC, all 5 discharge measures (HBIPS-1, 4, 5, 6, and 7) would be processed, but when submitting HBIPS-NORX only HBIPS-1, 6 and 7 are processed.
  • The HBIPS-RX sub-population requires the use of pharmacy logs to identify patients discharged on one or more antipsychotic medications (Number of Antipsychotic Medications Prescribed at Discharge) for HBIPS-4 and 5. The purpose of using pharmacy logs is to exclude cases where no antipsychotic medications were prescribed at discharge from the initial population, thus providing a more accurate measurement of the initial population and potentially increasing the denominator count for these measures. Both electronic and paper pharmacy logs may be used; however, it is expected that hospitals will not implement this approach if, during their analysis, they determine the approach will create unreasonable additional burden. All cases reported for the HBIPS-RX sub-population are reported to HCD and PaS using ‘HBIPS-RX’ as the measure set, sampled using the HBIPS-RX sample size table, and processed through HBIPS-4 and 5.

Both sub-populations in this second approach can be sampled and a different sample size table (HBIPS-NORX and HBIPS-RX) is used for the two sub-populations. DO NOT automatically include cases sampled for HBIPS-RX into the HBIPS-NORX sampled group. These two sup-populations must be sampled separately; therefore, sampled cases may be reported as part of HBIPS-RX and/or HBIPS-NORX.

In summary, this second approach uses the data element Number of Antipsychotic Medications Prescribed at Discharge to determine which sub-population the case belongs:

• If the case has no antipsychotic medications, the case belongs only in the HBIPS-NORX sub-population, is sampled using the HBIPS-NORX sample size table, and only processed through measures HBIPS-1, 6 and 7. This case is not processed through HBIPS-4 and 5.
• If the case has one or more antipsychotic medications, the case belongs to both sub-populations, HBIPS-NORX and HBIPS-RX, and is sampled separately for each sub-population.

Note: A single hospital cannot mix these two approaches when reporting for one quarter of data. However, the approach could be changed from one quarter to next. Therefore, for one quarter a hospital can report their HBIPS cases either as HBIPS-DSC measure set OR as HBIPS-RX and HBIPS-NORX measure sets.

There are four distinct strata within the HBIPS Discharge Topic Population; each is identified by a specific age range. The patients in each stratum are counted in the HBIPS Initial Patient Population for discharge measures of multiple measures.

Discharge Measures	Age Strata	Initial Patient Population definition
HBIPS-1a, 4a, 5a, 6a, and 7a (overall measures)	Age greater than and equal to 1 year	The count of all patients in strata 1, 2, 3, and 4
HBIPS-1b, 4b, 5b, 6b, and 7b	Age 1 year through 12 years	The count of all patients in stratum 1
HBIPS-1c, 4c, 5c, 6c, and 7c	Age 13 years through 17 years	The count of all patients in stratum 2
HBIPS-1d, 4d, 5d, 6d, and 7d	Age 18 years through 64 years	The count of all patients in stratum 3
HBIPS-1e, 4e, 5e, 6e, and 7e	Age greater than and equal to 65 years	The count of all patients in stratum 4

Patients discharged from the hospital with an ICD-9-CM Principal or Other Diagnosis Code for Psychiatric Services as defined in Appendix A, Table 10.01 are included in one of the HBIPS Strata Initial Populations for discharge measures and are eligible to be sampled if they have:

Discharge Stratum 1 – Age 1 year through 12 years stratum – A Patient Age at Discharge (Discharge Date -– Birthdate) >= 1 year and < 13 years

Discharge Stratum 2 - Age 13 years through 17 years stratum – A Patient Age at Discharge (Discharge Date -– Birthdate) >= 13 years and < 18 years

Discharge Stratum 3 - Age 18 years through 64 years stratum – A Patient Age at Discharge (Discharge Date -– Birthdate) >= 18 years and < 65 years

Discharge Stratum 4 - Age greater than and equal to 65 years stratum – A Patient Age at Discharge (Discharge Date -– Birthdate) >= 65 years

---

Initial Patient Population for Event Measures (HBIPS-2 and HBIPS-3)

The population of the HBIPS event measures can be identified by using two data elements that are common to the event performance measures in the HBIPS set:

• Event Date
• Psychiatric Care Setting

The HBIPS Event Topic Population (common to all HBIPS event measures) is defined as patients with an event (Event Date exists) while they are in the hospital with a Patient Age at Time of Event (Event Date -– Birthdate) >= 1 year and the patient was in a Psychiatric Care Setting (=’Y’) when the event occurred. There are four distinct strata or sub-populations within the HBIPS Event Topic Population, each identified by a specific age range. The patients in each stratum are counted in the HBIPS Initial Patient Population for event measures of multiple measures.

Event Measures	Age Strata	Initial Patient Population definition
HBIPS-2a and 3a (overall measures)	Age greater than and equal to 1 year	The count of all patients in strata 1, 2, 3, and 4
HBIPS-2b and 3b	Age 1 year through 12 years	The count of all patients in stratum 1
HBIPS-2c and 3c	Age 13 years through 17 years	The count of all patients in stratum 2
HBIPS-2d and 3d	Age 18 years through 64 years	The count of all patients in stratum 3
HBIPS-2e and 3e	Age greater than and equal to 65 years	The count of all patients in stratum 4

Patients for which an event occurs (Event Date exists) while in a Psychiatric Care Setting (=’Y’) in the hospital are included in one of the Strata Initial Populations for the event measures. There is no sampling for the HBIPS event measures. All patients in the Initial Population for HBIPS event measures are automatically sampled.

Event Stratum 1 – Age 1 year through 12 years stratum – A Patient Age at Time of Event (Event Date -– Birthdate) >= 1 year and < 13 years

Event Stratum 2 - Age 13 years through 17 years stratum – A Patient Age at Time of Event (Event Date –- Birthdate) >= 13 years and < 18 years

Event Stratum 3 - Age 18 years through 64 years stratum – A Patient Age at Time of Event (Event Date -– Birthdate) >= 18 years and < 65 years

Event Stratum 4 - Age greater than and equal to 65 years stratum – A Patient Age at Time of Event (Event Date –- Birthdate) >= 65 years

---

---

---

---

---

---

Sample Size Requirements

Note For Joint Commission purposes, the HBIPS measure set is not included in the aligned Global Sampling methodology. All patients meeting the definition of the HBIPS Initial Patient Populations are eligible to be sampled, abstracted, and transmitted to the Joint Commission’s Data Warehouse.

Sample Size Requirements for HBIPS Discharge Measures (HBIPS-1, HBIPS-4, HBIPS-5, HBIPS-6, and HBIPS-7)
Hospitals that choose to sample have the option of sampling quarterly or sampling monthly. A hospital may choose to use a larger sample size than is required. Hospitals whose Initial Patient Population size is less than the minimum number of cases per quarter/month for the stratum cannot sample that stratum.

Regardless of the option used, hospital samples must be monitored to ensure that sampling procedures consistently produce statistically valid and useful data. Because the sample for a measure set will rarely be equal to the effective sample due to exclusions and contraindications, hospitals selecting sample cases MUST submit AT LEAST the minimum required sample size.

The following sample size tables for each option automatically build in the number of cases needed to obtain the required sample sizes. For information concerning how to perform sampling, refer to the Population and Sampling Specifications section in this manual. Furthermore, since the availability or lack of pharmacy logs provides two different population sizes for HBIPS-4 and HBIPS-5 (grouped as HBIPS-NORX and HBIPS-RX), two different sample size tables have been developed.

Hospitals that choose to use Pharmacy logs to determine their HBIPS-4 and HBIPS-5 population would need to use the sample size tables for those with access to HBIPS pharmacy data (HBIPS-RX). This assumes that the hospital can identify their initial patient population and then use pharmacy data to identify those patients discharged with one or more antipsychotic medications. Using this as their initial population for HBIPS-4 and HBIPS-5 they would then use the sampling tables to determine the minimum sample size and sample from that identified population for these two measures.

Sampling for Hospitals Not Using Pharmacy Logs

Hospitals not using pharmacy logs need to use the following tables when determining their sample size for HBIPS-DSC cases. Every case sampled from this group is submitted once with all measures in HBIPS-DSC (HBIPS-1, HBIPS-4, HBIPS-5, HBIPS-6, and HBIPS-7).

Quarterly Sampling

For hospitals selecting sample cases for the HBIPS discharge measures, a modified sampling procedure is required. Hospitals selecting sample cases for this set must ensure that each individual stratum’'s population and effective quarterly sample size meets the following conditions:

• Select within each of the four individual measure strata. The effective quarterly sample size within a stratum is at least 44 cases per quarter. Cases are placed into the appropriate stratum based upon the patient’'s age.
• The required quarterly sample size is at least 20% of the stratum population for the quarter.

Quarterly Sample Size
Based on Initial Patient Population for the HBIPS-DSC Discharge Measures

Hospital's Measures
Average Quarterly Stratum Initial Patient Population Size “N”	Minimum Required Stratum Sample Size “n”
> 877	176
221 –- 877	20% of the Initial Patient Population
44 -– 220	44
< 44	No sampling; 100% of the Initial Patient Population is required

Monthly Sampling

For hospitals selecting sample cases for HBIPS discharge measures, a modified sampling procedure is required. Hospitals selecting sample cases for this set must ensure that each individual strata population and effective monthly sample size meets the following conditions:

• Select within each of the four individual measure strata. The effective monthly sample size within a stratum is at least 15 cases per month. Cases are placed into the appropriate stratum based upon the patient'’s age.
• The required monthly sample size is at least 20% of the stratum population for the month.

Monthly Sample Size
Based on Initial Patient Population for the HBIPS-DSC Discharge Measures

Hospital's Measures
Average Monthly Stratum Initial Patient Population Size “N”	Minimum Required Stratum Sample Size “n”
> 295	60
76 –- 295	20% of the Initial Patient Population
15 –- 75	15
< 15	No sampling; 100% of the Initial Patient Population is required

Sampling for Hospitals Using Pharmacy Logs

Hospitals using pharmacy logs need to use two sets of tables when determining their sample size for two different sub-populations. One set of sample size tables for HBIPS-NORX cases and a different set of sample size tables for HBIPS-RX cases.

Sampling for HBIPS-NORX Sub-population

Hospitals will use the following tables when determining their sample size for HBIPS-NORX cases. Every case sampled from this group is submitted once with all measures in HBIPS-NORX (HBIPS-1, HBIPS-6, and HBIPS-7). Note: the HBIPS-NORX sample size tables are the same as those used when pharmacy logs are not used and HBIPS-DSC is submitted.

Quarterly Sampling

For hospitals selecting sample cases for the HBIPS-NORX discharge measures, a modified sampling procedure is required. Hospitals selecting sample cases for this set must ensure that each individual stratum's population and effective quarterly sample size meets the following conditions:

• Select within each of the four individual measure strata. The effective quarterly sample size within a stratum is at least 44 cases per quarter. Cases are placed into the appropriate stratum based upon the patient's age.
• The required quarterly sample size is at least 20% of the stratum population for the quarter.

Quarterly Sample Size
Based on Initial Patient Population for the HBIPS-NORX Discharge Measures

Hospital's Measures
Average Quarterly Stratum Initial Patient Population Size “N”	Minimum Required Stratum Sample Size “n”
> 877	176
221 -– 877	20% of the Initial Patient Population
44 -– 220	44
< 44	No sampling; 100% of the Initial Patient Population is required

Monthly Sampling

For hospitals selecting sample cases for HBIPS-NORX discharge measures, a modified sampling procedure is required. Hospitals selecting sample cases for this set must ensure that each individual strata population and effective monthly sample size meets the following conditions:

• Select within each of the four individual measure strata. The effective monthly sample size within a stratum is at least 15 cases per month. Cases are placed into the appropriate stratum based upon the patient’s age.
• The required monthly sample size is at least 20% of the stratum population for the month.

Monthly Sample Size
Based on Initial Patient Population for the HBIPS-NORX Discharge Measures

Hospital's Measures
Average Monthly Stratum Initial Patient Population Size “N”	Minimum Required Stratum Sample Size “n”
> 295	60
76 -– 295	20% of the Initial Patient Population
15 –- 75	15
< 15	No sampling; 100% of the Initial Patient Population is required

Sampling for HBIPS-RX Sub-population

Based on the Initial Patient Population and pharmacy data for HBIPS-4 and HBIPS-5, hospitals will use the following tables when determining their sample size for HBIPS-RX cases. Every case sampled from this group is submitted once with all measures in HBIPS-RX (HBIPS-4 and HBIPS-5).

Quarterly Sampling

For hospitals selecting sample cases for the HBIPS-RX discharge measures, a modified sampling procedure is required. Hospitals selecting sample cases for this set must ensure that each individual stratum's population and effective quarterly sample size meets the following conditions:

• Select within each of the four individual measure strata. The effective quarterly sample size within a stratum is at least 26 cases per quarter. Cases are placed into the appropriate stratum based upon the patient's age.
• The required quarterly sample size is at least 20% of the stratum population for the quarter.

Quarterly Sample Size
Based on Initial Patient Population and Pharmacy Data for the HBIPS-RX Discharge Measures

Hospital's Measures
Average Quarterly Stratum Initial Patient Population Size “N”	Minimum Required Stratum Sample Size “n”
> 515	104
131 - 515	20% of the Initial Patient Population
26 - 130	26
< 26	No sampling; 100% of the Initial Patient Population is required

Monthly Sampling

For hospitals selecting sample cases for HBIPS-RX discharge measures, a modified sampling procedure is required. Hospitals selecting sample cases for this set must ensure that each individual strata population and effective monthly sample size meets the following conditions:

• Select within each of the four individual measure strata. The effective monthly sample size within a stratum is at least 9 cases per month. Cases are placed into the appropriate stratum based upon the patient’s age.
• The required monthly sample size is at least 20% of the stratum population for the month.

Monthly Sample Size
Based on Initial Patient Population and Pharmacy Data for the HBIPS-RX Discharge Measures

Hospital's Measures
Average Monthly Stratum Initial Patient Population Size “N”	Minimum Required Stratum Sample Size “n”
> 175	36
46 - 175	20% of the Initial Patient Population
9 - 45	9
< 9	No sampling; 100% of the Initial Patient Population is required

Sample Size Examples

All sampled strata in HBIPS should be used in the calculation of all HBIPS discharge measures. All of the HBIPS discharge measures' specific exclusion criteria are used to filter out cases that do not belong in the measure denominator. Using HBIPS-1b as an example, include cases covering all sampled strata, although the measure-specific exclusion criteria would only allow cases with an age of 1 year through 12 years to be included in the denominator.

• Example 1: Quarterly sampling for a hospital not using Pharmacy logs for HBIPS-4 and HBIPS-5 :
  When applicable, larger hospitals must also abide by the required quarterly sample sizes for the four individual measure strata a minimum of 44 or 20% of population required sample cases per stratum when Initial Patient Population size is 44 or greater.
  
  • The HBIPS Initial Patient Population sizes for a hospital are 5, 100, 221, and 876 patients for each stratum respectively per quarter. The required quarterly sample sizes would be 5, 44, 45, and 176.
    • The 1st stratum is less than the minimum required quarterly sample size, so 100% of this stratum is sampled.
    • The 2nd stratum has 100 patients per quarter, which falls in the average quarterly population size of 44 to 220 patients, so 44 cases are sampled.
    • The 3rd stratum has 221 patients per quarter, which requires a 20% sample size, of 45 cases (twenty percent of 221 equals 44.2 rounded to the next highest whole number = 45).
    • The 4th stratum has 876 patients per quarter, which is more than the maximum condition, so a minimum of 176 cases are required to be sampled.

• Example 2: Monthly sampling for a hospital not using Pharmacy logs for HBIPS-4 and HBIPS-5 :
  When applicable, larger hospitals must also abide by the required monthly sample sizes for the four individual measure strata a minimum of 15 required sample cases per stratum when Initial Patient Population size is 15 or greater.
  
  • The HBIPS Initial Patient Population sizes for a hospital are 5, 45, 294 and 400 patients respectively in July. The required monthly sample sizes would be 5, 15, 59, and 60.
    • The 1st stratum is less than the minimum required monthly sample size, so 100% of this stratum is sampled.
    • The 2nd stratum has 45 patients per month, which falls in the average monthly population size of 15 to 75 patients, so 15 cases are sampled.
    • The 3rd stratum has 294 patients per month, which requires a 20% sample size, of 59 cases (twenty percent of 294 equals 58.8 rounded to the next highest whole number = 59).
    • The 4th stratum has 400 patients per month, which is more than the maximum condition, so a minimum of 60 cases are required to be sampled.

• Example 3: Quarterly sampling for a hospital which is using Pharmacy logs for HBIPS-4 and HBIPS-5 :
  When applicable, larger hospitals must also abide by the required quarterly sample sizes for the four individual measure strata a minimum of 44 or 20% of population required sample cases per stratum when Initial Patient Population size is 44 or greater for HBIPS-NORX Sub-population, and minimum of 26 or 20% of population required sample cases per stratum when Initial Patient Population size is 26 or greater for HBIPS-RX Sub-population.
  
  • The HBIPS Initial Patient Population sizes for a hospital are 5, 100, 221, and 876 patients for each stratum respectively per quarter. From these general HBIPS Populations 2, 65, 132 and 525 have at least one antipsychotic medication respectively per quarter, Therefore the required quarterly sample sizes would be 2, 26, 27, and 104 for HBIPS-RX sub-population and for HBIPS-NORX sub-populations the required quarterly sample sizes would be 5, 44, 45, and 176, Additional detail of to calculating the sample size:
    
    • The 1st stratum is also less than the minimum required quarterly sample size of 45 for HBIPS-NORX, so 100% of this stratum is sampled and 5 cases have to be submitted.
    • The 1st stratum is less than the minimum required quarterly sample size of 26 for HBIPS-RX, so 100% of this stratum is sampled and 2 cases have to be submitted.
      
      
    • The 2nd stratum has 100 patients per quarter, which falls in the average quarterly HBIPS-NORX sub-population size of 44 to 220 patients, so 44 cases are sampled for HBIPS-NORX sub-population.
    • The 2nd stratum has 65 patients per quarter for HBIPS-RX sub-population (patients with at least one antipsychotic medication) and 26 cases are sampled for HBIPS-RX.
      
      
    • The 3rd stratum has 221 patients per quarter for HBIPS-NORX sub-population, which requires a 20% sample size of 45 cases (twenty percent of 221 equals 44.2 rounded to the next highest whole number = 45).
    • The 3rd stratum has 132 patients per quarter for HBIPS-RX sub-population (patients with at least one antipsychotic medication) , which requires a 20% sample size of 27 cases (twenty percent of 132 equals 26.4 rounded to the next highest whole number = 27).
      
      
    • The 4rd stratum has 876 patients per quarter for HBIPS-NORX sub-population, which requires a 20% sample size of 175 cases (twenty percent of 876 equals 175.2 rounded to the next highest whole number = 176).
    • The 4rd stratum has 525 patients per quarter for HBIPS-RX sub-population (patients with at least one antipsychotic medication) , which requires a 20% sample size of 105 cases (twenty percent of 525 equals 105 )

• Example 4: Monthly sampling for a hospital which is using Pharmacy logs for HBIPS-4 and HBIPS-5 :
  When applicable, larger hospitals must also abide by the required monthly sample sizes for the four individual measure strata a minimum of 15 or 20% of population required sample cases per stratum when Initial Patient Population size is 15 or greater for HBIPS-NORX Sub-population, and minimum of 9 or 20% of population required sample cases per stratum when Initial Patient Population size is 9 or greater for HBIPS-RX Sub-population.
  
  • The HBIPS Initial Patient Population sizes for a hospital are 5, 45, 102 and 400 patients for each stratum respectively in July. From these general HBIPS Populations 2, 27, 61 and 240 have at least one antipsychotic medication respectively for each stratum in July, And the required monthly sample sizes would be 5, 15, 21, and 60 for HBIPS-NORX sub-population. The required monthly sample sizes would be 2, 9, 13, and 36 for HBIPS-RX sub-population. Following is the details of calculating the sample size:
    
    • The 1st stratum is also less than the minimum required quarterly sample size of 15 for HBIPS-NORX, so 100% of this stratum is sampled and 5 cases have to be submitted.
    • The 1st stratum is less than the minimum required monthly sample size of 9 for HBIPS-NORX, so 100% of this stratum is sampled and 2 cases have to be submitted.
      
      
    • The 2nd stratum has 45 patients in July , which falls in the average monthly HBIPS-NORX sub-population size of 15 to 75 patients, so 15 cases are sampled for HBIPS-NORX sub-population.
    • The 2nd stratum has 9 patients in July for HBIPS-RX sub-population (patients with at least one antipsychotic medication) and 9 cases are sampled for HBIPS-RX.
      
      
    • The 3rd stratum has 102 patients in July for HBIPS-NORX sub-population, which requires a 20% sample size of 21 cases (twenty percent of 102 equals 20.4 rounded to the next highest whole number = 21).
    • The 3rd stratum has 61 patients in July for HBIPS-RX sub-population (patients with at least one antipsychotic medication) , which requires a 20% sample size of 13 cases (twenty percent of 61 equals 12.2 rounded to the next highest whole number = 13).
      
      
    • The 4rd stratum has 400 patients in July for HBIPS-NORX sub-population, which is more than the maximum condition, so a minimum of 60 cases are required to be sampled.
    • The 4th stratum has 240 patients in July for HBIPS-RX sub-population (patients with at least one antipsychotic medication), which is more than the maximum condition, so a minimum of 36 cases are required to be sampled.

Sampling Size Requirements for HBIPS Event Measures (HBIPS-2 and HBIPS-3)
The measures in HBIPS-EVT (HBIPS-2 and HBIPS-3) are not eligible for sampling and will use the entire Initial Patient Population for reporting.